Ecologia Mediterranea - Université d`Avignon

Transcription

6 mm
(96 p)
ecologia mediterranea
Vol. 41 (2) – 2015
00-ecol-med-vol41(2)-couv-corTC_Mise en page 1 27/01/16 08:33 Page1
Vol. 41 (2) – 2015
Sommaire – Contents
Éditorial – Editorial
..........................................................................................
3
MEDECOS special issue
Seed Bank Divergence Between Arctostaphylos Adans. (Ericaceae)
and Ceanothus L. (Rhamnaceae) Suggests Different Seed Predator Interactions
Developing Allometric Volume-Biomass Equations to Support Fuel Characterization
in North-Eastern Spain
V. THOMAS PARKER
...............................................................................................
Studying Shoot and Root Architecture and Growth of Quercus ithaburensis
subsp. macrolepis Seedlings; a Key Factor for Successful Restoration
of Mediterranean Ecosystems
T. TSITSONI, N. GOUNARIS, A. B. KONTOGIANNI, V. XANTHOPOULOU-TSITSONI
33
.......................................................
Creation of an Integrated System Model for Governance in Urban MTEs
(Mediterranean-Type Ecosystems) and for Adapting Cities
to Climate Change – Preliminary Results
T. TSITSONI, M. TSAKALDIMI, M. GOUSIOPOULOU
.........................................................
.......................
Vol. 41 (2) – 2015
Revue internationale d’écologie méditerranéenne
International Journal of Mediterranean Ecology
5
15
B. D. PEDRA, J. GODOY PUERTAS, L. FUENTES LOPEZ
ecologia
mediterranea
25
Vegetation Dynamics of Coastal Dunes with Juniperus spp. in Crete, Gavdos
and Chrysi Islands (Greece)
Caractérisation du fonctionnement des steppes d’Alfa marocaines
par la méthode de l’analyse fonctionnelle du paysage
............
45
M. DERAK, F. T. MAESTRE, J. L. QUERO, V. OCHOA, C. ESCOLAR,
S. SOLIVERES, P. GARCÍA-PALACIOS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Floristic Diversity Patterns in the Beni-Haoua Forest (Chlef, Algeria)
61
......................................
Insight into the Dietary Habits of the Eurasian Otter, Lutra lutra,
in the East of Algeria (El-Kala National Park)
73
............................................................................
Résumés de thèses – Ph. D summaries
85
..........................................................................................
Hommage à Jacques Gamisans
92
..........................................................................
95
P.DELIPETROU, D. GHOSN, G. KAZAKIS, P. NYKTAS, E. REMOUNDOU, I.N. VOGIATZAKIS
A. ABABOU, M. CHOUIEB, A. BOUTHIBA, D. SAIDI, K. MEDERBAL
R. LIBOIS, R. GHALMI, A. BRAHIMI
C. CALVET, T. FRÉJAVILLE
Revue indexée dans Pascal-CNRS et Biosis
ISSN 0153-8756
RESEARCH PAPERS
Editors-in-Chief: Dr Élise Buisson & Dr Brigitte Talon
Institut méditerranéen de biodiversité et écologie (IMBE)
Mediterranean Institute of Biodiversity and Ecology
Naturalia Publications
Éditrices en chef : Dr Élise Buisson
et Dr Brigitte Talon
UMR CNRS IRD IMBE
Université d’Avignon, IUT
Site Agroparc, BP 1207
84911 Avignon cedex 09
France
Instructions aux auteurs
ecologia mediterranea publie des articles de recherche originaux sur
des sujets se rapportant à l’écologie fondamentale ou appliquée des
régions méditerranéennes. La revue exclut les articles purement descriptifs ou de systématique. ecologia mediterranea privilégie les
domaines scientifiques suivants : bioclimatologie, biogéographie, biologie de la conservation, biologie marine, biologie des populations,
écologie des communautés, écologie forestière, écologie génétique,
écologie marine, écologie microbienne, écologie du paysage, écologie de la restauration, écologie végétale et animale, écophysiologie,
paléoclimatologie, paléoécologie. La revue accepte également des
articles de synthèse, des notes/communications courtes, des comptes
rendus d’ouvrages, des résumés de thèses, ainsi que des commentaires
sur les articles récemment parus dans ecologia mediterranea. La revue
publie aussi des actes de colloques faisant l’objet d’un numéro spécial. Dans ce cas, prendre contact avec les éditrices.
Comité éditorial
Dr Pierre CHEVALDONNÉ, CNRS, Université
Aix-Marseille, Marseille, France
Dr Marc CHEYLAN, EPHESS, Montpellier,
France
Dr Cécile CLARET, Université AixMarseille, Marseille, France
Dr Bruno FADY, INRA, Avignon, France
Pr Thierry GAUQUELIN, Université AixMarseille, Marseille, France
Dr Grant WARDELL-JOHNSON, Université
Western, Australie
Dr Raphaël GROS, Université AixMarseille, Marseille, France
Dr Frédéric GUITER, Université AixMarseille, Marseille, France
Pr Serge KREITER, SupAgro, Montpellier,
France
Dr Audrey MARCO, École nationale
supérieure du paysage, Marseille,
France
Pr Frédéric MÉDAIL, Université AixMarseille, Marseille, France
Pr François MESLÉARD, Université
Avignon-Tour du Valat, France
Dr Tom PARKER, San Francisco State
University, États-Unis
Dr Philippe PONEL, CNRS, Université AixMarseille, Marseille, France
Dr Roger PRODON, EPHE, Montpellier,
France
Dr Sandra SAURA-MAS, Autonomous
University of Barcelona, Espagne
Dr Isabelle SCHWOB, Université AixMarseille, Marseille, France
Dr Thekla K. TSITSONI, Aristotle
University of Thessaloniki, Grèce
Dr Éric VIDAL, IRD, France
Dr Mercedes VIVAS, Universidad of
Concepción, Chili
Dr Ioannis VOGIATZAKIS, Open University
of Cyprus, Chypre
Les manuscrits sont soumis à des lecteurs spécialistes du sujet. La
décision finale d’accepter ou de refuser un article relève des éditrices.
L’article proposé doit être envoyé en version électronique à [email protected] (version doc(x) ou rtf). Pour la mise en
forme du document, voir les instructions qui suivent. Une fois leur
article accepté, les auteurs devront tenir compte des remarques des lecteurs, puis ils renverront leur texte corrigé sous deux mois toujours
sous format électronique (doc(x) ou rtf). Passé ce délai, la seconde
version sera considérée comme une nouvelle proposition.
TYPES DE MANUSCRIT
À préciser sur la première page lors de la soumission d’un manuscrit.
Article de recherche : contribution inédite découlant d’une étude complète. Ce type d’article fait typiquement une vingtaine de pages et environ 6 000 à 8 000 mots.
Note/communication courte : observation nouvelle ou rapport d’expérience dans un contexte pertinent avec les sujets visés par la revue.
Ce type d’article fait typiquement une dizaine de pages et environ
3 000 à 4 000 mots.
Article de synthèse : revue critique et originale de sujets spécifiques
d’actualité ou d’un champ de recherche de pointe dans le domaine de
l’écologie méditerranéenne. Ce type d’article fait typiquement une
vingtaine de pages et environ 6 000 à 8 000 mots.
Commentaire : avis sur des sujets déjà publiés dans ecologia mediterranea ou réflexion critique sur des problèmes d’intérêt général en
écologie méditerranéenne. Ce type d’article fait typiquement une à
cinq pages et environ 1 000 à 3 000 mots.
Compte rendu d’ouvrage : revue critique d’ouvrages (livres, monographies, manuels, etc.) dans le domaine de l’écologie méditerranéenne. Les auteurs d’ouvrages souhaitant voir un compte rendu publié
dans ecologia mediterranea doivent envoyer un exemplaire de l’ouvrage en question aux éditrices en chef.
Résumé de thèse : résumé d’une thèse soutenue récemment dans le
domaine de l’écologie méditerranéenne. Auteur, année, titre, spécialité et université (e.g. Thèse de doctorat en écologie soutenue le 3 avril
2012 à l’université de Padoue, Italie, laboratoire xxx), composition du
jury, mots clés, résumé de 1 000 mots maximum.
TEXTE
Les articles (dactylographiés en Times 12, double interligne, format
A4) doivent être rédigés en anglais ou en français. Si l’article soumis
n’est pas rédigé en anglais, il est demandé (en plus des résumés) une
version anglaise abrégée ainsi qu’une traduction en anglais des titres
des figures et tableaux. L’article doit être complet : type de manuscrit,
titres anglais et français, auteur(s) et adresse(s), résumés en anglais et
en français (au minimum), version anglaise abrégée (si le texte n’est
pas en anglais), mots clés anglais et français, texte, puis remerciements, bibliographie, liste des titres des figures et tableaux puis les
figures et tableaux (un(e)/page). Ainsi, pour la soumission du manuscrit, les illustrations seront intégrées au document et non envoyées
ISSN 0153-8756
http://ecologia-mediterranea.
univ-avignon.fr
ABONNEMENT
(contact : [email protected])
France
Europe
Monde
Abonnement à adresser à :
SARL Transfaire
Immeuble Wanad
F-04250 TURRIERS
Frais de port
Total
60 €
60 €
60 €
6€
12 €
16 €
66 €
72 €
76 €
Code banque
19106
Code guichet
00839
numéro de compte
13995626000
Domiciliation : CA SISTERON
IBAN : FR76 1910 6008 3913 9956 2600 062
BIC : AGRIFRPP891
PREMIÈRE PAGE
La première page contient :
1) le type de manuscrit visé (article de recherche, communication
courte, etc.) ;
2) le titre de l’article ;
3) le nom et prénom des auteurs ;
4) l’adresse de chaque auteur sera indiquée avec le courriel de l’auteur pour la correspondance. Dans le cas où la publication est le fait
de plusieurs auteurs, il doit être précisé lors du premier envoi la personne à qui doit être retourné l’article après lecture ;
5) le nombre total de mots de l’introduction à la bibliographie.
RÉSUMÉS, MOTS CLÉS ET VERSION ABRÉGÉE
Les résumés doivent comporter 300 mots au maximum et la version
abrégée (français si le manuscrit est en anglais et anglais si le manuscrit est en français) 1 000 mots (environ une page). Le nombre de mots
clés est limité à six, dans la langue des résumés ; ils ne doivent généralement pas figurer dans le titre.
BIBLIOGRAPHIE
La bibliographie regroupera toutes les références citées et elles seules.
Les références seront rangées dans l’ordre alphabétique des auteurs et
de façon chronologique. Les abréviations internationales des titres des
revues doivent être utilisées (ISI Journal Abbreviations Index). Vérifier attentivement le manuscrit pour s’assurer que toutes les références
citées dans le texte apparaissent bien en bibliographie et inversement.
Article
Andow D.A., Karieva P., Levin S.A. & Okubo A., 1990. Spread of
invading organisms. J. Ecol. 4: 177-188.
Ouvrage
Harper J.L., 1977. Population biology of plants. Academic Press, London, 300 p.
Chapitre d’ouvrage
May R.M., 1989. Levels of organisation in ecology. In: Cherret J.M. (ed.),
Ecological concepts. Blackwell Scientific Public, Oxford: 339-363.
Acte de conférence
Grootaert P., 1984. Biodiversity in insects, speciation and behaviour
in Diptera. In: Hoffmann M. & Van der Veken P. (eds.), Proceedings
of the symposium on “Biodiversity: study, exploration, conservation”.
Ghent, 18 November 1992: 121-141.
Rapport et thèse
Jaouadi W., 2011. Écologie et dynamique de régénération de l’Acacia tortilis (Forsk.) Hayne subsp. raddiana (Savi) Brenan var. raddiana
dans le parc national de Bouhedma (Tunisie). Thèse de doctorat de
l’Institut national agronomique de Tunisie, 180 p.
Editors-in-Chief: Dr Élise Buisson
& Dr Brigitte Talon
UMR CNRS IRD IMBE
France
ecologia mediterranea publishes original research reports and
syntheses in the fields of fundamental and applied ecology of
Mediterranean areas, except for descriptive articles or articles
about systematic. The editors of ecologia mediterranea invite
original contributions in the fields of: bioclimatology, biogeography, conservation biology, marine biology, population biology,
community ecology, forest ecology, marine ecology, genetic ecology, landscape ecology, microbial ecology, restoration ecology,
plant and animal ecology, ecophysiology, palaeoecology, palaeoclimatology. The journal also publishes reviews, short communications, book reviews, Ph. D. thesis abstracts and comments on
papers recently published in the journal. ecologia mediterranea
invite conference organizers to get in touch with the editors for
special issues as part of conference/symposium proceedings.
Manuscripts are peer-reviewed by appropriate referees. The final
decision to accept or reject the manuscript is made by the editors.
To submit a paper, please send an electronic version of your paper
to [email protected] (doc(x) or rtf). Please read the
following guidelines to prepare your manuscript. When the article is accepted, the authors should take reviewer’s comments into
consideration. They will send back to the journal Editorial Office,
within 2 months. After this deadline, the manuscript will be considered as a new submission.
Editorial Board
France
Western, Australie
France
France
France
Concepción, Chili
of Cyprus, Chypre
ISSN 0153-8756
Les mots « figures » et « tableaux » annoncés dans le texte sont écrits
en toutes lettres et en minuscules. Indiquer le nom d’auteur et l’année
de publication (mais indiquer tous les auteurs dans la bibliographie).
Exemples : “Since Dupont (1962) has shown that...”, or “This is in
agreement with previous results (Durand et al. 1990; Dupond &
Dupont 1997) ...”. Le numéro de page de la citation n’est mentionné
que dans le cas où elle est entre guillemets. Si la publication est écrite
par plus de deux auteurs, le nom du premier doit être suivi par et al.
univ-avignon.fr
L’usage d’un acronyme ou d’une abréviation technique doit être précédé de sa signification lors de sa première apparition. Les codes de
nomenclature doivent être respectés selon les conventions internationales. Les mots latins doivent être mis en italiques (et al., a priori,
etc.), et en particulier les noms de plantes ou d’animaux. Lors de la
première apparition du nom d’une espèce, il est demandé d’y faire
figurer le nom d’auteur (exemple : Olea europaea L.).
FIGURES ET TABLEAUX
clé RIB
62
Guidelines for authors
CITATIONS ET RENVOIS APPELÉS DANS LE TEXTE
ABRÉVIATIONS, NOMENCLATURE ET MOTS LATINS
1 an = 2 numéros
Abonnement
séparément. Le texte des articles de recherche doit comporter quatre
parties non numérotées : introduction, méthodes, résultats, discussion.
Les sous-titres ne sont pas numérotés. Par exemple :
Introduction
Méthodes
Site d’étude
Plan expérimental
Analyses des données
Résultats
Discussion
L’emploi de mots soulignés est à proscrire. Les noms d’auteurs cités
figureront en minuscules dans le texte comme dans la bibliographie.
En français, n’utilisez les majuscules que pour les noms propres, sauf
exception justifiée. Les ponctuations doubles ( : ; ? ! ) sont précédées
d’un espace, contrairement aux ponctuations simples ( , . ). En
revanche, toutes les ponctuations sont suivies d’un espace. La mise en
forme définitive du texte sera assurée par la revue.
Pour la soumission du manuscrit, les illustrations seront intégrées au
document, à la suite des références bibliographiques (voir la partie
« Texte »). Une fois l’article accepté, les illustrations doivent être
envoyées séparément du texte, prêtes à l’impression (résolution 300
dpi). Tous les documents devant être insérés doivent être annoncés dans
le texte, numérotés dans l’ordre d’apparition et légendés.
TIRÉS À PART
Il n’est plus fourni de tirés à part mais un pdf par article.
TYPES OF MANUSCRIPTS
Please, mention the type of manuscript you are submitting on the
first page of your submission.
Research article: research report of broad scope that is an original contribution to science. The typical length of research papers
is about 6,000 to 8,000 words.
Short communication: brief report of new observations or of
important findings that deserve publication before broader studies are completed. The typical length of short communications is
about 3,000 to 4,000 words.
Reviews: critical appraisal of broad areas of investigation or research in Mediterranean ecology. The typical length of reviews is
Commentaries: opinion on topics recently published in ecologia
mediterranea or essays on topics of general interest in Mediterranean ecology. The typical length of short communications is
Book review: critical appraisal of a Book interesting to the readers of ecologia mediterranea. Editors who wish to see their book
reviewed in ecologia mediterranea should get in touch with the
editors of the journal.
Ph. D. thesis abstracts: abstract of the Ph. D. thesis recently
defended in the fields covered by ecologia mediterranea. Author,
Year, Title, University (e.g. Ph. D. thesis defended on April, 4th
2014 at University of California, Irwin, USA, Lab xxx), Composition of board, Keywords, Abstract (1,000 words max.).
MANUSCRIPT PREPARATION
Manuscripts (typewritten Times 12, with double line spacing)
must be written in English or in French. If the language is not
English, you should join an English short version and English
titles of figures and tables. The manuscript must be complete: e.g.
title in English and French, author(s) and address(es), abstract in
English and French, an English short version (if English is not the
language used in the article), keywords in English and French,
text, acknowledgements, references, figures and tables (one /
page). For the submission, figures and tables are integrated in the
SUBSCRIPTION
(contact: [email protected])
1 year = 2 issues
word document and not sent separately. For research papers, the
text should normally consist of 4 sections: introduction, methods,
results, discussion which are not numbered. Subtitles are not be
numbered either:
Introduction
Methods
Study site
Experimental design
Statistical analyses
Results
Discussion
Use lower-case letter type for names in the text and in the reference section. Do not underline any word. In English, there is one
blank after any punctuation, never before. Copy editing of manuscripts is performed by the journal.
FIRST PAGE
The first page contains:
1) the type of manuscript (research paper, short communication,
etc.);
2) the title of the article;
3) the names of the authors;
4) the address of each author + an email at least of the corresponding author;
5) the total number of words from the introduction to the references included.
ABSTRACTS, KEYWORDS AND SHORT VERSION
Abstracts should be no longer than 300 words. The English short
version should not exceed one page long (1,000 words). Do not
use more than six keywords. Keywords should not be present in
the title.
REFERENCES
All publications cited in the text should be presented in a list of
references following the text of the manuscript. The list of references should be arranged alphabetically on author’s names, and
chronologically for each author. You should abbreviate the titles
of periodicals mentioned in the list of references (except if you
are not sure of it; ISI Journal Abbreviations Index). Check the
manuscript to make sure that all references are cited and that all
citations in the text are included in the references. Use following
system to write the references:
Journal article
Andow D.A., Karieva P., Levin S.A. & Okubo A., 1990. Spread
of invading organisms. J. Ecol. 4: 177-188.
Book
Harper J.L., 1977. Population biology of plants. Academic Press,
London, 300 p.
Book chapters
May R.M., 1989. Levels of organisation in ecology. In: Cherret
J.M. (ed.), Ecological concepts. Blackwell Scientific Public., Oxford: 339-363.
Conference proceedings
Grootaert P., 1984. Biodiversity in insects, speciation and behaviour in Diptera. In: Hoffmann M. & Van der Veken P. (eds.), Proceedings of the symposium on “Biodiversity: study, exploration,
conservation”. Ghent, 18 November 1992: 121-141.
IN-TEXT CITATIONS
The words “figure” and “table” announced in-text should be written in extenso and with lower-case letter type. In the text refer to
the author’s name and year of publication (followed by pages only
if it is a quotation). If a publication is written by more than two
authors, the name of the first author should be used followed by
“et al.” (this indication, however, should never be used in the list
of references: first author and co-authors should be mentioned in
it). Examples: “Since Dupont (1962) has shown that…”, or “This
is in agreement with previous results (Durand et al. 1990; Dupond
& Dupont 1997)…”.
ABBREVIATES, NOMENCLATURE AND LATIN WORDS
Subscription
Postage
Total
60 €
60 €
60 €
6€
12 €
16 €
66 €
72 €
76 €
France
Europe
World
Explanation of a technical abbreviate is required when the first
use. International convention codes for nomenclature should be
used. Latin words should be in italic (et al., a priori, etc.), particularly for plants or animals’ denomination (the first time, please
precise author’s name: for example, Olea europaea L.).
FIGURES AND TABLES
Mail this subscription to:
SARL Transfaire
Immeuble Wanad
F-04250 TURRIERS
Bank: CA SISTERON
IBAN: FR76 1910 6008 3913 9956 2600 062
BIC: AGRIFRPP891
All illustrations should be included in the word document after
the section “References”. Once the paper is accepted for publication, illustrations should be sent separately (3,000 dpi). All the
illustrations should be cited and should have a legend.
REPRINTS
A pdf version will be supplied free of charge for each paper.
00-ecol-med-vol41(2)-debut-3-corTC_Mise en page 1 26/01/16 11:03 Page1
Sommaire/Contents
Vol. 41 (2) – 2015
Editorial – Éditorial
.............................................................................
3
and Ceanothus L. (Rhamnaceae) Suggests Different
Seed Predator Interactions
V. THOMAS PARKER
.....................................................................................
5
Developing Allometric Volume-Biomass Equations to Support Fuel
Characterization in North-Eastern Spain
B.DUGUY PEDRA, J.GODOY PUERTAS, L. FUENTES LOPEZ
............................................
15
Studying Shoot and Root Architecture and Growth of Quercus
ithaburensis subsp. macrolepis Seedlings; a Key Factor
for Successful Restoration of Mediterranean Ecosystems
...................................................
25
....................
33
RESEARCH PAPERS
Vegetation Dynamics of Coastal Dunes with Juniperus spp. in Crete,
Gavdos and Chrysi Islands (Greece)
...........
45
S. SOLIVERES, P. GARCÍA-PALACIOS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
61
P. DELIPETROU, D. GHOSN, G.KAZAKIS, P. NYKTAS, E.REMOUNDOU, I.N. VOGIATZAKIS
ecologia mediterranea – Vol. 41 (2) – 2015
1
.................................
73
R. LIBOIS, R. GHALMI, A. BRAHIMI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
85
...............................................................................
Hommage à Jacques GAMISANS
............................................................
Journal indexed in PASCAL-CBRS and Biosis
http://ecologia-mediterranea.univ-avignon.fr/
Acknowledgments – Remerciements
The editorial committee thanks the associate editors and reviewers who have
participated in this volume for their advices, corrections and opinions.
Le comité éditorial de la revue remercie les éditeurs associés et les relecteurs
qui ont participé à ce numéro pour leurs conseils, corrections et avis.
Editorial Board – Comité éditorial
© ecologia
mediterranea
Fabrication :
Transfaire,
04250 Turriers
Imprimé en Europe
Dr Pierre CHEVALDONNÉ, CNRS, Université Aix-Marseille, Marseille, France
Dr Marc CHEYLAN, EPHESS, Montpellier, France
Dr Cécile CLARET, Université Aix-Marseille, Marseille, France
Pr Thierry GAUQUELIN, Université Aix-Marseille, Marseille, France
Dr Grant WARDELL-JOHNSON, University of Western Australia, Australie
Dr Raphaël GROS, Université Aix-Marseille, Marseille, France
Dr Frédéric GUITER, Université Aix-Marseille, Marseille, France
Pr Serge KREITER, SupAgro, Montpellier, France
Dr Audrey MARCO, École nationale supérieure du paysage, Marseille, France
Pr Frédéric MÉDAIL, Université Aix-Marseille, Marseille, France
Pr François MESLÉARD, Université Avignon-Tour du Valat, France
Dr Tom PARKER, San Francisco State University, États-Unis
Dr Philippe PONEL, CNRS, Université Aix-Marseille, Marseille, France
Dr Roger PRODON, EPHE, Montpellier, France
Dr Sandra SAURA-MAS, Autonomous University of Barcelona, Espagne
Dr Isabelle SCHWOB, Université Aix-Marseille, Marseille, France
Dr Thekla K. TSITSONI, Aristotle University of Thessaloniki, Grèce
Dr Mercedes VIVAS, Universidad of Concepción, Chili
Dr Ioannis VOGIATZAKIS, Open University of Cyprus, Chypre
92
95
Editorial – Éditorial
Élise BUISSON
et Brigitte TALON
Éditrices en chef
Editors-in-Chief
This issue, the second of the year marking the 40th anniversary of ecologia
mediterranea, half consists of articles from the MEDECOS conference held in
Olmué, Chile, from Oct. 6 to 9, 2014. This is an opportunity to remind readers of
ecologia mediterranea that ISOMED, the International Society for Mediterranean
Ecology promotes research, conservation and public awareness of the biological
diversity of the world’s Mediterranean-climate regions since 1971
(http://www.incomme.org/isomed.html). The next ISOMED conference, known as
MEDECOS conference, will take place in Seville (Spain) from 31 January to
4 February 2017. The rest of the issue is devoted to research articles, which from
Crete to Algeria and Morocco, take us to forests and alfa steppes and tell us about
the dietary habits of the otter.
This 40th anniversary is also tarnished by the demise, this fall, of two great
botanists: Jacques Gamisans and Pierre Quézel, the latter being the founder in
1975 of the journal ecologia mediterranea of which he was the editor-in-chief
from 1975 to 1996. Both have greatly contributed to the knowledge on the flora
and to the development of plant ecology in the Mediterranean Basin. Both were
attached to ecologia mediterranea, in which they published numerous articles
since the 1970s. Both will greatly be missed by the scientific community.
We would like to conclude with a thought for the victims of Nov. 13 attacks in
Paris, which included nationals of many countries that contribute to ecologia
mediterranea: Algeria, Australia, Chile, Spain, USA, Italy, Morocco, Portugal,
Tunisia.
Ce nouveau numéro, le deuxième de l’année marquant le 40e anniversaire
d’ecologia mediterranea, est pour moitié constitué d’articles issus de la conférence
MEDECOS qui s’est tenue à Olmué, au Chili, du 6 au 9 octobre 2014. Ceci est
l’opportunité de rappeler aux lecteurs d’ecologia mediterranea qu’ISOMED, la
Société internationale d’écologie méditerranéenne promeut la recherche, la
conservation et la sensibilisation du public à la diversité biologique depuis 1971
(http://www.incomme.org/isomed.html). La prochaine conférence d’ISOMED, connue
sous le nom de conférence MEDECOS, portant sur des problématiques liées aux
régions au climat méditerranéen, aura lieu à Séville (Espagne) du 31 janvier au
4 février 2017. Le reste du numéro est consacré à des articles de recherche, qui, de
la Crète au Maroc en passant par l’Algérie, nous conduisent des forêts aux steppes à
Alfa et nous renseignent sur les habitudes alimentaires de la loutre.
Mais ce quarantième anniversaire de la revue s’est teinté d’une grande tristesse cet
automne avec la disparition à un mois d’intervalle de deux grands botanistes,
Jacques Gamisans et Pierre Quézel, ce dernier étant le fondateur en 1975 de la
revue ecologia mediterranea, dont il fut l’éditeur en chef de 1975 à 1996. Tous
deux ont grandement contribué à la connaissance de la flore et au développement
de l’écologie végétale du bassin méditerranéen. Tous deux étaient fidèles à ecologia
mediterranea, dans laquelle ils ont publié de nombreux articles dès les années
1970. Tous deux vont beaucoup manquer à la communauté scientifique.
Nous finirons par une pensée pour les victimes des attentats du 13 novembre à
Paris, parmi lesquelles figurent des ressortissants des pays qui contribuent à faire
vivre ecologia mediterranea : Algérie, Australie, Chili, Espagne, États-Unis, Italie,
Maroc, Portugal, Tunisie.
3
Seed Bank Divergence Between
Arctostaphylos Adans. (Ericaceae)
and Ceanothus L. (Rhamnaceae)
Suggests Different Seed Predator Interactions
La divergence de banque de semences entre
Arctostaphylos Adans. (Ericaceae) et Ceanothus L. (Rhamnaceae)
suggère différentes interactions de prédateur de semences
V. Thomas PARKER
Department of Biology, San Francisco State University,
1600 Holloway Avenue, San Francisco, CA 94132, U.S.A.
Corresponding author: [email protected]
Received: 15 January, 2015; First decision: 11 February, 2015; Revised: 5 March, 2015; Accepted: 19 April, 2015
Abstract
Two of the dominant shrub genera in California
chaparral, Arctostaphylos and Ceanothus, utilize persistent soil seed banks for post-fire recovery of their populations. Seed production and
seed rain is subject to considerable predation
by the animal community. These genera differ
significantly in their seed size, and contrary to
seed bank theory, Arctostaphylos with the larger
seeds are able to accumulate substantially larger
seed banks. Literature and field data are used
to propose a model to account for these differences in seed bank densities, specifically, that
the animal community involved in seed predation differs between the two genera.
Résumé
Deux genres d’arbustes dominants dans le
chaparral de Californie, Arctostaphylos et Ceanothus, utilisent les banques persistantes de
semences dans le sol pour le rétablissement de
leurs populations après un incendie. La production de semences et la pluie de semences
est sujette à une prédation considérable par
la communauté animale. Ces genres diffèrent
de manière significative dans la taille de leur
Keywords: scatter-hoarding, mutualism, bird
community, wildfire, mammal community.
semence. Contrairement à la théorie de banque
de semences, les Arctostaphylos, avec les
semences les plus grosses, peuvent accumuler
des banques de semences sensiblement plus
grandes. Des données provenant de la littérature et du terrain sont utilisées afin de proposer un modèle permettant d’expliquer les différences dans les densités de banque de semences,
et notamment le fait que la communauté animale impliquée dans la prédation de semences
diffère entre les deux genres.
Introduction
Arctostaphylos and Ceanothus are the only
two genera in California chaparral that have
developed both fire-stimulated persistent soil
seed banks and a large number of obligate
seeders. Research on life history evolution
and fire regimes suggests these species are
quite similar in their responses (Keeley et al.
2012). However, early studies of chaparral
Mots clés : banque de graines, caches de
nourriture, communauté de mammifères,
communauté d’oiseaux, feu, incendie, mutualisme.
5
V. Thomas Parker
seed banks indicated that Arctostaphylos and
Ceanothus have substantially different seed
bank densities (Keeley 1977; Parker & Kelly
1989). Arctostaphylos species produce relatively large seed banks ranging from around
300 to 10,000 seeds m-2 (Parker & Kelly
1989). Ceanothus species, in contrast, exhibit
much smaller seed banks, ranging from close
to no seed in some years (Quinn 1994) up to
only several hundred seeds m-2 (Keeley 1977;
Zammit & Zedler 1988; Parker & Kelly 1989;
O’Neil & Parker 2005). No hypotheses have
been put forward to explain these differences
nor their implications.
While chaparral seed banks are a key stage
for life history dynamics, they are subject to
considerable seed predation (Keeley & Hays
1976; Kelly & Parker 1990; Quinn 1994;
O’Neil & Parker 2005). Consequently chaparral seed banks can be used as a probe into
the granivore community and to determine the
post-dispersal fate of seed (Thompson 2000,
Vander Wall et al. 2005). Ceanothus greggii
seed banks, for example, show no consistent stand age-correlated pattern (Zammit &
Zedler 1988), while no change occurred in
persistent soil seed bank densities for a pair
of Arctostaphylos species over a 10 years
period (Keeley 1987), both studies implying
considerable post-dispersal predation. Overall, seed predation studies indicate variable
but significant removal of from 30% in short
term studies to nearly all fruit or seed of Arctostaphylos and Ceanothus offered (Keeley
& Hays 1976; Conard et al. 1985; Kelly &
Parker 1990; Quinn 1994; O’Neil & Parker
2005; Deveny & Fox 2006). While studies
generally lack definitive evidence, researchers
have assumed that rodents are the most important seed predators (Quinn 1994; Deveny &
Fox 2006; Huffman 2006).
With respect to most studies in shrublands,
research has tended to focus on single species or to treat the animal community as a
whole, not attempting to differentiate among
different types. However, studies distinguishing between invertebrates and vertebrates
indicate that there can be quite different
impacts (Hulme 1997) or that impacts are
additive (Zammit & Westoby 1988; O’Neil
& Parker 2005). Additional evidence suggests
that herbivores and granivores may differentially impact plant genera, with differential
herbivory on resprouting plants or post-fire
germination; for example, herbivores consume seedlings and browse young plants of
6
Ceanothus more than Adenostoma Hook. &
Arn. (Mills 1983; Quinn 1994).
The difference in seed bank densities between
Arctostaphylos and Ceanothus suggest the
potential for differential animal community
dynamics. Animals play a major role in the
dynamics of shrub-dominated communities
by impacting plant species either positively,
for example by dispersal, or negatively by
herbivory or seed consumption (Quinn 1994).
Animal communities are a well-established
component within chaparral (Quinn 1994;
Schwilk & Keeley 1998; Parker et al. 2015)
and other Mediterranean-shrub communities
(e.g., Bond et al. 1980; Bond 1984; Zammit
& Westoby 1988; Cowling et al. 1996; Hulme
1997; Forget & Vander Wall 2001; Midgley
et al. 2002; Rusch et al. 2014) and can have
differential impacts across landscapes (e.g.,
Pons & Pausas 2007a, b).
Animal foraging also may be critical to the
burial of seed for these plant community dominants. Seeds must be deep enough to escape
the killing heat pulse of a fire in the upper
portions of the soil (Odion & Davis 2000).
Rodent species found in chaparral are frequently scatter-hoarders (Vander Wall 1990;
2001). If these animals represent a large proportion of the animal community, their activity potentially could provide a mechanism
of burial. However, many birds frequently
found in chaparral are considered direct seed
consumers (e.g. California and Spotted Towhees, California Quail). Seed predators, for
example, may be more effective at reducing
the density of post-dispersal Ceanothus seeds
compared to Arctostaphylos. In field experiments, birds appear to be the dominant seed
predator of Ceanothus compared to rodents
with Arctostaphylos (Warzecha & Parker
2014).
This study examines the potential for differential impact on post-dispersal seed predation and seed banks of two dominant genera
of California chaparral, Arctostaphylos and
Ceanothus, both of which depend on seed
banks for post-fire recruitment. Ironically,
most Arctostaphylos seed structures (seeds
within nut-like endocarps) (2-12 mm diameter) are much larger than Ceanothus seed
(1.5-3 mm diameter) and Arctostaphylos
having greater seed bank density contradicts
general seed bank theory (Thompson et al.
1993; Bekker et al. 1998; but see Leishman
& Westoby 1994; Moles et al. 2000).
Seed Bank Divergence Between Arctostaphylos Adans. (Ericaceae) and Ceanothus L. (Rhamnaceae)
Figure 1 – Ceanothus and Arctostaphylos species differ in the type of fruit produced. Illustrated is A.) an example of the ballistic fruit
found in Ceanothus species (in this case C. papillosus Torr. & A. Gray) with three seed chambers, and B.) an example of the
dry drupes produced by Arctostaphylos species (in this case, A. gabilanensis V.T. Parker and M.C. Vasey).
My principal objective is to develop a hypothesis that differences in seed banks reflect differences in the animal granivore community.
Differential animal impact could be critical for
long-term plant dynamics (Moreno & Oechel
1991; Tyler 1996), and reciprocally, if animals
generally associate with different plant genera, then there is also the potential for a longterm impact on animal populations. Animal
community structure has an influence on the
assembly of plant communities through direct
and indirect interactions such as differential
herbivory, granivory, and different foraging
behavior and preferences (Davidson 1993;
Hulme 1998). How animal community composition impacts plant community composition is complex and variable yet important to
predict how plant communities will respond
to future environments (Götzenberger et al.
2012). Here I ask if the animal community
differentiates by different types of foraging
behaviors in Ceanothus versus Arctostaphylos
dominated chaparral.
Methods
Study sites: three study sites in California
were used, UC Ft. Ord Reserve near Marina in
a stabilized dune area, Monterey Co.; Bolinas
ridge of the Golden Gate National Recreation
Area, Marin Co.; and Bonny Doon Ecological
Reserve (BDER), Santa Cruz Co.
Seed sizes and seed bank density: Arctostaphylos produces drupes that range in diameter from 3-15 mm. These drupes have a dry
exocarp, dry, pulpy mesocarp (or mesocarp
absent), and a hardened endocarp divided into
multiple small nuts surrounding individual
seeds. Ceanothus produces explosive fruit; the
fruit exhibit three segments that each usually
contain a single seed (Figure 1).
To assess seed size and seed bank sizes, data
were collected from published sources. Algorithms were developed for both genera based
on measurements of sample fruit and seed, and
a constant was developed to multiply into all
species fruit diameters to estimate seed sizes.
For Arctostaphylos, constants were based on
measurements of fruit and seed for five species (A. sensitiva, A. canescens, A. viscida, A.
glandulosa, A. glauca) that provided a range
of fruit and seed size. Different constants were
developed for fruit with separate seed from
those with partially or fully fused fruit. For
Ceanothus, a similar procedure used measurements of seed from five species representing a
range of seed sizes and from both subgenera
(C. papillosus, C. thyrsiflorus, C. cuneatus,
C. jepsonii, C. prostratus). Seed sizes for both
genera were then estimated by multiplying the
constants into published fruit diameters from
the most recent treatments (Parker et al. 2012;
Wilken 2012). Seed bank sizes were taken
from available published literature sources
(Keeley 1977; Zammit & Zedler 1988; Parker
& Kelly 1989; Quinn 1994; O’Neil & Parker
2005; Warzecha & Parker 2014). Values were
tested for differences in means using a t-test
for unequal variances and unequal sample
sizes (Welch two sample t-test) (R Development Core Team 2013).
7
V. Thomas Parker
Figure 2 – S
ize differences between dispersing propagules of Ceanothus
(seed) versus Arctostaphylos (seed enclosed in nut-like endocarps).
Differences were highly significant using a t-test accounting for
differences in variance and sample size (p = 5.591E-10). Boxes are
25–75% quantile, horizontal line indicates median, open circles
indicate outliers. See methods for how data were estimated.
The animal community was estimated using
preliminary field data and literature. Fruit
from Arctostaphylos were collected 4-6 weeks
before experiments. Fruit were placed out in
petri dishes. During these experiments, live
traps were set out on one night (Gage Dayton,
University of California, Santa Cruz) to help
identify species. In the Ceanothus stand, seed
were collected using seed traps. In the study
sites, seed was offered to animals for one day
to several weeks and video surveillance used
motion-sensored Bushnell 119455C Trophy
Night Vision Trail Cameras (Bushnell Corporation, Overland Park, Kansas, USA). In
earlier experiments, seed or fruit was placed
in petri dishes in the middle of large 35 cm
diameter pans covered with a layer of fluorescent paint and animal activity was analyzed by
examining footprints. These methods are limited in their ability to reliably identify many of
the smaller rodents, but these animals generally fall into the same functional group.
The field portion was restricted to coastal
sites at relatively low elevation (< 450 m)
and restricted latitudinal distribution
(~38°N – 36°N). Chaparral, however, is a
geographically widespread community in
California and it has considerable elevation
range as well (Parker et al. 2015). Consequently, a literature survey was conducted of
common granivores or frugivores in chaparral
to generalize the results of this study, even
though they could not be assigned to dominants (sources were Lawrence 1966; Soule
et al. 1988; Fellers 1994; Price et al. 1995;
Schwilk & Keeley 1998; Laakkonen 2003)
(see Parker et al. 2015).
Results
Figure 3 – D
ensity differences between seed banks of Ceanothus versus
Arctostaphylos. Data are from published sources (Keeley 1977,
Zammit & Zedler 1988, Parker and Kelly 1989, Quinn 1994, O’Neil
& Parker 2005, Warzecha & Parker 2014). Differences in density
were highly significant using a t-test accounting for differences in
variance and sample size (p = 0.002). Boxes are 25–75% quantile,
horizontal line indicates median, open circles indicate outliers.
8
Estimated seed sizes differed significantly
between Ceanothus and Arctostaphylos (Figure 2) (t = -7.1689; df = 71.432, p = 5.591E10). The average diameter for Ceanothus
seed was 2.46 mm (+ 0.10 S.E.) and for
Arctostaphylos was 5.16 mm (+ 0.36 S.E.).
From published data, thirteen seed bank density values were found for Ceanothus and
eleven for Arctostaphylos, and for these data,
seed bank densities were significantly different between Ceanothus and Arctostaphylos
(Figure 3) (t = -4.0894, df = 10, p = 0.002).
Ceanothus seed banks were smaller (226.69 +
55.20 seeds m-2) compared to Arctostaphylos
seed banks (4382.83 + 1014.81 seeds m-2).
Chaparral contains a diversity of animals and
yet there seems to be some differentiation
between Arctostaphylos-dominated compared
to Ceanothus-dominated chaparral (Table 1,
Table 2). Both types of chaparral contain a
diverse group of seed consumers, most of
which are recognized as scatter-hoarders.
Even species thought to be principally herbivorous, for example Neotoma fuscipes Baird
(wood rat) will take fruit of Arctostaphylos or
seed of Ceanothus when available. One trail
camera caught a wood rat burying a small
cache, which was not expected. The table
includes other species common in chaparral, but in the sites investigated, chipmunks,
pocket mice, California mice, and Boyle’s
mice were the most common visitors, in addition to wood rats. These species all exhibit
scatter-hoarding to some extent. Other mammals were observed, but only occasionally,
and these are thought to be principally herbivorous (e.g., the brush rabbit).
One major difference between the stands was
the common occurrence of granivorous birds
in stands of Ceanothus, especially the California and Spotted Towhees. Quail were observed
on multiple occasions, but towhees were the
most frequent of all representing over 80% of
bird samples. During daylight, towhees were
present the greatest number of times, followed
by quail and sparrows. These types of birds
were rarely seen in stands of Arctostaphylos.
Other animals were also observed such as
the fox sparrow (Passerella iliaca Merrem),
the hermit thrush (Catharus guttatus Pallas)
and the wrentit (Chamaea fasciata Gambel).
These birds eat mixed diets but only the fox
sparrow eats seed to any extent. Other vertebrates included the coast garter snake (Thamnophis elegans terrestris Fox), the Pacific
coast rattlesnake (Crotalus oreganus Holbrook), brush rabbits (Sylvilagus bachmanii
Waterhouse), bobcats (Lynx rufus Mearns),
and evidence of coyotes (Canis latrans Say).
The brush rabbits are generally strictly herbivorous and so do not directly impact seed
banks. The snakes and bobcats are essentially
carnivores, and also would only affect seed
banks indirectly. Coyotes are not seed predators, but they do consume fruit and frequently
eat fruit of Arctostaphylos. The seed passes
through unharmed, however, and so they act
as long-distance dispersal agents, but do not
otherwise directly affect seed banks.
Table 1 – Common small mammals found in chaparral by species and common name organized by their general food habits.
Frequency refers to how widespread and common species is reported in the literature, while the last column refers to whether
evidence was found in this study for their presence in different stands.
C = Ceanothus stands; A = Arctostaphylos stands; A “?” means the determination is uncertain.
Species
Common name
Frequency
Tamias merriami J.A. Allen
T. sonomae Grinnell
T. quadrimaculatus Gray
Chipmunks
Frequent,
common at higher
elevations
Callospermophilus lateralis Say
Ground squirrel
Higher elevations
Otospermophilus beecheyi Richardson
Ground squirrel
Occasional
Chaetodipus californicus Merriam
C. fallax Merriam
Pocket mouse
Frequent to common
Dipodomys venustus Merriam
D. agilis Gambel
D. heermanni Le Conte
Kangaroo rat
Coastal Ranges
Peromycus boylii Baird
P. maniculatus Wagner
P. californicus Gambel
P. truei Schufeldt
Boyle’s mouse
Deer mouse
California mouse
True’s mouse
Common
Reithrodontomys megalotis Baird
Harvest mouse
Frequent
Microtus sp. Shrank
Vole
Frequent post-fire
Neotoma sp. Say & Ord
Woodrats
Common
Sylvilagus sp. Gray
Brush rabbit
Common
Lepus sp. L.
Jackrabbit
Common
Ceanothus or
Arctostaphylos
Principally frugivores or granivores
C/A
C
C/A
C
C/A
C/A?
Principally herbivores but also consume seed
C/A
Sources are from this study or from published data (see Methods).
9
V. Thomas Parker
Table 2 – Birds found in chaparral by species and common name organized by their general food habits. Frequency refers to how
widespread and common species is reported, the last column refers to whether evidence was found in this study for their
presence in different stands.
C = Ceanothus stands; A = Arctostaphylos stands.
Species
Common name
Frequency
Ceanothus or
Arctostaphylos
Callipepla californica Shaw
California quail
Common
C
Piplio maculates Swainson
Spotted towhee
Common
C
Melozone crissalis Vigors
California towhee
Common
C
Melozone fuscus Swainson
Canyon towhee
Frequent
Chamaea fasciata Gambel
Wrentit
Common
Toxostoma redivivum Gambel
California thrasher
Common
Aphelocoma californica Vigors
Western scrub jay
Common
Passerina amoena Say
Lazuli bunting
Occasional
Amphispiza belli Cassin
Sage sparrow
Occasional
Aimophila ruficeps Cassin
Rufous-crowned sparrow
Occasional
Spizella atrogularis Cabinis
Black-chinned sparrow
Occasional
Thryomanes bewickii Audubon
Bewick’s wren
Common
Psaltriparus minimus Townsend
Bushtit
Frequent
Polioptila californica Brewster
California gnatcatcher
Occasional
Geococcyx californianus Lesson
Road runner
Occasional
Principally granivores
Principally omnivorous
C/A
A
C
Principally carnivorous
C/A
Sources are from this study or from published data (see Methods).
Discussion
Seed bank densities of Arctostaphylos species average about 20 times that of Ceanothus species (Figure 2). Because Arctostaphylos has significantly larger seeds on average
(Figure 1), such differences require attention
because that appears to violate seed bank
theory (Thompson et al. 1993; Bekker et al.
1998). This suggests that some other process
is involved that inverts the seed bank results.
Differential dominance of principal seed predators may be the source of that process. The
major difference between animal communities in stands of Arctostaphylos and Ceanothus dominated chaparral is the frequency and
abundance of granivorous birds. Both types
of chaparral contain a variety of rodents that
are also seed consumers, but generally no
consistent difference in rodent composition
exists between the two chaparral types. The
similar mammal composition may still have
differential impacts because rodents might
directly consume the smaller seed of Ceanothus more frequently than cache them. Caches
of Ceanothus do occur, but apparent caches
are quite rare in low elevation coastal sites
10
(Parker, pers. obs.), although they do occasionally occur in higher elevation sites, either
as scatter-hoarded caches or larder-hoarded
caches (Parker, pers. obs.; Vander Wall, pers.
comm.). Scatter-hoarding is being recognized
as a common process among Mediterraneanclimate shrublands (Forget & Vander Wall
2001; Midgley & Anderson 2005).
Historically, most researchers have assumed
rodents are the principal seed predators for
both genera (e.g., Quinn 1994; Deveny &
Fox 2006; Huffman 2006). Generally, studies have not attempted definitive experiments
to distinguish between rodents and birds (but
see Warzecha and Parker 2014), and have
assumed rodents were the most important
or made no distinction among animals (e.g.,
O’Neil and Parker 2005). This distinction is
important as granivorous birds are visual seed
predators and rigorously work through the litter during and after seed dispersal. Evidence
from this study and another indicates that
they can be more effective seed predators of
Ceanothus than rodents (Warzecha and Parker
2014), and thus may be the missing dimension
explaining the differences in seed bank densities between the two genera.
Escaping seed predators for Ceanothus seed
and becoming incorporated into soil seed
banks is based potentially on the small size
allowing them to percolate into the soil,
while for Arctostaphylos, scatter-hoarding
burial would be the critical process. Because
the seeds of both genera are essentially dormant until stimulated by fire, the differential
animal composition and activity in chaparral
suggest a number of predictions. One is that
seed banks should come into a rough equilibrium based on their detectability to the granivore community, and generally, larger seeds
should be easier to detect, suggesting lower
seed bank densities for larger fruit/seed for
both genera. Because of the larger range in
fruit/seed size, if detectability is a principal
process, this further suggests that there should
be a larger range in the density of Arctostaphylos seed banks relative to the density range of
Ceanothus seed banks. Some evidence exists
to support that contention (Parker & Kelly
1989; O’Neil & Parker 2005), but further
research will be necessary to substantiate it.
Seed size may be simultaneously under selection by animals to remain undetected, yet also
selected to have enough reserves to establish
seedlings able to survive summer drought,
both within and among species. Both rodents
and birds can have selective preferences on
seed sizes and shapes in species, complicating the dimensions of these interactions (Pons
& Pausas 2007c; Muñoz et al. 2012; Rusch
et al. 2013). Finally, overall lower seed bank
densities suggest Ceanothus seedlings should
be under selection for faster growth rates or
tolerance of water stress to assure post-fire
establishment more than Arctostaphylos.
Some data may support this last suggestion as
Frazer & Davis (1988) found greater tolerance
to summer drought for Ceanothus seedlings
compared to Adenostoma seedlings. Also,
in general, seedlings emerging from smaller
seeds or suffering higher rates of first-year
mortality generally tend to have faster relative growth rates (Gilbert et al. 2006; Kitajima
& Myers 2008; Moles & Leishmann 2008)
The interactions between the plant community and the resident animal community range
from facilitory to inhibitory in terms of the
long-term success of the dominant plants. The
role of soil seed banks in these two genera is
fundamental to the dynamics and persistence
of their populations. Given that most species
in both Arctostaphylos and Ceanothus are
obligate seeders, their ability to replace themselves post-fire from seed banks is critical and
their interactions with animals subject to considerable selection. In the context of changing
climates, which in California is predicted to
be considerably warmer and drier (Hayhoe et
al. 2004; Cayan et al. 2008), how both plant
seed banks and animal community behavior
and composition respond may determine the
long-term dynamics of these communities
(Keeley et al. 2012). Already the composition
of animal communities in California is changing (Moritz et al. 2008); thus we might expect
future local dynamics could include both the
enhancement or the extirpation of obligate
seeder populations in conjunction with wildfire or extreme weather events.
Animal activity is intense within chaparral
with respect to interactions with fruit and
seed and the difference in seed bank densities may result from the shifts in dominance
of the animal communities. Seed burial processes appear to differ between these genera
due to the behavioral differences of the principal seed predators, the effective granivorous
birds versus scatter-hoarding rodents. These
compositional shifts suggest predictions about
seed size:seed bank relationships within genera and predict differences in relative growth
rates between the two genera. These results
and the review of published data support the
hypothesis that Arctostaphylos and Ceanothus dominated chaparral stands differ in the
composition and abundance of their animal
communities. However, because few stands
were investigated and the number of survey
days differed considerably among sites, these
conclusions should be considered preliminary.
Acknowledgements
Critical field help was provided by E. Herbert,
J. van den Berg, B. Warzecha, and B. Peterson. Support for the work in part came from
the U.S. National Science Foundation and the
Office of Research and Sponsored Projects at
San Francisco State University.
11
V. Thomas Parker
References
Bekker R.M., Bakker J.P., Grandin U., Kalamees R.,
Milberg P., Poschlod P., Thompson K. & Willems
J.H., 1998. Seed size, shape and vertical distribution
in the soil: indicators of seed longevity. Funct. Ecol.
12: 834-842.
Bond W.J., 1984. Fire survival of Cape Proteaceae –
influence of fire season and seed predators. Vegetatio
56: 65-74.
Bond W.J., Ferguson M. & Forsyth G., 1980. Small
mammals and habitat structure along altitudinal
gradients in the southern Cape mountains. S. Afr.
J. Zool. 15: 34-43.
Cayan D.R., Maurer E.P., Dettinger M.D., Tyree M.,
Hayhoe K., 2008. Climate change scenarios for
the California region. Climatic Change 87 (Suppl
1):S21-S42.
Conard S.G., Jaramillo A.E. & Cromack K., 1985. The
role of the genus Ceanothus in western forest ecosystems. USDA For. Serv. gen. tech. rep. PNW-182,
Oregon State University, Corvallis, 72 p.
Cowling R.W., MacDonald I.A.W. & Simmons M.T.,
1996. The Cape Peninsula, South Africa: physiographical, biological and historical background to
an extraordinary hot-spot of biodiversity. Biodivers.
Conserv. 5: 527-550.
Davidson D.W., 1993. The effects of herbivory and
granivory on terrestrial plant succession. Oikos 68:
23-35.
Deveny A.J. & Fox L.R., 2006. Indirect interactions
between browsers and seed predators affect the seed
bank dynamics of a chaparral shrub. Œcologia 150:
69-77.
Fellers G.M., 1994. Species diversity, selectivity, and
habitat associations of small mammals from coastal
California. Southwest. Nat. 39:128-136.
Forget P-M. & Vander Wall S.B., 2001. Scatter-hoarding
rodents and marsupials: convergent evolution on
diverging continents. Trends Ecol. Evol. 16: 65-67.
Frazer J., Davis S., 1988. Differential survival of chaparral seedlings during the first summer drought after
wildfire. Œcologia 76: 215-221.
Gilbert B., Wright S.J., Muller-Landau H.C., Kitajima
K. & Hernandéz A., 2006. Life-history trade-offs
in tropical trees and lianas. Ecology 87: 1281-1288.
Götzenberger L., de Bello F., Bråthen K.A., Davison J.,
Dubuis A., Guisan A., Lepš J., Lindborg R., Moora
M., Pärtel M., Pellissier L., Pottier J., Vittoz P.,
Kristjan Zobel K. & Zobel M., 2012. Ecological
assembly rules in plant communities – approaches,
patterns and prospects. Biol. Rev. 87: 111-127.
Hayhoe K., Cayan D., Field C.B., Frumhoff P.C., Maurer E.P., Miller N.L., Moser S.C., Schneider S.H.,
Cahill K.N., Cleland E.E., Dale L., Drapek R.,
Hanemann R.M., Kalkstein L.S., Lenihan J., Lunch
C.K., Neilson R.P., Sheridan S.C. & Verville J.H.,
2004. Emissions pathways, climate change, and
impacts on California. Proc. Natl. Acad. Sci. USA
101: 12422-12427.
Huffman D.W., 2006. Production, losses, and germination of Ceanothus fendleri seeds in an Arizona
ponderosa pine forest. West. N. Am. Naturalist 66:
365-373.
Hulme P.E., 1997. Post-dispersal seed predation and
the establishment of vertebrate dispersed plants in
Mediterranean scrublands. Œcologia 111: 91-98.
12
Hulme P.E., 1998. Post-dispersal seed predation:
consequences for plant demography and evolution.
Perspect. Plant Ecol. 1: 32-46.
Keeley J.E. & Hays R.L., 1976. Differential seed predation on two species of Arctostaphylos (Ericaceae).
Œcologia 24: 71-81.
Keeley J.E., 1977. Seed production, seed populations
in soil, and seedling production after fire for two
congeneric pairs of sprouting and nonsprouting chaparral shrubs. Ecology 58: 820-829.
Keeley J.E., 1987. Ten years of change in the seed banks
of the chaparral shrubs, Arctostaphylos glauca and
A. glandulosa. Am. Midl. Nat. 117: 446-448.
Keeley J.E., Bond W.J., Bradstock R.A., Pausas J.G. &
Rundel P.W., 2012. Fire in mediterranean ecosystems. Cambridge Univ. Press, Cambridge. 522 p.
Kelly V.R. & Parker V.T., 1990. Seed bank survival in
sprouting and nonsprouting Arctostaphylos species.
Am. Midl. Nat. 124: 114-123.
Kitajima L. & Myers J.A., 2008. The seedling as part
of a plant’s life history strategy. In: Leck M.A., Parker V.T. & Simpson R.L. (eds.), Seedling ecology
and evolution. Cambridge Univ. Press, Cambridge:
172-188.
Laakkonen J., 2003. Effect of arboreal activity on
species composition and abundance estimates of
rodents in a chaparral habitat in southern California
Am. Midl. Nat. 150:348-351.
Lawrence G., 1966. Ecology of vertebrate animals in
relation to chaparral fire in the Sierra Nevada foothills. Ecology 47: 278-291.
Leishman M.R. & Westoby M., 1998. Seed size and
shape are not related to persistence in soil in Australia in the same way as in Britain. Funct. Ecol.
12: 480-485.
Midgley J.J. & Anderson, B., 2005. Scatterhoarding in
Mediterranean shrublands of the SW Cape, South
Africa. In: Lambert J.E., Hulme P.E., & Vander Wall
S.B (eds.), Seed fate: Predation, dispersal, and
seedling establishment. CABI Publ., Wallingford:
197-204.
Mills J.N., 1983. Herbivory and seedling establishment
in post-ﬁre southern California chaparral. Œcologia
60: 267-270.
Mills J.N., 1986. Herbivores and early postﬁre succession in Southern California chaparral. Ecology 67:
1637-1649.
Moles A.T. & Leishman M.R., 2008. The seedling as
part of a plant’s life history strategy. In: Leck M.A.,
Parker V.T. & Simpson R.L. (eds.), Seedling ecology
and evolution. Cambridge Univ. Press, Cambridge:
217-238.
Moles A.T., Hodson D.W. & Webb C.J., 2000. Seed size
and shape and persistence in the soil in the New
Zealand flora. Oikos 89: 541-545.
Moreno J.M. & Oechel W.C., 1991. Fire intensity and
herbivory effects on postfire resprouting of Adenostoma fasciculatum in southern California chaparral.
Œcologia 85: 429-433.
Moritz C., Patton J.L., Conroy C.J., Parra J.L., White
G.C. & Beissinger S.R., 2008. Impact of a century
of climate change on small-mammal communities
in Yosemite National Park, USA. Science 322: 261264.
Muñoz A., Bonal R. & Espelta J.M. 2012. Responses of
a scatter-hoarding rodent to seed morphology: links
between seed choices and seed variability. Anim.
Behav. 84: 1435-1442.
O’Neil S.E. & Parker V.T., 2005. Factors contributing
to the seed bank size of two obligate seeding Ceanothus species in Northern California. Madroño 52:
182-190.
Odion D.C. & Davis F.W., 2000. Fire, soil heating, and
the formation of vegetation patterns in chaparral.
Ecol. Monogr. 70: 149-169.
Parker V.T. & Kelly V.R. 1989. Seed banks in California chaparral and other mediterranean-climate shrublands. In: Leck M.A., Parker V.T. & Simpson R.L.
(eds.), Ecology of soil seed banks. Academic Press,
New York: 231-255. Parker V.T., Pratt J.B. & Keeley J.E., 2015. Chaparral.
In: Mooney H.A. & Zavaleta E. (eds.), Ecosystems
of California – A source book. Univ. of California
Press, Berkeley: 479-507.
Parker V.T., Vasey M.C. & Keeley J.E., 2012. Arctostaphylos. In: Baldwin B.G., Goldman D.H., Keil D.J.,
Patterson R., Rosatti T.J. & Wilken D.H. (eds.), The
Jepson manual: Vascular plants of California (2nd
ed.). Univ. of California Press, Berkeley: 689-699.
Pons J. & Pausas J.G., 2007a. Acorn dispersal estimated
by radio-tracking. Œcologia 153: 903-911.
Pons J. & Pausas J.G., 2007b. Rodent acorn selection
in a mediterranean oak landscape. Ecol. Res. 22:
535-541.
Pons J. & Pausas J.G., 2007c. Not only size matters:
acorn selection by the European jay (Garrulus glandarius). Acta Oecol. 31: 353-360.
Price M.V., Waser N.M., Taylor K.E. & Pluff K.L., 1995.
Fire as a management tool for Stephen’s kangaroo
rat and other small mammal species. In: Keeley J.
E. & Scott T. (eds.), Brushfire in California: ecology and resource management. Intn’l Assoc. of
Wildland Fire, Fairfield: 51-61.
Quinn R.D., 1994. Animals, fire, and vertebrate herbivory in californian chaparral and other mediterranean-type ecosystems. In: Moreno J.M. & Oechel
W.C. (eds.), The role of fire in mediterranean-type
ecosystems. Springer, New York: 46-78.
R Development Core Team, 2013. R: a language and
environment for statistical computing. The R Foundation for Statistical Computing, Vienna, Austria.
ISBN 3-900051-07-0, URL http://www.R-project.
org/.
Rusch U.D., Midgley J.J. & Anderson B., 2013. Rodent
consumption and caching behaviour selects for specific seed traits. S. Afr. J. Bot. 84: 83-87.
Rusch U.D., Midgley J.J. & Anderson B., 2014. Seasonal
fluctuations in rodent seed caching and consumption
behaviour in fynbos shrublands: Implications for
fire management. S. Afr. J. Bot. 93: 217-221.
Schwilk D.W. & Keeley J.E., 1998. Rodent populations
after a large wildfire in Calfornia chaparral and coastal sage scrub. Southwest. Nat. 43: 480-483.
Schwilk D.W. & Keeley J.E., 1998. Rodent populations after a large wildfire in California chaparral
and coastal sage scrub. Southwest. Nat. 43:480-483.
Soule M.E., Bolger D.T., Alberts A.C., Wright J., Sorice
M. & Hill S., 1988. Reconstructed dynamics of
rapid extinctions of chaparral-requiring birds in
urban habitat islands. Conserv. Biol. 2: 75-92.
Thompson K., 2000. The functional ecology of soil seed
banks. In: Fenner M. (ed.), Seeds: The ecology of
regeneration in plant communities (2nd ed.). CABA
Publishing, Wallingford: 215-235.
Thompson K., Band S.R. & Hodgson J.G., 1993. Seed
size and shape predict persistence in soil. Funct.
Ecol. 7: 236-241.
Tyler C.M., 1996. Relative importance of factors contributing to postfire seedling establishment in maritime
chaparral. Ecology 77: 2182-2195.
Vander Wall S.B., 1990. Food hoarding in animals.
Univ. Chicago Press, Chicago, 445 p.
Vander Wall S.B., 2001. The evolutionary ecology of
nut dispersal. Bot. Rev. 67: 74-117.
Vander Wall S.B., Forget P-M., Lambert J.E. & Hulme
P.E., 2005. Seed fate pathways: filling the gap
between parent and offspring. In: Forget P-M.,
Lambert J.E., Hulme P.E. & Vander Wall S.B. (eds.),
Seed fate: Predation, dispersal and seedling establishment. CABI Publ., Wallingford: 1-8.
Warzecha B. & Parker V.T., 2014. Differential postdispersal seed predation drives chaparral seed bank
dynamics. Plant Ecol. 215: 1313-1322.
Wilken D.H., 2012. Ceanothus. In: Baldwin B.G., Goldman D.H., Keil D.J., Patterson R., Rosatti T.J. &
Wilken D.H. (eds.), The Jepson manual: Vascular
plants of California (2nd ed.). Univ. of California
Press, Berkeley: 1153-1162.
Zammit C.A. & Westoby M., 1988. Pre-dispersal seed
losses, and the survival of seeds and seedlings of
two serotinous Banksia shrubs in burnt and unburnt
heath. J. Ecol. 76: 200-214.
Zammit C.A. & Zedler P.H., 1988. The influence of
dominant shrubs, fire and time since fire on soil seed
banks in mixed chaparral. Vegetatio 75: 175-187.
Zammit C.A. & Zedler P.H., 1994. Organization of the
soil seed bank in mixed chaparral. Vegetatio 111:
1-16.
13
Developing Allometric Volume-Biomass
Equations to Support Fuel Characterization
Développement d’équations allométriques volume-biomasse pour aider
à la caractérisation du combustible dans le nord-est de l’Espagne
Beatriz DUGUY PEDRA*, Jesús GODOY PUERTAS, Laura FUENTES LOPEZ
Departament de Biologia Vegetal, Facultat de Biologia, Universitat de Barcelona,
Av. Diagonal, 643, Barcelona, Spain.
* Corresponding author: [email protected]
Received: 26 January, 2015; First decision: 11 March, 2015; Revised: 14 June, 2015; Accepted: 14 July, 2015
Abstract
As a consequence of the drastic land use changes
which occurred in Spain, as in other northern
Mediterranean countries, throughout the 20th
century, high amounts of fuels accumulated
in landscapes and large fires increased in the
last decades causing the degradation of ecosystems and landscapes. This trend will likely be
enhanced by climatic change, which makes necessary the design at the landscape level of new
fire management strategies. Fire models appear
as crucial tools for supporting that task.
FARSITE is a spatially-explicit fire growth and
behavior model initially validated in the USA
and successfully calibrated for Mediterranean
conditions afterwards. The goodness of its predictions is however strongly dependent on the
accuracy of the fuel-related inputs.
In that sense, the description of customized fire
behavior fuel models for Mediterranean plant
communities is essential. When describing custom fuel models a crucial parameter is fuel load
for which plant aboveground fine biomass can
be considered as a proxy. The development
of allometries is a rapid and nondestructive
method used for estimating aboveground biomass. The main objective of this study is to establish allometric equations accurately describing
Keywords: allometric relationship, fire model,
customized fuel model, Mediterranean ecosystems.
the relationship between the apparent volume
and aerial total biomass, as well as fine and
coarse biomass fractions, for two shrub species,
which are very common in dry Mediterranean
shrublands of Eastern Spain: Pistacia lentiscus
and Ulex parviflorus.
Several volume approaches (total and canopy
volume) and geometric shapes were explored
for estimating the apparent volume based on
field-collected dimensional variables.
For a given species and a given volume approach,
all shapes led to very similar volume-biomass relationships. In most cases, the apparent volume was
a very good predictive variable for the aerial biomass, with r2 values generally larger than 0.7. The
power-function model applied with the canopy
apparent volume predicted better the total and
fine biomass fractions. The smallest predictive
power was observed for allometric equations
predicting the U. parviflorus coarse aerial biomass, which is likely due to the small relative presence of this fraction in this species.
The volume-fine biomass allometric relationships appear as suitable tools for contributing
to the description of shrubland-type custom fuel
models that would result in more accurate fuel
model maps and more reliable fire modeling
predictions.
Mots clés : relation allométrique, programme de
simulation du feu, modèle de combustible adapté,
écosystèmes méditerranéens.
15
Beatriz Duguy Pedra, Jesús Godoy Puertas, Laura Fuentes Lopez
16
Résumé
Introduction
Les changements drastiques d’usage des terres
qui ont eu lieu tout au long du xxe siècle en
Espagne, comme dans d’autres pays de la Méditerranée nord, ont provoqué l’accumulation
de grandes quantités de combustible dans les
paysages et l’augmentation des grands feux de
forêt dans les dernières décennies, ce qui a causé
la dégradation des écosystèmes et des paysages.
Cette tendance sera probablement accentuée
par le changement climatique, ce qui rend
nécessaire la conception de nouvelles stratégies
de gestion du feu à l’échelle du paysage. Les
modèles d’incendies sont des outils essentiels
pour contribuer à cette tâche.
FARSITE est un programme spatialement explicite de simulation de la propagation et du comportement du feu qui a été initialement validé
aux États-Unis, puis calibré avec succès pour
les conditions méditerranéennes. Néanmoins,
l’adéquation de ses prédictions est fortement
dépendante de la qualité des entrées associées
aux combustibles.
En ce sens, la description de modèles de combustible adaptés (i.e. non standard) pour les
communautés végétales méditerranéennes est
essentielle. Pour la description de modèles de
combustible adaptés, la charge de combustible
est un paramètre essentiel et la biomasse en
éléments fins en surface peut être considérée
comme un proxy. Le développement d’allométries est une méthode rapide et non destructive
utilisée pour estimer la biomasse en surface.
L’objectif principal de ce travail est d’établir des
équations allométriques qui décrivent avec précision la relation entre le volume apparent et la
biomasse totale en surface, ainsi que celle des
fractions fines et grossières, pour deux espèces
d’arbustes qui sont très communes dans les formations arbustives méditerranéennes de l’est de
l’Espagne : Pistacia lentiscus et Ulex parviflorus.
Plusieurs approches du volume (volume total
et de canopée) et formes géométriques ont été
explorées pour estimer le volume apparent à
partir des variables dimensionnelles mesurées
sur le terrain.
Pour une espèce et un type de volume donnés, toutes les formes ont mené à des relations
volume-biomasse très similaires. Dans la plupart
des cas, le volume apparent a été une très bonne
variable prédictive de la biomasse en surface,
avec des valeurs de r2 généralement supérieures
à 0,7. Le modèle potentiel appliqué avec le
volume apparent de canopée est celui qui prédit
le mieux les biomasses totales et éléments fins.
La plus faible capacité de prédiction a été observée avec les équations allométriques de la biomasse grossière en surface d’U. parviflorus, ce
qui est probablement dû à la faible présence
relative de cette fraction dans cette espèce.
Les relations allométriques volume-biomasse
en éléments fins semblent donc des outils très
appropriés pour contribuer à la description de
modèles de combustible adaptés qui conduiraient à des cartes de modèles de combustibles
plus précises et à des prédictions plus fiables des
programmes de simulation du feu.
The profound socioeconomic changes that
occurred in Spain, as in other northern Mediterranean countries during the 20th century,
led to intense land abandonment (FernándezAles et al. 1992; Vallejo et al. 2012). The
decrease in agricultural and grazing activities combined with large-scale afforestation
policies often caused the disappearance of the
former mosaic-like landscape and the accumulation of fuel in landscapes (Lloret et al.
2002; Duguy 2003), which in turn resulted in
a drastic alteration of fire regimes (Fernandes
et al. 2014; Moreno et al. 2014). Large severe
fires increased in the 1970s becoming a major
cause of environmental degradation in most
Mediterranean ecosystems and landscapes
(Díaz-Delgado et al. 2002; Vallejo et al. 2012).
The magnitude of the problem, which has very
negative economic and ecological effects, will
likely be enhanced by climatic change (Piñol
1998; Pausas & Keeley 2009), making necessary the implementation of landscape-level
designed fire management strategies (Duguy
et al. 2013).
Given the difficulty and obvious limitations
in implementing large-scale and long-term
experiments on fuel treatments and in assessing their performance in relation to fire control, the design of effective landscape-level
fuel management policies requires the use of
theoretical and modeling-based approaches
that may allow to minimize the arbitrariness
in the planning processes while optimizing the
effectiveness of fuel treatments (Hiers et al.
2003; Finney 2004; Fernandes 2006).
Spatially-explicit fire models, such as FARSITE (Finney 1998), have shown to correctly
project fire growth and behavior of hypothetical fires through real landscapes and have
become very powerful decision-support tools
for land managers (Stratton 2006; Thompson
et al. 2011). FARSITE was developed and initially calibrated in the USA (Finney & Ryan
1995; Finney 1998) where it has been widely
applied for evaluating the effects of different
fuel treatment options on reducing fire hazard (Stephens 1998; Finney 2001; Finney
et al. 2007; Schmidt et al. 2008). It has also
been successfully parameterized for Mediterranean landscapes (Arca et al. 2007; Duguy
et al. 2007). The reliability of its predictions
strongly depends, however, on the accuracy
of fuel-related inputs (Arca et al. 2007) and
that of the fuel model map, in particular. In
Developing Allometric Volume-Biomass Equations to Support Fuel Characterization in North-Eastern Spain
that sense, the characterization of custom fuel
models developed for Mediterranean plant
communities is crucial (Arca et al. 2007; Rodríguez y Silva & Molina-Martínez 2012) and
more efforts have to be devoted to the rigorous
calibration of these models aiming to maximise their performance, i.e. the fit between
observed and predicted fire behaviour (Cruz
& Fernandes 2008; Ascoli et al. 2015).
Such calibrated custom fuel models are
expected to improve fire models predictions
and thus to promote the use of fire modelingbased approaches for assessing the effectiveness of landscape-level alternative fuel treatment strategies in relation to fire control in the
Mediterranean region.
The accurate characterization of custom fuel
models requires extended datasets of fieldcollected information about several fuel structure variables. The compilation of such datasets is often impossible due to temporal and
economical limitations. One of the essential
structural parameters required to characterize
a plant community as a fuel complex is the
aerial fuel load for which aboveground biomass, and more specifically its fine fraction
(i.e. branch diameter < 6 mm), may be often
considered as a suitable proxy. It is important, thus, to develop rapid and nondestructive methods, such as allometric relationships,
for estimating aboveground biomass and its
fine and coarse fractions in the most extended
plant communities across the Mediterranean
region.
In the framework of a broader project, in which
this study is included, we intend to develop
allometric regression equations to predictaboveground biomass fractions (total, fine and
coarse) as a function of the volume for various
shrub species, which are often among the most
abundant in dry Mediterranean shrublands of
eastern Spain. Such allometric relationships
are very scarce for Mediterranean shrubs so
far. The first stage of this work is presented in
this study and focuses on Pistacia lentiscus L.
(Anacardiaceæ) and Ulex parviflorus Pourr.
(Fabaceæ). The former species reaches the
largest average specific cover and phytovolume in the shrub layer of the studied plant
community, whereas the latter is critical in
relation to fuel modeling since it accumulates
fine dead phytomass with development stage,
which strongly contributes to the increase in
fire risk (Baeza et al. 2006).
Methods
Study site
The study site is located in the municipality of
El Perelló, Tarragona province, north-eastern
Spain (40º 54’ 19.3’’ N, 00º 40’ 53.0’’ E). The
height ranges from 230 to 245 m.a.s.l. (ICGC
2015). It has a dry Mediterranean climate,
with a mean annual temperature of 15.9 ºC
and a mean annual precipitation of 565 mm,
after data registered by the nearest weather
station, El Perelló, for the 1999-2013 period
(SMC 2015).
The soil studies carried out by Mora Betancort
(2013) in the study area showed that the dominant soils are Lithic xerorthents and Lithic
haploxerepts, after keys to soil taxonomy
(Soil Survey Staff, 2010), over limestone, and
are generally shallow (< 50 cm).
The area is dominated by Pinus halepensis
Mill. forests planted in 1970, with an understory of either macchia of Querco-Lentiscetum Br.-Bl. et al. 1935 em. A. et O. Bolòs
1950 (dominated by Quercus coccifera L.,
Chamaerops humilis L. and Pistacia lentiscus
L.) or Rosmarino-Ericion Br.-Bl. 1931 shrublands (dominated by Rosmarinus officinalis
L. and Erica multiflora L.). The herbaceous
layer is dominated by Brachypodium retusum (Pers.) Beauv. Other abundant species
are Aphyllanthes monspeliensis L. and Carex
halleriana Asso.
The digital fire perimeter maps produced
by the regional administration (ICC 2015)
showed that no fire occurred in the study site
during the 1986-2013 period. This information combined with the revision of the available digital orthophotos showed that the studied plant communities have been present for
at least the last 28 years.
Experimental design
Field sampling
Field work was carried out in February, March
and April 2013 in the shrub understory of a
Pinus halepensis stand in which Pistacia lentiscus and Ulex parviflorus were two of the
dominant species.
Twenty individuals of each of these two shrub
species were randomly selected taking into
account the whole dimensional range (focusing on the apparent volume, more specifically)
17
of each species in the studied plant community. Each individual was measured (maximum height, crown base height, minimum
and maximum crown diameters), cut (at the
basis of the stem), weighted (for obtaining the
total aboveground biomass) and reweighted in
the field after removing the fine fuel fraction
(< 6 mm diameter).
The aerial plant biomass could be thus separated into fine and coarse categories. The former is required for the characterization of fire
behavior fuel models, which do not consider
coarse live vegetation (Scott & Burgan 2005).
One U. parviflorus individual had to be discarded later due to measurement incongruities.
Estimation of the apparent volume and
search of the best regression equations
For each species, three geometric shapes
(elliptical cylinder, rectangular prism and
elliptical cone) and two volume approaches
(total and canopy volume) were used for
estimating the apparent volume (explanatory
or independent variable) of each individual,
based on the four field-collected dimensional
variables (Table 1). The selection of these
geometric shapes was based on the similarity
between them and the studied species’ morphologies as well as on previous studies (Usó
et al. 1997).
For each species and each geometric shape,
four regression functions (linear, power, exponential and logarithmic) were tested with the
two volume approaches, looking for the ‘best’
regression equation between the volume and
any of the three considered aboveground biomass fractions (total, fine and coarse). This
“best-fitting” equation is defined as the one
with the largest coefficient of determination,
or r2 value. Two additional statistics on model
performance were also computed: the mean
absolute error (MAE) and the mean absolute
Table 1 – Geometric shapes, plant apparent
volumes and regression functions used for
developing the allometric equations.
Function
Geometric
shape
Y = a + bX
elliptical
cylinder
Y = aXb
Y = c + aexp(bX)
Apparent
Volume
total volume
rectangular
prism
canopy volume
Y = aLogb(x) + c
18
elliptical cone
percentage error (MA%E) (Cruz & Fernandes
2008). We thus explored 24 allometric equations for each combination of a given species
and geometric shape (Tables 2 and 3), that is
72 equations for a given species considering
the three tested shapes. The assumptions of
normality and homocedasticity were tested
for all variables (independent and dependent) in SPSS Statistics v22.0.0. The level of
significance for all tests was set to α = 0.05.
Both assumptions could only be met through
transformation of the variables.
In the case of P. lentiscus, the volume variables (total and canopy volumes) were fourth
root transformed as well as the total and coarse
aerial biomasses. The fine aerial biomass was
ln transformed. In the case of U. parviflorus, all volume and biomass variables were
square root transformed, except the total aerial
biomass, which was fourth root transformed
when predicted with the total volume. Results
are presented using the transformed variables
meeting the assumptions (Tables 2 and 3).
Results
For a given species, one type of apparent volume (total or canopy), a given biomass fraction and one of the four regression functions,
the three tested geometric shapes always led
to very similar volume-biomass relationships and coefficients of determination. We
will therefore only present and comment the
results obtained with the elliptical cylinder
(Tables 2 and 3).
For both species, for a given dependent variable and regression function, only slight differences were observed when comparing the
r2 values obtained with each type of apparent
volume (Tables 2 and 3). For a given biomass
variable, no significant differences were found
indeed when comparing the values predicted
with each of the two tested apparent volumes
(Table 4).
In the case of Pistacia lentiscus, the coefficient
of determination was always higher than 0.8
(Table 2). The best-fitting equations obtained
for predicting the total, fine and coarse aerial
biomasses reached r2 values of 0.971, 0.88
and 0.962, respectively, and were all developed with the power function and the canopy
apparent volume (Figure 1). The lowest r2
values, although still high, were observed
among the relationships predicting the fine
aboveground biomass and, more specifically,
Table 2 – Allometric equations obtained with the elliptical cylinder for Pistacia lentiscus (n = 20).
TAV and CAV (total and canopy apparent volume, respectively) are in cm3 and 4th root transformed.
MAE: mean absolute error; MA%E: mean absolute percentage error.
Dependent variable
(and transformation applied), in g
Allometric equation
Total aerial biomass, TAB (4th root)
Fine aerial biomass, FAB (ln)
Coarse aerial biomass, CAB (4th root)
r2
MAE
(g)
MA%E
(%)
TAB = 0.1684*TAV + 1.2708
0.958
0.48
6.10
TAB = 0.1737*CAV + 1.8488
0.958
0.53
7.04
TAB = 0.361*TAV0.843
0.967
0.46
5.66
TAB = 0.522*CAV0.772
0.971
0.47
5.56
TAB = 3.0868e0.0213*TAV
0.906
0.85
10.36
TAB = 3.3329e0.0218*CAV
0.897
0.97
11.76
TAB = 6.3568*ln(TAV) – 14.562
0.927
0.78
10.62
TAB = 5.8303*ln(CAV) – 11.832
0.934
0.75
10.30
FAB = 0.0723*TAV + 4.2629
0.842
0.53
7.41
FAB = 0.0742*CAV + 4.5254
0.834
0.57
7.90
FAB = 1.667*TAV0.403
0.879
0.45
6.31
FAB = 1.990*CAV
0.880
0.47
6.55
FAB = 4.6485e0.0101*TAV
0.820
0.59
8.02
FAB = 4.8243e0.0104*CAV
0.809
0.63
8.53
FAB = 2.8195*ln(TAV) – 2.8553
0.869
0.45
6.51
FAB = 2.5803*ln(CAV) – 1.6248
0.872
0.46
6.60
CAB = 0.1642*TAV + 0.1486
0.954
0.58
10.62
CAB = 0.1697*CAV + 0.7036
0.956
0.60
11.45
CAB = 0.135*TAV1.0567
0.960
0.59
9.04
CAB = 0.215*CAV0.9664
0.962
0.59
9.22
CAB = 2.0341e0.0261*TAV
0.860
1.19
18.73
CAB = 2.2373e0.0267*CAV
0.849
1.20
19.31
CAB = 6.239*ln(TAV) – 15.432
0.934
0.72
12.27
CAB = 5.7265*ln(CAV) – 12.767
0.943
0.66
11.11
0.368
Figure 1 – B
est-fitting apparent volume-biomass regression equations for all aerial biomass fractions of Pistacia lentiscus (n = 20).
(Any transformation applied to any variable is indicated in the corresponding axis.)
for those obtained with the exponential function (0.82 and 0.809, using the total and the
canopy apparent volume, respectively). When
considering a given independent variable and
regression function, the smallest r2 was always
observed when predicting fine aerial biomass
(Table 2).
The statistic error confirms the good performance of all models (Table 2). For the three
dependent variables, and whatever the independent variable, the smallest MAE values
were generally obtained with the power function, although the logarithmic and linear functions led to very similar results, in the case of
fine and coarse aerial biomass, respectively
(Table 2).
In the case of Ulex parviflorus, r2 ranged
between 0.726 and 0.878, when considering
the total and the fine aerial biomasses, but it
was generally smaller, ranging between 0.363
and 0.779, for the coarse aerial biomass equations (Table 3). The highest r2 values for the
19
Table 3 – Allometric equations obtained with the elliptical cylinder for Ulex parviflorus (n = 19).
TAV and CAV (total and canopy apparent volume, respectively) are in cm3 and square root transformed.
MAE: mean absolute error; MA%E: mean absolute percentage error.
Dependent variable
(and transformation applied), in g
Allometric equation
Total aerial biomass, TAB
(4th root with TAV; sq. root with CAV)
Fine aerial biomass, FAB (sq. root)
Coarse aerial biomass, CAB (sq. root)
r2
MAE
(g)
MA%E
(%)
TAB = 0.0041*TAV + 2.0555
0.789
0.28
7.55
TAB = 0.0346*CAV + 2.7328
0.854
1.63
12.14
TAB = 0.2273*TAV0.4672
0.867
0.25
6.42
TAB = 0.1033*CAV0.8484
0.878
1.64
11.70
TAB = 2.2652e0.0012*TAV
0.767
0.29
8.06
TAB = 5.4205e0.0027*CAV
0.807
1.83
14.10
TAB = 1.6054*ln(TAV) – 5.804
0.853
0.23
5.81
TAB = 10.609*ln(CAV) – 46.27
0.854
1.64
10.58
FAB = 0.0239*TAV + 2.822
0.726
1.92
15.36
FAB = 0.0284*CAV + 2.9963
0.825
1.48
12.08
FAB = 0.0741*TAV0.8529
0.835
1.83
13.51
FAB = 0.1297*CAV
0.873
1.44
11.12
FAB = 4.9247e0.0021*TAV
0.739
2.15
16.68
FAB = 5.1031e0.0025*CAV
0.803
1.67
13.87
FAB = 9.2897*ln(TAV) – 42.505
0.762
1.67
12.28
FAB = 8.7465*ln(CAV) – 37.413
0.830
1.33
9.58
CAB = 0.0187*TAV – 0.8976
0.754
1.34
16.86
CAB = 0.0207*CAV – 0.1974
0.743
1.35
18.34
CAB = 3E-07*TAV2.7904
0.560
3.79
46.79
CAB = 7E-06*CAV2.3202
0.476
2.62
36.27
CAB = 0.319e0.0062*TAV
0.402
3.08
41.08
CAB = 0.4487e0.0066*CAV
0.363
2.79
36.73
CAB = 7.1924*ln(TAV) – 35.921
0.779
1.21
17.05
CAB = 6.3352*ln(CAV) – 29.452
0.742
1.27
19.00
0.7869
Figure 2 – B
est-fitting apparent volume-biomass regression equations for all aerial biomass fractions of Ulex parviflorus (n = 19).
(Any transformation applied to any variable is indicated in the corresponding axis.)
total, fine and coarse aerial biomass were
0.878, 0.873 and 0.779, respectively (Figure 2). They were obtained with the power
function and the canopy apparent volume for
the first two variables, but with the logarithmic function and the total apparent volume
in the case of the coarse aerial biomass. For a
given predictive variable and regression function, the lowest r2 value was generally found
with the equation predicting the coarse biomass fraction. For this biomass fraction, the
goodness-of-fit was particularly low with the
20
power and the exponential functions (with the
latter function, r2 was only 0.402 and 0.363
for the total and the canopy apparent volume,
respectively). In fact, considering both species, the smallest r2 value observed for a given
aerial biomass fraction was generally associated to the exponential function.
Generally, for a given species, apparent volume and regression function, the best-fitting
equation was the one established for predicting the total aerial biomass (Tables 2 and 3).
Table 4 – Results of the t-test applied for comparing the dependent variables predicted either with total
apparent volume or canopy apparent volume.
(In each case, the best-fitting volume-biomass regression was used.)
Dependent variable (transformation applied)
Levene’s test for
equality of variances
F
P
t-test for equality of means
t
df
P
P. lentiscus_total aerial biomass (4th root)
0.005
0.943
-0.007
38
0.994
P. lentiscus_fine aerial biomass (ln)
0.011
0.917
-0.002
38
0.999
P. lentiscus_coarse aerial biomass (4 root)
0.002
0.962
-0.006
38
0.995
U. parviflorus_total aerial biomass (sq. root)
0.018
0.894
-0.009
36
0.993
U. parviflorus_fine aerial biomass (sq. root)
0.047
0.830
-0.026
36
0.979
U. parviflorus_coarse aerial biomass (sq. root)
0.000
0.992
0.033
36
0.974
th
df: degrees of freedom
The differences observed in the goodness of
prediction among the various biomass fractions were often slight, though, except for U.
parviflorus and the coarse biomass fraction.
In that case, both the power and exponential
functions led to much smaller r2 values (from
0.363 to 0.56) than the ones reached by the
predictive equations of total or fine aerial biomass fractions with these same functions.
The statistic error showed, as the r2 values
did, a lower performance of the models for
U. parviflorus than P. lentiscus (Table 3). The
smallest goodness-of-fit was again observed
among the equations predicting U. parviflorus
coarse aerial biomass. The largest MAE and
MA%E values were found with the power and
the exponential functions (Table 3).
Discussion
Our results indicate that the total aerial biomass and the two explored biomass fractions
(fine, coarse) can be suitably predicted by the
apparent volume for both Pistacia lentiscus
and Ulex parviflorus. Only few studies that
focused on the development of allometries
worked so far with Mediterranean shrubs and
used the apparent volume as the explanatory
variable in their allometric models (Usó et al.
1997). Although obtaining rather poor results
for U. parviflorus (r2 value always ≤ 0.58),
Pereira et al. (1995) still considered their
volume-biomass allometric models as adequate for estimating U. parviflorus fuel loadings at the forest stand level and for mapping
understory fuels in Portuguese Pinus pinaster
stands dominated by this shrub species.
Working in shrublands under P. halepensis in
Alicante province (eastern Spain), Usó et al.
(1997) reported that the allometric regression
equations describing the relationship between
aerial biomass and apparent volume were a
sufficiently accurate method for estimating
the total aboveground biomass of Rosmarinus
officinalis and Cistus albidus, two common
Mediterranean shrubs.
Baeza et al. (2006) worked in shrublands of
U. parviflorus at different development stages
(young, mature and senescent) with no tree
layer and developed allometric equations for
predicting total aerial phytomass using the
stem diameter. They found high determination
coefficients for all stages, globally ranging
from 0.88 to 0.95. The r2 values corresponding to the young (n = 71) and mature (n = 44)
stages (0.88 and 0.94 respectively) were very
similar or slightly larger than the best r2 value
found in our study for estimating the total
aboveground biomass of this species (0.878).
Although our plant communities were different, most U. parviflorus in our study site could
be classified in the young (sometimes mature)
stage as defined by Baeza et al. (2006). The
use of stem diameter as the predictive variable for aerial biomass appears, thus, as very
suitable also. However, it should be pointed
out that the use of this variable may be more
difficult than the use of the apparent volume
in many situations, as previously remarked
by Usó et al. (1997). Its accurate measurement can be obviously quite complicated
when working with prickly species, such as
U. parviflorus, but also dealing with species
that have several stems.
We found no studies using either the apparent volume, or any dimensional variable, for
developing allometric equations predicting
21
the aerial biomass of P. lentiscus. Given the
high presence of this species in many types of
Mediterranean shrublands, which are usually
very extended, and, therefore, its high importance in relation to fire behavior studies, it
seems crucial to carry out more studies aiming to accurately estimate its aerial phytomass
through easy methods.
Our results also show that in the case of P. lentiscus the apparent volume-biomass equations
are slightly better at predicting the coarse
aerial biomass than the fine aerial biomass.
It is the contrary for U. parviflorus for which
the fine biomass is clearly better predicted.
Although more studies are required in order to
know if this result is a general rule that applies
for both species in any plant community or
not, we can still hypothesize that it might be
related to the relative abundance of these two
biomass fractions (fine, coarse) in each species. This relative abundance is clearly different between the two studied shrubs. Based
on the data measured in the field, the average
values of the ratio fine aerial biomass over
total aerial biomass were 0.52 (± 0.23) and
0.78 (± 0.1) for P. lentiscus and U. parviflorus,
respectively.
These results suggest that the aerial biomass
fraction reaching the largest proportion in a
given species would be better predicted by the
apparent volume-biomass equation approach.
The poor predictive power for the U. parviflorus coarse aerial biomass that we obtained
with several allometric equations would be,
therefore, a consequence of the rather small
relative presence of this fraction in the individuals of this species that we sampled.
Besides, the total apparent volume was a better predictive variable for this U. parviflorus
coarse aerial biomass than the canopy apparent volume. It is probably due to the fact that
a large part of the standing coarse fuels are
not considered when working with the canopy
apparent volume since the canopy of U. parviflorus is mostly made up by fine fuels.
It is noteworthy that we did not find many differences among the coefficients of determination of the various tested geometric shapes. It
is likely that we could have found more differences if we had worked with a larger number
of shapes. Usó et al. (1997) observed some differences between the circular or elliptical cylinders and the rotation paraboloid, but never
between the circular and elliptical cylinders.
In our study, we did not work with the rotation
paraboloid since it was reported that this shape
22
leads to an overestimate of the true volume for
shrubs, given that the radius is measured at the
base (Whittaker & Woodwell 1968).
Our results suggest that the three geometric
shapes that we tested were a good choice for
both P. lentiscus and U. parviflorus since they
were generally very appropriate for estimating
aerial biomass based on the apparent volume.
However, other shapes could maybe improve
the prediction of U. parviflorus coarse aerial
biomass and this aspect needs to be further
explored.
As for the type of apparent volume, the fact that
no significant differences were found between
the predictions obtained with either the total,
or the canopy apparent volume suggests that
both approaches are suitable for the two studied species. It is advisable, however, to always
explore both approaches since one or the other
can maybe lead to better results depending on
the plant species. This aspect is probably more
critical working with species for which one
type of fuel fraction (fine or coarse) is clearly
dominating in the aboveground biomass, as it
happens for U. parviflorus.
Given their good performance, the developed allometries, and more specifically the
ones predicting the fine aerial biomass, might
contribute to fuel characterization in Mediterranean plant communities dominated by the
shrub species considered in this study. Some
limitations will have to be taken into account
though, and solved on a case-by-case basis
with the suitable complementary datasets and
methodological approaches. The equations
do not apply to the community level but to
single plants. In our case study, an ongoing
vegetation sampling will result in a detailed
description of this understory shrub community through a number of composition and
structural variables estimated in 10 × 10 m
plots divided in 1-m2 quadrats. The grass and
the shrub (total and specific) covers as well
as the shrub phytovolumes (after height and
diameter measurements for all individuals of
the dominant species) are being estimated.
These data will allow, once combined with
the volume-fine biomass equations established for the dominant shrubs, to estimate
quite accurately the fine fuel load present in
the community.
The fact that the proposed allometric equations
for fine aerial biomass do not separate live and
dead fuels needs to be considered also. In the
studied plant community this aspect may be particularly limiting in the case of U. parviflorus,
which is the only species among the dominant
shrubs that accumulates large quantities of
standing fine dead phytomass. In this community, however, this accumulation is rather low,
suggesting that most U. parviflorus individuals
are young (the average height is smaller than
50 cm). Baeza et al. (2006) documented that
in young U. parviflorus shrublands the living
twigs fraction was the largest component in
the phytomass representing 91.8% of it. For
this species, however, the quotient between the
dead and live phytomass fractions shows high
variations with growth phases and a proper discrimination between live and dead fuels will be
needed when the fuel modeling work will be
implemented in later development stages or in
other communities.
Acknowledgements
We gratefully acknowledge funding from
the Spanish Ministry of Economy and Competitiveness to the ForBurn-Land project
(AGL2012-40098-C03-02). We thank Dr.
C. Vega-Garcia (University of Lleida), S.
Costafreda-Aumedes (University of Lleida),
M. Prendin Navarro, J. Garcia and A. Cardil
who collaborated in the field work. We also
sincerely acknowledge the useful comments
and suggestions of the anonymous referees
who helped improve an earlier version of the
manuscript. Dr. Duguy is a member of FORESTREAM (Research Group on Forest and
Stream Ecological Links: Watershed Management and Restoration) funded by Generalitat
de Catalunya (2014 SGR 0949).
Conclusions
The apparent volume of a shrub, which can
be easily estimated after few dimensional
variables measured in the field, appears as a
satisfactory predictor of total aboveground
biomass, but also of the fine and coarse biomass fractions, for both P. lentiscus and U.
parviflorus. The coarse aerial biomass of the
latter species seems harder to model with this
type of allometric equations, though, which
is likely due to the low relative abundance of
this biomass fraction in this species.
More studies including other shrub species
and a larger number of geometric shapes need
to be carried out in order to better understand
to which degree the relative abundances of
biomass fractions determine the goodnessof-fit of the apparent volume-biomass regressions and which combinations of geometric
shape, type of apparent volume and regression
function may be more appropriate for each
species.
In any case, our results confirm that apparent
volume-biomass allometric equations can be
quite an easy approach for accurately estimating shrub fuel loads, and fine fuel loads
in particular, and thus may be very useful
for the characterization of customized fire
behavior fuel models in Mediterranean ecosystems. Such fuel models, once calibrated,
would improve the reliability of fire spread
and behavior model projections and, therefore, help forest managers for designing more
effective landscape-level fire management
strategies under climate change.
References
Arca B., Duce P., Laconi M., Pellizzaro G., Salis M. &
Spano D., 2007. Evaluation of FARSITE simulator in Mediterranean maquis. Int. J. Wildland Fire
16(5): 563-572.
Ascoli D., Vacchiano G., Motta R. & Bovio G., 2015.
Building Rothermel fire behaviour fuel models by
genetic algorithm optimisation. Int. J. Wildland Fire
24(3): 317-328.
Baeza M.J., Raventós J., Escarré A. & Vallejo V.R.,
2006. Fire risk and vegetation structural dynamics
in Mediterranean shrubland. Plant Ecol. 187(2):
189-201.
Cruz M.G. & Fernandes P.M., 2008. Development of
fuel models for fire behaviour prediction in maritime
pine (Pinus pinaster Ait.) stands. Int. J. Wildland
Fire 17(2): 194-204.
Díaz-Delgado R., Lloret F. & Pons F.X., 2002. Satellite
evidence of decreasing resilience in Mediterranean
plant communities after recurrent wildfires. Ecology
83: 2293-2303.
Duguy B., 2003. Interacción de la historia de usos del
suelo y el fuego en condiciones mediterráneas.
Respuesta de los ecosistemas y estructura del paisaje. Tesis Doctoral. Facultad de Ciencias, Universidad de Alicante. 281 p.
Duguy B., Alloza J.A., Röder A., Vallejo R. & Pastor
F., 2007. Modelling the effects of landscape fuel
treatments on fire growth and behaviour in a Mediterranean landscape (eastern Spain). Int. J. Wildland
Fire 16(5): 619-632.
Duguy B., Paula S., Pausas J.G., Alloza J.A., Gimeno
T. & Vallejo V.R., 2013. Effects of climate extreme
events on wildfire regime and their ecological
impacts. In: Navarra A. & Tubiana L. (eds.), Regional assessment of climate change in the Mediterranean: Volume 2. Agriculture, forests and ecosystem
services and people. Springer, Dordrecht: 101-134.
Fernandes P.M., 2006. Silvicultura preventiva e gestão
de combustíveis: opções e optimização. In: Pereira
J.S., Pereira J.C., Rego F.C., Silva J.M.N. & Pereira
23
da Silva T. (eds.), Incêndios Florestais em Portugal.
ISA Press, Lisboa: 327-354.
Fernandes P.M., Loureiro C., Guiomar N., Pezzatti G.B.,
Manso F. & Lopes L., 2014. The dynamics and drivers of fuel and fire in the Portuguese public forest.
J. Env. Manag. 146: 373-382.
Fernández-Ales R., Martin A., Ortega F. & Ales E.,
1992. Recent changes in landscape structure and
function in a mediterranean region of SW Spain
(1950-1984). Landscape Ecol. 7(1): 3-18.
Finney M.A. & Ryan K.C., 1995. Use of the FARSITE
fire growth model for fire prediction in U.S. National
Parks. In: Proceedings of the International Emergency Management and Engineering Conference,
Sofia Antipolis, France, May 1995: 183-189.
Finney M.A., 1998. FARSITE: Fire Area Simulator–
model development and evaluation. Research Paper
RMRS-RP-4, USDA Forest Service, Fort Collins.
Finney M.A. 2001. Spatial strategies for landscape fuel
treatments. In: Bento J. & Botelho H. (eds.), Workshop on tools and methodologies for fire danger
mapping: proceedings of a workshop. Vila Real,
Portugal, Universidad de Traj-os-montes el AltoDouro, Departamento Forestal: 157-163.
Finney M.A., 2004. Landscape fire simulation and
fuel treatment optimization. In: Hayes J.L., Ager
A.A. & Barbour J.R. (eds.), Methods for Integrated
Modeling of Landscape Change. Interior Northwest
Landscape Analysis System. Ch. 9. Gen. Tech. Rep.
PNW-GTR-610, USDA Forest Service, Pacific
Northwest Research Station, Portland: 117-131.
Finney M.A., Seli R.C., McHugh C.W., Ager A.A.,
Bahro B. & Agee J.K., 2007. Simulation of longterm landscape-level fuel treatment effects on large
wildfires. Int. J. Wildland Fire 16: 712-727.
Hiers J.K., Laine S.C., Bachant J.J., Furman J.H.,
Greene Jr. W.W. & Compton V., 2003. Simple Spatial Modeling Tool for Prioritizing Prescribed Burning Activities at the Landscape Scale. Conserv. Biol.
17(6): 1571-1578.
ICC, 2015. Institut Cartogràfic de Catalunya – Cartographic Institute of Catalonia, maps of forest fires:
http://www.atlesnacional.cat/icc/atles-nacional/riscos-naturals/zones-cremades-per-incendi-forestal/
ICGC, 2015. Institut Cartogràfic i Geològic de Catalunya – Cartographic and Geological Institute of
Catalonia. Topographic maps (1:5.000). http://www.
icc.cat/vissir3/
Lloret F., Calvo E., Pons X. & Díaz-Delgado R., 2002.
Wildfires and landscape patterns in the Eastern Iberian Peninsula. Landscape Ecol. 17: 745-759.
Mora Betancort A., 2013. Efecto de una quema prescrita de intensidad moderada sobre el suelo bajo
matorral mediterráneo. Tesis de Master. Universitat
de Lleida. 49 p.
Moreno M.V., Conedera M., Chuvieco E. & Pezzatti
G.B., 2014. Fire regime changes and major driving
24
forces in Spain from 1968 to 2010. Environ. Sci.
Policy 37: 11-22.
Pausas J.G. & Keeley J.E., 2009. A burning story: the
role of fire in the history of life. BioScience 59(7):
593-601.
Pereira J.M.C., Sequeira N.M.S. & Carreiras J.M.B.,
1995. Structural Properties and Dimensional Relations of Some Mediterranean Shrub Fuels. Int. J.
Wildland Fire 5(1): 35-42.
Piñol J., Terradas J. & Lloret F., 1998. Climatic warming, wildfire hazard and wildfire occurrence in
coastal eastern Spain. Climatic Change 38: 345-357.
Rodríguez y Silva F. & Molina-Martínez J.R., 2012.
Modeling Mediterranean forest fuels by integrating field data and mapping tools. Eur. J. Forest Res.
131: 571-582.
Schmidt D.A., Taylor A.H. & Skinner C.N., 2008. The
influence of fuels treatment and landscape arrangement on simulated fire behavior, Southern Cascade
range, California. For. Ecol. Manag. 255(8–9):
3170-3184.
Scott J.H. & Burgan R.E., 2005. Standard fire behavior fuel models: a comprehensive set for use with
Rothermels surface fire spread model. Gen. Tech.
Rep. RMRS-GTR-153, USDA Forest Service,
Rocky Mountain Research Station, Ogden: 80 p.
SMC, 2015. Servei Meteorològic de Catalunya - Catalan Weather Service. El Perelló station (1999-2013):
http://www.meteo.cat/wpweb/climatologia/anuarisde-dades-meteorologiques/
Soil Survey Staff, 2010. Keys to Soil Taxonomy. 11th edition. USDA, NRCS, Washington DC: 338 p.
Stephens S.L., 1998. Evaluation of the effects of silvicultural and fuels treatments on potential fire
behaviour in Sierra Nevada mixed-conifer forests.
For. Ecol. Manag. 105(1-3): 21-35.
Stratton R.D., 2006. Guidance on spatial wildland fire
analysis: models, tools, and techniques. Gen. Tech.
Rep. RMRS-GTR-183. USDA Forest Service,
Rocky Mountain Research Station, Fort Collins:
15 p.
Thompson M.P., Calkin D.E., Finney M.A., Ager A.A.
& Gilbertson-Day J.W., 2011. Integrated nationalscale assessment of wildfire risk to human and ecological values. Stoch. Env. Res. Risk A. 25: 761-780.
Usó J.L., Mateu J., Karjalainen T. & Salvador P., 1997.
Allometric regression equations to determine aerial
biomasses of Mediterranean shrubs. Plant Ecol.
132(1): 59-69.
Vallejo V.R., Allen E.B., Aronson J., Pausas J.G., Cortina J. & Gutiérrez J.R., 2012. Restoration of Mediterranean-type woodlands and shrublands. In: Van
Andel J. & Aronson J., (eds.), Restoration Ecology:
The New Frontier. 2nd edition. Wiley-Blackwell,
Oxford, UK: 130-144.
Whittaker R.H. & Woodwell G.M., 1968. Dimension
and production relations of trees and shrubs in the
Brookhaven Forest, New York. J. Ecol. 56(1): 1-25.
Studying Shoot and Root Architecture
and Growth of Quercus ithaburensis subsp.
macrolepis Seedlings;
Key Factors for Successful Restoration
Étude de l’architecture des parties aériennes et racinaires
et de la croissance des semis de Quercus ithaburensis subsp.
macrolepis ; facteurs clés pour une restauration réussie
des écosystèmes méditerranéens
Thekla TSITSONI*, Marianthi TSAKALDIMI, Maria GOUSIOPOULOU
Laboratory of Silviculture, Department of Forestry & Natural Environment,
School of Agriculture, Forestry and Natural Environment Aristotle University of Thessaloniki
P.O. BOX 262, University Campus, 54124 Thessaloniki, Greece
*Corresponding author: [email protected]
Received: 19 January, 2015; First decision: 28 March, 2015; Revised: 2 June, 2015;
Second decision: 20 July, 2015; Revised: 28 August, 2015; Accepted: 30 September, 2015
Abstract
Quercus ithaburensis subsp. macrolepis is a
species of high ecological importance. It forms
some of the few deciduous oak forests in the
eastern Mediterranean zone, is well adapted to
Mediterranean conditions and warm dry periods, develops a deep root system, regenerates
easily after fire and is used in many restoration
projects. However, information on the nursery
production method of Quercus macrolepis seedlings and the growth and architectural development of these produced seedlings is limited. In
order to select the best cultivation method to
produce high quality seedlings, the objectives
of this study were to a) investigate shoot and
root architecture of seedlings of Quercus ithaburensis subsp. macrolepis and b) compare the
growth rates of the seedlings which were grown
as bareroot (i.e. propagated in raised beds) vs.
containerized in the nursery. The results of
the study showed that both aboveground and
Keywords: root growth rate, Valonia oak, first
order laterals, container, bareroot.
belowground morphological characteristics and
growth rates of container Quercus ithaburensis subsp. macrolepis seedlings outperformed
those of bareroot ones. Shoot height, root
collar diameter, shoot and root growth rates,
number of first order lateral roots and the tap
root of the container seedlings, showed significantly higher values than those of the bareroot
seedlings. The high number of first order roots
and the long central root of containerized seedlings would facilitate seedlings’ uptake of soil
water and nutrients and confirm that they have
a better chance of survival in degraded areas.
However, the results obtained are not enough
to discourage the production of Quercus macrolepis bareroot seedlings. The influence of root
pruning, seedbed density and fertilization on
the bareroot seedlings quality, deserve future
investigation.
Mots clés : taux de croissance des racines,
chêne Valonia, racines latérales de premier ordre,
contenant, pot, semis en plein sol, semis en
parterre ou plate-bandes.
25
Thekla Tsitsoni, Marianthi Tsakaldimi, Maria Gousiopoulou
Résumé
Quercus ithaburensis, sous-espèce macrolepis est une espèce d’importance écologique
majeure. Elle est à la base des quelques forêts
de chênes à feuilles caduques de l’est du bassin
méditerranéen ; elle est bien adaptée aux conditions chaudes et sèches de la Méditerranée,
développe un système de racines profondes, se
régénère facilement après un incendie et est utilisée dans certains projets de restauration écologique. Pourtant, peu d’informations existent sur
ses besoins écologiques, et encore moins sur tout
ce qui concerne sa propagation et méthodes
de production en pépinières. Afin de produire
des semis de très bonne qualité, pouvant être
utilisés pour des projets de reforestation ou
de restauration réussis, les objectifs de cette
étude étaient de a) enquêter sur l’architecture
des parties aériennes et racinaires de Quercus
ithaburensis sous-espèce macrolepis et b) comparer les vitesses de croissance des semis ayant
crû plein sol vs. en pot durant l’étape de pépinière. Les résultats de l’étude ont montré que
les caractéristiques morphologiques aériennes
et racinaires ainsi que le taux de croissance des
semis ayant grandi en pot ont surpassé de loin
les performances de ceux de plein sol.
Plus spécifiquement, les conclusions de cette
recherche sont les suivantes :
–– la hauteur des parties aériennes, le diamètre
des collets ainsi que le taux de croissance des
parties aériennes des semis en pot ont montré
des valeurs supérieures à celles des semis en
plein sol, à toutes les dates de prélèvement
des données ;
–– le taux de croissance des parties aériennes
(Shoot Growth Rate – SRG) de tous les semis
étudiés a diminué tout au long de la période
d’étude, mais celui des semis en pot est resté
supérieur à celui des semis en plein sol, à
toutes les dates de prélèvement ;
–– les semis en plein sol ont montré un taux supérieur de croissance des racines (Root Growth
Rate – RGR) au début, mais par la suite le RGR
des semis en pot a dépassé ce premier ;
–– le grand nombre de racines de premier ordre
et la longue racine centrale des semis en pot
assurent ses apports en eau et nutriments
provenant du sol et confirment qu’ils ont des
meilleures chances de survie dans des zones
dégradées ;
–– afin d’éviter que des racines ne s’enroulent
pendant la production des semis dans des
pots cylindriques, l’utilisation de pots plus
profonds est conseillée ;
–– cependant, plus de recherches sur l’influence
1) des dommages faits aux racines lors de
la transplantation, 2) de la densité du lit de
semences et 3) des apports de nutriments sur
les semis en plein sol sont nécessaires.
Introduction
Q. ithaburensis subsp. macrolepis grows
mainly in the central and eastern part of
Mediterranean basin and particularly in SE
Italy, S. Albania, Greece, Turkey, Israel, Palestine, Jordan, Syria and Lebanon (Figure 1).
Large forests with old Q. ithaburensis subsp.
macrolepis trees existed during the past in the
Mediterranean zone of Greece, in Peloponnesus, Attica, and the Aegean islands (Grispos 1973). During the last decades, activities
such as conversion of forests to agricultural
land, illegal lumbering and overgrazing have
confined Q. ithaburensis subsp. macrolepis
to small-forested patches or to isolated
Figure 1 – Geographical distribution of Quercus ithaburensis subsp. macrolepis.
26
Studying Shoot and Root Architecture and Growth of Quercus ithaburensis subsp. macrolepis Seedlings;
Key Factors for Successful Restoration of Mediterranean Ecosystems
individuals in the interior of forested islets
in lowland and semi-mountainous agricultural fields (Pantera et al. 2008; Pantera &
Papanastasis 2012). Compared to the past,
the species currently has an ecological rather
than economic importance; it forms some
of the few deciduous oak forests in the eastern Mediterranean zone (eu-mediteranean,
Quercetalia ilicis zone) and certain of these
forests are included in the “Natura 2000”
network. It is well adapted to Mediterranean
conditions and warm dry periods, develops
a deep root system, and regenerates easily
after fire (Pantera & Papanastasis 2012). In
the past few years, there has been a growing interest for the species to be included in
reforestation as well as in restoration projects
(Zaady and Perevolotsky 1995; Tsakaldimi
et al. 2000; Tsitsoni et al. 2011). However,
information on the nursery production methods of Quercus macrolepis seedlings and the
growth and architectural development of these
produced seedlings is limited. Actually, no
relevant research on seedling production and
quality of the subsp. macrolepis seedlings was
conducted. A detailed knowledge of how the
production methods affects seedlings quality
of the species could contribute to better seedlings’ outplanting performance.
The two important factors that limit the establishment and growth of woody seedlings in
Mediterranean environments, particularly in
abandoned cropland and in deforested areas,
are excessive irradiation and reduced water
availability (Rey Benayas & Camacho-Cruz
2004; Valdecantos et al. 2006). Moreover,
predation is also a limiting factor in case of
direct oak seeding in the field, while herbivory
has been showed to strongly impact Quercus
saplings in Mediterranean forests. Most seedling mortality occurs in the first dry season of
their life cycle and it has been attributed, in
addition to other factors, to poor stock quality
(Gazal et al. 2004). In Mediterranean Greece,
the harsh site conditions in relation to the fact
that degraded Mediterranean areas occupy
extensive areas in Greece (Ganatsas et al.
2012) increase the need for the production of
high planting stock quality, able to cope with
the environmental conditions and ensure a
successful reforestation or restoration.
The poor survival of oak plantings has been
linked to factors, such as slow growth, rapid
growth of competing vegetation, poor planting
methods, and poor seedling quality (McGee
& Loftis 1986; Pope 1993; Tsakaldimi et al.
2005; Pantera & Papanastasis 2012).
Oak plantings may be established by direct
seeding in the field, but the success of such
operations has been inconsistent for many
species. For that reason, the vast majority of
oak plantings are established using seedlings
grown in nurseries (Allen et al. 2001; Jacobs
2003; González-Rodríguez et al. 2011).
The establishment of transplanted seedlings of
woody species, especially oaks, in degraded
Mediterranean environments may be aided by
nursery treatments that promote the development of a deep and well-structured root system
(Green et al. 2005; Tsakaldimi et al. 2009).
Several authors argue that root morphology
and growth may provide an effective indicator of seedling performance (Davis & Jacobs
2005; Tsakaldimi et al. 2005) despite the fact
that these measurements are destructive, laborious and time consuming. The elongation of
the taproot and new root growth during the
first season after field planting is extremely
important for seedling survival under the
Mediterranean climatic conditions. Moreover,
root regeneration is of critical importance to
the establishment of planted seedlings. New
root growth enables the seedling to establish
a functional connection with the soil and
thereby overcome the water stress imposed
by transplanting (Burdett 1990; Krasowski
2003; Grossnickle 2005).
Among the commonly assessed root system attributes of hardwood seedlings are the
number of primary first order lateral roots
(FOLRs) and root system fibrosity (Davis &
Jacobs 2005). These parameters are broadly
indicative of the structural framework (i.e.
mainly involved in support and transport functions) and the fine root component (i.e. mainly
involved in water and mineral nutrient uptake)
of a seedling root system, respectively. Several
authors argue that the FOLRs could be one of
the best indicators of seedlings competitive
capacity in the field (Schultz & Thompson
1997; Teclaw & Isebrands 1993; Kormanik et
al. 1994). A large number of FOLRs is linked
to rapid early establishment, improved growth
rates and survival of oak seedlings (Ruehle &
Kormanik 1986; Schultz & Thompson 1997).
The two main types of nursery operations for
oak seedling production are bare-root and
containerized. Although many technological advances have been made over the last
twenty years in production of both bareroot
and containerized planting stock to improve
planting stock quality, oak containerized
seedlings dominate the forest nurseries in
27
Greece, possibly due to shorter production
cycles, stock uniformity and better field performance on harsh planting sites (Tsakaldimi
et al. 2005). However, few studies have compared shoot and root growth and architecture
of oak seedlings produced by these two nursery treatments. Wilson et al. (2007) reported
that container Quercus rubra seedlings had a
larger number of FOLRs and greater relative
growth rate (RGR) than bareroot seedlings.
The objectives of this study therefore were a)
to compare shoot and root architecture and
growth rates in time of Quercus ithaburensis subsp. macrolepis seedlings which were
grown as bareroot or containerized in the nursery phase, b) to ascertain through which production technique we have better seedlings’
quality. This knowledge can contribute to a
better understanding of what is the indicated
species production technique in order to produce high quality seedlings capable of ensuring a successful reforestation or restoration.
Materials and Methods
Study Site
Acorns of Quercus ithaburensis subsp. macrolepis were collected in January 2012 from
the Forest Botanic Garden (6.2 hectares) of
Aristotle University of Thessaloniki, Northern Greece (Latitude 40.566266, Longitude
22.969896, 32.3 m.a.s.l).
According to data from the meteorological
station of Aristotle University of Thessaloniki, the climate of the region is Mediterranean
with dry hot summers and mild winters. The
average annual air temperature is 15.8 °C
with minimum average monthly temperature 5.9 °C (January) and maximum 25.9 °C
(July). The average annual precipitation is
449.3 mm and relative humidity is 66.7%
(Samara & Tsitsoni 2014)
Experimental design
Acorns were transported to the laboratory for
measurements and experimental trials. Prior
to experiments, acorns were visually checked
and any acorns that appeared abnormal or
obviously defective were discarded (Ganatsas & Tsakaldimi 2013). Then acorns were
immersed in water for 24 hours to eliminate
dead acorns.
28
120 viable acorns were linearly sown in seedbeds (sowing distance 15 × 15 cm) and as many
were sown in plastic bags (25 cm depth and
1.5 L volume) filled with peat-perlite (1:1) for
the production of bareroot and container seedlings respectively. Soil bulk density in seedbeds is high enough (1.58 g/cm3 at 0-20 cm
depth) and it increases with the increase of
soil depth (1.98 g/cm3 at 20-40 cm). Soil pH
is neutral (pH 6.77-6.88) and its organic matter ranges from 2.82 (0-20 cm) to 1.74 (2040 cm).
All acorns were sown in mid March 2012 in
an open nursery and were irrigated with an
overhead irrigation system. No shading was
used and no nutrients were added.
Three and a half months after sowing, five
seedlings per treatment were randomly
extracted for destructive sampling and the
same sample size was extracted every fifteen
days for a period of two months (five samples
of five seedlings each per treatment, totally).
After each destructive sampling, samples were
transferred to the laboratory for the following measurements: length of the central root,
number of first order lateral roots (FOLRs),
length of the longest first order lateral, root
collar diameter (RCD), stem height, number
of leaves.
For root measurements, each sampled root
system was separated from the soil. More
specifically, before the excavation of bareroot
seedlings, sufficient water was poured around
the seedlings to loose and wet the soil. In the
laboratory, roots from container and bareroot
seedlings were repeatedly submerged in water.
A sieve was used to collect any root fragments
detached from the root system (Tsakaldimi et
al. 2013). Height and length measurements
and root collar diameters were taken using a
ruler and a vernier calliper, respectively. The
length of the central root (taproot) and the
length of the longest first order lateral were
measured on a millimeter paper for more precise measurements. Root growth rate (RGR)
estimation was based on the growth of the
longest first order lateral root.
Statistical analysis
Distribution was tested for normality by
Kolmogorov-Smirnov criterion. Because the
homogeneity of variances was not assumed
the non parametric Mann-Whitney test was
used, comparing two groups (bareroot vs. container seedlings) in each analysis. Variables
Table 1 – Above and below ground growth characteristics of containerized vs. bareroot seedlings of Quercus ithaburensis subsp.
macrolepis in different sampling dates (sowing date mid-March).
For each sampling date, the means followed by asterisk (*) are significantly different (P < 0.01).
Given values are mean (std error of the mean).
Sampling
dates
Seedlings
type
Shoot
height
(cm)
Leaf
number
Root collar
diameter
(mm)
Length of the
longest FOLR
(cm)
FOLRs
number
Tap root
length
(cm)
Bareroot
19.90
(1.68)
19.40
(2.65)
4.60
(0.24)
34.02
(3.18)
7.40
(1.63)
20.02
(5.65)
Container
22.40
(2.11)
19.60
(2.42)
4.80
(0.48)
55.74
(5.96)
7.20
(1.57)
23.82
(5.32)
Bareroot
21.12
(1.53)
19.80
(2.51)
5.04
(0.42)
46.22*
(2.82)
10.00
(0.77)
21.16
(3.96)
Container
23.78
(0.59)
21.40
(3.52)
5.80
(0.20)
63.94*
(1.94)
12.00
(0.83)
24.38
(3.35)
Bareroot
21.50
(1.50)
20.60
(2.50)
5.62
(0.29)
47.46*
(2.84)
16.40
(2.73)
22.44
(2.46)
Container
24.46
(0.99)
22.20
(3.81)
6.36
(0.49)
66.58*
(2.62)
18.00
(2.48)
26.74
(3.36)
Bareroot
21.87
(1.48)
21.40
(2.50)
5.86
(0.13)
50.16*
(2.54)
18.40*
(1.39)
23.36*
(0.50)
Container
24.98
(1.18)
22.60
(3.73)
6.94
(0.68)
79.50*
(6.01)
36.40*
(5.23)
35.16*
(7.97)
20/06/2012
05/07/2012
20/07/2012
04/08/2012
analyzed were: shoot height, shoot growth
rate, leaf number, root collar diameter, length
of the longest FOLR, number of FOLRs, taproot length and root growth rate. Statistical
analysis was performed with SPSS (SPSS
Inc., version 21).
Results and Discussion
During five month growth in the nursery, the
containerized seedlings, raised in 1.5 L container, exhibited significantly greater growth at
all sampling dates than the bareroot seedlings.
At the final extraction date, the five-month old
containerized seedlings were 3.11 cm taller
and 1.08 mm greater in root collar diameter
(Table 1). The aerial development of container
seedlings was probably favoured by both the
large volume and the favourable properties
of the artificial growing medium peat:perlite
1:1 (a very porous, well-drained, with high
exchange capacity and friable medium) compared to those of the nursery field soil. Williams and Strupe (2002) found that bareroot
water oak (Quercus nigra) and willow oak
(Quercus phellos) seedlings grew taller and
had larger RCD than the container seedlings,
but the latter were raised in plastic cone containers of very limited volume (vol. 164 cm3).
However, much more important was the statistically significant increase in root parameters
of the container seedlings; after five months
they presented 29.34 cm longer longest
FOLR, 18 more FOLRs and 11.8 cm longer
central root than bareroot ones (Table 1).
Similar to our results, Wilson et al. (2007)
found that container seedlings of red oak (Q.
rubra L.) had higher number of FOLRs and
were significantly more fibrous than bareroot
seedlings throughout the first growing season.
They also reported that container seedlings
acclimated faster in the field, creating a better
relationship between soil and main root than
bareroot seedlings due to the high difference
in the length of the tap root.
The rapid growth of the root system improves
the uptake and transport of water which in
turn favour the survival and growth of plants
(Tsakaldimi et al. 2005). Under dry conditions, trees with large root system can survive
better than seedlings with small root system
(Hobbs 1984). Johnson (1979) noted that
an increased number of first order laterals
(FOLRs) in container seedlings compared to
bareroots contributed to the increase in total
length of roots ten weeks after outplanting.
This facilitated water absorption when environmental conditions were suitable for the
growth and minimized water stress. Many
researchers have noted that a large number
of FOLRs and a fibrous root system with
large root surface area are important features
of high-quality seedling planting (Ruehle &
29
Figure 2 – S
hoot growth rate of bareroot (♦) and container seedlings (○). In
each sampling date, statistical significant differences were detected
between the bareroot and container seedlings (P < 0.05). The error
bars are the std error of the mean.
Figure 3 – L ongest FOLR growth rate of bareroot (♦) and container seedlings
(○). In each sampling date, statistical significant differences were
detected between the bareroot and container seedlings (P < 0.05).
The error bars are the std error of the mean.
Kormanik 1986; Thompson & Schultz 1995;
Tsakaldimi et al. 2005).
Concerning the length of the central root in
container seedlings, after four months growth,
it was found longer (26.74 cm) than the container depth because roots were twisting at the
bottom of the container. Generally the roots of
oaks tend to exhibit twist root around the interior or the base of the containers when these
are cylindrical (Landis et al. 1990). Seedlings
of most oak species preferentially invest into
producing roots, which initially have a taproot or carrot-like configuration. Such rapid
growth means that seedlings’ root systems
can quickly exceed the depth of the container
30
and become pot-bound. In general, for better
quality of oak seedlings narrow, deep bottom
opened containers are recommended. The
root deformities such as the twisted roots can
persist for decades after planting and create
problems of weakness, poor growth and lack
of stability years later (McCreary & Canellas
2005). Similarly, Wilson et al. (2007), despite
the good performance of red oak container
seedlings, observed a great number of twisted
roots and they suggested the use larger pots
than those commonly used.
The shoot growth rate (SGR) of all studied
seedlings decreased throughout the study
period but was significantly higher in container seedlings than that of bareroot, in
all sampling dates (Figure 2). The first 3.5
months after sowing, the bareroot seedlings
showed significantly higher root growth
rate (RGR) than the containerized seedlings
(0.81 cm/day and 0.55 cm/day respectively),
however, then the RGR was significantly
increased in container-seedlings (P < 0.05).
After 4.5 months, the RGR of container
seedlings was found 0.86 cm/day while the
RGR of bareroots remained close to 0.2 cm/
day (Figure 3). The RGR change of bareroots
could be affected by the seedlings’ density in
seedbeds. Possibly the competition for space
and moisture in seedbeds was increased
between bareroot seedlings by that time and
had a negative influence in seedlings RGR.
On the contrary, such competition was absent
in container seedlings where seedlings grew
in an autonomous growth space. Similarly,
container fir seedlings presented higher RGR
than bareroot ones (Scagel et al. 1993). Also,
a successful seedling establishment is dependent on the capacity of seedlings to initiate new
roots quickly (Grossnickle 2005). Seedlings
with higher RGR are expected to survive and
perform better, than those with lower RGR
because they are more prone to damage under
adverse site conditions.
The results obtained showed that container
seedlings were enhanced by the favourable
growth conditions; their better performance
could be attributed to the moist and friable
artificial medium of sufficient volume that
they were grown in relation to the nursery
field soil. The artificial medium peat:perlite
(1:1) has better structure and aeration than the
nursery field soil so it was easier for the roots
of container seedlings to absorb the necessary
nutrients and water for their growth. Pantera
et al. (2010) found that acorn germination
of Quercus ithaburensis subsp. macrolepis
is not affected by soil compaction but under
high soil compaction the seedlings’ growth
was negatively affected. Recent studies show
that field performance of bareroot seedlings
can be improved by increasing the amount of
organic matter in beds or decreasing nursery
bed seedling densities to produce larger seedlings. Salifu et al. (2009) reported that nutrient loading of oak seedlings in the nursery
has the potential to increase seedling performance especially on harsh or degraded sites.
Moreover, the undercutting of bareroots during the growing season can cause an increased
production of new roots, especially FOLRs
(Jacobs 2003, Dey et al. 2004). However, it
still remains unknown whether the undercutting of taprooted species may favour or not
their outplanting survival and growth in dry
Mediterranean soils.
From all the above, it seems that both aboveground and belowground morphological
characteristics and growth rates of container
Quercus ithaburensis subsp. macrolepis
seedlings outperformed those of bareroot
ones. However, it is unclear if the superior
growth performance of container seedlings
will persist beyond the establishment phase.
Andersen et al. (1989) found that container
seedlings had higher survival than bareroot
red oak seedlings during the first four years
after planting in the field. However, seven
years after planting, survival did not greatly
differ between the two types of seedlings.
Further work is needed to investigate both the
influence of pot design in root twisting of container seedlings and the influence of root pruning, seedbed density and nutrients in bareroot
seedling growth. Enhanced seedling quality
could help young trees of Q. ithaburensis
subsp. macrolepis, establish themselves more
rapidly, thereby giving them a better chance
for successful growth in the field.
References
Allen J.A., Keeland B.D., Stanturf J.A., Clewell A.F. &
Kennedy H.E., 2001. A guide to bottomland hardwood restoration. Gen. Tech. Rep. SRS-40, USDA
Forest Service, pp.132.
Andersen C.P., Bussler B.H., Chaney W.R., Pope P.E.
& Byrnes W.R., 1989. Concurrent establishment
of ground cover and hardwood trees on reclaimed
mined land and unmined reference sites. For. Ecol.
Manage. 28: 81-99.
Burdett A.N., 1990. Physiological processes in plantation establishment and the development of
specifications for forest planting stock. Can. J. For.
Res. 20: 415-427.
Davis A.S. & Jacobs D.F., 2005. Quantifying root system quality of nursery seedlings and relationship
to outplanting performance. New For. 30: 295-311.
Dey D.C., Lovelace W., Kabrick J.M. & GoldIn M.A.,
2004. In: Michler, C.H.; Pijut, P.M.; Van Sambeek,
J.W.; Coggeshall, M.V.; Seifert, J.; Woeste, K.;
Overton, R.; Ponder, F., Jr., (eds.), Proceedings of
the 6th Walnut Council research symposium: Black
walnut in a new century. Gen. Tech. Rep. NC-243,
USDA, Forest Service, North Central Research Station, St. Paul: 188 p.
Ganatsas P., Tsitsoni T., Tsakaldimi M. & Zagas T.,
2012. Reforestation of degraded Kermes oak shrublands with planted pines: effects on vegetation
cover, species diversity and community structure.
New For. 43: 1-11.
Ganatsas P. & Tsakaldimi M., 2013. A comparative
study of desiccation responses of seeds of three
drought-resistant Mediterranean oaks. For. Ecol.
Manage. 305: 189-194.
Gazal R.M. & Kubiske M.E., 2004. Influence of initial root characteristics on phsiological responses
of cherrybark oak and Shumard oak seedlings to
field drought conditions. For. Ecol. Manage. 189:
295-305.
González-Rodríguez V., Navarro-Cerrillo R.M. & Villar R., 2011. Artificial regeneration with Quercus
ilex L. and Quercus suber L. by direct seeding and
planting in southern Spain. Annals of Forest Science
68(3): 637-646.
Green J.J., Baddeley J.A., Cortina J. & Watson C.A.,
2005. Root development in the Mediterranean shrub
Pistacia lentiscus as affected by nursery treatments,
J. Arid Environ. 61: 1-12.
Grispos P., 1973. Greek forest history. Publication 25 of
the Forest Applications Service, Athens, Ministry of
Nation. Econ., Agricultural Sector, Forest Division,
385 p. (In Greek).
Grossnickle S.C., 2005. The importance of root growth
in overcoming planting stress. New For. 30: 273294.
Hobbs S.D., 1984. The influence of species and stocktype selection on stand establishment: An ecophysiological perspective. In: Duryea M.L. & W. Brown
W. (eds.), Seedling physiology and rReforestation
success. NijhofB/Dr. W. Junk Pub. Boston: 179-224.
Jacobs D.F., 2003. Nursery production of hardwood seedlings. FNR 212, Purdue University, Cooperative
extension Service, West Lafayette, USA, 8 p.
Johnson P.S., 1979. Growth potential and field performance of planted oaks. In: Holt H.A. & Fisher B.C.
(eds), Proceedings: Regenerating oaks in upland
hardwood forests. The 1979 John S. Wright Forestry
Conference. Purdue University, Indianna: 113-119.
Kormanik P.P., Sung S.S. & Kormanik T.L., 1994.
Toward a single nursery protocol for oak seedlings.
In: Proceedings of 22nd southern forest tree improvement conference, Atlanta, GA 14-17 Jun 1993:
89-98.
Krasowski M.J., 2003. Root system modifications by
nursery culture reflect on post-planting growth and
development of coniferous seedlings. For. Chron.
79(5): 882-891.
Landis T.D., Tinus R.W., McDonald S.E. & Barnett J.P.,
1990. The container tree nursery manual, Vol. 2.
Agriculture Handbook 674. USDA Forest Service,
Washington DC: 88 p.
31
McCreary D. & Canellas I., 2005. Restoration of oak
woodlands in Mediterranean ecosystems. In: Stanturf J.A. & Madsen P. (eds), Restoration of boreal
and temperate forests, CRC Press: 294-295.
McGee C.E. & Loftis D.L., 1986. Planted oaks perform
poorly in North Carolina and Tennessee. North.
J.Appl. For. 3: 114-115.
Pantera A., Papadopoulos A., Fotiadis G. & Papanastasis V., 2008. Distribution and phytogeographical
analysis of Quercus ithaburensis ssp. macrolepis in
Greece. Ecologia Mediterranea 34: 73-82.
Pantera A., Genitsaropoulos C., Fotiadis G. & Mertzanis
A., 2010. Soil compaction effects on the germination and growth of valonia oak seedlings (Quercus
ithaburensis ssp. macrolepis). In: Proceedings of the
7th panhellenic rangeland congress, Xanthi, 14-16
October 2010: 129-134.
Pantera A. & Papanastasis V.P., 2012. Competitive
effects of herbaceous species on water potential and
growth of Quercus ithaburensis ssp. macrolepis seedlings. Global Nest J. 14(4): 525-531.
Pope P.E., 1993. A historical perspective of planting
and seeding oaks: progress, problems, and status.
In: Loftis D. & McGee C.E., (eds.), Oak regeneration: serious problems, practical recommendations
(symposium proceedings). Gen. Tech. Rep. SE-84.
USDA, Forest Service, Southeastern Forest Experiment Station, Asheville: 224-240.
Rey Benayas J.M. & Camacho-Cruz A., 2004. Performance of Quercus ilex saplings planted in abandoned Mediterranean cropland after long-term interruption of their management. For. Ecol. Manage.
194: 223-233.
Ruehle J.L. & Kormanik P.P., 1986. Lateral root morphology: a potential indicator of seedling quality
in northern red oak. Research Note SE-344, USDA
Forest Service, Southeastern Forest Experiment Station, New Orleans: 1-5.
Salifu K.F., Jacobs D.F., & Birge Z.K.D., 2009. Nursery nitrogen loading improves field performance of
bareroot oak seedlings planted on abandoned mine
lands. Restor. Ecol. 17: 339-349.
Samara T. & Tsitsoni T., 2014. Selection of forest species for use in urban environment in relation to their
potential capture to heavy metals. Global NEST
Journal. 16(5): 966-974.
Scagel R.K., Binder W.D., Krumlik G.J., 1993. Sitespecific relationships of root growth potential to
outplanting performance of coastal Douglas-fir
(Pseudotsuga menziesii [Mirb.] Franco) under
controlled irrigated field conditions. Research
Report RR 93002-HQ, B.C. Ministry of Forests,
Research Branch, Victoria: 12 p.
Schultz R.C. & Thompson J.R., 1997. Effect of density
control and undercutting on root morphology of
32
1 + 0 bareroot hardwood seedlings: five-year field
performance of rootgraded stock in the central USA.
New For. 13: 301-314.
Teclaw R.M. & Isebrands J.G., 1993. An artificial regeneration system for establishing northern red oak on
dry-mesic sites in the Lake States, USA. Ann. Sci.
Forest. 50: 543-552.
Thompson J.R. & Schultz R.C., 1995. Root system
morphology of Quercus rubra L. planting stock
and 3-year field performance in Iowa. New For. 9:
225-236.
Tsakaldimi M., Ganatsas P., Gounaris N., Zagas T., Tsitsoni T. & Chatzistathis A., 2000. Water stress effect
on the quality of one-year old container seedlings.
In: Proceedings of the 9th Hellenic forestry association conference, Kozani 17-20 October 2000:
170-179.
Tsakaldimi M., Zagas T., Tsitsoni T. & Ganatsas P.,
2005. Root morphology, stem growth and field performance of seedlings of two Mediterranean evergreen oak species raised in different container types.
Plant Soil 278: 85-93.
Tsakaldimi M., Tsitsoni T., Ganatsas P. & Zagas T.,
2009. A comparison of root architecture and shoot
morphology between natural regenerated and
container seedlings of Quercus ilex L. Plant Soil
324: 103-113.
Tsakaldimi M., Ganatsas P. & Jacobs D.F., 2013. Prediction of planted seedling survival of five Mediterranean species based on initial seedling morphology.
New For. 44: 327-339.
Tsitsoni K.T., Tsakaldimi M. & Prodofikas Ch., 2011.
Root growth potential and seedling morphological
attributes of four Mediterranean hardwood species.
In: Book of abstracts of the 12th international Mediterranean ecosystems conference. MEDECOS XII.
“Linking Science with Resource Management”, Los
Angeles, 6-9 September: 220.
Valdecantos A., Cortina J. & Vallejo R., 2006. Nutrient
status and field performance of tree seedlings
planted in Mediterranean degraded areas. Ann.
Forest Sci. 63: 249-256.
Williams H. & Stroupe M., 2002. First-year survival
and growth of bareroot and container water oak
and willow oak seedlings grown at different levels of
mineral nutrition. Faculty Publications. Paper 203.
http://scholarworks.sfasu.edu/forestry/203.
Wilson E.R., Vitols K.C. & Park A., 2007. Root characteristics and growth potential of container and
bare-root seedlings of red oak (Quercus rubra L.)
in Ontario, Canada. New For. 34: 163-176.
Zaady E. & Perevolotsky A., 1995. Enhancement of
growth and establishment of oak seedlings (Quercus
ithaburensis Decaisne) by inoculation with Azospirillum brasilense. For. Ecol. Manage.72: 81-83.
Creation of an Integrated System Model
for Governance in Urban MTEs and
for Adapting Cities to Climate Change:
Preliminary Results
Création d’un modèle intégré pour la gouvernance
des écosystèmes méditerranéens urbains et pour adapter
les villes aux changements climatiques : premiers résultats
Thekla TSITSONI1*, Nikolaos GOUNARIS2, Aimilia B. KONTOGIANNI1,
Valia XANTHOPOULOU-TSITSONI3
1. School of Forestry & Natural Environment, Laboratory of Silviculture,
Aristotle University of Thessaloniki, University Campus,
P.O. Box: 262, 54124 Thessaloniki, Greece.
2. CEO of Homeotech Co Company, Grigoriou Lampraki 210, 54352 Thessaloniki, Greece.
3. School of Economics, Department of Development and Planning,
Aristotle University of Thessaloniki, University Campus,
P. O. Box: 178, 541 24 Thessaloniki, Greece
*
Received: 20 January, 2015; First decision: 30 April, 2015; Revised: 4 December, 2015; Second decision: 14 December, 2015
Abstract
Urban trees grow under adverse conditions,
governed by the combinatorial effect of multiple natural and anthropogenic factors. Global
climate change adds new challenges regarding
urban green management that should be taken
into account when designing future green
urban policies. The development of a smart
Information and Communications Technology
(ICT) system and the establishment of a continuously up to date information system regarding
urban trees is a key issue for future management which aims at:
–– climate change adaptation, by providing an
instrument for measuring the cities CO2 emission offsets by CO2 sequestration of the tree
biomasses;
–– efficient utilization of resources that are
spent for urban tree management in order to
decrease the cities’ environmental footprint;
–– enhancement of the cities’ social fabric by promoting citizen participation during the decision making process regarding urban trees.
The project was selected for funding after a
competitive process by the General Secretary
of Research and Development. The Municipality of Thessaloniki was selected as the key study
area. The core of the project is the development
of a software suite named GreenTree. Through
the GreenTree client Android application, 105
different sets of data are collected for each
urban tree. The urban tree inventory includes
37,328 tree sites on the pavements, from which
we found 1,239 dead trees, 2,787 empty sites
and 937 trees which had to be removed because
they had been planted in inappropriate location and they disturbed the circulation of cars
or pedestrians. The numbers above clearly show
the significant need for the establishment of a
reliable and smart monitoring system for urban
tree management. This system could also help
manage the decision making process. Finally,
the numbers show that urban environment can
be easily improved by applying fast and cheap
measures of tree replanting and replacement.
Keywords: urban forestry, green spaces, urban
tree management, street trees, GreenTree Software.
33
Thekla Tsitsoni, Nikolaos Gounaris, Aimilia B. Kontogianni, Valia Xanthopoulou-Tsitsoni
Résumé
Les arbres urbains se développent dans des
conditions difficiles, régies par l’effet combiné
de multiples facteurs naturels et anthropiques.
Le changement climatique mondial ajoute de
nouveaux défis en matière de gestion urbaine
“verte” qui devrait être prise en compte lors de
l’élaboration des futures politiques urbaines
écologiques. Le développement d’un système
intelligent en Technologie de l’Information
et de la Communications (TIC) et la mise en
place d’un système d’information, mis à jour en
continu, concernant les arbres urbains est un
enjeu clé pour la gestion future qui vise à :
–– l’adaptation au changement climatique, en
fournissant un instrument pour mesurer les
émissions de CO2 que les villes compensent
par la séquestration du CO2 dans la biomasse
des arbres ;
–– l’utilisation efficace des ressources qui sont
dépensées pour la gestion des arbres en milieu
urbain afin de réduire l’empreinte environnementale des villes ;
–– le renforcement des liens sociaux dans les villes
en favorisant la participation des citoyens
pendant le processus de prise de décisions
concernant les arbres urbains.
Le projet a été sélectionné pour financement
après un processus compétitif organisé par le
Secrétaire Général de la Recherche et du Développement. La municipalité de Thessalonique a
été choisie comme zone d’étude principale. Le
cœur du projet est le développement d’une suite
de logiciels nommée GreenTree. Via l’application android GreenTree, 105 ensembles distincts
de données ont été collectés pour chaque arbre
urbain. L’inventaire des arbres en milieu urbain
comprend 37 328 sites de plantation d’arbres
sur les trottoirs, parmi lesquels ont été trouvés
1 239 arbres morts, 2 787 absents de leur site
(trous vides sur les trottoirs) et 937 arbres devant
être retirés. Les chiffres ci-dessus montrent clairement la nécessité de la mise en place d’un
système de suivi fiable et intelligent pour la
gestion des arbres urbains. Ce système pourrait
aussi faciliter la gestion du processus de prise
de décisions. Enfin, les chiffres montrent que
l’environnement urbain peut être facilement
amélioré par l’application de mesures rapides
et bon marché pour la replantation et le remplacement des arbres.
Introduction
The world is undergoing the largest wave of
urban growth in history, as it is calculated that
above half of the world population lives in
urban regions while by 2030 the urban population is expected to be twice as large as the
Mots clés : foresterie urbaine, espaces verts,
gestion des arbres en milieu urbain, arbres de rue.
34
corresponding rural (Samara & Tsitsoni 2010,
Lang 1999, Sandberg 1999, OECD-ECMT
1995, Lambert & Vallet 1994). Increasingly,
urban green space is seen as an integral
part of cities, providing a range of services
to both the people and the wildlife living in
urban areas (James et al. 2009, Wolf 2004,
Nowak & Dwyer 2000). So, for the last 50
years there has been a growing realization that
the solutions to most of environmental problems reside in making cities more efficient in
their consumption of energy and materials and
disposing of waste products, and in altering
patterns of urban development to reduce the
amount of impervious “grey” infrastructure
and to increase the amount of “green” infrastructure, particularly trees (Carreiro 2006).
With this recognition and resulting from the
simultaneous provision of different services,
there is a real need to identify a research
framework in which to develop multidisciplinary and interdisciplinary research on urban
green space (James et al. 2009). This realization has been expressed in the concepts of the
eco-cities movement, adopted by many environmentalists and urban designers throughout
the world (Register 2002).
A major problem that globally has to be faced
is the climate change and mostly what deals
with global warming. The definition of climate
change refers to any significant change in the
measures of climate lasting for an extended
period of time (EPA 2015). Climate change
is caused by factors, such as biotic processes,
variations in solar or volcanic eruptions etc.,
but the main threat, mostly because of human
activity is referred to as global warming and
that is the recent and ongoing global average
increase in temperature near the Earth’s surface. According to NASA, the global temperature increased by 0.77 oC since 1880 while
nine of the ten warmest summers on record
have occurred the last 15 years (NASA 2015).
Natural and human factors cause changes in
Earth’s energy balance, leading, among other
things to increased greenhouse effects. Within
this context, urban green spaces can play a
central role in both climate-proofing cities and
in reducing the impacts of cities on climate
(Gill et al. 2007). Urban vegetation through its
physiological features could provide potential
help to the cities’ adaptation and mitigation
strategies for climate change. Global climate
change adds new challenges regarding urban
green management that should be taken into
account when designing future green urban
policies.
and for Adapting Cities to Climate Change: Preliminary Results
Adaptation deals with preparing for inevitable
climate change while mitigation is the act of
limiting further climate change, for example
by reducing emissions of greenhouse gases.
Urban green spaces can help to alleviate
the consequences of climate change mostly
by cooling. Trees, especially when located
close to buildings can reduce temperatures,
acting as natural air conditioners through
their evapotranspiration and providing shade
with their foliage, therefore reducing energy
consumption required to maintain comfortable climatic conditions. Even small green
spaces can have a cooling effect – parks of 1
or 2 hectares only have been found to be 2 oC
cooler than surrounding areas (Shashua-Bar
& Hoffman 2000). The extent of the cooling
effect is greatest when temperatures beyond
the park are highest (Handley & Carter 2006).
At the same time, urban trees and green spaces
can mitigate the impacts of climate change
through the absorption of the CO2 that takes
part in greenhouse gases and act as carbon
sinks. Additionally trees that are placed close
to buildings lead to lower consumption of
energy for mechanical cooling and heating,
decreasing CO2 emissions.
All green spaces help urban areas adapt to
the impact of climatic change regardless
of whether they are park, private garden or
street trees, with location, structure, composition and spatial configuration of them to
influence their ecological qualities and functions (Turner et al. 2005). At the same time
vegetation type that is characterized by the
plant species, the size of the trees and the
mixture of the species and the proportion of
urban tree canopy as well influence the impact
level. The challenge is to find functional and
as low budget as it is possible ways for adaptation and mitigation solutions based on urban
greening providing the total of physiological,
sociological, economic and aesthetic benefits.
Identifying and describing the benefits of
urban trees to a community is the first step in
gaining support for an urban forestry program
of tree planting, maintenance and replacement. An urban tree inventory is essential
because the information about the quantity
and quality of the existing vegetation and its
characteristics are important for urban greening management. Nevertheless, there is a lack
of urban tree inventory and monitoring protocols and standards. These protocols could
detect change over time and across cities,
while providing flexibility required by diverse
users.
A great proportion of the urban infrastructure
consist of urban streets where people walk,
shop, meet and generally participate in many
social and recreational activities that make
urban living enjoyable (Wolf 2004). The
use of trees as an element of the landscape
is an important design concept that has been
used throughout the world, and continues to
shape the aesthetics and function of the streets
(Gezer & Gül 2009).
Street trees are one of the most important components of urban green space and they play an
important role in city’s aesthetics as people’s
first impression of a city comes from its street
landscape (Jacobs 1993). Street trees are complex to study because this entails technical,
aesthetical, biological and ecological knowledge (Küçük & Gül 2005). Moreover there is
no homogeneous urban environment or site, as
it is a conglomeration of soils, microclimates
and other site conditions. Street trees should
possess strong apical growth, strong branching angles, overall high aesthetic values, predictable growth rates and, in general, have a
potential for a long life span (Gezer & Gül
2009), although they also sometimes grow in
adverse environments. Over recent decades,
a growing proportion of the commonly used
species have shown increasing difficulties in
coping with the conditions at urban paved
sites (Sjöman et al. 2010). Overall, trees in
these environments tend to be greatly exposed
to heat, low air humidity, periods of critical
water stress, high lime content and high soil
pH, limited soil volume, pollutants and deicing salts (Pauleit 2003, Sieghardt et al. 2005)
Under this perspective a project has been compiled using as a study case the metropolitan
area of Thessaloniki in northern Greece. The
project was selected for funding by the General Secretary of Research and Development
after a competitive process. The main goal of
the project is to create a standard system for
the exercise of governmental and decision
making in the practice of urban forestry in
a holistic and integrated way. The concept
of holistic lays in a comprehensive manner,
examining the needs and problems of urban
green infrastructures taking into consideration
as many factors as possible, anthropogenic or
not, each of which has different weight. This
integrated system includes all processes, services and products considered necessary for
the existence of the holistic approach to the
exercise of urban forestry.
35
The policy that is followed for the goal to be
achieved contains the following objectives:
1) to create the appropriate guidelines for the
management of urban greening, through the
proper treatment of the wooden species;
2) adaptation strategies for the inevitable climate change and mitigation of the factors that
lead to climate change;
3) to provide an efficient appliance for measuring the city CO2 emission offsets by CO2
sequestration of the tree biomasses;
4) to provide an efficient utilization of
resources that are spent for urban tree management in order to decrease the cities’ environmental footprint; and
5) to enhance the cities’ social fabric by promoting citizen participation during the decision making process regarding urban trees.
The main tool that is used is a smart Information and Communications Technology (ITC)
system that had been developed for that purpose and the establishment of a continuously
up to date information system regarding urban
greening. The aim of this paper is to show the
preliminary results of this research and try to
make an estimation of the sufficiency and the
functionality of the existing urban greening
of Thessaloniki.
Methods
Study area
The area where the specific project is taking
place is the whole territory of the municipality
of Thessaloniki, the second-largest city of the
country and the capital of central Macedonia,
in northern Greece. According to the preliminary results of the 2011 census, the municipality of Thessaloniki today has a population of
322,240, while its urban area has a population
of 790,824.
Thessaloniki lies on the northern fringe of
the Thermaic Gulf on its eastern coast and is
bound by Mount Chortiatis on its southeast.
Its proximity to imposing mountain ranges,
hills and fault lines, especially towards its
southeast have historically made the city
prone to geological changes.
Thessaloniki’s climate is directly affected
by the sea it is situated on. The city lies in
a transitional climatic zone, so its climate
displays characteristics of a mosaic of microclimates. As reported by Hellenic National
36
Meteorological Service, the total character of
the climate is humid subtropical climate (Cfa)
that borders on a semi-arid climate (BSk)
– according to Köppen climate classification.
The total annual precipitation is 458.4 millimetres due to the Pindus rain shadow drying
the westerly winds. However, the city has a
summer precipitation between 21 to 31 millimeters – with August the driest month and
November the wettest – which borders it close
to a hot-summer Mediterranean climate (Csa).
The mean annual temperature in Thessaloniki
is 15.6 °C. During the coldest winters, temperatures can drop to -10 °C, while the minimum temperature ever recorded is -14 °C. On
average, Thessaloniki experiences frost for
about 30 days per year. The coldest month of
the year is January. Thessaloniki’s summers
are hot with rather humid nights. Maximum
temperatures usually rise above 30 °C and
sometimes over 40 °C. The maximum temperature ever recorded is 42 °C. On average,
Thessaloniki experiences hot waves for about
30 days per year. The hottest month of the year
is July. The average wind speed for June and
July is 20 km/h while in winter the average
wind speed is about 26 km/h (Samara & Tsitsoni 2014). Papakostas et al. (2014) represent
the variation of temperature and annual average temperature during the period from 1983
to 2012 as it is shown to the Figures 1a and
1b. The same authors conclude that the total
increase in the annual average temperature
from the first to the third decade was 1.1 °C
with an upward trend.
Experimental design
From August 2013 to October 2014 and during two growing seasons, every single street
tree of the municipality of Thessaloniki was
recorded in order to build an integrated street
tree inventory. In the inventory qualitative
and quantitative information was included.
Hence every above-ground characteristic of
each individual as well as site features were
estimated or measured (Table 1).
The development of a smart Information and
Communications Technology (ITC) system
and the establishment of a continuously up to
date information system regarding urban trees
is a key issue for future management. The core
of the project is the development of a software
suite named GreenTree. Its use makes it easy
to collect data and make a tree inventory that is
based on field measurements and estimations
for both trees and their site features. Through
the GreenTree client Android application, 105
different sets of data were collected as mentioned above.
Breast height diameter and tree height data
were measured directly in situ using callipers and laser measurement instruments.
Tree crown projection extracted using ellipse
equation and four radii from the tree trunk R1
and R2 along and R3 and R4 vertically the
sidewalk. The crown volume derived using
standard geometry shapes equations (Troxel
et al. 2013).
30
1983-1992
1993-2002
2003-2012
Ambient Temperature [°C]
25
20
15
10
5
0
Jan
Fer
Mar
Apr
Mai
Jun
Jyl
Aug
Sep
Okt
Nov
Dec
Month
Figure 1a – V
ariation of temperature for the period from 1983 to 2012 (Papakostas et al. 2014).
18
Ambient Temperature [°C]
Average per year
Fitter curve
17
16
15
1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011
Year
Figure 1b – A
nnual average temperature for the period from 1983 to 2012 (Papakostas et al. 2014).
Table 1 – Set of tree measurements and evaluation.
Tree Species
Tree Location
Height
Crown height
Crown diameters
Breast height diameter
Tree health data
Soil data
Proposed future measurements
Existence of Utilities’ elements
Tree importance
(i.e. historical, monumental tree etc)
Maintenance data
(i.e. existence of grade, irrigation etc)
37
Results
Street tree characteristics
The preliminary analysis of the project data
showed that the area of total green spaces
under the jurisdiction of the municipality
services covers about 906,000 m2, less than
5% of the total city’s area, as analytically it is
represented to the Table 2.
The total number of tree sites of the pavements
that were surveyed is 41,672 belonging on
76 different species. To facilitate data processing, the species with population less than 50
individuals did not participate to the calculations, thus they eliminated from this exhaustive inventory. The revised inventory includes
37,328 tree sites of the pavements. They represented 45 wooden species from 24 families
– with the majority of them being forest species. 74.34% of the individuals belonged to 12
species, while 25.66% belonged to 33 species
(Tables 3 and 4). From these 33 species, 25
species were not much represented (< 1%).
At the family level, from the total number of
Table 2 – U
rban green area per category and per resident for the territory of the municipality of Thessaloniki.
Urban greening Categories
Area (m2)
m2 / resident
% of Municipality’s
Total area
Parks, road islets
527,475
1.64
2.85%
Street trees
378,480
1.17
2.05%
905,955
2.81
4.90%
University Campus
132,551
0.41
0.72%
Military basis
93,321
0.29
0.50%
Hospitals
28,378
0.09
0.15%
Open spaces
43,193
0.13
0.23%
Streams
216,593
0.67
1.17%
Stadiums
63,278
0.20
0.34%
Cemeteries
8,898
0.03
0.05%
Churches
15,000
0.05
0.08%
Total
601,212
1.87
3.25%
Private green spaces (yards, gardens)
970,940
3.01
5.25%
2,478,107
7.69
13.40%
Municipality’s urban green spaces
Total
Urban green spaces under special
management status
Total green area of the municipality of Thessaloniki
Table 3 – S
pecies occurrence proportion.
Species
38
Presence %
Species
Presence %
Species
Presence %
Sophora japonica
15.52
Cercis siliquastrum
1.91
Alnus glutinosa
0.41
Robinia pseudoacacia
12.78
Albizia julibrissin
1.88
Paulownia tomentosa
0.40
Acer negundo
9.62
Fraxinus sp.
1.83
Ficus carica
0.39
Ligustrum japonicum
7.62
Olea europaea
1.68
Pinus nigra
0.35
Koelreuteria paniculata
6.20
Aesculus hippocastanum
1.34
Broussonetia papyrifera
0.33
Platanus prientalis
4.72
Acer pseudoplatanus
0.91
Ginkgo biloba
0.29
Citrus x aurantium
4.31
Cupressus sp.
0.89
Pinus brutia
0.27
Celtis australis
2.81
Morus sp.
0.81
Magnolia grandiflora
0.25
Liquidambar orientalis
2.81
Catalpa bignonioides
0.77
Quercus sp.
0.24
Tilia sp.
2.73
Prunus cerasifera
var. pissardii
0.70
Pittosporum tobira
0.23
Populus x canadensis
2.63
Nerium oleander
0.69
Pinus sp.
0.17
Ulmus sp.
2.59
Chamaerops humilis
0.66
Prunus domestica
0.16
Populus alba
2.22
Populus nigra
subsp. thevestina
0.54
Eriobotrya japonica
0.15
Acer campestre
1.99
Laurus nobilis
0.50
Thuja plicata
0.14
Hibiscus syriacus
1.97
Ailanthus altissima
0.43
Acacia sp.
0.13
Table 4 – Number of species and individuals on each family.
Family
Species
Individuals
%
Species
Individuals
%
Fabaceae
6
12,115
32.46
Cupressaceae
2
384
1.03
Aceraceae
3
4,672
12.52
Rosaceae
3
379
1.02
Oleaceae
3
4,154
11.13
Pinaceae
3
294
0.79
Sapindaceae
2
2,814
7.54
Bignoniaceae
1
286
0.77
Ulmaceae
2
2,016
5.40
Apocynaceae
1
259
0.69
Salicaceae
3
2,011
5.39
Arecaceae
1
248
0.66
Platanaceae
1
1,763
4.72
Lauraceae
1
188
0.50
Malvaceae
2
1,757
4.71
Simaroubaceae
1
162
0.43
Rutaceae
1
1,609
4.31
Betulaceae
1
154
0.41
Altingiaceae
1
1,050
2.81
Paulowniaceae
1
150
0.40
Moraceae
3
574
1.54
Gingoaceae
1
108
0.29
24, 3 of them hold 56.1% of individuals (12
species), 91% of individuals belonged to 10
families, and 11 families had a representation
lower than 1%. The 45 species consisted of 5
gymnosperms and 40 angiosperms – 64.5%
deciduous and 22.3% evergreen. Approximately 50% of the planted species were large
and medium sized trees (Table 5). Finally,
44.5% were introduced species – mainly from
Asia – while 48.9% were native to Mediterranean region, Balkan Peninsula or generally
to Europe (Table 6).
Table 5 – Number of species in life form categories.
Life form
Number of
Species
%
Large Tree
10
22.22
Large – Medium tree
10
22.22
Medium tree
3
6.67
Medium – Small tree
8
17.78
Small tree
5
11.11
Small tree – Shrub
6
13.33
Shrub
1
2.22
Depend on species
2
4.44
Family
The cumulative crown volume that came from
378,480 m2 of urban street trees equalled
2,633,959 m3. For all 45 species, population
diagrams of height, breast height diameter
(dbh), crown projection and crown volume
created are shown on Figures 2, 3, 4 and 5.
Breast height diameter (Figure 2) and crown
projection area (Figure 3) followed an ascending trend, showing that tree growth in diameter was not affected by pruning and that crown
projection areas were less affected than the
other two silvicultural characteristics measured (i.e. tree height and crown volume).
Tree height and crown volume diagrams
showed an abrupt reduction in number of
existing trees in the 3rd class in comparison
with the 3rd of dbh class, 7.203 vs. 15.401. This
result can easily be explained by the intense
height reduction interventions on street trees
and hence on the crown volume. It is commonly known that these interventions are not
always appropriate; the actions of pruning
are greatly exaggerated and inappropriate but
happen because of lack of personnel expertise and low maintenance cost needs. On the
other hand the greenhouse gases sequestration
is directly related to total leaf surface and thus
directly to the volume of the crown.
Table 6 – Species origin.
Origin
Species
Origin
Species
Mediterranean region
13
Europe, Asia, Africa
1
Asia
13
Australia
1
Northern America
5
Balkans
1
Europe, Asia
3
Northern hemisphere
4
Europe, Asia and North America.
1
Hybrids
2
When only the genre is known
depend on species
1
39
Figure 2 – Frequency in diameter classes.
Figure 3 – Frequency in crown area classes.
Figure 4 – Frequency in height classes.
Figure 5 – Frequency in crown volume classes.
Estimation of CO2 sequestration
At this point an effort was made to obtain a
coarse image about the potential augmentation of CO2 sequestration that the replacement
of approximately 12,000 trees would provide;
this concerns the dead individuals and the
empty pavement tree sites. This theoretical
study examined the effect that these 12,000
trees would play in greenhouse gases sequestration through a correlation of crown volume
distribution with the existing distribution of
dbh. A wide range of tolerance of ± 20% was
used to cover the great variety of conditions
prevailing in city trees. By using complex
My SQL queries in database of 36,089 living trees it was found the cases where crown
volume distribution of each species deviates
from the dbh diameter distribution more than
± 20%. As a result 22 of the 40 tree species
showed discrepancies greater than ± 20%.
40
A theoretical model was developed where a
virtual transfer from volume categories that
show surplus, to them of deficit crown volume
categories was performed following species
dbh distribution. If there were no surplus categories, the transfer occurred proportionally
from the other categories in volume deficit,
provided that this should not be exceeded ±
20%. As a result a theoretical increased in
crown volume equal of 375,388 m3 or 14.56%
of the current crown volume of those 40 species derived (Figure 6). Even if this result is
not a spectacular increase in the total urban
trees crown volume, it gives an idea of what
is achievable by applying low cost managerial
measures. For this analysis, My SQL queries
and PHP programming language are used.
Table 7 – General view of street tree health condition.
Excellent
Condition
Good Condition
Moderate
Condition
Bad Condition
Worst
Condition
Dead
Empty pavement
tree site
5,554 (13.32%)
13,068
(31.35%)
9,537
(22.88%)
6,236
(14.96%)
3,257
(7.82%)
1,239
(2.98%)
2,785
(6.68%)
Table 8 – Preliminary management recommendations.
Additionally, tree health was to be excellent
to moderate for 68% of the trees, 10% of the
tree needed to be replaced immediately, while
there are 2,787 empty pits (Table 7). Finally,
in situ management recommendations were
proposed (Table 8).
Conservation
Maintenance
Removal
Replacement
9,849 (23.63)
18,974 (45.53)
937 (2.25)
11,912 (28.60)
Discussion
Every community tree planting program coordinates several processes in order successfully
manage public trees. To support a vigorous
population of trees, these programs plan and
design planting areas, select species for individual sites, coordinate planting activity, perform regular maintenance and pest management, and remove hazardous trees in a timely
manner (Burcham 2009).
Figure 6 – Frequency in potential crown volume classes.
GreenTree specifications
The specific ongoing project involves the collection of reliable and accurate information
for each tree managed by the municipality of
Thessaloniki, the organization of this information in a functional geospatial database,
which will be updated and will be editable
for completion of new data fields, if necessary. This database will be analyzed to derive
statistics and indicators to 1) identify problematic individuals requiring urgent intervention, for example pruning, removal, etc.,
2) assess the health situation and the goal
achievement level in a management period
and to plan maintenance works, supplementing or replacing individuals and calculating
the corresponding cost, 3) compare work cost
and quality that has to be done by private contractors, and 4) examine alternative scenarios
for the distribution of the annual municipal
resources by prioritizing the importance of the
planned projects according to their contribution to achieving the objectives of the strategic
management plan. The project will also take
place actions to inform the public and publicize the results of the project.
In order to implement all the above acts,
GreenTree has been developed, as a special
software to be a basic tool in urban greening management, scheduled to enter into pilot
operation at the end of the project. The purpose
of this tool is to help municipality services to
manage their urban greening specifically in
relation to the potential CO2 sequestration and
the regulation of temperatures.
As referred above, each entry in GreenTree
contains information for each individual like:
tree record number, species code and name,
health status and management needs, dbh,
height, crown measurements and site features, etc. The data provided by the inventory
will lead to the evaluation of the silvicultural
characteristics that describe the tree as carbon
sink. A number of different kinds of tables can
be printed that will help the visualization of
the data and summarize the results. The software is not designed only for professionals
but also for beginners or unskilled workers;
as skill levels improve the quality of the input
and output increases as well.
41
GreenTree stages of implementation
During our collaboration with the municipality services, we were faced with 1) a disappointing lack of data and 2) disorganization of
the operations related to urban forestry, both
due to the absence of proper staff numbers and
financial resources. The gap of this procedure
is expected to be solved by GreenTree system,
as it will make data collecting process much
easier.
So, the first step that had to be done was to put
a set of guiding principles of urban greening
management, starting by standardizing the
monitoring protocols and the construction of
the tree inventory. The challenge was to make
protocols as simple as possible and at the same
time flexible enough for different situations.
These technical guidelines for long-term data
collection and urban tree inventory development followed Miller (1997): management
planning for street tree population involves
an inventory of trees and community values;
then, this inventory is used to develop management goals, the next step is to develop
a management plan to achieve these goals
(selection, establishment and maintenance
of street trees), and finally a feedback allows
monitoring the entire process. Additionally,
guidelines for urban green planning and
management under the perception of climate
change will be provided, including a national
network of public and private organizations as
well that deal with urban forestry.
General considerations
One practical concern faced by all community tree planting programs is the need for
biological diversity. The heavy, often exclusive, reliance on a small number of species
contributed to the proliferation of speciesspecific landscape pests (Burcham 2009).
In the case of Thessaloniki, there are more
than 70 tree species participate in urban vegetation. Nevertheless, on the one hand there
is an uncontrolled distribution and on the
other hand about 30 species are represented
with population less than 50 individuals and
11 families have rare appearance, less than
1% of the total population of street trees. So,
even though species diversity enhances landscape ecological balance and value, it makes
it difficult to organize the spatial and temporal
planning of management treatments.
Additionally, after in situ observation, there
were evidence of inadequate tree species
42
selection for particular features of a site, like
damages to pavements or other infrastructure:
the size of the mature plant had not been taken
into consideration in relation with the size of
the available growth space. The life form of
approximately 50% of the planted trees in the
streets of Thessaloniki belongs to the category
of large tree, making the individual inappropriate in the most cases. The large proportion
of introduced species shows that no ecological
properties and their relations with the environmental conditions were taking into consideration when the selection was made.
The inadequate plant selection for urban use
also cause health problems to the citizens as
the pollen of specific species such as poplar,
pines, olive and plane trees can have an allergic effect (Papageorgiou 2003).
The existence of power and telephone lines in
urban areas is indisputable. Planting trees that
are expected to grow high in their maturity
leads to inevitable trimming and pruning, that
in most of the case is inadequate and without
any spatial and temporal planning. Topping
and lopping is very often with irreversible
impact on tree health.
Conclusion
In Thessaloniki the basic rules took into
account as Miller’s (1997) proposed model
for selecting species for urban uses were
never applied. The main factors to take into
account are site conditions, such as cultural
and environmental elements, economic factors like planting and maintenance costs and
social factors as functional utility, landscape
enhancement and public safety. So, the basic
properties of the trees must be: 1) climate
adaptation, 2) resistance to diseases, 3) large
phenotypic plasticity in the plant materials,
4) root quality, 5) growth potential and form at
maturity phase, 6) wind and snow resistance,
7) drought resistance and 8) tolerance of air
pollution (Sæbø et al., 2003).
Species selection, planting location, and
cultural practices all have an impact on the
ultimate visual quality, health, and cost of
street tree maintaining. The use of appropriate species, the proper location of plantings,
and the implementation of a program of preventative maintenance of the street trees, will
allow a cost effective tree management system. Action programs related to trees in the
urban are: (I) policy making, planning and
designing, (II) technical focus, such as selection programs and establishment techniques
and (III) management aspects (Konijnendijk
& Randrup 2002).
Apparently the ecological and functional species selection is of utmost importance but the
general point is how to apply an efficient management plan with the lower ecological and
social cost. Urban greening, when sustainably
managed, can have a central role in climate
change mitigation and adaptation.
Specifically, from the preliminary results of
this project can be concluded that:
––the potential crown volume of the existing
trees can be increased to 14.5%;
––the replacement of dead individuals will
increase the number of trees within 3%;
––the filling of the empty pavement spaces
will increase the number of trees at 6.7%.
The numbers above clearly show the significant need for the establishment of a reliable
and smart monitoring system for urban tree
management. This system could also help
in managing the decision making process.
Finally, the numbers show that urban environment can be easily improved by applying
fast and cheap measures of tree replanting and
replacement.
References
Burcham D.C., 2009. Urban Forest Management for
Multiple Benefits: an Analysis of Tree Establishment Strategies Used By Community Tree Planting
Programs. Thesis submitted to the Faculty of the
University of Delaware.
Carreiro M.M., 2006. The growth of cities and Urban
Forestry. In: Carreiro M.M. (eds. ), Ecology, Planning, and Management – International perspectives.
Springer, New York.
EPA, 2015. Glossary of Climate Change Terms. Unites
States Environmental Protection Agency. Available
on: http://www.epa.gov/climatechange/glossary.
html#Climate_change (Accessed in 25-12-2014)
Gezer A. & Gül A., 2009. Kent Ormancılıgı-KavramsalTeknik ve Kulturel Boyutu. SDU Orman Fakültesi,
Kitap Yayın No 86. Isparta. 29 p. (In Turkish).
Gill S., Handley J., Ennos R., & Pauleit S., 2007. Adapting cities for climate change: the role of the green
infrastructure. Journal of the Built Environment 33
(1):115-133.
Handley J. & Carter J., 2006. Adaptation Strategies for
Climate Change in the Urban Environment (ASCCUE). Available on: http://www.sed.manchester.
ac.uk/research/cure/downloads/asccue_final_
report_national_steering_group.pdf. (Accessed in
25-12-2014)
Jacobs J., 1993. The death and Life of Great American
Cities. New York, 449 p.
James P., et al. 2009. Urban Green – Towards an integrated understanding of greenspace in the built
environment. Urban Forestry and Urban Greening
8(2): 65-76.
Konijnendijk C. C. & Randrup T. B., 2002. Editorial.
Urban Forestry & Urban Greening 1(1): 1-4
Küçük V. & Gül A., 2005. Isparta kentiçi yol
agaçlandirmalari üzerine bir arastirma, Süleyman
Demirel Üniversitesi, Fen Bilimleri Enstitüsü
Dergisi. 9-3:111-118 (In Turkish).
Lambert, J., & Vallet, M., 1994. Study Related to the
Preparation of a Communication on a Future
EC Noise Policy. INRETS LRN Report No 9420,
INRETS- Institut National de Recherche sur les
Transports et leur Securite, Bron.
Lang, W.W., 1999. Is noise policy a global issue, or is it a
local issue? In: Cuschieri J., Glegg S. and Yan Yong
(eds.), Proceedings of Internoise 99 - The 1999
International Congress on noise Control Engineering, December 1999, Fort Landerdale, 1939-1943.
Miller R. W., 1997. Urban Forestry: Planning and Managing Urban Greenspaces. Prentice Hall, Englewood Cliffs.
NASA. 2014. Global Climate Change. Vital signs of
the Planet. Available on: http://climate.nasa.gov/.
(Accessed in 25-12-2014)
Nowak D. J., & Dwyer J. F., 2000. Understanding the
benefits and costs of urban forest ecosystems. In:
Kuser, J. (ed.), Urban and Community Forestry in
the Northeast. Springer, New York: 11-25.
OECD-ECMT, 1995. Urban Travel and Sustainable
Development. European Conference of Ministers
of Transport, Organization for Economic Co-operation and Development, OECD, Paris.
Papageorgiou V., 2003. Inventory of plant polulations
and their allergic pollen types in region of N. Moudania Chalkidiki and their clinical importance.
PhD Thesis, Aristotle University of Thessaloniki.
(In Greek)
Papakostas K.T., Zagana-Papavasileiou P., & Mavromatis T., 2014. Analysis of 3 Decades Temperature
Data for Athens and Thessaloniki, Greece – Impact
of Temperature Changes On Energy Consumption
For Heating And Cooling Of Buildings. International Conference ADAPT to CLIMATE. Nicosia,
27-28 March 2014.
Pauleit, S. 2003. Urban street tree plantings: Identifying
the key requirements. Proceedings of the Institute of
Civil Engineers-Municipal Engineers 156(1):43-50.
Register, R., 2002. Eco-cities: Building Cities in Balance with Nature. Berkeley Hills Books, Berkeley,
320 p.
Samara T., & Tsitsoni T., 2010. The effects of vegetation on screening road traffic noise from a city ring
road. Noise Control Engineering Journal. 59 (1),
pp. 68-74
Samara T., & Tsitsoni T., 2014. Selection of forest species for use in urban environment in relation to their
potential capture to heavy metals. Global NEST
Journal. 16 (5): 966-974.
Sandberg, U., 1999. Abatement of traffic, vehicle and
tire/road noise- the global perspective. In: Cuschieri
J., Glegg S. and Yan Yong (eds.), Proceedings of
Internoise 99 - The 1999 International Congress on
noise Control Engineering, December 1999, Fort
Landerdale: 37-42.
43
Shashua-Bar L. & Hoffman M.E., 2000. Vegetation as a
climatic component in the design of an urban street:
an empirical model for predicting the cooling effect
of urban green areas with trees. Energy and Buildings 31 (3): 221-235
Sieghardt, M., E. et al. 2005. The abiotic urban environment: Impact of urban growing conditions on urban
vegetation. In: Konijnendijk C.C. et al. (eds.), Urban
Forests and Trees. Springer, New York: 281-323.
Sæbø A., et al. 2005. The selection of plant materials
for street trees, park trees and urban woodlands, In:
Konijnendijk C.C. et al. (eds.), Urban Forests and
Trees. Springer, New York: 257-280.
Sjöman H., et al., 2010. Habitat Studies Identifying
Potential Trees for Urban Paved Environments: A
44
Case Study from Qinling Mt., China. Arboriculture
& Urban Forestry 36(6): 261-271
Troxel B., Piana M., Ashtona M.S., & Murphy-Dunninga C., 2013. Relationships between bole and
crown size for young urban trees in the northeastern USA. Urban Forestry & Urban Greening 12
(2): 144-153.
Turner K., Lefler L. & Freedman B., 2005. Plant communities of selected urbanised areas of Halifax,
Nova Scotia, Canada. Landscape and Urban Planning 71:191-206.
Wolf K. L., 2004. Economics and public value of urban
forests. Urban Agriculture Magazine. Issue on
Urban and Peri-urban Forestry 13: 31-33.
Vegetation Dynamics of Coastal Dunes
with Juniperus spp. in Crete, Gavdos
Pinelopi DELIPETROU1, Dany GHOSN2, George KAZAKIS2, Panagiotis NYKTAS2,
Electra REMOUNDOU2, Ioannis N. VOGIATZAKIS3, *
1. Department of Biology, University of Athens, Greece
2. Department of Geoinformation in Environmental Management,
Mediterranean Agronomic Institute of Chania, Greece
3. School of Pure & Applied Sciences, Open University of Cyprus, Cyprus
* Corresponding author: [email protected],
Tel: +357 22411933
Received: 1 March, 2015; First decision: 15 October, 2015; Revised: 26 October, 2015; Accepted: 10 November, 2015
Abstract
We aimed to determine the composition, structure and ecological processes of the vegetation of the coastal dunes with Juniperus spp. in
Crete, Gavdos and Chrysi in the South Aegean,
Greece, in order to apply sound habitat management and restoration. Vegetation composition,
structure and zonation were investigated with
plots and transects. Data from seven study sites
were classified using TWINSPAN. The major patterns of the vegetation data and their relation
to environmental variables were explored by
DCA and CCA ordination techniques. Ellenberg
indicators for moisture, nutrients, and salt were
used to characterise the community types identified. The habitat’s floristic composition includes
142 plant species. Five principal community
types were identified. Vegetation distribution
was related to geomorphology and disturbance
gradients. The analysis of transect data identified 20 vegetation units on incipient dune, foredune, interdune and hind dune. Vegetation and
geomorphological data were used to construct
sand dune profiles for each site, while a set of
36 keystone and 76 indicator species were identified. The sites examined have varied levels of
dune development and face different threats.
Habitat management should address grazing
and trampling at the local level but also land
use changes at the catchment level.
Keywords: CCA, DCA, Ellenberg values,
geomorphologic units, Natura 2000, PCA.
Introduction
Coastal dunes are found on shores all over the
world, from polar latitudes to the tropics (Martinez et al. 2004). Plant growth forms, sea and
wind are the key factors that create, mould or
destroy these structures. Among these factors,
plant growth plays a vital role in all stages of
coastal dune formation by reducing the effect
of the wind, trapping sand, and thus encouraging further dune growth (Musila et al. 2001).
Mediterranean coastal dunes are characterized by strong seasonality, with warm wet
winters and hot dry summers (Castillo et al.
2002). Vegetation is exposed to stress caused
by drought, high evaporative demand, and
both high irradiance and temperatures (Garcia
Novo et al. 2004). Coastal dunes with Juniperus spp. (juniper) are widespread on the coasts
of Europe, but not common.
Maritime juniper woodlands (Juniperus macrocarpa Sm.) represent the late successional
stage of Mediterranean dunes and cliffs.
According to Rivas-Martinez et al. (1980),
maritime juniper woodlands with Juniperus
phoenicea, represent the mature ecosystem
on outer dunes and cliffs of the Mediterranean coasts. They have a high ecological value
in relation to their ability to retain sand, and
45
Pinelopi Delipetrou, Dany Ghosn, George Kazakis, Panagiotis Nyktas, Electra Remoundou, Ioannis N. Vogiatzakis
provide habitat for flora and fauna. Coastal
dunes with Juniperus spp. are vulnerable due
to their extreme ecological position (Gehu
1993), withstanding the effects of wind,
drought, salt, erosion and pH (Brown &
McLachlan 1990).
At a european level, coastal dunes with Juniperus spp. are threatened by logging, urban
and tourism development, fires, invasive species, erosion, grazing, fragmentation, pollution but also by juniper’s restricted natural
regeneration (Anon 1992). Global climate
changes may exacerbate these threats through
changes in minimum winter temperatures.
These changes can modify the distribution of
plant species, and can combine with frequent
storms resulting in damage to the seaward side
of the dune slacks. Changes in dune structure
and ecosystems are often cyclical, with periods of loss (erosion) balanced by periods of
gain (sand deposition) so these tendencies will
be obvious only in the long term (Corre 1991).
Junipers usually occur in isolated stands of
different extent, but large populations still survive in natural or semi-natural conditions. The
protection of coastal woodlands is a priority
due to the range of functions they perform for
biodiversity, recreation and sand stabilization.
In the European Union, coastal dunes with
Juniperus spp. are a priority habitat (2250*) of
the Habitats Directive (Council of European
Community 1992).
Worldwide studies of the relationship between
vegetation and dune formation (Castillo et al.
1991) show that coastal dunes form a complex system of habitats for plants due to the
combined effect of steep environmental gradients which are related to the distance from the
shoreline, elevation but also to salinity, nutrients, humidity, wind, and inundation (Ranwell
1972). Plants suited to the dune habitat are
highly specialized and able to cope with limiting factors such as water shortage, soil fertility,
and salt spray. In this prograding shoreline, a
temporal and spatial dune succession is found.
In the late 20th century, conservation efforts
were focused on plant inventories, threatened
species and maintaining as many natural habitats as possible. Today, there is a paradigm
shift with emphasis placed on understanding
the physical and ecological processes within
and beyond these habitats as a means for
improved conservation. This is a necessity
in the South Aegean which contains a high
concentration of sand dunes with junipers,
46
currently facing increased pressures (Heslenfeld et al. 2008).
Vegetation is by far the most important component of the biota on sand dunes because it
is directly involved in establishing the dune
formation and structure (Wolfe & Nickling
1993) while there is constant interaction of
dune vegetation and geomorphology (Maun
2009). Despite their ecological value, baseline
information on juniper dune woodlands composition, structure and ecology are lacking in
Greece. This is crucial if this priority habitat is to be effectively managed and restored
where necessary. Moreover, describing plant
communities is challenging, due to the lack
of characteristic species and the resulting difficulties in phytosociological classification
(Mayer 1995).
This study focuses on the sand dune systems
of the islands of Crete, Gavdos and Chrysi in
South Aegean, which are part of the European
network of protected areas Natura 2000, as a
precursor to sound conservation actions. The
focus is on coastal juniper woodlands, which
are important indicators of the general physical conditions of the coastal dune environment. The objectives of this paper are to: a)
identify and describe the plant communities
of juniper shrub and vegetation zonation in the
sand dunes in the islands of Crete, Gavdos and
Chrysi, b) investigate the relationship between
plant communities and environment; c) assess
the implications of the results for dune habitat
conservation and management.
Methods
Study areas
In Greece this habitat is found in 15 sites
mainly in the south covering 1,077 ha, i.e.
4.26% of the total surface of Natura 2000 in
the country (Picchi 2008). This study was carried out in all Natura 2000 sites designated
as coastal dunes with Juniperus spp. habitats
on the islands of Crete (sites of Kedrodasos
and Falasarna), Gavdos (sites Agios Ioannis,
Lavrakas and Sarakiniko) and Chrysi (Figure 1).
Data collection
Fieldwork took place in May 2009 (Kedrodasos and Gavdos) and in April-May
2010 (Chrysi and Falasarna). Vegetation
Vegetation Dynamics of Coastal Dunes with Juniperus spp. in Crete, Gavdos and Chrysi Islands (Greece)
Figure 1 – L ocation of the 2250* priority habitats in Chrysi, Kedrodasos-Elafonisi, Gavdos and Falasarna (upper left) and the sampling
scheme employed for vegetation plots (bottom left) and transects (bottom right).
composition and structure were studied in 38
plots of 30 m × 30 m established within the
juniper scrub forest in all sites: 6 in Kedrodasos-Elafonisi (E1 – E6), 4 in SarakinikoGavdos (Gs1 – s4), 6 in Ag. Ioannis-Gavdos
(Gi1 – Gi4), 8 in Lavrakas-Gavdos (Gl1 –
Gl8), 6 in East Chrysi (Ca1 – Ca6), 4 in West
Chrysi (Cw1 – Cw4), and 2 in Falasarna (F1
– F2). Vegetation plot locations were selected
subjectively so as to a) span the studied areas;
b) represent various levels of disturbance;
and c) include both Juniperus macrocarpa
and Juniperus phoenicea shrub. Plot locations were recorded with a GPS. Two subplots
were established in each 30 m × 30 m plot: a
10 m × 10 m subplot including juniper shrub
or trees (Figure 1) and a 3 m × 3 m subplot
in the open (outside the canopy of shrubs or
trees). Percentage cover of vegetation layers (trees, tall shrubs, low shrubs, and herbs)
and woody vegetation height were visually
estimated in all subplots. The abundance
of all plants in each subplot was recorded
using the modified Braun-Blanquet 9-grade
cover-abundance scale (Braun-Blanquet
1932; Wikum & Shanholtzer 1978). Environmental data included slope and aspect, geology from 1:50000 scale maps, grazing pressure assessed in the field qualitatively (high,
medium, low). Plot use was also assessed
qualitatively (high, medium, low) based on
the presence of broken branches observed.
Vegetation zonation was studied in 12 line
transects perpendicular to the seashore, 2 of
which in Kedrodasos-Elafonisi, 1 in Sarakiniko-Gavdos, 3 in Agios Ioannis-Gavdos, 2
in Lavrakas-Gavdos, and 4 in Chrysi. Their
length varied from 55 to 225 m depending on
dune morphology while their locations were
selected subjectively so that they were representative of vegetation zonation at each site
and were geo-referenced using a GPS. Total
plant cover, rock cover and vascular species
cover-abundance, using the modified BraunBlanquet 9-grade scale (Braun-Blanquet
1932), were recorded in contiguous 5 m × 5 m
quadrats along each transect.
47
Life forms were based on Raunchier (1937)
as modified by Ellenberg & Müller-Dombois
(1967) while chorology was based on Jahn
& Schönfelder (1995). Voucher specimens of
plants were collected and stored in the Herbarium of the Mediterranean Agronomic Institute at Chania, Crete. A list of the total species recorded together with their functional
attributes (Böhling et al. 2002; Jahn & Schönfelder 1995; Espejel et al. 2004) is presented
in Appendix I. Data analysis included only
the Ellenberg indices for moisture (F), nutrients (N), and salt (S), because they presented
significant variation among the species, while
the indices for light (L, values 6 to 9, semilight to light plants), pH (R, values 7 to 9,
weakly basic to basic soils), and temperature
(T, values 7 to 8, fairly hot to hot sites) did not
present significant variation.
Data analysis
TWINSPAN analysis (Hill 1979a) in PCORD (Ver. 4, McCune & Mefford 1999) and
JUICE© (Ver. 7.0.61, Lubomír Tichý 19982010) was used to classify the vegetation
and transect plot data. The 10 m × 10 m and
3 m × 3m plots were analyzed separately. The
phi-coefficient (Chytrý et al. 2002) with data
standardization estimated by JUICE was used
as a fidelity measure. Species with significance
< 0.05 according to the Fischer’s exact test
were given zero fidelity value. Species were
assigned to vegetation classes (Appendix II)
according to, Rivas-Martínez et al. (2002).
Detrended Correspondence Analysis (DCA;
Hill 1976) was employed to evaluate the type
of response model for selection. Based on
the DCA results, a unimodal response model
for the data was accepted (Jongman et al.
1987). Therefore Canonical Correspondence
Analysis (CCA) in CANOCO© (ter Braak &
Smilauer 1998) was used to assess vegetation
environment relationships The CCA model
and the significance of the fitted environmental variables were evaluated by the Monte
Carlo permutation test.
Table 1 – Number of species (in bold) in sand dune juniper shrub vegetation in
the studied sites and floristic similarity (Sørensen index) between them.
Kedrodasos
Falasarna
Kedrodasos
Falasarna
Gavdos
92
0.14
0.43
37
0.25
62
Chrysi
0.43
0.31
0.48
48
Gavdos
Chrysi
75
Results
Floristic analysis
The total number of species in sand dune juniper shrub based on the 30 m × 30 m plots in
all the sites was 142. Species numbers varied
among sites and floristic similarity was relatively low (< 50%) (Table 1). The highest species richness, in terms of absolute numbers,
was observed in Kedrodasos and the lowest
in Falasarna. Floristic similarity was higher
among Chrysi, Gavdos, and Kedrodasos and
lower between Falasarna and all the other
sites. The dominant families were Poaceae
(14%), Compositae (13%) and Fabaceae
(11%). The dominant chorological element
was the Mediterranean (26%) followed by
the East-Mediterranean (12%), South-Mediterranean (10%) and Mediterranean-Atlantic
(9%). The Greek endemic element accounted
for 5% of the flora. Chrysi had the highest
number of endemics.
Vegetation composition and
structure/Vegetation plot data
analysis
The species with the highest frequency in all
the plots were the phryganic shrubs Coridothymus capitatus and Phagnalon graecum
(26 and 21 plots, respectively); a dry grassland
small annual, Hedypnois cretica (22 plots);
and the ammophilous annuals Pseudorlaya
pumila, Triplachne nitens, Vulpia fasciculata and Lotus halophilus (29, 26, 21, and 24
plots, respectively). These, along with J. macrocarpa and J. phoenicea were considered as
characteristic of all the juniper shrub communities of the area of Crete.
TWINSPAN analysis of the 10 m × 10 m
plots resulted in 5 distinct groups (Figure 2,
Appendix II). The vegetation units corresponding to these groups are referred to as
“community types” (cf. Mueller-Dombois
& Ellenberg 1974), since they have not been
placed into existing phytosociological categories. Species with a high constancy (frequency > 50%) and a high degree of fidelity
(phi-coefficient > 50%) in each group were
termed characteristic (or diagnostic) and were
used to name each community type. The life
form spectra (Table 2) and Ellenberg number
spectra (Table 3) were also used for the characterisation of the community types (Table 5)
identified, namely:
A1: Juniperus phoenicea-Periploca angustifolia community type.
A2: Malcolmia flexuosa-Nigella stricta community type.
B1: Silene colorata-Ononis natrix community
type.
B2: Silene succulenta-Cutandia maritima
community type.
B3: Elytrigia juncea-Medicago marina community type.
Ecological gradients/Vegetation plot
data analysis
Based on the DCA preliminary analysis a
unimodal response model for the data was
assumed (ter Braak & Smilauer 1998). The
full CCA model explained 1.837 of the
total inertia of the species data (3.930) and
selected automatically five significant variables (p < 0.05), in decreasing order of significance: conglomerate, dune depth × interdune
depth, grazing, plot use, and marl. Thus, the
environmental variables best explaining the
species data were related to geomorphology
and anthropogenic disturbance (Figure 3).
In Figure 3 the upper quadrants include sites
on marl (e.g. Chrysi), and the lower sites on
sandstone (e.g. Gavdos), and conglomerate
(e.g. Kedrodasos). Dune-interdune depth and
slope increase from top to bottom. On the
contrary grazing and plot use increase from
bottom right to top left. Juniperus macrocarpa
is located at the bottom left quadrant but close
to the start of the axes along with the species
Figure 2 – Hierarchical classification of the vegetation data (community types
A1, A2, B1, B2, B3 as in text).
Table 2 – Life form spectra of juniper shrub plant community types.
The number of plants and percentage cover (in parenthesis) for each
life form are shown.
Plant Community
Life Form
A1
A2
B1
B2
B3
Therophyte scapose
23 (8)
39 (32)
22 (18)
18 (15)
14 (13)
Therophyte caespitose
5 (4)
12 (10)
11 (5)
9 (8)
5 (4)
2 (1)
2 (0)
2 (1)
2 (1)
1 (1)
Therophyte parasitic
Therophyte rosette-forming
2 (1)
5 (2)
Hemicryptophyte
3 (1)
3 (2)
1 (4)
4 (4)
Hemicryptophyte biennial
2 (0)
2 (1)
1 (0)
1 (0)
2 (0)
Geophyte bulbous
3 (1)
3 (2)
1 (1)
1 (0)
1 (1)
Chamaephyte fruticose
5 (5)
3 (4)
9 (8)
4 (7)
3 (4)
Chamaephyte suffruticose
7 (8)
3 (2)
5 (3)
3 (1)
3 (3)
Phanerophyte caespitose
4 (73)
4 (44)
5 (58)
3 (63)
3 (68)
Geophyte rhizomatous
4 (2)
Phanerophyte scapose
1 (5)
Table 3 – Ellenberg number spectra for moisture (F), nutrients (N), and salinity (S) of juniper shrub community types.
The number of plants and percentage cover (in parenthesis) for each Ellenberg number are shown.
Plant Community Type
Ellenberg
Number
A1
F
N
A2
S
F
N
B1
S
F
11 (6)
N
F
2 (1)
6 (4)
2 (0)
1 (0)
46 (31)
6 (5)
29 (23)
4 (9)
17 (13)
4 (5)
13 (9)
7 (7)
12 (56)
9 (8)
6 (67)
6 (7)
3 (3)
2 (0)
1
4 (3)
25 (15)
2
13 (78)
1 (0)
11 (11) 15 (42)
3
12 (7)
5 (34)
7 (35) 21 (22) 4 (30)
6 (37) 18 (21) 5 (53)
4
10 (4)
7 (3)
1 (0)
1 (2)
5
5 (1)
14 (8)
1 (0)
4 (3)
9 (5)
2 (1)
6
7
1 (0)
8
4 (1)
4 (1)
20 (16) 10 (7)
5 (6)
11 (11) 8 (65)
3 (1)
7 (54) 10 (14) 2 (67)
S
F
N
S
1 (0)
8 (71) 10 (12) 2 (68) 11 (76)
11 (10)
5 (5)
11 (8)
2 (1)
2 (1)
11 (9)
1 (1)
2 (1)
12 (10)
3 (1)
9 (6)
3 (1)
7 (6)
1 (4)
5 (3)
8 (7)
3 (2)
1 (0)
16 (14)
1 (0)
11 (9)
1 (0)
11 (14)
(0)
1 (1)
8 (7)
1 (1)
1 (0)
15 (13)
1 (0)
12 (10)
1 (0)
9 (7)
1 (0)
8 (10)
1 (1)
1 (0)
1 (0)
4 (1)
1 (1)
10 (6)
1 (0)
10 (12)
3 (7)
4 (3)
1 (3)
3 (2)
1 (1)
1 (0)
4 (1)
9
x
5 (2)
N
B3
S
0
5 (4)
B2
1 (0)
6 (3)
10 (47) 4 (38)
8 (13)
8 (22)
5 (16)
5 (6)
3 (4)
3 (3)
49
50
F
E
27
2110,
2230
F2.2
2260
2110,
2230
2250
F3.1
F3.2
B2.4
F3
2260
F2
26
2250
2110
2250
2250
2250
2250
2250
2250
2250
2250
2250
2190
2230
2110
2250
2110
2260
2110
Habitat
Type
F2.1
B2.3
B*
19,20,
21
F1
B3.3
18
25
B2
16, 17
24
B2.2
23
A1.1
6
B2.1
B3.2
11
15
D4
10
B1
D3
8,9
12
D2
7
A2
B3.1
4
13, 14
D1
5
C2
3
D
C1
1,2
C
Vegetation
Unit
Sub
group
Group
foredune
incipient dune,
interdune
foredune, hind
dune
incipient dune,
interdune
foredune, hind
dune
foredune, hind
dune
incipient dune
foredune, hind
dune
foredune
foredune, hind
dune
hind dune, rarely
foredune
foredune
hind dune
foredune, hind
dune
hind dune
foredune
dune slack
interdune
incipient dune,
rarely interdune
foredune
incipient dune,
interdune
incipient dune,
foredune
incipient dune
Zone
Gavdos (Lavrakas, Sarakiniko)
Gavdos (Lavrakas, Sarakiniko)
Kedrodasos, Gavdos (Lavrakas, Agios
Ioannis, Sarakiniko)
Chrysi (East), Kedrodasos, Gavdos
(Agios Ioannis)
Gavdos (Lavrakas)
Gavdos (Agios Ioannis, Lavrakas)
Chrysi (West)
Gavdos (Agios Ioannis)
Chrysi (East), Gavdos (Lavrakas)
Gavdos (Agios Ioannis)
Gavdos (Lavrakas)
Kedrodasos
Chrysi (West)
Chrysi (West)
Chrysi (West)
Chrysi (East, West)
Chrysi (mainly West), Gavdos (Agios
Ioannis)
Chrysi (East)
Chrysi (mainly East)
Kedrodasos, Gavdos (Agios Ioannis)
Kedrodasos
Site
Silene succulenta, Plantago squarrosa,
Medicago marina, Medicago littoralis,
Hedypnois cretica
Mostly low cover vegetation with few ammophilous species, lacking any other shrub species and dry
grassland species. Divided into 2 subcommunities based on the presence of shrubs.
Silene succulenta-Pseudorlaya pumila
Juniper shrub (community type B2) with Ononis natrix, Coridothymus capitatus, lacking any other
shrub species on foredune or more inland at the margins of erosion corridors.
Ammophilous primary dune vegetation at the frontal zone of Sarakiniko and at large shrub openings
on deep dune at Agios Ioannis.
Dune shrub with Ononis natrix-Coridothymus capitatus on foredune, behind the frontal zone or on
hind dune.
Mostly low cover vegetation with an array of ammophilous and dry grassland species. Divided into 3
subcommunities based on the presence of shrubs.
Very low cover, fragmental vegetation of frontal dune or shrub openings.
Dune shrub with Ononis natrix-Coridothymus capitatus on foredune, behind the frontal zone, or on
hind dune.
Juniper shrub (community type B2) representing an aspect of B2.1 at more stabilised dune.
A group of juniper shrub quadrats with floristic composition between communities B1 and B2
(apparently at transition zones).
Juniperus macrocarpa, Plantago
squarrosa
Frontal, very low cover primary dune vegetation.
Juniper shrub (community type B3), on foredune.
Juniperus macrocarpa, Limoniastrum
monopetalum
Juniperus macrocarpa, Limonium
elaphonesicum, Ononis natrix,
Coridothymus capitatus
Juniper shrub (community type B2) with few species.
Juniperus macrocarpa, Cutandia maritima
Limonium elaphonesicum
Juniper shrub (community type B2) with a few ammophilous species and stable presence of
synanthropic vegetation species, mainly at hind dunes highly disturbed by human use.
Juniperus macrocarpa, Mercurialis annua,
Cakile maritime
Juniper shrub (community type A1) at inland dunes.
Juniperus phoenicea, Asparagus
stipularis, Periploca angustifolia
Juniper shrub (community type B2) , lacking sclerophyllous shrub and dry grassland species, on
foredune (frontal) or more inland at the margins of erosion corridors.
Juniper shrub (community type B3), on foredune (frontal) or more inland at the margins of erosion
corridors or at steep slopes.
Juniperus macrocarpa, Pancratium
maritimum, Silene sedoides
Juniperus macrocarpa, Silene succulenta,
Limonium elaphonesicum
Dune slack with Juncus heldreichianus, at flat wet dunes, behind the front vegetation zone.
Juncus helreichianus
Juniper shrub (community type B1).
Low cover dune grassland occurring on loose sand at inland shrub openings; Juniperus macrocarpa
seedlings are frequent.
Triplachne nitens, Erodium lacianiatum
Juniperus macrocarpa, Ononis natrix,
Erica manipuliflora
Ammophilous primary dune vengetation similar to D1 but lacking ammonitrophilous plants (e.g.,
Zygophyllum album, Cakile maritima) at the front zones and rarely at interior large shrub openings
with moving sand or blowouts.
Silene succulenta, Euphorbia paralias
Juniper shrub (community type A2).
Juniper shrub (community type B3), on foredune (frontal) or more inland at the margins of erosion
corridors or at steep slopes.
Juniperus macrocarpa, Zygophyllum
album
Juniperus macrocarpa, J. phoenicea,
Nigella stricta
Ammophilous primary dune vegetation including sand binders (Elytrigia juncea, Medicago marina,
Limonium graecum, Limoniastrum monopetalum) at the front zones or at interior large shrub openings
with moving sand or blowouts.
Dune shrub, occasional presence of ammophilous plants (Centaurea pumilio, Limonium
elaphonesicum)
Anthyllis hermaniae
Silene succulenta, Zygophyllum album
Ammophilous primary dune vegetation on flat dune with very low cover.
Description
Centaurea pumilio, Pancratium
maritimum
Characteristic species
Table 4 – Vegetation units identified by the transect plot data. Habitat type code as in Appendix III.
that occur in many plots and are not highly
affected by the environmental variables tested
such as Pseudorlaya pumila, Lotus halophilous and Pistacia lentiscus. These are candidates for keystone species due to their abundant and stable presence in the habitat. The
species closely related to larger dune depths
are candidates for indicators of deep dunes
and for keystone species are those that characterize communities B2 (Silene succulenta,
Cutandia maritima) and B3 (Elytrigia juncea, Medicago marina) and also Salsola kali,
Cakile maritima and Lycium schweinfurthii.
Juniperus phoenicea is located at the upper
quadrant since it occurs at shallower dunes
along with Periploca angustifolia and Ononis
reclinata. It is not an indicator of dunes on
marls since it also occurs on sandstone. The
species closely related to marls (e.g. Paronychia macrosepala, Helianthemum stipulatum) are candidates for indicators and for keystone species of shallow and moderate depth
dunes on marl. The species closely related to
conglomerates and higher rock cover are candidates for indicator and keystone species and
are the characteristic species of community
type A2, Malcolmia flexuosa, Nigella stricta,
and Minuartia mediterranea.
Vegetation zonation
TWINSPAN analysis of the transect plot
data resulted in four well distinguished large
groups, whose segregation was strongly
influenced by the locality: Group C – Anthyllis hermaniae-Centaurea pumilio (mostly
Kedrodasos); Group D – Silene succulentaPancratium maritimum (mostly Chrysi); and
two Silene succulenta-Pseudorlaya pumilaCutandia maritima groups; Group E – Juniperus macrocarpa (juniper shrub, all sites);
Group F – Ononis natrix-Coridothymus capitatus (mostly Gavdos). Silene succulenta,
Table 5 – Community types resulting from TWINSPAN analysis of plot data.
Community Type
Description
A1: Juniperus phoeniceaPeriploca angustifolia
Shrub community with Juniperus phoenicea often mixed with J. macrocarpa on shallow/moderately deep
dunes with shallow/ moderately deep interdunes, on marls. It is typical of flat, inner dunes and of the
transition zones to the shrublands of the inner non dunal vegetation. It was found in ungrazed sites of
Chrysi, but also in Gavdos. Floristically characterised by phanerophytes (notably P. angustifolia, restricted to
Gavdos and Chrysi in Greece), and chamaephytes (notably the rare in Greece Helianthemum stipulatum),
by high frequency/low cover of dry grassland species and by low participation of ammophilous species. The
ecological profile, based on Ellenberg indicators, shows dominance of plants of extremely dry to dry sites
(values 0–2), salt indifferent to medium halotolerant (values x, 0–2) and increased participation of plants of
nutrient extremely poor to poor sites (values 2–4) or indifferent.
A2: Malcolmia flexuosa-Nigella
stricta
Recorded only at Kedrodasos, at mostly grazed moderately deep dunes with moderately deep interdunes,
on conglomerate. It is characterised by reduced participation of shrubs (including the resistant to grazing
Verbascum spinosum), high frequency/high cover of annuals including few ammophilous species (notably
the rare Aegean endemic Nigella stricta), and a large number of dry grassland and synanthropic vegetation
species. The ecological profile, shows dominance of plants of extremely dry to dry sites (values 0–3)
and increased participation of plants of intermediate to extremely nutrient-rich sites (values 5–9) and of
halophobe to medium halotolerant (values 0–2) plants. The increased nutrient indicator values are mainly
due to the increased participation of synanthropic species (e.g., Urospermum picroides, Sonchus oleraceus,
Mandragora autumnalis).
B1: Silene colorata-Ononis natrix
This is the main Juniperus macrocarpa community type of moderately deep dunes on Gavdos and on Chrysi
and at the deep dunes of Kedrodasos. Floristically characterized by the dominance of phanerophytes and
chamaephytes (pines, sclerophyllous shrubs, phrygana), high frequency of annuals, and an array of dry
grassland species. It often represents the innermost dune zone preceding pine forest or shrub on stable
substrate. The ecological profile shows dominance of plants of extremely dry to dry sites (values 0–3)
and slightly to very halotolerant (values 1–3). Plants of nutrient poor sites (values 2–4) dominate, with an
increased frequency of plants of intermediate to nutrient rich sites (values 5–8). The increased nutrient
indicator values are due to both ammophilous (e.g. Vulpia fasciculata) and synanthropic species (e.g.
Sonchus oleraceus).
B2: Silene succulenta-Cutandia
maritima
The main Juniperus macrocarpa community type of deep dunes at all sites (rare at Kedrodasos). Floristically
characterized by a less prominent shrub layer and high frequency/high cover of annual and increased cover
of perennial herbs including ammophilous species (notably Silene succulenta occurring only in the project
sites in Greece) and fewer dry grassland species. The ecological profile is similar to community B1, but very
halotolerant plants are prominent and there is increased participation of salt stressed plants (values 7–8)
and nutrient rich sites (value 8). The increased nutrient indicator values are mainly due to ammophilous and
ammonitrophilous drift line species (e.g. Cakile maritima, Salsola kali).
B3: Elytrigia juncea-Medicago
marina
It occurs on foredunes and was found on deep dunes at Chrysi, and degraded, medium deep or flat dunes,
at Falasarna. Floristically characterized by a less prominent shrub layer and by increased perennial versus
annual herbs including mainly ammophilous species, especially sand binders (e.g. Limonium graecum,
Sporobolus pungens) and few dry grassland species. The ecological profile is similar to community B2, but
there is more increased frequency and cover of plants of salt stressed (values 6–8) and nutrient rich sites
(value 8). This attribute is exemplified by the nitrophilous, halotolerant shrub Lycium schweinfurthii
51
Figure 3 – C
CA plot of 36 vegetation plots (123 species), with symmetric biplot scaling, axes I (eigenvalue 0.400, p = 0.005) and II
(eigenvalue 0.342, p = 0.005). The two axes explain 18.9% of the variance of species data and 40.4% of the variance of
species-environment relation. Triangles: sites; circles: species; diamonds: characteristic species of the plant community
types; stars: most frequent species in all community types. Black arrows: variables with non significant contribution to the
model; coloured arrows: variables with significant contribution to the model. Coloured species’points represent the species
with a strong positive relation to the respective environmental variables (top 10 weighted averages of species with respect
to the variable).
Triplachne nitens and Lotus halophilous
were the species with the highest frequency
in all transect plots of all sites, except from
Kedrodasos. The analysis further separated
27 subgroups (maximum dissimilarity 0.508)
which were assigned to the 20 vegetation units
(Figure 4, Table 4). These represent the spatial
52
succession of sand dune community types
from the sea landwards, but since the transect
plots may represent fragments or marginal
zones of communities, the term “community
type” is not used in the phytosociological
sense. The juniper shrub vegetation units were
assigned to the community types identified
Figure 4 – H
ierarchical classification of the transect plot data. Primary dune/dune grassland community types (habitat types 2110,
2230): C1, D1, D2, D3, D4, F1, and parts of F2 and F3; dune slack (habitat type 2190): D4; low shrub community types
(habitat type 2260): C2, F2, F3; juniper shrub community types (habitat type 2250): A1, A2, B1, B2, B3, and part of F3.
For community type abbreviations see text.
by the plot data and named correspondingly.
Based on both the vegetation and the transect
plot data, the vegetation succession profiles
in the study sites were constructed (Figures
4, 5, 6, and 7).
Geomorphology, vegetation
and dune development
In this section we present the most characteristic relationships recorded between vegetation units and dune development while details
are given in Table 4. In Chrysi (Figure 5) the
dune system presents well developed primary,
secondary, and tertiary vegetation zones. At
the northwest part, a wide, moderately deep
incipient dune zone is formed with high cover,
typical ammophilous vegetation (mainly v.u.
D2). Subsequently, a moderately high foredune with juniper shrub (v.u. B3.2) and a low
and moderately high hind dune (v.u. A1) are
formed, with moving sand (v.u. D2) and dune
grassland openings (v.u. D3). At the northeast
part the incipient dune zone is also wide, but
ammophilous communities (v.u. D1, F2.2)
form a high cover but narrow front zone followed by wide openings with scarce vegetation (v.u. D1, D2).
At Kedrodasos (Figure 6) the primary vegetation zone is not well developed. Sparse, almost
flat incipient dunes with scattered ammophilous plants of low sand-binding ability of the
vegetation unit (v.u. C1) (Centaurea pumilio,
Pancratium maritimum, Pseudorlaya pumila)
or shrubs (v.u. C2) are formed at the front zone
of the east part and behind the rocky beach at
the west part.
At Sarakiniko-Gavdos (Figure 8), the incipient dune area is characterised by mostly low
cover ammophilous vegetation (v.u. F3.2) followed by a low foredune with phrygana (v.u.
F3.1) and then juniper shrub (v.u. B2.4). At
the west part of Agios Ioannis, Gavdos (Figure 7), a front almost flat zone with scarce
ammophilous plants (v.u. F1) is succeeded by
rising, moderately deep foredune with juniper
(v.u. B*, B2.1) and by deep dune areas with
juniper (v.u. B2.1) and large openings with
very low vegetation cover (v.u. F2.2, F3.2).
At the extended dune system of Lavrakas,
Gavdos (Figure 8) the incipient dune zone is
not well developed and includes low cover
ammophilous communities (v.u. F1) which
are succeeded by low cover phrygana (v.u.
F2.1) marking the transition to the moderately high foredune which is stabilised by the
same phrygana (v.u. F2.1, F3.1) and by juniper shrub (v.u. B2.2, B2.4, B2.3). The hind
dune develops on low dunes and shallow sand
deposits with often thick juniper shrub (v.u.
B*, B1).
53
Figure 5 – S
and dune vegetation profiles in Chrysi.
54
Figure 6 – S
and dune profiles in Kedrodasos.
Keystone and Indicator species
On the basis of vegetation analysis we identified keystone species for the habitat and indicator species for the geomorphological characteristics and the quality of the habitat with
regards to plot use and grazing. The indicator
species were selected by CCA. The species
with the top 10 positive weighted averages
with respect to each variable were selected as
showing positive relation and the species with
the bottom 10 negative weighted averages
were selected as showing negative relation.
The general literature for each species was
taken into account for the final selection (i.e.,
obviously generalist species were excluded).
The criteria for the selection of keystone species were: a) species identified as characteristic (diagnostic and/or constant) of plant communities resulting from TWINSPAN; b) the
most frequent species in habitat 2 250*, i.e.
those occurring in more than 40% of all the
plots; c) species that were identified as indicators of the geomorphological environmental
variables; d) species with high Ellenberg
indicator value; e) the set of keystone species should contain representatives of all the
functional attributes (see Appendix I). A set of
36 keystone species (31 for the habitat 2250)
and a set of 79 indicator species were identified (Appendix IV). Both sets can be used for
habitat quality assessment.
Discussion
The sand dune systems investigated in this
study, represent an array of geomorphological
conditions, a gradient of human influence and
geographical isolation as reflected in the range
of community types recorded. The plant species number of the juniper shrub community
types in each site (Table 1), was comparable
to that found in similar communities in Greece
(unpublished relevé data of NATURA 2000
sites) and western Mediterranean (Acosta et
55
Figure 7 – S
and dune profiles in Gavdos-Ag. Ioannis.
al. 2009). The community types of Kedrodasos which had the highest species numbers
also had the highest number of synanthropic
elements (sensu Mucina 1997; Acosta et al.
2006) (Appendix II).
Floristically similarity is higher between
Gavdos and Chrysi indicating stronger biogeographic affinities. The comparatively
low floristic similarity of the juniper shrub
56
community types among islands can be attributed partly to geographic isolation and partly
to the different environmental variables. Geographical vicariation in the Mediterranean
Juniperus macrocarpa communities was also
found by Géhu et al. (1990). In fact, the community types identified in the study sites seem
to represent yet another geosynvicariant(s) of
the Junipereta macrocarpae which cannot be
Figure 8 – S
and dune profiles in Gavdos-Lavrakas.
classified with the existing associations in
the South Aegean (Géhu et al. 1989). The life
form and Ellenberg number spectra (Tables 2
and 3) differentiated the juniper shrub communities primarily according to the abiotic
factors (foredune vs hind dune, dune depth,
transition to stable substrate) and secondarily
according to human impact (grazing). These
two functional attributes can detect essential
qualitative changes in species composition of
this priority habitat and can be used effectively
in monitoring and assessment of its conservation status.
The ecological gradient analysis of the juniper shrub communities was strongly influenced by the presence of species with narrow
distribution (e.g. Limonium elaphonisicum
and Limonium graecum) and this resulted
57
in biogeography masking to some extent the
influence of the ecological factors tested. This
supports the notion that random biogeographical events may affect community composition at a large scale (Forey et al. 2008). Also,
the effects of natural disturbance and possible
differences in precipitation and soil moisture
among the sites, important factors for dune
community composition (Forey et al. 2008;
Miller et al. 2009), were not tested. Nevertheless, the analysis highlighted the well documented relationship between geomorphology
and vegetation establishment in the dune systems (Provoost et al. 2011).
Most importantly, the analysis showed that
vegetation composition in the study sites is
affected by anthropogenic disturbance and
grazing. The effects of trampling on vegetation composition and structure in sand dunes
are well documented (Santoro et al. 2012) as
is the increased resilience of mobile and fixed
dunes compared to semifixed ones (Lemauviel & Rozé 2003). Grazing, when practiced
in a non-intensive manner, can be beneficial for the typical fixed dune communities
(Halada et al. 2011) and juniper regeneration
(Pihl et al. 2001). Although, this practice is
advocated in central and northern Europe
where the habitat’s natural successional trend
is always towards climax woodland, it may
not be beneficiary in the Mediterranean where
stable dunes face severe exposure (Picchi
2008). In our study, grazing favored synanthropic species at the expense of dry grassland
and ammophilous ones. Also, although these
effects were not quantified, observations in the
field revealed that grazing by goats may also
change the architectural structure of the juniper trees, especially of Juniperus macrocarpa.
Vegetation zonation was studied and profiles
constructed for each site since front vegetation
zones protection is crucial for the conservation of the fixed dunes where juniper develops (Hesp 2002). The plant community types
identified exhibit similarities with sand dune
communities described elsewhere in the Mediterranean (Mayer 1995; Sýkora et al. 2003)
and show a typical linear succession (zonation) (e.g. Cassar & Stevens 2002).
Of all three study areas only the dune systems in Chrysi, demonstrate well-developed
primary, secondary, and tertiary vegetation
zones. Vegetation zonation is better preserved in the western area which is less frequented by visitors compared to the eastern part, where there is degradation due to
58
trampling, sea-shell overcollection, and erosion. Extended drying and an unbalanced sex
ratio have been observed in Juniperus macrocarpa trees mainly in the eastern part, probably due to extreme drought stress episodes
(Thanos et al. 2010). It seems that vegetation
has recovered naturally in the western part
but that human disturbance has impeded this
process in the east part. Aeolian sand deposition on a rugged terrain is more important for
dune development in Gavdos and Kedrodasos. The typical primary dune vegetation with
ammophilous species is not well developed
while a shrub zone plays the main protective
role for the juniper shrub dunes. It is only in
Gavdos-Sarakiniko, the site most affected by
human disturbance, that the abiotic conditions
favor the development of typical primary dune
vegetation, but this is degraded due to trampling and grazing. In Kedrodasos, the whole
front vegetation is rather thin and this may be
due partly to the rugged terrain and partly to
excess trampling.
Habitat Management
Coastal sand dunes play an important role in
providing a range of ecosystem functions and
services (Barbier et al. 2011). The conservation of these habitats requires more detailed
knowledge of their ecology and condition.
To adhere to Habitats Directive (Council of
European Community 1992) European member states should report every six years on the
progress achieved in protected habitat conservation. Towards that end, the keystone and
indicator species for the habitat identified in
this study can be employed. In addition, the
different processes and dune development
stages that occur in the study sites require
habitat management plans tailored to individual sites’needs.
The conservation of the natural processes is
a major target in dune system management
(Van der Meulen & Salman 1996) due to
their effect (Ciccarelli et al. 2012; Miller et
al. 2009). Usually management interventions
to tackle the above effects should be combined
with relief from anthropogenic pressures (Picchi 2008). In this light and according to the
results of our study, it seems that interventions including restoration of the front dune
vegetation are warranted only in the east part
of Chrysi, in Kedrodasos and in Gavdos-Sarakiniko where vegetation recovery is improbable or may take too long.
The processes affecting the dune systems
of our study sites operate at different scales
from local (e.g. anthropogenic disturbance)
to catchment scale (e.g. land use changes)
which affects sediment transfer and therefore
habitat dynamics (Cassar & Stevens 2002).
Gavdos and Chrysi islands belong to the first
case where human impact is localized while
Falasarna and Kedrodasos are additionally
affected by agricultural and related road building activities at the catchment level.
In Gavdos, grazing should be excluded from
the priority habitat sites, while trampling
should be controlled with the use of designated paths. Controlling of camping should
be addressed by future management efforts
while grazing should be restricted particularly
in Falasarna and Kedrodasos. The proximity
of the habitat in Falasarna and Kedrodasos to
agricultural areas and grazing lands result in
increased grazing pressure. In Chrysi, there
are no grazing animals, while in Gavdos, grazing is not uniform across sites (i.e. intensive in
Sarakiniko, localized in Ag Ioannis and absent
in Lavrakas).
Mitigation measures to address habitat pressures may include sand trapping, transplanted
vegetation, fencing, dune walkovers and
environmental education campaigns. Most
of these actions were carried out as part of
the JUNICOAST LIFE+ Project. While these
activities at the site level are easier to implement, what still remains a challenge is controlling factors affecting the habitat at the catchment scale. Although a stakeholders group
was established early on in the LIFE+ project
the implementation of some of the actions
is still faced by obstacles at the community
level. This is yet again proven to be the major
challenge which nature conservation faces in
Europe, that is to persuade the public about
the importance of Natura 2000 (Papageorgiou
& Vogiatzakis 2006; Paloniemi et al. 2009).
Acknowledgements
This research was funded by the LIFE +
Nature Programme “Actions for the conservation of coastal dunes with Juniperus spp.
in Crete and the South Aegean (Greece)”
(LIFE07NAT/GR/000296). We are grateful
to Dr. Christina Fournaraki, curator at the
Herbarium of MAICh, for her help in species
identification.
References
Acosta A., Carranza .M.L. & Izzi .C.F., 2009. Are there
habitats that contribute best to plant species diversity
in coastal dunes? Biodivers Conserv 18: 1087-1098.
Acosta A., Ercole S., Stanisci A. & Blasi C., 2006. Sandy
coastal ecosystems and effects of disturbance in
Central Italy. JCoastal Res, SI 39: 985-989.
Anon., 1992. Council Directive 92/43/EEC on the conservation of natural habitats and of wild fauna and
flora. Official Journal European Commission 35.
Barbier, E.B., Hacker, S.D., Kennedy, C., Koch, E.W.,
Stier, A.C. & Silliman B.R. 2011. The value of estuarine and coastal ecosystem services. Ecol Monogr
81: 169-193.
Böhling N., Greuter W. & Raus T., 2002. Zeigerwerte
der Gefäßpflanzen der Südägäis (Griechenland).
Braun-Blanquetia. Review of Geobotanical Monographs 32: 1-108.
Braun-Blanquet J., 1932. Plant sociology. McGrawHill, London.
Brown AC &McLachlan A., 1990. Ecology of Sandy
Shores. Elsevier, Amsterdam.
Cassar L.F., & Stevens D.T., 2002. Coastal dunes under
siege: a guide to conservation for environmental
managers. International Environment Institute,
University of Malta.
Castillo S, Popma J, & Moreno-Casasola J. 1991.
Coastal sand dune vegetation of Tabasco and
Campeche, Mexico. J Veg. Sci 2: 73-88.
Castillo JM, Rubio Casal AE, Luque CJ, Luque T, &
Figueroa ME. 2002. Comparative field summer
stress of three tree species co-occurring in Mediterranean coastal dunes. Photosynthetica 40: 49-56.
Ciccarelli D., Bacaro G., & Chiarucci A. 2012. Coastline
Dune Vegetation Dynamics: Evidence of No Stability. Folia Geobot. 47: 263-275.
Chytrý M., Tichý L., Holt J., & Botta-Dukát Z. 2002.
Determination of diagnostic species with statistical
fidelity measures. J Veg. Sci 13: 79-90
Corre JJ. 1991. The sand dunes and their vegetation
along the Mediterranean coast of France. Their
likely response to climatic change. Landscape Ecol
6: 65-75
Council of the European Community, 1992. Council
Directive 92/43 EEC of 21 May 1992 on the conservation of natural habitats and wild fauna and flora.
Official Journal of the European Communities.
Espejel I., Ahumada B., Cruz Y., & Heredia A., 2004.
Coastal vegetation as indicators for conservation.
In Coastal Dunes (pp. 297-318). Springer Berlin
Heidelberg.
Ellenberg, H. & D. Mueller-Dombois. 1967. A key to
Raunkiaer plant life forms with revised subdivisions. Ber. Geobot. Inst. ETH Rübel, 37: 56-73.
Forey, E., Chapelet, B., Vitasse, Y., Tilquin, M., Touzard,
B. & Michalet, R. 2008. The relative importance of
disturbance and environmental stress at local and
regional scales in French coastal sand dunes. J Veg.
Sci 19: 493-502.
Garcia Novo F, Diaz Barradas MC, Zunzunegui M, Garcia Mora R, & Gallego Fernandez JB 2004. Plant
Functional Types in Coastal Dune Habitats. Chapter
Coastal Dunes, Springer, Series Ecological Studies
171: 155-169.
Géhu JM, Costa M, & Biondi E. 1990. Les Junipereta
macrocarpae sur sable. Acta Bot. Malacitana 15:
303- 309.
59
Géhu JM. 1993. The coastal woodlands of Europe. In:
Tekke R, Salman A. (eds.). Coastal Dune Woodlands along the Atlantic and North Sea Shores.
EUCC Internal Report Series 4, Leiden, pp. 6-16.
Géhu JM., Apostolidis N., Géhu-Franck J., & Arnold K.
1989. Premières données sur la végétation littorale
des îles de Rodhos et de Karpathos. Colloques phytosociologiques XIX: 545-582.
Halada L., Evans D., Romão C., & Petersen J. E., 2011.
Which habitats of European importance depend on
agricultural practices? Biodiver Conserv, 20: 23652378.
Heslenfeld P., Jungerius P., & Klijn J., 2008. European
coastal dunes: ecological values, threats, opportunities and policy development. In: Martínez, M. L.;
Psuty, Norbert P. (eds.) Coastal Dunes, 335-351.
Hesp P., 2002. Foredunes and blowouts: initiation, geomorphology and dynamics. Geomorphology 48(1):
245-268.
Hill MO., 1979a. TWINSPAN - a FORTRAN Program
for arranging Multivariate Data in Ordered Two
way Table by Classification of the Individuals and
the Attributes. Cornell University, Department of
Ecology and Systematics, Ithaca, New York.
Hill MO., 1979b. DECORANA - a FORTRAN Program
for detrended correspondence analysis and reciprocal averaging. Cornell University, Department of
Ecology and Systematics, Ithaca, New York.
Jahn R. & Schönfelder P., 1995. Exkursionsflora für
Kreta. Verlang Eugen Ulmer, Stuttgart (Hohenheim).
Jongman R. H., ter Braak C. J. F., & Tongeren O. F. R.,
1987. Data analysis in community and landscape
ecology. Pudoc, Wageningen. 299 p.
Lemauviel S., Gallet S., & Rozé F., 2003. Sustainable
management of fixed dunes: example of a pilot site
in Brittany (France). CRBiol 326: 183-191.
Martinez ML & Psuty NP. (eds.), 2004. Coastal dunes:
ecology and conservation. 386 pp. Berlin: SpringerVerlag. Ecological Studies 171.
Maun M. A., 2009. The biology of coastal sand dunes.
Oxford University Press.
Mayer A., 1995. Comparative study of the coastal vegetation of Sardinia (Italy) and Crete (Greece) with
respect to effects of human influence. IHW-Verlag,
Munich.
McCune B., & Mefford M. J., 1999. PC-ORD: multivariate analysis of ecological data; Version 4 for
Windows; [User‘s Guide]. MjM software design.
Miller TE., Gornish ES., Buckley HL., 2009. Climate
and coastal vegetation: disturbance, recovery, and
succession. Plant Ecol. 206: 97-104.
Mueller-Dombois D. & Ellenberg H., 1974. Aims and
methods of vegetation analysis. New York: Ed.
J. Wiley & Sons, 547.
Mucina L., 1997. Classification of vegetation: Past, present and future. J Veg Sci: 751-760.
Musila WM., Kinyamario JI. & Jungerius PD., 2001.
Vegetation dynamics of coastal sand dunes near
Malindi, Kenya. Afr J Ecol 39: 170-177.
Paloniemi R. & Vilja V., 2009. Changing ecological and
cultural states and preferences of nature conservation policy: the case of nature values trade in SouthWestern Finland. J Rural Stud 25: 87-97.
60
Papageorgiou K. & Vogiatzakis I.N., 2006. Nature Protection in Greece: an appraisal of the factors shaping
integrative conservation and policy effectiveness.
Environ. SciPolicy 9: 476-486.
Picchi S., 2008. Management of Natura 2000 habitats.
2250* Coastal dunes with Juniperus spp. European
Commission.
Pihl S., Ejrns R., Sgaard B., Aude E., Nielsen K.E., Dahl
K., & Laursen J.S., 2001. Habitat and species covered by the EEC Habitats Directive – A Preliminary
assessment of distribution and conservation status
in Denmark. National Environmental Research Institute, Denmark. 121 pp. - NERI Technical Report
no. 365. Available on: http://faglige-rapporter.dmu.
dk
Provoost S., Jones M. L. M., & Edmondson S. E., 2011.
Changes in landscape and vegetation of coastal
dunes in northwest Europe: a review. J Coastal
Conserv 15: 207-226.
Ranwell D. S., 1972. Ecology of Salt marshes and Sand
dune. Chapman Hall, London. Botanische Jahrbucher Systematik, 75, 563-583.
Rivas-Martínez S.,Díaz T.E., Fernández-González
F., Izco J., Loidi J., Lousã M., & Penas A., 2002.
Vascular plant communities of Spain and Portugal:
Addenda to the syntaxonomical checklist of 2001.
Itinera Geobotanica 15: 5-922.
Santoro R., Jucker T., Prisco I., Carboni M., Battisti C.,
& Acosta A. T., 2012. Effects of trampling limitation
on coastal dune plant communities. EnvironManage
49: 534-542.
Sýkora K.V., Babalonas D. & Papastergiadou E.S.,
2003. Strandline and sand-dune vegetation of coasts
of Greece and of some other Aegean countries. Phytocoenologia 33: 409-446
ter Braak CJF., 1994. Canonical community ordination.
Part 1. Basic theory and linear methods. Ecoscience
1: 127-140.
ter Braak CJF. & Smilauer P., 1998. CANOCO Reference Manual and User’s Guide to Canoco for Windows: Software for Canonical Community Ordination (version 4). Microcomputer Power, Ithaca, New
York.
Thanos C.A., Kaltsis A., Koutsovoulou K., Skourti E.,
& Sarris D., 2010. Deliverable A.3: Report on the
composition and structure of Juniperus macrocarpa
subpopulations, Chania, Greece. The JUNICOAST
project, Actions for the conservation of coastal
dunes with Juniperus spp. in Crete and the South
Aegean (www.junicoast.gr).
Tichý L., 2002. JUICE, software for vegetation classification. JVeg Sci 13: 451-453.
van der Meulen F. & Salman AHPM., 1996. Management of Mediterranean coastal dunes. Ocean Coast
Manag 30: 177-195.
Wikum D.A. & Shanholtzer G.F., 1978. Application
of the Braun-Blanquet cover abundance scale for
vegetation analysis in land development studies.
Environ Manage 2: 323-329.
Wolfe SA & Nickling WG., 1993. The protective role of
sparse vegetation in wind erosion. ProgPhys Geog
17: 50-68.
Caractérisation du fonctionnement
des steppes d’alfa marocaines par la méthode
de l’analyse fonctionnelle du paysage
Characterization of the Moroccan Alfa Steppe Functioning
Using the Landscape Function Analysis Method
Mchich DERAK1, Fernando T. MAESTRE2, José L. QUERO3, Victoria OCHOA3, Cristina
ESCOLAR3, Santiago SOLIVERES3, Pablo GARCÍA-PALACIOS3
1. Direction Régionale des Eaux et Forêts et de la Lutte Contre la Désertification du Rif,
Avenue Mohamed V, BP 722, 93000, Tétouan, Maroc.
2. Área de Biodiversidad y Conservación, Departamento de Biología y Geología,
Física y Química Inorgánica, Escuela Superior de Ciencias Experimentales y Tecnología,
Universidad Rey Juan Carlos, Calle Tulipán Sin Número, 28933 Móstoles, España.
3. Área de Biodiversidad y Conservación, Departamento de Biología y Geología,
Escuela Superior de Ciencias Experimentales y Tecnología,
Universidad Rey Juan Carlos, Calle Tulipán Sin Número,
28933 Móstoles, España.
Auteur correspondant : [email protected]
Received: 14 October, 2014; First decision: 9 January, 2015; Revised: 25 February, 2015;
Second decision: 6 July, 2015; Revised: 13 July, 2015; Accepted: 1 September, 2015
Résumé
Le suivi des processus écologiques est devenu un
outil nécessaire dans les efforts de lutte contre
la désertification. Les techniques de suivi disponibles sont généralement basées sur les indicateurs du fonctionnement de l’écosystème. Parmi
ces techniques, la méthode de l’analyse fonctionnelle du paysage, connue sous le nom LFA
(Landscape Function Analysis), permet d’évaluer
le fonctionnement géochimique de l’écosystème de manière facile, précise, consistante et
peu chère. Cette méthode n’a jamais été appliquée pour caractériser le fonctionnement des
steppes d’alfa (Stipa tenacissima L.) marocaines.
Dans la présente étude, nous avons appliqué la
méthode LFA pour caractériser le fonctionnement des steppes d’alfa dans douze sites situés
à l’est du Maroc. Nous avons testé la corrélation
entre les indices LFA et les variables du fonctionnement du sol. Les résultats ont montré
que les indices LFA peuvent être utilisés comme
Mots clés : aménagement, alfa, Maroc, méthode
LFA.
équivalents des variables du fonctionnement du
sol comme le C organique, le N total, le pH et
la salinité. Comparées aux steppes espagnoles,
les steppes marocaines ont montré des niveaux
fonctionnels bas justifiant des actions adéquates
de conservation et de restauration écologique.
L’application de la méthode LFA pour les steppes
d’alfa est susceptible d’améliorer nos connaissances sur le fonctionnement de ces écosystèmes
et d’aider les gestionnaires à mieux concevoir les
plans d’aménagement y afférents.
Abstract
The monitoring of ecosystem processes has
become an essential tool for combating desertification. Monitoring techniques are often
based on indicators of ecosystem functioning.
Among them, the Landscape Function Analysis
(LFA) method allows the assessment of ecosystem geochemical functioning by means of an
Keywords: Management, alfa, Morocco, LFA
method.
61
Mchich Derak, Fernando T. Maestre, José L. Quero, Victoria Ochoa, Cristina Escolar, Santiago Soliveres, Pablo García-Palacios
easy, precise, consistent and cheap approach.
This method has never been used to assess the
functioning of the Moroccan alfa steppes (Stipa
tenacissima L.). In the present study, we applied
the LFA method to characterize the functioning
of alfa steppes in 12 sites distributed throughout Eastern Morocco. We checked for correlation between the LFA indices and variables of
soil functioning. Our results showed that LFA
indices can be used as surrogates of soil properties related to soil functioning such as organic
C, total N, pH and salinity. Moroccan steppes
showed a low level of functionality compared
to similar steppes in Spain, and require conservation and ecological restoration measures.
The use of the LFA method in alfa steppes can
be used to increase our understanding of their
functioning, and may help managers to design
more efficient management plans.
Abridged English Version
The monitoring of ecological processes has
become a necessity for combating desertification (Herrick et al. 2005 ; Reynolds et al.
2007b). In the last decades, several monitoring techniques based on ecosystem functioning indicators have been developed. Among
them, the Landscape Function Analysis (LFA)
is based on the assessment of soil surface
indicators, and provides useful information
on ecosystem geochemical functioning. The
method generates three numerical indices
(Table 2) related to soil stability, water infiltration and nutrient cycling (Tongway 1995;
Tongway & Hindley 2004). This simple,
cheap and consistent technique has been
widely applied for assessing the functioning
of Spanish alfa (Stipa tenacissima L.) steppes
(Maestre & Cortina 2004; 2006; Cortina et al.
2006; Maestre & Puche 2009; Mayor & Bautista 2012), but it has never been employed
for the analysis of Moroccan alfa steppes. The
main objective of this study is to assess the
functioning of the Moroccan alfa steppes on
the basis of the LFA indices.
Our study was conducted in 12 semiarid sites
in Eastern Morocco (Figures 1, 2; Table 1). We
compared LFA indices values between bare
soil and under alfa tussocks, and checked for
correlation between the three LFA indices and
variables associated to soil functioning (pH,
salinity, organic C, Total N and available P).
The values of LFA indices were lower in bare
soil than under alfa tussocks (Figure 3, Table
4). This confirms the role of the spatial distribution of patches of vegetation in semiarid
62
ecosystems (Maestre et al., 2002). The LFA
indices were correlated to the main soil
functioning variables, except for available P
(Table 5). This concurs with previous studies
(Tongway & Hindley 2003; Ata Rezaei et al.
2006; Maestre & Puche 2009), and suggests
that LFA indices can be used as surrogates of
steppes functioning variables. The functional
state of Moroccan steppes (expressed by LFA
indices) was lower when compared with the
state of Spanish steppes. This confirms previous field observations which had highlighted
the advanced degradation level in Moroccan
steppes (DREFLCD-O 2007).
The LFA method represents a suitable tool for
assessing the functional status of alfa steppes
and thus, for improving their management.
Moroccan steppes showed a low functioning
level and may require urgent intervention to
control erosive processes, restore nutrient
cycling and also on controlling the human
pressure on the landscape.
Introduction
La désertification est l’un des problèmes environnementaux les plus sérieux qui menacent
la persistance des zones arides et leur capacité
à fournir les biens et services écosystémiques
(Évaluation des écosystèmes pour le millénaire : MEA 2005 ; Reynolds et al. 2007a). Le
suivi des processus écologiques est devenu de
plus en plus une composante nécessaire dans
les plans de lutte contre la désertification en
zones arides (Herrick et al. 2005 ; Reynolds et
al. 2007b). Ce type de suivi permet d’évaluer
l’état et l’évolution des processus écosystémiques et de prévenir des possibles risques
de désertification, ce qui permettrait d’adopter
des mesures de conservation et de gestion de
plus en plus efficaces (Fernández et al. 2002).
En outre, le suivi améliore nos connaissances
sur la réponse de l’écosystème et du paysage
aux actions de restauration et contribue par
conséquent à la réduction de l’incertitude
autour de ces actions (Navarro et al. 2009 ;
Cortina et al. 2011).
Dans les dernières décades, plusieurs techniques de suivi basées sur les indicateurs du
fonctionnement de l’écosystème ont été mises
en œuvre. Combinant les caractéristiques du
sol et de la végétation, ces techniques permettent d’évaluer la résilience de l’écosystème face à l’érosion ainsi que sa capacité de
Caractérisation du fonctionnement des steppes d’alfa marocaines par la méthode de l’analyse fonctionnelle du paysage
retenir et de recycler l’eau et les nutriments
(Tongway 1995 ; Tongway & Hindley 2004 ;
Maestre & Puche 2009). Parmi les techniques utilisées, on cite la méthode de l’analyse fonctionnelle du paysage, connue sous
le nom méthode LFA (Landscape Function
Analysis, Tongway 1995 ; Tongway & Hindley 2004). Cette méthode, basée sur des
indicateurs mesurés au terrain, vise à évaluer
le fonctionnement géochimique des écosystèmes à l’échelle du versant. Les écosystèmes
ayant un état fonctionnel élevé ont tendance à
conserver les ressources en sol, eau et nutriments. Par contre, ceux à état fonctionnel bas
ont tendance à perdre les ressources existantes
et à ne retenir qu’une faible fraction des précipitations incidentes (Tongway & Hindley
2004). La méthode LFA présente l’avantage
d’être précise, consistante dans le temps et
applicable à une large gamme d’écosystèmes.
En outre, son application est simple, rapide
et peu chère et elle peut être utilisée par les
techniciens et les gestionnaires du terrain avec
un effort moindre d’entraînement.
La méthode LFA a été conçue et validée pour
des écosystèmes semi-arides australiens (Tongway 1995 ; McR. Holm et al. 2002 ; Bartley
et al. 2006), où elle a été fréquemment utilisée
pour évaluer l’état fonctionnel du paysage et
sa relation avec la structure de l’écosystème
(p. ex. Ludwig et al. 2007 ; Kwock et al.
2010). Grâce à ses multiples caractéristiques
et utilités, son usage s’est étendu à différentes
régions du monde. À titre d’exemple, elle a
été utilisée pour caractériser les pelouses en
Iran (Ata Rezaei et al. 2006), pour étudier les
effets des reboisements en Tunisie (Derbel et
al. 2009 ; Jeddi et al. 2009), pour suivre le
phénomène de désertification en Argentine
(Oliva et al. 2010), et pour évaluer les effets
des plantations d’Atriplex sur les fonctions
du sol et du paysage au Maroc (Zucca et al.
2013), entre autres.
L’application de la méthode LFA aux steppes
de Stipa tenacissima L. a bien évolué durant
ces dernières années en raison de l’étendue
et de l’importance de ces écosystèmes à
l’échelle du bassin méditerranéen et du risque
permanent de dégradation d’origine humaine
auquel ils sont soumis dans toute son aire de
distribution (Cortina et al. 2009 ; Cortina et al.
2012). C’est en Espagne que cette méthode a
été le plus appliquée pour évaluer la relation
entre la structure et la fonction des steppes
d’alfa (Maestre & Cortina 2004 ; 2006), pour
étudier la relation entre la fonctionnalité du
sol et la capacité de restauration de ces steppes
(Cortina et al. 2006), pour caractériser leur
fonctionnement hydrologique à différentes
échelles (Mayor & Bautista 2012), et pour
étudier la relation entre les indices LFA et les
différentes variables du sol qui caractérisent
mieux le fonctionnement de l’écosystème
(Maestre & Puche 2009). Au Maroc, l’un des
pays d’Afrique du Nord où les steppes d’alfa
s’étendent sur de très grandes superficies (Le
Houérou 2001), l’étude du fonctionnement de
ces écosystèmes est souvent réalisée à base de
mesures qualitatives et d’opinions d’experts.
Rares sont les études qui se sont focalisées sur
la caractérisation des steppes d’alfa moyennant des techniques basées sur les indicateurs
du fonctionnement de l’écosystème comme la
méthode LFA.
La présente étude a pour objectifs : (1) d’évaluer l’état fonctionnel des steppes marocaines
à l’échelle du microsite moyennant la méthode
LFA, (2) d’examiner la relation entre les
indices LFA et les variables clés du cycle des
nutriments des steppes d’alfa étudiées (pH,
salinité, C organique, N total, P disponible),
et (3) de proposer un cadre d’application de la
méthode LFA dans les plans d’aménagements
des steppes marocaines.
Méthodes
Zone d’étude
Notre étude a porté sur 12 parcelles expérimentales situées à l’est du Maroc (figures 1
et 2). Le tableau 1 montre les valeurs des
principales caractéristiques topographiques,
climatiques, pédologiques et de végétation
des parcelles échantillonnées. C’est dans cette
région de l’Est que se concentre la majorité
des 3,3 millions d’hectares de steppes marocaines (Benabid & Fennane 1999). L’alfa peut
s’y trouver dans des ambiances semi-arides,
arides ou sahariennes. L’aridité augmente en
direction O-E et N-S. Le sol est généralement
bien drainé et relativement rocheux (Benabid
2000). La principale vocation du terrain est le
parcours suivi par l’agriculture. En général, la
population locale souffre de difficultés économiques et exerce une forte pression sur les
steppes qui se manifeste par le surpâturage, le
labour et l’exploitation anarchique de l’alfa
(Benabid 2000 ; El Rhazi 2003). L’industrie
de l’alfa a connu un essor spectaculaire dans
la période 1925-1975, mais à partir des années
1980, et à cause de la fermeture des usines
63
Figure 1 - L ocalisation des parcelles échantillonnées (étoiles) à l’est du Maroc. Les limites intérieures correspondent aux limites des
régions marocaines. Les principales caractéristiques des 12 parcelles sont consignées dans le tableau 1.
Figure 1 - Location of the sampling plots (stars) in eastern Morocco. Internal limits correspond to Moroccan regions’ limits.
Main characteristics of the 12 plots are reported in table 1.
Tableau 1 – V
aleurs des principales caractéristiques des 12 parcelles échantillonnées. LAT = latitude (en décimal) ;
LONG = longitude (en décimal) ; ALT = altitude (m) ; PENT = pente (°) ; TMA = température moyenne annuelle (°C) ;
PMA = précipitation moyenne annuelle (mm) ; TSA = teneur en sable (%) ; TAR = teneur en argile (%) ; TLI = teneur en
limons (%) ; CON = conductivité électrique (µS/cm) ; RS = richesse spécifique (N) ; CT = couvert total (%) ; pH = pH du sol ;
COR = C organique (%) ; NTO = N total (%) ; PDI = P disponible (mg P g-1 sol).
Table 1 – Main characteristic values of the 12 sampled plots. LAT = latitude (in decimal); LONG = longitude (in decimal);
ALT = elevation (m); PENT = slope (°); TMA = annual mean temperature (°C); PMA = annual mean precipitation (mm);
TSA = Sand content (%); TAR = clay content (%); TLI = silt content (%); CON = electrical conductivity (µS/cm); RS = specific
richness (N); CT = total cover (%); pH = soil pH; COR = organic C (%); NTO = total N (%); PDI = available P (mg P g-1 soil).
SITE
LAT
LONG
1
33.977
– 3.374
2
34.633
– 3.414
3
34.626
4
33.872
5
PENT
TMA
PMA
TSA
TAR
TLI
776
18
16
265
57
8
36
756
20
16
339
56
7
38
– 3.465
922
17
15
385
43
16
41
– 3.634
1 001
7
15
307
66
4
29
33.933
– 3.559
726
11
16
289
73
5
6
34.442
– 3.593
1 028
14
15
399
54
7
34.473
– 3.639
867
4
16
401
8
34.431
– 2.193
946
6
15
321
9
34.310
– 1.999
1 144
7
14
10
34.159
– 2.373
1 002
5
11
33.052
– 2.423
1 339
12
33.068
– 2.729
1 427
64
ALT
CON
RS
CT
pH
COR
NTO
PDI
99
7
17
8,3
0,7
0,088
0,005
114
16
28
8,5
1,4
0,156
0,005
108
11
11
8,3
2,1
0,182
0,005
99
10
33
8,4
1,8
0,157
0,005
22
67
9
26
8,5
0,6
0,068
0,004
5
41
120
14
46
8,3
1,6
0,167
0,006
43
5
52
121
7
37
8,4
0,8
0,113
0,003
53
6
41
90
11
37
8,3
2,2
0,207
0,007
377
52
5
44
127
18
19
8,4
2,1
0,269
0,010
15
294
59
4
37
86
11
23
8,3
2,1
0,216
0,006
2
15
283
82
4
15
75
7
28
8,5
0,5
0,043
0,008
6
15
310
68
3
29
90
13
18
8,4
0,7
0,071
0,005
A
B
Figure 2 - Exemples des steppes d’alfa échantillonnés dans le site 4 (A) et le site 11 (B) à l’est du Maroc.
Figure 2 – Examples of alfa steppes sampled in site 4 (A) and site 11 (B) in eastern Morocco.
locales destinées à la confection de la fibre
d’alfa et à l’arrêt de leurs exportations, l’alfa
a perdu son rôle socio-économique localement. Actuellement, son exploitation se limite
à l’usage quotidien comme combustible ou
complément de fourrage, et à l’usage artisanal pour la confection des cordes, paniers,
et autres ustensiles (Mhirit & Benchekroun
2006 ; El Rhazi 2003). Au cours des dernières
années, le Haut-commissariat aux eaux et
forêts et à la lutte contre la désertification
marocaine a accordé un intérêt particulier à ces
écosystèmes à travers l’adoption d’une série
de mesures : réalisation d’un inventaire des
ressources alfatières de la région de l’est du
Maroc, élaboration des plans d’aménagement
et des steppes d’alfa pour la même région, et
appui aux coopératives locales pour la restauration de l’usage traditionnel de l’alfa, entre
autres (DREFLCD-O 2007).
Estimation des indices
du fonctionnement du paysage
Les travaux de terrain ont été réalisés dans
des parcelles de 30 × 30 m choisies de telle
manière à couvrir une large gamme de conditions climatiques tout en réduisant la variabilité inter-site en termes de type de végétation,
topographie et type du sol. Toutes les parcelles
ont été établies dans des steppes d’alfa à
65
Tableau 2 – D
escription et interprétation des caractéristiques de la superficie du sol utilisées pour le calcul des indices LFA.
La valeur d’un indice LFA, exprimée en %, correspond à la somme des notes obtenues pour les différentes caractéristiques
impliquées dans le calcul de cet indice (troisième colonne), divisée par la note maximale pouvant être attribuée à cet
indice, soient 40, 57 et 43 respectivement pour les indices de stabilité, infiltration et recyclage des nutriments.
Adaptation à partir de Tongway & Hindley (2004).
Table 2 – Description and interpretation of soil surface characteristics used in the calculation of LFA indices. The value of an index,
expressed in %, corresponds to the sum of the obtained marks for the different characteristics involved in the calculation
of this index (third column), divided by the maximum mark assignable to this index, being 40, 57 and 43 for the stability,
infiltration and nutrient cycling indices, respectively. Adaptation from Tongway & Hindley (2004).
Caractéristique de la
superficie du sol
Description et interprétation
Note
maximale
Couverture du sol
Pourcentage de la couverture projetée de la végétation pérenne à une
hauteur de 0,5 m, en addition aux roches et matériels boisés de plus
de 2 et 1 cm de diamètre respectivement, ou tout autre objet longévive
et immobile. Cet indicateur estime la vulnérabilité à la formation de la
croûte physique.
Couvert basal des herbacées pérennes et/ou couvert de la canopée des
arbres et matorral. Cet indicateur permet d’évaluer la contribution de
biomasse racinaire au processus du recyclage des nutriments.
Couvert basal des
herbacées pérennes et
couvert de la canopée
du matorral
Couvert de litière et degré Quantité, origine et degré de décomposition de la litière. Estime la
de décomposition
disponibilité de la matière organique superficielle à la décomposition
et au recyclage des nutriments.
Couvert de la croûte
Couvert des cryptogames (mousses, lichens et cyanobactéries) visibles
biologique
à la surface du sol. C’est un indicateur de la stabilité superficielle, de la
résistance à l’érosion et de la disponibilité des nutriments.
Degré de brisement
Estime le degré de brisement de la croûte superficielle et la génération
de la croûte
du matériel édaphique vulnérable à l’érosion.
Type et sévérité de l’érosion Évalue la nature et la sévérité du processus érosif actuel.
Indice LFA concerné
5
Stabilité
4
Infiltration
Recyclage des nutriments
30
Stabilité
Infiltration
Stabilité
4
4
Stabilité
4
Stabilité
Matériel déposé
Évalue la nature et la quantité des dépôts alluviaux.
4
Stabilité
Rugosité de la surface
du sol
Évalue la rugosité de la surface du sol en fonction de sa capacité
à capturer et à retenir les ressources mobiles telles que l’eau, les
propagules, les sédiments et la matière organique.
5
Infiltration
Résistance de la surface
aux perturbations
Test d’humectation
Mesure la facilité avec laquelle le sol peut être mécaniquement altéré
pour générer un matériel vulnérable à l’érosion hydrique ou éolienne.
Évalue la stabilité des fragments naturels du sol à l’humectation rapide.
5
4
Stabilité
Infiltration
Stabilité, Infiltration
Texture du sol
Classifie la texture de la surface du sol.
4
Infiltration
orientation SE-SO et en versant de pente inférieure à 30°. Pour chaque parcelle, la méthode
LFA a été appliquée moyennant dix points
quadrats de 50 × 50 cm distribués au hasard,
cinq au sol nu et cinq sous les touffes d’alfa.
Pour chaque point quadrat, nous avons évalué onze variables de la superficie du sol qui
définissent son fonctionnement (tableau 2).
L’évaluation a été réalisée sur une échelle
semi-quantitative en suivant les indications
de Tongway & Hindley (2004). En utilisant
le modèle Excel développé par David Tongway (http://www.csiro.au/services/EcosystemFunctionAnalysis), les onze variables ont
été combinées pour obtenir les trois indices
LFA : stabilité, infiltration et recyclage des
nutriments. L’indice de stabilité est lié à la
résistance du sol à l’érosion et à sa capacité
de récupération après perturbation. L’indice
d’infiltration renseigne sur la faculté du sol
à répartir l’eau de pluie en eau disponible
pour les plantes et en eau de ruissellement
66
perdue par le système. L’indice de recyclage
des nutriments estime l’efficience de recycler
la matière organique et de la retourner au sol
en nutriments assimilables par les plantes.
Les trois indices se présentent sous forme de
pourcentage et leur valeur est inversement
proportionnelle à l’état de détérioration d’une
fonction de l’écosystème (Tongway & Hindley 2004).
Estimation des variables associées
au fonctionnement du sol
Pour caractériser le fonctionnement du sol,
nous avons utilisé cinq variables : pH, salinité (estimée par la conductivité électrique),
C organique, N total et P disponible. Ces
variables sont fortement corrélées avec
les processus de recyclage des nutriments
(Forster 1995 ; Chapin III et al. 2002) et
déterminent en grande mesure le fonctionnement des écosystèmes arides et semi-arides
(Whitford 2002). Ces variables ont été évaluées dans chacun des points quadrats de
50 × 50 cm utilisés pour l’estimation des
indices LFA. Ainsi, pour chaque point quadrat, nous avons extrait approximativement
250 g du sol (0-7,5 cm de profondeur) en utilisant un tube cylindrique. Une fois extrait, le
sol a été tamisé puis laissé à l’air libre pour
sécher. Les analyses de C organique et P disponible ont été réalisées dans le laboratoire du
département de biologie et géologie de l’université Rey Juan Carlos (Móstoles, Espagne),
et celles du N total dans l’université de Jaén
(Jaén, Espagne). Le C organique a été déterminé par colorimétrie après oxydation avec
le mélange du bichromate de potassium et
de l’acide sulfurique (Anderson 1993). Le
N total a été obtenu à l’aide d’un analyseur
CN (Leco CHN628 Series, Leco Corporation,
St Joseph, MI, USA) et le contenu en P disponible a été déterminé selon une extraction
0,5 M NaHCO3 (pH : 8,5) (Olsen & Sommers
1982). Les valeurs des cinq variables obtenues
pour chaque parcelle sont consignées dans le
tableau 1.
Tableau 3 – Statistiques descriptives des trois indices de LFA calculés à l’échelle
des microsites.
Table 3 – Descriptive statistics for the three LFA indices calculated at the
microsite scale.
Indice LFA (%)
Moyenne ± ES
(n = 12)
Min.
Max.
Stabilité
63 ± 1,0
40
78
Infiltration
37 ± 1,5
12
60
Recyclage nutriments
29 ± 1,6
9
52
Analyse statistique
Pour évaluer la signification statistique des
différences entre les valeurs des indices LFA
obtenus pour le sol nu et sous les touffes
d’alfa, nous avons procédé à une analyse
de variance (ANOVA) à deux facteurs, en
considérant la parcelle comme facteur aléatoire et le type du microsite (sol nu et sous
touffes d’alfa) comme facteur fixe. Les cinq
variables associées au fonctionnement du sol
ont montré une distribution non normale (Test
de Kolmogorov-Smirnov, p < 0,05), et par
conséquent leur corrélation avec les indices
LFA a été mesurée à l’aide du coefficient de
corrélation de Spearman.
Résultats
Les valeurs moyennes des indices LFA calculées à l’échelle du microsite ont été de 63,
37 et 29 respectivement pour les indices de
stabilité, infiltration, et recyclage de nutriment (tableau 3). En comparant les valeurs
des indices LFA obtenues pour les deux types
de microsites, on remarque que les sols nus
ont montré des valeurs significativement inférieures à celles en dessous des touffes d’alfa
(figure 3). La différence était bien saillante
Figure 3 - Représentation des trois indices LFA en % obtenus à l’échelle des
microsites.
Figure 3 – Representation of the three microsite LFA indices in %.
67
Tableau 4 - R
ésultats de l’ANOVA à deux critères pour les trois indices de LFA
comparés entre sol nu et sous les touffes d’Alfa.
Table 4 – Two ways ANOVA results for the three LFA indices compared
between bare-ground areas and Alfa tussocks.
Indice LFA (%)
Stabilité
Infiltration
Facteur
Microsite
F
p-value
424,278
< 0,0001
Parcelle
5,926
0,003
Microsite* Parcelle
3,041
0,002
315,883
< 0,0001
0,604
0,792
Microsite* Parcelle
13,309
< 0,0001
Microsite
554,479
< 0,0001
Parcelle
0,300
0,971
Microsite* Parcelle
13,274
< 0,0001
Microsite
Parcelle
Recyclage nutriments
Statistiques
Tableau 5 – C
oefficients de corrélation de Spearman à l’échelle de microsites
entre les trois indices LFA et variables du sol évaluées.
Les valeurs de p et de n sont montrées au-dessous de chaque
coefficient. Les corrélations significatives sont montrées en
caractère gras. pH = pH du sol ; Corg = carbone organique (%) ; Ntot
= nitrogène total (%) et Pdis = phosphate disponible (mg P g-1 sol).
Table 5 – Spearman correlation coefficients obtained at the microsite scale
between the three LFA indices and the assessed soil variables.
The p values and n are showed below each coefficient. Significant
correlations are highlighted with bold character. pH = soil pH; Corg
= organic carbon (%); Ntot = total nitrogen (%) et Pdis = available
phosphate (mg P g-1 soil).
Indices LFA
Stabilité
Infiltration
Recyclage
nutriment
– 0,252
0,005
120
– 0,261
0,004
120
– 0,258
0,004
120
Cond
0,271
0,003
118
0,328
< 0,001
118
0,361
< 0,001
118
Corg
0,265
0,003
120
0,350
< 0,001
120
0,326
< 0,001
120
Ntot
0,201
0,028
120
0,216
0,018
120
0,250
0,006
120
Pdis
– 0,097
0,290
120
0,063
0,492
120
– 0,018
0,846
120
Variables du sol
pH
surtout pour les indices d’infiltration et du
recyclage des nutriments. Cette différence
s’est montrée significative en comparant
tous les microsites étudiés entre eux (facteur
Microsite) et aussi en comparant les microsites à l’intérieur de chaque parcelle (facteur
Microsite*Parcelle) (tableau 4).
68
Les trois indices LFA ont montré une certaine
relation avec les variables du fonctionnement
du sol. Ils se sont montrés négativement corrélés avec le pH, positivement corrélés avec
la salinité, le C organique et le N total, et sans
corrélation avec le P disponible (tableau 5).
En analysant ces relations pour chacun des
deux types de microsites, nous remarquons
que les corrélations ont changé de force et de
direction, en devenant non significatives pour
le pH, faibles pour la salinité, le C organique
et le N total et relativement significatives avec
le P disponible (tableau 6).
Discussion
Les sols nus ont montré des niveaux fonctionnels inférieurs à ceux en dessous des touffes
d’alfa. Cette différence était significative en
comparant les deux types de microsites dans
des conditions topographiques, lithologiques
et de végétation hétérogènes (toutes les parcelles confondues) et aussi en effectuant la
même comparaison pour des conditions
homogènes (à l’intérieur de chaque parcelle). Ce résultat concorde avec les résultats d’autres études menées sur les steppes
d’alfa en Espagne (Maestre & Puche 2009 ;
Mayor & Bautista 2012) et en Tunisie (Derbel
et al. 2009). Cette différence est due essentiellement au fait que le sol nu agit comme
source des ressources en eaux et sédiments,
et les touffes d’alfa comme collecteur de ces
ressources (Puigdefábregas et al. 1999). Ceci
confirme l’importance de la distribution spatiale en tâches de la végétation dans les écosystèmes semi-arides (Maestre et al. 2002).
Les trois indices LFA se sont montrés corrélés aux variables du fonctionnement du sol
étudiées, sauf au P disponible. Ceci confirme
les résultats des études réalisées dans des
milieux semi-arides de l’Australie (Tongway
& Hindley 2003), de l’Iran (Ata Rezaei et
al. 2006) et de l’Espagne (Maestre & Puche
2009), et montre que les indices LFA peuvent
être utilisés comme équivalents aux variables
estimatives directes du fonctionnement des
steppes d’alfa des zones semi-arides. Ces
indices renseignent sur les principaux processus fonctionnels de l’écosystème en relation
avec le recyclage des nutriments, la fertilité
du sol, l’activité microbienne, etc. Cependant, la capacité des indices LFA à caractériser le fonctionnement des steppes est plus
évidente quand on considère une gamme
Tableau 6 – C
oefficients de corrélation de Spearman à l’échelle du sol nu et sous les touffes d’alfa entre
les trois indices LFA et les variables du sol évaluées. Les valeurs de p et de n sont montrées audessous de chaque coefficient. Les corrélations significatives sont montrées en caractère gras.
Pour les abréviations, voir tableau 5.
Table 6 – Spearman correlation coefficients obtained at the bare soil and under alfa between the three LFA
indices and the assessed soil variables. The p values and n are showed below each coefficient.
Significant correlations are highlighted with bold character. For abbreviations, see table 5.
Variables
du sol
Sol nu
Sous alfa
Indices LFA
Indices LFA
Stabilité
Infiltration
Recyclage
nutriment
Stabilité
Infiltration
Recyclage
nutriment
pH
– 0,092
0,484
60
– 0,009
0,948
60
– 0,090
0,496
60
– 0,108
0,412
60
– 0,088
0,501
60
0,011
0,933
60
Cond
– 0,043
0,746
59
– 0,194
0,141
59
0,003
0,984
59
0,064
0,628
59
0,352
< 0,006
59
0,259
< 0,047
59
Corg
– 0,072
0,583
60
– 0,067
0,610
60
0,025
0,852
60
0,040
0,764
60
0,292
0,024
60
0,104
0,429
60
Ntot
0,017
0,898
60
– 0,149
0,256
60
0,089
0,501
60
0,165
0,209
60
0,288
0,026
60
0,162
0,217
60
Pdis
– 0,336
0,009
60
0,262
0,043
60
0,099
0,453
60
– 0,064
0,627
60
– 0,118
0,369
60
– 0,383
0,003
60
variée et hétérogène des microsites. En effet,
quand les microsites des sols nus et des touffes
d’alfa sont considérés séparément, on obtient
certains changements dans la signification
des corrélations entre les indices LFA et les
variables du fonctionnement du sol. Des
changements similaires ont été observés par
Mayor (2008) et Maestre & Puche (2009), ce
qui montre que les indices LFA permettent
de détecter des variations fonctionnelles entre
des microsites de différents types et microconditions mais pas autant à l’intérieur de
chaque type de microsites.
La méthode LFA a été amplement utilisée et
validée pour caractériser le fonctionnement
des steppes d’alfa en Espagne. En considérant
comme références les valeurs des indices LFA
obtenues pour les steppes espagnoles, nous
pouvons approcher l’état du fonctionnement
des steppes marocaines étudiées. De cette
manière, en comparant les valeurs moyennes
des indices de stabilité, infiltration et recyclage des nutriments obtenues à l’échelle du
microsite pour les steppes marocaines (63,
37 et 29, respectivement, tableau 3) et celles
obtenues par Maestre & Puche (2009) pour
les steppes espagnoles (70, 44 et 35, respectivement), nous pouvons considérer que les
steppes d’alfa marocaines se caractérisent par
un état fonctionnel bas. Ceci est probablement
dû aux conditions climatiques qui sont plus
sévères au Maroc qu’en Espagne (tableau 7)
et qui constituent un facteur limitant au fonctionnement des steppes marocaines (Le Houérou 1992, 1995 ; El Rhazi, 2003) et aussi à la
grande pression humaine à laquelle ces dernières sont soumises continuellement. L’état
fonctionnel bas constaté pour les steppes
marocaines confirme les observations faites
sur le terrain par les chercheurs et gestionnaires marocains qui font état d’un état de
Tableau 7 – Moyennes ± ES des principales caractéristiques climatiques des
12 sites d’alfa marocaines échantillonnés dans cette étude et les
steppes d’alfa espagnoles étudiées par Maestre & Puche (2009).
Table 7 – A
verages ± SE of main climatic characteristics of the 12 Moroccan
alfa sites sampled in this study and of Spanish alfa steppes studied
by Maestre & Puche (2009).
Pays
TMA
PMA
IA
Maroc
15,1 ± 0,2
330,8 ± 13,9
0,26 ± 0,01
Espagne
14,11 ± 0,2
418,4 ± 5,4
0,36 ± 0,01
TMA = température moyenne annuelle (°C) ; PMA = précipitation moyenne annuelle
(mm) ; IA : indice d’aridité.
TMA = annual mean temperature (°C); PMA = annual mean precipitation (mm);
IA: aridity index.
69
dégradation avancé de ces steppes justifiant
des mesures urgentes pour leur conservation
et restauration écologique (DREFLCD-O
2007).
En raison de l’intérêt que présente la méthode
LFA, nous pensons que son application peut
être de grande utilité dans les plans d’aménagement et de gestion des steppes d’alfa marocaines. Dans ces plans, la tendance générale
est d’établir des parcelles d’échantillonnage
où l’on collecte des données quantitatives en
relation avec les caractéristiques de la végétation comme la couverture, la biomasse et l’état
de régénération, tout en mesurant certains
paramètres des touffes comme la hauteur, le
diamètre, la densité et le poids. Dans ce sens,
nous recommandons de compléter ces données par l’estimation des indices LFA dans les
mêmes parcelles afin de mieux comprendre
l’état du fonctionnement des steppes et d’obtenir une information pertinente pour suivre
les processus de dégradation de l’écosystème
avec le temps. Cette information peut être très
utile pour définir des mesures de gestion et de
restauration encore plus précises et justifiées
et pour établir l’ordre de priorité aux zones
objet d’intervention.
Dans de futures études, et afin de mieux comprendre le fonctionnement des steppes marocaines, nous suggérons d’établir un nombre
suffisamment grand de parcelles d’échantillonnage reflétant la grande variabilité des
conditions écologiques et socio-économiques
existantes. En outre, nous suggérons que, dans
les mêmes parcelles d’échantillonnage, les
mesures portent simultanément sur les facteurs abiotiques, la structure et la composition de la végétation, le fonctionnement de
l’écosystème (à travers les indices LFA), et
la pression humaine, et ce dans un contexte
de changements globaux qui pèsent sur les
écosystèmes terrestres.
Conclusion
La méthode LFA s’est montrée comme un
outil adéquat pour caractériser le fonctionnement des steppes d’alfa en milieux arides.
L’intégration de cette technique dans les
plans d’aménagement des steppes d’alfa
marocaines est susceptible d’améliorer leur
mode de gestion. En comparaison avec les
steppes espagnoles, les steppes marocaines
se caractérisent par un état fonctionnel plus
70
bas, et nécessitent des interventions urgentes
qui doivent être focalisées sur la stabilité du
sol, les processus érosifs, le recyclage des
nutriments et sur le contrôle des pressions
humaines sur le milieu naturel.
Remerciements
Cette recherche a été financée par le Conseil
Européen de la Recherche à travers le Septième Programme-Cadre de la Communauté
Européenne (FP7/2007-2013)/ERC Grant
agreement 242658 (BIOCOM). Nous adressons nos vifs remerciements au Haut-commissariat aux eaux et forêts et la lutte contre la
désertification du Maroc et à ses Directions
Régionales du nord-est à Taza et de l’est à
Oujda pour leur inestimable appui logistique
et technique, et à MM. Mokaram Jabir, Kanafi
Aziz, Mouflih Boujemàa et Cortina Jordi pour
leur précieuse collaboration dans cette étude.
Références bibliographiques
Anderson J.M. & Ingramm J.S.I., 1993. Tropical Soil
Biology and Fertility: A Handbook of Methods.
Wallingford, UK: CABI.
Ata Rezaei S., Arzani H. & Tongway D., 2006. Assessing rangeland capability in Iran using landscape
function indices based on soil surface attributes. J.
Arid. Environ. 65: 460-473.
Bartley R., Roth C.H., Ludwig J.A., McJannet D., Liedloff A., Corfield J., Hawdon A. & Abbott B., 2006.
Runoff and erosion from Australia’s tropical semiarid rangelands: influence of ground cover for differing space and time scales. Hidrol. Process. 20:
3317-3333.
Benabid A. & Fennane M., 1999. Principales formations
forestières au Maroc. In: Le Grand Livre de la forêt
marocaine, Mardaga, Sprimont.
Benabid A., 2000. Flore et écosystèmes du Maroc :
Évaluation et préservation de la biodiversité. Ibis
Press, Paris.
Chapin III F.S., Matson P. & Money H.A., 2002. Principles of Terrestrial Ecosystem Ecology. Springer,
New York.
Cortina J., Maestre F.T., Vallejo R., Baeza M.J., Valdecantos A. & Pérez-Devesa M., 2006. Ecosystem
structure, function, and restoration success: are they
related?, J. Nat. Conserv. 14: 152-160.
Cortina J., Maestre F.T. & Ramírez D., 2009. Innovations in semiarid restoration. The case of Stipa tenacissima L. grass steppes. In: Bautista S, Aronson
J, Vallejo R (eds.), Land Restoration to Combat
Desertification: Innovative Approaches, Quality
Control and Project Evaluation. Fundación CEAM,
Valencia.
Cortina J., Amat B., Castillo V., Fuentes D., Maestre
F.T., Padilla F. & Rojo L., 2011. The restoration of
vegetation cover in the Iberian southeast. J. Arid.
Environ. 75: 1377-1384.
Cortina J., Ruiz-Mirazo J., Amat B., Amghar F., Bautista
S., Chirino E., Derak M., Fuentes D., Maestre F.T.,
Valdecantos A. & Vilagrosa A., 2012. Bases para la
restauración ecológica de espartales. Gland, Suiza
y Málaga, España, UICN.
Derbel S., Cortina J. & Chaieb M., 2009. Acacia saligna
plantation impact on soil surface properties and vascular plant species composition in central Tunisia.
Arid Land Res. Manag. 23: 28-46.
DREFLCD-O : Direction régionale des eaux et forêts
et de la lutte contre la désertification de l’Oriental,
2007. Étude de la nappe alfatière des communes
rurales de Maatarka, Tendrara et Béni Guil. Renseignements Généraux. Rapport interne, 35 p.
El Rhazi M., 2003. Phytoécologie, production et valorisation de l’alfa (Stipa tenacissima) de la région de
l’Oriental. Thèse de doctorat, université d’Oujda.
Fernández R.J., Archer E.R.M., Ash A.J., Dowlatabadi
H., Hiernaux P.H.Y., Reynolds J.F., Vogel C.H.,
Walker B.H. & Wiegand T., 2002. Degradation and
recovery in socio-ecological systems: a view from
the household/farm level. In: Reynolds J.F., Stafford
Smith M. (eds.), Do Humans Create Deserts? The
Ecological, Meteorological, and Human Dimensions of Global Desertification. Dahlem University
Press, Berlin.
Forster J.C., 1995. Soil physical analysis. In: Alef K,
Nannipieri P (eds.), Methods in Applied Soil Microbiology and Biochemistry. Academic Press Inc, San
Diego.
Herrick J.E., Van Zee J.W., Havstad K.M. & Whitford
W.G., 2005. Monitoring Manual for Grassland,
Shrubland, and Savanna Ecosystems. Volume II:
Design Supplementary Methods and Interpretation.
USDA-ARS, Las Cruces.
Jeddi K., Cortina J. & Chaieb M., 2009. Acacia salicina, Pinus halepensis and Eucalyptus occidentalis improve soil surface conditions in arid southern
Tunisia. J. Arid. Environ. 73: 1005-1013.
Kwok A.B.C., Eldridge D.J. & Oliver I.A.N., 2010. Do
landscape health indices reflect arthropod biodiversity status in the eucalypt woodlands of eastern
Australia ? Aust. Ecol. 36: 800-813.
Le Houérou H.N., 1992. Relations entre la variabilité
des précipitations et celles des productions primaire et secondaire en zone arides. Dans Le Floc’h
E, Gouzis A, Cornet A, Bille JC, (eds.), L’aridité,
une contrainte au développement. ORSTOM, Paris.
Le Houérou H.N., 1995. Bioclimatologie et Biogéographie des steppes arides du Nord de l’Afrique.
Diversité biologique, développement durable et
désertisation. Dans Le Houérou HN (ed.), Bioclimatologie et biogéographie des steppes arides du Nord
de l’Afrique : diversité biologique, développement
durable et désertisation. CIHEAM, Montpellier.
Le Houérou H.N., 2001. Biogeography of the arid steppe
land north of the Sahara. J. Arid. Environ. 48: 103128.
Ludwig J.A., Bastin G.N. & Wallace J.F., Mcvicar TR,
2007. Assessing landscape health by scaling with
remote sensing : when is it not enough ? Landscape
Ecol. 22: 163-169.
Maestre F.T., Huesca M.T., Zaady E., Bautista S. & Cortina J., 2002. Infiltration, penetration resistance and
microphytic crust composition in contrasted microsites within a Mediterranean semi-arid steppe. Soil
Biol. Biochem. 34: 895-898.
Maestre F.T. & Cortina J., 2004. Insights into ecosystem
composition and function in a sequence of degraded
semiarid steppes. Restor. Ecol. 12: 494-502.
Maestre F.T. & Cortina J., 2006. Ecosystem structure
and soil surface conditions drive the variability in
foliar d13C and d15N of Stipa tenacissima in semiarid Mediterranean steppes. Ecol. Res. 21: 44-53.
Maestre F.T. & Puche M.D., 2009. Indices based on
surface indicators predict soil functioning in Mediterranean semi-arid steppes. Appl. Soil. Ecol. 41:
342-350.
Mayor A.G., 2008. El papel de la dinámica fuente-sumidero en la respuesta hidrológica, a varias escalas,
de una zona mediterránea semiárida. Ph.D. Thesis,
University of Alicante.
Mayor A.G. & Bautista S., 2012. Multi-scale evaluation
of soil functional indicators for the assessment of
water and soil retention in Mediterranean semiarid
landscapes. Ecol. Indic. 20: 332-336.
McR. Holm A., Bennet L.T., Loneragan W.A. & Adams
M.A., 2002. Relationships between empirical and
nominal indices of landscape function in the arid
shrubland of Western Australia. J. Arid. Environ.
50: 1-21.
MEA: Millennium Ecosystem Assessment, 2005. Ecosystems and Human Well-Being: Current State and
Trends. Island Press, Washington DC.
Mhirit O. & Benchekroun F., 2006. Les écosystèmes
forestiers et périforestiers : situation, enjeux et perspectives pour 2025. Disponible sur : www.rdh50.
ma/fr/pdf/contributions/GT8-7.pdf.
Navarro R.M., Guzman J.R., Herrera R., Lara P.A.,
Torres M., Ceacero C., Del Campo A. & Bautista
S., 2009. Monitoring Guidelines for the Implementation of Forest Restoration Projects in Mediterranean Regions. In: Bautista S, Aronson J, Vallejo R
(eds.), Land Restoration to Combat Desertification:
Innovative Approaches, Quality Control and Project
Evaluation. Fundación CEAM, Valencia.
Oliva G., Gaitán J., Bran D., Nakamatsu V., Salomone J., Buono G., Escobar J., Frank F., Ferrante
D., Humano G., Ciari G., Suarez D., Opazo W. &
Adema E., 2010. Manual para la instalación y lecturas de Monitores MARAS (Monitoreo Ambiental para Regiones Áridas y Semiáridas). INTA,
Proyecto PNUD GEF 07/G35.
Olsen S.R., Sommers L.E., 1982. Phosphorus. In: Page
A.L., Miller R.H., Keeney D.R. eds. Methods of
Soil Analysis. Part 2. Chemical and Microbiological Properties. Madison, WI, American Society of
Agronomy and Soil Science Society of America.
Puigdefábregas J., Solé-Benet A., Gutiérrez L., Del
Barrio G. & Boer M., 1999. Scales and processes
of water and sediment redistribution in drylands:
results from the Rambla Honda field site in Southeast Spain. Earth-Sci. Rev. 48: 39-70.
Reynolds J.F., Stafford Smith D.M., Lambin E.F., Turner
II B.L., Mortimore M., Batterbury S.P.J., Downing
T.E., Dowlatabadi H., Fernández R.J., Herrick J.E.,
Huber-Sannvald E., Leemans R., Lynam T., Maestre
F.T., Ayarza M. & Walker B., 2007a. Global desertification: building a science for dryland development. Science 316: 847-851.
Reynolds J.F., Maestre F.T., Kemp P.R., Stafford-Smith
D.M. & Lambin E., 2007b. Natural and human
dimensions of land degradation in drylands: causes
and consequences. In: Canadell J., Pataki D., Pitelka
L.F. (eds.), Terrestrial Ecosystems in a Changing
World. Springer-Verlag, Berlin.
71
Tongway D.J., 1995. Monitoring soil productive potential. Environ. Monit. Assess. 37: 303-318.
Tongway D.J. & Hindley N., 2003. Indicators
of ecosystem rehabilitation success. Unpublished report. Available from http://www.cse.csiro.
au/publications/2003/IndicatorsOfMinesiteRehabilitationSuccessStage2-100703.pdf.
Tongway D.J. & Hindley N., 2004. Landscape Function
Analysis: procedures for monitoring and assessing
landscapes. CSIRO Sustainable Ecosystems, Camberra.
72
Whitford W.G., 2002. Ecology of Desert Systems. Academic Press, London.
Zucca C., Pulido-Fernández M, Fava F., Dessena L. &
Mulas M., 2013. Effects of restoration actions on
soil and landscape functions: Atriplex nummularia
L. plantations in Ouled Dlim (Central Morocco).
Soil Til. Res.133: 101-110.
Floristic Diversity Patterns
in the Beni-Haoua Forest (Chlef, Algeria)
Organisation de la diversité floristique
dans la forêt de Beni-Haoua (Chlef, Algérie)
Adda ABABOU1,*, Mohammed CHOUIEB2, Abdelkader BOUTHIBA3,
Djamel SAIDI1, Khalladi MEDERBAL4
1. Department of Biology, Faculty of Sciences, University Hassiba Ben Bouali, Chlef, Algeria.
2. Department of Agronomy, Faculty of Sciences and Engineering, University Abd El Hamid Ibn Badis,
Mostaganem, Algeria.
3. Departement of Hydraulic, Institute of Agronomical Sciences, University Hassiba Ben Bouali,
Chlef, Algeria.
4. Department of Biology, Faculty of natural and life sciences, University Ibn Khaldoun Tiaret, Algeria.
*
Received: 9 January, 2015; First decision: 11 February, 2015; Revised: 28 February, 2015;
Second decision: 7 April, 2015; Revised: 13 April, 2015; Accepted: 8 May, 2015
Abstract
Résumé
The purpose of this study was to provide an
inventory and an analysis of plant species occupying Beni-Haoua, a mountainous coastal ecosystem, with a rarely studied well-developed
forest. As a result, 87 species were recorded in
7 sites, the Jaccard classification resulted in 4
groups of sites with significantly different diversities. According to the ϕ-coefficient of association, which can be used as a measure of fidelity,
among the 87 species, 34 diagnostic species were
distributed over four plant communities, with a
fidelity value ranging from 55 to 100%, 28 differential species, among which 16 species were
common to 2 plant communities and 12 common to 3 plant communities. The redundancy
analysis (RDA) showed that among the studied
environmental variables, altitude and pH were
the most important ones. Indeed, according to
the detrended correspondence analysis (DCA),
plant species occurrence and distribution in the
study area were affected by a strong altitudinal
gradient.
La région de Beni-Haoua est un écosystème littoral, montagneux, avec un patrimoine forestier très développé, qui n’a fait l’objet que de
très rares études. Le principal objectif de cette
étude est l’inventaire des espèces végétales
ainsi que l’identification des principaux groupements végétaux que comporte cet écosystème
par le biais des méthodes statistiques comme le
coefficient de fidélité ϕ et l’analyse canonique
des redondances (RDA). Ainsi, le coefficient de
Jaccard a permis l’identification de 4 groupes
de relevés (sites) avec des diversités floristiques
significativement différentes. Selon, le coefficient de fidélité ϕ, parmi les 87 espèces recensées, 34 espèces sont réparties sur 4 unités de
végétations avec un degré de fidélité allant de
55 % à 100 % et considérées comme espèces
caractéristiques, 28 espèces différentielles parmi
lesquelles 16 espèces communes à 2 unités de
végétation et 12 espèces communes à 3 unités
de végétations. La RDA a montré une très forte
corrélation entre les espèces végétales et les
paramètres environnementaux : l’altitude et le
pH sont les paramètres environnementaux les
plus importants conditionnant la distribution
de la végétation. En effet la DCA a montré que
l’occurrence et la distribution de la végétation
dans la région de Beni-Haoua sont influencées
par un fort gradient altitudinal.
Keywords: Beni-Haoua, RDA, Phi coefficient,
plant communities, diversity.
Mots clés : Beni-Haoua, RDA, coefficient Phi,
communautés végétales, diversité.
73
Adda Ababou, Mohammed Chouieb, Abdelkader Bouthiba, Djamel Saidi, Khalladi Mederbal
Introduction
Plant communities are characterized by a definite floristic composition, presenting physiognomic and structural uniformity (Flahault
& Schroter 1910), with a highly complex
and variable composition throughout ecosystems (Tozer 2003, Buckley et al. 2004,
Sherman et al. 2008). Since, ecosystems are
often described by their plant communities
and environmental conditions (Wisheu &
Keddy 1989, Kingston & Waldren 2003), the
environmental control model is considered
to be the primary factor affecting species
occurrence and communities structure and
composition (Collins et al. 1989, Cornwell
& Ackerly 2009, Flinn et al. 2010). In this
context, understanding the occurrence and
the spatial patterns of species in connection
with environmental conditions, is a major goal
in ecological studies (Legendre 1993). The
achievement of this goal relies largely upon
the right choice of statistical methods, including classification methods, statistical fidelity
measures, modelling methods (McCullagh &
Nelder 1989) and multivariate analyses dedicated to the analysis of data sets with more
than one variable. Among a large set of multivariate techniques, constrained ordination
analysis has been widely used by ecologists to
predict vegetation composition and response
to environmental conditions (Collins et al.
1989, Zhu et al. 2005, Ababou et al. 2009,
2010), and as mentioned by Ramette (2007)
plant studies rank first with insect studies for
their use of constrained ordination.
The aim of this study was to investigate the
area of Beni-Haoua, a coastal ecosystem with
a well-developed forest, rarely surveyed for
its flora, in order to: (1) inventory the existing plant species within this ecosystem; (2)
identify for each species the most influential
environmental gradient; (3) identify the plant
communities and species showing high fidelity to each community; (4) investigate the
environmental gradients influencing these
communities.
Material and methods
Study area
Covering approximately 400 km2, the study
area is a mountainous, coastal ecosystem
with a highly-developed forest, located in the
eastern part of Dahra mountains and comprised between 36°19’35”- 36°33’16” N of
latitude and 1°22’50”-1°41’59” E of longitude (Figure 1). The area is characterized by
a very rugged relief, steep slopes and an average height of 500 meters, the highest point
stands over 1,000 m at Bissa Mountain. It is a
Figure 1 – Location of the study area.
74
typical Mediterranean area in terms of landscape structure, composition and climate, distinguished by hot, dry summers and relatively
rainy winters with an annual rainfall ranging
from 400 to 600 mm, an average temperature
of 6 °C during winter and 30 °C during summer. In terms of vegetation, the landscape is
covered with natural sclerophyllous vegetation alternating with bare soils, dominated in
the maritime face by Pinus halepensis and
Pinus pinaster, but above a certain altitude
Quercus suber and Quercus ilex are the most
significant species, whereas, Juniperus oxycedrus and Pistacia lentiscus are less imposing.
Within this area, Bissa is the densest forest,
including Quercus suber, Quercus ilex, Pinus
halepensis, holly and wild fruit trees, such as
Prunus avium, Arbutus unedo and Rubus fruticosus, whereas the mattoral part of this forest is dominated by Calluna vulgaris, Cistus
monspeliensis and Lavandula dentata within
a very dense Pistacia lentiscus cover.
Soil and vegetation sampling
The vegetation inventory was carried out
during spring 2013 (March - May) in 7 sites
(Bissa, Briera, Kadaian, Bouriach, Boucheral,
Souhalia and Souamer) by exploring most of
the area accessible in each site and considering only the presence/absence. A total of
87 species was recorded (Appendix), among
which 25 species common to all sites were
excluded from the analysis; species identification was done based on Quezel & Santa (1962,
1963) and botanical experts. Also, in order to
study the effects of environmental variables
on species composition and occurrence, for
each of the 7 sites, we recorded the altitude
(using a topographical map and a GPS), as
well as the electrical conductivity (EC) (conductivity meter), the percentage of organic
matter (OM) (Walkley and Black method,
1934) and the pH (pH meter) of the soil.
Data analysis
Initially, a co-linearity test was performed
between environmental variables; as a result
the variance inflation factors (Montgomery &
Runger 2006, O’brien 2007) were less than
10 for the 4 selected variables, indicating no
significant correlation. In order to minimize
data-normality problems, the Shapiro-Wilk’s
normality test showed that, with the exception
of altitude, which was log-transformed, the
remaining variables were normally distributed. The most important floristic gradients
were assessed using a Detrended Correspondence Analysis (DCA) (Hill & Gauch 1980)
which is an ordination technique that reveals
independent gradients with maximum species
dispersion (Schmidtlein & Sassin 2004). In
order to relate the inventoried vegetation to the
measured environmental variables we opted
for direct gradient analysis (constrained ordination), where species are directly related to
measured environmental factors. In this context, the two most commonly used constrained
ordination are either the Redundancy Analysis
(RDA) (ter Braak 1994, Zuur et al. 2007) or
Canonical Correspondence Analysis (CCA)
(ter Braak 1986), to that end, DCA showed a
gradient length almost equal to 3 (3.38) suggesting that RDA is more appropriate than
CCA (Jongman et al. 1996). However, RDA
are strongly affected by double zeros (Zuur et
al. 2007), thus, an alternative is to apply the
Hellinger transformation (Rao, 1995); according to Legendre & Gallagher (2001) this
approach is less sensitive to double zeros. In
connection with RDA, a linear regression (Zar
1999, Montgomery 2007) was also applied
to evaluate the most effective environmental
variable on each individual species.
To cluster the plant species listed in the study
area the Jaccard coefficient of similarity
(Magurran 1988) was firstly applied, followed
by the ϕ-coefficient of association. This coefficient is a statistical measure of association
which can be used as a measure of fidelity;
it can be calculated as follows (Bruelheide
2000):
ϕ
Where, N = total number of sites in the data set
Np = number of sites in a particular group of
sites obtained through Jaccard classification;
n = number of occurrences of the species in
the data set; np = number of occurrences of the
species in a particular group of sites.
Finally, we performed a Shannon-Wiener
diversity index (Shannon index) followed by
a t-test to evaluate biodiversity among sites
and an analysis of similarity test (ANOSIM)
(Clarke 1993) to examine the differences
among the communities identified through
ϕ-coefficient.
75
Results
Species assemblage
and environmental factors
Gradient analysis of vegetation
assemblage (DCA)
The first 4 axes of the RDA explained 92.5%
of the cumulative variance of species data and
97.1% of the cumulative variance of speciesenvironment relationship, the Monte Carlo
permutation test showed that all canonical
axes of RDA were significant (p < 0.05),
meaning that species composition was significantly related to the measured variables
(Figure 3). The highest cumulative percentage of variance (68.1%) was explained by
the first 2 canonical axes, the first axis with
40.4% of variance explained was significantly
(p < 0.05) related to pH (r = - 0.82) and altitude (r = 0.81), the variance explained by
each of these 2 variables according to forward
selection was respectively 33.2% and 32%,
the combined variance explained by pH and
altitude was 56.6%. The second axis (27.7%)
was negatively related to diversity (r = - 0.85),
EC (r = - 0.68), OM (r = - 0.61) and positively
related to pH (r = 0.46) (Table 1).
According to the marginal effects (λ1), the
eigenvalues expressed by each variable used
individually in RDA (Table 2) indicated that
the best explanatory variables were pH (0.33)
and altitude (0.32). As indicated by the conditional effects (λA) (Table 2), these 2 variables showed the most significant (p < 0.05)
increases in the total sum of eigenvalues during the forward selection, whereas the contribution of OM and EC was not significant.
The first two ordination axes of the DCA
explained 43.7% variability of vegetation
assemblages. With an explained variance of
38.1%, the first axis symbolizing the longest
gradient (3.38) in species composition, represented the low-high altitudinal gradient, so
species dispersion along this axis reflected an
altitudinal gradient of species assemblage.
The second axis with only 5.6% of explained
variance, reflected the variability caused by
the remaining environmental gradient, mainly,
OM and EC (Figure 2).
ϕ-coefficient of association
Figure 2 – D
etrended correspondence analysis (DCA) ordination diagram of
the plant species in Benin-Haoua. Species names are indicated in
Table 4 and Table 5.
76
Initially, the classification of sites based on
Jaccard’s similarity coefficient and cluster
analysis resulted in 4 different groups (Group
A = Bissa and Souamer; Group B = Briera and
Kadian; Group C = Boucheral and Souahlia;
Group D = Bouriach) (Figure 4), the highest
level of diversity according to Shannon index
were shown by groups A (4.38) and C (4.11)
whereas group D showed the lowest diversity (3.5), when comparing the 4 groups. The
t-test (Table 3) showed significant differences
in diversity between groups A and B, A and
D, B and D and C and D, but no significant
differences between A and C and B and C.
Subsequently, the ϕ-coefficient of association related to each species was calculated in
accordance with the Jaccard classification.
The ϕ-coefficient resulted in 4 vegetation
units A, B, C and D differentiated successively
by 22, 10, 1 and 1 diagnostic species (species
Figure 3 – R
edundancy analysis (RDA) ordination diagram of 62 species, 4 environmental variables
and 7 sites. Species names are indicated in Table 4 and Table 5 column code.
Table 1 – Eigenvalues and percentage of variance explained by RDA, with Pearson correlations (r)
between environmental variables and the 4 canonical axes.
Axes
Axis 1
Axis 2
Axis 3
Axis 4
Eigenvalues
0.384
0.265
0.172
0.104
Species-environment correlations
0.99
0.989
0.991
0.929
Species data
38.4
64.9
82.1
92.5
Species-environment relation
40.4
68.1
86.3
97.1
– 0.823 **
0.455
0.279
– 0.186
EC
0.056
– 0.684 *
– 0.434
0.523
OM
– 0.172
– 0.606 *
0.468
– 0.198
0.814 **
0.238
– 0.528
0.039
0.251
– 0.851 **
0.194
– 0.408
F
P
3.985
0.034
Cumulative percentage variance of:
Environmental variables
pH
Altitude
Shannon index
Monte Carlo test (999 permutations)
Test of significance of all canonical axes
77
Table 2 – M
arginal and conditional effects of each environmental variable
obtained through forward selection in the RDA ordination.
Marginal Effects
Conditional Effects
λ1
λA
P
F
pH
0.33
0.33
0.036
2.49
Altitude
0.32
0.26
0.048
2.44
Shannon index
0.24
0.18
0.202
2.31
EC
0.19
0.14
0.070
3.26
OM
0.16
0.04
0.560
0.86
Variable
Table 3 – V
alue of Shannon index (H) for each group (A, B, C, D) of sites,
with value of t-test (higher part of the matrix) comparing between
groups diversity and significance level (lower part of the matrix)
(*** p < 0.001; ** p < 0.01;
* p < 0.05), with, Group A = Bissa and Souamer; Group B = Briera and
Kadian; Group C = Boucheral and Souahlia; Group D = Bouriach.
(H)
Group A
4.38
Group A
Group B
3.99
Group C
4.11
Group D
3.50
P-value
t-test
Group A
Group B
Group C
Group D
2.66
1.84
4.20
0.74
2.15
Group B
**
Group C
NS
NS
Group D
***
*
occurring in a single vegetation unit), each
with the highest fidelity value to the vegetation unit to which it was assigned (Table 4).
It was also remarkable that many differential
species (species occurring in a few vegetation
units) were shared between the 4 vegetation
units, 6 species between A and B, 5 species
between A and C, 5 species between B and C
and 12 species between A, B and C (Table 5).
According to the results of regression analysis
(Table 4), vegetation unit A showed two major
subgroups of diagnostic species significantly
(r2 = 0.9; p < 0.01) related to environmental
conditions. The first subgroup was negatively
related to pH and the second was positively
related to altitude. Vegetation unit B was
mostly positively (r2 = 0.8; p < 0.05) related to
pH and EC. Vegetation unit C was negatively
related to pH and EC. Vegetation unit D was
positively related to pH and altitude, and both
vegetation unit C and D were not significantly
related to the studied environmental factors.
2.76
**
Figure 4 – D
endrogram showing the classification of 7 reference sites in Beni-Haoua based on binary data according to the Jaccard’s
similarity coefficient.
78
Table 4 – Synoptic table of 34 diagnostic species distributed over 4 vegetation units (A, B, C, D), based
on Jaccard classification and Φ-coefficient of association. In addition the results of linear
regression analysis showing the most influential environmental factors on each species with r2
(determination coefficient) and (P-value) (significance level) (** p < 0.01; * p < 0.05).
Φ-coefficient
Species
Code
Φ
value
Urginea maritima (L.) Baker
U.Mar
100
Rubia peregrina L.
R.Per
Erica cinerea L.
E.Cin
Hedysarum spinosissimum L.
Vegetation
Unit
Linear regression
pH
EC
A
– **
100
A
–
**
100
A
– **
OM
Altitude
r2
P
–
0.9
0.01
–
0.9
0.01
–
0.9
0.01
**
H.Spi
100
A
–
–
0.9
0.01
M.Com
100
A
– **
–
0.9
0.01
Centella asiatica L.
C.Asi
100
A
–
–
0.9
0.01
Quercus ilex L.
Q.Ile
73
A
–
0.6
0.14
Anagallis arvensis L.
A.Arv
73
A
0.5
0.21
Alnus glutinosa (L.) Gaertn.
A.Glu
73
A
–
0.6
0.14
Cistus monspeliensis L.
C.Mon
73
A
–
0.9
0.01
E.Arb
73
A
–
0.6
0.14
–
Muscari comosum (L.) Mill.
Erica arborea L.
–
–
*
–
–
–
–
Linum corymbiferum Desf.
L.Cor
73
A
0.6
0.14
Quercus suber L.
Q.Sub
64.6
A
+
+ **
0.9
0.01
Salvia verbenaca L.
S.Ver
64.6
A
+
+
**
0.9
0.01
Asperula arvensis L.
A.Ars
64.6
A
+
+ **
0.9
0.01
Crataegus laevigata Poir.
C.Lae
64.6
A
+
+
0.9
0.01
Schinus molle L.
S.Mol
64.6
A
0.6
0.15
Anthyllis tetraphylla L.
A.Tet
64.6
A
Bromus madritensis L.
B.Mad
64.6
A
–
Marrubium multibracteatum H&M.
–
–
–
–
**
–
–
**
0.6
0.14
–
0.6
0.15
–
0.6
0.15
+ **
0.9
0.01
*
M.Mul
64.6
A
Moehringia trinervia L.
M.Tri
54.8
A
Quercus coccifera L.
Q.Coc
54.8
A
–
–
0.6
0.17
Artemisia absinthium L.
A.Abs
100
B
+*
+*
0.8
0.05
Althaea officinalis L.
A.Off
100
B
+
+
*
0.8
0.05
Gladiolus byzantinus Mill.
G.Byz
100
B
+*
+*
0.8
0.05
Helichrysum arenarium (L.) Moench
H.Are
100
B
+
+
*
0.8
0.05
Helichrysum stoechas (L.) Moench
H.Sto
100
B
+*
+*
0.8
0.05
Jacobaea maritima (L.) Pel.Mei.
+
+ **
*
*
J.Mar
100
B
+
0.8
0.05
Chamaerops humilis L.
C.Hum
73
B
–*
–*
0.8
0.06
Chamaelirium luteum L.
C.Lut
73
B
–
–
*
0.8
0.06
Daucus carota L.
D.Car
73
B
–*
–*
0.8
0.06
Picris echioides L.
P.Ech
73
B
0.6
0.19
Opuntia ficus–indica (L.) Mill.
O.Ind
73
C
–
0.5
0.21
Lavandula angustifolia Mill.
L.Ang
64.6
D
+
0.3
0.46
*
*
*
+
+
–
+
Analysis of similarity (ANOSIM)
The ANOSIM showed a highly significant difference (p < 0.001) in taxonomic composition
between vegetation unit A and B, significant
differences (p < 0.05) between vegetation unit
A and D, B and D and C and D: as indicated
by the R value close to 1 there was a clear
difference in taxonomic composition between
these vegetation units, whereas no significant
differences were observed between A and C
and B and C (Table 6), for which the R values
almost equal to 0, indicative of an important
overlap.
79
Table 5 – Synoptic table of 28 differential species common to 2 or 3 vegetation units. In addition to the
results of linear regression analysis showing the most influential environmental factors on each
species with r2 (determination coefficient) and (P-value) (significance level)
(** p < 0.01; * p < 0.05).
Φ-coefficient
Species
Code
Φ
value
Lavandula dentata L.
L.Den
54.8
Pallenis spinosa (L.) Casso
Chrysanthemum myconis L.
Ceratonia siliqua L.
Vegetation
Unit
Linear regression
pH
EC
OM
A, B
+*
*
Altitude
r2
P
+
0.80
0.03
P.Spi
54.8
A, B
+
+
0.80
0.03
C.Myc
54.8
A, B
+*
+
0.80
0.03
+
+
0.80
0.03
0.30
0.46
C.Sil
54.8
A, B
G.Rob
40.0
A, B
Quercus robur L.
Q.Rob
25.8
A, B
0.50
0.26
Asplenium adiantum-nigrum L.
A.Adn
54.8
A, C
–*
–*
0.70
0.07
Asplenium ceterach L.
A.Cet
54.8
A, C
–
–
*
0.70
0.07
Astragalus monspessulanus L.
A.Mon
54.8
A, C
–*
–*
0.70
0.07
Thymus vulgaris L.
T.Vul
40.0
A, C
–
–
Cistus albidus L.
C.Alb
25.8
A, C
Capsella bursa-pastoris (L.) Medik
C.Bup
40.0
B, C
Cynoglossum creticum Mill.
C.Cre
40.0
B, C
Melica minuta L.
M.Min
40.0
Raphanus raphanistrum L.
R.Rap
40.0
Geranium robertianum L.
Tetragonolobus biflorus Desr.
*
–
–
+
*
–
0.80
0.05
+
0.73
0.07
+
–
0.80
0.03
+
–
0.80
0.03
B, C
+
–
0.80
0.03
B, C
+
–
0.80
0.03
+
*
*
–*
T.Bif
40.0
B, C
+
0.10
0.83
E.Bon
25.8
A, B, C
–
+
0.40
0.40
Echium plantagineum L.
E.Pla
25.8
A, B, C
–
+
0.40
0.40
Avena sterilis L.
A.Ste
25.8
A, B, C
–
+
0.40
0.40
Arisarum vulgare Targ.Tozz.
A.Vul
25.8
A, B, C
–
+
0.40
0.40
Bellis annua L.
B.Ann
25.8
A, B, C
–
+
0.40
0.40
Borago officinalis L.
B.Off
25.8
A, B, C
–
+
0.40
0.40
Oxalis corniculata L.
O.Cor
25.8
A, B, C
–
+
0.40
0.40
Papaver rhoeas L.
P.Rho
25.8
A, B, C
–
+
0.40
0.40
Scorpiurus muricatus L.
S.Mur
25.8
A, B, C
–
+
0.40
0.40
Taraxacum officinale F.H.
T.Off
25.8
A, B, C
–
+
0.40
0.40
Tanacetum parthenium (L.) Sch.
Bip.
T.Par
25.8
A, B, C
–
+
0.40
0.40
Valeriana tuberosa L.
V.Tub
25.8
A, B, C
–
+
0.40
0.40
Erigeron bonariensis L.
Table 6 – ESimilarity analysis (ANOSIM) between the 4 vegetation units (Veg. Unit) with P-value (higher
part of the matrix) and R-value (lower part of the matrix).
P values (*** p < 0.001; * p < 0.05)
Veg. Unit A
Veg. Unit A
R-Values
80
Veg. Unit B
Veg. Unit C
Veg. Unit D
0.0002 ***
0.169
0.019 *
0.060
0.028 *
Veg. Unit B
0.156
Veg. Unit C
0.034
0.062
Veg. Unit D
0.781
0.978
0.041 *
0.977
Discussion
The goal of this study was to inventory and
examine the plant diversity in relation to the
environmental conditions in the coastal Mediterranean forest of Beni-Haoua. Overall, as
indicated by the DCA, the study area was
characterized by a strong altitudinal gradient.
Indeed, the forward selection in the RDA ordination showed that plant species distributions
were strongly affected by altitude and pH. The
effect of these two variables on plant species
distributions was consistent with RDA ordination results, the highest proportion of variation in species data (38.4%) was explained
by the first canonical axis, correlation analysis
indicated that this axis was highly positively
related to altitude and negatively related to
pH, so this axis was interpreted as altitudinal
gradient axis, on which a change in species
composition from low to high altitude was
observed, the amount of variability in species-environment relation along this axis was
40.4%, suggesting that the influence exerted
on the distribution of vegetation in the study
area by all the remaining unmeasured biotic
and abiotic factors was less important in
comparison to altitude, these findings were
highlighted by several authors in previous
studies (Reeder & Riechert 1975, Gutierrez
et al. 1998, Erschbamer et al. 2006). The
proportion of variation in species data and
species-environment relation explained by
the second canonical axis was less important,
this axis was positively correlated with pH and
negatively correlated to OM, EC and Shannon index, meaning that the diversity in the
study area was influenced by OM in a positive way (R = 0.7), and pH in a negative way
(R = – 0.46), the latter observation was as also
reported by Isermann (2005).
The 87 species recorded as binary data (presence/absence) and used in the Jaccard classification enabled us to identify 4 different
groups of sites, significantly distinct in species richness and diversity according to t-test.
Through the ϕ-coefficient of association, these
4 groups of sites have been successfully differentiated by 4 vegetation communities. The
first vegetation community (vegetation unit A)
encompasses 22 diagnostic species closely
related to high-altitude, this community was
distinguished by three species of the genus
Quercus (Quercus ilex L., Quercus suber L.
and Quercus coccifera L.), one of the most
important woody angiosperms in the area,
positively correlated with altitude according
to the results of this study, also mentioned
by Barbero at al. (1992) & Ducousso et al.
(1996), the second vegetation community
(vegetation unit B) observed at low-altitude,
includes 10 diagnostic species, mostly significantly correlated with pH and EC, the third
and fourth community less rich in diagnostic
species have not shown significant correlations with the studied environmental factors.
The results of regression analysis suggested
in general that altitude and pH were the most
important factors in species occurrence and
community composition in the coastal region
of Beni-Haoua. In this study, ANOSIM test
demonstrated a clear difference in species
composition among the vegetation communities, especially between the community
related to high-altitude (vegetation unit A)
and the community related to low-altitude
(vegetation unit B) (p < 0.001) implying that
changes in flora were connected to the change
of altitude, vegetation is therefore an important ecological indicator of the local topography (Pinder et al. 1997, Qiu et al. 2012).
Conclusion
Through the 7 sites selected for this study, we
inventoried 87 plant species among which 62
species closely related to the selected environmental factors, nevertheless, the statistical
analysis revealed that species occurrence and
community composition were mainly affected
by an altitudinal gradient, so that the composition of plant communities related to high
altitude was highly different than that related
to low altitude. Finally, among very few studies carried out in Beni-Haoua forest our study
improved understanding species distributions
and occurrence in a southern Mediterranean
forest.
References
Ababou A., Chouieb M., Khader M., Mederbal K., Saïdi
D. & Bentayeb Z., 2009. Multivariate analysis of
vegetation in the salted lower-Cheliff plain, Algeria.
B. Soc. Bot. Mex. 85: 59-69.
Ababou A., Chouieb M., Khader M., Mederbal K. &
Saïdi D., 2010. Using vegetation units as salinity
predictors in the lower-Cheliff, Algeria. Turk. J. Bot.
34: 1-10.
Barbero M., Loisel R., & Quezel P., 1992. Biogeography, ecology and history of Mediterranean Quercus
ilex ecosystems. Vegetatio 100: 19-34.
81
Bruelheide H., 2000. A new measure of fidelity and its
application to defining species groups. J. Veg. Sci.
11: 167-178.
Buckley H.L., Burns J.H., Kneitel J.M., Walters E.L.,
Munguia P. & Miller T.E., 2004. Small-scale patterns in community structure of Sarracenia purpurea
inquilines. Community Ecol. 5: 181-188.
Clarke K.R., 1993. Non-parametric multivariate analysis of changes in community structure. Aust. J.
Ecol. 18: 117-143.
Collins S.L., Mitchell G.S. & Klahr S.C., 1989. Vegetation-Environment Relationships in a Rock Outcrop
Community in Southern Oklahoma. Am. Midl. Nat.
122: 339-348.
Cornwell W.K. & Ackerly D.D., 2009. Community
assembly and shifts in plant trait distributions across
an environmental gradient in coastal California.
Ecol. Monogr. 79: 109-126.
Ducousso A., Guyon J.P. & Kremer A., 1996. Latitudinal and altitudinal variation of bud burst in western
populations of sessile oak Quercus petraea (Matt
Liebl). Ann. Sc. Forest. 53: 775-782.
Erschbamer B., Mallaun M. & Unterluggaue P., 2006.
Plant diversity along altitudinal gradients in the
Southern and Central Alps of South Tyrol and Trentino (Italy). Gredleriana 6: 47-68
Flahault C. & Schroter C., 1910. Report on phytogeographical nomenclature. Proceedings of the 3rd
International Botanical Congress, Brussels. 1: 131164.
Flinn K.M., Gouhier T.C., Lechowicz M.J. & Waterway
M.J., 2010. The role of dispersal in shaping plant
community composition of wetlands within an oldgrowth forest. J. Ecol. 98: 1292–1299.
Gutierrez J.R., Lopez-Cortes F. & Marquet P.A,. 1998.
Vegetation in an altitudinal gradient along the Rıo
Loa in the Atacama Desert of northern Chile. J. Arid.
Environ. 40: 383-399.
Hill, M.O. and H.G. Gauch. 1980. Detrended correspondence analysis, an improved ordination technique.
Vegetatio 42: 47-58.
Isermann M., 2005. Soil pH and species diversity in
coastal dunes. Plant Ecol. 178: 111-120.
Jongman R.H.G., Ter Braak C.J.F. & van Tongeren
O.F.R., 1996. Data analysis in community and landscape ecology. Cambridge Univ. Press, Cambridge.
Kingston N. & Waldren S., 2003. The Plant Communities and Environmental Gradients of Pitcairn Island:
The Significance of Invasive Species and the Need
for Conservation Management. Ann. Bot. 92: 31-40.
Legendre P. & Gallagher E., 2001. Ecologically meaningful transformation for ordination of species
data. Oecologia 129: 271-280.
Legendre P., 1993. Spatial autocorrelation: trouble or
new paradigm. Ecology 74: 1659-1673.
Magurran A.E., 1988. Ecological diversity and its measurement. Princeton Univ. Press, Princeton.
McCullagh P. & Nelder J.A., 1989. Generalized Linear
Models. 2nd ed. Chapman and Hall, London.
Montgomery D.C., Peck E.A. & Vining G.G., 2006.
Introduction to Linear Regression Analysis. 4th ed.
Wiley-Interscience, New Jersey.
82
Montgomery D.C. & Runger G.C., 2006. Applied Statistics and Probability for Engineers. 4th ed. John
Wiley & Sons, New Jersey.
O’Brien R.M., 2007. A caution regarding rules of thumb
for variance inflation factors. Qual. Quant. 41: 673690.
Pinder J.E., Kroh G.C., White J.D. & Basham May
A.M., 1997. The Relationships between Vegetation
Type and Topography in Lassen Volcanic National
Park. Plant. Ecol. 131:17-29.
Quezel P. & Santa S. 1962. New flora of Algeria and
southern desert regions. Volume 1. ed. CNRS, Paris.
Quezel P. & Santa S. 1963. New flora of Algeria and
southern desert regions. Volume 2. ed. CNRS, Paris.
Qiu B., Zhong M., Zeng C., Tang Z. & Chen C., 2012.
Effect of topography and accessibility on vegetation
dynamic pattern in Mountain-hill Region. J. Mt. Sci.
9: 879-890.
Ramette A., 2007. Multivariate analyses in microbial
ecology. FEMS Microbiol. Ecol. 62:142-160.
Rao C.R., 1995. A review of canonical coordinates and
an alternative to correspondence analysis using Hellinger distance. Questiio 19: 23-63.
Reeder W.G. & Riechert S.E., 1975. Vegetation change
along an altitudinal gradient, Santa Cruz Island,
Galapagos. Biotropica 7: 162-175.
Schmidtlein S. & Sassin J., 2004. Mapping of continuous floristic gradients in grasslands using hyperspectral imagery. Remote Sens. Environ. 92: 126-138.
Sherman R., Mullen R., Haomin L., Zhendong F. &
Yi W., 2008. Spatial patterns of plant diversity and
communities in Alpine ecosystems of the Hengduan
Mountains, northwest Yunnan, China. J. Plant. Ecol.
1: 117-136.
Ter Braak C.J.F., 1986. Canonical correspondence analysis: a new eigenvector technique for multivariate
direct gradient analysis. Ecology 67: 1167-1179.
Ter Braak C.J.F., 1994. Canonical community ordination. Part I: Basic theory and linear methods. Ecoscience 1: 127-140.
Tozer M., 2003. The native vegetation of the Cumberland Plain, western Sydney: systematic classification and field identification of communities. Cunninghamia 8: 1-75.
Walkley A. & Black I.A., 1934. An examination of Degtjareff method for determining soil organic matter
and a proposed modification of the Chromic acid
titration method. Soil Sci. 37: 29-37.
Wisheu I.C. & Keddy P.A., 1989. Species richnessstanding crop relationship along four lakeshore
gradients: constraints on the general model. Can. J.
Botany 67: 1609-1617.
Zar J.H., 1999. Biostatistical analysis. 4th ed. Prentice
Hall, New Jersey.
Zhu M., Hastie T.J. & Walther G., 2005. Constrained
ordination analysis with flexible response functions.
Ecol. Model. 187: 524-536.
Zuur A.K., Ieno E.N. & Smith G.M., 2007. Analysing
Ecological Data. Springer, New York.
Appendix – L ist of species recorded in the Mediterranean region of Beni-Haoua during spring (March May) 2013. In bold are the species used in the analysis.
Class
Family
Species
Cycle of life
Life-form
Anacardiaceae
Schinus molle L.
Perennial
Phanerophyte
Pistacia lentiscus L.
Perennial
Nanophanerophyte
Centella asiatica L.
Perennial
Hemicryptophyte
Daucus carota L.
Annual
Therophyte
Eryngium triquetrum Vahl.
Biennial
Hemicryptophyte
Foeniculum vulgare (Mill) Gaertn.
Perennial
Hemicryptophyte
Torilis nodosa Gaertn.
Annual
Therophyte
Apocynaceae
Nerium oIeander L.
Perennial
Phanerophyte
Asteraceae
Anacyclus pyrethrum (L.) Casso.
Annual
Therophyte
Artemisia absinthium L.
Perennial
Chamaephyte
Bellis annua L.
Annual
Therophyte
Calendula arvensis L.
Annual
Therophyte
Centaurea melitensis L.
Annual
Therophyte
Chrysanthemum myconis L.
Annual
Therophyte
Erigeron bonariensis L.
Annual
Therophyte
Galactites tomentosa (L.) Moench.
Annual
Therophyte
Helichrysum arenarium (L.) Moench.
Perennial
Hemicryptophyte
Helichrysum stoechas (L.) Moench.
Perennial
Chamaephyte
Hypochaeris glabra L.
Annual
Therophyte
Jacobaea maritima (L.) Pel.Mei.
Perennial
Chamaephyte
Pallenis spinosa (L.) Casso.
Annual
Therophyte
Phagnalon saxatile (L.) Casso.
Perennial
Chamaephyte
Picris echioides L.
Annual
Therophyte
Scolymus maculatus L.
Annual
Therophyte
Silybum marianum (L.) Gaertn.
Biennial
Hemicryptophyte
Tanacetum parthenium (L.) Sch. Bip.
Perennial
Hemicryptophyte
Taraxacum officinale F.H.
Perennial
Hemicryptophyte
Betulaceae
Alnus glutinosa (L.) Gaertn.
Perennial
Phanerophyte
Boraginaceae
Borago officinalis L.
Annual
Therophyte
Cynoglossum creticum Mill.
Biennial
Hemicryptophyte
Echium plantagineum L.
Biennial
Hemicryptophyte
Brassica nigra (L.) Koch.
Annual
Therophyte
Capsella bursa-pastoris (L.) Medik
Annual
Therophyte
Magnoliopsida
Apiaceae
Brassicaceae
Raphanus raphanistrum L.
Annual
Therophyte
Cactaceae
Opuntia ficus-indica (L.) Mill.
Perennial
Phanerophyte
Caryophyllaceae
Moehringia trinervia L.
Annual
Therophyte
Cistaceae
Cistus albidus L.
Perennial
Chamaephyte
Cistus monspeliensis L.
Perennial
Chamaephyte
Convolvulaceae
Convolvulus althaeoides L.
Perennial
Hemicryptophyte
Ericaceae
Erica arborea L.
Perennial
Nanophanerophyte
Erica cinerea L.
Perennial
Nanophanerophyte
Anthyllis tetraphylla L.
Annual
Therophyte
Astragalus monspessulanus L.
Perennial
Hemicryptophyte
Calycotome spinosa (L.) Lamk
Perennial
Nanophanerophyte
Ceratonia siliqua L.
Perennial
Phanerophyte
Hedysarum spinosissimum L.
Annual
Therophyte
Scorpiurus muricatus L.
Annual
Therophyte
Tetragonolobus biflorus Desr.
Annual
Therophyte
Fabaceae
83
Class
Family
Species
Cycle of life
Life-form
Fagaceae
Quercus coccifera L.
Perennial
Phanerophyte
Quercus ilex L.
Perennial
Phanerophyte
Quercus robur L.
Perennial
Phanerophyte
Quercus suber L.
Perennial
Phanerophyte
Erodium cicutarium L’Her.
Annual
Therophyte
Geranium robertianum L.
Annual
Therophyte
Lavandula angustifolia Mill.
Perennial
Chamaephyte
Lavandula dentata L.
Perennial
Chamaephyte
Marrubium multibracteatum H&M.
Perennial
Hemicryptophyte
Salvia verbenaca L.
Perennial
Hemicryptophyte
Geraniaceae
Pinopsida
Liliopsida
Magnoliopsida
Lamiaceae
Filicopsida
84
Thymus vulgaris L.
Perennial
Chamaephyte
Malvaceae
Althaea officinalis L.
Perennial
Hemicryptophyte
Oleaceae
Olea europea L.
Perennial
Phanerophyte
Oxalidaceae
Oxalis corniculata L.
Perennial
Hemicryptophyte
Papaveraceae
Papaver rhoeas L.
Annual
Therophyte
Plantaginaceae
Plantago coronopus L.
Annual
Therophyte
Primulaceae
Anagallis arvensis L.
Annual
Therophyte
Rosaceae
Crataegus laevigata Poir.
Perennial
Mesophanérophyte
Rubiaceae
Asperula arvensis L.
Annual
Therophyte
Rubia peregrina L.
Perennial
Hemicryptophyte
Valerianaceae
Valeriana tuberosa L.
Perennial
Cryptophyte
Agavaceae
Agave americana L.
Perennial
Hemicryptophyte
Araceae
Arisarum vulgare Targ.Tozz.
Perennial
Cryptophyte
Arecaceae
Chamaerops humilis L.
Perennial
Nanophanerophyte
IIridaceae
Gladiolus byzantinus Mill.
Perennial
Cryptophyte
Liliaceae
Chamaelirium luteum L.
Perennial
Cryptophyte
Muscari comosum (L.) Mill.
Perennial
Cryptophyte
Urginea maritima (L.) Baker
Perennial
Cryptophyte
Linaceae
Linum corymbiferum Desf.
Annual
Therophyte
Poaceae
Ampelodesma mauritanicum (Poir.)
Perennial
Hemicryptophyte
Avena sterilis L.
Annual
Therophyte
Bromus madritensis L.
Annual
Therophyte
Melica minuta L.
Perennial
Hemicryptophyte
Juniperus oxycedrus L.
Perennial
Phanerophyte
Tetraclinis articulata (Vahl)
Perennial
Phanerophyte
Pinus halepensis Mill.
Perennial
Phanerophyte
Pinus pinaster Soland.
Perennial
Phanerophyte
Asplenium adiantum-nigrum L.
Perennial
Hemicryptophyte
Asplenium ceterach L.
Perennial
Hemicryptophyte
Cupressaceae
Pinaceae
Aspleniaceae
Insight into the Dietary Habits
of the Eurasian Otter, Lutra lutra,
in the East of Algeria
(El-Kala National Park)
Aperçu du régime alimentaire de la loutre d’Europe, Lutra lutra,
dans l’est de l’Algérie (Parc national d’El-Kala)
Roland LIBOIS1, *, Rachida GHALMI2, Amina BRAHIMI1, 3
1. Labo de zoogéographie, ULG, Bâtiment B22, Chemin de la vallée, 4, 4000 Sart Tilman, Belgique
2. École nationale supérieure des sciences de la mer et de l’aménagement du littoral ;
Campus universitaire de Dely-Ibrahim Bois des Cars, BP19, 16320 Alger, Algérie
3. Département des sciences agronomiques, Université Mohammed Kheider, 7000 Biskra, Algérie
* Corresponding author: [email protected]
Received: 1 Avril, 2015; First decision: 20 October, 2015; Revised: 29 November, 2015; Accepted: 30 November, 2015
Abstract
In 1997, faeces samples (spraints) (n = 175) from
the European otter (Lutra lutra) were taken in
seven localities of the El-Kala region. This is a
restricted eco-geographic complex in which
freshwater hydrographical systems, comprising
rivers, ponds and wet coastal zones (brackish
lagoons, shipping channels), are highly interwoven. The frequency of occurrence and relative
abundance of consumed taxons were calculated
on the basis of 493 identified prey. Fish made
up more than 88% of the otter’s catch (relative
abundance), with a strong predominance of
Pseudorasbora parva, Luciobarbus callensis and
undetermined Cyprinidae. Anuran amphibians
made up 9% of the prey. The remainder were
represented, in order of significance, by mullets, eels (Anguilla anguilla), bleaks (Alburnus
alburnus), perciformes, gobies, insects, birds,
crustaceans, etc. However, Cyprinidae, the eel
and the barbel dominated in terms of ingested
biomass. The otter’s diet varies with local conditions: lagoons and channels have marine or
migratory fish; rivers are dominated by barbels; ponds by Cyprinidae. Finally, small-sized
Keywords: otter, diet, National Park El Kala,
Algeria, Pseudorasbora.
fish (topmouth gudgeon and barbel) dominated the diet in terms of numbers: 62% are
smaller than 12.5 cm. The introduction of the
topmouth gudgeon into the region could have
catastrophic consequences for endemic fish,
such as Pseudophoxinus callensis.
Résumé
En 1997, des échantillons de fèces (épreintes)
(n = 175) de la loutre d’Europe (Lutra lutra) ont
été récoltés dans sept localités dans la région
d’El-Kala. C’est un complexe écogéographique
restreint où les systèmes hydrographiques dulcicoles, constitués de rivières, d’étangs et de
zones humides littorales (lagunes saumâtres,
chenaux maritimes), sont assez imbriqués. Les
fréquences d’occurrence et l’abondance relatives des taxons consommés ont été calculées
à partir des 493 proies identifiées. Les poissons
constituent plus de 88 % des prises (abondance
relative) avec une très large prédominance de
Pseudorasbora parva, Luciobarbus callensis et
des cyprinidés indéterminés. Les amphibiens
anoures constituent 9 % des proies. Le reste
Mots clés : loutre, régime alimentaire, Parc
national d’El Kala, Algérie, Pseudorasbora.
85
Roland Libois, Rachida Ghalmi, Amina Brahimi
est représenté, selon l’importance, respectivement par des mugilidés, des anguilles (Anguilla
anguilla), l’ablette (Alburnus alburnus) des perciformes, des gobiidés, des insectes, des oiseaux,
des crustacés... Cependant, les cyprinidés, l’anguille et le barbeau dominent pour la biomasse
ingérée. Le régime alimentaire est fonction
des conditions locales : les lagunes et chenaux
avec des poissons marins ou amphihalins ; les
rivières où dominent les barbeaux ; les étangs
avec les cyprinidés. Enfin, les poissons de petite
taille (pseudorasbora et barbeau) dominent le
régime en nombre : 62 % font moins de 12,5 cm.
L’introduction de pseudorasbora dans la région
pourrait être catastrophique pour des poissons
endémiques, comme Pseudophoxinus callensis.
Introduction
Since the beginning of the 1980’s, whether it
is related to fresh, brackish or marine waters,
the diet of the Eurasian otter (Lutra lutra L.)
has been described in numerous publications.
Broyer & Erome (1982) and then Mason &
Macdonald (1986) and Kruuk (2006) summarised the findings of most of these publications. In its category of specialised, semiaquatic predators, the Eurasian otter has been
shown to be relatively opportunistic and euryphagous, feeding mainly on fish, but also on
various animals found in aquatic environments, such as cyclostomes, crustaceans and
amphibians.
Only two studies have been published on the
diet of otters in North Africa, on the basis
of their spraints: the first of these was carried out in Morocco (Broyer et al. 1988) and
deals with 389 droppings collected mainly in
Saharan river wadis, and wadis of the High
Atlas and Middle Atlas. This data is characterised by strong variations in the class of prey,
depending on seasons and localities. Although
fish traditionally play a major dietary role, the
analysis made by these authors was limited to
the identification of specimens up to the level
of zoological classes (fish, amphibians, etc.).
The second study was carried out in the Beth
river wadi, which is a typical river of the
Middle Atlas (Morocco) (Libois et al. 2015).
It was based on 760 droppings, methodologically collected at six stations, over two
annual cycles, derived from one collection
campaign per season (2,442 identified prey).
Fish made up more than 70% of the catches
(relative abundance) with a very strong predominance of barbels (Luciobarbus labiosa
[Pellegrin], Labeobarbus fritschii [Günther]
86
and Labeobarbus paytoni [Boulenger]). The
remaining prey corresponded to (in decreasing order): anuran amphibians, insects, ophidians, Mediterranean tortoises (Mauremys
leprosa Schweigger), birds, crustaceans and
small mammals. Over time, the otter’s diet has
evolved for various reasons, of both climatic
and anthropogenic origins: major floods and
their consequences profoundly modified the
composition of the fish population and leading
to the local extinction of the small barbels;
then, in the autumn of 2010, fishermen introduced Cichlidae into the river. In addition, fish
of small size dominated their diet in terms of
numbers: 80% are less than 10 cm in size.
The present study deals with the dietary habits of the Eurasian otter, in the rivers and wet
brackish zones of El-Kala in Algeria. It is proposed to provide an overview of the otter’s
choice of prey in a restricted eco-geographic
complex, in which freshwater hydrographical systems, comprising rivers, ponds and
wet coastal zones (brackish lagoons, shipping
channels) are highly interwoven.
Study area
In the spring of 1997, in the region of ElKala, a series of spraints was collected
around the Mellah lagoon (860 ha; 36.9°N
8.34°E) (n = 18), the Oubeira (2,174 ha;
36.8°N 8.4°E) (n = 19) and Tonga (2,392 ha;
36.9°N 8.5°E) (n = 17) ponds, in the Messida
channel (36.88°N 8.53°E) (n = 26), as well
as in the El-Kébir (36.7°N 8.4°E) (n = 20),
Bougous (36.66°N 8.39°E) (n = 74) and ElAreug (36.86°N 8.33°E) (n = 7) river wadis
(Ghalmi 1997). The brackish Mellah lagoon
collects the waters of a small number of rivers
that flow directly into the sea, via a relatively
narrow channel. The lagoon is lined with a
coastal strip colonised by French Tamarisk
(Tamarix gallica L.), where the otter leaves
its droppings on large branches at a height
of approximately 1.5 m above ground. The
vegetation in the Oubeira pond is dominated
by helophytes (Phragmites australis [Cav.])
and hydrophytes (Potamogeton sp. and
Myriophyllum sp.). In order to restrict the
development of aquatic plants, some species
of fish have been introduced, grass carp in
particular (Aristichthys nobilis [Richardson],
Ctenophayngodon idella [Valenciennes],
Hypophtalmichthys molitrix [Valenciennes]).
The Tonga pond is a vast P. australis and Scirpus lacustris L. reed bed, which is becoming
Insight into the Dietary Habits of the Eurasian Otter, Lutra lutra, in the East of Algeria (El-Kala National Park)
silted and communicates with the sea via the
Messida channel, a backwater blocked by
fallen trees, in which the banks are covered
by a dense shrubby vegetation. The tailrace
of the El-Kebir wadi, slightly downstream
from the Mexena dam, is lined with dense
riverside vegetation comprising ash (Fraxinus angustifolia Vahl), alder (Alnus glutinosa
[L.]) and oleander (Nerium oleander L.). Its
course and flow rate are highly irregular, with
violent flooding in winter. Faecal matter was
collected in a zone that is strongly disturbed
by the dam site. The Bougous wadi, a tributary of the El-Kebir wadi, is a torrential river
whose bed comprises large pebbles or rocky
slabs. Its banks are covered by a degraded forest of cork oaks (Quercus suber L.) and zeen
oaks (Quercus canariensis Willd.), accompanied by oleander (N. oleander) and tamarisk
(T. gallica and Tamarix africana [Poir.]). The
El-Areug wadi flows through a narrow ravine,
with a width not exceeding one meter. This
river, with its torrential regime, is situated in
a highly degraded cork oak forest and feeds
the Mellah lagoon.
Methods
In the laboratory, the analysis of the spraint
contents involved the identification of undigested remains of various prey. For this, a
standardised method for dropping treatment
was followed (Libois et al. 1987a). Teleosts
were determined through the recognition of
characteristic bone fragments, using reference
collections and previous studies: Libois et al.
(1987a, b); Libois & Hallet-Libois (1988).
Finally, reference collections were prepared
for barbels from the El-Kala region, ophidians
and amphibians. The feathers were identified
by Dr. R. Rosoux, on the basis of the ornithological reference collection of the natural
sciences museum in Orléans, France.
To optimise data processing, three methods
were used: occurrence, abundance and relative biomass, in accordance with the recommendations of Libois et al. (1987a, 1991);
Libois & Rosoux (1989, 1991) and Libois
(1995, 1997). The characteristic bone fragments were methodically measured, and the
size of the biomass of the consumed fish was
thereafter estimated on the basis of the studies
of Wise (1980) for vertebrae, and the studies of Libois & Hallet (1988) for cephalic
fragments. In the case of amphibians, their
mass was estimated to be 10, 15, 20, 30 and
40 g, depending on the estimated size. We
estimated the weighted mass of ophidians to
be 100 g and that of birds, depending on the
species: the respective values were consulted
in the “Handbook of the birds of the world”
(Anatidae: Carboneras 1992; Rallidae: Taylor
1996).
Results
A total of 493 prey are identified for a total of
294 occurrences (Tables 1, 2).
In general, the El-Kala otters have a “classical” diet with diversified prey: this can be
broken down into fish, amphibians, reptiles,
birds, arthropods and even molluscs. In terms
of occurrence and abundance, fish are by far
the main type of prey: barbels (Luciobarbus callensis [Valenciennes]), undetermined
Cyprinidae and an introduced fish: the topmouth gudgeons (Pseudorasbora parva
[Schlegel]). The biomass represented by the
undetermined Cyprinidae, barbels and eels
(Anguilla anguilla [L.]) make up two thirds
of the prey found in the otter’s droppings (Figure 1).
The dietary intake is specific to the seven
sites. Indeed, in the surroundings of the Mellah lagoon, mainly small mullets (< 225 mm),
eels, and also other diadromous fish: gobies
(Gobiidae), big-scale sand smelts (Atherina
boyeri Risso) and some Sparidae. Near to
the Oubeira pond, large Cyprinidae dominate, although the species could not be identified. On the other hand, on the Tonga pond,
the dominant prey are P. parva and anuran
amphibians.
In terms of biomass, birds predominate, with
purple swamp hens (Porphyrio porphyrio
[L.]) and two other rallidae, followed by eels
(5 individuals more than 45 cm in length).
The specific prey richness is greater in the
Messida channel: 12 taxons were counted,
including the P. parva, which makes up the
main component of the food (70%). However, as a consequence of the small size of
this invasive species, the corresponding food
intake is low. The biomass is dominated by
large eels, perciformes and mullets. As the
channel feeds into the sea, marine prey (chitons, crustaceans, decapods) or diadromous
fish (mullets and perciformes) can be found
there. In the rivers (El-Kebir and Bougous),
87
Table 1 – Prey occurrence of otter’s spraints by locality in the El-Kala National Park
Total
Mellah
Oubeira
Tonga
Messida
El Kebir
Bougous
El Areug
Polyplacophora
1
1
Crustacea
3
1
1
1
Insecta
4
2
1
1
Anguilla anguilla
25
11
5
5
4
Cyprinidae ind.
57
Alburnus alburnus
7
1
Carassius sp.
3
1
Ctenopharyngodon idella
1
1
Cyprinus carpio
1
Luciobarbus callensis
70
Pseudorasbora parva
43
Atherina boyeri
2
2
Gambusia sp.
2
1
1
Mugilidae
20
11
1
Perciformes
8
3
Gobiidae
2
2
Teleostei ind.
5
2
1
Anoura
36
3
5
Ophidia
1
1
Aves
3
2
1
294
37
27
34
52
27
102
15
Bougous
El Areug
15
2
12
26
2
5
1
2
1
1
3
11
8
20
59
2
9
1
6
2
4
2
9
9
5
3
2
Table 2 – Prey abundance of otter’s spraints by locality in the El-Kala National Park
Total
Mellah
Polyplacophora
1
Crustacea
3
1
Insecta
5
2
Anguilla anguilla
25
11
5
Cyprinidae ind.
65
Alburnus alburnus
12
1
Carassius sp.
3
1
Ctenopharyngodon idella
1
1
Tonga
16
Messida
El Kebir
1
1
1
1
2
5
4
2
14
30
3
9
2
2
1
Luciobarbus callensis
112
Pseudorasbora parva
162
Atherina boyeri
2
2
Gambusia sp.
2
1
1
Mugilidae
30
21
1
Perciformes
9
4
Gobiidae
5
5
Teleostei ind.
5
2
1
Anoura
45
3
7
Ophidia
1
1
Aves
4
3
1
493
51
39
81
101
31
168
22
Cyprinus carpio
88
Oubeira
1
1
12
54
10
69
99
2
27
1
6
2
4
2
12
9
5
3
6
Figure 1 – Prey relative biomass of otter’s spraints (El-Kala National Park) (21.5 kg).
barbels play a key role in the otters’diet. In the
El-Areug wadi, close to the Mellah lagoon,
the brackish waters receive prey, such as mullets from the marine environment, and prey
such as barbels from the freshwater environment. The frequency distribution of size
classes has been studied for the fish that are
most abundantly consumed by the otter: L.
callensis, P. parva, A. anguilla, mullets and
other Cyprinidae (Figure 2). A very high proportion of the fish are small in size: 62% are
smaller than 125 mm.
Figure 2 – Fish distribution size in the otter’s spraints (El-Kala National Park).
89
Discussion
In the El-Kala region, the otter’s diet basically resembles that found in studies carried out in Europe (Broyer & Erome 1982;
Mason & Macdonald 1986; Libois 1995)
and in Morocco (Broyer et al. 1988; Libois
et al. 2015). In fact, it is made up essentially
from fish, but also amphibians, reptiles, birds,
mammals, crustaceans and insects, most of
which are aquatic or semi-aquatic animals, or
animals that are temporarily related to water
in their biological cycle, as in the case of the
artificial studies carried out by Heptner &
Naumov 1974; Mason & Macdonald 1986;
Libois 1995; Kruuk 2006.
Although aquatic habitats can be highly
diverse, ranging from small rivers to ponds
and lagoons, the otter demonstrates considerable trophic plasticity, as well as a remarkable
adaptive capacity, depending on the various
environments in which it lives, which provide
a large variety of prey.
It demonstrates a clear pattern of opportunistic behaviour, since the proportions of prey
classes remain the same, whether the results
be expressed in terms of presence or relative
abundance. This confirms the conclusions of
similar studies carried out in Europe: Erlinge
(1967); Fairley & Wilson (1972); Webb
(1975); Callejo-Rey et al. (1979); Jenkins et
al. (1979); Chanin (1981); Wise et al. (1981);
Gormally & Fairley (1982); Green et al.
(1984); Bouchardy (1986); Delibes & Adrian
(1987); Libois et al. (1987a); Callejo (1988);
Libois & Rosoux (1989); Libois (1995, 1997).
In the El-Kala region, we observed strong
variations at different sites, like for example
in Scotland (Kruuk & Moorhouse 1990) or
in the French Massif Central (Libois 1997).
On the other hand, we are not able to draw
clear conclusions concerning the otter’s possible preference for one particular class of
size. We have noticed that most of the fish it
catches, barbels and P. parva in particular, are
small in size (less than 10 cm), an observation
that had already been made in other regions
by various authors (Webb 1975: stickleback,
Gasterosteus gymnurus [Cuvier], loach,
Barbatula barbatula [L.], and sculpin, Cottus gobio L.; Jenkins & Harper 1980: pike,
Esox lucius L. and perch, Perca fluviatilis
L.; Jenkins et al. 1979: pike, perch, salmonidae; Chanin 1981; Green et al. 1984; Libois
1997; Libois et al. 2015: barbels, Luciobarbus labiosa and Labeobarbus fritschii). In
90
hydrosystems, small-sized fish generally
dominate in terms of numbers. Nevertheless,
it is necessary to compare the frequency distributions of the size of a prey-species in a
predator’s food intake and in its habitat (capture using electro-fishing, traps or fishways),
which was not possible to implement in the
context of the present study.
The introduction of P. parva appears to have
had significant repercussions within the ElKala fish community. Although Pseudophoxinus callensis (Guichenot) lives in the
Oubeira pond (Paris: MNHN-16-2000-5725,
det. Daget), its absence in the otter droppings
could be indicative of competition between
the two species, to the advantage of P. parva.
The latter species is zooplanctivorous, but
consumes the eggs and fry of other fish and
at the international level it is thus considered to be noxious (Keith et al. 2011; Witkowski 2011). For the El-Kala region and
the Kroumirie, where Pseudophoxinus are
endemic, this poses a serious problem for the
conservation of biodiversity.
Acknowledgement
We wish to thank Mr. K. Djeffel, Director of
the El-Kala National Park, together with his
entire team, and in particular Mr. K. Bousentouh and A. Boukrabouza for their assistance in the field. Mr. René Rosoux, Scientific Director of the Orléans natural science
Museum, provided us with the identification
of feathers in spraints, and corrections to the
first draft of this manuscript.
References
Bouchardy C., 1986. La loutre. Sang de la terre, Paris.
174 p.
Broyer J. & Erome G., 1982. Eléments d’écologie de la
loutre Lutra lutra (L). Premières données bibliographiques. Bièvre 4 : 33-58.
Broyer J., Aulagnier S. & Destre R., 1988. La loutre,
Lutra lutra angustifrons Lataste, 1885, Maroc.
Mammalia 52: 361-370.
Callejo A., 1988. Le choix des proies par la loutre (Lutra
lutra) dans le nord-ouest de l’Espagne, en rapport
avec les facteurs de l’environnement. Mammalia
52: 11-20.
Callejo-Rey A., Rivera J.G., Bas-Lopez S., SanchezCanals J.L. & De Castro-Lorenzo A., 1979. Primeros datos sobre la dieta de la Nutria, Lutra lutra (L.),
en aguas continentales de Galicia. Donãna Acta
Vert. 6: 191-202.
Carboneras C., 1992. Family Anatidae. In: del Hoyo
J., Elliott A. & Sargatal J. (eds.), Handbook of the
Birds of the world, vol. 1. Lynx Edicions, Barcelona,
536-628.
Chanin P., 1981. The diet of the Otter and its relations
with the feral Mink in two areas of Southwest England. Acta Theriol. 26: 83-95.
Delibes M. & Adrian I., 1987. Effects of crayfish introduction on otters, Lutra lutra, food in the Donãna
National Park, SW. Spain. Biol. Conserv. 42: 153159.
Erlinge S., 1967. Food habits of the fish otter, Lutra
lutra, in South Swedish habitats. Viltrevy 4: 371443.
Fairley J.S. & Wilson S.C., 1972. Autumn food of otter
(Lutra lutra) on the Agivery river, Co. Londonderry,
Northern Ireland. J. Zool. 166: 468-469.
Ghalmi R., 1997. Étude préliminaire du régime alimentaire de la loutre (Lutra lutra) dans le nord-est
algérien (Parc national d’El-Kala). DES Sciences
Naturelles, Université de Liège, Liège. 57 + 13 p.
Gormally M.J. & Fairley J.S., 1982. Food of otters, Lutra
lutra, in freshwater lough and an adjacent brackish
lough in the West of Ireland. J. Zool. 197: 313-321.
Green J., Green R. & Jefferies D.J., 1984. A radiotracking survey of otters, Lutra lutra, on a Pertshire river
system. Lutra 27: 85-145.
Heptner V.G. & Naumov N.P., 1974. Die Säugetiere der
Sowjetunion. Band II. Fischer Verlag, Iena. 1009 p.
Jenkins D. & Harper R.J., 1980. Ecology of otters in
Northern Scotland. II. Analyses of otter and mink
faeces from Deeside (N.E. Scotland) in 1977-1978.
J. Anim. Ecol. 49: 737-754.
Jenkins D., Walker J.G.K. & Mccowan D., 1979. Analyses of otter (Lutra lutra) faeces from Deeside, NE
Scotland. J. Zool. 187: 235-244.
Keith P., Persat H., Feunteun E. & Allardi J., 2011. Les
poissons d’eau douce de France. Biotope et Muséum
national d’histoire naturelle, Mèze et Paris. 552 p.
Kruuk H., 2006. Otters: ecology, behaviour and conservation. Oxford University Press, Oxford. 265 p.
Kruuk H. & Moorhouse A., 1990. Seasonal and spatial
differences in food selection by otters (Lutra lutra)
in Shetland. J. Zool. 221: 621-637.
Libois R.M., 1995. Régime et tactique alimentaire de
la loutre (Lutra lutra) en France : synthèse. Cahiers
Ethol. 15: 251-274.
Libois R., 1997. Régime et tactique alimentaire de la
loutre (Lutra lutra) dans le Massif central. Vie et
Milieu 47: 33-45.
Libois R.M. & Hallet-Libois C., 1988. Eléments pour
l’identification des restes crâniens des poissons
dulçaquicoles de Belgique et du nord de la France.
II. Cypriniformes. Fiches d’ostéologie animale pour
l’archéologie. Sér. A. n° 4. CRA-CNRS. Valbonne,
24 p.
Libois, R.M. & Rosoux R., 1989. Écologie de la loutre
(Lutra lutra) dans le marais poitevin. I. Étude de la
consommation d’anguilles (Anguilla anguilla). Vie
et Milieu 39: 191-197.
Libois R.M. & Rosoux R., 1991. Ecologie de la loutre
(Lutra lutra) dans le marais poitevin. II. Aperçu
général du régime alimentaire. Mammalia 55:
35-47.
Libois R.M., Hallet-Libois C. & Lafontaine L., 1987a.
Le régime alimentaire de la loutre (Lutra lutra) en
Bretagne inférieure. Rev. Ecol. (Terre & Vie) 42:
135-144.
Libois R.M., Hallet-Libois C. & Rosoux R., 1987b.
Eléments pour l’identification des restes crâniens
des poissons dulçaquicoles de Belgique et du nord
de la France. I. Anguilliformes, Gastérostéiformes,
Cyprinodontiformes, Perciformes. Fiches d’ostéologie animale pour l’archéologie. Sér. A. n° 3. CRACNRS. Valbonne, 15 p.
Libois R.M., Rosoux R. & Delooz E., 1991. Écologie
de la loutre (Lutra lutra) dans le marais poitevin. III.
Variation du régime et tactique alimentaire. Cahiers
Ethol. 11: 31-50.
Libois R., Fareh M., Brahimi A. & Rosoux R., 2015.
Régime alimentaire et stratégie trophique saisonnière de la Loutre d’Europe, Lutra lutra, dans le
moyen-Atlas (Maroc). Rev. Ecol. (Terre & Vie) 70:
314-327.
Mason C.F. & Macdonald S.M., 1986. Otters: ecology
and conservation. Cambridge University Press,
Cambridge. 236 p.
Taylor P.B., 1996. Family Rallidae. In: del Hoyo J., A.
Elliott A. & Sargatal J. (eds.), Handbook of the Birds
of the world, vol. 3. Lynx Edicions, Barcelona: 108209.
Webb J.B., 1975. Food of the Otter (Lutra lutra) on the
Somerset levels. J. Zool. 117: 486-491.
Wise M.H., 1980. The use of fish vertebrae in scats for
estimating prey size of otters and mink. J. Zool. 192:
25-31.
Wise M.H., Linn I.J. & Kennedy C. R., 1981. A comparaison of feeding biology of Mink, Mustela vison
and Otter, Lutra lutra. J. Zool. 195: 181-213.
Witkowski A. 2011: NOBANIS – Invasive alien species
fact sheet – Pseudorasbora parva. – From: Online
Database of the European Network on invasive
alien species – NOBANIS www.nobanis.org. Date
of access 1/2/2015.
91
Résumés de thèses
Coralie CALVET
2015
Analyse de l’utilisation de la
compensation écologique dans
les politiques publiques comme
outil de conciliation des intérêts
économiques et des objectifs de
conservation de la biodiversité
Analysis of the use of biodiversity
offsetting in public policies as a
balancing tool of economic interests
and biodiversity conservation
objectives.
Thèse de doctorat en sciences économiques soutenue le
17 décembre 2015 à l’université d’Avignon.
Jury - Valérie Boisvert (Pr, université de Lausanne, Suisse), Sophie
Thoyer (Pr, Supagro, Montpellier, France), rapporteurs ; John
Thompson (directeur de recherches Umr Cefe-Cnrs, Montpellier,
France), François Mesléard (Pr, université d’Avignon, France)
examinateurs ; Philippe Puydarrieux (chef de bureau, Ministère
de l’écologie, Paris, France) membre invité ; Claude Napoléone
(ingénieur de recherches, Ecodev-Inra, Avignon, France), Thierry
Dutoit (directeur de recherches, Umr Imbe-Cnrs, Avignon, France)
codirecteurs.
Mots-clés : compensation écologique, conservation de la biodiversité, politiques publiques, économie néo-institutionnelle.
Keywords: biodiversity offsetting, biodiversity conservation,
public policies, new-institutional economics.
Les gouvernements se sont récemment engagés à enrayer
l’érosion de la biodiversité. Dans ce contexte, la compensation écologique est apparue comme une réponse politique en permettant, en principe, de répondre à l’exigence
de conciliation de deux intérêts souvent antagonistes : le
développement économique et la conservation de la biodiversité. L’objectif de ce travail de thèse est d’analyser si la
compensation écologique peut accomplir cette promesse.
J’envisage cette problématique avec trois angles d’analyse
complémentaires et de façon interdisciplinaire en mobilisant
les apports de l’économie et de l’écologie. Premièrement,
dans une approche théorique, je pose la question de la com92
patibilité du principe de la compensation avec son objectif de
conservation de la biodiversité. Je pose ensuite la question de
la possibilité d’atteindre l’objectif d’absence de perte nette
de biodiversité dans la mise en œuvre de la compensation.
Pour cela, j’étudie empiriquement deux mécanismes de
compensation au travers de deux cas d’étude français : une
banque de compensation, et la contractualisation agro-environnementale. J’utilise principalement les outils de l’économie néo-institutionnelle pour analyser l’efficacité de ces
mécanismes pour la réalisation des objectifs écologiques
de la compensation. Au travers d’une approche épistémologique, ma troisième interrogation porte sur le rôle des
dynamiques politiques dans la diffusion et dans la promotion de la compensation écologique dans la communauté
scientifique. L’analyse théorique met en évidence des limites
intrinsèques au principe de la compensation pour atteindre
ses objectifs de conservation de la biodiversité, notamment
au regard de l’impossibilité d’adopter une approche écologique complexe de la biodiversité dans le processus de
la compensation. L’étude empirique montre que les modes
d’organisation de la compensation comportent également
des limites qui obligent à des compromis susceptibles de
remettre en cause l’atteinte des objectifs écologiques de
la compensation. Ces résultats mettent en évidence le rôle
et l’importance des institutions dans la mise en œuvre des
compensations, notamment pour limiter l’apparition de
comportements opportunistes, responsables des principaux
problèmes d’efficacité identifiés. Enfin, l’analyse épistémologique révèle que le développement et la promotion de la
compensation écologique répondent à un agenda politique
principalement porté par les politiques anglo-saxonnes et
certains acteurs de conservation. Ainsi, la compensation écologique n’est pas un objet neutre car elle sert à la diffusion
d’une certaine idéologie sur la pratique de la conservation
de la biodiversité dans le sillage du développement durable
et de l’économie verte. Pour conclure, ce travail permet de
souligner que la conciliation des intérêts économiques et
écologiques constitue une problématique complexe dont
la voie du consensus ne semble pas permettre de répondre
aux enjeux d’érosion de la biodiversité. La compensation
offre en somme une occasion de penser les conditions de
possibilités et d’impossibilités de la protection de la nature
aujourd’hui.
Governments have recently pledged to halt the loss of biodiversity. In this context, biodiversity offsetting (BO) appeared
as a political response by allowing, in principle, to reconcile two often conflicting interests: economic development
and biodiversity conservation. The objective of this work is
to analyse whether BO can fulfill that promise. I am conecologia mediterranea – Vol. 41 (1) – 2015
sidering this issue from an interdisciplinary perspective by
mobilizing the contributions of economy and ecology and
using three complementary approaches. First, in a theoretical perspective, I ask the question of the compatibility of the
offsetting principle with its biodiversity conservation goal.
Second, in an empirical approach, I investigate the strengths
and limits of using BO mechanisms in conservation policies,
particularly two specific mechanisms analysed through two
French case-studies: habitat bank and agri-environmental
contracts. I mainly use the new institutional economics
framework to analyse the effectiveness of these mechanisms
for achieving the environmental objectives of BO. Finally,
I adopt an epistemological approach to question the role of
political dynamics in the diffusion and the promotion of BO
in the scientific community. The results of the theoretical
analysis highlights the inherent limits to the BO principle
to achieve its conservation goals, especially with regard to
the impossibility to adopt a complex ecological approach
to biodiversity in the process of BO. My empirical study
shows that the mechanisms of BO also have limitations that
force compromises likely to jeopardize the achievement of
environmental objectives of BO. These outcomes highlight
the role and the importance of institutions in the implementation of BO, especially to provide clear and precise rules
in order to limit the emergence of opportunistic behaviors,
responsible for the major problems identified. Finally, my
epistemological analysis shows that the development and
the promotion of BO actually respond to a political agenda
driven by the Anglo-Saxon policies and some conservation
stakeholders. Overall my work emphasizes that BO is not a
neutral object as it serves to spread a certain ideology on the
practice of biodiversity conservation through the lens of sustainable development and green economy. Finally, this work
allows stressing that reconciling economic and conservation
interests is a complex problem that cannot be solved by using
idealized consensus. Rather, the concept of BO offers an
opportunity to think about the conditions of possibilities and
impossibilities of the protection of nature today.
Thibaut FRÉJAVILLE 2015
Vulnérabilité des forêts
de montagne des Alpes occidentales
au changement de régime d’incendie
Vulnerability of mountain forests
to changing fire regime
in the western Alps
Thèse de doctorat en sciences de l’environnement soutenue le 25 juin 2015 à l’université d’Aix-Marseille, École doctorale 251.
Encadrement - Thomas Curt (directeur de recherches, Irstea, Aixen-Provence, France), Christopher Carcaillet (Directeur d’études,
école pratique des hautes études) codirecteurs.
Mots clés : Alpes, biogéographie, incendies, inflammabilité,
résistance, traits fonctionnels.
Key-words: Alps, biogeography, fire, flammability, functional
traits, resistance to fire.
Les forêts d’altitude connaissent une émergence croissante
des feux. Dans le sud de l’Europe, les effets du réchauffement du climat sont accrus par les changements d’usage qui
génèrent une accumulation forte du combustible végétal et
une reconnexion des massifs forestiers de montagne. Via
une analyse rétrospective des feux et du climat des dernières
décennies dans le sud-est de la France, cette thèse montre
d’abord que l’augmentation de l’activité des feux a été largement restreinte aux zones de montagne, en particulier dans
les Alpes du sud, les politiques de lutte contraignant les feux
au sein des paysages méditerranéens. Pour quantifier la vulnérabilité au feu des écosystèmes et des espèces d’arbre de
montagne, j’ai ensuite croisé des données de climat et de
végétation des Alpes occidentales afin de caractériser les
attributs d’inflammabilité et de résistance du végétal qui
gouvernent ses effets et sa réponse au feu. Ces informations
ont été utilisées pour simuler le comportement des feux de
surface et leurs dommages (mortalités des arbres) au sein
des niches environnementales des espèces dominantes. Les
résultats obtenus montrent que l’abondance et les propriétés
d’inflammabilité des litières, des strates herbacées et arbustives varient au sein des Alpes en fonction de l’ouverture
de canopée, sa hauteur et la saisonnalité des précipitations.
Ces gradients biotiques et abiotiques gouvernent ainsi les
niches d’inflammabilité des espèces, depuis les forêts denses
et humides peu inflammables des Alpes externes du nord
aux forêts ouvertes inflammables périméditerranéennes des
Alpes du sud et subalpines des Alpes internes. Par ailleurs,
les traits de résistance au feu des arbres varient entre espèces
en lien avec l’inflammabilité des communautés, selon un
compromis entre tolérance (écorce épaisse en climat sec)
et évitement (houppier haut en climat humide), suggérant
une coévolution entre espèces et feux. De ces interrelations
entre climat, composition et structure de la canopée résultent
des patrons biogéographiques de vulnérabilité des écosystèmes et des espèces au feu. Le chêne, le pin sylvestre et
le pin noir connaîtraient des taux de mortalité élevés. Leur
vulnérabilité est néanmoins réduite au sein des forêts fermées et plus humides du montagnard, constituant, ainsi,
une marge favorable. À l’inverse, les espèces qui dominent
ces marges favorables (le sapin et l’épicéa) sont peu vulnérables (excepté le hêtre et son écorce fine), bien que les
taux de mortalité s’accroissent rapidement au sein de leurs
marges sèches, à basse (forêts supraméditerranéennes) ou
haute altitude (forêts subalpines des Alpes internes). Enfin,
l’absence d’attributs de résistance pour le pin à crochet et
en particulier le pin cembro définit ces espèces comme les
plus vulnérables aux changements globaux. En effet, leur
niche est restreinte aux forêts subalpines les plus alticoles et
93
particulièrement inflammables en conditions extrêmes, alors
que l’écorce épaisse du mélèze qui codomine ces forêts lui
permettrait d’outrepasser les feux, comme cela a été observé
par le passé. Ainsi, du fait des propriétés de leur niche et de
leurs traits d’histoire de vie, les arbres Alpins ne sont pas
tous égaux face à l’émergence des feux, mettant en lumière
l’importance des résultats de cette thèse dans l’anticipation
possible des dommages des feux futurs au sein de la diversité
des conditions de végétation et de climat des montagnes du
sud de l’Europe.
Mountain forests are experiencing an increase in fire occurrence. In southern Europe, land use changes enhance the
effects of global warming by promoting a fuel build-up and
a greater connectivity of forests. Through a retrospective
analysis over the past decades of fire and climate in southeastern France, this dissertation firstly shows that Mediterranean ecosystems have experienced a strong decrease in fire
activity, emphasizing the strong efficiency of fire suppression policies, while fires have been more frequent and large
in mountains like the southern Alps. I used vegetation and
climate data to characterize the flammability and fire resistance traits of mountain forests of the western Alps that determine their effects and responses to fire. Then, assessing the
vulnerability to fire of mountain ecosystems and tree species
was carried out by simulating the behaviour of surface fires
and their impacts (tree mortality) within the environmental
niches of dominant tree species. Results show that tree cover
and height, and the seasonality of precipitation largely drive
the amount and flammability properties of surface fuels (litter, grass and shrubs). Therefore, these environmental gradients govern the flammability niches of tree species, from
the few flammable dense and moist forests of the northern
Alps, to the open and highly flammable sub-Mediterranean
and subalpine forests of the southern and inner Alps, respectively. Otherwise, fire resistance traits of trees vary according
to community flammability through a trade-off between tolerance (thick bark in dry climates) and avoidance strategies
(high crown in moist climates), suggesting a coevolution
between species and surface fires. These interrelationships
between climate, composition and canopy structure highlight biogeographic patterns of ecosystem and tree species vulnerability to fire in the western Alps. Fire simulations indicate that Quercus pubescens, Pinus sylvestris and
P. nigra should suffer high mortality rates under extreme fire
weather conditions. However, their vulnerability is strongly
reduced within the close and moist forests of the montane
belt, which therefore constitute favourable margins for these
species. On the contrary, tree species that dominate these low
flammable margins (Abies alba, Picea abies) are less vulnerable to fire (except Fagus sylvatica having a too thin bark),
although they should suffer high mortality rates towards
their dry margins at low (sub-Mediterranean forests) and
high elevation (subalpine forests). Finally, having thin barks
and low crowns, the subalpine pines (P. uncinata, P. cembra)
seem the most vulnerable to global changes. Indeed, their
narrow niches are mostly restricted in the most elevated forests that are highly flammable under extreme fire weather. In
contrast, their co-dominant species Larix decidua displays
a thick bark and the lowest vulnerability, highlighting that
this species may resist to surface fires as shown in the last
millennia. To conclude, the niche properties and the life history traits of Alpine trees make them unequally exposed to
an increasing fire risk, highlighting the importance of the
findings of this thesis to anticipate the impacts of future fires
across the diversity of climate and vegetation of southern
European mountains.
Jacques GAMISANS, un botaniste catalan en Corse (1944-2015)
« En abordant en 1966, l’étude de la végétation des montagnes corses, je pensais,
étant donné la liste impressionnante des travaux de mes prédécesseurs, avoir affaire à
un ensemble floristique parfaitement connu [...]. L’entrée dans le vif du sujet m’a montré
qu’il n’en était rien [...]. Certains secteurs montagneux n’avaient encore jamais reçu de
visite de botanistes et la prospection systématique des massifs corses pendant sept années
(1966-1972) m’a permis de découvrir quelques taxons inédits ou nouveaux pour l’île. »
Jacques Gamisans, 1976. Phytocoenologia, 3 (4).
C
très fier de ses origines, Jacques Gamisans est
né en 1944. Après ses études secondaires, il acheva la
première phase de son cursus universitaire par la soutenance à l’université de Toulouse de son diplôme d’études
supérieures (Des) portant sur les Caractères anatomiques des
Calamagrostis de la flore française.
Jacques Gamisans a été enseignant-chercheur (maître
de conférences) à l’université d’Aix-Marseille III (faculté des
sciences et techniques de Marseille-Saint-Jérôme) jusqu’en
1995, puis à l’université Paul-Sabatier de Toulouse où il termina
sa carrière universitaire en 2004.
C’est lors d’un premier voyage d’agrément à l’invitation
d’un ami d’Oletta, en 1962, qu’il découvrit cette remarquable
île-montagne qu’est la Corse. Dès lors, son orientation était
toute trouvée, il allait s’attacher à étudier la flore et la végétation
orophiles de cette île car les domaines alticoles étaient encore
peu connus.
À partir de 1966, il commença à arpenter assidûment la
région de Vizzavona et à en étudier la végétation alticole dans
le cadre du projet de création du parc national de Corse, projet
qui ne verra pas le jour mais qui fut l’une des prémices à la
création du parc naturel régional de Corse. Cette recherche le
conduisit à soutenir en 1968 sa thèse de spécialité « Étude phytosociologique de la zone montagneuse correspondant au projet
de Parc national de Corse », à l’université d’Aix-Marseille sous
la direction du professeur Pierre Quézel, responsable du laboratoire de botanique et d’écologie méditerranéenne. Intégré à
ce laboratoire comme assistant puis maître-assistant, Jacques
Gamisans poursuivit ses recherches dans le cadre de son doctorat consacré à la végétation de la haute montagne corse, en
développant des approches en floristique, phytosociologie et
biogéographie, toujours sous la houlette de P. Quézel.
atalan
En 1975, il soutint sa thèse de doctorat d’État ès sciences
naturelles, « La végétation des montagnes corses », devant un
jury prestigieux composé de MM. P. Quézel, H. Ellenberg,
M. Grison, P. Ozenda, A. Pons, Ch. Sauvage et R. Tomaselli.
Cette thèse fut publiée in extenso dans la revue internationale
Phytocoenologia. Ce travail conduisit à la description et à
l’analyse détaillées de 42 associations de végétations de l’étage
supraméditerranéen à l’étage alpin, dont 30 associations inédites, et à la mise en évidence de 15 taxons nouveaux pour
la science et 15 nouveaux pour la Corse. La soutenance, le
4 mars 1975, fut d’ailleurs un jour marquant pour la connaissance de la végétation corse et de sa dynamique, puisque son
collègue et ami Maurice Reille, un paléoécologue spécialisé
dans l’étude des pollens fossiles, soutint aussi, l’après-midi
même, son doctorat d’État sur la végétation tardiglaciaire et
holocène de Corse ! Leurs terrains communs et les résultats
complémentaires obtenus ont conduit à des conceptions novatrices à l’échelle méditerranéenne, avec une mise en perspective
de la dynamique et de la biogéographie du peuplement végétal
de ces hautes montagnes méditerranéennes encore marquées
par la présence de taxons de souche européenne, notamment
le fameux élément artico-alpin.
Jacques Gamisans continuait en parallèle ses travaux de
botanique systématique et il publia, entre 1970 et 1985, dix
« Contributions à l’étude de la flore corse » dans la revue suisse
Candollea. Plusieurs taxons nouveaux pour la science furent
décrits (Adenostyles briquetii, Erigeron paolii, Seseli djianeae,
Trisetum conradiae, etc.) et de nombreuses espèces inconnues
en Corse furent découvertes. Ces herborisations lui permirent
de réaliser en 1985 une première mouture du Catalogue des
plantes vasculaires de la Corse. Cette synthèse sera ensuite
approfondie et complétée dans une seconde édition publiée en
95
Hommage à Jacques GAMISANS
1993 avec son complice Daniel Jeanmonod, conservateur aux Conservatoire et Jardin botaniques de la
ville de Genève. À partir de 1986, il présida le comité
scientifique « Flore corse », qui permit d’achever le
monumental Prodrome de la flore corse initié par
John Briquet, grâce à la publication de plusieurs
monographies de familles botaniques. Sa collaboration fructueuse avec D. Jeanmonod aboutit à la réalisation du Flora Corsica, la première flore complète
moderne de la dition publiée en 2007 et mise à jour
dans une seconde édition parue en 2013, un ouvrage
remarquable par sa clarté et sa concision. Ses activités floristiques n’ont pas faibli comme le montre la
description en 2011 avec Laetitia Hugot, directrice
du Conservatoire botanique national de Corse, d’un
remarquable endémique du massif du Cintu, Hippocrepis conradiae.
Au cours des années 1980 et 1990, Jacques
Gamisans étendit ses études phytoécologiques corses
aux zones de basse et moyenne altitudes, avec plusieurs travaux détaillés de cartographie de la végétation effectués dans le Niolu, le Haut-Venacais, les
réserves naturelles de la presqu’île de Scandola et de
l’étang de Biguglia, les îles Lavezzi et Cerbicales,
etc. Plusieurs travaux furent aussi consacrés à l’étude
de la distribution et de la dynamique des divers types
de forêts ou de ligneux caractéristiques des paysages
végétaux de Corse (Pin laricio, Aulnes, Genévrier
thurifère, etc.). Tout ce corpus approfondi de relevés
détaillés facilita la rédaction de la première monographie phytoécologique complète de l’île, La Végétation de la Corse, publiée en 1991 et rééditée en 1999.
Si ses recherches se sont presque exclusivement
focalisées en Corse, Jacques Gamisans a participé à
d’autres travaux en Méditerranée, notamment dans
le cadre de programmes conduits avec des collègues
du laboratoire de botanique et d’écologie méditerraMission dans la haute montagne corse en juillet 1971, avec Maurice Reille.
En haut, au sommet du Monte Rotondo ;
néenne de l’université d’Aix-Marseille. À l’étranger,
en bas, dans les pozzines de Vaccaja. (Clichés DR)
il aida ainsi, durant les années 1970, Gilles Bonin
dans la réalisation de son doctorat d’État portant sur
la végétation forestière de l’Italie méridionale, et
avec Jean-Pierre Hébrard il étudia les forêts du nord de la Grèce corse. En 1981, tous deux publient un article intitulé « À pro(Épire, Macédoine et Thrace) lors de deux missions (1976- pos de certaines espèces de la flore corse menacées de dis1977) financées par le Cnrs. Mais c’est en France méridionale parition », prémices d’actions plus importantes initiées par le
(Provence, Drôme, Cévennes) et dans les Pyrénées, notamment parc naturel régional de Corse, le Conservatoire botanique de
avec son collègue et ami Michel Gruber, lui aussi maître de Porquerolles et l’Agenc, dans le cadre du programme européen
conférences à l’université d’Aix-Marseille, que Jacques Gami- Medspa en 1989-1993, « Inventaire permanent et protection des
sans fut le plus actif, hors de sa terre de prédilection. Enfin, il plantes menacées, rares ou endémiques de la Corse » ; Jacques
participa à plusieurs études phytochimiques sur les activités Gamisans participa activement à ce projet pilote et novateur à
antioxydantes et sur les corps gras de divers végétaux.
l’échelle de la Méditerranée qui permit d’élaborer les premières
Cette activité soutenue s’est traduite par la publication listes rouges de la flore menacée et diverses actions de conserde 170 articles scientifiques dont 8 ouvrages principaux, et de vation in situ ou ex situ des végétaux les plus vulnérables.
nombreux rapports d’études.
Lors de la création en 2007 du Conservatoire botanique
Sans doute au contact de Marcelle Conrad, qu’il avait national de Corse, Jacques Gamisans devint tout naturellement
rencontrée dès 1966 à Vizzavona, Jacques Gamisans s’est président du conseil scientifique qu’il présida jusqu’en 2014. Il
rapidement attaché à la préservation du patrimoine végétal siégea aussi de longues années au conseil scientifique régional
96
Hommage à Jacques GAMISANS
Les principaux ouvrages
de Jacques Gamisans
BRUN B., BRUN L., CONRAD M. &
GAMISANS J., 1975. La Nature
en France : Corse. Horizons de
France, Paris : 224 p.
GAMISANS J., 1985. Catalogue des
plantes vasculaires de la Corse.
Précédé de données statistiques et
d’un exposé synthétique sur l’origine de cette flore et son organisation en ensembles de végétation.
Parc naturel régional de la Corse,
Ajaccio : 231 p.
GAMISANS J., 1991. La végétation
de la Corse. In : Compléments au
Prodrome de la flore corse. Annexe
2. Éd. Conservatoire & Jardin botaniques de Genève : 391 p. (seconde
édition publiée en 1999 parÉdisud,
Aix-en-Provence).
Jacques Gamisans lors de l’excursion du colloque international
« Connaissance et conservation de la flore des îles
de la Méditerranée », octobre 1993. (Cliché F. Médail)
du patrimoine naturel (Csrpn) de Corse et au Conseil scientifique du parc naturel régional de Corse.
Par ses écrits et ses conférences qu’il donnait régulièrement en été, il a su sensibiliser la population locale, les touristes, les administrations et les politiques à la préservation de
ce patrimoine naturel unique, n’hésitant pas à fustiger certaines
pratiques délétères : « Dans un pays méditerranéen montagneux comme la Corse, aux fortes pentes, la gestion par le feu
est actuellement une aberration écologique qui conduit à une
érosion irréversible des sols, à la dominance des pyrophytes
et à une notable perte de valeur pastorale et de biodiversité »
(Gamisans, 2010, p. 327).
Jacques Gamisans avait à cœur de fournir au grand public
une information botanique de qualité et il publia plusieurs
ouvrages qui connaissent de grands succès éditoriaux : La Flore
endémique de la Corse (1996, 2014), Le Paysage végétal de la
Corse, publié en 2010 et la Flore des maquis de Corse et des
végétations associées, parue en 2014.
Durant près de cinquante années d’activités ininterrompues en Corse, Jacques Gamisans a réalisé une œuvre considérable qui a permis d’asseoir définitivement les connaissances
sur la flore vasculaire et la typologie de la végétation de l’île.
Sa modestie et sa discrétion étaient au service de sa passion pour la montagne méditerranéenne et pour ce fier pays
catalan dont il suivait avec bonheur les évolutions identitaires.
GAMISANS J. & JEANMONOD D.,
1993. Catalogue des plantes vasculaires de la Corse (ed. 2). In : Compléments au Prodrome de la flore
corse, Annexe 3. Éd. Conservatoire
& Jardin botaniques de Genève :
258 p.
GAMISANS J. & MARZOCCHI J.-F.,
1996. La Flore endémique de la
Corse. Édisud, Aix-en-Provence :
207 p. (seconde édition publiée en
2014 par Édisud, Saint Rémy-deProvence).
JEANMONOD D. & GAMISANS J.,
2007. Flora Corsica. Édisud, Aixen-Provence : 1 008 p. (seconde
édition publiée en 2013 dans le
Bull. Soc. Bot. Centre-Ouest,
numéro spécial 39 : 1 072 p.).
GAMISANS J., 2010. Le Paysage
végétal de la Corse. Albiana, Ajaccio : 343 p.
GAMISANS J., 2014. Flore des
maquis et des végétations associées
de Corse. Albiana, Ajaccio : 303 p.
La liste complète des publications
de Jacques Gamisans figure
sur son site internet :
http://jacquesgamisans.blogspot.fr
Frédéric Médail, 29 septembre 2015.
97
Fabrication :
04250 Turriers
www.transfaire.com
www.naturalia-publications.com
Achevé d’imprimer : février 2016
Éditrices en chef : Dr Élise Buisson
et Dr Brigitte Talon
UMR CNRS IRD IMBE
France
Instructions aux auteurs
ecologia mediterranea publie des articles de recherche originaux sur
des sujets se rapportant à l’écologie fondamentale ou appliquée des
régions méditerranéennes. La revue exclut les articles purement descriptifs ou de systématique. ecologia mediterranea privilégie les
domaines scientifiques suivants : bioclimatologie, biogéographie, biologie de la conservation, biologie marine, biologie des populations,
écologie des communautés, écologie forestière, écologie génétique,
écologie marine, écologie microbienne, écologie du paysage, écologie de la restauration, écologie végétale et animale, écophysiologie,
paléoclimatologie, paléoécologie. La revue accepte également des
articles de synthèse, des notes/communications courtes, des comptes
rendus d’ouvrages, des résumés de thèses, ainsi que des commentaires
sur les articles récemment parus dans ecologia mediterranea. La revue
publie aussi des actes de colloques faisant l’objet d’un numéro spécial. Dans ce cas, prendre contact avec les éditrices.
Comité éditorial
France
Western, Australie
France
France
France
Concepción, Chili
of Cyprus, Chypre
Les manuscrits sont soumis à des lecteurs spécialistes du sujet. La
décision finale d’accepter ou de refuser un article relève des éditrices.
L’article proposé doit être envoyé en version électronique à [email protected] (version doc(x) ou rtf). Pour la mise en
forme du document, voir les instructions qui suivent. Une fois leur
article accepté, les auteurs devront tenir compte des remarques des lecteurs, puis ils renverront leur texte corrigé sous deux mois toujours
sous format électronique (doc(x) ou rtf). Passé ce délai, la seconde
version sera considérée comme une nouvelle proposition.
TYPES DE MANUSCRIT
À préciser sur la première page lors de la soumission d’un manuscrit.
Article de recherche : contribution inédite découlant d’une étude complète. Ce type d’article fait typiquement une vingtaine de pages et environ 6 000 à 8 000 mots.
Note/communication courte : observation nouvelle ou rapport d’expérience dans un contexte pertinent avec les sujets visés par la revue.
Ce type d’article fait typiquement une dizaine de pages et environ
3 000 à 4 000 mots.
Article de synthèse : revue critique et originale de sujets spécifiques
d’actualité ou d’un champ de recherche de pointe dans le domaine de
l’écologie méditerranéenne. Ce type d’article fait typiquement une
vingtaine de pages et environ 6 000 à 8 000 mots.
Commentaire : avis sur des sujets déjà publiés dans ecologia mediterranea ou réflexion critique sur des problèmes d’intérêt général en
écologie méditerranéenne. Ce type d’article fait typiquement une à
cinq pages et environ 1 000 à 3 000 mots.
Compte rendu d’ouvrage : revue critique d’ouvrages (livres, monographies, manuels, etc.) dans le domaine de l’écologie méditerranéenne. Les auteurs d’ouvrages souhaitant voir un compte rendu publié
dans ecologia mediterranea doivent envoyer un exemplaire de l’ouvrage en question aux éditrices en chef.
Résumé de thèse : résumé d’une thèse soutenue récemment dans le
domaine de l’écologie méditerranéenne. Auteur, année, titre, spécialité et université (e.g. Thèse de doctorat en écologie soutenue le 3 avril
2012 à l’université de Padoue, Italie, laboratoire xxx), composition du
jury, mots clés, résumé de 1 000 mots maximum.
TEXTE
Les articles (dactylographiés en Times 12, double interligne, format
A4) doivent être rédigés en anglais ou en français. Si l’article soumis
n’est pas rédigé en anglais, il est demandé (en plus des résumés) une
version anglaise abrégée ainsi qu’une traduction en anglais des titres
des figures et tableaux. L’article doit être complet : type de manuscrit,
titres anglais et français, auteur(s) et adresse(s), résumés en anglais et
en français (au minimum), version anglaise abrégée (si le texte n’est
pas en anglais), mots clés anglais et français, texte, puis remerciements, bibliographie, liste des titres des figures et tableaux puis les
figures et tableaux (un(e)/page). Ainsi, pour la soumission du manuscrit, les illustrations seront intégrées au document et non envoyées
ISSN 0153-8756
univ-avignon.fr
ABONNEMENT
(contact : [email protected])
France
Europe
Monde
Abonnement à adresser à :
SARL Transfaire
Immeuble Wanad
F-04250 TURRIERS
Frais de port
Total
60 €
60 €
60 €
6€
12 €
16 €
66 €
72 €
76 €
Code banque
19106
Code guichet
00839
numéro de compte
13995626000
Domiciliation : CA SISTERON
IBAN : FR76 1910 6008 3913 9956 2600 062
BIC : AGRIFRPP891
PREMIÈRE PAGE
La première page contient :
1) le type de manuscrit visé (article de recherche, communication
courte, etc.) ;
2) le titre de l’article ;
3) le nom et prénom des auteurs ;
4) l’adresse de chaque auteur sera indiquée avec le courriel de l’auteur pour la correspondance. Dans le cas où la publication est le fait
de plusieurs auteurs, il doit être précisé lors du premier envoi la personne à qui doit être retourné l’article après lecture ;
5) le nombre total de mots de l’introduction à la bibliographie.
RÉSUMÉS, MOTS CLÉS ET VERSION ABRÉGÉE
Les résumés doivent comporter 300 mots au maximum et la version
abrégée (français si le manuscrit est en anglais et anglais si le manuscrit est en français) 1 000 mots (environ une page). Le nombre de mots
clés est limité à six, dans la langue des résumés ; ils ne doivent généralement pas figurer dans le titre.
BIBLIOGRAPHIE
La bibliographie regroupera toutes les références citées et elles seules.
Les références seront rangées dans l’ordre alphabétique des auteurs et
de façon chronologique. Les abréviations internationales des titres des
revues doivent être utilisées (ISI Journal Abbreviations Index). Vérifier attentivement le manuscrit pour s’assurer que toutes les références
citées dans le texte apparaissent bien en bibliographie et inversement.
Article
Andow D.A., Karieva P., Levin S.A. & Okubo A., 1990. Spread of
invading organisms. J. Ecol. 4: 177-188.
Ouvrage
Harper J.L., 1977. Population biology of plants. Academic Press, London, 300 p.
Chapitre d’ouvrage
May R.M., 1989. Levels of organisation in ecology. In: Cherret J.M. (ed.),
Ecological concepts. Blackwell Scientific Public, Oxford: 339-363.
Acte de conférence
Grootaert P., 1984. Biodiversity in insects, speciation and behaviour
in Diptera. In: Hoffmann M. & Van der Veken P. (eds.), Proceedings
of the symposium on “Biodiversity: study, exploration, conservation”.
Ghent, 18 November 1992: 121-141.
Rapport et thèse
Jaouadi W., 2011. Écologie et dynamique de régénération de l’Acacia tortilis (Forsk.) Hayne subsp. raddiana (Savi) Brenan var. raddiana
dans le parc national de Bouhedma (Tunisie). Thèse de doctorat de
l’Institut national agronomique de Tunisie, 180 p.
Editors-in-Chief: Dr Élise Buisson
& Dr Brigitte Talon
UMR CNRS IRD IMBE
France
ecologia mediterranea publishes original research reports and
syntheses in the fields of fundamental and applied ecology of
Mediterranean areas, except for descriptive articles or articles
about systematic. The editors of ecologia mediterranea invite
original contributions in the fields of: bioclimatology, biogeography, conservation biology, marine biology, population biology,
community ecology, forest ecology, marine ecology, genetic ecology, landscape ecology, microbial ecology, restoration ecology,
plant and animal ecology, ecophysiology, palaeoecology, palaeoclimatology. The journal also publishes reviews, short communications, book reviews, Ph. D. thesis abstracts and comments on
papers recently published in the journal. ecologia mediterranea
invite conference organizers to get in touch with the editors for
special issues as part of conference/symposium proceedings.
Manuscripts are peer-reviewed by appropriate referees. The final
decision to accept or reject the manuscript is made by the editors.
To submit a paper, please send an electronic version of your paper
to [email protected] (doc(x) or rtf). Please read the
following guidelines to prepare your manuscript. When the article is accepted, the authors should take reviewer’s comments into
consideration. They will send back to the journal Editorial Office,
within 2 months. After this deadline, the manuscript will be considered as a new submission.
Editorial Board
France
Western, Australie
France
France
France
Concepción, Chili
of Cyprus, Chypre
ISSN 0153-8756
Les mots « figures » et « tableaux » annoncés dans le texte sont écrits
en toutes lettres et en minuscules. Indiquer le nom d’auteur et l’année
de publication (mais indiquer tous les auteurs dans la bibliographie).
Exemples : “Since Dupont (1962) has shown that...”, or “This is in
agreement with previous results (Durand et al. 1990; Dupond &
Dupont 1997) ...”. Le numéro de page de la citation n’est mentionné
que dans le cas où elle est entre guillemets. Si la publication est écrite
par plus de deux auteurs, le nom du premier doit être suivi par et al.
univ-avignon.fr
L’usage d’un acronyme ou d’une abréviation technique doit être précédé de sa signification lors de sa première apparition. Les codes de
nomenclature doivent être respectés selon les conventions internationales. Les mots latins doivent être mis en italiques (et al., a priori,
etc.), et en particulier les noms de plantes ou d’animaux. Lors de la
première apparition du nom d’une espèce, il est demandé d’y faire
figurer le nom d’auteur (exemple : Olea europaea L.).
FIGURES ET TABLEAUX
clé RIB
62
Guidelines for authors
CITATIONS ET RENVOIS APPELÉS DANS LE TEXTE
ABRÉVIATIONS, NOMENCLATURE ET MOTS LATINS
1 an = 2 numéros
Abonnement
séparément. Le texte des articles de recherche doit comporter quatre
parties non numérotées : introduction, méthodes, résultats, discussion.
Les sous-titres ne sont pas numérotés. Par exemple :
Introduction
Méthodes
Site d’étude
Plan expérimental
Analyses des données
Résultats
Discussion
L’emploi de mots soulignés est à proscrire. Les noms d’auteurs cités
figureront en minuscules dans le texte comme dans la bibliographie.
En français, n’utilisez les majuscules que pour les noms propres, sauf
exception justifiée. Les ponctuations doubles ( : ; ? ! ) sont précédées
d’un espace, contrairement aux ponctuations simples ( , . ). En
revanche, toutes les ponctuations sont suivies d’un espace. La mise en
forme définitive du texte sera assurée par la revue.
Pour la soumission du manuscrit, les illustrations seront intégrées au
document, à la suite des références bibliographiques (voir la partie
« Texte »). Une fois l’article accepté, les illustrations doivent être
envoyées séparément du texte, prêtes à l’impression (résolution 300
dpi). Tous les documents devant être insérés doivent être annoncés dans
le texte, numérotés dans l’ordre d’apparition et légendés.
TIRÉS À PART
Il n’est plus fourni de tirés à part mais un pdf par article.
TYPES OF MANUSCRIPTS
Please, mention the type of manuscript you are submitting on the
first page of your submission.
Research article: research report of broad scope that is an original contribution to science. The typical length of research papers
is about 6,000 to 8,000 words.
Short communication: brief report of new observations or of
important findings that deserve publication before broader studies are completed. The typical length of short communications is
Reviews: critical appraisal of broad areas of investigation or research in Mediterranean ecology. The typical length of reviews is
Commentaries: opinion on topics recently published in ecologia
mediterranea or essays on topics of general interest in Mediterranean ecology. The typical length of short communications is
Book review: critical appraisal of a Book interesting to the readers of ecologia mediterranea. Editors who wish to see their book
reviewed in ecologia mediterranea should get in touch with the
editors of the journal.
Ph. D. thesis abstracts: abstract of the Ph. D. thesis recently
defended in the fields covered by ecologia mediterranea. Author,
Year, Title, University (e.g. Ph. D. thesis defended on April, 4th
2014 at University of California, Irwin, USA, Lab xxx), Composition of board, Keywords, Abstract (1,000 words max.).
MANUSCRIPT PREPARATION
Manuscripts (typewritten Times 12, with double line spacing)
must be written in English or in French. If the language is not
English, you should join an English short version and English
titles of figures and tables. The manuscript must be complete: e.g.
title in English and French, author(s) and address(es), abstract in
English and French, an English short version (if English is not the
language used in the article), keywords in English and French,
text, acknowledgements, references, figures and tables (one /
page). For the submission, figures and tables are integrated in the
SUBSCRIPTION
(contact: [email protected])
1 year = 2 issues
word document and not sent separately. For research papers, the
text should normally consist of 4 sections: introduction, methods,
results, discussion which are not numbered. Subtitles are not be
numbered either:
Introduction
Methods
Study site
Experimental design
Statistical analyses
Results
Discussion
Use lower-case letter type for names in the text and in the reference section. Do not underline any word. In English, there is one
blank after any punctuation, never before. Copy editing of manuscripts is performed by the journal.
FIRST PAGE
The first page contains:
1) the type of manuscript (research paper, short communication,
etc.);
2) the title of the article;
3) the names of the authors;
4) the address of each author + an email at least of the corresponding author;
5) the total number of words from the introduction to the references included.
ABSTRACTS, KEYWORDS AND SHORT VERSION
Abstracts should be no longer than 300 words. The English short
version should not exceed one page long (1,000 words). Do not
use more than six keywords. Keywords should not be present in
the title.
REFERENCES
All publications cited in the text should be presented in a list of
references following the text of the manuscript. The list of references should be arranged alphabetically on author’s names, and
chronologically for each author. You should abbreviate the titles
of periodicals mentioned in the list of references (except if you
are not sure of it; ISI Journal Abbreviations Index). Check the
manuscript to make sure that all references are cited and that all
citations in the text are included in the references. Use following
system to write the references:
Journal article
Andow D.A., Karieva P., Levin S.A. & Okubo A., 1990. Spread
of invading organisms. J. Ecol. 4: 177-188.
Book
Harper J.L., 1977. Population biology of plants. Academic Press,
London, 300 p.
Book chapters
May R.M., 1989. Levels of organisation in ecology. In: Cherret
J.M. (ed.), Ecological concepts. Blackwell Scientific Public., Oxford: 339-363.
Conference proceedings
Grootaert P., 1984. Biodiversity in insects, speciation and behaviour in Diptera. In: Hoffmann M. & Van der Veken P. (eds.), Proceedings of the symposium on “Biodiversity: study, exploration,
conservation”. Ghent, 18 November 1992: 121-141.
IN-TEXT CITATIONS
The words “figure” and “table” announced in-text should be written in extenso and with lower-case letter type. In the text refer to
the author’s name and year of publication (followed by pages only
if it is a quotation). If a publication is written by more than two
authors, the name of the first author should be used followed by
“et al.” (this indication, however, should never be used in the list
of references: first author and co-authors should be mentioned in
it). Examples: “Since Dupont (1962) has shown that…”, or “This
is in agreement with previous results (Durand et al. 1990; Dupond
& Dupont 1997)…”.
ABBREVIATES, NOMENCLATURE AND LATIN WORDS
Subscription
Postage
Total
60 €
60 €
60 €
6€
12 €
16 €
66 €
72 €
76 €
France
Europe
World
Explanation of a technical abbreviate is required when the first
use. International convention codes for nomenclature should be
used. Latin words should be in italic (et al., a priori, etc.), particularly for plants or animals’ denomination (the first time, please
precise author’s name: for example, Olea europaea L.).
FIGURES AND TABLES
Mail this subscription to:
SARL Transfaire
Immeuble Wanad
F-04250 TURRIERS
Bank: CA SISTERON
IBAN: FR76 1910 6008 3913 9956 2600 062
BIC: AGRIFRPP891
All illustrations should be included in the word document after
the section “References”. Once the paper is accepted for publication, illustrations should be sent separately (3,000 dpi). All the
illustrations should be cited and should have a legend.
REPRINTS
A pdf version will be supplied free of charge for each paper.
6 mm
(96 p)
Vol. 41 (2) – 2015
Vol. 41 (2) – 2015
Sommaire – Contents
Éditorial – Editorial
..........................................................................................
3
and Ceanothus L. (Rhamnaceae) Suggests Different Seed Predator Interactions
Developing Allometric Volume-Biomass Equations to Support Fuel Characterization
V. THOMAS PARKER
...............................................................................................
Studying Shoot and Root Architecture and Growth of Quercus ithaburensis
subsp. macrolepis Seedlings; a Key Factor for Successful Restoration
33
.......................................................
.........................................................
.......................
Vol. 41 (2) – 2015
5
15
B. D. PEDRA, J. GODOY PUERTAS, L. FUENTES LOPEZ
ecologia
mediterranea
25
Vegetation Dynamics of Coastal Dunes with Juniperus spp. in Crete, Gavdos
............
45
S. SOLIVERES, P. GARCÍA-PALACIOS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
61
......................................
73
............................................................................
85
..........................................................................................
Hommage à Jacques Gamisans
92
..........................................................................
95
P.DELIPETROU, D. GHOSN, G. KAZAKIS, P. NYKTAS, E. REMOUNDOU, I.N. VOGIATZAKIS
R. LIBOIS, R. GHALMI, A. BRAHIMI
ISSN 0153-8756
RESEARCH PAPERS
Editors-in-Chief: Dr Élise Buisson & Dr Brigitte Talon
Institut méditerranéen de biodiversité et écologie (IMBE)
Mediterranean Institute of Biodiversity and Ecology
Naturalia Publications

Ecologia Mediterranea - Université d`Avignon

Transcription

Documents pareils

vocabulaire des problemes environnementaux

Guillaume Greff

Threatened, Endangered and Extirpated Species Regulation, MR 25

CIESM Congress 1990, Perpignan.

Costa Mediterranea

boucle d`1h depuis le camping - Camping du Léman Saint Disdille

Les battures et les rivages - Fondation de la faune du Québec

this brochure - Grands Lacs de Champagne Forêt d`Orient

Inventaire pour la vente d`arbres 2016 Tree sale inventory 2016