Éditeur(s) : HAL CCSD Cambridge University Press (CUP)
Résumé : Devil's gardens are one of the most remarkable mutualistic associations between ants and plants. Myrmelachista ants eliminate all vegetation from around their host plants, resulting in wide forest clearings which have intrigued scientists from the start. Despite their noticeability, here we report the discovery of devil's gardens in remote highland cloud forests of the Eastern Amazon, more than 2000 km away from their nearest known analogues in Western Amazonia. We describe the ecological characteristics of these gardens and consider what factors could have produced the geographic isolation of Eastern Amazonian devil's gardens. Three hypotheses are investigated: (1) the host plant distribution restricts the distribution of the mutualism, (2) the ecological tolerances of Myrmelachista explain the isolation, and (3) the devil's gardens of the Eastern Amazon constitute relicts from ancient forest refugia. The distribution of the possible associated myrmecophytes and previously described ecological ranges of devil's gardens cannot explain their ecological restriction to cloud forests in Eastern Amazonia, but our discovery is consistent with the biogeographic refuge hypothesis (i.e. highlands along the Amazon Basin constitute refugia for humid forests that spread during the Cenozoic). Our finding opens exciting perspectives for comparative studies of the origin, ecology and evolutionary history of this ant-plant mutualism.
ISSN: 0266-4674

hal-01532640
https://hal.archives-ouvertes.fr/hal-01532640
DOI : 10.1017/S0266467416000195
PRODINRA : 384619

Éditeur(s) : HAL CCSD
Résumé : Land-use changes have deep consequences on species diversity, community structure and ecosystem functioning. Consequently, many works have tried to understand the effects of such changes on the diversity and functional properties of organisms. It has been demonstrated that species traits are oftenly more important than the identity of species per se in order to account understand the ecological roles of species. Besides, while the majority of assessments about land-use change effects on ecosystem functioning have focused on primary producers, the contribution of other organisms is essential to maintain the functioning and resilience of ecosystems.Ants are interesting organisms to track land-use changes due to their abundance, ubiquity and ecological dominance. Ants are present in most of terrestrial biomes and participate in several ecosystem processes through their consumption of food resources. They are also easy to sample and standard methods exist to provide accurate comparisons between studies. While a number of studies have demonstrated abrupt changes in ant species composition along environmental gradients, only a reduced number of such studies have tried to understand how ant community changes affect ecosystem processes. Moreover, the traits responsible for the stability of ant communities in response to land-use changes are little known. It is therefore necessary to develop a methodology that enables a proper identification of ant participation to different ecosystem processes and their contribution to ecosystems’ resistance and resilience.In this thesis I aim to bring some light about i) how land-use changes affect ant community structure? ii) what traits are responsible of the ecological success of a species or it’s extinction from a given environment? iii) how changes in the species or trait composition affect the participation intensity of ants in different ecosystem processes ?
Les perturbations d’origine anthropique sont à l’origine de l’érosion de la biodiversité avec des conséquences sur la structure des communautés et pour le fonctionnement des écosystèmes. La vitesse de ces changements a stimulé de nombreuses recherches portant sur les effets de ces modifications sur la biodiversité des organismes et leurs propriétés fonctionnelles. En effet il a été montré que les traits des espèces sont plus importants que leur identité elle-même pour tenir compte de leurs rôles écologiques. Même si la majorité de ces travaux ont étudié la réponse des producteurs primaires sur les écosystèmes terrestres, la contribution d’autres organismes est toutefois essentielle pour maintenir un fonctionnement et une résilience des écosystèmes corrects.Les fourmis sont des organismes cibles intéressants pour étudier de telles modifications du fait de leur abondance et de leur dominance écologique. En effet elles ont une distribution cosmopolite et participent à de nombreuses fonctions dans l’écosystème terrestre. Elles sont également faciles à étudier et des méthodes d’échantillonnage standard existent. Bien que de nombreuses recherches aient démontré des modifications dans la composition spécifique des fourmis dans des gradients environnementaux, un nombre réduit d’entre elles ont étudié la contribution de ces organismes sur le fonctionnement des écosystèmes. De plus, les traits responsables de la stabilité des communautés de fourmis en réponse à ces modifications sont encore peu connus. Il est donc nécessaire de développer une méthodologie capable d‘identifier la participation des fourmis dans des processus écosystémiques permettant également de quantifier leur activité afin de comprendre comment ces organismes agissent sur les capacités de résistance et résilience des écosystèmes.Ainsi, cette thèse vise mettre en lumière i) en quoi la simplification des habitats liée aux activités humaines a des répercussions sur les communautés de fourmis, ii) quels sont les traits responsables du succès écologique d’une espèce ou de son déclin au sein d’un environnement donné, et iii) comment ces variations en composition spécifique impactent l’intensité de participation des fourmis aux processus écosystémiques.
https://hal.archives-ouvertes.fr/tel-01389102

NNT : 2016YANE0004
tel-01389102
https://hal.archives-ouvertes.fr/tel-01389102
https://hal.archives-ouvertes.fr/tel-01389102v2/document
https://hal.archives-ouvertes.fr/tel-01389102/file/these.pdf

Éditeur(s) : HAL CCSD Springer Verlag
Résumé : The diversity and functional identity of organisms are known to be relevant to the maintenance of ecosystem processes but can be variable in different environments. Particularly, it is uncertain whether ecosystem processes are driven by complementary effects or by dominant groups of species. We investigated how community structure (i.e., the diversity and relative abundance of biological entities) explains the community-level contribution of Neotropical ant communities to different ecosystem processes in different environments. Ants were attracted with food resources representing six ant-mediated ecosystem processes in four environments: ground and vegetation strata in cropland and forest habitats. The exploitation frequencies of the baits were used to calculate the taxonomic and trophic structures of ant communities and their contribution to ecosystem processes considered individually or in combination (i.e., multifunctionality). We then investigated whether community structure variables could predict ecosystem processes and whether such relationships were affected by the environment. We found that forests presented a greater biodiversity and trophic complementarity and lower dominance than croplands, but this did not affect ecosystem processes. In contrast, trophic complementarity was greater on the ground than on vegetation and was followed by greater resource exploitation levels. Although ant participation in ecosystem processes can be predicted by means of trophic-based indices, we found that variations in community structure and performance in ecosystem processes were best explained by environment. We conclude that determining the extent to which the dominance and complementarity of communities affect ecosystem processes in different environments requires a better understanding of resource availability to different species.
ISSN: 0029-8549

hal-01606497
https://hal.archives-ouvertes.fr/hal-01606497
DOI : 10.1007/s00442-016-3741-z
PRODINRA : 407457