Human activities are altering species' environments, consequently driving many to extinction and changing biotic communities worldwide. Functional traits are species’ intrinsic characteristics that shape their roles in an environment. Loss of functional diversity compromises ecosystem processes and potentially the resilience of communities facing further change. Biological homogenization reflects the non-random loss of species and possibly also of trait distributions within community, leading to the increasing ubiquity of some traits and growing rarity of others. By changing the composition and configuration of species’ local habitat and their regional surroundings, land use can alter community dynamics. The extent to which land uses within habitats and across the surrounding landscape matrix alter the distribution of functional traits in biotic communities remains highly uncertain but could determine how to design management strategies intended to aid conservation. Here, I investigate the relative contributions of compositional and configurational landscape characteristics at local and regional scales on the diversity and composition of functional traits within butterfly communities. I constructed models to identify landscape predictors of functional trait diversity but found no significant associations with individual traits. Managing habitat quality by fostering compositional and configurational heterogeneity in the local landscape can improve functional diversity. However, efforts to facilitate conservation of species with rare traits necessitates maintaining compositional and configurational variety within habitats and in the surrounding matrix.

No description available.

Environmental gradients structure ichthyofaunal communities longitudinally along river networks via the selective filtering of species’ traits. In many instances, these environmental influences have created distinct zones of co-occurring fish species. Zonation studies have most often been conducted with taxonomic data (species x site matrices), but the increasing availability of functional trait data creates an opportunity to build more rigorous understanding of species’ co-occurrence patterns. Notably, zonation studies that use taxonomic data may not reveal the same patterns as studies based on trait data. In this study, we tested for distinct ichthyofaunal zonation in James River Basin, VA using a combination of historical (1950-1987) and contemporary fish occurrence records (1986-2016) that were aggregated within 12-digit hydrologic units (HU). Zonation tests were performed separately for taxonomic data and functional trait data, using a combination of non-metric multidimensional scaling and k-means cluster analysis. We detected three distinct taxonomic zones and three functional trait zones within the James River Basin. In addition, through identifying that taxonomic dissimilarity between HUs was strongly correlated with functional dissimilarity, these zonation patterns were determined to not be significantly different.

A l’heure où l’on cherche à optimiser les services éco-systémiques fournis par les prairies, une meilleure compréhension du fonctionnement des systèmes racinaires apparait comme nécessaire. Nous avons donc décidé de caractériser les stratégies d’espèces issues des familles des Fabaceae et Poaceae prairiales par la mesure de traits fonctionnels racinaires. Le principal résultat de ce travail est la mise en évidence de plusieurs axes majeurs de différentiation des stratégies racinaires et de leurs relations avec les exigences écologiques des espèces. Le plus important est le compromis entre stratégies de capture et de conservation des ressources. Les espèces à stratégie de capture sont mieux adaptées aux milieux peu stressants notamment pour l’eau, mais sont peu impactées par le manque de phosphore, ce sont aussi des compétitrices efficaces. Bien que les plantes à stratégie de conservation soient mieux à même de se développer dans des milieux continentaux, elles sont plus impactées que les autres par la présence de compétiteurs et le manque de phosphore. Le type de relations existant entre les stratégies aériennes et racinaires apparait aussi comme un marqueur de l’adaptation des plantes aux stress. L’étude du système racinaire est donc un moyen efficace de prédire le comportement des espèces de plantes prairiales face à des contraintes abiotiques et biotiques. Cela ouvre des perspectives particulièrement intéressantes notamment pour la construction de mélanges d’espèces mieux adaptées à leurs environnements. / In order to optimise grassland ecosystem services we need to improve our understanding of root system functioning. As a result, we decided to characterise strategies of species coming from grassland’ Fabaceae and Poaceae families, by root functional traits measurement. The main result of this work is the establishment of several axes of root strategies differentiation. The main axis is the trade-off between resources capture and conservation strategies. Species with capture strategies appear to be adapted to non-water stressful habitat and are barely impacted by phosphorus shortage; they also happen to be strong competitors. On the opposite, species with conservation strategies are adapted to water stressful continental climates, but are strongly impacted by phosphorus shortage and happen to be weak competitors. The kind of relation between above-ground and below-ground strategies also appears to be a marker of the plants adaptation to stress. We showed that the root system study is a good way to predict grassland species comportment to face abiotic and biotic constraints. The present work widens interesting perspectives for the sowing of mixed grassland species better adapted to their environments.

京都大学 / 0048 / 新制・課程博士 / 博士(農学) / 甲第20586号 / 農博第2238号 / 新制||農||1052(附属図書館) / 学位論文||H29||N5075(農学部図書室) / 京都大学大学院農学研究科森林科学専攻 / (主査)教授 神﨑 護, 教授 北島 薫, 教授 井鷺 裕司 / 学位規則第4条第1項該当 / Doctor of Agricultural Science / Kyoto University / DGAM

A robust ecosystem requires a functionally heterogeneous community of organisms with a wide range of traits that permits broad resource partitioning. Understanding community diversity patterns can help investigate drivers of community assembly and how different metrics reflect the success of restoration in grassland or weed control in cropland. The objectives of my study were to identify assembly drivers influencing community taxonomic, phylogenetic, and functional diversity. The first study examined the effect of different ecotypes of dominant species on grassland community diversity along a spatial aridity gradient during restoration. This study showed that ecotype significantly affected species richness and shaped taxonomic, phylogenetic, and functional diversity. Accordingly, restorations should consider ecotypic variation as a critical biological filter to community assembly in grassland ecosystems. Given the community response across the tallgrass prairie, restoration practitioners should draw attention to regional sources of dominant species because ideal ecotypic sources can affect species richness and even sustain the phylogenetic and functional patterns. The restoration efficacy of existing ecosystems should be evaluated to large-scale diversity patterns to detect gaps and limitations that will reveal which diversity components to highlight for further restoration investment. The second study investigated temporal variations in diversity metrics among dominant species ecotypes and a drought treatment on the importance of external and internal filters in shaping the assembly of grassland communities. In this study, species richness decreased significantly during the early stages of restoration while phylogenetic and functional diversity was maintained over eight years. I also found no significant effect of experimental drought treatment on community biodiversity. Moreover, ecotypic variation as an internal filter played a key role in grassland assembly but the external filter was less strong because of high trait overlap among species within a community. In general, this study highlighted the consequence of integrating both interspecific and intraspecific trait variabilities and the value of concentrating on functional traits to comprehend better how trait variability is coupled with species coexistence. Future investigations are necessary to examine the distinctive origins of variability in plant traits and how they contribute to grassland community assembly. The third study focused on whether weed management tactics such as a glyphosate-resistant (GR) cropping system were beneficial to control weed diversity in the soil seedbank of the agroecosystem. Both phylogenetic and functional diversity of the soil seedbank was relatively stable with different GR cropping systems across six years. The neutral assembly of the soil seedbank may imply that the belowground weed community could be restrained by stochastic mechanisms, such as dispersal and demographic stochasticity of seeds, during agricultural activities. Therefore, I recommend integrated weed management with a sustainable perspective to fight against the evolutionary feedback due to weed herbicide resistance. While the GR cropping system still seems beneficial, future weed control cannot overlook the extensive impact of GR systems on biodiversity variations, the shifts in weed composition, and the resistance evolution of weed species to herbicides in the agroecosystem.

The trait based approach has been proposed as a way to reconcile community ecology. Despite recent advances in trait based ecology, such as the development of global trait databases and standardized methodology for trait collections, it remains unclear to what degree traits vary across individuals, species, and communities. In addition, the drivers of trait variation may shed light on the underlying processes that maintain species diversity and community assembly at local to continental scales yet these have been poorly studied. In this study, I examine both the magnitude of trait variation as well as the patterns of trait variation at local to continental scales in order to understand the drivers of diversity patterns across environmental gradients. First, I quantified the magnitude of trait variation at local scales in a dry tropical forest and determined that intraspecific variation is not negligible and can be quite large for compound-leaved species. However, I showed that the sample sizes necessary for quantifying trait variation are tractable and should encourage the adoption of trait variation in trait based ecology. Second, I tested whether climatic variables are predominantly responsible for observed trait variation across dry tropical forests in the Americas. I showed that climatic variability, specifically variability in precipitation, explained a large degree of observed trait variation across dry tropical forests and may provide a unique approach for classifying dry tropical forests based on their inherent degree of climatic seasonality. Third, I quantified patterns of trait variation at continental scales across elevational gradients at high to low latitudes. I showed that climatic variables largely drive patterns of trait variation at high latitudes while biotic factors largely drive patterns of trait variation at low, tropical latitudes. This finding has implications for understanding large-scale patterns of species diversity across elevational and latitudinal gradients. Finally, I apply trait variation to life history theory by quantifying variation in two life history traits (growth and reproduction) in a tropical tree species using a legacy dataset. I showed that variation in these two life history traits is due to both resource availability and allometric related effects on both traits. In sum, this study advances our understanding of the magnitude and underlying drivers of trait variation at local to continental scales.

This thesis deals with the effects of changing land use, following landscape abandonment, on butterfly communities in the Mediterranean Basin. It consists of three case studies. The first focuses on the effects of forest encroachment on butterflies in the Southern Balkans; the second studies butterfly communities in Portuguese 'montados' and the third explores demography and life histories of three co-occurring Papilionidae butterfly species (Archon apollinus, Zerynthia polyxena and Zerynthia cerisy) in Greek Thrace. The results describe shifts in butterfly communities, detectable even at the level of individual species life history traits, with increasing forest encroachment. The preference of range-restricted Mediterranean endemics for either grasslands or open woodland formations contributes to falsifying the forested Mediterranean hypothesis, favouring a hypothesis of finely grained landscape mosaic instead. This mosaic is currently threatened by land use change and biodiversity homogenisation. Maintaining habitat and landscape heterogeneity is crucial for conserving the Mediterranean biodiversity hot-spot.

Submitted by Bruna Rodrigues (bruna92rodrigues@yahoo.com.br) on 2016-10-03T12:26:59Z No. of bitstreams: 1 DissNA.pdf: 1582958 bytes, checksum: 96d6a6328b468f669e8215074d435b9f (MD5) / Approved for entry into archive by Marina Freitas (marinapf@ufscar.br) on 2016-10-10T18:52:06Z (GMT) No. of bitstreams: 1 DissNA.pdf: 1582958 bytes, checksum: 96d6a6328b468f669e8215074d435b9f (MD5) / Approved for entry into archive by Marina Freitas (marinapf@ufscar.br) on 2016-10-10T18:52:15Z (GMT) No. of bitstreams: 1 DissNA.pdf: 1582958 bytes, checksum: 96d6a6328b468f669e8215074d435b9f (MD5) / Made available in DSpace on 2016-10-10T18:52:24Z (GMT). No. of bitstreams: 1 DissNA.pdf: 1582958 bytes, checksum: 96d6a6328b468f669e8215074d435b9f (MD5) Previous issue date: 2016-03-08 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / The so-called "Plant Ecology Strategy Scheme" groups species into categories along the spectrum of their functional traits. When two or more important functional traits are correlated between concurrent species, they define a strategic dimension. There are four main dimensions of ecological strategies for establishing the relationships among functional traits of plants: maximum plant height, size of branches and leaves, leaf structure and physiology, and seed production. Traits that may define these dimensions include: maximum plant height, wood density, leaf area, specific leaf area, fruit size, and seed size. We analysed the relationships among the traits in cerrado woody species and compared the results with those found in tropical forests. In this dissertation, we presented our results in a single chapter, written in english. We measured these traits from 28 woody species at the Vaçununga State Park, southeastern Brazil. We found two significant relationships: between fruit size and seed size and between fruit size and leaf area. We also identified two ecological strategies: species with high specific leaf area and species with low specific leaf area. Cerrado and forest woody species seem to have distinct ecological strategies, probably due to the different environmental pressures to which they are submitted. / Os chamados “Esquemas de Estratégias Ecológicas Vegetais” agrupam as espécies em categorias ao longo do espectro de seus traços funcionais. Quando dois ou mais traços funcionais importantes estão correlacionados entre espécies coocorrentes, eles definem uma dimensão estratégica. Há quatro principais dimensões de estratégias ecológicas para estabelecer relações entre os traços: altura máxima da planta, tamanho dos ramos e das folhas, estrutura e fisiologia foliar, e produção de sementes. Os traços que podem definir essas dimensões são: altura máxima da planta, densidade da madeira, área foliar, área foliar específica, tamanho do fruto, e tamanho da semente. Nessa dissertação, apresentamos os resultados em um único capítulo, escrito em inglês. Coletamos os seis traços para 28 espécies arbustivo-arbóreas no Parque Estadual de Vaçununga, sudeste do Brasil. Analisamos as relações entre esses traços para espécies lenhosas de cerrado e comparamos nossos resultados com os encontrados para florestas tropicais. Encontramos duas correlações significativas: entre tamanho do fruto e tamanho da semente e entre tamanho do fruto e área foliar. Identificamos, também, duas estratégias ecológicas: espécies com baixas áreas foliares específicas, e espécies altas áreas foliares específicas. Espécies lenhosas de cerrado e floretas parecem ter estratégia ecologias distintas, provavelmente pelas diferentes pressões a que estão submetidos.

No description available.