Near infrared spectroscopy (NIR) has become increasingly widespread throughout various fields as an alternative method for efficiently phenotyping crops and plants at rates unparalleled by conventional means. With growing reliability, the convergence of NIR spectroscopy and modern machine learning represent a promising methodology offering unprecedented access to rapid, high throughput phenotyping at negligible costs, representing prospects that excite agronomists and plant physiologists alike. However, as is true of all emergent methodologies, progressive refinement towards optimization exposes potential flaws and raises questions, one of which is the cornerstone of this study. Spectroscopic determination of plant functional traits utilizes plants' morphological and biochemical properties to make predictions, and has been validated at the community (inter-family) and individual crop (intraspecific) levels alike, yielding equally reliable predictions at both scales, yet what lies amid these poles on the spectrum of taxonomic scale remains unexplored territory. In this study, we replicated the protocol used in studies of the aforementioned taxonomic scale extremes and applied it to an intermediate scale. Interestingly, we found that predictive models built upon hyperspectral reflectance data collected across three genera of woody plants: Cornus, Rhododendron, and Salix, yielded inconsistent predictions of varying accuracy within and across taxa. Identifying the potential cause(s) underlying variability in predictive power at this intermediate taxonomic scale may reveal novel properties of the methodology, potentially permitting further optimization through careful consideration.

Comprendre l'influence de l'environnement sur la distribution des espèces végétales est une préoccupation à la base même de l'écologie végétale. L'objectif de cette thèse est de comprendre comment les communautés végétales de parcours des Grands Causses s'assemblent en fonction de la disponibilité en ressources édaphiques. Pour cela, la structure de niche des communautés a été caractérisée en termes de gestion des ressources, de stratégie de régénération, de niche temporelle et d'histoire évolutive des espèces le long d'un gradient édaphique. Par une approche basée sur les traits fonctionnels, nous avons mis en évidence (i) des processus de filtres, d'origine abiotique et biotique, qui restreignent localement la gamme de variation des traits et trient les espèces le long du gradient, ainsi que (ii) des patrons de divergence ou de convergence des traits au sein des communautés qui révèlent les conditions locales de coexistence des espèces. En milieux peu contraints et productifs, nous observons une convergence des stratégies d'utilisation des ressources, probablement en réponse à une forte compétition aérienne, qui est associée à une divergence des stratégies de reproduction et de régénération. A l'inverse, vers les milieux plus contraignants, une diversité de stratégies de gestion de la ressource coexiste, entrainant un maximum de diversité fonctionnelle en conditions de contraintes intermédiaires. Nous montrons par ailleurs une forte convergence phylogénétique dans ces parcours, associée à la dominance des espèces graminoïdes, qui s'atténue dans les milieux les plus contraints, où une diversité de lignées évolutives adaptées à la sécheresse coexiste. De plus, nous avons pu mettre en évidence que la dominance dans ces parcours est associée à des caractères fonctionnels généraux – tels qu'une teneur en matière sèche élevées, de grosses graines et une hauteur reproductive élevée – mais que celle-ci est ensuite modulée par les conditions édaphiques à une échelle plus fine via d'autres traits tels que la surface spécifique foliaire. Enfin, nous montrons que la phénologie des communautés joue un rôle essentiel dans l'assemblage de ces communautés le long du gradient, à la fois en réponse aux contraintes abiotiques saisonnières, particulièrement la précocité de la sécheresse édaphique, mais également aux interactions biotiques qui limitent le chevauchement des floraisons dans les milieux productifs. La combinaison des différentes approches fonctionnelle, phénologique et phylogénétique de la structure des communautés nous permet ainsi de proposer une vision intégrative des processus complexes d'assemblage des communautés dans ces parcours. / Understanding how the environment influences plant species distribution is a fundamental question in plant ecology. This work aims at understanding how soil resource availability influences plant community assembly and structure in Mediterranean rangelands of Southern France. To do this, the niche structure of plant communities has been described in terms resource use, regeneration strategy, phenology and evolutionary history along a soil resource gradient. Using a trait-based approach, we show that (1) filtering processes, both abiotic and biotic, may restrain trait ranges within communities and sort species along a gradient of soil resource availability, and (2) patterns of functional convergence and divergence among species within communities may reveal different processes of local species coexistence under different soil conditions. Within productive habitats, we found a strong convergence in resource use strategies, possibly resulting from strong aboveground competition, which was accompanied by a divergence in reproductive and regenerative strategies. By contrast, towards more constrained habitats, and despite a strong abiotic filter, a diversity of resource use strategies coexisted, creating a maximum of functional diversity at intermediate levels of constraints. Moreover, the strong phylogenetic convergence in these rangeland communities, mainly related to the dominance of graminoid species, diminished towards the more constrained soils where a diversity of drought-adapted lineages coexisted. In addition, we were able to relate dominance in these rangelands to a few general characters – namely high leaf dry matter content, large seeds and high reproductive heights – which were modulated at a finer spatial scale by local soil conditions influencing different criteria such as specific leaf area. Finally, we highlight the strong phenological response of communities to soil resource availability and the timing of summer drought. However, results also suggest a role of biotic factors, such as competition, in limiting flowering overlap among coexisting species in productive habitats. Combining functional, phenological and phylogenetic approaches allowed us to provide an integrative understanding of the complex processes driving community assembly in these rangelands.

No description available.

Différents modes de réhabilitation forestière des sites agricoles abandonnés peuvent affecter la diversité végétale des sites. L’analyse des traits fonctionnels des plantes pourrait permettre de révéler l’effet des différentes pratiques sylvicoles suggérées. L’étude porte sur deux communautés de friches arbustives ayant reçu la plantation de trois espèces d’arbres feuillus. La préparation des sites par un débroussaillement total ou par bande, combinés ou non d’herbicide offre l’occasion de mesurer l’effet de ces traitements sur la distribution des traits fonctionnels des communautés végétales après onze ans. Les résultats d’une analyse du 4e coin montrent un effet des traitements sur les traits fonctionnels des communautés et ce, davantage sur le site où la transmission de la lumière est supérieure. Un débroussaillement par bande permet un recul successionnel moins grand que total, avec la présence de plusieurs traits fonctionnels liés aux espèces de fin de succession tels que les phanérophytes et les espèces à semences de plus grande taille Un débroussaillement total résulte plutôt en une présence accrue des espèces exogènes et des intolérantes à la lumière. L’application d’herbicide influence peu la distribution des traits mais augmente la croissance du noyer noir lors de débroussaillement total et dans une moindre mesure lors de débroussaillement en bande. Le peu de différenciation significative de survie et de croissance en hauteur des arbres entre les traitements permet de proposer un débroussaillement par bande plutôt que total, afin de diminuer le recul successionnel, tout en nécessitant un moins grand recours à l’herbicide. / Different kinds of forest rehabilitation on abandoned farms could have an impact on plant diversity. Plant functional traits analyses could reveal effects of different management strategies. Two shrublands were planted with three hardwood species. They received the following treatments in 1998: total or striped clearing, with or without herbicide. The goal of this study is to determine the effects of these treatments, 11 years later, on functional traits distribution within the plant communities. According to a fourth corner analysis, the functional traits communities variation between treatments was significant, even higher, for the shrubland that had higher light availability. Striped clearing caused a less important successional retreat than total clearing, selecting for traits representative of shaded habitats like the presence of phanerophyte species and species with big seeds. Total clearing allowed a greater light availability in the following years, which selected for functional traits such as presence of exogenous and shade-intolerant species in the sampling units. Herbicide application had few effects on traits but increased black walnut growth when total clearing was applied, as well with strip clearing, but to a lesser extent. With low differentiation of tree growth and survival between treatments, we suggest using the striped clearing on planted shrublands in order to decrease the succession decline, while minimising herbicide use and their impacts on the surrounding plant communities.

Urbanisation is an ever-growing global phenomenon which creates altered environments characterised by increased human habitation, exotic species, impermeable surfaces, artificial structures, landscape fragmentation, habitat loss, and modified energy– and resource pathways. The vulnerable Rand Highveld Grassland vegetation unit in the Tlokwe Municipal area, South Africa, has been extensively degraded and transformed by urbanisation and agriculture. Only 1% of this endangered ecosystem is currently being actively conserved. Grassland fragments in urban areas are considered to be less species rich and less functional than their more “natural” counterparts, and are therefore not a priority for conservation. In this study the effects of landscape matrix quality on intra patch variables, namely plant species diversity and functional diversity, and fine-scale biogeochemical landscape function (as determined by Landscape Function Analysis or LFA) of 30 fragments of the Rand Highveld Grassland vegetation unit were explored. Four urbanisation measures (percentage urban land cover, percentage grass land cover, edge density, and density of people), acting as indicators for patterns and processes associated with urban areas, were calculated for matrix areas with a 500m radius surrounding each selected grassland fragment to quantify the position of each grassland remnant along an urban-to-rural gradient. Using the specific urbanisation measures, the grassland fragments were objectively classified into two classes of urbanisation, namely “rural/peri-urban” and “urban”, to allow for statistical comparisons between intra-patch variables for grassland remnants exposed to similar urbanisation pressures. Plant species composition and diversity were determined in the selected grassland fragments and nine functional traits were described for each species. Plant functional diversity was determined by five functional diversity indices, namely functional richness, evenness, divergence, dispersion, and specialisation. Fine-scale biogeochemical landscape function was determined by executing the LFA method. LFA assesses fine-scale landscape patchiness and 11 soil surface indicators to produce three main LFA parameters (stability, infiltration, and nutrient cycling), which indicates how well a system is functioning in terms of resource conservation and soil processes. Possible relationships between fine-scale biogeochemical landscape function and plant species- and functional diversity were also investigated. NMDS ordinations and basic statistics were used to determine trends and effects within the data. The results indicated that urban grassland remnants had lower mean plant species richness, Shannon species diversity (significantly), and Pielou species evenness than rural/peri-urban grassland fragments. Urban grassland fragments also contained significantly higher percentage of exotic species. Correlations were found between the four urbanisation measures and percentage species of the total species richness possessing certain functional attributes. This indicated that increased urbanisation may influence the species composition and the occurrence of certain plant traits in the selected grassland fragments. Urbanisation seems to have no effect on fine-scale landscape heterogeneity of the selected grassland fragments. Rural/peri-urban grassland fragments had higher infiltration capacity, nutrient cycling potential, and total SSA functionality (although not significantly), which may be ascribed to differences in management practices, such as mowing in urban areas and grazing in rural areas. Rand Highveld Grassland fragments in the urban landscape matrix of Potchefstroom city are just as conservable in terms of plant species diversity and functional diversity, as well as on a biophysical function level involving soil processes than rural/peri-urban grassland fragments. High plant species diversity and the presence of certain plant traits did not contribute to high soil surface stability, infiltration capacity, nutrient cycling potential and total soil surface functioning. / MSc (Environmental Sciences), North-West University, Potchefstroom Campus, 2013

Urbanisation is an ever-growing global phenomenon which creates altered environments characterised by increased human habitation, exotic species, impermeable surfaces, artificial structures, landscape fragmentation, habitat loss, and modified energy– and resource pathways. The vulnerable Rand Highveld Grassland vegetation unit in the Tlokwe Municipal area, South Africa, has been extensively degraded and transformed by urbanisation and agriculture. Only 1% of this endangered ecosystem is currently being actively conserved. Grassland fragments in urban areas are considered to be less species rich and less functional than their more “natural” counterparts, and are therefore not a priority for conservation. In this study the effects of landscape matrix quality on intra patch variables, namely plant species diversity and functional diversity, and fine-scale biogeochemical landscape function (as determined by Landscape Function Analysis or LFA) of 30 fragments of the Rand Highveld Grassland vegetation unit were explored. Four urbanisation measures (percentage urban land cover, percentage grass land cover, edge density, and density of people), acting as indicators for patterns and processes associated with urban areas, were calculated for matrix areas with a 500m radius surrounding each selected grassland fragment to quantify the position of each grassland remnant along an urban-to-rural gradient. Using the specific urbanisation measures, the grassland fragments were objectively classified into two classes of urbanisation, namely “rural/peri-urban” and “urban”, to allow for statistical comparisons between intra-patch variables for grassland remnants exposed to similar urbanisation pressures. Plant species composition and diversity were determined in the selected grassland fragments and nine functional traits were described for each species. Plant functional diversity was determined by five functional diversity indices, namely functional richness, evenness, divergence, dispersion, and specialisation. Fine-scale biogeochemical landscape function was determined by executing the LFA method. LFA assesses fine-scale landscape patchiness and 11 soil surface indicators to produce three main LFA parameters (stability, infiltration, and nutrient cycling), which indicates how well a system is functioning in terms of resource conservation and soil processes. Possible relationships between fine-scale biogeochemical landscape function and plant species- and functional diversity were also investigated. NMDS ordinations and basic statistics were used to determine trends and effects within the data. The results indicated that urban grassland remnants had lower mean plant species richness, Shannon species diversity (significantly), and Pielou species evenness than rural/peri-urban grassland fragments. Urban grassland fragments also contained significantly higher percentage of exotic species. Correlations were found between the four urbanisation measures and percentage species of the total species richness possessing certain functional attributes. This indicated that increased urbanisation may influence the species composition and the occurrence of certain plant traits in the selected grassland fragments. Urbanisation seems to have no effect on fine-scale landscape heterogeneity of the selected grassland fragments. Rural/peri-urban grassland fragments had higher infiltration capacity, nutrient cycling potential, and total SSA functionality (although not significantly), which may be ascribed to differences in management practices, such as mowing in urban areas and grazing in rural areas. Rand Highveld Grassland fragments in the urban landscape matrix of Potchefstroom city are just as conservable in terms of plant species diversity and functional diversity, as well as on a biophysical function level involving soil processes than rural/peri-urban grassland fragments. High plant species diversity and the presence of certain plant traits did not contribute to high soil surface stability, infiltration capacity, nutrient cycling potential and total soil surface functioning. / MSc (Environmental Sciences), North-West University, Potchefstroom Campus, 2013

Abstract Plant community composition and diversity are determined by the balance between rates of immigration and extinction. Processes of immigration to a local community, i.e. propagule availability and dispersal of propagules between and within habitats, set the upper limit for the pool of species potentially capable of coexisting in a community, while local biotic interactions, i.e., competition, facilitation, herbivory and interactions with below-ground ecosystem components, and environmental factors control colonisation and establishment, and determine the persistence and dynamics of already existing species. In this thesis, I studied (1) the interactions between propagule availability, biotic and environmental constraints on colonisation, and (2) the interdependence between biotic and environmental factors regulating community processes in already established resident vegetation. First, I found that both propagule availability and competition with adult plants limited the rates of colonisation and total community diversity in a relatively low-productive tundra ecosystem. Long-term exclusion of mammalian herbivores and alleviation of nutrient limitation by fertilization increased the intensity of competition with established vegetation, and diminished immigration rates. In addition, I also found that community openness to colonization depended on the initial community properties, i.e., the functional composition and the traits of dominant plants in resident vegetation, which mediate the effects of nutrient addition and biomass removal on immigration rates. Second, adult plants in the resident vegetation experienced an increased extent of neighbourhood competition and herbivory in nutrient enriched conditions and in naturally more fertile habitats. However, the effects were also species-specific. On a community level, release from heavy grazing favoured lichens over graminoids and increased species richness. Furthermore, I also showed that plant community composition was strongly linked with soil organic matter quality and microbial community composition, and that these vegetation-soil-microbe interactions varied along a gradient of soil pH. Overall, my results emphasise that propagule availability, biotic and environmental control over community processes are strongly interconnected in tundra ecosystems. Especially, my findings highlight the role of plant competition and herbivory and their dependence on soil nutrient availability in governing colonisation and resident community dynamics. My results also indicate that plant functional composition and traits of dominant plants are of great importance in channelling community responses to external alterations and dictating plant-soil interactions.

Abstract Air temperatures in high-latitude regions are anticipated to rise by several degrees by the end of the century and result in substantial northward shifts of species. These changes will likely affect the source and sink dynamics of greenhouse gases and possibly lead to a net carbon release from high-latitude soils to the atmosphere. However, regional differences in carbon cycling depend highly on the vegetation community composition, which may be controlled by the abundance of herbivores. I investigated whether mammalian herbivores, mainly reindeer and rodents, alter ecosystem carbon storage through their impacts on vegetation and on dominant plant functional traits. I combined observations of recent changes in ecosystem carbon with experimental field manipulations of both herbivory and climate change and measured carbon storage in vegetation and soil, the uptake and release of carbon dioxide, microbial activity and compared these to plant community composition. Results of my PhD thesis show that under ambient conditions, the impacts of herbivory on both above- and belowground carbon storage ranged from positive to negative. Herbivory altered dominant plant functional traits and these were fairly good predictors of the changes in soil carbon. When combined with experimental warming, herbivory continued to exert control on the dominant plant functional traits but the strong effects of warming on ecosystem carbon storage mostly concealed the impact of herbivory. Interestingly, herbivory–nutrient interactions that were not linked to dominant functional traits determined the consequences of warming on soil carbon. Taken together, I show clear and site-specific impacts of herbivores on vegetation and ecosystem carbon storage and the processes that govern them. Therefore, I suggest that an improved understanding of the role of herbivory in the global carbon cycle could improve estimations of global carbon–climate feedbacks. / Tiivistelmä Vuosisadan loppuun mennessä arktisten alueiden lämpötilan odotetaan nousevan usealla asteella ja johtavan lajien siirtymiseen yhä pohjoisemmaksi. Nämä muutokset todennäköisesti muuttavat pohjoisten ekosysteemien kykyä vapauttaa ja sitoa ilmakehän hiiltä ja saattavat johtaa siihen, että yhä enemmän hiiltä vapautuu tundramailta ilmakehään. Kuitenkin paikallisesti hiilenkierto on riippuvainen kasviyhteisöstä ja erityisesti kasvien funktionaalisista ominaisuuksista. Väitöskirjassani tutkin, voivatko herbivorit, pääasiassa porot sekä jyrsijät, muokata hiilenkiertoa muuttamalla kasvillisuutta. Tutkimuksissani seurasin kuinka alueen laidunnushistoria on muokannut hiilivarastoja ja hiilenkiertoa tällä hetkellä ja pyrin arvioimaan herbivorien vaikutusta lämpenevässä ilmastossa kokeiden avulla, joissa manipuloidaan sekä herbivoriaa että lämpötilaa tai ravinteiden saatavuutta. Tulokseni perustuvat arvioihin hiilen varastoista, hiilidioksidin vapautumisesta ja sitoutumisesta sekä mikrobien aktiivisuudesta, joita vertaan kasviyhteisöön. Tulokseni osoittavat, että herbivoria voi joko lisätä tai vähentää ekosysteemin hiilivarastoja sekä maan päällä että maan alla. Muutokset hiilivarastoissa selittyivät varsin hyvin herbivorien tuottamilla kasvillisuusmuutoksilla ja valtalajien funktionaalisilla ominaisuuksilla. Herbivoria muokkasi kasviyhteisöä myös kokeellisen lämmityksen yhteydessä, mutta lämmityksen välittömät vaikutukset hiilivarastoihin peittivät suureksi osaksi alleen herbivorian vaikutukset. Kuitenkin herbivorian ja lannoituksen kasvillisuusmuutoksista riippumattomat yhdysvaikutukset määrittivät lämpenemisen seuraukset maan hiileen. Kaiken kaikkiaan, tutkimukseni osoittaa, että herbivorit voivat paikkakohtaisesti muokata kasvillisuutta, ekosysteemin hiilivarastoja sekä hiilenkierron prosesseja. Näiden tulosten myötä ehdotan, että parempi ymmärrys herbivorian vaikutuksista maailmanlaajuisesti voisi parantaa nykyisiä ennusteita siitä, kuinka ilmaston lämpeneminen muuttaa hiilenkiertoa.

No description available.

Différents modes de réhabilitation forestière des sites agricoles abandonnés peuvent affecter la diversité végétale des sites. L’analyse des traits fonctionnels des plantes pourrait permettre de révéler l’effet des différentes pratiques sylvicoles suggérées. L’étude porte sur deux communautés de friches arbustives ayant reçu la plantation de trois espèces d’arbres feuillus. La préparation des sites par un débroussaillement total ou par bande, combinés ou non d’herbicide offre l’occasion de mesurer l’effet de ces traitements sur la distribution des traits fonctionnels des communautés végétales après onze ans. Les résultats d’une analyse du 4e coin montrent un effet des traitements sur les traits fonctionnels des communautés et ce, davantage sur le site où la transmission de la lumière est supérieure. Un débroussaillement par bande permet un recul successionnel moins grand que total, avec la présence de plusieurs traits fonctionnels liés aux espèces de fin de succession tels que les phanérophytes et les espèces à semences de plus grande taille Un débroussaillement total résulte plutôt en une présence accrue des espèces exogènes et des intolérantes à la lumière. L’application d’herbicide influence peu la distribution des traits mais augmente la croissance du noyer noir lors de débroussaillement total et dans une moindre mesure lors de débroussaillement en bande. Le peu de différenciation significative de survie et de croissance en hauteur des arbres entre les traitements permet de proposer un débroussaillement par bande plutôt que total, afin de diminuer le recul successionnel, tout en nécessitant un moins grand recours à l’herbicide. / Different kinds of forest rehabilitation on abandoned farms could have an impact on plant diversity. Plant functional traits analyses could reveal effects of different management strategies. Two shrublands were planted with three hardwood species. They received the following treatments in 1998: total or striped clearing, with or without herbicide. The goal of this study is to determine the effects of these treatments, 11 years later, on functional traits distribution within the plant communities. According to a fourth corner analysis, the functional traits communities variation between treatments was significant, even higher, for the shrubland that had higher light availability. Striped clearing caused a less important successional retreat than total clearing, selecting for traits representative of shaded habitats like the presence of phanerophyte species and species with big seeds. Total clearing allowed a greater light availability in the following years, which selected for functional traits such as presence of exogenous and shade-intolerant species in the sampling units. Herbicide application had few effects on traits but increased black walnut growth when total clearing was applied, as well with strip clearing, but to a lesser extent. With low differentiation of tree growth and survival between treatments, we suggest using the striped clearing on planted shrublands in order to decrease the succession decline, while minimising herbicide use and their impacts on the surrounding plant communities.