La version actuelle du modèle de surface continentale ISBA représente la couche de sol superficiel et la végétation comme un composite dont les propriétés physiques résultent de pondérations entre les propriétés du sol et de la végétation (albédo, rugosité par exemple). Une nouvelle version, ISBA-MEB (Multi-Energy-Balance), représente le sol et la végétation au travers de deux bilans séparés. Ce développement permet notamment une approche plus réaliste de la modélisation des flux au-dessus, dans et sous de la canopée avec un traitement propre à chacune des couches et la prise en compte de nouveaux processus (effet d'ombrage de la canopée sur le sol, transfert radiatif à travers la végétation, interception de la neige par la canopée). Le sol superficiel est lui désormais caractérisé, pour les surfaces forestières, par une couche de litière agissant sur le cycle énergétique comme sur le cycle hydrologique. Pour évaluer l'impact de ces nouveaux développements sur les forêts, quatre sites Français instrumentés sont dans un premier temps utilisés. Le large panel de données disponible permet une évaluation précise permettant de mettre en évidence certaines améliorations apportées par les développements, En particulier sur le bilan d'énergie en réduisant l'amplitude du flux de chaleur par conduction dans le sol et en réinjectant tout ou partie de cette énergie dans le flux sensible, mais aussi sur le bilan d'eau en améliorant le partitionnement de l'évapotranspiration en ses différentes composantes. Ensuite, leréseau Fluxnet, standard international, se révèle être outil idéal confirmer ces améliorations. Celui ci contient en effet des données relatives à des sites instrumentés correspondant à un large panel de climats et de types de végétation indispensable à la validation d'un modèle destiné à des simulations à échelle globale. 42 sites forestiers sont retenus pour procéder à l'évaluation des nouveaux développements sur la base d'un critère de fermeture du bilan énergétique. Lesrésultats de cette étude permettent non seulement de confirmer à grande échelle les résultats de la partie précédente mais aussi de caractériser des améliorations plus marquées sur les forêts de conifères ainsi que les sites enneigés. Enfin, l'impact sur le bilan hydrologique d'ISBA-MEB est évalué par l'utilisation de la chaîne opérationnelle hydrométéorologique SIM (SAFRAN-ISBAMODCOU).Cette chaîne permet de modéliser le débit des rivières françaises sur une longue période (jusqu'à un demi siècle) et par extension d'obtenir une estimation de la climatologie spatiallement distribuée de l'évapotranspiration. L'impact des développements, et nottament laprise en compte de la litière dans le modèle, modifient l'amplitude annuelle moyenne de l'évaporation du sol et impacte ainsi la modélisation des débits des rvières simulée. D'autre part, une augmentation significative de la sublimation de la neige est à l'origine dans les bassins de montagne d'une réduction du ruisselement total mais ceux-ci ne représente qu'une faible fraction du domaine d'étude. / The current version of ISBA land surface model represents the superficial layer of soil and vegetation as a composite with physical properties modeled as a weighted combination between soil and vegetation properties (e.g. albedo, roughness). A new version, ISBA-MEB (Multi-Energy-Balance), represents the soil and vegetation energy budgets separately. This development consists in a more realistic modeling of the flux exchanges above, within and below the canopy with an explicit treatment of each energy budget and the associated processes (canopy shading of the forest floor, canopy radiative transfer, canopy snow interception and loading, etc.). In addition, the superficial soil layer can be characterized as being comprised of forest litter thereby modulating both thermal and hydrological processes. In order to assess the impact of these developments, first, data from four French well-instrumented sites were used. The wide range of available data permitted a thorough assessment of the impact of the improvements. In particular, the reduced amplitude of the surface ground conduction flux has a profound impact on the energy budget by transferring the resulting residual energy into the sensible heat flux. The water balance was also mainly impacted by improving the partitioning of evapotranspiration into its various components and making it depend more directly on the temporal evolution of the vegetation properties. Next, the Fluxnet network database, which is an international standard for model evaluation, proved to be ideal for confirming these improvements since it is comprised of data from instrumented sites for a wide range of climates and vegetation types. A sub-set of 42 forest sites were selected for assessing the new developments using classical energy balance closure-based criteria. The results of this study not only confirmed the results from the first part of this study for the three French sites, but it also helped characterize the biggest improvements which occurred for moderately open canopies, the pine forest class, and sites with significant snow cover. Finally, the impact of MEB on the hydrological cycle is evaluated by using the SIM operational hydrometeorological chain (SAFRAN-ISBA-MODCOU). This system is used to model and evaluate river discharge over all of France over long periods of time (up to half a century), and by extension, obtain spatially distributed climatological evapotranspiration estimates. The impact of the new developments, especially the combination of the canopy and the explicit forest litter treatment, modified the temporal evolution of bare soil evaporation which had a notable impact on river flow. Finally, significantly increased snow sublimation reduced both total runoff and evapotranspiration in mountain catchments (which comprised a relatively small part of the studied domain).

A varved sediment sequence spanning the past ~11.0 ka was collected from Lac Noir (45.77N, 75.13W, 168 m a.s.l.) in southwestern Québec. A high-resolution pollen record documents the post-glacial vegetation history of the region over the course of the Holocene. The record shows an initial open spruce woodland, the establishment of the boreal and mixed conifer-hardwood forest into the area, as well as the expansion and contractions of tree populations in response to climate variability during the Holocene. The well known Tsuga decline at Lac Noir lasted 500 years starting at 5.5 ka and it took 1 460 years for hemlock to recover. The highest frequency of fire activity occurred during the early Holocene, and the lowest in the mid-Holocene. The late Holocene saw an increase in fire frequency, which could be attributed to a drier climate in eastern North America during this period. The impact of climate variability, fire disturbances and possible biotic factors on the Lac Noir vegetation are examined.

n/a

Thesis (Ph. D.)--Oregon State University, 1964. / Typescript. Mounted photographs. Includes bibliographical references (leaves 99-109). Also available on the World Wide Web.

The purpose of this study was to document vegetation on "The Island", a Research Natural Area at the confluence of the Crooked River and the Deschutes River in central Oregon's Juniperus occidentalis Zone and to compare the results with an earlier study reported in 1964 from 1960-'61 data. Present-day comparisons were also made between "The Island" vegetation and three nearby sites. Percent cover and constancy of major tree, shrub, grass, and forb species were considered along with percent cover of litter, moss/lichen, rock and bare ground. Climatic data from the Metolius, OR Station were examined, and the literature of succession especially succession in the juniper and sagebrush steppe of the Great Basin was reviewed. Data from 1992-'93 show more woody vegetation on "The Island", both tree and shrub, than was measured thirty years ago. The only tree species present is Juniperus occidentalis, while major shrub species are Artemisia tridentata and Purshia tridentata. Grass cover appeared to be less, with a more even mix of the native perennial bunchgrass species Agropyron spicatum, Festuca idahoensis, Poa sandbergii, and Stipa thurberiana, than in the past when Agropyron spicatum and the alien annual grass, Bromus tectorum dominated. The two plant associations identified in the 1964 report by Driscoll, Juniperus occidentalis / Artemisia tridentata / Agropyron spicatum and Juniperus occidentalis / Purshia tridentata / Agropyron spicatum were still identifiable, but the shrub, Artemisia tridentata appeared to be entering areas where Purshia tridentata had dominated in the past study. The present-day comparison sites showed many similarities with sites on "The Island." The comparison sites in the Juniperus occidentalis / Artemisia tridentata / Agropyron spicatum association measured slightly more tree and shrub cover but similar grass cover when compared to "The Island." The comparison site in the Juniperus occidentalis / Purshia tridentata / Agropyron spicatum association had more tree cover and more Purshia tridentata cover, but less shrub cover generally and more grass cover than the same association on "The Island." Forbs represented less than one percent cover on all study sites. The differences recorded in 1992-'93 from that of the study thirty years ago may reflect successional processes at work and a lack of any major natural fires in the system. / Graduation date: 1996

Thesis (M.S.)--University of Washington, 2006. / Includes bibliographical references (leaves 52-57). Issued also electronically via World Wide Web in PDF format.

Graduation date: 1992

The wetlands of the semi-arid savanna ecosystem of Amboseli are the critical dry-season range for a diverse wildlife population, as well as for livestock and humans. Changes in the migratory patterns and increases in the population of the elephants in the ecosystem, as well as a shift in the lifestyle of the growing human population from predominantly nomadic pastoralist society to a sedentary agrarian community, has had far-reaching effects on the ecosystem and the wetlands. In this thesis, the current status of the wetlands and the changes in vegetation over the last four decades were examined using satellite imagery and long-term census data. Studies were also conducted to determine the impact of grazing by elephants and other large mammals on wetland vegetation, and the flooding patterns and extents of some of the wetlands were examined. <br /><br /> Long-term aerial count data indicated that the wetlands that are currently protected by the Amboseli National Park have been under increasing use year-round by elephants since 1975. These wetlands showed a rapid increase in extent between 1950 and 1976. This corresponds to the elephant-driven decline in <I>Acacia xanthophloea</I> woodlands, which may have changed the hydrology of the area. Since then, only one of the wetlands, Longinye, has shown dynamic changes in extent. This is either the effect of the blockage of water flow by vegetation or the creation of new channels by elephants and hippopotami. The wetlands had a diverse range of wetland habitat from areas of open water with scattered tall <I>Cyperus papyrus</I> communities in the centre of one wetland to large expanses of seasonally inundated <I>C. laevigatus</I> and <I>Cynodon dactylon</I> dominated habitats fringing the wetlands. Most of the wetland habitats were composed of short sedges and grasses, which was shown to be the direct result of elephant grazing. Elephants tended to keep the vegetation of a control wetland short and of low biomass during both wet and dry seasons, whereas the vegetation in a treatment wetland, where elephants were excluded, showed a rapid increase in biomass and height. Other herbivores were shown to maintain conditions of short vegetation of low biomass through grazing pressure in the dry season. The impact of natural and simulated grazing on growth of the vegetation was studied and only under simulated grazing pressure was growth increased in the seasonally inundated <I>C. dactylon</I> grasslands, as well as the floating mat communities dominated by <I>C. laevigatus</I> that occur in the shallow water wetlands. This indicates that natural grazing may be detrimental either to the plants or the soil. <br /><br /> The wetlands that are under human use had predominantly shallow water and seasonal wetland habitat. Deep water habitat with <I>C. papyrus</I> communities was only present in one of these wetlands, Kimana, which is also the only wetland outside the park that is used by livestock and wildlife, as well as to irrigate a large area of land around the wetland. Two other wetlands, Namelok and Lenkir that were predominantly used for irrigated agriculture, were largely composed of seasonally inundated wetland habitat. A fourth wetland, Esoitpus, has been almost completely drained and this has most likely resulted in the development of <I>A. xanthophloea</I> and <I>C. dactylon</I> dominated riverine and <I>C. laevigatus</I> / <I>C. dactylon</I> dominated seasonally inundated wetland communities. <br /><br /> Overall, the wetlands in the park provide a measure of habitat diversity that may be useful for various invertebrate and vertebrate communities. However, the present lack of tall sedges may negatively impact the bird community. This state can be reversed by the exclusion of elephants from some wetlands for short periods of time. On the other hand, the wetlands outside the park appear to be facing rapid draw-down. Hence, there is an urgent need for sound water management practices for these wetlands.

The wetlands of the semi-arid savanna ecosystem of Amboseli are the critical dry-season range for a diverse wildlife population, as well as for livestock and humans. Changes in the migratory patterns and increases in the population of the elephants in the ecosystem, as well as a shift in the lifestyle of the growing human population from predominantly nomadic pastoralist society to a sedentary agrarian community, has had far-reaching effects on the ecosystem and the wetlands. In this thesis, the current status of the wetlands and the changes in vegetation over the last four decades were examined using satellite imagery and long-term census data. Studies were also conducted to determine the impact of grazing by elephants and other large mammals on wetland vegetation, and the flooding patterns and extents of some of the wetlands were examined. <br /><br /> Long-term aerial count data indicated that the wetlands that are currently protected by the Amboseli National Park have been under increasing use year-round by elephants since 1975. These wetlands showed a rapid increase in extent between 1950 and 1976. This corresponds to the elephant-driven decline in <I>Acacia xanthophloea</I> woodlands, which may have changed the hydrology of the area. Since then, only one of the wetlands, Longinye, has shown dynamic changes in extent. This is either the effect of the blockage of water flow by vegetation or the creation of new channels by elephants and hippopotami. The wetlands had a diverse range of wetland habitat from areas of open water with scattered tall <I>Cyperus papyrus</I> communities in the centre of one wetland to large expanses of seasonally inundated <I>C. laevigatus</I> and <I>Cynodon dactylon</I> dominated habitats fringing the wetlands. Most of the wetland habitats were composed of short sedges and grasses, which was shown to be the direct result of elephant grazing. Elephants tended to keep the vegetation of a control wetland short and of low biomass during both wet and dry seasons, whereas the vegetation in a treatment wetland, where elephants were excluded, showed a rapid increase in biomass and height. Other herbivores were shown to maintain conditions of short vegetation of low biomass through grazing pressure in the dry season. The impact of natural and simulated grazing on growth of the vegetation was studied and only under simulated grazing pressure was growth increased in the seasonally inundated <I>C. dactylon</I> grasslands, as well as the floating mat communities dominated by <I>C. laevigatus</I> that occur in the shallow water wetlands. This indicates that natural grazing may be detrimental either to the plants or the soil. <br /><br /> The wetlands that are under human use had predominantly shallow water and seasonal wetland habitat. Deep water habitat with <I>C. papyrus</I> communities was only present in one of these wetlands, Kimana, which is also the only wetland outside the park that is used by livestock and wildlife, as well as to irrigate a large area of land around the wetland. Two other wetlands, Namelok and Lenkir that were predominantly used for irrigated agriculture, were largely composed of seasonally inundated wetland habitat. A fourth wetland, Esoitpus, has been almost completely drained and this has most likely resulted in the development of <I>A. xanthophloea</I> and <I>C. dactylon</I> dominated riverine and <I>C. laevigatus</I> / <I>C. dactylon</I> dominated seasonally inundated wetland communities. <br /><br /> Overall, the wetlands in the park provide a measure of habitat diversity that may be useful for various invertebrate and vertebrate communities. However, the present lack of tall sedges may negatively impact the bird community. This state can be reversed by the exclusion of elephants from some wetlands for short periods of time. On the other hand, the wetlands outside the park appear to be facing rapid draw-down. Hence, there is an urgent need for sound water management practices for these wetlands.

With the increase in human population, water resources have become more and more precious. A comprehensive study of water yield characteristics is imperative, especially in water-limited semiarid regions. The objective of this study is to examine spring flow and vegetation cover in a first-order watershed and investigate the herbaceous community structure of upland riparian zones. This study consists of two major components: (1) the effects of environmental factors and vegetation cover on spring flow at Pedernales River upland catchments, and (2) the ecological responses of vegetation to altered flow regimes that result from brush management at the upland riparian zones. The study finds that an average of 3.67% of the monthly water budget of first-order catchments in central Texas is made up of spring flow. The influence of woody plant cover on streamflow was evaluated by comparing spring sites with different percentages of woody cover three times during 2003 and 2004. Our findings indicate that changes in woody plant cover had no influence on the amounts of streamflow from these catchments, and the surface catchment area had only a minor influence. This suggests that the real spring catchment area might be different from the surface watershed boundaries that have been delineated by topography. Plant species richness and diversity gradually decreased with increasing lateral distances from the stream bank. Herbaceous richness and diversity declined with increasing Ashe juniper cover in the riparian zone. Ashe juniper canopy cover had a larger effect on the understory composition than the cover of other woody species. Herbaceous diversity and production was greater in areas with sparse tree density than in areas with no trees, but was lowest at high tree densities. The complete removal of Ashe juniper in the riparian zones is not recommended because of the potential loss of grass cover. The recommended management would be to leave a sparse cover of canopy trees to maintain understory plants.