Matching Items (1,853)
Filtering by
- Creators: Debussy, Claude, 1862-1918
Description
Two critical limitations for hyperspatial imagery are higher imagery variances and large data sizes. Although object-based analyses with a multi-scale framework for diverse object sizes are the solution, more data sources and large amounts of testing at high costs are required. In this study, I used tree density segmentation as the key element of a three-level hierarchical vegetation framework for reducing those costs, and a three-step procedure was used to evaluate its effects. A two-step procedure, which involved environmental stratifications and the random walker algorithm, was used for tree density segmentation. I determined whether variation in tone and texture could be reduced within environmental strata, and whether tree density segmentations could be labeled by species associations. At the final level, two tree density segmentations were partitioned into smaller subsets using eCognition in order to label individual species or tree stands in two test areas of two tree densities, and the Z values of Moran's I were used to evaluate whether imagery objects have different mean values from near segmentations as a measure of segmentation accuracy. The two-step procedure was able to delineating tree density segments and label species types robustly, compared to previous hierarchical frameworks. However, eCognition was not able to produce detailed, reasonable image objects with optimal scale parameters for species labeling. This hierarchical vegetation framework is applicable for fine-scale, time-series vegetation mapping to develop baseline data for evaluating climate change impacts on vegetation at low cost using widely available data and a personal laptop.
ContributorsLiau, Yan-ting (Author) / Franklin, Janet (Thesis advisor) / Turner, Billie (Committee member) / Myint, Soe (Committee member) / Arizona State University (Publisher)
Created2013
Description
Species distribution modeling is used to study changes in biodiversity and species range shifts, two currently well-known manifestations of climate change. The focus of this study is to explore how distributions of suitable habitat might shift under climate change for shrub communities within the Santa Monica Mountains National Recreation Area (SMMNRA), through a comparison of community level to individual species level distribution modeling. Species level modeling is more commonly utilized, in part because community level modeling requires detailed community composition data that are not always available. However, community level modeling may better detect patterns in biodiversity. To examine the projected impact on suitable habitat in the study area, I used the MaxEnt modeling algorithm to create and evaluate species distribution models with presence only data for two future climate models at community and individual species levels. I contrasted the outcomes as a method to describe uncertainty in projected models. To derive a range of sensitivity outcomes I extracted probability frequency distributions for suitable habitat from raster grids for communities modeled directly as species groups and contrasted those with communities assembled from intersected individual species models. The intersected species models were more sensitive to climate change relative to the grouped community models. Suitable habitat in SMMNRA's bounds was projected to decline from about 30-90% for the intersected models and about 20-80% for the grouped models from its current state. Models generally captured floristic distinction between community types as drought tolerance. Overall the impact on drought tolerant communities, growing in hotter, drier habitat such as Coastal Sage Scrub, was predicted to be less than on communities growing in cooler, moister more interior habitat, such as some chaparral types. Of the two future climate change models, the wetter model projected less impact for most communities. These results help define risk exposure for communities and species in this conservation area and could be used by managers to focus vegetation monitoring tasks to detect early response to climate change. Increasingly hot and dry conditions could motivate opportunistic restoration projects for Coastal Sage Scrub, a threatened vegetation type in Southern California.
ContributorsJames, Jennifer (Author) / Franklin, Janet (Thesis advisor) / Rey, Sergio (Committee member) / Wentz, Elizabeth (Committee member) / Arizona State University (Publisher)
Created2014
Description
Climate and land use change are projected to threaten biodiversity over the coming century. However, the combined effects of these threats on biodiversity and the capacity of current conservation networks to protect species' habitat are not well understood. The goals of this study were to evaluate the effect of climate change and urban development on vegetation distribution in a Mediterranean-type ecosystem; to identify the primary source of uncertainty in suitable habitat predictions; and to evaluate how well conservation areas protect future habitat in the Southwest ecoregion of the California Floristic Province. I used a consensus-based modeling approach combining three different species distribution models to predict current and future suitable habitat for 19 plant species representing different plant functional types (PFT) defined by fire-response (obligate seeders, resprouting shrubs), and life forms (herbs, subshurbs). I also examined the response of species grouped by range sizes (large, small). I used two climate models, two emission scenarios, two thresholds, and high-resolution (90m resolution) environmental data to create a range of potential scenarios. I evaluated the effectiveness of an existing conservation network to protect suitable habitat for rare species in light of climate and land use change. The results indicate that the area of suitable habitat for each species varied depending on the climate model, emission scenario, and threshold combination. The suitable habitat for up to four species could disappear from the ecoregion, while suitable habitat for up to 15 other species could decrease under climate change conditions. The centroid of the species' suitable environmental conditions could shift up to 440 km. Large net gains in suitable habitat were predicted for a few species. The suitable habitat area for herbs has a small response to climate change, while obligate seeders could be the most affected PFT. The results indicate that the other two PFTs gain a considerable amount of suitable habitat area. Several rare species could lose suitable habitat area inside designated conservation areas while gaining suitable habitat area outside. Climate change is predicted to be more important than urban development as a driver of habitat loss for vegetation in this region in the coming century. These results indicate that regional analyses of this type are useful and necessary to understand the dynamics of drivers of change at the regional scale and to inform decision making at this scale.
ContributorsBeltrán Villarreal, Bray de Jesús (Author) / Franklin, Janet (Thesis advisor) / Fenichel, Eli P (Committee member) / Kinzig, Ann P (Committee member) / Collins, James P. (Committee member) / Arizona State University (Publisher)
Created2012
Description
Climate change has the potential to affect vegetation via changes in temperature and precipitation. In the semi-arid southwestern United States, heightened temperatures will likely lead to accelerated groundwater pumping to meet human needs, and altered storm patterns may lead to changes in flood regimes. All of these hydrologic changes have the potential to alter riparian vegetation. This research, consisting of two papers, examines relationships between hydrology and riparian vegetation along the Verde River in central Arizona, from applied and theoretical perspectives. One paper investigates how dominance of tree and shrub species and cover of certain functional groups change along hydrologic gradients. The other paper uses the Verde River flora along with that river's flood and moisture gradients to answer the question of whether functional groups can be defined universally. Drying of the Verde River would lead to a shift from cottonwood-willow streamside forest to more drought adapted desert willow or saltcedar, a decline in streamside marsh species, and decreased species richness. Effects drying will have on one dominant forest tree, velvet ash, is unclear. Increase in the frequency of large floods would potentially increase forest density and decrease average tree age and diameter. Correlations between functional traits of Verde River plants and hydrologic gradients are consistent with "leaf economics," or the axis of resource capture, use, and release, as the primary strategic trade-off for plants. This corresponds to the competitor-stress tolerator gradient in Grime's life history strategy theory. Plant height was also a strong indicator of hydrologic condition, though it is not clear from the literature if plant height is independent enough of leaf characteristics on a global scale to be considered a second axis. Though the ecohydrologic relationships are approached from different perspectives, the results of the two papers are consistent if interpreted together. The species that are currently dominant in the near-channel Verde River floodplain are tall, broad-leaf trees, and the species that are predicted to become more dominant in the case of the river drying are shorter trees or shrubs with smaller leaves. These results have implications for river and water management, as well as theoretical ecology.
ContributorsHazelton, Andrea Florence (Author) / Stromberg, Juliet C. (Thesis advisor) / Schmeeckle, Mark W (Committee member) / Franklin, Janet (Committee member) / Arizona State University (Publisher)
Created2011
ContributorsDebussy, Claude, 1862-1918 (Composer)
ContributorsDebussy, Claude, 1862-1918 (Composer)
ContributorsDebussy, Claude, 1862-1918 (Composer)
ContributorsDebussy, Claude, 1862-1918 (Composer)
ContributorsDebussy, Claude, 1862-1918 (Composer)
Description
Often, when thinking of cities we envision designed landscapes, where people regulate everything from water to weeds, ultimately resulting in an ecosystem decoupled from biophysical processes. It is unclear, however, what happens when the people regulating these extensively managed landscapes come under stress, whether from unexpected economic fluctuations or from changing climate norms. The overarching question of my dissertation research was: How does urban vegetation change in response to human behavior? To answer this question, I conducted multiscale research in an arid urban ecosystem as well as in a virtual desert city. I used a combination of long-term data and agent-based modeling to examine changes in vegetation across a range of measures influenced by biophysical, climate, institutional, and socioeconomic drivers. At the regional scale, total plant species diversity increased from 2000 to 2010, while species composition became increasingly homogeneous in urban and agricultural areas. At the residential scale, I investigated the effects of biophysical and socioeconomic drivers – the Great Recession of 2007-2010 in particular – on changing residential yard vegetation in Phoenix, AZ. Socioeconomic drivers affected plant composition and increasing richness, but the housing boom from 2000 through 2005 had a stronger influence on vegetation change than the subsequent recession. Surprisingly, annual plant species remained coupled to winter precipitation despite my expectation that their dynamics might be driven by socioeconomic fluctuations. In a modeling experiment, I examined the relative strength of psychological, social, and governance influences on large-scale urban land cover in a desert city. Model results suggested that social norms may be strong enough to lead to large-scale conversion to low water use residential landscaping, and governance may be unnecessary to catalyze residential landscape conversion under the pressure of extreme drought conditions. Overall, my dissertation research showed that urban vegetation is dynamic, even under the presumably stabilizing influence of human management activities. Increasing climate pressure, unexpected socioeconomic disturbances, growing urban populations, and shifting policies all contribute to urban vegetation dynamics. Incorporating these findings into planning policies will contribute to the sustainable management of urban ecosystems.
ContributorsRipplinger, Julie (Author) / Franklin, Janet (Thesis advisor) / Collins, Scott L. (Thesis advisor) / Anderies, John M (Committee member) / Childers, Daniel L. (Committee member) / York, Abigail (Committee member) / Arizona State University (Publisher)
Created2015