Multi-scale habitat patch modeling: Integrating landscape pattern, habitat suitability and population dynamics with implications for ecology and conservation.

摘要

It is well recognized by landscape ecological theory that habitat patches are hierarchically organized within landscapes, and that different ecological processes will occur within patches at different spatial scales. However, it is difficult to analyze relationships between ecological processes and habitat patches across spatial scales because currently available geographic patch delineation techniques are insufficient. This dissertation presents a novel habitat patch delineation algorithm, called PatchMorph, which uses geographic information systems (GIS) analysis to hierarchically delineate patches from a habitat map based on organism-specific thresholds for patch perception and utilization (Chapter 1). This algorithm is based on the theory of organism-specific thresholds for gaps between patches and thin spurs extruding out from the core of patches. The PatchMorph algorithm is then applied to delineate habitat patches at two spatial scales relevant to the California state-listed threatened western yellow-billed cuckoo (Coccyzus americanus occidentalis) nesting in forest patches along the Sacramento River, California (Chapter 2). By statistically relating patch characteristics to the presence of this species, this chapter shows that the area of cottonwood forest measured at the finest spatial scale of patches was the most important factor determining yellow-billed cuckoo presence. Then, to establish an explicit relationship between the PatchMorph algorithm and population ecological theory, neutral landscapes were used to compare PatchMorph to an individual-based and spatially-explicit population model that simulates randomized movement, reproduction, and mortality (Chapter 3). The results show that the best correlating patch gap threshold values were most influenced by the population movement distance and the level of habitat fragmentation (i.e., contagion), while the best correlating patch spur threshold values were most influenced by movement distance and reproductive probability. Finally, Chapter 4 presents a population viability analysis (PVA) for bank swallow (Riparia riparia) nesting along the Sacramento River. This PVA is based on river bank patches delineated using GIS analysis, and incorporates the effects of habitat loss, density dependence, site fidelity, and stochasticity in survival and fecundity. The results show that the estimated viability of this population has decreased 40-60% due to habitat loss caused by the installation of bank erosion control projects (bank revetment). Taken together, these chapters provide a step forward in the integration landscape ecological patterns with ecological processes. They show that habitat suitability and population dynamics can be explicitly related to, and explained by habitat patches delineated at specific spatial scales.