| Summary: | <p>Migratory wildebeest are a key component of the Serengeti-Mara ecosystem. Their population dynamics and movement patterns depend on, and contribute to, a complex set of interactions between the soils, wildlife, vegetation, fire, topography, geology, and precipitation patterns. Yet climate change is projected to alter precipitation patterns across the globe. This thesis set out to explore how future mean precipitation patterns may change over the SME, and how this may impact the migratory wildebeest. </p>
<p>The first paper of the thesis quantified the past mean climatological state of the regional precipitation, and identified the key atmospheric processes governing its spatial and temporal distribution. Precipitation over the SME was heterogeneous; different parts of the region were found to have distinct annual cycles, peak precipitation months, dry season precipitation totals and time of peak daily precipitation. These were found to be controlled by interactions between the complex rift topography, mid-tropospheric large-scale easterlies (at approximately 700 hPa), and lower tropospheric lake breeze westerlies from Lake Victoria (present between the surface and 800 hPa). </p>
<p>The second paper assessed the suitability of mechanistic ecological model for use in ecological climate change impact studies, by driving the model with a spectrum of novel precipitation conditions. The study found that modelled wildebeest population size was most sensitive to changes in annual precipitation, dry season precipitation, and precipitation over the south-east of the SME. As long as the annual cycle was present, simulated movement only ceased when all environmental gradients were removed, suggesting climate change would unlikely to stop the migration (although increased rainfall could weaken the environmental gradients that drive it). </p>
<p>The third paper evaluated the performance of a convection-permitting climate model over the SME, examined the projected future changes to the regional climate, and modelled the impact of the future changes on the SME wildebeest. The climate model captured the zonal component of the precipitation gradient across the ecosystem and the annual cycle of precipitation, although unrealistically high totals were simulated in wet season months. Overly strong lake breeze winds were suggested as a causal mechanism for a dry annual bias over Lake Victoria’s shoreline, and abnormally high precipitation totals over the rift highlands. Future mean precipitation was projected to increase over the SME in every month except August. Analysis of mid-tropospheric winds suggested this was due to anomalous southerlies over the Lake Victoria basin. The projected increase in future precipitation resulted in an increase in the modelled wildebeest population size. This suggests that climate change may cause a mean state that is favourable to the SME wildebeest, although changes to interspecific competition or interannual variability of precipitation could undermine these favourable mean conditions. </p>
|
|---|