Modeling habitat selection and population dynamics of Mongolian gazelle

A classic discussion of large mammalian herbivore population dynamics would focus on top-down and bottom-up drivers. Yet what is often forgotten, is that many of these species are also highly mobile, covering hundreds or thousands of kilometers in a year, and that this mobility can also influence population dynamics, although the mechanisms are still understudied. While the top-down and bottom-up drivers are more researched, global change will alter how all three drivers impact population dynamics. Of particular concern are habitat fragmentation, which alters movement patterns, and climate change through its direct impact on large herbivore physiology but also its indirect influence through impacts on vegetation dynamics. Understanding these potential effects remains challenging, so the goal of my dissertation was to show that one approach to understanding the effects of global change is by modeling both plant and herbivore ecophysiology. To do this I made new additions to a dynamic global vegetation model coupled to a physiological model of herbivores, using Mongolian gazelle in the steppes of eastern Mongolia as a case study. To parameterize the ecophysiological model for Mongolian gazelle, I needed to better understand how gazelle move to select for forage during the growing season and snow cover during winter. I also wanted to understand if selection differs between the individual and population level. To do this I combined gazelle movement data with satellite data on vegetation greenness and snow cover and used resource and step-selection functions to test for selection. At the population level, gazelle selected for higher-than-average vegetation greenness during the growing season indicating that they select areas of higher forage cover in a landscape where forage cover is often sparse. In winter, at the population level, gazelle selected for intermediate snow cover, striking a balance between staying hydrated and being able to move through the snow. At the individual level, in both seasons and across various spatial scales, I was not able to detect selection for most individuals. This was likely because vegetation, even up to 35 km away, is still very similar to where a gazelle currently is. Therefore, once gazelle are in a good foraging patch, they can move within it for a long time before they must decide where to move to next. In such a landscape, random searches might be the best foraging strategy. For the ecophysiological herbivore-vegetation model, this meant using the ~45 x 45km grid cell size of the vegetation model and a simple random search movement pattern was adequate to describe gazelle movements. Based on the results of the habitat selection study I added movement to an existing ecophysiological herbivore-vegetation model and adapted it to work for temperate ungulates. I used this model it to ask how movement drives the population dynamics of Mongolian gazelle. I did this by running the model once allowing gazelle to move freely within the landscape and once restricting movement to ~45 x 45km areas. Not only were gazelle more than two times more abundant when they were allowed to move, their population also increased more during years of abundant forage and decreased less during drought, indicating that movement also stabilized population dynamics. Restricting movement resulted in local extinctions because gazelle were vulnerable to boom-bust dynamics or harsh winters. The results suggest that for highly mobile species, protected areas are not an adequate conservation measure and that the focus must be on creating permeable landscapes. For many arctic and temperate herbivores, including Mongolian gazelle, harsh winters decrease survival. Yet with warmer winters due to climate change this fundamental population control might change. Simultaneously, many areas are experiencing vegetation greening trends which have been linked to the plant-physiological effects of increased atmospheric CO2 concentrations (CO2 fertilization). Both changes could positively influence temperate herbivores but are not well studied and so I examined their effects with the ecophysiological herbivore-vegetation model by modifying it to better account for large herbivore energy expenditures like thermoregulation. I then ran with model with climate data for two contrasting socio-economic future scenarios. Gazelle abundance increased in both future scenarios, driven equally by increases in forage biomass and decreases in winter thermoregulation costs. Increases in forage biomass were due to increases in growing season length and CO2 fertilization effects. While ultimately negative consequences of climate change might cancel out these positive effects, the results show these positive effects are large and cannot be ignored like they currently are. While there are detailed ecophysiological models of herbivores or vegetation, the combination is still rare. My PhD shows that the combination is key. Gazelle respond to environmental conditions by moving and these movements influence energy intake and expenditure, scaling up to influence population abundance and stability. Under climate change, accounting for both the physiological response of plants and herbivores, showed that both contribute equally to increases in gazelle abundance. Therefore, unlike most climate change studies which examine distribution, I was able to examine abundance. Most importantly, because the herbivore part of the model can simulate any terrestrial ungulate and the vegetation part works globally, I hope the model, whose code is freely available, will be applied to a variety of other questions and herbivore systems in the future.