Characterizing Habitat and Densities of the Mojave Desert Tortoise (Gopherus agassizii) at Multiple Spatial Scales

Determining what environmental and anthropogenic factors have the greatest influence on the distribution of the Mojave desert tortoise (Gopherus agassizii) is key to improving population estimates and better understanding how this species uses their habitat around them. The Mojave Desert tortoise is a federally listed threated species and a key component to their delisting is ensuring that they are well distributed throughout their range. Currently range wide surveys of desert tortoises are conducted over large regions of suitable habitat with little consideration of the patchy nature of their distribution. Models were constructed to evaluate influences on desert tortoise densities across their range and are capable of informing conservation managers on where to conduct surveys in the future as well as what habitat to allocate for protection based on desert tortoise habitat preferences. At the range wide scale satellite data is readily available for input into models; however, when evaluating tortoise densities at smaller scales satellite derived remote sensing imagery proves to be too coarse for analyses at these scales. Remote sensing imagery derived from unmanned aerial vehicles (UAV) has recently become a viable option for obtaining data at these scales for various types of analyses. With high resolution imagery obtained with UAVs density models were constructed to evaluate influences on tortoise densities at local scales and with this I was also able to evaluate if tortoise habitat preferences differ amongst scales and regions. Detailed plant and soil data are collected at these small scales through field methods such as the Assessment Inventory and Monitoring (AIM) protocol but do not have the ability to capture the heterogeneity across the landscape; thus, UAV derived imagery has the potential to bridge the gap between satellite derived imagery and field collected data. Here I’ve shown that UAVs can capture a more accurate representation of shrub cover than field methods such as the AIM protocol. Shrub cover is important to Mojave desert tortoise habitat as it provides protection from high temperatures and predators. Though the UAV imagery proved useful for obtaining shrub cover, data collected through field methods will still be necessary for obtaining specific plant and soil data as well as for calibration with remotely sensed imagery. The density models constructed at both the range wide and local scales revealed that desert tortoises do show preferences in habitat selection and these preferences vary from region to region and amongst scales. On the other hand, more work is needed to improve the types of data available for collection with UAVs. These results demonstrate the importance of evaluating Mojave desert tortoise densities at different scales and with data collected through various different means