Understanding the impacts of climate change on habitats, and the species they support, is a necessity for effective natural resource planning and management. Because climate change is likely to create novel environments, the capacity for responding effectively to the impacts of climate change in managed habitats will rely increasingly on gridded, predictive ecological models. A major challenge of using spatially-explicit models is spatial mismatch: the area represented by each grid square is typically much larger than both the scale at which microclimate conditions vary and at which most organisms experience their environments. As a result, model predictions capture little of the micro-environmental variation that drives critical ecological processes. However, the understanding of how model spatial resolution affects predictive models is limited, impairing the ability to make evidence-based management decisions. Thus, quantifying the limitations of current modeling approaches as well as developing new, more accurate methods will be key to meeting the Department of Defense's (DoD) goals for natural resource management in climate-modified environments, while optimizing time and financial investments and maximizing training capacity.

This research will focus on mapping the above-ground and below-ground microclimate landscapes that support terrestrial ecosystem health and species persistence. In this study, the research team will quantify how spatial resolution and data availability interact in complex landscapes to mediate uncertainty in ecological models and impact the makeup of the three-dimensional (3D) microclimate landscape itself. By modeling both microclimate conditions and their associated spatially-mediated uncertainty, we will improve managers’ ability to effectively target resource conservation by taking full advantage of available remotely-sensed data.

The research will combine emerging methods in remote-sensing with mechanistic modeling to map the 3D microclimate landscape and quantify the spatially-mediated uncertainty in those models. Researchers will use terrain layers derived from unmanned aerial system (UAS)-collected aerial imagery to drive a series of mechanistic microclimate and physiological models for two sites with different terrain profiles. By using a mechanistic approach to microclimate modeling, the spatial resolution of model output will be limited only by the spatial resolution of the terrain input that drives the microclimate model, rather than by the sampling resolution of field-collected data. Researchers will map the microclimate landscape at extremely high spatial solutions and compare predictions with those generated using lower-resolution input layers.

For the physiological components of this study, researchers will focus on a reptile with temperature dependent sex determination (TSD) as a model system. Including TSD as a model thermal trait will allow the team to link model spatial resolution (and the associated spatially-mediated uncertainty) directly to multiple, fundamental physiological processes of embryonic development and sexual differentiation that can have critical, climate-mediated consequences at individual, population, and species levels. Using both broad-scale and fine resolution climate and weather data, researchers will generate microclimate and physiological predictions for the current climate, then define developmental microrefugia: sites that are predicted to support embryonic metabolism, while maintaining balanced hatchling sex ratios.

By addressing key issues in using gridded data in ecological modeling, the results will provide DoD resource managers with insight into the spatially-mediated drivers of microclimate landscapes. The modeling approach emphasizes spatial and temporal transferability of results, a framework that can improve targeting of management efforts and reduce costs associated with conservation-focused data collection.