Scaling Issues and Spatio-Temporal Variability in Ecohydrological Modeling on Mountain Topography: Methods and Future of the VELMA Model

The interactions between vegetation and hydrology in mountainous terrain are difficult to represent in mathematical models. There are at least three primary reasons for this difficulty. First, expanding plot-scale measurements to the watershed scale requires finding the balance between computational intensity and physical significance. Second, parameters that affect soil, plant and hydrologic processes are distributed heterogeneously across mountain landscapes, and these patterns and processes may be spatially connected. Third, temporal variation in water availability (particularly in seasonal rainfall climates) may involve a “topographical memory” that may be expressed as “lags” between biological and hydrological processes. A unique opportunity for examining the implications of scaling and spatio-temporal variability on ecohydrological models exists at the H.J. Andrews Experimental Forest (HJA) in Blue River, Oregon. HJA is a National Science Foundation Long Term Ecological Research (LTER) site, and has been monitoring climate, stream, and vegetation characteristics of small watersheds for more than 50 years. A recent LIDAR (Light Detection and Ranging) reconnaissance has produced watershed scale estimations of vegetation and soil surface parameters at a very high spatial resolution, allowing spatially-explicit expansion of long-term data. An ecohydrological model, Visualizing Ecosystems for Land Management Assessments (VELMA) developed by the Stieglitz lab at Georgia Tech in collaboration with EPA has also been calibrated specifically for watershed topographies in HJA. VELMA is a coupled ecohydrological model that simulates the cycling and transport of water and nutrients in three dimensions by specific parameterization of hydrological and biogeochemical functions. It contains submodels for plant, soil, and water processes including surface and sub-surface flow on a daily time step. We are using the VELMA model to explore three sequential and fundamental questions in ecohydrological modeling in mountainous terrain. 1) How does the topographical structure of mountains (elevation, slope, and aspect) impact hydrological parameters such as temperature, rainfall, soil depth, canopy structure, and airflow? 2) To what degree are the model results from high-resolution, spatially-explicit parameterization different from results based on broadly distributed means, and when different, on what scale(s) are the discrepancies most pronounced? 3) Is there an optimal scale for the process-based ecohydrological modeling, and if so, what are the computational limits at this scale? This poster will present our overall experimental plan and initial findings. Experimentation on and establishment of a standard procedure for spatial and temporal partitioning in ecohydrological models is a fundamental step from which advancement towards more comprehensive, dynamic models can be developed.