Developing a Forest Gap Model to Be Applied to a Watershed-scaled Landscape in the Cross Timbers Ecoregion Using a Topographic Wetness Index
Description: A method was developed for extending a fine-scaled forest gap model to a watershed-scaled landscape, using the Eastern Cross Timbers ecoregion as a case study for the method. A topographic wetness index calculated from digital elevation data was used as a measure of hydrologic across the modeled landscape, and the gap model modified to have with a topographically-based hydrologic input parameter. The model was parameterized by terrain type units that were defined using combinations of USDA soil series and classes of the topographic wetness index. A number of issues regarding the sources, grid resolutions, and processing methods of the digital elevation data are addressed in this application of the topographic wetness index. Three different grid sizes, 5, 10, and 29 meter, from both LiDAR-derived and contour-derived elevation grids were used, and the grids were processed using both single-directional flow algorithm and bi-directional flow algorithm. The result of these different grids were compared and analyzed in context of their application in defining terrain types for the forest gap model. Refinements were made in the timescale of gap model’s weather model, converting it into a daily weather generator, in order to incorporate the effects of the new topographic/hydrologic input parameter. The precipitation model was converted to use a Markov model to initiate a sequence of wet and dry days for each month, and then daily precipitation amounts were determined using a gamma distribution. The output of the new precipitation model was analyzed and compared with a 100-year history of daily weather records at daily, monthly, and annual timescales. Model assumptions and requirements for biological parameters were thoroughly investigated and questioned. Often these biological parameters are based on little more than assumptions and intuition. An effort to base as many of the model’s biological parameters on measured data was made, including a new ...
Date: August 2014
Creator: Goetz, Heinrich
Partner: UNT Libraries