Get reliable rangeland science

Systems analysis has aided in analyzing and predicting the impacts of various management strategies to many rangeland issues, including grazing effect on annual net primary production, animal health (e.g., livestock body condition scores), ranch profitability (i.e., from alternative stocking rates), or wildlife populations (e.g., interaction between climate and harvest strategies). Based on previously published models, evidence suggests that rangeland models could be enhanced through incorporation of ecohydrologic concepts, particularly as it relates to forage supply and water balance on rangelands. Previous rangeland models have relied on: a) empirical relationships of precipitation and plant production; and b) coupling these estimates to assumed coefficients about range condition and previous rainfall trends; in order to c) model forage supply usable for grazing or wildlife through changes in range condition, irrespective of changes in plant community composition. On the other hand, ecohydrology models have focused on the importance of soil texture and the basic water-balance equations to model infiltration, excess runoff, and changes in plant community composition through changes of evapotranspiration, which is partly driven by available soil moisture. By combining approaches, rangeland models could account not only for grazing impacts on production and profitability, but also the impacts to site-specific hydrologic function, which should prove useful given uncertain climate changes and increasing awareness of ecosystem goods and services. In this poster, I present a simple plant-soil-water model created in Stella� (iSeeSystems, Lebanon, NH) that illustrates how ecohydrology concepts could be incorporated to new or existing rangeland models. The model is calibrated to observed data from four locations of diverse soil properties and climate characteristics in Texas (Seymour, Palestine, San Marcos, and Edinburg; TAMU North America Soil Moisture Database). Early diagnostic and sensitivity tests will be presented. Lastly, some model limitations are described along with directions of future work.