A Quantitative Energy Model for Predicting Pedogenic Environments.
Craig Rasmussen, Univ of Arizona, Soil, Water, and Environmental Science Dept, 1177 E. Fourth St. Shantz Bldg #38, PO Box 210038, Tucson, AZ 85721-0038
A quantitative energy model for predicting pedogenic environments and energy flow through soil systems is presented. Energy inputs from precipitation and Net Primary Production (NPP) were calculated based on the temperature of effective precipitation (Peff) and the temperature of the months with Peff, respectively. This energy calculation presents an improvement of previous energy based approaches to modeling soil genesis (e.g., Runge, 1973) in that it allows for quantification of energy input (kJ m-2 yr-1) from both Peff (EPPT) and NPP (ENPP). We suggest these parameters may be used to segregate similar pedogenic regimes and may be used to predict specific soil properties. The model was developed using the PRISM climate dataset (at a scale of 1:250,000 or 4 km by 4 km pixels) for the continental United States. Monthly temperature data was used to calculate potential evapotranspiration (ETp) using the Thornthwaite equation and Peff calculated as the difference between monthly precipitation and ETp. Using the specific heat of water, Peff was assumed to have been heated from 0°C to the average temperature of that month, facilitating the conversion from cm of water to EPPT. Likewise, it was assumed that NPP occurred primarily in months of Peff and that NPP was then controlled by the temperature of that month. An empirical equation was used to estimate mass of NPP produced, which was subsequently converted to ENPP assuming a set amount of energy per gram of NPP. The sum of ENPP and EPPT is termed EIN, and represents the total input of energy to the soil system. We also utilized global weather station compiled by the International Atomic Energy Agency (IAEA) to examine the relationship between mean annual precipitation (MAP), mean annual temperature (MAT), and EIN. We were able to fit a 2-dimensional Gaussian model to the MAP, MAT, and EIN data and developed an equation for predicting EIN based on MAP and MAT. We used this equation to calculate EIN in the U.S. and compared the results to EIN estimated from the PRISM dataset (r2 = 0.89; P<0.0001). Results suggest the global equation accurately estimates energy input into U.S. soil systems. Using watershed data of Si flux from granitoid watersheds in the continental U.S. (White and Blum, 1995), we observed a significant relationship between EIN and Si flux (r2 = 0.71; P<0.001), indicating EIN may provide an estimate of current rates of silicate weathering. We also compared EIN and the percent of EIN derived from ENPP by soil order for the continental U.S. Soil orders in all states showed differences in EIN and the percent of EIN from NPP (%ENPP)(e.g., Ultisols EIN =29,915, % ENPP =49%; Mollisols EIN =5,880, % ENPP =90%), suggesting that the model may be used to isolate pedogenic environments at the regional scale. We further tested the model using data from a series of climosequences in the western Sierra Nevada, California. Climosequence data indicates significant relationships between EIN and chemical weathering indices. In addition, EIN provided a better prediction of weathering indices relative to MAP and MAT (r2 =0.72 versus r2 = 0.18 and 0.49 for MAP and MAT, respectively). Results from multiple scales suggest the potential for using this energy based approach to predict pedogenic regimes, as well as estimate regional and local rates of silicate weathering. The model requires further testing to establish its usefulness in landscape scale applications.