|Van Donk, Simon|
|Evett, Steven - Steve|
Submitted to: Transactions of the ASAE
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/4/2003
Publication Date: 7/2/2004
Citation: Van Donk, S.J., Tollner, E.W., Steiner, J.L., Evett, S.R. 2004. Soil temperature under a dormant bermudagrass mulch: Simulation and measurement. Transactions of the American Society of Agricultural Engineers. 47:91-98.
Interpretive Summary: The ENergy and WATer BALance (ENWATBAL) computer model simulates soil water and temperature, evaporation from soil, and transpiration from crops. It has been used to evaluate the effect of management on water use and savings, which is especially important in dry climates. ENWATBAL can also be used to assess the effect of management on soil temperature, which is important for such things as nutrient availability, disease development, and crop growth. It does not simulate the effects of a mulch layer, limiting the model's applicability since surface vegetative mulches are becoming more common, especially in reduced-tillage systems. For example, the presence of a mulch modifies soil temperature, which may be favorable or unfavorable for the growth of a specific crop. Our objective was to enhance ENWATBAL to enable simulation of mulch effects. Soil temperatures simulate with the mulch enhanced model were compared with those measured at Watkinsville, Georgia, USA under dormant bermudagrass having a thatchy layer that acted as a mulch. Soil temperature was simulate much better with the mulch enhanced model than with the original ENWATBAL model. We also found a method, slightly modifying the model's mathematical equations, that makes the model run 50 times faster on a computer than the method that uses the exact equations.
Technical Abstract: Simulation models are useful for analyzing agricultural systems, but to be widely applicable, models must accommodate common phenomena. The ENergy and WATer BALance (ENWATBAL) model is a mechanistic, numerical model that simulates soil water and temperature profiles, evaporation from soil, and transpiration from crops, but it does not simulate the effects of a mulch layer. Surface vegetative mulches are becoming more common, especially in reduced-tillage systems, limiting the model's applicability. Our objective was to modify ENWATBAL to enable physically based simulation of the effects of a dense mulch. As a preliminary evaluation of the model, soil temperatures simulated with the modified model were compared with those measured at Watkinsville, Georgia, USA in Cecil sandy loam (clayey, kaolinitic, thermic, Typic Kanhapludult) under dormant bermudagrass (Cynodon dactylon, [L.] Pers.) having a thatchy layer that acted as a muclh during the simulation period. Measured daily soil termperature amplitudes at 0.04 m depth were about 2 C during an 8-day period in December 1995. Simulated amplitudes were 12 C with the original ENWATBAL model, and 3.5 C with the mulch enhanced model. Average absolute difference between measured and simulated soil temperature was 3.9 C using the original ENWATBAL model and 1.2 C using the mulch enhanced model. Measured soil temperatures lagged behind those simulated, indicating that conduction may be an important process of heat transfer through the mulch. Two solution methods were tested, one an iterative solution for mulch and soil surface temperatures implicit in the energy balance equations, and the other a linearized, explicit solution of the energy balances. The latter method was 50 times faster than the iterative method without compromising accuracy: the largest linearization error was only 0.01 C.