|Jabro, Jalal "jay"|
|Jabro, Ann -|
|Steven, Fales -|
Submitted to: Soil Science Society of America Journal
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: August 8, 2012
Publication Date: October 19, 2012
Repository URL: http://handle.nal.usda.gov/10113/56096
Citation: Jabro, J.D., Jabro, A.D., Steven, F.L. 2012. Models performance and robustness for simulating drainage and nitrate-nitrogen fluxes without recalibration. Soil Science Society of America Journal. 76(6): 1957–1964. Interpretive Summary: This study sought to assess the robustness of LEACHM, NCSWAP, and SOIL SOIL-N models for simulating drainage and NO3 fluxes. The statistical results suggest that for most cases in the years studied, the models provide reasonable monthly and annual simulations of drainage fluxes and NO3-N losses from the soil profile collected below the 1 m depth under orchardgrass pasture without recalibration. However, the models failed to produce accurate simulations of drainage and NO3-N fluxes due to restricted water movement linked with snow accumulation, frozen soil, soil thaw and snow melt during winter months. Both LEACHM and SOIL-SOILN performed better than NCSWAP based on simulated versus measured NO3-N leaching losses and SOIL-SOILN performed somewhat better than LEACHM based on statistical indices used in this study. Small differences existed among the performances of these three models. A cogent review of the models’ programs suggests the differences among these three models’ performances may be related to the simulation of N and C transformation and pools embedded in the code of the models. Each of these models uses different equations that govern water flow, N and C fractionation pools, cycling, and transformations in the soil, water, and plant system. The results of this study are a monumental progression in the validation of computer models which will lead to continued diffusion across diverse stakeholders. The models were calibrated under corn under different conditions and validated under pasture without recalibration. LEACHM, NCSWAP and SOIL SOIL-N’s ability to predict drainage and NO3-N fluxes is robust. Future research is necessary to further test and validate these models under diverse soils, cropping systems and weather conditions to continue to test for models’ robustness.
Technical Abstract: Intensively grazed pastures generate nitrate-nitrogen (NO3-N) contaminated groundwater creating grave environmental concern. Computer models simulate and forecast appropriate agricultural practices to reduce environmental impact. Model validation is an essential process in evaluation and field application of computer simulation models. The objectives of this study were to assess and compare robustness and performance of three models -- LEACHM, NCSWAP, and SOIL-SOILN--for simulating drainage flux and NO3-N leaching in an intense pasture system without recalibration. A 3-yr study was conducted on a Hagerstown silt loam to measure drainage and NO3-N fluxes below 1 m depth from N-fertilized orchardgrass using intact core lysimeters. Five N-fertilizer treatments were replicated five times in a randomized complete block experimental design. The models were validated under orchardgrass using soil, water and N transformation rate parameters and C pools fractionation derived from a previous study conducted on similar soils under corn. The model efficiency (MEF) of drainage and NO3-N fluxes were 0.53, 0.69 for LEACHM; 0.75, 0.39 for NCSWAP; and 0.94, 0.91for SOIL-SOILN. The models failed to produce reasonable simulations of drainage and NO3-N fluxes in January, February and March due to limited water movement associated with frozen soil and snow accumulation and melt. The differences between simulated and measured NO3-N leaching and among models’ performances may also be related to soil N and C transformation processes embedded in the models. Generally, the results demonstrate the potential of these three models to reasonably simulate annual and monthly drainage and NO3-N fluxes under pasture system without recalibration.