Skip to main content
ARS Home » Research » Publications at this Location » Publication #202050
Title: Using the RZWQM to Simulate the Fate of Nitrogen in Field Soil - Crop Environment in the Mediterranean Region

Author
item CAMEIRA, M - TECHNICAL UNIVERSITY
item FERNANDO, R - TECHNICAL UNIVERSITY
item Ahuja, Lajpat
item Ma, Liwang

Submitted to: Agricultural Water Management
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/5/2007
Publication Date: 4/2/2007
Citation: Cameira, M.R., Fernando, R.M., Ahuja, L.R., Ma, L. 2007. Using the RZWQM to Simulate the Fate of Nitrogen in Field Soil - Crop Environment in the Mediterranean Region. Agricultural Water Management. 90:121-136. 2007.

Interpretive Summary: Water and Nitrogen (N) balances in agricultural systems are important for evaluating management effects on environmental quality. This study presents an evaluation of the Root Zone Water Quality Model (RZWQM) for assessing the fate of N in the soil-crop environment at the field scale in Portugal under the Mediterranean conditions, dealing specifically with N transformations in the soil, crop uptake and transport in the soil vadose zone. It was designed to evaluate the RZWQM for two distinct agricultural systems; one consists of a grain corn planted in a silty loam soil with level basins (flood) irrigation and the other a forage corn planted in a sandy soil with sprinkler irrigation. In this study, we emphasized the calibration of the soil organic matter pools and selected soil N transformation processes (mineralization, hydrolysis and nitrification), using two years of data (1996 and 1997). A third year (1998) was used to validate the model performance under four different fertilization management practices, with predicted crop yield within an error of 1.1 and 2.8 % for grain and forage corn, respectively. Predicted N uptake was within an error of 2.8 and 13 % for grain and forage corn, respectively. For the silty loam soil and during the crop season, soil profile nitrate-N was predicted within the experimental errors. For the sandy soil predicted soil profile nitrate-N was within experimental error for one fertilizer treatment and higher than experimental error for the other two treatments. Predicted soil residual nitrate-N, after crop harvest, has errors ranging from 18 to 37%.

Technical Abstract: Water and Nitrogen (N) balances in agricultural systems are important for evaluating management effects on environmental quality. This paper presents an evaluation of the Root Zone Water Quality Model (RZWQM) for assessing the fate of N in the soil-crop environment at the field scale in Portugal under the Mediterranean conditions, dealing specifically with N transformations in the soil, crop uptake and transport in the soil vadose zone. The study was designed to evaluate the RZWQM for two distinct agricultural systems; one consists of a grain corn planted in a silty loam soil with level basins (flood) irrigation and the other a forage corn planted in a sandy soil with sprinkler irrigation. In this study, we emphasized the calibration of the soil organic matter pools and selected soil N transformation processes (mineralization, hydrolysis and nitrification), using two years of data (1996 and 1997). An iterative method was necessary to account for the strong interaction between the different model components, based upon detailed experimental data on soils and crops. The purposed calibration procedure allowed the model to predict N uptake, crop yield and nitrogen balance, with an accuracy that is acceptable in practical applications for complex and spatially variable field conditions. A third year (1998) was used to validate the model performance under four different fertilization management practices, with predicted crop yield within an error of 1.1 and 2.8 % for grain and forage corn, respectively. Predicted N uptake was within an error of 2.8 and 13 % for grain and forage corn, respectively. For the silty loam soil and during the crop season, nitrate-N was predicted for the entire soil profile with a RMSE (22.1 kg ha-1) lower than the average standard deviation of the measured data (MSD) (44.7 kg ha-1). For the sandy soil RMSE was lower than MSD for one fertilizer treatment and higher for the other two treatments. The prediction of the residual nitrate-N, after crop harvest, presented errors ranging from 18 to 37%.