Skip to main content
ARS Home » Research » Publications at this Location » Publication #195747
Title: SIMULATING MANAGEMENT EFFECTS ON CROP PRODUCTION, TILE DRAINAGE, AND WATER QUALITY USING RZWQM-DSSAT

Author
item ANAPALLI, S.A - COLORADO STATE UNIVERSITY
item Ma, Liwang
item Malone, Robert - Rob
item Heilman, Philip - Phil
item Ahuja, Lajpat
item KANWAR, R.S - IOWA STATE UNIVERSITY
item Karlen, Douglas
item HOOGENBOOM, G. - UNIVERSITY OF GEORGIA

Submitted to: Geoderma
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 10/15/2006
Publication Date: 5/23/2007
Citation: Anapalli, S., Ma, L., Malone, R.W., Heilman, P., Ahuja, L.R., Kanwar, R., Karlen, D.L., Hoogenboom, G. 2007. Simulating management effects on crop production, tile drainage, and water quality using RZWQM-DSSAT. Geoderma 140:3 (2007)297-309. Published on-line 5/23/2007. doi:10.1016/j.geoderma.2007.04.013.

Interpretive Summary: Agricultural system models are viable tools for synthesizing information from long-term field experiments. Our objective was to calibrate and evaluate the Root Zone Water Quality Model (RZWQM) with the Decision Support for Agrotecchnology Transfer (DSSAT v3.5) plant growth modules (RZWQM-DSSAT) for simulating tillage, crop rotation, and N and manure management effects on crop production and water quality. Data from 1978 to 2003 from a water quality experiment near Nashua, IA were used. The model correctly simulated management effects of tillage effects, crop rotation effects, and N application timing on yearly tile flow, yearly N load in tile flow and yearly flow weighted N concentration in tile flow. However, the treatment effects between SA (single N application) and fall injected swine manure (SM) treatments were simulated with relatively less accuracy. On average, the LSNT (late-spring soil N test) plots were simulated to have 337 kg ha-1 higher corn yield compared to SA, when the observed increase was 812 kg ha-1. In all the CS (corn-soybean), SC (soybean-corn), and CC (continuous corn) rotations, the observed trends in yield loss due to NT (no-till) over MP (moldboard plow) or CP were simulated correctly. In the CS and SC rotations the NT treatments were simulated to have 74 kg ha-1 less corn yield on average compared to MP and CP (chisel plow), when the observed decrease was 349 kg ha-1. In the CC rotation, simulated corn yield loss due to NT over MP and CP was 334 kg ha-1 against the observed value of 624 kg ha-1.

Technical Abstract: Well-calibrated and evaluated agricultural system models are viable tools for synthesizing information from long-term field experiments and extrapolating the results to other agricultural systems that have different soil and climate conditions. Our objective was to calibrate and evaluate the Root Zone Water Quality Model (RZWQM) with the Decision Support for Agrotecchnology Transfer (DSSAT v3.5) plant growth modules (RZWQM-DSSAT) for simulating tillage, crop rotation, and N and manure management effects on crop production and water quality. Data from 1978 to 2003 from a water quality experiment near Nashua, IA were used. The model was calibrated using data from one treatment plot and validated for the rest of the treatment plots. The model correctly simulated 74, 80 and 80% (tillage effects); 88, 100 and 99% (crop rotation effects); and 99, 83 and 58% (N application timing effects) of the corresponding observed management effects or trends (increase or decrease) in yearly tile flow, yearly N load in tile flow and yearly flow weighted N concentration in tile flow. However, the treatment effects between SA (single N application) and fall injected swine manure (SM) treatments were simulated with relatively less accuracy (only 64 to 68% of the simulations matched the observed effects). On average, the LSNT (late-spring soil N test) plots were simulated to have 337 kg ha-1 higher corn yield compared to SA, when the observed increase was 812 kg ha-1. In all the CS [corn (Zea mays L.)-soybean [Glycine max (L.) Merr.]), SC (soybean-corn), and CC (continuous corn) rotations, the observed trends in yield loss due to NT (no-till) over MP (moldboard plow) or CP were simulated correctly. In the CS and SC rotations the NT treatments were simulated to have 74 kg ha-1 less corn yield on average compared to MP and CP (chisel plow), when the observed decrease was 349 kg ha-1. In the CC rotation, simulated corn yield loss due to NT over MP and CP was 334 kg ha-1 against the observed value of 624 kg ha-1. Grain yield simulations were not sensitive to differences between RT (ridge till) and NT, SM and SA treatments, and SC or CS and CC. We conclude that the RZWQM-DSSAT model has enough potential for simulating management effects on tile flow and associated N leaching, but needs further improvement for simulating SM and crop rotation effects on yield.