Skip to main content
ARS Home » Midwest Area » Ames, Iowa » National Laboratory for Agriculture and The Environment » Agroecosystems Management Research » Research » Publications at this Location » Publication #239138
Title: RZWQM predicted effects of soil N testing with incorporated automatic parameter optimization software (PEST) and weather input quality control

Author
item Malone, Robert - Rob
item Jaynes, Dan
item Ma, Liwang
item NOLAN, B.T. (TOM) - Us Geological Survey (USGS)
item Meek, David

Submitted to: ASA-CSSA-SSSA Annual Meeting Abstracts
Publication Type: Abstract Only
Publication Acceptance Date: 11/5/2009
Publication Date: 11/5/2009
Citation: Malone, R.W., Jaynes, D.B., Ma, L., Nolan, B., Meek, D.W. 2009. RZWQM predicted effects of soil N testing with incorporated automatic parameter optimization software (PEST) and weather input quality control. American Society of Agronomy Annual Meetings [abstracts]. ASA-CSSA-SSSA Annual Meeting, Nov. 1-4, 2009, Pittsburgh, PA. CD-ROM.

Interpretive Summary:

Technical Abstract: Among the most promising tools available for determining precise N requirements are soil mineral N tests. Field tests that evaluated this practice, however, have been conducted under only limited weather and soil conditions. Previous research has shown that using agricultural systems models such as the Root Zone Water Quality Model (RZWQM) to extrapolate beyond the limitations of field experiments has been successful for many treatments but not soil N testing. But, determining optimum input for mechanistic agricultural systems models can be very difficult with the vast number of soil, hydrological, crop, and carbon cycling parameters that are interrelated. Another difficulty is that these models are very sensitive to meteorological input data, but the quality of raw meteorological observations is known to be suspect. Therefore, we adapted the well known parameter optimizer PEST to help calibrate RZWQM and then we use the calibrated model to investigate the long-term effects of soil testing for N rate on corn yield and N loss in subsurface drainage from the Walnut Creek Watershed in central Iowa. This watershed data is important because it is one of the very few watershed scale studies that compare the water quality effects of two agricultural treatments. Early in this analysis, incorrect meteorological inputs were identified as a factor that reduced the quality of simulated crop yield, drainage, and N loss. Therefore, quality assurance procedures were implemented. Based on an autoregressive model with monthly time series, after thorough parameter optimization and quality checking the input data RZWQM accurately simulates significantly lower nitrate concentrations in tile drainage from spring soil testing (about 11 mg N/L) compared to fall N application (about 16 mg N/L).