Manure Management in an Irrigated Silage Corn Field: Experiment and Modeling

Abstract

On agricultural lands, animal waste disposal as fertilizer has been practiced since the beginning of agriculture. However, the practice has been an environmental concern in recent years due to over disposal of animal waste in some instances. This study evaluated soil NO3 response to beef‐manure application on a corn (Zea mays L.) field and tested the Root Zone Water Quality Model (RZWQM) for manure management. The experiment site was located in Northeastern Colorado on a silage‐corn field with a history of fertilization with beef manure every fall after corn harvest. To study the residual effect of long‐term manure application, 582 kg ha‐1 of manure‐N was applied to the east side of the field in the Fall of 1993, 1994, and 1995, while the west side received manure in 1993 only. Average silage‐corn yields from the west site were 25.4, 31.9, and 22.5 Mg ha‐1 for 1994, 1995, and 1996, respectively, which were not significantly different from that harvested from the east site (25.1, 30.9, and 24.3 Mg ha‐1, respectively). Average soil NO3 concentrations decreased significantly from 14.9 to 8.5 mg N kg‐1 in the top 30 cm of soil, and from 5.4 to 3.7 mg N kg‐1 in the 30‐ to 60‐cm soil profile after stopping manure application. No significant difference in soil NO3 concentrations between the manured and not‐manured sites was found below 60 cm. Average plant N uptake ranged from 140 to 362 kg N ha‐1 and was not significantly different between the two sites. The RZWQM was calibrated on the basis of the measured silage‐corn yield and plant N uptake, and was then used to predict soil NO3 concentration and total water storage in the soil profile. Generally, the calibrated model provided adequate predictions for both NO3 and soil water content with r2 > 0.83. The model was further used to evaluate alternative scenarios of manure and water management.

Simulated N management effects on corn yield and tile-drainage nitrate loss

“…The model overpredicted Nloss and Nconc compared to field data for fall applied swine manure prior to both corn and soybean (Table 4 and Fig. 2), possibly because RZWQM over-predicts winter nitrification (Ma et al, 2007a-this issue;1998). Predicted average October root zone soil nitrate after soybean was 76 kg N/ha for fall swine manure application prior to corn and soybean and 33 kg N/ha for fall application prior to only corn (150 kg N/ha).…”

“…2). Over-predicted Nconc could result from RZWQM over-predicted winter nitrification (Ma et al, 2007a-this issue;1998), which may be especially problematic under fall injected swine manure application to soybean as discussed below.…”

“…(Randall and Vetsch, 2005a, b). If RZWQM over-predicts winter nitrification (Ma et al, 2007a-this issue;1998), the benefit of fall AA compared to fall UAN to reduce Nloss and increase corn yield is under-predicted and the difference between fall and spring AA Nloss is over-predicted.…”

Sensitivity of tile drainage flow and crop yield on measured and calibrated soil hydraulic properties

“…These estimates of soil hydraulic properties were adequate in some applications of RZWQM (Ma et al, 1998;Starks et al, 2003), while they were inadequate under other conditions (Malone et al, 2004). Although it was recommended to use measured values whenever it was possible (Ma et al, 1998), it was a major concern for model users on how to deal with the experimental errors and spatial variability in soil hydraulic property measurements since most models took only one value rather than a distribution for a given parameter.…”

“…Although it was recommended to use measured values whenever it was possible (Ma et al, 1998), it was a major concern for model users on how to deal with the experimental errors and spatial variability in soil hydraulic property measurements since most models took only one value rather than a distribution for a given parameter. Users also would like to know how the experimental errors in input parameters were transferred to errors in model outputs.…”