Submitted to: Agronomy Journal
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: December 27, 2010
Publication Date: March 9, 2011
Citation: Jaradat, A.A., Weyers, S.L. 2011. Statistical modeling of yield and variance instability in conventional and organic cropping systems. Agronomy Journal. 103(3):673-684. Interpretive Summary: Cropping systems in the northern Corn Belt have been changing over time in response to several interacting factors. The current system, which is based on corn-soybean crop rotation, conventional tillage, and chemical inputs, may not be sustainable in the long run. We evaluated the impact of current and alternative management approaches and several phases of crop rotation options on crop rotation yield and its variation over four years under conventional and organic cropping systems. We developed a classification scheme of cropping systems, crop rotations, and management practices to help identify appropriate combinations of these factors that would result in large and stable yields. The scheme can be used as a guideline to identify stability in crop production, to plan the use of appropriate management practices for a given cropping system or crop rotation, and to maximize the chances of obtaining the largest and most stable yield over time. Results of the study are useful in identifying causes of yield variation under certain soil conditions and management practices, and will help researchers, crop consultants, and farmers optimize input use, maximize the ability to detect true responses to management factors, determine trends and understand changes in yield over time, and assess long-term sustainability of various cropping systems.
Technical Abstract: Cropping systems research was undertaken to address declining crop diversity and verify competitiveness of alternatives to the predominant conventional cropping system in the northern Corn Belt. To understand and capitalize on temporal yield variability within corn and soybean fields, we quantified and modeled the cumulative effects of management practices and covariates (physical, chemical and biological variables, as surrogates of spatial variation) on total rotation yield (TY), temporal yield variances (TYV) and coefficients of variation (CV) under conventional (CNV) and organic (ORG) cropping systems using all phases of a two-year (corn-soybean) or a four-year (corn-soybean-wheat-alfalfa) crop rotation, conventional or strip-tillage, and with or without fertilizer. To help determine yield goals and plan field operations when faced with significant temporal variation, we then modeled TY as a function of TYV or CV. Soil covariates differed as to their impact on TY, TYV, and CV, in both cropping systems. However, spatial variation, as quantified by soil covariates, did not fully explain variation in TY or TYV. TYV explained up to 86% of variation in TY, both of which were less variable in ORG than in CNV. Multivariate relationships identified between TYV, CV, management factors, and soil covariates using partial least squares regression indicated that TYs of the four-year crop rotation were more stable than the predominant corn-soybean crop rotation, especially under ORG. The relationship between TY and TYV, but not between TY and CV differed among and within cropping systems; it was positive, but not always significant, and less statistically significant in ORG than in CNV. The largest and most stable yields were obtained when TYV and CV, respectively, reach their optimum and minimum values. We developed a classification scheme of cropping systems, crop rotation phases, and management practices based on the three-way relationship between TY, TYV and CV, and deviations from their respective means. In addition to its utility in selecting the largest and most stable yield, the scheme can be used as a measure of stability in crop production, and to plan strategic deployment of appropriate management practices for a given crop rotation or cropping system.