The relationships between the crop moisture index and wheat production in Kansas

Abstract

One crop reporting district (CRD) in each of the three former climatic divisions of Kansas is picked for agroclimatological analysis of the annual wheat yields per planted, non-irrigated hectare, and the weekly Crop Moisture Index (CMI) values. The goal of this research is to determine an equation giving a relationship between the weekly CMI values and the annual wheat yields. In particular, the question investigated is how crops respond to moisture status (linearly and/or non-linearly). The first part of this research is a verification of the CMI values. The results of this first analysis show that the measured soil moisture contents at various soil layers correlated highly with the CMI values, the correlation coefficients on the average being about 0.90. In particular, the moisture content of the second 30 cm of the soil layer is found to be the most highly correlated with the CMI values. The second part of this study investigates the impact of technology on wheat yields. A highly significant difference is found between the mean of the wheat yields before and after World War II, and this difference is attributed to technological adaptations over the years. More accurate predictions of wheat yields are made on the basis of technological trends (either linear or non-linear) than those based on the mean of the wheat yields after World War II. Furthermore an arctangent technological function is found to be better than a linear equation as an estimator of the average technological component of wheat yields. The difference, however, is not statistically significant. With reference to the question of linear or non-linear crop response to moisture status, it is found that the closest estimates of wheat yields are obtained by relating wheat yields to CMI values (a) linearly in a CMI range of about -2.5 to 1.25 and (b) non-linearly below and above these CMI values. Optimum non-linear relationships below and above the CMI values are derived.