Spatial Variability of Soil Properties, Cotton Yield, and Quality in a Production Field

Abstract

Successful precision agriculture requires an understanding of spatial variability of soil properties, crop growth, and their interactions. The objectives of this study were 1) to examine the spatial variability of soil properties, cotton lint yield, and fiber quality and 2) to evaluate the spatial variability within mapping units at a production‐field scale. This research was conducted on an irrigated 49‐ha cotton field in Texas from 1998 through 2000. Samples were collected from regular 1‐ha grids, triangular, and random points. Results indicated that soil properties had stronger spatial dependence than did cotton lint yields and fiber quality. Soil properties, except nitrate (NO3 −) nitrogen (N) and Olsen phosphorus (P), were strongly spatially dependent, whereas lint yield was moderately to strongly spatially dependent. Fiber quality was moderately spatially dependent. Spatial appearance of higher yield in drier years was associated with the distribution of soil properties favoring cotton growth, including lower pH and calcium (Ca) and higher P and sand content. There were differences in the means and spatial structure parameters of the selected measurements among the three soil mapping units, suggesting the need for careful sampling design to make reasonable estimation of soil properties under a nonstationary situation. Furthermore, the spatial variability of soil properties and cotton yield were intensified under limited water supply, implying that site‐specific cotton management may be appropriate for cotton producers in the Texas high plains. Contribution of the College of Agricultural Sciences and Natural Resources at Texas Tech University Scientific J. Series Paper Number T‐4‐572.