Site-Specific Management Zones: Soil-Color Based and Yield-Based.
Rajiv Khosla, Danny Inman, Andrew Hornung, Dwayne Westfall, and Robin Reich. Colorado State University, C013 Plant Sciences Bldg., Dept. of Soil & Crop Sci-CSU, Fort Collins, CO 80523-1170
Development of the variable rate application (VRA) map is an important step in conducting variable rate fertilizer application. Intensive grid soil sampling was among the first methods used to map soil fertility levels and remains a commonly practiced means of generating VRA maps. Grid soil sampling has been widely reported as a method that is both expensive and time consuming. Management zones have been studied as an alternative method of developing VRA maps. Numerous techniques of management zone delineation have been studied; however, few comparisons between techniques exist in the literature. The objectives of this study were: (i) to determine how consistently two management zone delineation techniques (a soil color-based management zone (SCMZ) technique and a yield-based management zone (YBMZ) technique) characterize regions of high, medium, and low grain yield; and (ii) to compare the relative accuracies with which the two management zone delineation techniques characterize the grain yield within low, medium, and high productivity potential management zones. This study was conducted over three site years in Northeastern Colorado. Prior to planting, each season grid soil sampling was performed to measure soil properties used during the yield-based management zone (YBMZ) delineation technique. Management zones were also delineated prior to planting. The SCMZ technique used: (i) bare-soil imagery, (ii) topography and (iii) farmer's experience. These data layers were then incorporated into a geographic information system database to generate mathematical interpolation surfaces to develop three management zones (Khosla et al., 2002). Traits such as regions of dark color, areas of low-lying topography, and areas of historic high yields as reported by the farmer were designated as a zone of potentially high productivity or high zone. The YBMZ relied on: (i) bare-soil imagery, (ii) soil organic matter, (iii) cation exchange capacity, (iv) soil texture, and (v) the previous season's yield map. Surfaces for selected soil properties (soil organic matter, cation exchange capacity, sand, silt, and clay determined from grid based (2.5 sample ha-1) soil sampling) were created using median-polish kriging. Yield monitor data were interpolated using ordinary kriging. Yield maps, aerial imagery, and interpolated soil surfaces were brought to the same spatial resolution using cubic convolution. These data layers were then used with K-means clustering to produce three management zones. Experimental strips were randomly allocated and consisted of 24 rows that spanned the length of the entire field (i.e., > 700 m). Treatments were replicated once and were nested within management zones. Nitrogen treatments were made at the six-leaf crop growth stage using undiluted urea ammonium nitrate 32% applied with an eight-row cultivator. Nitrogen treatments were based on the Colorado State N rate algorithm. Treatments were (i) the recommended N rate, (ii) half the recommended N rate, and (iii) a control of 0 kg N ha-1. The management zone delineation techniques were evaluated using three approaches: (i) farmer's approach, (ii) quantitative approach, and (iii) a subjective approach. The farmer's approach was the simplest. and was based on grain yield production between the two techniques. The quantitative approach used K-means clustering algorithm to group the grain yield into three clusters of high, medium, and low grain yield. Grain yield clusters were then compared to the management zones using percent areal agreement and Kappa statistics and percent areal agreement. The subjective approach involved grouping the grain yield data into three subjectively determined yield classes, classes were compared to the management zones using percent areal agreement and Kappa statistics. Grain yields ranged from 6.9 to 15.5 Mg ha-1 across all site years. Grain yields were significantly different between SCMZ zones for all site years. Based on the farmer's approach, the high and medium zones delineated using the SCMZ exhibited higher grain yields than the corresponding management zones delineated with the YBMZ technique indicating that the SCMZ technique is accurately characterizing the high and medium yielding areas of the field. Using the quantitative approach, Kappa statistics and percent areal agreement were higher for the SCMZ technique across all site years, again suggesting that the SCMZ technique is relatively better than the YBMZ technique. The subjective approach indicated that the SCMZ technique compared better to yield classes than the YBMZ technique. Overall, grain yield in the SCMZ high zones were up to 1.88 Mg ha-1 higher than YBMZ high zones. Based on the approaches used in this study to classify grain yield patterns, the SCMZ technique was found to be relatively better than the YBMZ technique.