Geostatistical Analyses of Field Spatial Variability of Cotton Yield

Abstract

More information is needed on the spatial variability of soil properties and plant characteristics at the field strip plot experiment scale for accurate evaluation of treatment effect significance. The objective of this study was to examine the pattern and degree of field spatial variability of cotton yield and the relationship between cotton yield and canopy Normalized Difference Vegetation Index (NDVI). A strip plot trial was carried out on a private farm near Brazil, Gibson County, TN from 2009 to 2011. Five side dress N treatments of 0, 45, 90, 134, and 179 kg N ha−1 were imposed on cotton in strip plots under a RCB design with three replications after 45 kg N ha−1 was applied as pre-plant N in the form of chicken litter. Spatial variability was high in lint yield although its pattern and degree varied with year. The correlation of lint yield with NDVI was almost always statistically significant but not strong during early square to late bloom irrespective of year. There was significant global spatial autocorrelation of residual lint yield (N treatment effects on yield excluded) within the test field in 2010 and 2011 based on the Moran’s I statistic. The Localized Indicators of Spatial Autocorrelation (LISA) cluster map showed that there were some significant local clusters of residual lint yield within the field each year. In conclusion, spatial variability needs to be included in data analyses of N treatment effects on cotton yield in strip plot field studies. Cotton yield from farmers’ fields could be expected to have noticeable annual and within field spatial variations in the region, which will significantly influence cotton yields.