WITHIN FIELD VARIABILITY IN WHEAT GRAIN YIELDS OVER NINE YEARS IN OKLAHOMA

ABSTRACT Wheat grain yield levels are known to vary from 1 year to the next and this variability is frequently attributed to differences in rainfall. However, within field variability of wheat grain yields as a function of time has not been extensively evaluated. Wheat grain yields from selected fields in Oklahoma were monitored over a 9-year period using satellite imagery. Yields for each 25 m × 25 m area within each field were estimated from NDVI measurements obtained from LANDSAT scenes in north-central Oklahoma. Minimum, maximum, mean, standard deviation, and coefficient of variation (CV) values were collected from six locations from 1991 to 1999. Coefficient of variations for wheat grain yield ranged between 16 and 38% for the same field. The wide range in CVs could be partly explained by the changes in average grain yield where CVs tended to be greater when mean yields were lower. Because CVs for the same field had such a wide range (from year to year), these results suggested that the expression of spatial variability was a function of the environment in which wheat was grown. Therefore, if within field CVs could be predicted (mid-season satellite images of variable growth using NDVI), the potential response to added nutrients may also be established, and in-season nutrient additions adjusted accordingly. Furthermore, knowledge of the CV mid-season for a particular field could be equated to the response index which various researchers have used to determine topdress fertilizer needs. Contribution from the Oklahoma Agricultural Experiment Station.