ASSESSING CORN YIELD AND NITROGEN UPTAKE VARIABILITY WITH DIGITIZED AERIAL INFRARED PHOTOGRAPHS

Methods for large-scale mapping of crop variability are needed for pre cision farming applications. We hypothesized that aerial-infrared phot ographs can predict corn yield and N uptake variability. A 2.6-ha area in Minnesota was cropped with corn in 1991 and 1992. Photographs of b are soil and of the mature crops were color scanned. Corn yield and N uptake were determined at 58 locations, and were related to phototone data using multiple regression. Spatial analysis of imagery showed tha t the crop canopy was more influenced by soil conditions in 1991 than in 1992. This is attributed to the cooler climate in 1992. Consequentl y, harvest data were better predicted in 1991 than in 1992. In 1991, 6 5 percent of yield and 59 percent of N-uptake variability were capture d by phototones; respective values for 1992 data were 47 percent and 3 7 percent. Use of aerial photographs for spatial modelling of crop gro wth can work, but multi-year studies are recommended.