U.S. Department of Agriculture: Agricultural Research Service, Lincoln, Nebraska
Document Type
Article
Date of this Version
2013
Citation
Precision Agric (2013) 14:71–85; DOI 10.1007/s11119-012-9301-6
Abstract
Active crop canopy sensors make possible in-season fertilizer nitrogen (N) applications by using the crop as a bio-indicator of vigor and N status. However, sensor calibration is difficult early in the growing season when crops are rapidly growing. Studies were conducted in the United States and Mexico to evaluate procedures to determine the vegetation index of adequately fertilized plants in producer fields without establishing a nitrogen-rich reference area. The virtual-reference concept uses a histogram to characterize and display the sensor data from which the vegetation index of adequately fertilized plants can be identified. Corn in Mexico at the five-leaf growth stage was used to evaluate opportunities for variable rate N fertilizer application using conventional tractor-based equipment. A field in Nebraska, USA at the twelve-leaf growth stage was used to compare data interpretation strategies using: (1) the conventional virtual reference concept where the vegetation index of adequately fertilized plants was determined before N application was initiated; and (2) a drive-and-apply approach (no prior canopy sensor information for the field before initiating fertilizer application) where the fertilizer flow-rate control system continuously updates a histogram and automatically calculates the vegetation index of adequately fertilized plants. The 95-percentile value from a vegetation-index histogram was used to determine the vegetation index of adequately fertilized plants. This value was used to calculate a sufficiency index value for other plants in the fields. The vegetation index of reference plants analyzed using an N-rich approach was 3–5 % lower than derived using the virtual-reference concept.