Late-season digital aerial imagery and stalk nitrate testing to estimate the percentage of areas with different nitrogen status within fields

Abstract

Precision agriculture technologies offer potential economic and environmental benefits from site-specific management of nitrogen (N) fertilizer and animal manure sources for corn (Zea mays L.). However, lack of knowledge and reliable methodology for developing and evaluating site-specific N fertilizer recommendations are the major obstacles for realizing these potential benefits. The objective of this study was to evaluate corn N status at the field scale and across many fields using late-season digital aerial imagery and the end-of-season corn stalk nitrate test in large-scale on-farm evaluation studies. About 30 groups of farmers, lead by agronomists and crop consultants, were formed across Iowa to evaluate different N management practices. Late-season color digital aerial imagery and digital soil maps were used to guide the collection of the corn stalk nitrate test samples within 683 cornfields in 2006, 824 fields in 2007, and 828 fields in 2008. Four areas—one from each of the three predominant soil types and one within a target-deficient area—were sampled in each field. Multilevel binary logistic regressions were used to quantify the relationship between green reflectance of the corn canopy and corn N status, expressed as deficient and sufficient (a com- bination of marginal, optimal, and excessive categories of the corn stalk nitrate test), within and across the fields. Percentages of areas within fields with deficient and sufficient N status were estimated using distributions of pixel counts of green reflectance of the corn canopy. Multiple regression analysis was used to identify factors affecting percentage-deficient area within the fields. Results showed that N management category (a combination of N form and timing of application) and early season rainfall (May, June, or cumulative from March through June) had the largest effects on percentage-deficient area. Fields with liquid swine manure applied in the fall or urea-ammonium nitrate solution applied in the spring (before planting) or urea-ammonium nitrate solution applied at sidedress had larger areas of N defi- ciency than fields with anhydrous ammonia applied in the spring. Larger early season rainfalls also increased percentage-deficient area during each year. The results of large-scale evaluations can be used to develop more accurate site-specific N recommendations based on knowledge of differences between management practices and effects of soil properties and rainfall on N status within fields. Future evaluations can identify areas that persistently have excessive N status and quantify potential N fertilizer reductions within those areas or fields.