Precision nitrogen management across site-specific management zones in irrigated maize production systems

Abstract

In the United States, crop nitrogen-use efficiency (NUE) is very low. Approximately 33% of all N applied towards cereal crop production is captured in the harvested grain. Precision agricultural practices have shown potential for increasing crop NUE. This dissertation investigates three aspects of precision N management: management zones, aerial remote sensing, and active remote sensing. The specific objectives were: (i) to characterize the within field spatial variability of N uptake across irrigated com production fields, (ii) to quantify and compare N uptake and grain yield across three site specific management zones (SSMZs), (iii) to compare grain yield response to applied N between management zones, (iv) to examine the relationships between normalized difference vegetation index (NDVI) determined early in the growing season, site-specific management zones, and relative maize yield; (v) to determine if NDVI can be used to estimate relative maize yield; (vi) to determine if site-specific management zones can be used in conjunction with remote sensing to provide yield estimates in irrigated maize, and (vii) to evaluate the effectiveness of using a hand-held active remote sensing instrument to estimate yield potential in irrigated maize. This study was conducted on commercially-operated irrigated production maize fields throughout northeastern Colorado. For objectives i, ii, and iii, fields were classified into high, medium, and low site specific management zones. Treatments consisted of a control and two uniform N application rates over three site years (one field over two consecutive years and another field over one year). Nitrogen fertilizer rates varied with site year and ranged from 56 kg N ha-1 to 268 kg N ha-1. Above ground biomass samples were collected at physiological maturity and analyzed for total N. For objective iv, v, and vi, aerial imagery was acquired at approximately the eight-leaf crop growth stage. Grain was harvested using a commercial-combine outfitted with a yield monitor at the crop's physiological maturity. Objective iv was analyzed using percent areal agreement, kappa statistics, and regression analysis. Objectives v and vi were analyzed using regression analysis with cross-validation and indicator variables. For objective vii, the GreenSeeker™ active remote sensing unit was used to measure red and near infrared reflectance of the crop canopy. Grain was harvested by hand at physiological maturity. Presidedress soil nitrate samples (PSNT) were collected from the plots at the time of sensing. The crop was hand-harvested at physiological maturity. NDVI was calculated from the reflectance data and normalized by dividing by the number of growing-degree days from planting to sensing. A response index (RI) was calculated the ratio of the reflectance of an area of interest to the reflectance of an N-rich portion of the field. Regression analysis was used to model grain yield. Cross validation was used to validate regression models. Nitrogen uptake and grain yield within management zones was found to be less spatially variable than the whole field. Nitrogen uptake, grain yield, and grain yield response to applied N were found to be statistically different (p < 0.05) across management zones. NDVI and grain yield had a slight to substantia] areal association, with kappa statistics ranging between 0.10 to 0.63 and % areal agreement from 13 to 67. Models estimating grain yield from NDVI had coefficients of determination as high as 82%. Management zones resulted in only marginal improvements in the yield estimations. A strong relationship was found between NDVI determined from the GreenSeeker and observed grain yield (R2 = 0.76). Overall, results from this dissertation highlight the potential of site-specific management zones, aerial remote sensing, and active remote sensing to characterize N needs and/or yield limiting factors across irrigated maize fields.