Project Number: 5356-13210-002-02
Project Type: Trust

Start Date: Jun 01, 2006
End Date: Aug 31, 2008

Objective:
This cooperative research project will evaluate John Deere's Optigro multispectral imaging system to meet the high resolution, high precision data needs and operational conditions of precision nitrogen (N) management. Specific objectives include: 1) quantify the relationship between multispectral canopy reflectance and mid-season N status in hard red spring wheat; 2) use the spectral reflectance-crop N relationship (from Obj. 1) to improve the prediction of grain protein response to late season foliar N application, grain protein differences within fields as needed to develop harvesting zones of low vs. high quality grain, and biomass and straw yield levels at harvest; and 3) assess accuracy of on-combine optical sensors for measuring protein concentration at harvest.

Approach:
Field experiments with hard red spring wheat will examine the viability of multispectral remote sensing to predict crop N status, straw yield and grain protein, and grain protein response to late-season N application. This work will augment ongoing work and be undertaken by D. Long of the USDA-ARS Columbia Plateau Conservation Research Center (Pendleton, OR) who has expertise in remote sensing and precision nitrogen management. Experiments will consist of a strip trial design with foliar N treatments of zero and 20- or 40-lb N/ac of UAN solution (32% N) applied between heading and flowering in wheat. The width of these strips will be in accordance with the spray boom width of typical liquid-fertilizer applicators. An aircraft or ground vehicle equipped to handle liquid fertilizer applications will be used. Early season soil samples will be collected for determination of NO-3-N and other relevant soil characteristics. Soil samples will be positioned at numerous points on a regular grid to capture a wide possible range in soil characteristics within the field. Flag leaf N content, SPAD chlorophyll, and canopy reflectance will be measured before and after application of foliar N. Nitrogen content in the plant tissue will be determined via dry combustion analysis. These variables will be correlated with image reflectance as needed to derive linear regression models predicting crop N at mid-season, and protein and straw yield at harvest. Crop reflectance models will be used to develop maps of biomass and harvesting zones. Sequential imagery will also be used to direct crop scouting to anomalous areas within planted fields where it is important to identify causes of the crop variability at early, middle, and late stages of growth that may be due to pests, poor fertility, shallow/eroded soils, and other deleterious factors. Variables to be measured include stand density, soil depth, soil water, soil N, crop N, and canopy reflectance. If possible, some of this information may serve as a set of GPS-reference test observations that can be compared with modeled crop N attributes derived from experimental data. Measurements of grain yield and grain protein will be obtained at harvest using a combine equipped with a GPS receiver, yield monitor, and grain quality sensor. Sensors will include the in-line ProSpectraä instrument and on-line Cropscan instrument. Grain will also be hand sampled from the combine, laboratory analyzed for protein, and compared with sensor readings to determine accuracy. Grain protein response will be determined by contrasting between the zero treatment and N treatments. Documents Trust with Deere and Company. Log 31738. Formerly 5356-13210-001-02T (11/06).