1a. Objectives (from AD-416)
(1) Determine crop coefficients, adjustment algorithms, and transfer capabilities for improving crop water use estimation of common and alternative crops in arid Southwestern U.S. climates; (2) Develop and verify remote sensing methods and techniques for predicting near real-time evapotranspiration and plant water stress at spatial scales relevant for single fields to watersheds; (3) Develop high resolution remote sensing decision support tools for managing spatially and temporally variable water and nutrient applications to crops.
1b. Approach (from AD-416)
Remote sensing experiments will be conducted in small plots and on production-size fields at The University of Arizona, Maricopa Agricultural Center (MAC). Remote sensing data, including reflectance and emittance, will be collected at regular intervals during the crop growing seasons with aircraft-mounted sensors, as well as with ground-based, handheld instruments. Meteorological and air quality parameters will be obtained from micrometeorological stations installed within the fields. Assessment of air quality at experimental sites will be made in conjunction with remote sensing data collection times. Reference evapotranspiration (ET) data will be provided by local weather stations of the Arizona Meteorological (AZMET) Network. Field data collection will include measurements of soil water contents, soil physical and chemical properties, crop ET, and irrigation water applications. Plant growth observations will include crop growth stage, percent canopy cover, biomass, and crop yield.
3. Progress Report
The wheat plot experiment was completed in FY2009. Data from this experiment will be used to validate our remote sensing crop coefficient and surface energy balance approaches for estimating spatially distributed wheat ET. The experiment also tested adjustment procedures for developing a standard FAO-56 crop coefficient curve for a given plant density. Water stress treatments, also included in the experiment, allowed rigorous testing of crop growth models over a range of water stress conditions. Measurements during the wheat plot experiment included frequent monitoring of NDVI with both passive and active ground-based sensors and periodic collection of aerial remote sensing imagery. Evaluation of these experimental data will be used to improve our wheat irrigation scheduling procedures using remote sensing and crop growth models, approaches that will be applied during the large wheat field experiment planned for FY2011-12. The large-field cotton experiment will be completed in October 2009. Cotton irrigations in large borders are scheduled based on spatial information of crop soil water needs. These estimations are determined from three primary sources of data collected in the field. These include, periodic monitoring of spatially distributed NDVI using a tractor-mounted sensor and aerial survey, spatial information of soil water holding based on electrical conductivity surveys, and estimation of furrow infiltration using field data combined with simulation of furrow irrigation hydraulics. Procedures and algorithms were developed for incorporating these data within a spatial irrigation model developed in collaboration with the University of Arizona. Four treatments included in the experiment are: two treatments using NDVI-based crop coefficients, one treatment using a standard crop coefficient curve, and one treatment based on a farm management irrigation schedule. Twenty-four, three-band radiometers built by the CRIS were installed in the cotton field as part of a wireless mesh network system. This season we will evaluate the performance of these radiometers and their ability to provide continuous quality NDVI data at fixed locations in the large field. During FY2009, the CERES-Wheat plant growth model was evaluated for local conditions using experimental data from previous wheat experiments. The model was evaluated for its ability to simulate plant growth and soil moisture conditions under varying irrigation, plant population, and nitrogen levels. Techniques were also explored for implementing the model as a real-time predictor of crop water use for irrigation scheduling. Data assimilation techniques were tested to determine whether model simulations of ET and wheat yield could be improved by updating the model using remote sensing-based estimates of leaf area index. Spectral reflectance characteristics of lesquerella canopies were evaluated during a collaborative 2009 field experiment in Maricopa. During FY2009, team scientists met with collaborators who took part in the 2008, multi-agency remote sensing field campaign in Bushland, TX. The meeting established data sharing and manuscript authorships.
1. A remote sensing method to help manage efficient irrigation of Crops. A measure called a crop vegetation index can be obtained for every section of a grower’s field with remote sensing imagery. This information provides a valuable input for managing efficient and precise irrigation to crops. Although remote sensing from satellites and aircraft can provide the vegetation index data, the number of times aerial imagery is needed during the season is not cost-effective. ARS scientists at Maricopa, AZ used a statistical procedure to select the best field locations for placing 12 low-cost sensors to collect daily vegetation index data. Information from this small number of field sensors accurately predicted the vegetation index at all 17,000 locations within the field. The research demonstrates that a few well-placed sensors can provide an effective tool for irrigation water management.
2. Remote sensing technique to improve simulations of crop water use and crop yield. Practical application of crop growth models for support of farm management decisions has been limited by the vast model input requirements and the model sensitivity to input parameter uncertainty. ARS scientists at Maricopa, AZ developed a technique to reduce this sensitivity by updating model simulations with estimates of leaf area index from remote sensing data. The technique improved crop model simulations of both yield and crop water use by 2% to 4%. Incorporation of this technique into operational decision support systems for regional and field-scale crop assessments is necessary to improve management of water and nutrients and to forecast crop yields.
3. Land temperatures for modeling water use. Accurate high-resolution remote sensing techniques can improve measurement and forecasting of agricultural water use. One promising technique is to measure surface temperatures of land from space. ARS researchers at the Water Management and Conservation Unit in Maricopa, Arizona, in collaboration with NASA, developed and implemented ways to combine satellite observations that result in hourly surface temperature images over regional arid land areas in the U.S. Southwest and Australia. The research provided detailed temperatures that could be helpful to hydrologists, climatologists, and agronomists who need better ways to monitor water use at regional to global scales.
Inamdar, A.K., French, A.N. 2009. Disaggregation of goes land surface temperatures using surface emissivity. Geophysical Research Letters. Vol. 36, L02408, doi:10.1029/2008GL036544.