Skip to main content
ARS Home » Research » Publications at this Location » Publication #154074
Title: CROP CONDITION AND YIELD SIMULATIONS USING LANDSAT AND MODIS IMAGERY

Author
item Doraiswamy, Paul
item Hatfield, Jerry
item Jackson, Thomas
item Prueger, John
item AKHMEDOV, BAKHYT - USDA ARS CONTRACTOR
item Stern, Alan

Submitted to: Remote Sensing of Environment
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/24/2003
Publication Date: 9/1/2004
Citation: Doraiswamy, P.C., Hatfield, J.L., Jackson, T.J., Prueger, J.H., Akhmedov, B., Stern, A.J. 2004. Crop condition and yield simulations using Landsat and MODIS imagery. Remote Sensing of Environment. 92:548-559.

Interpretive Summary: The use of operational satellites for monitoring crop condition during the season and predicting yields at regional scales is an important for USDA. Accurate and timely assessment of crop yields is used for studying the economic impact of agricultural production on the world market. The National Agricultural Statistics Service (NASS) of USDA is responsible for providing the crop production estimates for the U.S. and we have been working in cooperation with NASS to develop methods to assess seasonal crop progress using satellite imagery and predicting crop yields before the harvest. Imagery from the new MODIS sensor onboard the Terra satellite offers an excellent opportunity for daily coverage at 250 m resolution. A field study was conducted in the predominantly corn and soybean area of central Iowa during the 2002 crop season. Integrating information derived from satellite imagery was combined with weather driven crop yield models to predict the crop yields. The results of the simulations provided a spatial distribution of corn and soybean yields. There was a good agreement between predicted yields and NASS county average estimated yields.

Technical Abstract: Monitoring crop condition and yields at regional scales using imagery from operational satellites remains a challenge because of the problem in scaling small scale yield models to the larger regional scales. NOAA AVHRR satellite imagery has been traditionally used to monitor the vegetation dynamic and incorporated into indirect assessments of crop condition and yields. The 1 km spatial resolution of NOAA AVHRR with two primary spectral bands is not adequate for monitoring crops at field level. Imagery from the new MODIS sensor onboard the Terra satellite offers an excellent opportunity for daily coverage at 250 m resolution, which is adequate where field sizes are larger than 36 ha. The objective of this research is to investigate the applicability of the 8-day MODIS composite data in the operational assessment of crop condition and yield was conducted as part of the SMEX02 soil moisture investigations in Iowa during the 2002 crop season. A field study was conducted in the predominantly corn and soybean area of Iowa. Ground-based canopy reflectance and LAI measurements were made to calibrate the models at local scales. The MODIS 250 m resolution data was used in a radiative transfer model to map the progression of leaf area index (LAI) through the season. LAI was integrated in a climate-based crop simulation model to scale up from local simulation of crop development and responses to a regional scale. The weekly changes in soil moisture are also mapped as part of the crop simulation model. Temporal and spatial variability of soil moisture is one of the critical parameters that influences the crop condition and yields. The spatial distribution of yields were mapping at 250 m resolution and the mean of simulated yields for corn overestimated the reported yields by about 3 % and the soybean crop was underestimated by 6.6 % of the NASS reported yields.