Skip to main content
ARS Home » Midwest Area » Columbia, Missouri » Cropping Systems and Water Quality Research » Research » Publications at this Location » Publication #134882
Title: USING AIRBORNE HYPERSPECTRAL AND SATELLITE MULTISPECTRAL DATA TO QUANTIFY WITHIN-FIELD SPATIAL VARIABILITY

Author
item HONG, SUK - KOREAN RURAL DEV ADM
item Sudduth, Kenneth - Ken
item Kitchen, Newell
item PALM, H - UNIV OF MO
item WIEBOLD, W - UNIV OF MO

Submitted to: Korean Journal of Precision Agriculture
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/1/2002
Publication Date: 6/1/2002
Citation: Hong, S.Y., Sudduth, K.A., Kitchen, N.R., Palm, H.L., Wiebold, W.J. 2002. Using airborne hyperspectral and satellite multispectral data to quantify within-field spatial variability. Korean Journal of Precision Agriculture. 1(1):82-98.

Interpretive Summary: Precision farming relies on the ability to efficiently and economically collect and interpret data describing the variability within cropped fields. Remote sensing from satellites or airplanes is one approach to obtaining the data needed for precision farming. Depending on the wavelength bands of data collected and the ground conditions, remote sensing data may be used to examine variation in soil properties, crop condition, or crop yield. Multispectral images, where only a few data bands are obtained, are available commercially from both aircraft and satellites. Recently hyperspectral remote sensing data, which includes tens to hundreds of channels, has become available to the agricultural research community. While this large amount of data may allow us to develop better relationships between the remote sensing data and actual field conditions, it also greatly increases the complexity of the data processing and interpretation procedure. In this study, we examined both aircraft hyperspectral and satellite multispectral data collected over two Missouri corn/soybean fields on multiple dates in 1999 and 2000. A number of statistical procedures were employed to relate the remote sensing data to soil properties and crop yield. We found that bare soil images obtained in the spring were significantly correlated with soil properties. We also found that image data acquired at the proper growth stage (usually in August for these fields and crops) were highly correlated with grain yield. This research will benefit other researchers, producers, and consultants who may be interested in using remote sensing data to understand crop and soil differences for precision agriculture.

Technical Abstract: The relationship between hyperspectral and multispectral remotely sensed images and ground based soil and crop information was investigated for two central Missouri experimental fields in a corn (Zea mays L.)soybean (Glycine max L.) rotation. Multiple airborne hyperspectral and IKONOS satellite images were obtained during the 1999 and 2000 growing seasons. Hyperspectral images covered 120 bands from 457 to 823 nm with a spatial resolution of 1 m. Multispectral IKONOS images included four bands (blue, green, red, and near-infrared) with a 4 m spatial resolution. Geometric distortion of the pushbroom-type hyperspectral sensor caused by aircraft attitude change during image acquisition was corrected with a rubber sheeting transformation. Within-field data collection included crop yield, soil electrical conductivity (ECa), and soil chemical properties. Simple correlation, multiple regression, and principal component analysis were used to identify those remotely sensed data most highly related with field measured soil and crop properties. Blue wavelengths were most highly correlated with ECa measurements. For corn, the early reproductive stage provided the best relationships between final yield data and spectral signatures in both years. For soybeans, yield data were highly correlated with wavelengths in the near infrared region from August images in both 1999 and 2000. Maps estimating soil ECa and corn yield from hyperspectral and multispectral images were derived.