Skip to main content
ARS Home » Research » Publications at this Location » Publication #102332
Title: ASSESSING SOURCES OF CORN FIELD VARIABILITY USING AIRBORNE IMAGERY COINCIDENT WITH SURFACE MEASUREMENTS

Author
item Walthall, Charles
item LOECHEL, SARA - UNIVERSITY OF MARYLAND
item Dulaney, Wayne
item Daughtry, Craig
item WATSON, LEE - 3DI

Submitted to: Proceedings of the International Airborne Remote Sensing Conference
Publication Type: Proceedings
Publication Acceptance Date: 10/1/1999
Publication Date: N/A
Citation: N/A

Interpretive Summary: Successful remote sensing image interpretation for agricultural applications requires an understanding of the agronomic sources of image tone variability. An experiment was conducted to compare airborne imagery with interpolated surface measurements of corn foliage density and mean leaf tip angle (MTA: an indicator of stress and a contributing factor of vegetation reflectance) and a digital elevation model (DEM) of the field. Comparisons between the foliage density and the imagery showed some correspondence. Better correspondence between the imagery and the DEM was found: areas of higher NIR reflectance (indicative of higher productivity) coincided with low-lying areas of the field. Areas of lower foliage density coincided with areas of greater tonal variability of the imagery. The relationships between MTA and the imagery were less clear. This analysis suggests that stratification or correction of imagery for topography may be useful for isolating problem areas and explaining yield variability.

Technical Abstract: Successful remote sensing image interpretation for agricultural applications requires an understanding of the agronomic sources of image tone variability. A coordinated airborne and surface data collection was conducted on a 6 ha cornfield in Beltsville, MD. Surface measurements of foliage density and mean leaf tip angle (MTA: an indicator of stress and a contributing factor of canopy reflectance) were measured at 64 locations within the field, interpolated and rasterized to an image that could be geo-registered with the airborne imagery. A digital elevation model (DEM) for the field was also co-registered with these data. Qualitative comparisons between the foliage density and the imagery showed some correspondence. Even better correspondence between the imagery and the DEM was found: areas of higher NIR reflectance coincided with low-lying areas of the field. Areas of lower foliage density coincided with areas of greater tonal variability of the imagery. The relationships between MTA and the imagery were less clear. This analysis suggests that stratification or correction of imagery for topography may be useful for isolating problem areas and explaining yield variability.