Skip to main content
ARS Home » Research » Publications at this Location » Publication #91013
Title: REMOTE SENSING FOR THE ENHANCEMENT OF CROP SIMULATION MODELING EFFORTS

Author
item Barnes, Edward
item Pinter Jr, Paul
item Moran, Mary
item Clarke, Thomas

Submitted to: Biological Systems Simulation Group Proceedings
Publication Type: Abstract Only
Publication Acceptance Date: 3/19/1997
Publication Date: N/A
Citation: N/A

Interpretive Summary:

Technical Abstract: Techniques to estimate the physical characteristics of plants using multispectral observations are well established; therefore, these observations can be very useful for both validation of and incorporation in crop simulation models. Non-destructive sampling allows the ability to revisit the same location during the course of the season and to assess larger areas than would be possible through traditional approaches. Measurements of red and near-infrared reflectance can be related to plant canopy properties such as leaf area index (LAI). The possibility also exists to empirically derive other relationships between plant components and reflectance data for particular sites. These relations can then be used to rapidly and non-destructively sample other treatments and areas of the field. Similar techniques can be also applied with image-based data. Remotely-sensed images of bare soils can be used to interpolate soil data and aid in forming this important base layer of a geographic information system. Furthermore, images acquired at different times during the season can be used to infer canopy density and reset the model's predictions to agree with actual field conditions. In this manner a means is provided to incorporate the actual field-scale variability into the model, leading to more realistic analysis of projected yield.