Note: You are reading this message either because you can not see our css files, or because you do not have a standards-compliant browser.

LOGO: Journal of Cotton Science


Application of Remote Sensing to Strategic Questions in Cotton Management and Research

Authors: Richard E. Plant, Daniel S. Munk, Bruce R. Roberts, Ronald N. Vargas, Robert L. Travis, D. William Rains, and Robert B. Hutmacher
Pages: 30-41
Molecular Biology and Physiology

Remote sensing can be a relatively inexpensive source of data for site-specific crop management. Many potential applications of remote sensing are tactical, in that they involve responses to particular conditions or situations that arise during the course of the season. Other potential applications, however, both in research and in management, involve strategic questions that concern the integrated whole of the crop production system. Strategic decision making generally occurs before the season begins. The use of remotely sensed images in addressing strategic vs. tactical questions differs in that strategic questions may involve patterns of spatial variability only and tactical questions may involve temporal as well as spatial variability. This difference may have several practical consequences. In the strategic use of remote-sensing data, extreme speed in the delivery of the image or image data after acquisition may be unnecessary. Speed may be necessary for tactical management uses. Using remote sensing in strategic situations may not require calibrated image data and may not require as many images. If true, this could result in considerably lower data-collection costs. The objective of this research was to seek answers to two questions: (i) can uncalibrated data be used for strategic management? (ii) what is the inter-temporal relationship among sequences of images? Analyses of chronological sequences of images of irrigation and N stress indicate that uncalibrated data are useful for addressing strategic questions that involve spatial variability of crop status and that locations within the field are highly autocorrelated, so that relatively few images are necessary to determine crop spatial reflectance properties.