SELECTING THE BEST REMOTE SENSING PLATFORM FOR AGRICULTURAL ASSESSMENT USING MULTI-OBJECTIVE ANALYSIS

Abstract

Remote sensing has been identified as a cost-effective technology for site-specific management. This technology offers geographically extensive and continuous assessment of plants, soil and water resources, and other land surface phenomena. Selection of the best available remote sensing technology is determined in most cases by platform and sensor attributes. If these attributes can be quantified, then a multi-objective analysis may be performed to assess quantitatively the tradeoffs between different sensor and platform attributes, identifying the best overall technology. Pioneer work was presented by applying multi-objective analyses to remote sensing technology. Experts were surveyed to identify the best overall technology for agricultural purposes at three different pixel sizes: very fine (<5 cm), fine to moderate (0.5 m – 1.0 m), and moderate to coarse resolutions (0.1 km – 1.0 km). Platform technologies included hand held sensors, booms, remotely piloted vehicles, unmanned aerial vehicles, manned aircraft, Quickbird, Landsat, AVHRR, MODIS, ASTER, and SPOT. Plurality voting, Borda count, Hare system, and pairwise voting were used to analyze survey responses. Results suggest hand held sensors and manned aircraft platforms were favored for applications requiring very fine and fine to moderate spatial resolutions. AVHRR and MODIS were rated equally as the best alternatives for applications requiring moderate to coarse resolution imagery.