REMOTE SENSING

Remote sensing of environment

Remote sensing of environment

Introduction

Remote feeling approximates of the exterior power balance and evapotranspiration (ET) in specific, provide a means to consider, in a spatially circulated manner, crop situation and end-of-season yields at large spatial levels (Moran et al., 1995 and Moulin et al., 1998). Such spatial information would furthermore be very helpful to agronomists who often assess the consequences of crop genotype and management practices in terms of the ratio of yield per a unit locality to water use or ET to make such a yield (Gregory et al., 2000), routinely called the crop's water use efficiency (WUE). To estimate WUE over a landscape comprised of various crop types, genotypes, soil conditions, and management practices requires spatial information only feasible with remote sensing (Bishop, 1999, 33).

The Land Remote Sensing Satellite (Landsat) and the sophisticated Space-borne Thermal Emission Reflectance Radiometer (ASTER) equipment on Terra supply remotely felt exterior warmth at pixel resolutions <100 m, but usual application is hindered by the reduced frequency of recurring treatment ( 16 days) and the detail that with cloud cover, monthly observations are likely. This severely bounds the utility of these sensors in providing routine supervising of ET.

With MODIS, which now can provide coverage two times a day from Terra and Aqua satellite stages, the interband resolution difference is a component of 4 having 250 m for VI musicians vs. 1000 m for surface temperature. Hence, exterior temperatures at 250-m tenacity are more expected to supply the level of spatial minutia essential for assessing fluxes of one-by-one land cover types. In this case, the most significant difficulty is to differentiate between dissimilarities due to change in land cover categories and dissimilarities due to variations in solar illumination, atmospheric situation or phonological development. In alignment to differentiate important dissimilarities from insignificant ones (Leva, 2003, 85), the distinction likeness supplied has to be suitably investigated utilizing methods for example implication and hypothesis checks or predictive models. In remote sensing, the investigation of the distinction likeness is classically founded on histogram thresholding methods as asserted by empirical schemes or trial-and-error methods (Klees, 1995, 28).

Discussion and Analysis

Of particular interest is to glimpse if fluxes from individual maize and soybean areas are distinguishable at 250-m resolution. This would show that ET estimation at finer resolutions may not be warranted to distinguish distinct land cover/crops. If true, MODIS data could be used with thermal sharpening techniques to provide a greater possibility for routine ET monitoring (Leva, 2003, 66).

The aim of this paper disagrees from preceding investigations evaluating the effect of sensor tenacity and/or aggregation techniques on remote feeling -based flux form yield (e.g., Friedl, 1997; Kustas & Humes, 1996; Moran et al., 1997; Sellers et al., 1997 and Su et al., 1999). At spatial resolutions that will encompass some cover kinds, errors in flux estimation can be important (Kustas & Norman, 2000b and Moran et al., 1997). However, because there was effectively no distinction in the area mean fluxes for the two images ...