Temporal variations in satellite reflectances at field and regional scales compared with values simulated by linking crop growth and SAIL models

The objective of this article is to present a methodology for modeling shortwave satellite measurements over crops. This step is a prerequisite before we can apply inversion techniques of satellite data in order to retrieve crop yields. Temporal profiles of satellite data were simulated for an agricultural area, at field (20 m, SPOT HRV) and regional (4 km, NOAA AVHRR) scales, using a vegetation process model (AFRCWHEAT2) and a radiative transfer model (SAIL). Reflectances for individual winter wheat fields observed in SPOT HRV scenes on five dates during the cereal growing season were simulated accurately. Low resolution NOAA AVHRR data used for the regional study included various vegetation assemblages. The fractional area for each type of vegetation and its reflectances were taken into account. Winter cereals reflectances were simulated by the methods validated at the field scale and empirically modeled reflectances were used for other species. Although an AVHRR pixel consists of the intermingled signals characteristic of several vegetation categories, the simulated and observed NOAA AVHRR scale reflectances agreed well.