However, if we can explain the processes behind the way land surfaces scatter light in specific wavelengths through a model of some kind, then we can correct the sensor data properly for these effects. In addition, we might hope to find additional information on the characteristics of the surface via the model's surface parameters, so this presents opportunities as well as problems. Such a model is known as a BRDF model because it tries to describe or explain the Bidirectional Reflectance Distribution Function.
These pages show some of the results of fitting such a model to AVHRR
observations over the semiarid grasslands of Xilingol Aimag,
Inner Mongolia
Autonomous Region, P.R. China; see also
this site and here.
The overall goal is to provide higher quality data from the AVHRR for community
type differentiation in the Inner Mongolia grasslands but there are
implications for all land remote sensing applications. The immediate aim of this
research is to assess the extent of the improvement in AVHRR data quality possible
operationally with BRDF correction and to evaluate the information content of the
BRDF model parameters.