Effects of the Relative Spectral Response on Vegetation Indices (NDVI) Derived from Disparate Remote Sensing Systems

Principal Investigator:

David J. Fleming

Remotely sensed data can provide the land cover information important for estimating levels and rates of deforestation, habitat fragmentation, urbanization, wetland degradation, and many other phenomena. A variety of space sensors of varying characteristics observe the Earth and the last decade has seen a large increase in the number and types of sensors that make available data sets that range in information content not previously seen by the scientific community. This wide range of operating scales of satellite sensors is necessary to capture the complex nature and dynamics of the phenomena at the surface of the Earth. Whether acquired data from disparate remote sensing systems can be brought into alignment or transposed between sensors, and under what conditions is of great importance to geographers and ecologists. Although there are other factors that must be considered in inter-sensor comparisons, the relative spectral response (RSR) function of an electro-optical sensor determines what portion and level of the electromagnetic spectrum is measured in a given spectral band of a sensor and is the fundamental basis of spectral measurements. Understanding the effect of the RSR is crucial to the understanding of remotely sensed data measurements and derived metrics, such as vegetation indices that are used in earth system science studies. Data from the Ikonos, Landsat 7 ETM+, and other sensors are being used in this research.