Mapping tropical forest biophysical properties from coarse spatial resolution satellite sensor data: applications of neural networks and data fusion

Forest biophysical properties are typically estimated and mapped from remotely sensed data through the application of a vegetation index. This generally does not make full use of the information content of the remotely sensed data, using only the data acquired on a limited number of spectral wavebands, and may provide a relatively crude spatial representation. Using NOAA AVHRR imagery, it is shown that an artificial neural network may use all the remotely sensed data available to derive more accurate estimates of the biophysical properties of tropical forests in Ghana than a series of vegetation indices and that the representation derived can be refined by fusion with finer spatial resolution imagery.