Remote sensing of tropical forests : the importance of shadow patters and multi-angle viewing in extracting canopy gap and forest structure information

This thesis examines principles surrounding the use of multi-angle remote sensing to exploit effects of canopy shadowing on tropical forest bi-directional reflectance and thus provide improved tropical forest structural variable estimates.;Remote sensing provides an invaluable source of information on inaccessible tropical forests. To date, the extraction of reliable and replicable forest canopy information from remotely sensed data has proved difficult. Recently, sensors have been developed which are able to view the Earth's surface from multiple angles. This extra dimension in reflectance data is expected to improve the estimation of canopy variables by exploiting the functional relationships between forest structure and anisotropic reflectance.;Shadow is an important scene component known to influence reflectance in the optical spectrum.