Variations of reflectance and vegetation indices as a function of the topographic modeling parameter of the Parque Estadual do Turvo, Rio Grande do Sul, Brasil

Remote sensing techniques have been widely used in forestry studies because they allow evaluation and monitoring of large areas. The Parque Estadual do Turvo (PET) (17.491 ha) is the largest fragment of preserved subtropical deciduous forest of South Brazil, representing an extension of the Misiones forest in Argentina (10.000 km²). This area has great environmental importance and is adequate for performing remote sensing studies using high or even coarse-to-moderate spatial resolution data as well as related vegetation indices. Both, the reflectance and vegetation indices, are affected by external factors that change the spectral response of the surface components. Among the factors that can introduce errors in the interpretation of the images, topographic effects add spectral variability in satellite products. In addition, previous studies in subtropical forests showed that the geometry of data acquisition affects significantly the estimates of vegetation parameters derived from images acquired at off-nadir viewing or by large field-of-view (FOV) sensors. This study aimed to evaluate the magnitude of the bidirectional reflectance variations and of the derived vegetation indices as a function of local topography using high spatial resolution data acquired by the RapidEye constellation of satellites.