Avaliação de metodologias de caracterização atmosférica na conversão de dados radiométricos digitais em valores físicos

AUTOR(ES)
DATA DE PUBLICAÇÃO

2008

RESUMO

- The conversion of digital numbers (DN) of orbital images to physical quantities of the Earth surface is dependent on a good atmospheric characterization, which has been performed through the use of data collected by meteorological balloons or by sun photometers positioned at the surface. Recently some orbital sensors have generated useful data that can be utilized in this characterization like those generated by the sensor MODIS. Here, surface Bidirectional Reflectance Factors (BRF) estimated from field data were compared to those calculated from TM/Landsat-5 images converted through radiative transfer codes. The input data to run these codes were estimated considering two different sources: a sun photometer and the MODIS sensor. In the visible region the BRF estimated by MODIS data were always quite similar to those based on the sun photometer. The RMSE values determined using MTR 5S based on the sun photometer data were 0.0316 (TM1), 0.0237 (TM2) and 0.0218 (TM3) and for those based on MODIS data 0.0246 (TM1), 0.0217 (TM2) e (0.0206) TM3. In the NIR and SWIR spectral regions, the BRF values estimated through MODIS data were also quite similar to those calculated through solar photometer data. Applying MTR 5S to the sun photometer the RMSE values were 0.0410 (TM4), 0.0614 (TM5) and 0.0433 (TM7); and 0.0442 (TM4), 0.0654(TM5) and 0.0451 (TM7) to the characterizations performed through MODIS data. The results have indicated that the MODIS data can be used to convert the DN into surface BRF.

ASSUNTO(S)

atmospheric correction landsat radiometria sun photometer modis radiometry sensoriamento remoto fotômetro solar landsat correção atmosférica modis

ACESSO AO ARTIGO

Documentos Relacionados