This work aims at modelling the morphological variable steam-volume of the macrophyte species as a function of the attributes derived from the ALOS/PALSAR polarimetric data using multiple regression technique. The study was carried out at Monte Alegre Lake, in the Amazon floodplain area, Pará State, Brazil. The modeled variable steam-volume was evaluated and the contribution of the phase information from the radar data was highlighted. The fieldwork was performed almost simultaneously to the radar acquisition. Macrophyte morphological variables were measured in the field and used to derive the stem-volume model regarding the attributes generated from the radar data. Although the adjusted coefficient of determination was not high ( 44%), the presented predictive ability and all validation elements were expected with 95% of confidence. Among the five independent variables of the model, four were generated from the phase information, which brings about this reliable information.
Polarimetric Radar; PALSAR data; Multiple Linear Regression; Macrophyte Monitoring; Amazon Floodplain
