Monitoring LULC dynamics in the Sao Paulo region through landsat and C-band SAR time series

The paper debates a novel approach for sugarcane identification and characterization based on multi-spectral and multi-temporal profile matching. A parametric model aimed at identifying sugarcane among pasture/grasses/shrubs, annual crops and forest is proposed. Differently from other supervised and unsupervised classification techniques, the discussed profile-based parametric model accounts for variability in growth date, that becomes valuable information to be extracted, rather than simply a nuisance parameter, and delivers an effective extrapolation of the cane vigor. The approach is then applied to Landsat 5 TM and ERS/ENVISAT SAR time-series over the Orindiuva area attaining preliminary promising although perfectible results.