Surface albedo as a proxy for land-cover clearing in seasonally dry forests: evidence from the Brazilian Caatinga

摘要: Ongoing increase in human and climate pressures, addition to the lack of monitoring initiatives, makes Caatinga one most vulnerable forests world. The is located semi-arid region Brazil its vegetation phenology highly dependent on precipitation, which has a high spatial temporal variability. Under these circumstances, satellite image-based methods are valued due their ability uncover human-induced changes from effects land cover. In this study, time series stack 670 Landsat images over period 31 years (1985–2015) was used investigate patterns land-cover clearing (LCC) removal an area Caatinga. We compared LCC detection accuracy three spectral indices, i.e., surface albedo (SA), Enhanced Vegetation Index (EVI) Normalized Difference (NDVI). applied residual trend analysis (TSS-RESTREND) attenuate seasonal signal detect only significant structural (breakpoints) monthly series. Our results show that SA able identify general occurrence year it occurred with higher (89 62%, respectively) EVI (44 22%) NDVI (46 22%). overall outcome study shows benefits using index incorporates short-wave infrared range, such as SA, visible near-infrared indices for seasonally dry