Assessment of spectral indexes for estimating soil water content in the Brazilian Pantanal

The Pantanal is the largest floodplain of globe occupying 138,183 km2 in Brazil. fundamental ecological factor interaction between ecosystems flooding regime. Connected to properties soil-plant-atmosphere system, knowledge on soil water content becomes increasingly necessary. high temporal and spatial variability caused by vast heterogeneity texture, vegetation, topography climate makes it a difficult physical variable be measured. However, its can determined recent modern techniques remote sensing based data obtained microwave or infrared sensors. Thus, aim this study was evaluate accuracy vegetation indexes through satellite images from Landsat 5 northern Brazilian Pantanal. conducted pasture, experimental site Northern Mato Grosso state. Soil moisture measured using TDR probe installed at 10 cm depth period 2009 2011. For comparison, spectral surface temperature provided Geological Survey (USGS) were used, these are derived bands ratios reflectance products TM. evaluation performed some indicators: - Willmott index, Root Mean Square Error Absolute Error. This demonstrated that application management resources very promising. correlated with measurement. Among NBR-2 showed related both types soils EVI had highest determination coefficient, lowest errors Willmott’s index agreement.