Karol Michał Przeździecki , Jarosław Zawadzki
Estimation of soil moisture condition on a local or regional scale is one of the most important problems in vegetation conditions monitoring. One of the most suitable approach to do this over vegetated areas is the Land Surface Temperature – Vegetation Index (LST-VI) Triangle Method. This method estimates water content in soil which is available for plants in contrast to methods using satellite observations in the microwave band, and it is superior to pixel by pixel soil moisture methods using observations in the optical band which neglect statistical relationship between VI and LST controlling evapotranspiration on vegetated area. This study tested a methodological modification of the method based on assumption that evapotranspiration depends on the difference between the air temperature and the terrain surface temperature. The spatial distribution of Temperature-Vegetation Dryness Indices, which reflect soil moisture in the root zone was calculated for vast areas of Central Europe using the classic Triangle Method and its modified version. Both NDVI and EVI indices were used as input data in these calculations. The air temperature at 2 m height was taken from SYNOP reports, and the terrain surface temperature from MODIS data. In addition, Matlab scripts were written by the authors to allow convenient access to free SYNOP data. Our results indicate that proposed modification increases the accuracy of soil moisture estimation. This was confirmed by comparing classic and modified Temperature-Vegetation Dryness Indices values with rainfall data.