Proceedings Article | 26 August 2020
KEYWORDS: MODIS, Atmospheric modeling, Earth observing sensors, Satellites, Meteorology, Remote sensing
Evapotranspiration (ET) constitutes a meteorological and hydrogeological parameter which plays a critical role to climate, crop irrigation and water management (aquifer and surface water supplies). The scarcity of ground-based data along with the call for regional and global scale research, led to the emergence of remote sensing methods and products. Thus, algorithms and adaptations of the empirical methods have been developed to incorporate satellite data or a combination of satellite and meteorological data. The current study reviews popular models, namely, surface energy balance (SEB) algorithms (i.e. SEBAL, S-SEBAL, SEBAL-Tasumi, SEBS, T-SEB, S-SEBI, SSEBop, ASEBAL, METRIC), models based on vegetation and soil moisture indices (e.g. EVI, NDVI, LAI, GVMI) such as NDVI-CWS and modifications of established ET equations such as Penman-Monteith, Priestley-Taylor, Hargreaves-Samani, Granger, Carlson and Buffum, Droogers-Allen, REMCT incorporating satellite data. The aforementioned models come with their limitations in ET estimation depending on parameters such as scale, landuse/landcover (LU/LC) of the study area, quality of meteorological and satellite data, spatiotemporal resolution of satellite products, number of inputs and usability. Uncertainties occur when coarse MODIS-retrieved data inputs are incorporated. In Greece few research attempts employing satellite data for ET estimation are numbered, the majority of which of field scale or local scale over Thessaly. Future research could implement interpolation of measurements derived from a wide number of meteorological stations distributed over the study area at regional scale. Moreover, comparisons of ET estimations to pan evaporation measurements are recommended, aiming at obtaining more sophisticated, area-tailored models and quantifiable physical relations.
