Evaluation of the remotely sensed actual evapotranspiration in the mosaic agricultural and forest landscape of the Czech Republic

Evapotranspiration (ET) represents the dominant component of the terrestrial water balance. Evapotranspiration is tightly related to the land surface cover, the surface radiation and energy balance, soil water availability, and to the evaporative demand of the atmosphere – which is in fact also related to the radiation and energy balance, yet at much larger scales. Despite the high significance of ET in water balance, its quantification at the field to landscape level still remains largely uncertain. This limits our ability to realistically predict e.g. precipitation and cloud formation patterns within local and regional climate and weather models as well as our understanding of how natural and managed ecosystems are responding to changing hydrological and climatic regimes. Remote sensing estimates of ET through land surface temperature (LST) offers complete land spatial coverage and does not require any explicit calibration, though validation is still needed. Although, there have been methods developed based on empirical relations with vegetation indices, the models based on LST provide a physically based technique with unique capability to capture actual response of vegetation such as partial stomatal closure and not only its long-term status mirroring in its development. In general, the high spatial resolution (order of m) of remote sensing methods is compromised by low time resolution (order of weeks), while the high time resolution (order of hours) is traded off by low spatial resolution (order of km). However, high spatiotemporal resolution can be attained by combining satellite retrievals from different platforms with different spatiotemporal characteristics. The presented contribution will focus on evaluation of Atmosphere Land EXchange Inverse (ALEXI) model for estimates of ET across the mosaic agricultural and forest landscape of the Czech Republic. ALEXI is a diagnostic model based on time-differential LST measurements and other primarily remote sensing inputs related to meteorological conditions and land cover properties. ALEXI quantifies the evaporative flux and cooling required to keep the soil and vegetation at the observed LST under the known solar radiation inputs and other meteorological conditions. The morning LST rise is directly monitored by geostationary satellite METEOSAT with relatively coarse spatial resolution of ~5 km. Although this resolution is appropriate to solving land surface-atmosphere feedback at scales commensurate with surface forcing of air temperature at the blending height, for landscape analysis as well as validation purposes, finer spatial resolution is needed. To satisfy this need, a disaggregation and data fusion with retrievals from finer resolution polar-orbiting satellites will be applied. The disaggregated ET will be validated by an array of eddy covariance stations collocated across the Czech Republic covering forest (4), cropland (3) and grassland (1) sites.  Opportunities for further application of remotely sensed ET in land surface modelling and drought monitoring will be featured and discussed.