Evapotranspiration in a forested mountain catchment of the humid temperate environment

Abstract

Evapotranspiration is an important component of the water budged. A direct estimate of evapotranspiration seems to be complicated namely because of significant temporal and spatial variations, particularly, in the mountain environment. The main aim of this study was to analyse and estimate evapotranspiration components within the temporal and spatial variations in the region of the Jizera Mountains. A distributed approach based on the evaluation of energy budget components of the active surface was applied here in the SVAT (soil-vegetation-atmosphere-transfer). This method was performed on 25 x 25 meters cell size level raster in daily time step. The approach uses recursive algorithm for active surface temperature as a main driver of upcoming long wave radiation and its role in energy balance. Such an algorithm cannot be solved in commercial geographical information systems, so all spatial components were transformed into numerical arrays and processed in the batch mode in scripting language Python and Numpy as its module for numerical calculations.Meteorological and hydrological time series of daily data (precipitation, air temperature, relative humidity and wind speed) collected in the period of 1999 to 2007, were processed. The records of point observations were interpolated and extrapolated into the entire area of interest. The additional measurement of vegetation characteristics and detailed fluxes of the energy budget in a vertical profile has been used to complete the detailed analysis of evapotranspiration. A simple modification of the Rutter model was used to clarify components of the evaporation loss: transpiration, interception and evaporation from the open water surface. The estimates of the Penman-Monteith method (mean potential evapotranspiration during the vegetation season, were compared with the alternative less complex methods of Makkink , Hamon a FAO method. Also a comparison with actual evapotranspiration taken from long term water balance of 9 experimental watersheds was done with average difference of 4.1% and standard deviation of 77.9 mm. Method of Penman-Monteith was found to be the most appropriate in this region; however also its accuracy depends on the level of spatial and temporal discretisation.

Evapotranspiration is an important component of the water budged. A direct estimate of evapotranspiration seems to be complicated namely because of significant temporal and spatial variations, particularly, in the mountain environment. The main aim of this study was to analyse and estimate evapotranspiration components within the temporal and spatial variations in the region of the Jizera Mountains. A distributed approach based on the evaluation of energy budget components of the active surface was applied here in the SVAT (soil-vegetation-atmosphere-transfer). This method was performed on 25 x 25 meters cell size level raster in daily time step. The approach uses recursive algorithm for active surface temperature as a main driver of upcoming long wave radiation and its role in energy balance. Such an algorithm cannot be solved in commercial geographical information systems, so all spatial components were transformed into numerical arrays and processed in the batch mode in scripting language Python and Numpy as its module for numerical calculations.Meteorological and hydrological time series of daily data (precipitation, air temperature, relative humidity and wind speed) collected in the period of 1999 to 2007, were processed. The records of point observations were interpolated and extrapolated into the entire area of interest. The additional measurement of vegetation characteristics and detailed fluxes of the energy budget in a vertical profile has been used to complete the detailed analysis of evapotranspiration. A simple modification of the Rutter model was used to clarify components of the evaporation loss: transpiration, interception and evaporation from the open water surface. The estimates of the Penman-Monteith method (mean potential evapotranspiration during the vegetation season, were compared with the alternative less complex methods of Makkink , Hamon a FAO method. Also a comparison with actual evapotranspiration taken from long term water balance of 9 experimental watersheds was done with average difference of 4.1% and standard deviation of 77.9 mm. Method of Penman-Monteith was found to be the most appropriate in this region; however also its accuracy depends on the level of spatial and temporal discretisation.