| dc.description.abstract | Evapotranspiration is a key hydrological process in Brazil’s 1,000,000-km2 tropical semiarid Caatinga biome, where monitoring is crucial, but data are scarce. We propose a method to estimate daily actual evapotranspiration (LE) based on the canopy-radiation balance, measuring air and canopy temperatures, relative humidity, wind speed, and global radiation. The method was applied at a location with preserved Caatinga forest [Aiuaba Experimental Basin, daily average net radiation (Rn) of 12 MJ/m2] during one hydrological year (2020–2021). The results, which agree with independently performed field measurements, identify the predominance of distinct evaporative processes throughout the year. In the dry season, net radiation is high (16.7 MJ/m2/day), but actual evapotranspiration is negligible (LE/Rn<0.01) because stress due to extremely low soil water content causes leave deciduousness and root shrinkage. Contrastingly, in the rainy season, soil water content increases, leaves recover, and roots expand, enhancing actual evapotranspiration (LE=6 mm/day and LE/Rn=0.89). Thus, in the Caatinga forest, during the dry and transition periods, actual evapotranspiration is ultimately ruled by the soil water content (source-limited), whereas during the rainy season, it is sink-limited (i.e., controlled by atmospheric demand). The authors developed a new method to assess actual evapotranspiration in forested areas. The actual evapotranspiration, which is fundamental to compute irrigation demand, is also essential to understand the impact of climate change on forests. This is particularly important for drylands, such as the 1,000,000-km2 Brazilian semiarid region, where water is scarce and actual evapotranspiration is high, consuming 70% of the precipitation. The method, which estimates the energy balance on the leaves of forest trees, demands the measurement of only a few variables, such as temperature, air relative humidity, wind speed, and radiation. The equipment to perform the measurement is much cheaper (less than 10%) than that of the reference methods for this kind of study. Therefore, with the same budget, the method can be applied at more locations, increasing the spatial representation of the measurements. The method hereafter presented is especially meant to help semiarid and arid regions, where evapotranspiration is high and monitoring is crucial, but data are often scarce. | |
