Variabilidade espaçotemporal da evapotranspiração em vegetação do bioma caatinga com uso do modelo hidrológico dicasm

• Ano de defesa: 2019
• Autor(a) principal: Soares, Nazaré Suziane
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Palavras-chave em Português: Modelagem hidrológica; Semiárido; Bacia experimental; Transpiração; Balanço hídrico; Transpiration; Semiarid; Hydrological modeling; Experimental basin; Water balance
• Link de acesso: http://www.repositorio.ufc.br/handle/riufc/46620

Evapotranspiration is inherently difficult to measure and predict, especially on large spatial scales as it is an atmospheric flow process. However, using hydrological modeling, reliable estimates of evapotranspiration and its components can be obtained. Therefore, the objective of this work was to study the spatial and temporal variability of plant evapotranspiration in Caatinga preserved through hydrological modeling. The DiCaSM hydrological model (Distributed Catchment Scale Model) was used for the evapotranspiration spatiotemporal analysis for the period from 2003 to 2017 in the study area, which is the Aiuaba Experimental Basin (BEA). The data obtained by the hydrological model were compared with data measured on an experimental scale and related to standard potential evapotranspiration according to FAO (ETPFAO). In addition, sensitivity analysis was performed for the parameters of leaf area index, maximum and minimum vegetation height, canopy resistance, maximum actual evaporation cut off point and for wind speed, net radiation, temperature and precipitation data. Sensitivity analysis emphasized that the maximum and minimum vegetation height and wind speed parameters caused the greatest changes in the potential evapotranspiration, actual evapotranspiration and transpiration processes. By evaluating spatial variability, it was observed that actual evapotranspiration and transpiration are represented by the model with a significant difference among the modeling units of the BEA while potential evapotranspiration showed no significant spatial variability. Evapotranspiration permanence curves allowed the analysis of the potential evapotranspiration in relation to the actual evapotranspiration modeled by DiCaSM. In addition, the relationship between modeled actual and potential evapotranspiration does not reach zero at any time of the year, while the relationship between modeled actual evapotranspiration and FAO standard potential evapotranspiration is null between August and November. BEA modeled and measured transpiration showed statistically similar permanence curves, evidencing DiCaSM's ability to model the periodicity of atmospheric water flow, especially in the months of the rainy season. The model could characterize evapotranspirative processes in a better way if it evaluated the influences of factors such as canopy resistance and wind speed that affect the water exchange of this vegetation.