Partitioning of rainfall in a seasonal dry tropical forest

Abstract Rainfall redistribution by forest cover has potential hydrological impacts in semi-arid regions due to continuous human intervention. Studying the process of interception loss by Caatinga vegetation and its changes due to deforestation is extremely important for local hydrology. However, such information is scarce in the literature. This study examined the partitioning of rainfall into throughfall (TF), stemflow (SF) and interception loss (I) in the Caatinga vegetation (CAA) and evaluated the influence of rainfall characteristics on this partitioning. The components of rainfall partitioning were measured from 2016 to 2017 to determine the TF and SF respectively, and their relationships with rainfall characteristics were evaluated based on linear regression models. For the Caatinga vegetation, TF and SF represent 89.2 and 0.5% of the gross rainfall, while interception loss was 10.3%, an expressive value that should be included in regional water balance models. The rainfall characteristics were able to explain the variations in water partitioning, showing that TF and SF increase for events of higher intensity and volume, whereas interception loss is mainly associated with events of longer duration. For the Caatinga, stemflow and throughfall usually occur for events of greater than 1.65 and 0.98 mm, while values for canopy water storage ranged from 0.88 to 1.16 mm. It is concluded that for semi-arid environments, these values are significant and cannot be ignored when managing local water resources.