Measurement and modeling of canopy interception loss of evergreen, deciduous and mixed forests in a subhumid watershed on the Loess Plateau, China

Abstract

Canopy interception, a significant yet inadequately comprehended hydrological phenomenon in terrestrial ecosystems, plays a crucial role in the water balance of forests. A profound understanding of the water retained and re-evaporated through interception storage is essential for developing a comprehensive understanding of forest hydrology. Integration this knowledge into hydrological models can help assess the effects of climate change on forests. In the Loess Plateau of China, extensive ecological restoration measures have been implemented to mitigate severe soil erosion and restore the fragile ecological environment. However, few studies have investigated the role of canopy interception in different tree species compositions (planted monoculture forests and mixed forests). This study monitored precipitation, throughfall, stemflow, and estimated canopy interception in three different forest stands during the 2021–2022 growing season in Shanxi Province, China. Canopy interception was quantified and simulated using the revised Gash model. The observed throughfall, stemflow and canopy interception for deciduous forest were 81.5 %, 1.6 % and 16.9 %, for evergreen forest were 84.2 %, 1.4 % and 14.4 %, respectively. The corresponding values for the mixed forest were 80.7 %, 2.0 % and 17.3 %. The revised model underestimated canopy interception to varying degrees in all three forest types, with the deciduous forest by 11.2 ± 1.8 %, evergreen forests by 21.1 ± 5.5 %, and mixed forest underestimating by, 16.7 ± 3.2 %. According to the statistical parameters (mean absolute error, mean bias error, root mean square error and Nash-Sutcliffe efficiency), the revised model can simulate the canopy interception dynamics of three forest types, with the best performance in simulating the deciduous forest, followed by mixed forest and evergreen forest. The study found that canopy interception loss was significantly influenced by a combination of canopy characteristics (including canopy storage capacity, canopy cover fraction, trunk storage capacity, and the percentage of precipitation diverted into stemflow) and climatic variables (such as average precipitation intensity and average evaporation rate). Among these factors, average precipitation intensity and canopy storage capacity were identified as the most influential variables across all three examined stands. Overall, the revised model was suitable for typical plantation forests and their mixed forests on Loess Plateau. Our study has important implications for understanding of the forest water balance in the Loess Plateau and also contributes to precipitation partitioning forecasts and efficient water resource management.