Quantifying throughfall, stemflow and interception loss in five vegetation communities in a maritime raised bog

Abstract

Throughfall and stemflow observations were used to quantify interception loss from five vegetation communities with complex multi-layered vegetative structure over the course of a year in a maritime raised bog in southwestern British Columbia, Canada. Separate plot-averaged interception loss estimates for trees and shrubs were computed from randomly relocated throughfall collectors and stemflow observations of trees. Independent throughfall observations from individual trees stratified by distance from the trunk were scaled to the plot extent based on tree configuration and canopy structure derived from LiDAR. Estimates from tree-based sampling compared favorably with plot-averaged tree throughfall means; however, results were biased towards underestimating throughfall compared to estimates from random observations possibly due to evaporation from the tree-based throughfall samplers, an issue that could be remedied using a different design. Nevertheless, this approach shows potential as a tool for estimating throughfall in forests with heterogeneous vegetation characteristics, and should be further tested and developed. Annual tree interception loss from birch was 15% of gross precipitation and average throughfall ranged from 75% to 83% during the leafed and leafless seasons, respectively. Tree interception loss in three pine stands of varying height and density was 13, 15 and 9% of gross precipitation. Stemflow from trees contributed between 0.4 and 3% of net precipitation, and smaller trees generated stemflow more efficiently. Interception loss from shrub ranged from 20% to 36% of gross precipitation. Tree throughfall increased with distance from trunks whereas shrub throughfall showed an opposite trend, and the relationship varied between stands. This study demonstrates the importance of shrub vegetation when quantifying net precipitation in vegetation communities typical of peatlands.