Quantification of Stemflow in Three Isolated Shrub Species in an Urban Environment

Abstract

As precipitation falls on to vegetation, it is partitioned into throughfall, vapor and stemflow. The stemflow component is frequently neglected in water budgeting for trees and shrubs due to its presumed small volume and limited research. Studies of stemflow in shrub species are especially rare. This study focuses on stemflow in shrubs and specifically examines its relationship to plant morphology and meteorological factors. Two individuals of three species, Prunus laurocerasus (common name ‘Otto Luyken'), Hydrangea quercifolia (common name ‘Alice’), and Itea virginica (common name ‘Little Henry’), were studied in an outdoor experiment in Philadelphia, PA during the 2015 growing season. Stemflow was collected using aluminum collars which had been attached to individual branches of each plant. Vinyl tubing conveyed the stemflow from the collars into collection bottles which were weighed after rain events. To relate stemflow to plant morphology, the vertical projected canopy area of each collared branch, branch attachment angle, stem circumference, and direction of collar were evaluated. To relate stemflow to meteorological conditions, the research also quantified rain depth, rain intensity, rain event duration, wind speed, wind direction, temperature, relative humidity, gust speed, and date and time of storm. The importance of each of these independent variables were studied using a forward stepwise linear regression. Stemflow averaged 7.6% of total incident rainfall on the canopy area of the monitored stems, with values ranging from 0% to 58% by branch and storm event. Species-specific averages were found to be 11.4%, 7.0%, and 4.4% for P. laurocerasus, H. quercifolia, and I. virginica, respectively. The results suggest a significant variation in stemflow by species. Statistical analysis determined that between 55% and 79% of the observed variation in stemflow can be explained by the regression model variables and gust speed and projected canopy area were determined to be significant across all regression models. Stemflow rates appear to be impacted by both inter-plant and local climate factors; however, additional research is needed to further refine the understanding of stem flow and canopy water budgeting.