Evapotranspiration of a willow cultivar (Salix miyabeana SX67) grown in a full-scale treatment wetland

Abstract

Abstract Since woody plants like willow are used increasingly in treatment wetlands, there is a growing need to characterize their ecophysiology in these specific growing conditions. For instance, evapotranspiration (ET) can be greatly increased in wetlands, due to factors like high water availability as well as oasis and clothesline effects. Few studies report willow ET rates measured in full-scale constructed wetland conditions, and fewer still in a temperate North-American climate. The objective of this study was to measure and model evapotranspiration of a commonly used willow cultivar, Salix miyabeana (SX67), to provide the ET rates and crop coefficient for this species. During two growing seasons, we studied a 48\u202fm2 horizontal subsurface flow willow wetland located in eastern Canada, irrigated with pretreated wood preservative leachate. Over two seasons, from May to October, we measured a mean monthly evapotranspiration rate of 22.7\u202fmm/day (16.5\u202fmm/d modelled), for a seasonal cumulative ET of 3954\u202fmm (2897\u202fmm modelled) and a mean crop coefficient of 6.4 (4.2 modelled). Both the evapotranspiration results and leaf area index (LAI) were greater than most results reported for open field willow plantations. Maximal stomatal conductance ( G ¯ s ) was higher than that expected for deciduous trees and even for wetland plants, and mean values correlated well with temperature, solar radiation, relative humidity and day of the year. We demonstrated that an ET model using G ¯ s , LAI and water vapor pressure deficit (VPD) as parameters could predict the evapotranspiration rate of our wetland. This simplification of traditional ET models illustrates the absence of evapotranspiration limitations in wetlands. Furthermore, this study also highlights some factors that can enhance ET in treatment wetlands. Our results should both improve the design of treatment wetlands using willows, and provide a simple ET predictive model based on major evapotranspiration drivers in wetlands.