Stable Water Isotopes Reveal Effects of Intermediate Disturbance and Canopy Structure on Forest Water Cycling

Abstract

Forests play an integral role in the terrestrial water cycle and link exchanges of water between the land surface and the atmosphere. To examine the effects of an intermediate disturbance on forest water cycling, we compared vertical profiles of stable water vapor isotopes in two closely located forest sites in northern lower Michigan. At one site, all canopy‐dominant early successional species were stem girdled to induce mortality and accelerate senescence. At both sites, we measured the isotopic composition of atmospheric water vapor at six heights during three seasons (spring, summer, and fall) and paired vertical isotope profiles with local meteorology and sap flux. Disturbance had a substantial impact on local water cycling. The undisturbed canopy was moister, retained more transpired vapor, and at times was poorly mixed with the free atmosphere above the canopy. Differences between the disturbed and undisturbed sites were most pronounced in the summer when transpiration was high. Differences in forest structure at the two sites also led to more isotopically stratified vapor within the undisturbed canopy. Our findings suggest that intermediate disturbance may increase mixing between the surface layer and above‐canopy atmosphere and alter ecosystem‐atmosphere gas exchange.

An isotopic approach to partition evapotranspiration in a mixed deciduous forest

“…Because LAI sets F T , Wang et al (2014) proposed that F T should be relatively consistent throughout the growing season. Although F T can vary with passing weather systems and precipitation (e.g., Aron et al, 2019; Wen, Yang, Sun, & Lee, 2016), periods of water stress (Good et al, 2014; Matheny et al, 2017) and the removal of biomass (e.g., harvesting or cutting grass) (Wang, Yamanaka, Li, & Wei, 2015), Berkelhammer et al (2016) demonstrated that forest F T was generally invariant on seasonal timescales. We come to the same conclusions on subdiurnal timescales (Figure 7), although this observation may be dependent on vegetation type, aridity and soil moisture.…”

“…Neither approach requires laborious leaf chamber measurements, and both are founded on a steady‐state assumption about δ T that is valid for midday (Figure 7) and seasonal (e.g., Wei et al, 2015) isotopic ET partitioning. In contrast, assumptions of steady‐state δ T may not suffice for questions related to isotope and water cycles on subdiurnal timescales (e.g., Aron et al, 2019; Simonin et al, 2013; Welp et al, 2012). On this relatively short timescale, nonsteady‐state δ T measurements inform how transpiration forces the isotopic composition of atmospheric water vapour and may help validate the Craig and Gordon (1965) model that is commonly used to estimate δ T and δ E (e.g., Dubbert et al, 2013; Dubbert et al, 2014; Good et al, 2012; Hu et al, 2014).…”

“…Air within the canopy is usually poorly mixed (Aron et al, 2019), so we used above-canopy measurements for the Keeling regression to derive ecosystem-scale δ ET . In contrast, δ T measurements are separated by species (e.g., δ T,maple , δ T,aspen and δ T,oak ).…”

“…This approach does not capture the spatial heterogeneity of δ s (Gazis & Feng, 2004;Hsieh, Chadwick, Kelly, & Savin, 1998) but is a common approach in most ET partitioning studies (e.g., Aouade et al, 2016;Yepez et al, 2005;Zhang, Shen, Sun, & Gates, 2011). The closed, thick canopy cover at our field site (Aron et al, 2019) likely reduces spatial variation in δ s . The kinetic fractionation factor in soil evaporation studies has long been a point of debate and varies with soil tortuosity, soil moisture and atmospheric conditions (Quade, Brüggemann, Graf, Vereecken, & Rothfuss, 2018;Xiao et al, 2018).…”

“…Further complicating our understanding of land-atmosphere water exchange, some ground-based observations of ET may not actually capture conditions at the transpiring or evaporating surfaces. For example, near-surface gradients of water vapour concentrations and vapour pressure deficits (VPDs) can make it difficult to relate ET measurements, most of which are made using eddy covariance above canopies, to leaf and soil fluxes within canopies (Aron, Poulsen, Fiorella, & Matheny, 2019;De Kauwe, Medlyn, Knauer, & Williams, 2017;Jarvis & McNaughton, 1986). Therefore, additional leaf-and soil-level flux measurements are needed to improve estimates of F T and predictions of terrestrial water fluxes.…”