Land–atmosphere carbon and water flux relationships to vapor pressure deficit, soil moisture, and stream flowby Stephen R. Mitchell, Ryan E. Emanuel, Brian L. McGlynn

Agricultural and Forest Meteorology


Global and Planetary Change / Agronomy and Crop Science / Atmospheric Science / Forestry


Agricultural and Forest Meteorology 208 (2015) 108–117

Contents lists available at ScienceDirect

Agricultural and Forest Meteorology journa l homepage: www.e lsev ier .com/ locate /agr formet

Land–atmosphere carbon and water flux relationships to vapor pressure deficit, soil moisture, and stream flow

Stephen R. Mitchell a,∗, Ryan E. Emanuelb, Brian L. McGlynnc a Nicholas School of the Environment, Duke University, Durham, NC 27708, USA b Department of Forestry and Environmental Resources, North Carolina State University, Raleigh, NC 27695, USA c Division of Earth and Ocean Sciences, Nicholas School of the Environment, Duke University, Durham, NC 27708, USA a r t i c l e i n f o

Article history:

Received 21 July 2014

Received in revised form 31 March 2015

Accepted 8 April 2015


Carbon dioxide flux

Net ecosystem production

Net ecosystem exchange


Climate change

Subalpine forests

Stream flow

Water-use efficiency a b s t r a c t

Climatic change is exerting considerable influence on the hydrologic and biogeochemical cycles of snowdominated montane forest ecosystems. Growing season drought stress is a common occurrence after snowmelt-derived soil water content (WC) and stream flow (Q) have declined, leading to an increase in atmospheric water demand (i.e., vapor pressure deficit, VPD). Here, we analyzed a 6-year record (2006–2011) of H2O and CO2 fluxes from the Tenderfoot Creek Experimental Forest, a montane forest in the northern Rocky Mountains to examine (1) how growing season evapotranspiration (ET), net ecosystem production (NEP), and water-use efficiency (WUE, NEP/ET) respond to changing WC and VPD, (2) how stream flow (Q), an integrated measure of catchment-level water availability, relates to NEP, and (3) how annual NEP is related to annual precipitation and the temperature-defined growing season length (GSL). Growing season NEP exhibited a linear relationship with WC and a log-linear relationship with Q, indicative of persistent water limitations when streamflow and soil moisture reach their annual minima late in the growing season. Nevertheless, years with long GSLs had relatively higher NEP, with a small net carbon sink maintained even at low levels of WC and Q, suggesting that trees are able to obtain water from deeper portions of the soil profile (>30 cm) during droughts. However, the warmer, drier climate projected for this region could bring this system closer to a critical threshold of GSL, WC, and VPD, introducing vegetation water stress that could alter the current relationship between GSL and annual NEP. © 2015 Elsevier B.V. All rights reserved. 1. Introduction

Subalpine forests are among the ecosystems with the greatest sensitivity to continued climatic change (Körner, 2003), and the high-elevation evergreen forests of the Intermountain West are already exhibiting effects of a warming climate, largely as a consequence of reduced water availability (Allen et al., 2010; Breshears et al., 2005; van Mantgem et al., 2009; Westerling et al., 2006).

Water inputs into these systems are largely supplied by the melting of winter snowfall, providing vegetation with a slow, transitory pulse of water that accounts for the majority of ecosystem water use (Hunter et al., 2006; Monson et al., 2002). Widespread warming has led to a decrease in winter snowpack in the Intermountain

West (Hamlet et al., 2005), and continued increases in surface temperatures in future years could result in earlier snow melt, thereby

Abbreviations: NEP, net ecosystem production; WUE, water use efficiency; VPD, vapor pressure deficit; GSL, growing season length. ∗ Corresponding author. Tel.: +1 919 491 0398.

E-mail address: (S.R. Mitchell). altering the timing and magnitude of water available for terrestrial ecosystems (Barnett et al., 2005; Mote, 2006).

Evapotranspiration (ET) is typically the largest flux of water out of these ecosystems. ET is constrainedby solar radiation, air and soil temperature, atmospheric vapor pressure deficit, and soil water content (Monteith, 1973). Thus the water balance of subalpine ecosystems is regulated by seasonal changes in these variables.

Like other ecosystem processes, ET can be limited by a single variable or co-limited by a combination of variables, resulting in a ‘switching’ sensu Baldocchi et al. (2006) among the variables that regulate ecosystemprocesses. For subalpine forest ecosystems that are dependent upon snowmelt, growing season ET is thought to respond to VPD until later in the growing season, when soil water content becomes more limiting to ET potentially inducing vegetation water stress (Emanuel et al., 2010).

Understanding the ecohydrology of water-limited subalpine forests during extended periods of low soil water content (WC) and high atmospheric water demand (VPD) could be crucial to projecting future patterns of CO2 uptake and storage under future climatic conditions (Huet al., 2010a,b;Monsonet al., 2010, 2002; Sackset al., 2007; Schimel et al., 2002). The components of net ecosystem pro 0168-1923/© 2015 Elsevier B.V. All rights reserved.

S.R. Mitchell et al. / Agricultural and Forest Meteorology 208 (2015) 108–117 109 duction of CO2 (NEP), namely photosynthesis (GPP) and ecosystem respiration (RE), are differentially influenced by soil water, humidity, temperature, and resulting VPD (Chapin and Matson, 2011), meaning that NEP may respond differently than ET to atmospheric water demands, especially given that NEP integrates vegetation and soil microbial fluxes of carbon acting in opposition to one another. The ratio of NEP to ET, also known as water use efficiency (WUENEP/ET), can be a useful metric for examining ecosystem stability, particularly during periods ofwater stress (Emmerich, 2007).

Mostof thedifferences inWUENEP/ET betweensites aredue todifferences in ET rather thanNEP (Ponton et al., 2006). Thus, higher levels of atmospheric water demand can increase water losses from the system via ET, leading to a decrease in WUENEP/ET as demonstrated by Monson et al. (2010) who found that growing season WUENEP/ET at the Niwot Ridge LTER decreased under conditions of low WC, a time during which daytime VPD is often highest.