Effects of Hot Versus Dry Vapor Pressure Deficit on Ecosystem Carbon and Water Fluxes

Abstract

Vapor pressure deficit (VPD) has increased and will likely continue increasing, with wide-ranging effects on ecosystems. Future VPD increases will largely be driven by warming, yet most experiments examining VPD effects on plants have done so by changing humidity. Here, we used meteorological data and carbon and water fluxes measured at 26 climatically-diverse eddy covariance sites to quantify the extent to which VPD has been driven by variation in air temperature versus humidity. We fit generalized additive models (GAMs) at each site to quantify effects of hotter (and wetter) and cooler (and drier) versus typical VPD on ecosystem-scale fluxes of carbon and water. We found that VPD has occurred under diverse combinations of temperature and humidity: >50% of a site's daytime growing season temperature range and >35% of its relative humidity range have often combined to define a particular VPD. We found moderate evidence that hotter versus drier VPD of the same magnitude differentially affect gross primary productivity (GPP), net ecosystem productivity (NEP), and latent heat flux (LE): Selected GPP and NEP GAMs at about half of sites and LE GAMs at about a third of sites included a VPD-temperature interaction. The magnitude of the interaction varied, but was generally 29%–57% of the effect attributable solely to VPD. The direction of the interaction also varied, but hot VPD was commonly associated with higher carbon fluxes. These effects were not strongly modified by soil moisture. Overall, results emphasize the relevance of VPD-temperature interactions at a critical time of rapid VPD increase.