Assessing the cross-site and within-site response of potential production to atmospheric demand for water in Eucalyptus plantations

Abstract

Plant water deficits arise from low soil water and high atmospheric demand for water (expressed as vapor pressure deficit; VPD). Soil water and VPD often covary making it difficult to examine the effect of only VPD on biomass production. We used four Eucalyptus plantation sites where one treatment maintained high soil water with irrigation to evaluate the response of forest production to VPD independent of soil water. We used two approaches: an empirical test and simulations with the 3-PG model. For the empirical test, we examined the VPD response of gross primary production (GPP), net primary production of aboveground wood biomass (ANPPW), photosynthesis per unit of light absorbed (GPP per unit of intercepted photosynthetically active radiation (APAR)), and wood growth per unit of light absorbed (ANPPW APAR-1). For modeling, we compared 3-PG model predictions of these variables using a constant VPD and VPD that varied with data from the sites. Photosynthesis per light absorbed and wood growth per light absorbed both decreased exponentially as VPD increased, but neither GPP nor ANPPW varied with VPD. Across sites, photosynthesis per light absorbed increased with VPD, but the other variables had no relationship with it; wood growth per light absorbed, flux to ANPPW, and partitioning of GPP to aboveground and ANPPW decreased with site mean annual temperature. Results from the 3-PG model simulations were similar to those in the data. Two factors explain the response of photosynthesis per light absorbed and wood growth per light absorbed to VPD and the lack of response of GPP and ANPPW to VPD. First, VPD is strongly correlated with APAR—clear days yield high APAR and high VPD. Second, the extra light absorbed when APAR is high cannot be used for GPP because leaf stomata are closed when VPD is high. We expect that similar results would apply across the wet tropics, and future studies linking aboveground production to water in the wet tropics should focus on soil water status, not VPD.