How to differentiate nighttime transpiration and water recharge in nocturnal sap flow?

<p>Nocturnal sap flow (<em>Q<sub>n</sub></em>) affect not only forest carbon and water budgets but also their responses to climate change as it consists of two ecohydrological and ecophysiological significant components: nighttime transpiration and water recharge. A vapor pressure deficit (<em>VPD</em>) based sap flow partitioning method has been developed to estimate nighttime transpiration, which is normally quantified through the discretely measured nighttime stomatal conductance, from the widely and continuously measured sap flow. However, given the increasing knowledge of <em>Q<sub>n</sub></em> mechanisms, whether <em>Q<sub>n</sub></em> could be partitioning simply by <em>VPD</em> and whether this method is valid in semi-arid regions remain unclear. We measured sap flow of <em>Pinus tabuliformis</em> and <em>Acer truncatum</em> in a middle-aged and a young monoculture forest stand, respectively, in a semi-arid mountainous area of northern China. We found the influence of <em>VPD</em> on <em>Q<sub>n</sub></em> conditioned by soil moisture. Meanwhile, a considerable impact of wind speed on <em>Q<sub>n</sub></em> was observed. In the stands with relatively dry soils, both increased and decreased soil moisture promoted <em>Q<sub>n</sub></em>, which might be due to enhanced nighttime water recharge for two distinct purposes, i.e., capacitance refilling and avoiding hydraulic failures. For these three environmental factors (i.e., <em>VPD</em>, wind speed, and soil moisture) that have been considered most in previous studies, their total effect explained less than 55% of the <em>Q<sub>n</sub></em> variations. This study highlights that physiological influences of <em>VPD</em> on nighttime stomatal water loss were uncertain. Furthermore, it suggests that there could exist considerable nighttime water loss induced by wind, possible region-specific patterns of nighttime water recharge, and limited concurrent environmental controls on <em>Q</em><sub>n</sub>. Our findings are helpful to improve the <em>VPD</em>-based sap flow partitioning method to differentiate nighttime transpiration and water recharge.</p>