Predictions for future climate suggest an increase in drought frequency and severity especially in the midlatitudes and low-latitudes [1]. This has brought up a concern about future agricultural and forest productivity. Reductions in productivity could be significant enough to have a major impact on human wellbeing [2,3]. Vegetation decline could also accelerate global warming through the altered carbon and water cycles [4]. These predictions are supported by a large number of observed ecosystem-scale forest mortality events during the past 20 years [5]. The scientific community has reacted by an increased interest in developing methods for predicting plant survival under drought [6–۱۰]. The most commonly used concept in these models is based on the theories about xylem vulnerability to embolism [11
], and its connection to stomatal closure [12,13]. These theories suggest that during drought water tension in the xylem increases leading to embolization of xylem conduits. To prevent catastrophic loss of xylem conductivity plants close their stomata before a water tension threshold isreached. Thisthreshold depends on plant species and is linked with xylem vulnerability to embolism [11
]. Even after stomatal closure, plants slowly lose water through the bark and cuticular tissue of leaves. Stomatal closure does not completely prevent additional embolism [14
,۱۵
], but it significantly reduces water loss rates and embolism propagation. The theories on xylem vulnerability to drought and its connection to stomatal closure point are robust and supported by a wealth of empirical evidence [11
,۱۳,۱۶,۱۷], but the predictive power of this approach concerning plant survival time is limited [9,11
,۱۸]. We lack knowledge on how to define the needed thresholds of catastrophic hydraulic failure [19,20]. Findings on a metaanalysis of 19 recent plant mortality studies on 26 species around the world suggests that 60% or higher loss of conductivity leads to mortality (defined as loss of leaves or cessation of respiration [21
]), while many other studies have used thresholds on 50–۸۸% [۱۹,۲۰]. Other open questions include how fast plants would die once a threshold is reached [8], and how availability of new or stored carbohydrates, and their use impact these thresholds, and survival or revival capacity after drought [22,23].