COS 86-10 - Antecedent soil moisture and VPD interactively control water potential and hydraulic thresholds in Larrea tridentata

Thursday, August 11, 2016: 4:40 PM
Floridian Blrm A, Ft Lauderdale Convention Center
Jessica S. Guo1,2 and Kiona Ogle1,2,3, (1)Center for Ecosystem Science & Society, Northern Arizona University, Flagstaff, AZ, (2)Department of Biological Sciences, Northern Arizona University, Flagstaff, AZ, (3)School of Informatics, Computing, and Cyber Systems, Northern Arizona University, Flagstaff, AZ
Background/Question/Methods

Water potential (Ψ) is an integrated measure of plant water status that is regulated by stomatal responses to soil water content (SWC) and vapor pressure deficit (VPD).  Extreme SWC and VPD events are expected to drive plants past Ψ thresholds (Ψcrit) that lead to loss of hydraulic function. Understanding environmental conditions that push plants past Ψcrit are therefore important for predicting responses to a changing climate.  We propose that plant Ψ does not solely respond to concurrent environmental factors, but is influenced by an accumulation of extreme events. Our goal was to quantify the effects of antecedent SWC and VPD on Ψ in the desert shrub Larrea tridentata and to identify conditions leading to Ψcrit. From June 1998 to July 1999, predawn (ΨPD) and diurnal Ψ were measured on 16 Larrea shrubs at the Jornada LTER in New Mexico, USA.  Data were analyzed with two linked models, one for ΨPD and one for diurnal Ψ that modeled deviations relative to ΨPD. We employed a stochastic antecedent model to incorporate antecedent effects of maximum daily VPD, daily SWC at 0-30 and 30-60 cm, and half-hourly VPD. This approach allowed us to estimate time lags and relative importance of these factors for predicting conditions leading to Ψcritthresholds.  

Results/Conclusions

Variation in ΨPD was primarily explained by antecedent maximum daily VPD, antecedent SWC at two depths, and their two-way interactions (R2=0.76).  Importantly, the lag response of ΨPD to these drivers varied such that maximum VPD four days prior, shallow SWC two days prior, and deep SWC six days prior to the ΨPD measurement had the greatest influence on ΨPD. Furthermore, increased shallow soil moisture resulted in higher ΨPD, but more shallow soil moisture also increased the sensitivity of ΨPD to maximum VPD experienced four days ago. Variation in diurnal Ψ appears to be more stochastic (R2=0.33), but was partly explained by ΨPD, antecedent VPD (over 30-minute intervals), and its interaction with antecedent SWC at two depths. High VPD within the 30-min prior to measurement resulted in more negative Ψ, indicating a fast stomatal response to VPD, and increased deep soil moisture reduced sensitivity of Ψ to VPD.  These results suggest that extreme VPD conditions at multiple temporal scales (e.g., four days ago and 30-min ago) followed by precipitation events that recharge the top 30 cm of soil can lead to Ψ approaching Ψcrit.