COS 179-3 - Drivers, variability and memory in plant water potential of Juniperus osteosperma

Background/Question/Methods

Plant water potential is a key driver of water transport and carbon uptake, yet its response to environmental drivers remains poorly resolved. One limitation is the monthly to weekly temporal resolution of water potential necessitated by destructive sampling methods, which is generally insufficient to distinguish soil versus atmospheric moisture limitations. Furthermore, in water-limited ecosystems, vegetation responses can depend on past abiotic conditions at different strengths and temporal patterns. Here, we evaluated the effects of past environmental conditions on predawn water potential of Juniperus osteosperma. Seven mature trees were each instrumented with two automated stem psychrometers during the 2021 growing season in southeastern Utah. Using the stochastic antecedent model in a hierarchical Bayesian framework, we ask: 1) What is the relative sensitivity of predawn water potential (Ψ_PD) to atmospheric versus soil moisture? 2) On what timescales do the significant drivers operate? 3) What is the relative variability of environmental effects within versus among trees?

Results/Conclusions

Among individual trees, Ψ_PD reached seasonal minima ranging between -3.7 and -8.5 MPa; distinct water potentials were also evident during routine instrument rotation among branches of the same tree. Therefore, we designed the hierarchical model to account for the nested effects of branches within trees and trees within population (R² = 0.97). At the population level, shallow soil moisture (W₁₀) positively correlated with Ψ_PD, while deep soil moisture (W₅₀) primarily had a negative effect. Interestingly, maximum daily vapor pressure deficit (D_max) was not significantly associated with Ψ_PD at the population or tree level, as individual branches exhibited a range of positive, negative, and neutral responses to D_max. While both W₁₀ and D_max impacted Ψ_PD on relatively short time scales of 1-2 days prior, W₅₀ from 5 weeks ago was also strongly influential. Low winter precipitation resulted in minimal recharge of deep soil in 2021, yet J. osteosperma were nevertheless sensitive to deep soil moisture. Overall, Ψ_PD responses to environmental drivers were non-stationary, as evidenced by highly variability within and among trees and dependence on antecedent conditions. We confirm previous research that J. osteosperma are physiologically responsive to summer precipitation and highly efficient at extracting transient shallow soil moisture.