2020 ESA Annual Meeting (August 3 - 6)

COS 219 Abstract - Fitting anomalous water potential responses into the anisohyric/isohydric spectrum

Erik Hamerlynck, Eastern Oregon Agricultural Research Center, USDA-ARS, Burns, OR, Keith Reinhardt, Biological Sciences, Idaho State University, Pocatello, ID and Travis E. Huxman, Ecology and Evolutionary Biology, University of California, Irvine, Irvine, CA
Background/Question/Methods

Understanding how woody plants respond to extreme drought is a critical to understanding the structure and function of plant communities and ecosystems under ongoing global climate change. Recently, quantitative frameworks have been developed that allow using simple predawn and midday water potential measurements as reliable proxies to determine where on the anisohydric/isohydric spectrum woody plants fall, an important feature in observed patterns of drought-induced mortality in water-limited systems. However, these approaches have specifically excluded anomalous water potential responses (Ψanomaly), where predawn water potential (Ψpd) are markedly more negative than midday xylem potential (Ψmid), and have been observed under extreme drought conditions across Southwestern US deserts. We used a historical data set of water potential and photosynthetic gas exchange in the evergreen creosotebush (Larrea tridentata) and the drought deciduous silver bursage (Ambrosia dumosa) gathered under exceptional drought conditions to place Ψanomaly into the basic xylem potential-based anisohydric/isohydric framework to assess its applicability across the full range of drought conditions.

Results/Conclusions

Ψanomaly were observed at the low end of Ψpd for both species, they occurred even when maximum Ψpd was quite high (ca -2 MPa). Increasingly negative anomaly was associated with a tightening in the range of Ψpd, Ψmid and stomatal conductance (gs), but dynamic range of gs was maintained even into the anomalous range. Regression-based characterization of isohydric/anisohydric behavior using the traditional temporal context of Ψpd and Ψmid as indicators of maximum and minimum Ψ (Ψmax and Ψmin) were not consistent with observed variation in gs or patterns of canopy die-back and whole-plant mortality. However, removing the temporal context, and simply plotting Ψmax and Ψmin provided a clear picture of how Ψanomaly fits into plant drought responses. The slopes of Ψmin vs. Ψmax of anomalous observations (σ = 0.81 for L. tridentata and σ = 0.47, for A. dumosa, respectively) were lower than those of normal Ψ observations (σ = 0.88 and σ = 0.63, for L. tridentata and A. dumosa, respectively). These findings indicate that L. tridentata is more anisohydric than A. dumosa, and that during periods of Ψanomaly both become increasingly isohydric, consistent with local patterns of plant mortality. Our results show challenging some basic assumptions associated with plant water relations, such as the relationship of midday to predawn Ψ or periods of system equilibrium is useful in testing theory associated with plant function under extreme drought conditions.