EGU General Assembly 2020
© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Plant hydraulics accentuates the effects of atmospheric moisture stress on transpiration

Alexandra Konings1, Yanlan Liu1, Mukesh Kumar2, Xue Feng3, and Gabriel Katul4
Alexandra Konings et al.
  • 1Stanford University, Stanford, CA, United States of America (
  • 2University of Alabama, Tuscaloosa, AL, United States of America
  • 3University of Minnesota, Twin Cities, Minneapolis, MN, United States of America
  • 4Duke University, Durham, NC, United States of America

Transpiration directly links the water, energy and carbon cycles. It is commonly restricted by soil (through soil moisture) and atmospheric (through vapor pressure deficit, VPD) moisture stresses governed by the movement of water through plants, also known as plant hydraulics. These sources of moisture stress are likely to diverge under climate change, with globally enhanced VPD due to increased air temperatures but more variable and uncertain changes in soil moisture. In most Earth system and land surface models, the ET response to each of the two stresses is evaluated through independent empirical relations, while neglecting plant hydraulics. Comparison of these two models is challenged by the difficulty of ensuring any perceived differences are due to the model structure, not an imperfect parametrization. Here, we use a model-data fusion approach applied to long-term ET records collected at 40 sites across a diverse range of biomes to demonstrate that the widely used empirical approach underestimates ET sensitivity to VPD, but compensates by overestimating the sensitivity to soil moisture stress. The bias originates from the joint control of leaf water potential on plant response to soil moisture and VPD stress. To a lesser degree, it also overestimates from increased sensitivity to VPD under dry (low leaf water potential) conditions in the plant hydraulic model. As a result, a hydraulic model captures ET under high-VPD conditions for wide-ranging soil moisture states better than the empirical approach does. Our findings highlight the central role of plant hydraulics in regulating the increasing importance of atmospheric moisture stress on biosphere-atmosphere interactions under elevated temperatures.

How to cite: Konings, A., Liu, Y., Kumar, M., Feng, X., and Katul, G.: Plant hydraulics accentuates the effects of atmospheric moisture stress on transpiration, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-8472,, 2020

Display materials

Display file

Comments on the display material

AC: Author Comment | CC: Community Comment | Report abuse

Display material version 1 – uploaded on 02 May 2020
  • CC1: Comment on EGU2020-8472, Jasper Denissen, 08 May 2020

    Hey Yanlan,

    Thank you so much for the nice, self-explanatory slides. How do you account for spatial heterogeneity in the MCMC optimization of hydraulic parameters? Since hydraulic parameters depend on soil characteristics, which might vary within the footprint of the EC towers of the 40 FLUXNET sites you are using. So could you elaborate a bit more on the optimization of the hydraulic parameters?

    Very nice work, I'm looking forward to reading your publication. 

    • AC1: Reply to CC1, Yanlan Liu, 08 May 2020

      Hi Jasper,
      Thank you. This is a great point! The MCMC optimized the plant hydraulic parameters at each site against observed ET under given site-specific properties including soil characteristics. These site-specific properties were assigned based on literature information or additional datasets if the former was unavailable. One exception is that the uncertainty of the nonlinear parameter in the soil-water retention curve affects the soil-plant hydraulic system, especially under stress. So this soil hydraulic parameter was also retrieved together with plant hydraulic parameters using MCMC. These spatially heterogenous properties contribute to the variation of optimal hydraulic parameters across the 40 sites, which is included in the paper about to come out. However, we did not consider the variability within the footprint of each flux tower. All the retrieved parameters represent the integrated property of each site. I hope this helps answer your question. Thanks again for your interest and comment!