Predictive and Explanatory Modeling of Ecosystem Response to Soil Water Stress
2023-05
Loading...
View/Download File
Persistent link to this item
Statistics
View StatisticsJournal Title
Journal ISSN
Volume Title
Title
Predictive and Explanatory Modeling of Ecosystem Response to Soil Water Stress
Authors
Published Date
2023-05
Publisher
Type
Thesis or Dissertation
Abstract
Plant leaves close their stomatal apertures in response insufficient soil moisture, and, thus, control ecosystem water, carbon, and energy cycles. Many terrestrial biosphere models (TBMs) underlying general circulation models represent this ecosystem response to soil water stress with an empirical correction function (β) of soil moisture---a convenient approach that can produce large prediction uncertainties. To reduce this uncertainty, TBMs have increasingly incorporated physically-based Plant Hydraulic Models (PHMs) at the cost of more poorly-constrained parameters. Therefore, understanding why and when PHM and β predictions diverge would usefully inform model selection within TBMs. Here, we reconcile PHMs and β by illustrating that the soil-plant hydraulic transport in PHMs (parametrized by conductance) couples the effects of soil and atmospheric dryness on stomatal closure, and β emerges as an infinitely-conductive PHM, breaking the coupling. As a result, PHM and β transpiration predictions diverge most for soil-plant systems with low hydraulic conductance (transport-limited) that experience large variability in atmospheric dryness with moderate soil water stress. We confirm these results in a TBM case study at an eddy covariance site, and propose a `dynamic' β that compromises between β and PHMs. Our efforts to extend our analysis across biomes revealed a practical problem: which eddy covariance sites observed soil water stress? Many studies use explanatory models to infer stress signals from eddy covariance data, but the studies vary widely in their goals, selected sites, and inference assumptions. Furthermore, the sensitivity of the inferred stress signals to the numerous assumptions (i.e. robustness) are rarely quantified, giving pause to meta-analyses to identify stressed sites. Here, we develop a framework that quantifies the robustness of inferred soil water stress signals to the most prevalent data and modeling assumptions in literature, and apply this framework to 150 eddy covariance sites. Only 7 sites have a robust soil water stress signal due primarily to poor explanatory model performance. Furthermore, the results challenge common approaches of generalizing site-specific stress signals across biomes. We improve upon the robust stress signal detection by identifying several assumptions to improve explanatory model performance across most sites. Specifically, the response variable for parameter estimation and phenology considerations allow the robust stress framework to identify 30 sites with robust soil water stress signals. Lastly, we provide a user-friendly visual tool that rank-orders sites by the robustness of their soil water stress signals. Our research provides fundamental insights into ecosystem soil water stress prediction, while providing practical guidance on how to infer stress signals from data.
Description
University of Minnesota Ph.D. dissertation. May 2023. Major: Civil Engineering. Advisor: Xue Feng. 1 computer file (PDF); x, 201 pages.
Related to
Replaces
License
Collections
Series/Report Number
Funding information
Isbn identifier
Doi identifier
Previously Published Citation
Suggested citation
Sloan, Brandon. (2023). Predictive and Explanatory Modeling of Ecosystem Response to Soil Water Stress. Retrieved from the University Digital Conservancy, https://hdl.handle.net/11299/257056.
Content distributed via the University Digital Conservancy may be subject to additional license and use restrictions applied by the depositor. By using these files, users agree to the Terms of Use. Materials in the UDC may contain content that is disturbing and/or harmful. For more information, please see our statement on harmful content in digital repositories.