Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Accueil du site → Doctorat → États-Unis → 2021 → Terrestrial Water Storage Change Over Dryland Regions Using Observational Data and Model Simulations

University of Arizona (2021)

Terrestrial Water Storage Change Over Dryland Regions Using Observational Data and Model Simulations

Chang, Liling

Titre : Terrestrial Water Storage Change Over Dryland Regions Using Observational Data and Model Simulations

Auteur : Chang, Liling

Université de soutenance : University of Arizona

Grade : Doctor of Philosophy (PhD) 2021

Drylands occupy a large fraction of global terrestrial areas and are home to around 2.5 billion people. Dryland ecosystems are important constituents of global biodiversity and impact the global carbon cycle. However, drylands typically experience water shortages due to unbalanced atmospheric water demand and supply, being exacerbated by climate change and human water withdrawal. Water resources are critically important for both human societies and ecosystems to function. Therefore, quantifying water availability and changes is essential for dryland research, water management, and policy making. This dissertation comprises three studies aiming at quantifying and understanding terrestrial water storage (TWS) changes and flux partitioning patterns over global-, regional-, and catchment-scale drylands. The first study mainly evaluates TWS changes over global drylands from April 2002 to January 2017 using satellite observations. My collaborators and I find that water storage declines over global drylands, with an emphasis on arid and hyperarid regions. The drying trend is more pronounced over midlatitude subtropical drylands. Using a process-based model, the drying trend is attributed to human interventions and climate variations. The second study is focused on disaggregating the impacts of climate variations on the declining TWS over the Tigris-Euphrates river basin by analyzing observational datasets and conducting model simulations. Over the basin, interannual climate variability appears to be a dominant contributor, followed by decadal climate change, while the impacts of climate intraannual variability on the declining TWS are negligible. The interannual climate variability affects TWS mainly through a nonlinear response of monthly TWS changes to varying aridity conditions. During dry periods, TWS depletes sharply, mainly caused by rapid responses of plant transpiration (T) through root water uptake of subsurface water. However, TWS recovers slowly during wet periods due to the evaporative loss from the wetted surface soils. This study emphasizes the importance of evapotranspiration (ET) and its partitioning in controlling TWS dynamics over drylands. The third study analyzes ET and its partitioning over the Marshall Gulch catchment in Arizona. This study investigates four factors controlling ET partitioning to understand why the current state-of-art land surface models tend to underestimate the T/ET ratio. The two dominant contributors are lateral water flow driven by topography and surface soil resistance to evaporation. The lateral water flow spreads out soil moisture along hillslopes. Over the middle to upper slopes, soil surface evaporation is more suppressed than T because plant roots can access deeper water. A more process-based soil resistance to evaporation scheme enhances the T/ET ratio than an empirical formula. The other two factors, i.e., topographic shading and scattering effects and leaf dynamics, exert negligible impacts on the ET partitioning. However, their impacts are subject to further studies in catchments under other climates.


Version intégrale (48,4 Mb)

Page publiée le 1er décembre 2021