Department of Geosciences

Vegetation-Hydrology Feedbacks

How does the development of vegetation patterns depend on hydrological patterns and to which extent are these controlled by the vegetation patterns?

PhD Researcher: Shanghua Li
Supervisors: Katja Tielbörger (University of Tübingen), Sebastian Gayler (University of Hohenheim), Nandita Basu (University of Waterloo)

Introduction

In recent decades, dynamic vegetation models have emerged as a powerful tool for quantifying ecosystem dynamics in response to climate and hydrologic processes. However, state-of-the-art vegetation models typically assume static hydrological conditions, such that the effects of climate and climate change on near surface hydrology are not accurately represented. Vice-versa, hydrological models treat vegetation as a rather static 'green layer' and ignore its dynamic nature in space and time. This topic addresses this gap by quantifying the degree to which hydrologic transients impact predictions from vegetation dynamics models and vice-versa. Our approach involves a step-wise coupling of a dynamic vegetation model with a transient 3D hydrologic model.
The aim of this topic is to develop a general hydrological model which has the similar time and spatial scale with the current vegetation model. The interaction between this two models can be that hydrology model use vegetation properties as input variables (e.g. Leaf area index and root depth), and produce information about hydrodynamics that can feed into vegetation models (e.g. soil moisture, pressure head). After that, couple model types that have not been coupled before.

Approach

Our approach involves a step-wise coupling of a dynamic vegetation model with a transient 3D hydrologic model.

  1. Coupled a dynamic vegetation model and a plot-scale hydrological model at annual time step under one climate type.
  2. Increased coupling time scale from annual to seasonal. Coupled a dynamic vegetation model and a hillslope-scale hydrological model with stochastic climate input under three different climate types, from dry climate to wet climate.

Preliminary Results

Our preliminary results found that the dynamic plants simulated with the coupled model greatly influenced soil water content under drought water stress conditions, while the implementation of unrealistic static plants led to a dryer hydrological state. Vice-versa, we found increased soil water heterogeneity simulated with the coupled model decreased plant diversity. The heterogeneity-diversity relationship changed with extended time scale.