Master thesis - Global ecosystem flux modelling
Abstract
Eco-evolutionary optimality principles can be used to predict how plants adapt and acclimate to their environment. Using a model funded on such principles, you will investigate how of water use efficiency and rooting depth varies across the globe and governs ecosystem fluxes.
Main supervision: Prof. Dr. Benjamin Stocker Co-supervision: Francesco Giardina
As a result of the higher CO2 concentration in the atmosphere, plants have adapted their stomatal regulation to increase the water-use efficiency (WUE) at the leaf level. Another driver of plant transpiration is foliage surface area, and trends in land surface greenness are thus partly linked by enhanced WUE under elevated CO2. As climate change progresses, it is critical to understand the responses of terrestrial ecosystems and surface-atmosphere exchanges of CO2 and water vapour to global environmental change.
The objective of this project is to model spatial and seasonal variations in WUE to predict evapotranspiration (ET) within an established global photosynthesis and acclimation model (productivity model or simply P-model; Stocker et al. 2020). The core of this model employs a principle of eco-evolutionary optimality to simulate acclimation of photosynthesis and transpiration based on the trade-off between costs and benefits from water loss and C assimilation. You will use observed root zone moisture storage capacity (SR) to model ET at FLUXNET sites and investigate predictions under progressing droughts. We expect that using the combination of the two key quantities WUE and SR, which are under plant-control and subject to different acclimation and adaptation across the Earth?s climate zones, will improve global ET predictions. It has been demonstrated that rooting zone adapts its extent as a function of the phase lag between peak seasonal radiation and precipitation. Following this framework, we will investigate the effect of a spatially varying rooting depth on the ratio of actual ET over precipitation.
This thesis is a great starting point for mechanistic ecosystem modelling and working with large datasets of the terrestrial biosphere. Basic experience with working with R, Python, Fortran, or other programming languages are a requirement. You may start as soon as you like. Please contact me directly if you’re interested.
It is possible to carry out this thesis in teamwork, where the partner thesis would implement a predictive model based on machine-learning instead of a mechanistic model.
