Investigating links between vegetation and the water cycle using observations and models

Supervisors: Steve Arnold (SEE), Dominick Spracklen (SEE), Chris Taylor (Centre for Ecology and Hydrology)

Background and Motivation

The biosphere and atmosphere are intimately coupled aspects of the Earth system, and their interaction exerts strong control on climate. Vegetation plays a role in controlling energy, carbon and water vapour fluxes between the surface and atmosphere. While patterns of rainfall across the globe are known to exert control on the distribution of vegetation, there is much less understanding regarding the role of vegetation in controlling rainfall. There is some limited observational evidence, and experiments with climate models that suggest changes in vegetation due to e.g. deforestation may impact regional rainfall. However, a direct causal link or observationally-based constraint have not been made. Recent work in Leeds, using a novel combination of satellite observations and an atmospheric transport model has yielded some of the first observational evidence for a strong control from vegetation on rainfall over large regions of the globe. This evidence links evapotranspiration from plants and trees directly to daily variability in rainfall patterns in many tropical regions. This work ignites many intriguing questions regarding the effects of man-made or climate-driven changes in vegetation on moisture transport and rainfall.

Future climate projections for Europe show large uncertainty in how rainfall patterns will change under greenhouse-gas induced warming. Part of this uncertainty likely results from inadequate treatment of land surface-atmosphere coupling in climate models. Of particular interest for Europe is how the frequency of 'heat-wave' or drought periods may change into the future. Such events (for example, Summer 2003), have serious implications for human health, air quality, agriculture and wildfire activity.

As soils become dry during warm summers, evaporation of soil moisture into the atmosphere is limited. In addition, vegetation can become drought-stressed leading to reduced evapotranspiration. This can lead to a feedback where surface temperatures are increased further due to absorbance of solar radiation which is not being used to drive evaporation. Furthermore, if soil dries over a large region, this can lead to changes in atmospheric circulation patterns. These changes in surface moisture fluxes and atmospheric circulation will affect the formation and patterns of rainfall, and transport of emissions important for air quality.

In this project you will investigate links between soil moisture, vegetation, atmospheric water transport and rainfall over Europe, under drought and non-drought conditions. You will use observations from satellites, ground-based stations and data from European weather centres in conjunction with a Lagrangian transport model to identify land-atmosphere coupling under different conditions, and investigate atmospheric moisture sources and transport. You will  investigate the implications of these processes for our ability to simulate land surface-atmosphere exchange processes in the JULES land model, and the consequences for future projections of rainfall, surface temperature and air quality in a climate model.

Project Methodology and Objectives

In this project you will use state of the art computer models that simulate atmospheric transport, climate and atmospheric chemistry, and models of the global terrestrial biosphere and land surface. You will develop modifications to the transport model to include tracking of water vapour and terms of the atmospheric moisture budget. You will synthesise observations from surface stations and satellites to investigate land-atmosphere interaction processes and to evaluate model processes.

You will use the OFFLINE atmospheric back trajectory model (Methven, 1997) to calculate the 3-D transport of air parcels around the atmosphere, and combine these with satellite remote-sensed products of vegetation and rainfall. In our preliminary work we have used Moderate Resolution Imaging Spectroradiometer (MODIS) retrievals of leaf area index (LAI) to create a "vegetation footprint" of air masses circulating over Europe (see global example in Fig. 1). Using observations of precipitation, you will explore relationships between vegetation footprint and daily rainfall over different periods. Our preliminary work focused on limited regions of the tropics and demonstrates the technique (Fig. 2). A challenge will be to explore this systematically across Europe under different climatic conditions using a range of different vegetation satellite products.

You will then analyse water vapour origin and transport and links to precipitation. The aim is to extend our recent research to explore the water budget along 3D back trajectories. Whilst numerous previous studies have explored the simulated water budget in models (e.g., van der Ent & Savenije, 2011) none have linked such analysis to remote-sensed observations. You will include tracers in the OFFLINE model to enable us to track water vapour along back trajectories, and to identify the origin and transport pathways of water vapour and link this to observed patterns of precipitation. In conjunction with evapotranspiration and soil water diagnostics from the Global Land Data Assimilation System (GLDAS), a moisture budget for European air masses can be assembled to identify key differences between land-atmosphere coupling under drought and non-drought conditions.

Finally, you will evaluate land-atmosphere coupling in the JULES land surface model for heat-wave and non-heat-wave years by comparing with the observationally-derived processes above, and evaluate how these processes may impact future European rainfall, atmospheric circulation, climate and air quality using an Earth system model.

The project provides a unique breadth of experience and strong research environment for the student. The project is designed to offer scope for the student to develop their own ideas while exploring important scientific issues relating to land-use change and precipitation. The availability of long-term remote sensed rainfall data along with the availability of the necessary computer resources to calculate the large number of back trajectories required for this analysis make this project timely. The project will be conducted in collaboration with the Centre for Ecology and Hydrology who will act as a CASE sponsor.

For this project you should have a 1^st or 2:1 degree in a relevant physical science discipline (physics, maths, chemistry, meteorology). For informal enquiries and further information on this project, please contact Dr. Steve Arnold (s.arnold(at)leeds.ac.uk) or Dr. Dominick Spracklen (d.spracklen(at)see.leeds.ac.uk).