Climate and Atmospheric Science (ICAS) PhD Projects

Fig 1: (a) A cumulonimbus anvil over West Africa viewed from the BAe146 research aircraft August 2006; (b) an “arcus” cloud, marking the leading edge of the cold air outflow near the ground (also from the aircraft), and the shaft of heavy rain a few kilometres behind it (to the left, here).

From our recent field experiments in Africa and Europe, we know that after a period of dry conditions, the patterns of rainfall from a storm leave a signature in the soil moisture which can then control the occurrence of subsequent storms. For instance, Taylor et al. (2007, 2010) showed a dramatic case in which a region of wet soil left by a storm in the northern Sahel of West Africa caused the development of a major storm on the following day. This process is important for the local weather and climate, because it means that in principle, satellite remote sensing of the surface state will improve our ability to predict where new storms will occur. In essence, the development of the second storm was determined by observable features rather than being a random, stochastic event.

Fig 2: Contrasting regions of dry and wet soil in the Sahel, as seen from the air.

The feedback of soil moisture on atmospheric convection has a strong control on the climates of Africa and parts of Europe, with particular impact on agriculture, since it influences the seasonal distribution of rainfall on horizontal scales down to about 10km. Our work in Africa has shown how seasonal rainfall can vary by a factor of 2 over short horizontal distances of only 10 km or so – meaning that a wet season in one village can be a dry season for their neighbours. Understanding, monitoring and predicting this behaviour has the potential for great benefit to the local populations. In Europe, there has been major concern in recent years regarding the severity of droughts caused by “blocking” anticyclones, with the 2003 example being one in which many deaths in France and other parts of Western Europe were linked to the high temperatures. For the 2003 case, there was strong evidence that the high temperatures and the persistence of the drought were linked to the fact that the soil moisture had become very low. Again, the relationship between soil moisture and rainfall – in this case the suppression of rainfall – is of great significance. These feedback processes also have major relevance to the local impacts of climate change in Africa and Europe: in a drier future climate, soil moisture feedbacks are likely to be stronger in general.

Although there is now an accumulation of evidence of a feedback of soil moisture on subsequent convective storms, there are many unresolved theoretical and technical questions: for instance,

• How does the horizontal scale of a soil moisture feature influence its ability to initiate a new storm? For instance, we have evidence that the combination of dry and wet soils in proximity is favourable for storm generation, but a very large dry area will suppress storms.
• How does the degree, and pattern, of vegetation cover control the feedbacks, and how does this evolve through a season?
• How can these processes be represented in global weather and climate models?

In global weather and climate models, convective storms are generally unresolved by the coarse numerical grids used, meaning that they have to be parametrised in terms of the larger-scale atmospheric state. This kind of representation is still quite crude, but the new generation of models is starting to contain information about the structure of storms within a given gridbox – the types of storm and their internal morphology. For a number of years, a comparable approach has been taken in representing diverse land-surface characteristics within a grid-box (a “tiling” approach).

In this project, the student will develop idealised computer models of the feedback of the land surface on convective storms, using Met Office research and forecasting models. Through setting up idealised models, with very high resolution of the storms and the surface state, we will be able to isolate the physical processes in the atmosphere and surface state which control the feedbacks. These simulations will be used to develop theoretical models of the coupling, and thereby to develop a framework for the interpretation of satellite data and of field data from Africa and Europe. The student will also spend some time in the field, collecting new measurements for evaluation of the satellite and model results.

Fig. 3: Temperature (shaded), cloud extent (contoured) and wind vectors from a numerical simulation of the atmospheric boundary layer flow over a heterogeneous surface. The simulation comes from the Met Office Large Eddy Model (LEM), which was configured to represent a vegetated surface in West Africa. Cloud formation is enhanced, at around x=55km, at a boundary between cropland (left) and forest (right). These simulations are in good agreement with aircraft observations from AMMA.

The project will be a CASE studentship, co-supervised by Dr Chris Taylor at CEH Wallingford. It will be linked to the NERC-funded SWELTER project (led by Chris Taylor) analysing the influence of ground water on European climates, as part of the “Changing Water Cycle” programme, and will therefore be conducted in a collaborative, interdisciplinary environment. The work will focus on monsoon onset in Africa when the ground is initially dry and vegetation limited, and summertime blocking events in Europe. SWELTER will be analysing European soil moisture using satellite remote sensing techniques, and linking the observations to the occurrence of convective storms. SWELTER will also be running and testing large-scale models of the atmospheric weather and climate. This project will link to SWELTER by studying the detailed processes involved in the land-atmosphere coupling, through new surface measurements, closer examination of case-studies and through computer modelling.

The project will also be part of the African Monsoon Multidisciplinary Analysis (AMMA) programme (2000-2020; Redelsperger et al. 2006), which has collected many measurements from various climatic zones of West Africa. We will make use of data from AMMA, and will link the results to various AMMA studies of the climate and the water cycle. Closely related to AMMA, we will link the work with the high-resolution modelling of the Cascade project (NERC) in which we are developing models of convection over Africa, in collaboration with the University of Reading, UEA and the Met Office.

The work will be closely linked with the Met Office and Hadley Centre, through our partnership with their programme of research into African weather and climate.

Fig 4: A flux station in the Northern Sahel, installed for the AMMA project. Data from these systems will be used in this studentship.

The supervisors of this project have collaborated successfully for several years, and bring together expertise in atmospheric and land surface dynamics, numerical modelling and observational research. The School of Earth and Environment at Leeds has a well-knit group of around 40 atmospheric-science research students. The School provides an excellent programme of generic skills training for research students, as well as a number of subject specific activities including the Arran Summer School and a number of Masters-level taught modules.

Redelsperger, J-L., C. D. Thorncroft, A. Diedhiou, T. Lebel, D. J. Parker, and J. Polcher, 2006,'African Monsoon Multidisciplinary Analysis: An International Research Project and Field Campaign', Bull. Am. Met. Soc. 87, 1739-1746. doi:10.1175/BAMS-87-12-1739

Taylor, C. M., D. J. Parker, and P. P. Harris, 2007, 'An observational case study of mesoscale atmospheric circulations induced by soil moisture', Geophys. Res. Lett., 34, L15801doi:10.1029/2007GL030572

Taylor, C. M., P. P. Harris and D. J. Parker, 2010, 'Impact of soil moisture on the development of a Sahelian mesoscale convective system: a case-study from the AMMA Special Observing Period', Q. J. R. Meteorol. Soc., 136, 456-470. doi:10.1002/qj.465