- Analysis of precipitation from the worldwide first ensemble of convection permitting models
- Cutting-edge identification of how land-atmosphere interactions influence future changes in regional-scale precipitation
- Highly relevance to climate change impact assessment
Anthropogenic climate change is expected to lead to substantial increase in extreme precipitation with impact on flooding. However, simulating precipitation and how it may change in the future is still a major challenge, and associated with large uncertainties. One of the main problems is the important role of convection in generating precipitation. Until recently the spatial resolution of climate models was too coarse to resolve convection, and it was instead only approximately represented by so-called parameterisations. Over the last few years very high-resolution regional climate models, below 4km resolution, have been developed that resolve large storms and meso-scale organisation of convection explicitly. Simulations with such Convection-Permitting Models (CPMs) have shown much more realistic precipitation fields (Figure 1, Prein et al 2015; Chan et al 2014) and substantially stronger future changes in short-duration precipitation extremes compared to standard regional climate models (Kendon et al. 2014).
For reducing the uncertainties in regional climate change projections it is crucial to understand the processes that cause the climate change signal (Maraun et al. 2017, Maraun and Widmann 2018). CPMs allow for the first time to investigate in sufficient detail how land-atmosphere interactions, for instance related to soil-moisture variations or urban areas, influence precipitation and its future changes.
This project will exploit the first ensemble of CPM simulations over the UK within the UK Climate Projections UKCP18 project, which will feed into the Climate Change Risk Assessment (CCRA3). The aim is to compare the representation of soil moisture and surface hydrology in CPM and standard climate models, and investigate how it impacts on present and future convective rainfall. The ensemble will not only provide a large database for this analysis, but also allow for the first time an estimate of uncertainty in future changes at convection-permitting scale, thus providing UK risk assessment studies with more reliable climate change projections at local and hourly scales.
The CPM ensemble for the present climate will be used to investigate the sensitivity of soil moisture in the CPM’s land surface model JULES to the simulated precipitation. The project will also analyse the sensitivity of precipitation distributions, especially extremes, and spatial characteristics in CPMs to soil moisture and surface types. Implications for the UK will be assessed using the soil survey for England and Wales and the National Soil Inventory. The above-mentioned two processes will be separately assessed using temporally lagged analysis and sensitivity simulations with the CPMs and JULES. The simulations will be validated against multi-year evaporation flux data at several UK locations.
In the second phase the CPM ensemble and a standard regional model ensemble with 12km resolution for the future (2060-2080) will be used to analyse changes in precipitation distributions and spatial characteristics, as well as the role of the land-atmosphere interactions in these changes.
Training and Skills
The Met Office will provide training in convection-permitting models, in their interpretation and in handling large datasets using specific programming facilities. The student will have the possibility to become familiar with running the UM model on the super-computer Cray if required. CEH and Cranfield University will provide training in the use and interpretation of the land surface model JULES and surface flux measurements. University of Birmingham will provide training in downscaling and regional climate change assessment, and in urban climate studies. There is the opportunity to attend related lectures on the MSc program ‘Applied Meteorology and Climatology’.
CENTA students are required to complete 45 days training throughout their PhD including a 10 day placement. In the first year, students will be trained as a single cohort on environmental science, research methods and core skills. Throughout the PhD, training will progress from core skills sets to master classes specific to CENTA research themes.
Year 1: the PhD student will investigate the differences in the representation of soil moisture and surface hydrology between CPMs and standard regional climate models, and compare to observational data from surface flux stations, satellite remote sensing and research flights where available. He/she will identify misrepresented, missing processes and/or different model settings to be tested in the second year.
Year 2: the PhD student will implement and analyse sensitivity test simulations for identifying the role of different land-surface–atmosphere processes.
Year 3: the PhD student will investigate how the different representation of the water cycle and land-atmosphere interactions in CPMs and standard regional climate models affect the climate change projections. He/she will draw general conclusions on uncertainties in regional projections and on how to reduce them further.
Partners and collaboration (including CASE)
The student will be hosted jointly by the Met Office and the University of Birmingham, and also have extended visits at CEH and Cranfield University. This will ensure that the work will be undertaken in an efficient way, and will provide experience and networking possibilities in different research environments.
The project will bring together Met Office expertise in climate analysis and convective-scale climate modelling with CEH expertise in representing land surface hydrology in coupled land-atmosphere models, Cranfield University knowledge on soil properties and behaviour, and Birmingham experience with regional climate change, downscaling and urban modelling.