The UK experienced 12 major Atlantic depressions between mid-December 2013 and mid-February 2014. Individual storm events did not yield exceptional rainfall, but accumulated levels over the period led to extensive inland and tidal flooding of over 10,000 properties nationwide. The economic damages across England and Wales were estimated to be £1.3 billion, of which £320 million related to flood damage of 10,465 residential properties.
The Thames river basin district was one of the worst affected areas. Covering approximately 16,200 km2 across Greater London, Surrey, Berkshire and Oxfordshire, it is the most densely populated of all the UK river basins supporting over 13 million people.
In winter 2013/14 the Thames basin was subject to fluvial, pluvial (surface-water/flash) and groundwater flooding which led to widespread transport disruption and significant flooding of properties including local contamination from sewer surcharge. The 2013/14 floods were the fourth major flood event experienced in the catchment since 2003 and the Environment Agency estimate that more than 200,000 properties in the basin are at risk from a 1:100 year return fluvial flood event.
Flood prediction and prevention is thus a priority topic amongst local communities and stakeholders including the Environment Agency.
This case study targeted compound floods focusing on the co-occurrence of high rainfall, high river levels and high groundwater discharge in several sub-catchments of the Thames river basin, including:
- Lower Thames – greatest flood risk outside of Greater London and recognised as the largest concentration of properties not protected by flood defences in the UK
- Loddon – high concentration of fluvial and surface water flood risk plus areas known for groundwater flooding
- Thame and South Chilterns – geologically diverse catchment with a substantial number of high flood risk areas (see Fig. 1).
The focus was on reforecasting the 2013/14 flood event to see how far in advance we can predict compound flood events.
In this case study we:
- identified the risk of compound flood events in the Thames river basin by determining the drivers and the probability of occurrence of compound flood events for the sub-catchments of the Thames River basin (for today’s and the future climate).
- looked at the EFAS (European Flood Awareness System) monthly (up to 46 days ahead) and seasonal (up to 7 months ahead) hydrological re-forecasts of the 2013/14 flood event to determine whether they can give an early indication of compound flood events in the Thames river basin.
Tools and models
The following tools and models being implemented to address the two main outcomes mentioned above:
1) Monthly and seasonal hydrological reforecasts from the European Flood Awareness System (EFAS) were being run to determine whether they give an early indication of 2013/14 compound flood events in the sub-catchments.
EFAS uses the LISFLOOD hydrological model. LISFLOOD is a 5 x 5 km resolution model that uses meteorological inputs to produce hydrological forecasts at short to seasonal timescales (streamflow, groundwater storage and soil moisture).
The monthly (up to 46 days ahead) and seasonal (up to 7 months ahead) hydrological outlooks are a relatively new development in EFAS which are being led by the European Centre for Medium-Range Weather Forecasts (ECMWF). We anticipated furthering the development of extended and seasonal flood forecasts and worked closely with the ECMWF to produce ‘new’ EFAS flood forecasts for the Thames basin. The aim was to use meteorological inputs (from ECMWF experiments) that represent more accurately the real-life weather conditions observed over the UK during the 2013/14 floods. We hoped that the resulting flood forecasts would more closely match the river and groundwater levels observed during the 2013/14 floods when compared with the default EFAS outputs.
2) Bias corrected hydro-meteorological data (RACMO from KNMI) was statistically analysed to identify and characterise the occurrence of compound floods (as a combination of high groundwater discharge, high river level and high rainfall) in the Thames sub-catchments.
The compound flood risk was evaluated for today's climate (data tbc) and the future climate (using bias-corrected RACMO time series from 1950 to 2100).
The Environment Agency – their use of seasonal flood forecasts is currently limited, yet there is clear recognition that the Thames is sensitive to seasonal flooding due to changes in meteorological conditions and operational response. Advance detection of ‘at risk’ areas/catchments would be helpful when preparing for future flood events.
The Flood Forecasting Centre – a joint venture between the EA and UK Met Office, their aim is to provide improved flood risk guidance for England and Wales.
Local Authorities – responsible for developing and implementing local flood risk management strategies within the selected sub-catchments.
Partners and collaborators:
The ECMWF – home of the EFAS computational centre and leading the work in EFAS seasonal forecasts.
University of Reading – our work within the Thames (and specifically the Loddon sub-catchment) tied in closely with other collaborative research projects. These included catchment-based natural flood risk management, the impact of extreme rainfall events on the mobility of potentially toxic elements on floodplains and understanding the societal perception of flood risk (https://loddonobservatory.org/).
The Environment Agency (EA), our primary stakeholder, is a public body responsible for protecting and improving the environment in the UK. Alongside other responsibilities, a key priority of the EA is to manage the risk of flooding from main rivers, estuaries and the sea.
Their focus is on flood response up to 7 to 10 days in advance of a potential flood event. However, they also recognise that seasonal meteorological and operational factors (e.g. changing weather conditions and abstraction rates) are in play across the Thames basin. The detection of a potential flood signal weeks or months in advance could provide a head’s up and highlight ‘areas to watch’ as the seasons change.
The EA provides a local picture of the potential likelihood of flooding in England and issues flood alerts (floods possible, be prepared) and flood warnings (floods expected with possible risk to life, take immediate action) to the local community where needed.
They also deliver a wider scale strategic overview, which includes a daily flood guidance statement shared with the responder community, e.g. emergency services, police forces, county councils. If required, this strategic overview can go up to Government level (e.g. the COBRA emergency council) to facilitate high-level planning and to ensure timely movement of temporary flood barrier equipment in preparation for floods.
The EA would like to increase their knowledge on the seasonality of the sub-catchments hydrological responses and use this information to help identify areas of increased flood risk in advance of the current short to medium range forecasts.
Stuart Hyslop, Flood Modelling and Forecasting Team Leader, Environment Agency, UK:
IMPREX helped us improve our current understanding of the modelling processes, catchment responses and river level predictability in the Thames basin and provide positive changes to current operational forecasting techniques.
In December 2017, a focus group was held at the University of Reading in collaboration with the UK Environment Agency (EA; IMPREX stakeholders).
It included a decision-making activity designed to capture how different water sector users currently interpret and act on SHFs to inform decisions in the West Thames. Participants were shown progressively confident and locally tailored SHF out to four months to focus on a period of extreme stormy weather and flooding. To avoid participant bias through prior knowledge, they were not told which time period was covered by the SHFs.
Most West Thames local stakeholders present at the focus group reported using seasonal hydrological forecasts in their everyday job and believed that they could possibly forecast flood and drought risk. The stakeholders also indicated that the forecasts could be used to support decision-making and to increase preparedness. Still, many noted that the forecasts cannot yet be used in this way as the products that are available in the UK show too much uncertainty and have a coarse spatiotemporal resolution.
Read the full case study.
Longer range hydrological forecasts for stakeholders and society, to give an indication of possible flood events several months ahead.