On 3 December 2015, Storm Desmond formed, bringing record-breaking rainfall and flooding to the UK. The worst impacts were felt in Cumbria, although 19,000 homes were flooded across northern England as a whole and insured losses were estimated at £604m at the time of writing (PERILS, 2016).
With increased flooding expected across most of the country under future climate scenarios, it’s more important than ever to consider how we, as an industry, will respond to the next flood event of this scale in the UK.
One of the key challenges during and in the immediate aftermath of an event like Desmond is to identify the areas impacted and understand the extent of the damage. Despite recent advances in earth observation, there are still limitations in satellite coverage due to their revisit time and cloud cover can obscure the view for some satellite sensors (unfortunately a common occurrence when it has been raining). This means that data is intermittently available and can provide an inconsistent and incomplete view of flooding.
An alternative approach is to use forecast rainfall and river flows, in combination with real-time observations, to understand the potential for flooding and to model likely flood extents and depth before, during and after an event. This is the approach used by Flood Foresight® and, in this blog, we will take a look at some of the insights Flood Foresight would have provided for Storm Desmond.
There are three modules within Flood Foresight, the first of which is the Rainfall Screening Module. This module provides a forecast rainfall amount and associated return period for up to 6 days in advance. Values are updated daily and can be visualised either on a grid view or average by river catchment.
So, what did this module show in the days leading up to Storm Desmond?
Results from this module rely on the accuracy of weather forecasts which cannot be guaranteed, but the use of rainfall return period alongside forecast rainfall amount can be used to successfully identify areas likely to experience flooding.
Figure 1: Rainfall Screening forecast on 2 December for 5 December 2015 (3-day lead time) showing the forecast return period for each identified catchment.
The Flood Forecasting Module provides flood extent and depth maps for river (fluvial) flooding up to 10 days in advance. The maps are driven by streamflow forecasting models and are updated daily.
Did this module correctly identify flooding?
Figure 2: Flood Forecast flood extent and depth forecast for Kendal on 3 December for 5 December 2015 (2-day lead time).
The Flood Monitoring Module uses observed streamflow data from a network of river gauges, updated every three hours, to produce real-time flood extent and depth maps.
How did this module identify flooding?
Figure 3: Flood Monitoring extent and depth footprint for Kendal on 5 December 2015 18:00 UTC.
Overall, although rainfall amounts were challenging to forecast accurately due to uncertainty in weather forecasts and the record amounts of orographic rainfall to upland areas, Flood Foresight gave a comprehensive overview of Storm Desmond throughout all three modules. This type of early view into expected flooding has a range of benefits for insurance companies – from outreach to policy holders, giving them the opportunity to implement flood mitigation measures, to prioritisation of claims resources – all enabling better customer service and a minimisation of the costs.
As well as visualising hazard data through a web application, Flood Foresight data can be downloaded or integrated directly into risk management platforms to evaluate alongside exposure data and claims, estimating losses throughout the lifecycle of an event.
Producing flood footprints in the aftermath of an event has traditionally been a time-consuming process, dependent on manual steps and often limited observational data. Rainfall screening, flood forecasting using sophisticated streamflow modelling and near real-time monitoring provide a valuable new set of tools to understanding flood events as they evolve. And, importantly, access to these early insights into flooding and timely availability of reliable flood footprints can help insurers and reinsurers respond more effectively to the next large-scale event.
For more information on Flood Foresight® , get in touch or read our Flood Foresight Executive Briefing.
The validation of components and product throughout the development cycle of our flood maps gives us the confidence that we are creating the highest quality data. In this blog Cameron Whitwham takes us through some of the key steps.
Learn moreThe World Bank will have enhanced access to JBA Global Flood Maps following JBA's appointment to the Development Data Partnership.
Continue readingIn a country that has already experienced some challenging flood events, JBA's South Africa flood risk intelligence offers insight to help manage and mitigate risk.
Continue readingSpecialist insurer New National Assurance has partnered with JBA. The 50-year-old independent insurer will use the JBA flood data for South Africa, which includes climate change intelligence, to underwrite both commercial and personal lines property-risk.
Continue reading