Impact case study database

Beven and Lamb’s research on uncertainty since 2000 has improved flood risk management practice across the UK as evidenced by citations in 30 practitioner or policy reports [5.1], including government strategies for England, Scotland and Wales, evidence reports and policy updates. Of these reports, 12 were co-authored by Beven and/or Lamb. Beven’s research (as summarised in [3.1] and [3.2]) drove the development of Good Practice Guidelines within the EPSRC Flood Risk Management Consortium (FRMRC) and reached more than 200 stakeholders through the NERC-funded Catchment Change Network (from 2009 to 2012). Outputs from the FRMRC that cited [3.2] included guidance on assessing uncertainty in flood risk mapping, co-authored by Beven and Lamb, and published in 2014 by CIRIA, widely regarded as an authoritative provider of good practice guidance for the construction and built environment industries (CIRIA Guide C721, included in [5.1]).

Beven and Lamb’s research on environmental "models of everywhere" was adopted by their industry partner JBA, one of the UK’s leading flood risk management specialists, to improve national flood risk maps required by the government under the Flood and Water Management Act 2010. Paper [3.2] describes how JBA applied the models of everywhere concept by implementing a quality assurance portal for its flood models, which enabled local stakeholders to correct epistemic errors (by adjusting the JBA models) in the high-resolution (approximately 2m) national surface water flood risk maps they have produced for government agencies in the UK. A letter from the Environment Agency [5.2a], which in 2013 procured the maps from JBA in England for its Risk of Flooding from Surface Water (RoFSW) mapping, corroborates the importance of the local model corrections in delivering robust flood risk information at a higher level of detail than before. From 2017 onwards, Lamb and Beven’s interdisciplinary work with computer scientists, data scientists and technologists has further influenced the 2nd generation National Flood Risk Assessment (NaFRA2) in England. The NaFRA is used by the EA to target investment in flood mitigation measures, to support their funding submissions to the Treasury and to communicate flood risk to the public. Existing versions of NaFRA suffer from being fragmentary in their approach to reconciling risk information from flood models developed at different scales. A letter from the EA [5.2a] corroborates that Lamb and Beven’s research (reported in [3.5] and [3.6]) “has underpinned the design of our once in a generation national flood risk assessment replacement (NaFRA2)” was influential in the EA’s design for NaFRA2 by “testing of methods and technologies” to overcome this previously fragmentary approach. Although the research in [3.6] did not appear in a journal paper until 2020, [5.2a] confirms that it had already, since 2017, “directly influenced” the EA’s thinking whilst designing the NaFRA2 solution, which they describe as “one of our most significant modelling investments”, including an GBP8.0 million commercial contract [5.2b].

4.3 Improving the government’s understanding of extreme flood events

Lamb’s research to extend and apply novel statistical methods to model the spatial joint probability of extreme river flows and rainfall [3.4] improved the Government’s appreciation of flood resilience at a national scale following severe floods in 2013 to 2016, and changed the representation of risks in the National Security Risk Assessment (NSRA), which underpins central government emergency planning. It also changed the Government’s view of risk in the River Eden catchment, and it’s understanding of how open data can be used to better protect people from flooding.

After damaging floods occurred in many parts of the UK between 2013 and 2016, the Government established a National Flood Resilience Review (NFRR, [5.5]) “to assess how the country can be better protected from future flooding”. The NFRR was based on evidence from a Scientific Advisory Group (SAG), chaired by the Government’s Chief Scientific Adviser, Sir Mark Walport. Lamb was invited to join the SAG based on his research in [3.4] and related applications to the NSRA, which were on-going at the time and will be described further below. Drawing on the SAG’s evidence, the NFRR committed the Government to GBP12.5 million of spending on new temporary (i.e. mobile) flood defences, raising their number four-fold compared with 2015, as announced in Parliament and reported by the BBC [5.4a,b], whilst supporting the Government’s on-going capital spend of GBP2.3 billion to protect 300,000 homes [5.4a]. Lamb’s work in [3.4] was cited in the NFRR as “a method developed at Lancaster University and JBA” [5.5, p.12], showing that “while the probability of an extreme river flow that could result in a severe flood at any given location is very small, such flows are not unusual when considering the whole country”, hence motivating further immediate action to improve the flexibility of national flood defences. This improved understanding of flood risk was recognised by the London School of Economics in its response [5.7] to the NFRR, which stated that “… [the Government] deserves credit for admitting that Ministers have previously misunderstood and significantly under-estimated the probability of flooding.” [text removed for publication]

The Government plans for civil emergencies by maintaining a National Security Risk Assessment (NSRA), which is designed to compare and prioritise all major disruptive risks to national security over the subsequent five years. Prior to 2019 the NSRA was known as the National Risk Assessment (NRA), and this abbreviation will be used in what follows. The full risk assessment is classified, but there is a public-facing summary called the National Risk Register (NRR). The EA’s letter [5.2a] confirms that Lamb’s research at Lancaster, and subsequent applications by JBA, underpinned improvements in the 2017 NRA. The inland flooding scenarios that drove these improvements were developed in a project commissioned by the EA in which JBA applied the research in [3.4], which was extended and scaled to include all suitable river and rain gauges across the country (more than 1000 in number) and which “changed the representation of flood risks for the public facing National Risk Register (NRR) and the Cabinet Office’s National Risk Assessment ” [5.8]. Based on this extended application of [3.4], the Environment Agency announced [5.8] that “the National Risk Register now includes a new risk on widespread surface water flooding over a major city. The existing widespread river flooding scenario has been improved and provides a stretching, yet plausible planning assumption for national emergencies.”

Lamb’s work on modelling the joint probability of spatial extremes [3.4] was also combined with surface water flood risk models produced by JBA and incorporated local corrections for epistemic uncertainty discussed above in Section 4.2. JBA combined these models to win the top prize in a Defra innovation competition in recognition of their use of Defra’s open data in “innovative ways to tackle flooding focusing on the Eden catchment in Cumbria” through “the use of new data and whole-catchment modelling to generate a much wider range of realistic extreme event scenarios than is normally used ” [5.3a]. The winning report [5.3b], co-authored by Lamb and citing [3.4], further influenced Defra’s approach to flood risk management in Cumbria and beyond, with the Floods Minister Thérèse Coffey stating [5.3a] “We will now be able to use these new ideas and insight to help better protect not just Cumbria’s people, businesses and infrastructure from flooding but communities across the nation”.