Groundwater flooding – a hidden hazard

Flooding has become an increasing part of the national consciousness affecting a large section of the population, and an issue at the intersection of high-profile topics including climate change and net zero, housing development and planning. Of the four main recognised types of flooding – fluvial (from rivers), pluvial (from direct rainfall), coastal and groundwater – relatively little attention has been given to date on the integration of all forms of flooding, including from groundwater.

Following extensive flooding in the UK in 2007, the Pitt Review (Pitt, 2008) highlighted the need for integrated flood management. This led to the Flood and Water Management Act 2010. The act established governance approaches, including Lead Local Flood Authorities (LLFAs) who are responsible for all forms of flooding except from rivers and sewers. While many of the recommendations of the Pitt Review were implemented in the years immediately following the report (Defra, 2012), including the establishment of the Flood Forecasting Centre, which provides national forecasts for all forms of flooding through Flood Guidance Statements for the responder community and warnings and alerts for the public, there remain significant issues still to be addressed, in particular related to groundwater flooding (see review by Wagstaff et al. (2021)).

Groundwater flooding in the UK in the winter of 2013/2014 (the wettest winter on record at that time) led to new impetus in studies and actions related to groundwater flood management. In the current Flood and Coastal Resilience Innovation Programme (EA and Defra, 2023), three out of the total of 25 funded projects are focused on groundwater. These form the Project Groundwater Network, which includes Project Groundwater Northumbria (PGN, 2024). These, with other related projects (e.g. BGS, 2024), are helping to provide direction towards better integrated flood management, including groundwater.

The numbers of people and properties directly affected by groundwater flooding are generally considered to be less than those affected by other forms of flooding. However, a review following the 2013/2014 winter flooding (McKenzie and Ward, 2015) estimated that between 122 000 and 290 000 properties in England could be at direct risk from groundwater flooding. In addition, groundwater may play a role in fluvial and coastal flooding for close to a million additional properties, and close to four million properties are in areas that could be affected by groundwater flooding or high groundwater levels (McKenzie and Ward, 2015); although the role of subsurface infrastructure was not addressed, it was noted to be ‘a significant issue’. Engineered flood management solutions to groundwater flooding problems include pumping and diversionary drainage, but are generally difficult to design and implement effectively.

The reasons why groundwater flooding has received less attention than other forms of flooding may include

• lack of visibility of the problem (this has also been highlighted in relation to global groundwater depletion (Famiglietti, 2014))

• the numbers of people or properties affected

• lack of general understanding of the issue

• paucity of availability of information (both spatially and lack of appropriate long time-series data)

• difficulty in designing and implementing solutions.

In contrast, flooding from rivers is clearly visible and generally understandable, long time-series of river flows and levels are often readily available, and solutions such as flood walls and storage are well understood and can be engineered to be effective to defined criteria.

Groundwater flooding can be described as the emergence of groundwater at the ground surface or where it interacts with infrastructure such as basements, outside of ‘normal’ conditions. In contrast to fluvial and pluvial flooding, which typically last for a few days or hours, respectively (often caused by rainfall from frontal systems or convective storms), groundwater flooding often lasts for weeks or months and tends to be seasonal due to the large storage capacity in aquifers, with highest levels in the spring.

The key indicators for areas susceptible to groundwater flooding are permeable geological strata and low lying topography (although groundwater can emerge at higher locations, particularly as springs). The most commonly recognised sources of extensive (clearwater) groundwater flooding are from Chalk aquifers, which have low storage coefficients so can respond rapidly to recharge, and River Gravels or permeable alluvium. Indirect, secondary, or groundwater-induced flooding can occur where there is intense rainfall onto land with high water tables or where significant groundwater discharges to surface watercourses cause subsequent out-of-bank flooding downstream.

In addition to these natural controls, the role of interactions between groundwater and subsurface infrastructure is now becoming better recognised and understood, particularly at more local scales in urban areas (Figure 1). Ingress into stormwater and combined sewer drainage networks in areas of high groundwater levels can act to control groundwater level rise, but also reduces the capacity for conveying runoff from intense storm events (as well as creating ‘dry weather flows’ into sewage treatment works, resulting in significant treatment costs). Flood management solutions such as infiltration SuDS (sustainable drainage systems) may locally raise groundwater levels, leading to potential downstream impacts. Built infrastructure such as basements, retaining walls and so on create barriers to groundwater flow, potentially redirecting groundwater to more vulnerable areas. Inflow into leaky basements can be pumped, providing further hidden discharges into drainage networks.

Recognising that many flood events include interacting components from surface and groundwater, the phrase ‘multi-source flooding’ may describe the reality of flood causes in many cases more appropriately than the separate classifications.

The complexities and scales of interactions in groundwater and multi-source flooding provide challenges for monitoring and modelling (Smith, 2020). Observation boreholes provide information at specific points, but are expensive to install and maintain so are sparsely distributed, and may not capture spatial flow behaviour appropriately (for example, in karstic aquifers or mine workings). Modelling approaches have traditionally been taken from either surface water or groundwater perspectives, with key interactions represented as boundary conditions, for example infiltration being treated as a ‘loss’ term in two-dimensional surface water models. Important localised features such as deep basements acting as barriers may be difficult to represent at an appropriate scale in groundwater models.

Many of the approaches to flood management follow the separation of hydrological cycle components into groundwater and surface water, which has become embedded in UK and international institutional structures (e.g. the different professional bodies for hydrology and groundwater/hydrogeology). There has been significant movement in recent years towards more catchment-based approaches, including natural flood management (e.g. the recent Landwise project (Landwise, 2024)), combined with hard engineered solutions. There is a need to continue along this path towards a more holistic hydrological and engineering approach, to include groundwater flood management.

In Project Groundwater Northumbria, a systems approach was used to consider the factors affecting potential groundwater (and related) flooding from mine workings in the north-east of England. A simplified diagram showing some of the physical natural and infrastructure components with indicative cause–effect relationships is shown in Figure 2 (see PGN (2024) for further details). As hard engineered (reactive) methods of flood protection are often not feasible for groundwater flooding, proactive approaches of active groundwater level management can be more practicable. In the north-east of England (and other legacy mining areas), pumping has been used over periods of centuries during active mining, and levels continue to be managed since mine closures to control surface discharges. Now, there is increasing interest in use of water from flooded mine workings for ground-source energy schemes, with some already in operation. However, these require re-injection of water back into the ground. The integrated operation of such energy schemes with groundwater level control could have the potential for multiple benefits of flood risk reduction while moving towards reductions in costs and carbon dioxide emissions.

Groundwater is abstracted from most aquifers for domestic water supply and for use in agriculture, industry or heat management, resulting in depressed groundwater levels compared with a naturalised baseline. If land becomes developed under these conditions, the risk of flooding from groundwater may increase. Groundwater abstractions are balanced against the need to maintain environmental flows (e.g. to rivers and wetlands (Defra, 2017; Environment Act 2021)). There is a fine balance between the need to maintain high groundwater levels in some places while managing groundwater levels to avoid direct or indirect flooding impacts, especially under the uncertainties of future climate change.

Some wider issues that remain to be addressed include

• general awareness and understanding of groundwater flooding (e.g. in reporting of flood events and the need for appropriate hydrogeological knowledge to be embedded in LLFAs)

• the division of organisational responsibilities for different sources of flooding (noting that flooding is often multi-sourced)

• appropriate groundwater level monitoring (there are also potential opportunities in the increase in monitoring of storm sewers and combined sewer overflows (CSOs) that could provide more widespread information on groundwater–infrastructure interactions).

Understanding the role of groundwater flooding highlights the inter-relationships between flooding from different sources and between the natural and managed environment, requiring high-level systems thinking for multi-source flooding, combined with local knowledge. Key challenges towards inclusion of groundwater into integrated flood management are as follows.

• Recognise the role of groundwater as source, pathway and receptor in integrated flood management and land use planning.

• Develop appropriate training and educational resources, including hydrogeological understanding and flood mitigation guidance.

• Address governance arrangements to improve clarity of reporting and management of flooding from multiple sources.

• Ensure monitoring is implemented or continues at appropriate sites, to provide openly available data for groundwater flood forecasting and management.

• Support further research into opportunities for multi-source flood management including groundwater management for different uses.

These require understanding and close working between hydrologists, hydrogeologists, engineers, planners and related professions.