Most flood risk thinking starts at the riverbank. Public flood maps, underwriting rules and planning designations were largely built around fluvial flooding: water overtopping a river channel.

Yet a large share of flood losses comes from elsewhere. A flash flood can develop within hours in a valley with no mapped floodplain. Intense rain can overwhelm a city's drainage miles from any watercourse. Storm surge can push the sea inland, and a rising water table can flood basements for months.

In the United States, FEMA reports that more than 20% of flood insurance claims come from properties outside high-risk flood zones. That gap between where flood risk is mapped and where losses occur is where underwriters, lenders and asset owners carry exposure they may not have priced.

This article explains how each of the four main non-river flood types behaves, how flood alerts work across the European Union, the United States and Canada, and what these differences change for risk decisions.

Flash floods: fast, local and hard to map

The US National Weather Service defines a flash flood as a flood caused by heavy or excessive rainfall in a short period of time, generally less than six hours. The defining feature is speed. Water levels can rise from normal to dangerous within minutes to hours of the rain starting.

Several conditions make flash floods more likely:

• Short, intense rainfall, often from convective storms that stall over one area. Changes in precipitation variability shift how often these bursts occur.
• Steep terrain and narrow valleys, which channel runoff quickly into a small area.
• Sealed urban surfaces, where concrete and asphalt send rain straight into drains and streets.
• Burn scars left by wildfires, where scorched soil absorbs little water and debris flows can follow.
• Dam or levee failure, which releases stored water suddenly.

For European readers, the Mediterranean adds a specific case: ephemeral watercourses that are dry for most of the year and can carry torrents after autumn storms. Their channels often look harmless on a site visit and may not register as rivers on public maps.

Flash floods matter for risk because warning times are short, water moves fast and carries debris, and the damage can occur far from any mapped floodplain. Linear infrastructure is especially exposed. Roads, railways, pipelines and transmission corridors cross dozens of small catchments, and a culvert sized for historical rainfall can become the weak point. Low-lying sites at the bottom of narrow valleys face the same issue at a single location.

Surface water flooding: when drainage cannot keep up

Surface water flooding happens when rainfall overwhelms drains, sewers and the ground's ability to absorb it, so water ponds or flows over land before it ever reaches a river. The technical term is pluvial flooding.

Readers often confuse it with flash flooding, because the two overlap. The distinction is simple: 'flash' describes how fast a flood develops, while 'surface water' describes where the water comes from. A cloudburst over a city can cause a flash flood that is also surface water flooding. A slow build-up of water in a low-lying car park after a day of steady rain is surface water flooding, but it is not a flash flood.

Three characteristics make surface water flooding important for risk:

• It occurs away from rivers. A site can sit well outside any river flood zone and still flood from the street.
• It concentrates in dense urban areas, where sealed surfaces are extensive and drainage networks were sized for past rainfall.
• It is the type least captured by river-based flood zones. Some countries now publish separate surface water maps, but they often sit outside the zones used for planning, lending and insurance decisions.

For commercial real estate and urban portfolios, this is often the most relevant flood type. Exposure can change from one street to the next, depending on ground levels, kerb heights and the capacity of the local drainage network.

Coastal flooding: surge, tides and rising seas

Coastal flooding occurs when the sea reaches land that is normally dry. The main mechanisms are:

• Storm surge, where strong winds and low pressure, for example during a hurricane or a deep winter storm, push water towards the coast.
• High tides, which set the level a surge starts from. The same storm causes very different outcomes at low and high tide.
• Waves, which add height and force on top of the still-water level and can overtop defences.
• Sea level rise, which raises the baseline every event starts from. According to the IPCC's Sixth Assessment Report, global mean sea level rose by about 0.20 m between 1901 and 2018, and it will continue to rise over this century.

The less visible issue is compound flooding. When a storm surge arrives at the same time as heavy rainfall or high river flow, rivers and drains cannot discharge into the sea. Water backs up inland, and the combined event can exceed what either hazard would cause on its own. This is one of the clearest reasons why single-hazard views tend to understate flood exposure in coastal areas.

The assets most exposed include ports, energy terminals and coastal power plants, coastal real estate, and water and wastewater treatment works, which are often sited at low elevations near the shore by design.

Groundwater flooding: slow, long and often overlooked

Groundwater flooding occurs when the water table rises until it reaches the ground surface or the parts of a structure built below it. It usually follows a prolonged wet period, such as a wet winter, and is most common in permeable geology like chalk, limestone and sand and gravel deposits near rivers.

What sets it apart is its timing. Onset is slow, often over days or weeks. Once it arrives, it can persist for weeks or even months, because the water has nowhere to drain to until the water table falls. The damage profile is different too: less sudden structural damage, more long outages, water ingress and repeated pumping costs.

The assets most exposed are those that sit partly underground:

• Basements and underground car parks
• Buried cables and below-ground utilities
• Substations and pumping stations with below-ground equipment
• Foundations and structures where uplift pressure matters

Groundwater flooding is generally the flood type least covered by public flood mapping. Where it is mapped, it is often shown as susceptibility rather than as a probability, which makes it harder to bring into the same risk framework as other flood types.

Flash flood and coastal flood advisories: what alerts tell you, and what they don't

Flood alerts are designed to protect people in the hours and days before an event. Each region organises them differently.

United States

The National Weather Service uses a tiered system:

‍

'Flash flood advisory' is a common search term, but it is not an official NWS product. Lower-level rainfall flooding is covered by a general Flood Advisory, which the NWS consolidated from several older advisory types.

On the coast, the NWS issues Coastal Flood Watches, Warnings and Advisories, where an advisory signals minor coastal flooding. Coastal flood advisories are still in use as of October 2026. Under its Hazard Simplification initiative, the NWS has proposed replacing 'Advisory' headlines with plain-language headlines over the coming years.

Canadá

Environment and Climate Change Canada issues weather alerts, including rainfall warnings and storm surge warnings. Flood forecasting and flood warnings, however, are largely the responsibility of provincial and territorial agencies, so terminology and thresholds vary across the country.

European Union

National meteorological and hydrological services issue weather and flood warnings in each member state. Meteoalarm brings these together on one platform, using a common four-colour scale from green to red. Alongside this, the European Flood Awareness System (EFAS), part of the Copernicus Emergency Management Service, provides flood forecasts and early notifications to national authorities.

The horizon problem

Every one of these systems works on a horizon of hours to days. They are built to protect people tonight. They are not designed to tell an underwriter, a lender or a developer how often a site is likely to flood over the next 10 to 30 years. That is a different question, and it needs a different kind of data.

‍

What a complete flood exposure picture requires

If alerts cover the next few days and river flood zones cover one source, a long-term risk view has to fill the space in between. These are the criteria worth checking any flood analysis against:

1. Every relevant flood type. Flash, surface water, coastal and groundwater flooding assessed alongside river flooding, not river flooding alone.
2. Asset-level results. Surface water and flash flood exposure can change within a few hundred metres, so area-level indications are not enough.
3. Forward-looking projections under several scenarios. Heavy rainfall and sea levels are not stationary, so historical records alone will understate future exposure.
4. Return periods with uncertainty ranges, translated into financial terms. Damage and climate value at risk, so exposure can be priced and compared.
5. One methodology across the EU, the US and Canada. A cross-border portfolio can only be compared like for like if every site is assessed the same way.

EarthScan, Mitiga Solutions' climate risk intelligence platform, is one example of how these requirements can be met. It combines CMIP6 global climate models and CORDEX regional models with more than 40 years of observational data, and returns asset-level results with projections in five-year increments to 2100, across more than 180 countries. You can see how this applies to flood in our flood risk assessment approach.

What this changes for risk decisions

A risk view that looks beyond the river changes decisions differently depending on who is making them.

Underwriting and the protection gap

Losses from non-river flooding often fall outside mapped high-risk zones, and therefore outside the areas where flood cover is required or expected. FEMA's figure of more than 20% of claims coming from outside high-risk zones points to where the protection gap sits. Pricing these risks needs data that covers every flood type, at the level of the insured property.

Energy and infrastructure assets

Linear assets cross many small catchments and are exposed to flash flooding along their whole length. Substations and buried cables add groundwater and surface water exposure. Siting and design decisions lock this exposure in for decades, so it is far cheaper to address at the screening stage.

Commercial real estate

Urban portfolios carry surface water exposure and, on the coast, compound flood exposure that a river flood zone will not flag. Both belong in acquisition due diligence and in the assumptions behind the hold period and exit value.

Physical risk disclosure

CSRD, through ESRS E1, and IFRS S2 expect companies to assess physical risks across relevant hazards and scenarios. The EU Taxonomy's classification of climate-related hazards lists coastal, fluvial, pluvial and groundwater flooding separately. A river-only view does not meet that bar.

River flooding is the most mapped flood type, but it is only one of five. Flash floods strike fast and far from floodplains, surface water floods follow the street rather than the river, coastal floods compound with rain and rising seas, and groundwater floods last for months. Alerts protect people in the short term; they say little about how often a site will flood over the life of a loan, a policy or an asset.

A complete flood exposure picture covers every relevant type, at asset level, across scenarios and time horizons. That is what turns flood risk from a gap in the map into a number that can be priced.