Flood Risk Assessments for Electrical Substations

A substation flood risk assessment establishes whether a proposed electrical substation will be reached by flooding during its working life, and sets the levels and the protection the equipment needs to survive it. It can be required for planning. It can equally be required by the distribution network operator as part of the connection or adoption process, on a site where planning permission was granted years ago or is not needed at all. The two requests look alike, and they are answered against different standards.

Research published in April 2026 by Rebalance Earth with Dr Raghav Pant of the University of Oxford found that 20 centimetres of floodwater is enough to disable a secondary substation, exposing more than 27,000 businesses to power loss and costing roughly £90 million a day in national economic losses during severe weather.

A substation is a small building with a very low tolerance for water. Water reaching the switchgear or the transformer damages plant, interrupts supply and creates a safety risk, and the consequences are rarely confined to the site that paid for it. Network operators therefore assess substation flood risk to a standard of their own, separate from anything the planning system asks for, which is why a flood risk assessment for a proposed substation is worth commissioning while the layout can still move.

When does a substation need a flood risk assessment?

A substation needs a flood risk assessment when planning policy triggers one, when the network operator asks for one, or both. The planning trigger follows the ordinary rules that apply to any development. The operator's trigger is separate, is written into its own code of practice, and does not depend on the flood zones at all.

• The planning trigger. Policy F4 of the August 2026 National Planning Policy Framework requires a site-specific assessment for all development in Flood Zones 2, 3a and 3b, and in Flood Zone 1 where the site is a hectare or more, sits in a critical drainage area, or is shown at risk from any source. Our guide to when you need a flood risk assessment sets out the full position.
• The network operator's trigger. A DNO or IDNO may want an assessment before it accepts the design, energises the connection or adopts the asset. Northern Powergrid's code of practice asks for one prepared by specialist consultants for every new major substation and for new or replacement distribution substations.
• Existing assets count too. Electricity North West's policy calls for a risk assessment when distribution substation assets are replaced, when a major project is proposed at an existing site already at medium or high risk, and when a new substation is proposed within a flood zone.
• Neither trigger substitutes for the other. Planning permission does not satisfy the operator, and the operator's acceptance does not discharge a planning condition.

Grid, primary or distribution: why the type of substation matters

The three tiers of the electricity network are held to different flood standards, so the first question any substation assessment has to answer is which tier the asset belongs to. A grid or 132kV site is designed against a 1-in-1,000-year event. A primary substation below 10,000 customers is usually designed against a 1-in-100-year fluvial event and a 1-in-200-year coastal one. A distribution substation falls outside the main industry standard altogether.

That last point is the one that catches people out. Engineering Technical Report 138, the Energy Networks Association standard that most substation flood work is written against, covers grid and primary substations. Distribution substations are largely excluded from its scope, on the reasoning that they serve the area that is flooding anyway and can be restored once the water goes down. The gap is filled by each operator's own policy, and those policies differ; SP Energy Networks sets out its flooding resilience approach separately again.

So the small package substation going in beside a warehouse, a leisure centre or a solar farm is the one most likely to be assessed against the wrong criteria, or against none. It is also the one most likely to arrive late in a programme, once the layout is fixed and the levels are expensive to change.

Is a substation essential infrastructure under the NPPF?

Under the August 2026 Framework, yes. Annex F, Table 2 of the National Planning Policy Framework places electricity generation, storage and network infrastructure in the essential infrastructure class, and network infrastructure reaches substations at every level. That wording is broader than what it replaced, and the change matters for the smaller assets.

The old Annex 3 named grid and primary substations specifically. A distribution substation serving one commercial development was not named, and assessments frequently classified it as less vulnerable along with the general industry it served. The 2026 wording removes most of the room for that argument, and it is one of several changes worth reading in full in our summary of what the August 2026 NPPF changed.

Two consequences follow. Essential infrastructure needs the Exception Test in Flood Zone 3a and in the functional floodplain, where a warehouse would not, and the substation has to be designed to stay operational and safe in a flood rather than merely to survive one. Where a substation sits inside a larger scheme, the highest vulnerability class applies to the whole application unless the assessment argues the components separately. Our explainer on the sequential and exception tests covers how that argument is built, and we prepare sequential and exception test reports where a scheme needs one.

Which sources of flooding should a substation assessment cover?

Every source that can reach the site, which is a wider list than the mapped river and sea zones. A substation flood risk assessment should work through fluvial, tidal and coastal, surface water, groundwater, sewer and other artificial sources, and reservoir or infrastructure failure where the site is exposed to it. Each source reaches a substation differently, and the mitigation that answers one will not always answer another.

• Rivers and the sea. Taken from the Flood Map for Planning, refined by any detailed modelling available for the reach, and read against the site's own levels in metres above Ordnance Datum rather than the map extent alone.
• Surface water. The source that most often reaches a substation on a site the flood zones show as clear. Intense rainfall, a drainage system at capacity or an overland flow route crossing the compound will all do it, as our guide to surface water flooding and planning explains.
• Groundwater. High groundwater enters cable trenches and below-ground ducts, affects foundations and reduces the performance of any infiltration drainage. Where the available records suggest a risk, monitoring to establish a design groundwater level is the sensible next step, and we prepare groundwater flood risk assessments where one is needed.
• Sewers. Surcharge backing up through a connected drainage system, which is why non-return valves appear on almost every operator's mitigation list. Our explainer on sewer flooding covers how the risk is established.
• Reservoirs and other artificial sources. Canal breaches, burst mains and culvert failures are localised and are not mapped nationally, so reservoir failure and infrastructure flooding both need a site-specific review rather than a map check.

National Grid's substation flood defence programme protected 49 critical substations against river and coastal flooding by 2021, then identified roughly a further 100 sites needing protection from surface water alone.

Surface water deserves the extra attention. It is the source least likely to appear on the map the applicant has looked at, and the one most likely to be introduced by the development itself, because a substation compound adds sealed footprint and kerbs that change where water goes. Where a compound is bunded, the bund is a flood wall in one direction and an obstruction in the other. The Environment Agency's check for flooding service is a reasonable first look, though it carries no weight in a planning submission.

How is the design flood level set, and where does 600mm come from?

The design flood level is the predicted level of the design flood event at the substation, with a climate change allowance applied and a freeboard added on top. None of those three figures is fixed nationally. A 600mm freeboard is quoted often enough to be treated as a standard, and it is not one.

Read across the table and the pattern is consistent on the 300mm uncertainty allowance and inconsistent on everything else. Where 600mm appears, it is doing one of two different jobs: standing in for the climate change allowance under BS 8533:2017, or sitting on top of the 300mm as a separate safety factor. Treating it as a single universal freeboard produces the wrong level either way.

There is a second mismatch worth planning around. Most operator policies apply a flat 20% uplift for climate change, while a planning assessment uses the Environment Agency's climate change allowances for the catchment, epoch and percentile that apply. Those two routes can produce different design levels on the same site, and the assessment should say which it has used and why. Our guide to choosing the right climate change allowance works through the four allowances and when each applies.

Establishing the level properly usually needs surveyed ground levels rather than published terrain data, and on a constrained site it needs detailed hydraulic modelling rather than the published extents. A difference of 200mm decides whether a plinth works.

What flood mitigation do network operators accept?

The preferred answer is always to put the substation where the water does not reach. Where the connection design will not allow that, the measures operators accept run in a rough order of preference, from permanent and passive at the top to interventions that depend on somebody arriving in time at the bottom. The annex to ETR 138 sets out the measure types in more detail.

1. Site selection. Move the compound to the highest practical part of the site, outside the modelled extent if the cable route allows it. Every published policy puts this first.
2. Raising. Lift the plinth, the foundations and the vulnerable plant above the design flood level. In practice this ranges from 150mm above existing ground on a low-risk site to well over a metre where the compound sits below the modelled level.
3. Relocation. For distribution substations in particular, operators would rather move the asset than defend it. Northern Powergrid says so explicitly.
4. Permanent protection. Reinforced concrete flood walls or earth bunds around the compound or around individual structures, with flood gates and planks left in place.
5. Demountable and temporary defences. Barriers that need deploying before an event, which depend on a flood warning, a lead time and access.
6. Resistance detailing. Sealed duct entries rated for water pressure, flood doors and gates, sealed low-level vents, waterproof cable enclosures, non-return valves on storm and foul connections, and secured manhole covers.
7. Monitoring and recovery. Remote monitoring, gate-open alarms connected to control, and a plan for restoring supply once the water recedes.

Northern Powergrid's code of practice states that temporary barriers are not an acceptable permanent solution, which rules out a common assumption before the design starts.

Pumps sit awkwardly in this list. They can form part of a wider strategy, but their operation depends on power, which is the thing at risk, and they can fail mechanically or be overwhelmed by an event beyond the drainage system's capacity. Operators generally expect passive protection alongside pumping rather than instead of it.

Two details from the design side are worth carrying into the layout early. Palisade fencing, or anything similarly open, lets water pass through the compound rather than damming against it. And whatever is built has to avoid displacing floodwater onto neighbouring land, which on a floodplain site means accounting for the volume the plinth and the bund occupy. An exceedance route that runs through the compound, the cable trenches or the feeder pillars is a design fault, not a residual risk.

When the assessment is for the network operator, not the planning authority

A substation flood risk assessment is not always a planning document. It can be requested directly by the DNO, IDNO, electrical supplier or adopting authority before the design will be accepted, which means it can be needed on a development that already holds planning permission, on a substation covered by an existing consent, or on a scheme where no application is required at all.

That changes what the report has to do. A planning assessment answers Policy F7's test, that the development is safe for its lifetime and does not increase flood risk elsewhere. An operator's assessment answers a network reliability question, at the resilience level that applies to the site. The industry has been working to that second question since the 2007 floods, which the Energy Networks Association's own evidence to Parliament identifies as the event that produced ETR 138. A report written for one reader and submitted to the other tends to come back.

Written for an operator, the assessment should identify the flood risks at the site, state the design flood level and the standard it was derived from, give the proposed substation level and the freeboard achieved, and set out the resilience measures the design relies on. Naming the standard is the part most often left out, and it is the part the reviewer looks for first.

The failure mode when the two assessments never meet is predictable: permission is granted on a layout that treats the compound as ordinary hardstanding, the operator's requirement for a raised plinth arrives afterwards, and the redesign lands once the connection agreement is signed.

The same sequence plays out on charging schemes, which we cover in our guide to flood risk and drainage for EV charging infrastructure.

What a substation flood risk assessment should contain

Beyond the standard contents of any flood risk assessment, a substation report needs a handful of things a generic template will not carry, and they are the things both readers check first.

• The vulnerability position, stated. Which Annex F class the substation falls in, whether the scheme is being assessed as a whole or in component parts, and where that leaves the Exception Test.
• A named standard. ETR 138, the operator's own code of practice, or both, with the design event, the climate change allowance and the freeboard each traced to their source.
• One reconciled levels drawing. Existing ground level, design flood level, freeboard, and the proposed finished floor level of the plinth and the plant, shown together so the planning officer and the operator read the same section.
• An exceedance route that misses the electrical assets. Where water goes when the system is overwhelmed, and confirmation that the route avoids the compound, the ducts and the trenches.
• A position on residual risk and continuity. Whether the substation is expected to keep operating through a flood or only to be safe and recoverable. These are different design briefs and they are worth separating in writing.
• The offsite position. No loss of floodplain storage, no diversion of flows onto neighbouring land, and a surface water drainage strategy covering the new impermeable area the compound creates.

Get those settled at concept stage and the rest is ordinary work. Leave them to a planning condition or to the connection process and the levels argument arrives at the most expensive point in the programme. Unda prepares flood risk assessments for proposed electrical substations across England and Wales, for planning submissions and for technical submissions to DNOs, IDNOs, electrical suppliers and adopting authorities.

Common questions about substation flood risk