Reading a detailed Environment Agency flood data set

The flood zones on the planning map are broad-scale and indicative. When you need to set a finished floor level, argue a zone boundary, or show that a scheme is safe for its lifetime, you need the modelled data behind them, and that comes from the Environment Agency on request. The detailed flood risk data package, commonly referred to as Product 4, contains modelled levels by return period, model extents, information on defences and, where they exist, climate change scenarios. It is supplied to support flood risk assessments and it is the difference between a flood risk assessment that survives consultation and one that does not.

Request it early. The lead time is the single most common cause of a flood risk assessment being written on inadequate data.

What a detailed data request provides

The contents vary by location, because what exists depends on what has been modelled, but a typical package includes several of the following.

• Modelled flood levels at nodes along the watercourse, for a range of annual exceedance probabilities.
• Model extents and depth or hazard mapping where produced.
• Defended and undefended scenarios where the reach is defended.
• Climate change scenarios where the model has been run with an allowance applied.
• Historic flood records for the area, including dates and sources.
• Details of flood defences, their standard of protection and their condition where recorded.
• Flood warning area information.

What it will not always provide is a level at your site. Model nodes sit where the modeller put them, which may be several hundred metres upstream or downstream, and interpolating between them is the assessor's job, not the Agency's.

Requesting it, and the lead time

Requests go to the Environment Agency, usually through its enquiries service, and are supplied free where the data is needed to support a planning application or a flood risk assessment.

Send the site boundary as a drawing or a coordinate file rather than a postcode, and say what the proposal is and which watercourse you are interested in. A request that describes the development gets a package scoped to it. A request that gives an address gets whatever the default extract produces.

Lead times are commonly quoted at around twenty working days and vary by area and by workload. Treat that as a minimum and ask for a date when you submit the request. On a programme with a fixed committee date, the data request belongs in the first week, alongside the topographic survey, not after the drawings are finished.

While you wait, order the other site information. Sewer records, for instance, take their own time, as our note on pre-development enquiries explains, and the surface water picture usually needs the lead local flood authority as well.

Reading modelled levels by return period

Levels are given as metres above ordnance datum, by return period or annual exceedance probability. The usual set for a fluvial reach covers frequent events through to the 1 in 1,000 year event, with the 1 in 100 year event doing most of the work in planning terms. Tidal reaches are modelled on a different set, with the 1 in 200 year event as the reference.

Three checks before you use any figure.

Confirm the datum. Levels in the package and levels on your topographic survey must be on the same datum, and mixing them produces errors of a metre or more.

Confirm the node location relative to your site, and state the interpolation method you used in the assessment.

Confirm the model date and version. An older model superseded by a newer study is still in the archive, and using it will draw a consultation response.

Defended and undefended scenarios

Where the reach has formal defences, the package may include both scenarios. They answer different questions and both usually belong in the assessment.

The undefended run tells you the level that would occur if the defence were not there, which is what the flood zones are based on and what a breach or overtopping scenario has to consider. The defended run tells you what happens in the modelled event with the defence performing as designed.

Residual risk is the space between them. A site protected to a high standard still carries residual risk from breach, overtopping, blockage and defence failure, and a consultee will ask how the scheme responds to it: floor levels, safe refuge, escape routes, flood resilient construction and a flood warning and evacuation plan.

Where a breach scenario has not been modelled, say so rather than implying that none exists.

Climate change scenarios

Climate change allowances are published by the Agency and vary by river basin district, by allowance category and by the epoch relevant to the development's lifetime. The allowance applied depends on the vulnerability classification of the development and the flood zone, and the published guidance sets out which applies.

If the package includes a climate change run, check which allowance and which epoch it used. If it does not, the assessor applies the allowance to the modelled level using the published method, and shows the working.

These figures are updated periodically, so cite the version you used and its date. A flood risk assessment that quotes an allowance without saying which edition it came from is asking for a comment.

Setting finished floor levels

The usual approach is to take the design flood level, which is normally the 1 in 100 year level with the appropriate climate change allowance for fluvial sites, and add a freeboard.

Freeboard requirements are set locally. Three hundred millimetres is widely applied, but authorities and the Agency ask for more where the model is uncertain, where the watercourse is small, or where the consequences of exceedance are severe. Ask what the local expectation is rather than adopting a figure from another scheme.

Then check that the resulting levels work. Floor levels driven up by flood risk affect access thresholds, level access requirements, ramp gradients and the relationship to the street. That coordination belongs at concept stage, and it usually has to be resolved alongside the drainage strategy, because raising levels changes where surface water goes.

Limits of the model

Detailed data is only as good as the study behind it, and several limits recur.

Small watercourses, ditches and culverted streams are frequently unmodelled, because ordinary watercourses are the lead local flood authority's responsibility rather than the Agency's. A site on a minor tributary may have no detailed data at all, and a bespoke hydraulic model becomes the only route.

Surface water is not covered by fluvial models. It is a separate source with separate data and a separate consultee.

Models age. Channel geometry changes, defences are built, and catchment development alters runoff. A study from a decade ago may still be the best available and should be described as such.

And a modelled level is a modelled level. Where the stakes are high, the Agency may accept or require a site-specific model, which takes months and a hydraulic modeller.

Where the sequential test also applies, the data supports it but does not answer it, as our note on the flood sequential test sets out.

For specialist technical reports, qualified professional input may be required. PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval.