Rainfall data sets and which one to design with

Design surface water drainage with the FEH13 rainfall model unless there is a documented reason not to, and state in the drainage strategy which data set you used, where it came from and the date you obtained it. The choice matters because the FEH13 depth duration frequency model generally produces higher rainfall depths for short, rare events than the older FSR and FEH99 approaches, so the same site can need noticeably more attenuation storage depending on which numbers went into the model. FEH rainfall data for drainage design is a licensing and provenance question as much as a hydrological one.

Reviewers now ask which data set was used. Reports that do not say are sent back.

Why the rainfall data set changes the answer

Attenuation volume is driven by the difference between the inflow hydrograph and the permitted discharge rate, integrated over the critical storm duration. Change the rainfall depth for a given return period and duration and you change the inflow, the critical duration and the storage.

The effect is not uniform. For long-duration, frequent events the data sets can agree closely. For short, intense, rare events, which is exactly the 1 in 100 year 15 or 30 minute storm a small urban site is sized on, the difference can be substantial and in one direction. A scheme sized on FSR depths and then checked against FEH13 often shows the storage is undersized, and that discovery after the layout is fixed means losing a plot or redesigning the sub-base.

So the data set is a design decision taken at the start, not a reporting detail added at the end.

The data sets in general use and their basis

Three generations of statistical rainfall model are still encountered.

The Flood Studies Report method, from 1975, is the oldest and is still embedded in some older software defaults and in legacy sewer models. It is a national statistical approach with a growth curve applied to a short-duration index. The Flood Estimation Handbook 1999 model replaced it with a pooled statistical approach giving depth duration frequency estimates by catchment descriptor. The FEH 2013 model, usually written FEH13, revised the depth duration frequency statistics using a longer and larger rainfall record, and it is the current model.

Alongside these sits the Revitalised Flood Hydrograph method for rural catchment flows, which is a different tool answering a different question, and the Wallingford Procedure modified rational method, which still underpins a lot of pipe sizing. None of those replaces the rainfall input.

The practical position for a planning submission in England is that lead local flood authorities expect FEH13 for surface water design, with FSR retained only where a scheme is being modelled consistently with an existing adopted sewer model that uses it, and that exception has to be explained.

Obtaining catchment rainfall data for a site

Rainfall depth duration frequency data for a specific point is a licensed product, not an open dataset, and the licensing catches people out.

The FEH web service provides catchment descriptors and depth duration frequency estimates for a point or catchment, under a licence, and the output is a file that drainage software can import. Many drainage modelling packages include an interface to it, and a single site licence is inexpensive relative to the design fee. Some consultancies hold a wider licence.

Two practical points. Download the data for the correct point, which for a small site means the site itself rather than a catchment centroid several hundred metres away, and keep the output file with the calculation. And record the date of download, because the underlying model can be revised and a reviewer checking a calculation three years later needs to know which version was used.

Do not take rainfall depths from a textbook table, a spreadsheet inherited from a previous job, or software defaults that nobody has checked. Software defaults are the most common source of the wrong data set appearing in a submitted calculation.

Depth duration frequency curves explained

A depth duration frequency relationship gives the rainfall depth expected for a given duration at a given annual exceedance probability, at a given location.

Read it in three steps. Pick the return period, or more properly the annual exceedance probability: 1 in 30 for pipe surcharge checks, 1 in 100 for flooding checks, with the climate change allowance applied on top. Pick the duration: not one duration but a range, because the critical duration is the one that produces the greatest storage requirement and it has to be found by testing. Read the depth, and convert it to a profile using a design storm profile, typically a summer and winter profile, because peak flow and volume behave differently.

The output that matters is a table of storage requirement against duration, with the critical duration identified. A calculation that presents a single duration has skipped the step that finds the answer.

Urban creep and its allowance

Urban creep is the gradual increase in impermeable area over a development's life: patios, extensions, driveways, sheds, decking. Residential schemes are expected to allow for it, and ignoring it is a standard review comment.

The allowance is applied as an uplift to the impermeable area for residential development, with the percentage varying by housing density, higher for low-density housing with large gardens and lower for flats with little private external space. The Defra non-statutory technical standards for sustainable drainage and the lead local flood authority's own guidance set what is expected locally, so check the local design guide before assuming a figure.

Commercial schemes are usually treated differently, because there is less scope for incremental hard surfacing, though a service yard that might be extended deserves consideration.

Apply urban creep to the impermeable area and the climate change allowance to the rainfall. They are separate adjustments and applying one twice, or neither, are both common errors.

What a reviewing authority checks

Lead local flood authority engineers work through a short list, and knowing it makes the submission faster.

Which rainfall data set, with evidence. The climate change allowance against the current Environment Agency allowances for the relevant river basin district and epoch. The urban creep allowance. The range of durations tested and the critical duration. The permitted discharge rate and its justification, usually greenfield runoff for the site or an agreed betterment on a brownfield rate. Whether the 1 in 100 year plus climate change event is contained or routed safely in exceedance. And the maintenance regime for every component.

Where the site drains to a watercourse, the discharge structure needs its own consent, covered in land drainage consent for works to an ordinary watercourse. Where the site is in a flood zone, the Flood Risk Assessment and the drainage design have to share one set of levels and assumptions.

Recording the data set in the strategy

Put a short provenance section near the front of the Drainage Strategy, before the calculations. It takes a paragraph and a table, and it prevents the single most common round of review comments.

State the rainfall model and version, the source and download date, the point or catchment reference used, the climate change allowance with the guidance version and publication date, the urban creep allowance with its source, the software and version used for the modelling, and the permitted discharge rate with its derivation. Attach the raw data output as an appendix.

Then keep that appendix. When a condition is discharged two years later by a different engineer, the provenance is what lets them continue the design rather than start again, and the same reasoning applies to the surface water evidence described in reading the surface water flood risk map without overstating it.

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