Which rainfall dataset to cite in a drainage calculation

Every surface water calculation you submit rests on a rainfall dataset, and the reviewer's first question is which one. Name the model, the version, the grid reference and the date you extracted the data. In England the practical choice is between the Flood Estimation Handbook depth duration frequency models, the 1999 version and the 2013 revision, and the older Flood Studies Report rainfall used by some drainage software by default. Getting this wrong does not usually produce an obviously silly answer. It produces a plausible one that turns out to be undersized.

Why the rainfall source has to be on the page

A drainage calculation is a chain: rainfall depth for a given duration and return period, a storm profile, a runoff model, then storage sized to hold the difference between inflow and the permitted discharge rate. Change the first link and everything downstream moves.

Lead local flood authorities review hundreds of these a year. They are not recalculating your model. They are checking that the inputs are stated, current and defensible, and that the outputs follow. An unsourced rainfall figure means the reviewer cannot do their job, and the usual outcome is a holding objection rather than a refusal.

So state it. Dataset, version, grid reference, extraction date, and the return periods and durations tested. Four lines of text that remove an entire round of correspondence.

The national rainfall models and their revisions

The Flood Studies Report rainfall model dates from 1975 and still appears as a default in drainage software, often through the Wallingford Procedure routines. It is the oldest of the three and the hardest to defend on a new application.

The Flood Estimation Handbook depth duration frequency model published in 1999, usually shortened to FEH99, replaced it for most purposes. The 2013 revision, FEH13, rebuilt the model on a longer and denser record of rain gauge data. The difference between the two is not uniform across the country. In some locations FEH13 gives greater short duration depths than FEH99, in others less, and the direction of travel varies with return period as well as with place.

That variability is the reason a blanket rule fails. You cannot say the newer dataset is always more onerous. You can say that using a dataset from 1975 on a scheme designed in the 2020s invites a question you will struggle to answer.

Point rainfall data for a site is obtained by grid reference from the FEH web service, which is a licensed product. Keep the download, because a reviewer asking for the source wants the extract, not a screenshot of a results table.

Grid references and getting the location right

FEH point data is supplied on a one kilometre grid. For a small development site, the centroid of the site is the right reference. For a drainage catchment that extends beyond the red line, use the centroid of the contributing catchment instead, and say which you used.

Two common errors. The first is taking the grid reference from a postcode centroid, which can sit several hundred metres from the site and, in hilly country, in a different rainfall regime. The second is using the same extract for a phased scheme where later phases drain to a different outfall.

Neither error is large. Both are avoidable, and both look careless when the reviewer spots them.

Storm profiles and what they do to the answer

A rainfall depth is not a storm. The profile distributes that depth over the duration, and the choice of profile changes the shape of the inflow hydrograph.

Summer profiles are peakier and tend to be critical for systems limited by conveyance or by the rate of infiltration. Winter profiles are flatter, with antecedent conditions that generate more runoff volume, and they tend to be critical for storage dominated systems such as attenuation basins and tanks.

The practical answer is to test both and report the critical case for each element of the system. A scheme with a small orifice and a large basin can be sized by a winter event even though the peak flow arrives in a summer one. Presenting only the case that suits the design is the fastest way to lose a reviewer's confidence.

Run a range of durations too, from short intense events through to several hours, and show the critical duration for each return period rather than assuming the one hour event governs.

Areal reduction, and when it does not apply

Areal reduction factors account for the fact that heavy rainfall is not uniform over a large catchment. The factor reduces point rainfall to an average depth across the area.

For a typical development site the factor is one, because the site is small enough that the point value applies. Applying an areal reduction to a two hectare site is an error that reduces the rainfall and therefore the storage, and it is one of the few mistakes in this field that produces an undersized system while looking like diligence.

The factor becomes relevant on larger catchments, and the threshold above which it matters is a matter for the hydrologist rather than a rule of thumb. If your contributing catchment is measured in square kilometres rather than hectares, take advice.

When two datasets disagree

Sometimes FEH99 and FEH13 give materially different answers for the same site, and the design flips from comfortable to tight. Three responses work.

1. Check whether the lead local flood authority has published a preference. Several have, and if yours has, that settles it.
2. Present both, adopt the more onerous, and note the margin. This costs storage volume and buys an uncontested review.
3. Where adopting the more onerous case cannot be made to work on the site, set out the comparison, explain the choice and evidence the residual risk in the Flood Risk Assessment rather than burying it in an appendix of the drainage calculations.

Climate change allowances are a separate layer applied on top of the chosen rainfall dataset, using the Environment Agency's published peak rainfall intensity allowances. Those figures are revised from time to time, so check the current allowance at the point of design rather than copying one from a previous job.

An illustrative example. A 1.8 hectare brownfield site drains to a surface water sewer with a 5 litres per second discharge limit. Sized on the 1999 dataset the attenuation tank comes out at 310 cubic metres. Rerun on the 2013 dataset the same design needs about 8 per cent more volume, which the layout can absorb by deepening the tank rather than widening it. The consultant presents both, adopts the larger, and the drainage officer signs it off without comment. Figures here are illustrative rather than taken from a real scheme.

Writing it down so a reviewer can check it

Put the inputs in one table near the front of the Drainage Strategy, not scattered through appendices. It should carry:

• Rainfall model and version, with the extraction date and the grid reference used.
• Return periods and durations modelled, and the critical duration for each.
• Storm profiles applied, summer and winter, and which governed.
• Areal reduction factor, stated as one where no reduction applies.
• Climate change allowance applied, with the source and the epoch.
• Urban creep allowance, where the local standards require one.

The same discipline applies to the other evidence in a drainage submission. If you are citing the surface water map, read it without overstating what it shows, and where groundwater is part of the picture, set out the evidence for it properly.

Drainage calculations should be prepared by someone competent in hydraulic design, and councils expect a suitably qualified author. An AI-assisted draft can build the strategy structure, set out the policy framework and mark exactly where your rainfall inputs and discharge rates need to go, then flag the values it does not hold. The arithmetic remains yours.

Enter your postcode on PlanScope to see which reports your site is likely to need before you commission the drainage work.

This is general guidance, not legal advice; check the position with your lead local flood authority or a qualified professional before relying on it. PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval.