Resurveying levels and why a topographic survey matters

A flood risk assessment built on mapped ground levels is a draft. The national terrain models behind the published flood maps are made for catchment-scale work, and their stated vertical accuracy is wider than the freeboard you are trying to prove. So a topographic survey flood risk assessment pairing is not a luxury item on the fee proposal. It is the difference between a finished floor level you can defend and one an officer can take apart in a paragraph.

The commonest version of the failure is undramatic. The site turns out to be 350 millimetres higher than the model assumed, the scheme was designed to the wrong level, and nobody notices until the Environment Agency responds.

Why mapped levels are not good enough

Published flood zone mapping is generated from terrain models derived from airborne survey, and the metadata for those datasets states a vertical accuracy that is appropriate for identifying where flooding is likely across a catchment. It is not appropriate for setting a threshold level on one plot.

Three sources of error stack up. The terrain model has its own tolerance. Vegetation, walls and vehicles present at the time of capture can distort returns. And the model represents bare ground, so it knows nothing about the internal floor level of the building you are extending, which is the level that actually matters for a flood risk assessment.

Ordnance Survey mapping compounds the problem, because standard large-scale mapping carries no levels at all. Spot heights on older mapping are sparse and were never intended to support drainage design.

So the mapped data tells you whether to worry. A survey tells you what to do.

Datum, accuracy and what to specify

Everything in a flood risk assessment is quoted in metres above Ordnance Datum, so the survey has to be in the same datum or the numbers cannot be compared.

Specify Ordnance Datum Newlyn, established by satellite observation with the current national transformation applied, and ask for the method and the control points to be stated on the drawing. Ordnance Survey no longer maintains the benchmark network, so a level taken from a benchmark found on a wall needs checking rather than trusting.

The error that recurs is the local site datum. A survey issued on an assumed datum, with the lowest point set at 100.000 metres, is useful for cut and fill and useless for flood design. If a drawing shows levels in the nineties or the hundreds on a site that is 12 metres above sea level, the datum is assumed.

Specify the accuracy too. A tolerance in the region of plus or minus 25 millimetres on hard detail levels and a stated tolerance on soft ground is a reasonable ask for a development site, and the surveyor should state what was achieved. Published professional guidance on measured surveys sets out how to write the specification, and it is worth following rather than improvising.

Ask for the deliverables explicitly: a levelled plan at a stated scale, a digital terrain model, the point cloud or raw data if modelling will follow, and the control schedule.

Features the surveyor must pick up

A general topographic survey will not capture what a flood risk assessment needs unless the brief says so. Put these on the instruction.

• Internal finished floor levels and threshold levels of every building on and immediately adjoining the site
• Top of bank, toe and bed levels of any watercourse within or adjacent to the site, at several cross-sections
• Culvert inlet and outlet levels with barrel dimensions and condition
• Crest levels of any flood defence, embankment, wall or raised road acting as a defence
• Low spots in boundary walls, fences and gateways where water would enter or leave
• Manhole cover and invert levels, pipe sizes, materials and directions of flow
• Existing ground levels on a grid, tightened around the proposed building footprint
• Kerb, channel and road levels at the access, and the level of the nearest highway gully
• Tree positions with trunk diameters where root protection areas will constrain levels
• Any existing flood resilience measures, air brick positions and service entry points

The watercourse cross-sections and the culvert details are the two items most often left off, and they are the two that decide whether a hydraulic assessment can be done at all.

Tying levels to the flood model

Detailed model outputs from the Environment Agency give flood levels in metres above Ordnance Datum for the design event, usually with climate change allowances applied, at defined nodes or cross-sections.

Those nodes are rarely on your site. Someone has to interpolate between them along the watercourse, and the interpolation has to be stated and justified rather than performed silently. Where the site sits between two cross-sections 300 metres apart on a steep reach, the difference matters.

Then the finished floor level is set at the design flood level plus freeboard. Freeboard requirements vary by authority and by the confidence in the modelling, so take the figure from the council's own requirement or from the Agency's response rather than applying a default. Record the arithmetic on the drawing: design level, climate change allowance, freeboard, resulting finished floor level, existing ground level, and the step up.

Access and egress levels need the same treatment. A dry threshold reached by a 600 millimetre deep access route is not safe access, and that calculation also uses survey levels.

Where a basement is involved, the level work becomes the whole case, and the particular issues are set out in basements, flood risk and sleeping accommodation.

Where survey and model disagree

This happens often, and it is usually good news for the applicant.

If the survey shows ground levels materially higher than the terrain model used in the mapping, the mapped flood extent across the site may be wrong. The route is to request the detailed model data from the Environment Agency, have a hydraulic modeller review the local ground representation, and present the comparison. In some cases the Agency will accept that the site, or part of it, is not at the mapped risk.

Be careful with the opposite case. A survey showing the site lower than modelled means the mapped extent understates the risk, and that has to be disclosed. A flood risk assessment that quietly uses the survey where it helps and the model where it helps is the kind of document that gets a scheme refused and an agent's reputation damaged.

Three outcomes are possible and all are legitimate: the model is accepted, the model is reviewed and amended, or the site is shown to sit outside the extent. What is not legitimate is asserting a difference without the survey behind it. Surface water mapping has its own separate limitations, covered in reading the surface water flood risk map without overstating it, and groundwater is different again, as set out in evidencing groundwater flooding in a flood risk assessment.

Cost against the risk of getting it wrong

A topographic survey of a small development site is a fixed, quotable cost, obtained in a week, and it is the cheapest piece of certainty in the whole evidence base. Get three quotes against the same written specification so you are comparing like with like.

Set against that, the costs of not surveying are open-ended: redesign after consultee comments, a deferred committee date, an extension of time, an abortive drainage design, and in the worst case a scheme built to a level that fails a mortgage valuation or an insurance enquiry.

There is one honest uncertainty here. On a small infill plot in Flood Zone 1 with no watercourse and no surface water risk, a full survey may be more than the application needs, and threshold levels plus a few spot levels tied to datum will do. The judgement is about what the design depends on.

Specifying the survey brief

A workable one-page brief contains the site boundary with a margin of at least 10 metres beyond it, the datum and the required accuracy, the feature list, the deliverables and formats, the timescale and access arrangements, and a note of what the survey will be used for.

That last line matters. A surveyor told the data will feed a hydraulic model and a drainage design will pick up culvert inverts without being asked. A surveyor told nothing will survey what is visible.

Issue the brief before the Flood Risk Assessment is drafted, not alongside it, and make sure the Drainage Strategy uses the same levelled base as the architectural drawings. Divergence between the two is the single most common internal inconsistency consultees find.

To see which flood and drainage documents your site is likely to need before you commission anything, enter your postcode on PlanScope and work through the guided questions. For specialist technical reports, qualified professional input may be required.

This is general guidance, not legal advice. Check the position with your local planning 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.