FEH flood hydrology and 1D–2D modelling, as an associate or freelance specialist
I estimate design flows with the Flood Estimation Handbook methods and build linked 1D–2D river models, for consultancies that need an extra hydrologist and authorities that need a model checked. At Hamstall Ridware, below Blithfield Reservoir, the gauge record said 30.0 m³/s for the 1-in-100-year flood; pooled analysis said 44.6.
Real model output: the 2022 Flood Modeller 2D run, the first generation of the Hamstall Ridware model (1-in-100-year flood + 29% climate allowance, 6 m grid; 149 ha at its peak). Replayed from the model’s arrival times, not to time scale. How to read this: drag the slider to move through the flood.
Animated map: in the 2022 model of a 1% AEP flood with a 29% climate allowance, water spreads over 79 hours to cover 149 hectares of the River Blithe and Trent floodplain at Hamstall Ridware, Staffordshire.
- years of annual maximum flows at the village gauge, fitted to estimate the rare floods
- 50
- pooled 1-in-100-year flow, against 30.0 from the gauge record alone
- 44.6m³/s
- flood attenuation index (FARL) below Blithfield Reservoir, against 0.998 without it
- 0.87
- grid of the linked Flood Modeller–TUFLOW model, built on 1 m LiDAR
- 5m
About the author. Hisham Nuhu is a UK-based flood and disaster risk specialist. He worked at England’s Environment Agency from 2019 to 2025 and holds an MSc in Water Science, Policy and Management from the University of Oxford. He works remotely or in-country with governments, development banks, UN agencies and consulting firms, and in England with councils, communities, landowners and developers.
How do you estimate design flows when the gauge cannot be trusted?
Find out why it cannot be trusted, then lean on methods that do not share the flaw. At Hamstall Ridware the weir is bypassed above about 13–14.5 m³/s, so the record flattens just where it matters. Pooled analysis of similar gauged catchments, in WINFAP, gave 44.6 m³/s at 1 in 100, against 30.0 from the gauge.
The clue was in the station record: the National River Flow Archive notes that the weir is bypassed at high flows. The plot shows the effect. The largest annual maxima bunch together instead of climbing, and a single-site fit to that record sits a third below the pooled estimate at 1 in 100.
The hydrology was entirely my work: the single-site and pooled statistics, ReFH2, the routing through the reservoir, the hybrid design flows and the climate allowances.
Fifty years at the gauge: the record flattens where floods bypass the weir
How to read this: each dot is one year’s highest flow, with rarer floods to the right. The shaded band is where water starts to go round the weir. Add the pooled estimate to see how far the gauge falls short.
Annual maximum flows at NRFA station 28002, Blithe at Hamstall Ridware, water years 1959/60 to 2020/21, Gringorten plotting positions and a Gumbel fit; the pooled estimate for the whole catchment is from my flood estimation record (2025).
Sources
Gauge data: NRFA station 28002 © UKCEH; Environment Agency. My analysis.
Show the 50 annual maximum flows as a table
| Water year | Annual maximum flow, m³/s |
|---|---|
| 1959/60 | 17.10 |
| 1960/61 | 17.30 |
| 1961/62 | 7.88 |
| 1962/63 | 2.74 |
| 1963/64 | 10.70 |
| 1968/69 | 17.30 |
| 1969/70 | 11.60 |
| 1970/71 | 23.70 |
| 1971/72 | 14.60 |
| 1972/73 | 7.70 |
| 1973/74 | 12.50 |
| 1974/75 | 8.56 |
| 1975/76 | 2.92 |
| 1976/77 | 20.10 |
| 1977/78 | 11.80 |
| 1978/79 | 16.80 |
| 1979/80 | 15.20 |
| 1980/81 | 15.40 |
| 1981/82 | 18.80 |
| 1982/83 | 15.40 |
| 1983/84 | 13.50 |
| 1984/85 | 9.72 |
| 1985/86 | 17.40 |
| 1994/95 | 19.80 |
| 1995/96 | 2.50 |
| 1996/97 | 9.07 |
| 1997/98 | 17.80 |
| 1998/99 | 15.50 |
| 1999/00 | 9.79 |
| 2000/01 | 22.00 |
| 2001/02 | 11.40 |
| 2002/03 | 17.40 |
| 2003/04 | 11.40 |
| 2004/05 | 9.49 |
| 2005/06 | 4.31 |
| 2006/07 | 16.10 |
| 2007/08 | 13.80 |
| 2008/09 | 13.20 |
| 2009/10 | 6.68 |
| 2010/11 | 3.76 |
| 2011/12 | 18.70 |
| 2012/13 | 18.10 |
| 2013/14 | 12.20 |
| 2014/15 | 4.34 |
| 2015/16 | 14.30 |
| 2016/17 | 10.20 |
| 2017/18 | 13.50 |
| 2018/19 | 9.30 |
| 2019/20 | 21.90 |
| 2020/21 | 19.70 |
FEH statistical or ReFH2: which should we use?
Usually both, and then explain the difference. At Hamstall the single-site estimate came out 33% below the pooled value and ReFH2 46% above it: the gauge misses the biggest floods, and ReFH2 does not know about the reservoir. The final design flows used ReFH2 hydrograph shapes scaled to the pooled statistical peaks.
Each method answers a slightly different question. The statistical methods say how big the peak is, from records. ReFH2 turns design rainfall into a full flood hydrograph, from catchment properties. A 2D model needs both: the right peak, and a hydrograph shape with the right volume and timing.
So the hybrid takes the peak from the pooled statistics and the shape from ReFH2, for each of the four sub-catchments. The full-catchment estimates in the chart are a check on that, not the design flows themselves.
Three standard methods, three answers for the same rare flood
How to read this: each bar is one method’s estimate of the peak flow for the whole catchment. Switch between the 1-in-100 and 1-in-1,000-year flood; tap a bar to read it.
Whole-catchment checks from my flood estimation record v4 (2025): single-site statistics, pooled statistics (WINFAP) and the ReFH2 rainfall–runoff model.
| Estimate | 1 in 2 years | 1 in 100 years | 1 in 1,000 years | Method |
|---|---|---|---|---|
| Gauge record | 13.5 | 30.0 | 39.7 | Single-site statistics from the village gauge (FEH) |
| Pooled (WINFAP) | 17.7 | 44.6 | 58.3 | Pooled statistics from similar gauged catchments (FEH, WINFAP) |
| ReFH2 | 25.8 | 65.0 | 98.9 | ReFH2 rainfall–runoff model |
Show the final design flows for the four sub-catchments
| Sub-catchment | Area, km² | 1 in 2 years | 1 in 100 years | 1 in 1,000 years |
|---|---|---|---|---|
| Blithe above Blithfield Reservoir (inflow to the reservoir) | 119.7 | 19.2 | 53.2 | 75.4 |
| Ash Brook | 12.8 | 3.1 | 10.2 | 16.3 |
| Pur Brook | 15.7 | 3.9 | 12.7 | 20.1 |
| Intervening area to the village | 14.5 | 2.2 | 7.1 | 11.2 |
How are design floods estimated below a reservoir?
Split the catchment at the dam. Estimate the flood into the reservoir, route it through the reservoir, then add the floods from the tributaries below, timed correctly. At Hamstall three-quarters of the 163 km² catchment drains through Blithfield Reservoir, and the flood attenuation index falls from 0.998 without it to 0.87.
Timing matters as much as size. The brooks below the dam peak first, with critical storm durations of 8.5 to 11.5 hours. The catchment above the reservoir has a critical duration of 21 hours, and its contribution arrives later.
The floodplain then holds water back too. In the 2022 model, a peak of about 34 m³/s flowing into the 2D floodplain upstream of the village came out as about 19 m³/s, some 31 hours later. Those figures are approximate, read from the model output.
How do you apply climate change allowances to design flows?
With the Environment Agency’s peak river flow allowances for the management catchment, chosen by flood zone, vulnerability and lifetime. At Hamstall, in the Trent Valley Staffordshire catchment, the central allowances were +15% for the 2020s, +17% for the 2050s and +29% for the 2080s. The upper end for the 2080s is +61%.
The allowance is applied by scaling the design hydrograph, so the flood volume rises with the peak. In the 2024 model, adding the 2080s allowance to the 1-in-100-year flood took the buildings flooded from 14 to 21, and the deepest water at a building from 0.13 m to 0.34 m.
More buildings flood, and deeper, as the allowance grows
How to read this: each bar is one modelled flood; the gauge on the right (wide screens) or the readout (phones) shows the deepest water at any building. Tap a bar or a button to step through.
Hamstall Ridware, 2024 model (Flood Modeller–TUFLOW, 5 m grid), built jointly with an Environment Agency modelling colleague.
Show the data as a table
| Flood | Buildings flooded | Deepest at a building, m | 0.1 m or deeper | 0.3 m or deeper |
|---|---|---|---|---|
| 1% AEP flood (1 in 100) | 14 | 0.13 | 1 | 0 |
| 1% AEP + 15% (2020s allowance) | 16 | 0.24 | 8 | 0 |
| 1% AEP + 17% (2050s allowance) | 19 | 0.31 | 9 | 1 |
| 1% AEP + 29% (2080s allowance) | 21 | 0.34 | 9 | 1 |
| 0.1% AEP flood (1 in 1,000) | 29 | 0.46 | 17 | 5 |
What goes into a linked 1D–2D model?
A 1D model of the channel and its structures, a 2D model of the floodplain, and a clean link between them where water spills over the bank. At Hamstall: Flood Modeller 1D with TUFLOW 2D, a 5 m grid from 1 m LiDAR, a surveyed channel and bridge, the reservoir outflow upstream and the River Trent downstream.
The hydrology was entirely my work. I built the 2024 Flood Modeller–TUFLOW model jointly with an Environment Agency modelling colleague. The first generation, in 2022, used Flood Modeller 2D on a 6 m grid; the replay at the top is that run.
I build full 2D hydrodynamic models at 10–30 m resolution. I used TUFLOW and Flood Modeller 2D at Hamstall; HEC-RAS 2D and SFINCS are other options for river and coastal work. At Hamstall the grid is finer, 5 m, because the decision turns on a single bank line beside a village.
| Part | Set-up |
|---|---|
| Engines | Flood Modeller 1D in the channel, TUFLOW 2D on the floodplain |
| Ground | 5 m 2D grid built from 1 m LiDAR |
| Channel | Surveyed channel and structures, including the arch bridge |
| Roughness | Manning’s n from OS MasterMap land cover |
| Boundaries | Upstream: routed reservoir outflow plus the tributaries. Downstream: a fixed level on the River Trent |
| Events | 1-in-100-year flood; the same with +15%, +17% and +29% climate allowances; 1-in-1,000-year flood |
| Options tested | Opening the second bridge arch; raising the right-bank defence |
How do you check a model against a real flood?
With whatever the flood left behind: gauge levels, wrack marks, photos and residents’ accounts. The Hamstall gauge is bypassed in floods like February 2020’s, so I checked the first-generation model against a residents’ questionnaire instead: 11 homes reported flooded on three roads, 14 in the model. Two measured depths gave a further check.
Channel roughness was raised from 0.03 to 0.075 in the wooded reaches to match the observed flooding. A hindcast like this is not proof that the model is right everywhere. It shows the model is close enough to trust for the decision it served: which option keeps the village dry.
| Road | Reported by residents | Flooded in the model | Questionnaire |
|---|---|---|---|
| Street A | 5 | 7 | 1 household did not respond |
| Street B | 3 | 5 | 2 households did not respond |
| Street C | 3 | 2 | all households responded |
Totals: 11 reported, 14 in the model. Roads are labelled generically and no property is named.
Can you work as an associate hydrologist or modeller for our consultancy?
Yes. I can take a defined package and deliver it to your templates and quality process: an FEH calculation record, design hydrographs with climate allowances, a model build or update in Flood Modeller and TUFLOW, or an independent review of someone else’s hydrology or model. Remote, with site visits in England where the work needs them.
Stafford Common is a second example: hydrology for a heavily urbanised catchment with several watercourses, including climate change, written up to FCERM and Water Framework Directive standards and presented to partners. The modelling there was worth an estimated US$40k, within a project of more than US$500k.
I have also run two internal Environment Agency training courses, for 10 and 3 colleagues.
- Hydrology
- FEH statistical methods, single-site and pooled in WINFAP; ReFH2; reservoir routing; urban adjustments; climate allowances.
- Hydraulic modelling
- Flood Modeller 1D and 2D, TUFLOW 2D, linked 1D–2D; HEC-RAS 2D or SFINCS where they fit the job.
- Analysis
- Python, GDAL and QGIS; Monte Carlo uncertainty analysis, as in my national river flood risk work for Ghana, which was also my Oxford MSc dissertation.
- Reporting
- Calculation records, model reports, review notes, and technical reports to FCERM and Water Framework Directive standards.
Questions people ask
Why is ReFH2 so high at Hamstall?
Because it treats the catchment as if the reservoir were not there. ReFH2 turns design rainfall into a flood hydrograph from catchment properties; Blithfield Reservoir stores and delays the flood from the upper catchment, which the method does not see. That is why the catchment was split at the dam.
Can you review another consultant’s FEH calculations or model?
Yes. I can check the choice of method, the pooling group, the catchment descriptors, the hydrograph shapes and timing, the model schematisation, the boundaries and the calibration, and write it up as a review note.
Have you built coastal 2D models?
My hydraulic modelling record is on rivers. My coastal work in Ghana is a static connected-inundation screening with provisional forcing, not a 2D model; a 2D model is the next stage for the priority sites. Ghana coast.
Do you work for authorities as well as consultancies?
Yes. Councils and drainage boards ask for the same things: a design flow they can defend, a model checked independently, or the hydrology behind an options appraisal. FCERM funding and appraisal.
What should I send you?
The catchment or site, the deliverable, any existing model or calculation record, your templates and the deadline.
Sources
- UK Centre for Ecology & Hydrology. National River Flow Archive, station 28002: Blithe at Hamstall Ridware. The gauge record, including the note that the weir is bypassed at high flows.
- UK Centre for Ecology & Hydrology. Flood Estimation Handbook and ReFH2. The statistical and rainfall–runoff methods.
- Environment Agency. Flood risk assessments: climate change allowances. GOV.UK. Peak river flow allowances by management catchment.
- Defra and Environment Agency (2023). Using modelling for flood risk assessments. GOV.UK. Standards, method statements, quality assurance and the model report.
Work with me
Need an extra hydrologist or modeller?
Send me the package: the catchment, the deliverable, your templates and the deadline. I will reply with a scope, a fixed fee and a timescale. I don’t take on work on cases I handled while at the Environment Agency.