The Scottish Government

Contents |

Listen

CHAPTER 4: FLOOD ESTIMATION

Introduction

4.47 Important inputs into a flood risk assessment are the analysis of historic floods (where data is available) and estimation of flood flows for a range of annual probabilities or 'design' events. Flood estimates for the majority of catchments under investigation were undertaken using the Flood Estimation Handbook ⁷ ( FEH). The FEH offers three method categories for analysing design flood flows: the Statistical, the Rainfall-Runoff and hybrid methods. The Statistical method combines estimation of the median annual maximum flood ( QMED) at the subject site with a growth curve derived for a pooling group of gauged catchments that are hydrologically similar to the subject site. The Rainfall-Runoff method combines design rainfall with a unit hydrograph derived for the subject site. Hybrid methods involve a combination of the two. Both the Statistical and Rainfall-Runoff procedures require the derivation of catchment descriptors. For this study these were abstracted digitally using the FEHCDROM (1999 edition).

4.48 In a few cases, the FEH techniques were not applicable, usually due to the small size of the catchment. Here, flow estimates were made using alternative methods applicable to small catchments: Modified Rational Method, IH Report 124 ⁸ method for small catchments, ADAS 345 ⁹, and the best estimate chosen.

4.49 Flows are traditionally associated with a return period or probability and expressed as a 1 in a 100 year flood equivalent to a 1% annual exceedance probability flood. In each case a range of peak flows were calculated between 2 and 1000 years (Table 4-1) and an assessment of climate change was made using the 100 year flow plus an allowance for climate change to 2080.

Table 4-1: Return period and probability

4.50 All flood estimates were made using the HiFlows- UK peak flow dataset ¹⁰ up to the 2002 water year (October 2003) where applicable.

Methods Used

4.51 The main methods used in estimating the design flows are shown in Table 4-2 for each scheme. Additional details of the methods are included within Appendix B.

Table4-2: Summary of flood estimation methods used

Review and Comparison with Design Hydrology and Auto FEH

4.52 In all cases where scheme hydrology was undertaken, a comparison was made with estimates used for the SEPA Indicative River and Coastal Flood Map (Scotland) which used Auto FEH, with some local adjustments to assess the scale and reasons for differences in flow estimates. The Auto FEH dataset produced by CEH Wallingford provides flow estimates at 50 m intervals along every watercourse in the UK with catchment areas exceeding 3 km ². The estimates are produced by an automated FEH statistical methodology.

4.53 In some areas of Scotland SEPA have adjusted the flow grid using single site analysis on SEPA's river gauges. In some cases there is a substantial difference between these revised values and those used in our report. In most cases using single site analysis is unlikely to be appropriate for estimating 100 or 200 year flows unless the gauging station has a very long record.

4.54 The values used in generating the Indicative River and Coastal Flood Maps (ie the adjusted flow grid) are quoted in the FPS reports and within the database in order for the reader to distinguish between the two methods and help to explain why the flood outlines determined as part of this study and the SEPA flood maps may differ.

4.55 Table 4-3 illustrates how the flow estimates differ between the current estimates and the Auto FEH values for comparable schemes. Out of the 16 Flood Prevention Schemes where SEPA/ Auto FEH flow values were available, this study has found 6 of the schemes to have flows higher than Auto FEH and 10 of the schemes to have flows lower than Auto FEH.

Table 4-3: % difference between current flow estimates and SEPA / Auto FEH flow grid

Review and Comparison with Design Hydrology and Scheme Hydrology

4.56 Of the 7 FPS where hydrology was available for the 100 year flood from both the design and the updated analysis 5 were found to be lower than the original design flows and 2 greater. The details are shown in Table 4-4. For all the schemes studied in detail for which original design flows were known, updated flow estimates exceeded design flows in 65% of cases. The introduction of the FEH has increased flow estimation in a number of locations.

Table4-4: Comparison of current and previous estimates

Negative percentages show this studies flows are lower than previous design estimates and positive percentages the reverse.

Climate Change

4.57 The impacts of climate change have been assessed using the latest research published by Scottish Executive ¹¹. Scotland was divided into three climatically consistent regions (eastern, south west and north & north-west). "The analysis based on the UKCIP02 scenarios suggests that eastern Scotland will experience the largest change in fluvial risk with present 100 year flow becoming almost twice as common (40 to 50 year event) by the 2080's. This is approximately equivalent to a 20% - 30% increase in peak flows. South-western Scotland is projected to be similar though slightly less affected, with the present 100 year flow becoming about the 50 to 60 year event by the 2080s (approximately equivalent to a 15% to 20% increase in peak flows). North and north-western Scotland however is suggested to be less affected with the present 100 year flow only becoming about the 80 year flow by the 2080s (approximately equivalent to a 0% to 10% increase in flows)". The mid point climate change range for 2080 has been selected for use within this project as shown in Table 4-5 below.

Table 4-5: Climate Change % increases in peak flows by the 2080s

4.58 Taking this climate change data into consideration, the 100 year plus climate change peak flow was determined for each flow prediction point. This was then assessed to establish the implications of considering climate change.

Contents |