ESPACE Decision Making Framework and Tools
Phase 2 Piloting Report – Main Volume
June 2005
ESPACE — European Spatial Planning:
Adapting to Climate Events
Environment Agency
Halcrow Group Limited
ESPACE
ESPACE Decision Making Framework and Tools
Phase 2 Piloting Report – Main Volume
June 2005
Environment Agency Halcrow Group Ltd
The Thames Barrier Burderop Park
Eastmoor Street,Charton Swindon, Wiltshire
SE7 8LX SN4 0QD
Tel +44 (0)208 305 4803 Tel +44 (0)1793 812479
www.environment-agency.gov.uk www.halcrow.com
ESPACE
ESPACE Decision Making Framework and Tools
Phase 2 Piloting Report – Main Volume
Contents Amendment Record
This report has been issued and amended as follows:
Issue / Revision / Description / Date / Signed1 / 0 / Draft for comment / 01/04/05 / MGS
2 / 0 / Final draft, excluding executive summary / 29/04/05 / MGS
3 / 0 / Final report – Main Volume / 25/06/05 / JMW
Publishing Organisation
Environment Agency, Rio House, Waterside Drive, Aztec West, Almondsbury, BRISTOL, BS32 4UD.
Tel: 01454 624400 Fax: 01454 624409 Website: www.environment-agency.gov.uk
© Environment Agency 2005 June 2005
All rights reserved. No part of this document may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise without the prior permission of the Environment Agency.
The views expressed in this document are not necessarily those of the Environment Agency. Its officers, servants or agents accept no liability whatsoever for any loss or damage arising from the interpretation or use of the information, or reliance upon views contained herein.
Contents
Executive Summary 1
Glossary 8
1 Background 11
1.1 The ESPACE project 11
1.2 The Thames estuary 13
2 Decision Making Framework 16
2.1 Introduction 16
2.2 Guidance, procedures and scenarios 16
2.3 Decision Testing Tools 23
2.4 Visualisation and stakeholder engagement 26
3 Broad-scale Application 28
3.1 Introduction 28
3.2 Broad-scale piloting of the decision making framework 28
3.3 Results summary and uncertainties 31
4 Local Application 34
4.1 Introduction 34
4.2 Local piloting of the decision making framework 34
4.3 Results summary 37
5 Discussion 39
6 Conclusions and Recommendations 47
6.1 Conclusions 47
6.2 Recommendations 50
References 52
Executive Summary
Adaptation to climate change is essential if we want to minimise the impacts and take advantage of the opportunities that arise. The ESPACE research project is an ambitious four-year European project that aims to promote awareness of the importance of adapting to climate change and to encourage adaptation within spatial planning mechanisms at local, regional, national and European levels. Focussing on North West Europe, ESPACE is looking at how we manage our water resources and plan for a future with a changing climate.
One component of the ESPACE project is the development of a Decision Making Framework and Decision Testing Tools to aid the selection of spatial planning adaptation measures to cope with climate change. The Environment Agency through its Thames Estuary flood risk management project, TE2100, is leading on this component of the research project. The ESPACE Decision Testing Framework Phase 1 report (Environment Agency, 2004) presented the requirements of the framework and a thorough review of a set of candidate tools. The review concluded that a suitable generic Decision Making Framework is provided by the UKCIP Decision Making Framework (Climate adaptation: risk, uncertainty and decision-making, UKCIP 2003) as summarised in Figure 1.
Figure 1: UKCIP Decision Making Framework
The Phase 1 review did not identify a specific decision-testing tool that is appropriate for all sectors, locations and scales. Instead, it recognised that there is a range of tools that may be beneficial for particular studies. For application to the TE2100 project, the following tools were identified for piloting (the tools are described in more detail in subsequent sections of this Executive Summary):
· Source-Pathway-Receptor model to help identify the problem and objectives
· IPCC/UKCIP climate change scenarios to define climate change scenarios and their impact on the sources of flooding (primarily sea level rise and surges)
· TUFLOW and ISIS hydraulic modelling software to convert changes in extreme sea level to water depths at the receptors (primarily properties and people)
· MSDF software to calculate flood risk (consequences x probability) by translating scenario-neutral water depths at receptors into economic flood damage and social impact
· Excel Workbook to post process results and map the scenario-neutral data to specific strategic options
· FloodRanger Professional software as a strategic option exploration and visualisation tool for stakeholders
The Phase 2 report focuses on the piloting of the UKCIP Decision Making Framework and the above tools on the Thames Estuary at the broad and local scale (Figure 2).
Figure 2: Thames Estuary pilot area showing flood risk areas (‘embayments’)
The UKCIP Decision Making Framework provided clear structured guidance on decision making, highlighting both the sequential stages involved in decision making (stages 1 to 8 in Figure 1) and the need to iterate between stages (particularly stages 3, 4 and 5). Completion of the formal questions posed in the UKCIP Framework provide a very valuable audit trail which will encourage a systematic approach to the decision making and provide documentation suitable for stakeholder scrutiny.
The adoption of the Source-Pathway-Receptor (Figure 3) model helped to both identify the problem and objectives and to establish the decision-making criteria. Through the application of expert knowledge, a comprehensive list of risk components were identified and ranked. This process enabled the identification of tidal flood risk as the main ‘source’ of risk in the Thames estuary, and the resultant impact on properties and people as the main ‘receptor’ of this risk. The main ‘drivers’ were identified as climate change (increasing sea levels and surges) and land development pressures in the Thames Gateway. ‘Responses’ were identified at two levels of detail: at the broad scale responses were represented in terms of generic strategic options (such as maintain existing defence system or maintain existing standard of protection); whereas at the local scale responses included specific defence level and defence realignment options. For the piloting the decision-making criteria were the identification of cost-effective flood risk management strategies for properties and people given future likely climate change scenarios over the next 100years. (Note that full TE2100 project has a wider remit and will, for example, include environmental benefit in the criteria).
Figure 3: Drivers and responses can change the sources, pathways and receptors of risk
A distinctive feature of the UKCIP Decision Making Framework is the iterative application of stages 3 to 5 – assessing risk, identifying options and appraising options. This explicitly recognises that different approaches to risk assessment are required according to the level of understanding of the problem, structuring this approach through: risk screening; qualitative and generic quantitative risk assessment; and specific quantitative risk assessment. Within the Thames Estuary study area, the piloting focussed on the first two tiers of these stages at both the broad (Estuary wide) and local (Dartford embayment) scale.
For the risk assessment, a number of climate change scenarios based on IPCC SRES (Special Report on Emissions Scenarios) emissions scenarios were developed to provide a range of possible future tidal water levels to 2100. During piloting, it was recognised that UKCIP02 climate change scenarios provided a good starting point for the development of these scenarios, but did not include consideration of key components of Thames estuary tidal water levels, namely, storm surge and tidal propagation. These two components were therefore added to the sea-level rise climate change estimates.
Generic quantitative risk assessment was undertaken through the application of the selected Decision Testing Tools. The principal tool used during this stage of the piloting was the MDSF (Modelling and Decision Support Framework). This permitted the rapid estimation of direct economic damages associated with the flooding of residential and commercial properties, and an estimation of the number of people affected by flooding. This tool was supported by the use of the ISIS 1-dimensional and TUFLOW 2-dimensional hydraulic modelling software applied to the study area to provide information on estimated flood extent, depth and rate of flooding. These data were further processed to enable the calculation of risk of loss of life based on a rate of rise in flood water criterion.
Importantly, the application of the MDSF decision testing tool enabled the wide evaluation of strategic options and the identification and appraisal of options that were robust to climate change impacts. This appraisal was undertaken iteratively at a broad-scale to filter strategic options. During this process, a scenario-neutral approach was undertaken to modelling and application of the MDSF decision testing tool. An initial matrix of modelling was undertaken independently of climate change scenario and strategic option. This initial matrix was subsequently mapped across to particular strategic options using an Excel Workbook (the computationally intensive inundation modelling was thus decoupled from the economic damage calculation and strategic option). Such an approach enabled a wide variety of strategic options to be considered without the need for each strategic option to be explicitly modelled (see Figure 4).
Figure 4: Scenario-neutral database of modelling results supports appraisal of strategic options
Once a limited number of strategic options had been identified, a further iteration of the appraisal stage was undertaken at a more detailed spatial resolution for the Dartford ‘local’ pilot area within the wider study area (see Figure 2). This iteration included the explicit modelling of strategic option scenarios, enabling both a comparison of scale and method to be undertaken.
Further stages of the UKCIP Decision Making Framework were not applied during this piloting as a full assessment of preceding stages using specific quantitative risk assessment were not completed (ie the process stopped at step 6 of Figure 1). However, the development and trialling of ‘FloodRanger Professional’ as a visualisation and strategic option exploration tool was undertaken, both to assist with option appraisal and stakeholder engagement (Figure 5 shows example screen shots). The version of FloodRanger developed through the ESPACE project (called ‘Professional’ to differentiate it from the previous ‘educational game’ version) was able to import the MDSF generated Thames Estuary flood risk data (for current conditions, 2050 and 2100) and interpolate between these time slices to enable estimates of flood risks for 10-year time slices. The software concept is considered a significant innovation as it allows non-modellers to view outputs of potentially complicated modelling and risk assessment calculations in an intuitive and visually appealing software product. Further development of the concept is recommended to provide a simplified fit-for-purpose tool that will enable flood risk managers and other stakeholders to be able to assess, and to communicate to others, the positive and negative impacts of proposed development.
Figure 5: FloodRanger Professional visualises MDSF results for the Thames Estuary
Sensitivity analysis was undertaken to assess the sensitivity of results to variations in input data and calculation method. Findings with specific relevance to the Thames Estuary project are contained in the Main Volume discussion section and the Technical Appendix. Generic findings are listed below.
· Significant ‘short cuts’ in the analysis are possible by identifying and focussing on the major contributors to overall risk. For example, for the local pilot study, less than 20% of all properties contributed over 90% of the risk (measured in terms of annual average economic flood damage). Thus, results are likely to be insensitive to sensible changes in data quality for most of the property data set and any effort to improve data quality should only address the major contributors to overall risk.
· There are likely to be important ‘thresholds’ in the analysis beyond which there are changes in the relative importance of variability in input data. For example, the estuary wide annual average flood damage values are relatively insensitive to the sea level rise prediction changing from 7mm/year to 8.9mm/year for the strategic option to maintain a 1:1000 year standard of protection (damages increase by only 5%). However, the same variation in sea level rise for the ‘maintenance only – declining standards’ strategic option results in a 350% increase in flood damage.
· Similarly, there are also likely to be important calculation method ‘thresholds’. An example of this is the potentially time saving assumption of expressing flood damage per embayment as only a function of relative water level (expressed as local river water level minus flood defence crest level). Sensitivity testing showed that while this relationship was valid for some river and defence levels, it became inappropriate for the most important river levels (and thus flood damage had to be related to both river level and defence level in the scenario-neutral database).
The piloting of the Decision Making Framework and Tools on the Thames Estuary leads to the following generic findings.
· The application of the UKCIP ‘Risk, Uncertainty and Decision Making’ Framework provides excellent generic guidance and a set of procedures appropriate for assessing the impact of climate change on spatial planning. Despite its ‘UK’ title, it is appropriate for use throughout the ESPACE partner countries and outside flood risk management (eg for scarcity of water resources, threat to biodiversity, threat to water quality).
· The Framework proposes an iterative and tiered approach to the assessment of risk, identification of options and appraisal of options. This enables a level of analysis that is appropriate to both the level of decision and the level of understanding of the risk problems and objectives.
· The tiered approach is consistent with the development of the scenario-neutral approach to strategic option appraisal (as used in the broad scale piloting) which provides rapid quantitative estimates of risk. This approach enables the identification of sets of robust strategic options that can be further assessed using more detailed, scenario-specific quantitative methods (and the early screening out of ‘non-sensible’ options).