Assessment and measurement of asset deterioration including whole life costing

Transcription

1 Assessment and measurement of asset deterioration including whole life costing Science Report: SC060078/SR2 Product code: SCHO0509BQAV-E-P

2 The Environment Agency is the leading public body protecting and improving the environment in England and Wales. It s our job to make sure that air, land and water are looked after by everyone in today s society, so that tomorrow s generations inherit a cleaner, healthier world. Our work includes tackling flooding and pollution incidents, reducing industry s impacts on the environment, cleaning up rivers, coastal waters and contaminated land, and improving wildlife habitats. This report is the result of research commissioned and funded by the Environment Agency s Science Programme. Published by: Environment Agency, Rio House, Waterside Drive, Aztec West, Almondsbury, Bristol, BS32 4UD Tel: Fax: ISBN: Environment Agency June 2009 All rights reserved. This document may be reproduced with prior permission of the Environment Agency. The views and statements expressed in this report are those of the author alone. The views or statements expressed in this publication do not necessarily represent the views of the Environment Agency and the Environment Agency cannot accept any responsibility for such views or statements. This report is printed on Cyclus Print, a 100% recycled stock, which is 100% post consumer waste and is totally chlorine free. Water used is treated and in most cases returned to source in better condition than removed. Further copies of this report are available from: The Environment Agency s National Customer Contact Centre by ing: [email protected] or by telephoning Author(s): Marit Brommer Jaap Flikweert Peter Lawton Marta Roca Jonathan Simm Dissemination Status: Publicly available Keywords: deterioration, whole life costing, asset management, flood defences, coastal defences, maintenance Research Contractor: HR Wallingford Ltd Howbery Park, Wallingford, Oxon, OX108BA Tel: +44(0) Environment Agency s Project Manager: Stefan Laeger (Manley House, Exeter) Tim Hopkins (Cumbria House, Cardiff) Science Project Number: SC060078/SR2 Product Code: SCHO0509BQAV-E-P ii Science Report Assessment and measurement of asset deterioration including whole life costing

3 Science at the Environment Agency Science underpins the work of the Environment Agency. It provides an up-to-date understanding of the world about us and helps us to develop monitoring tools and techniques to manage our environment as efficiently and effectively as possible. The work of the Environment Agency s Science Department is a key ingredient in the partnership between research, policy and operations that enables the Environment Agency to protect and restore our environment. The science programme focuses on five main areas of activity: Setting the agenda, by identifying where strategic science can inform our evidence-based policies, advisory and regulatory roles; Funding science, by supporting programmes, projects and people in response to long-term strategic needs, medium-term policy priorities and shorter-term operational requirements; Managing science, by ensuring that our programmes and projects are fit for purpose and executed according to international scientific standards; Carrying out science, by undertaking research either by contracting it out to research organisations and consultancies or by doing it ourselves; Delivering information, advice, tools and techniques, by making appropriate products available to our policy and operations staff. Steve Killeen Head of Science Science Report Assessment and measurement of asset deterioration including whole life costing iii

4 Executive summary The Assessment and measurement of asset deterioration including whole-life costing research project aims to improve our understanding of the process of deterioration of common materials and components in flood risk management (FRM) assets and the way that deterioration interacts with asset condition and performance under service conditions. Understanding and quantifying deterioration rates is important for estimating and planning programmes of maintenance that contribute to an asset s whole-life costs (WLCs), and for the day-to-day maintenance and renewal intervention activities. As deterioration processes are closely related to the costs incurred over the life of assets, whole-life costing is another goal of the research. The main link between deterioration processes and WLCs is the relationship between periodic investments in managing the asset (maintenance, refurbishments) and the resulting performance of the asset, which includes the effect of reducing the deterioration. The project was conceived as being likely to require three phases: Phase 1 delivers the project scoping together with some interim deliverables. It includes the detailed assessment and collation of existing data and initiation of theories to tackle the process of developing deterioration curves and basics for whole life costing. A practical guide with typical deterioration curves for flood defence assets is also produced. Phase 1 also identifies and drafts out the scope of work suggested for the following phases. Phase 2 commences the implementation of the programme set down in Phase 1 and could include setting up the monitoring framework and potentially the carrying out of baseline surveys at monitoring or pilot sites. This phase could be progressed by a combination of Environment Agency Science (R & D) funds and funds from local Environment Agency areas and coast protection authorities in which assets will be monitored. Phase 3 could mainly comprise of ongoing monitoring of Environment Agency or local authority assets at pilot sites and be largely funded outside the Environment Agency science programme. This report presents the results achieved in Phase 1 of the project. The research project collected available knowledge from Environment Agency areas, from both Asset System Management and Operations Delivery, and from a variety of maritime local authorities. The collated information provides evidence of the rate of asset deterioration at the individual asset level and at a system level (for example, systems of groynes). Availability of information about costs has also been identified from interviews. The understanding of deterioration processes has been improved by the use of data collated both during the interviews and from the literature review, identifying the relevant failure mechanisms and key variables for a number of assets. The project has produced the Guidance on determining asset deterioration and the use of condition grade deterioration curves. This practical guide details the deterioration curves and how to use them to quantify the residual life of different types of assets. Vertical walls, embankments, culverts, dunes and shingle beaches are covered. These curves will be useful for guiding Environment Agency areas and local authorities in preparation of asset management plans. Changes in the dominant deterioration processes at different times in composite assets imply the use of limiting iv Science Report Assessment and measurement of asset deterioration including whole life costing

5 values of deterioration curves. Examples of how this should be considered are provided in the guide. The project has also identified ongoing developments in state of the art science related to whole life costing, which confirm the strong interrelationship with the asset deterioration element. The project has concluded that methods and tools for flood defence asset whole life costing can be developed in the next stage: the general concepts from asset management science can be applied, and examples from other fields of asset management can be used. This will require translation to the specific concepts and language of flood defence asset management. The project has found that there are various highly relevant ongoing developments in the Environment Agency s asset systems management (such as System Asset Management Plans [SAMPs] and Asset Management IT [AMIT] tools), where some of these specific concepts are being developed. Hence, any work on whole life costing needs to be fully embedded in those developments The project has identified the main goals and activities for Phase 2. These include activities related to deterioration, whole life costing and the setting-up of a long-term monitoring programme to facilitate the gathering of information on asset deterioration and the cost-effectiveness of different maintenance strategies. Monitoring activities (which should carry on beyond the end of Phase 2) would provide the required evidence base on deterioration and its relation to maintenance approaches. Phase 2 of the project would include measurement and understanding of component-level deterioration processes allowing more reliable and site specific deterioration and fragility curves to be created. It also could provide a vehicle within ongoing projects such as Performance-based Asset Management System (PAMS) for understanding the impact on flood risk of adopting various maintenance scenarios. Activities related to whole life costing methods and tools in Phase 2 would support flood defence asset management in the short, medium and long term. Science Report Assessment and measurement of asset deterioration including whole life costing v

6 Acknowledgements The authors wish to thank the Project Board (Linsay Hensman, Geoff Baxter, Tim Hopkins, Stefan Laeger) and the Environment Agency and Local Authorities representatives and operational staff which were consulted. vi Science Report Assessment and measurement of asset deterioration including whole life costing

7 Contents 1 Introduction 1 2 Background information General overview Types of assets and maintenance regimes General concepts about deterioration General concepts about whole life costing 14 3 Conceptual framework General framework Deterioration framework Whole life costing framework Coordination between ongoing programmes 26 4 Collation of available knowledge Interviews Summary of interviews Consultation with Environment Agency National Process team Refined end user requirements 34 5 Deterioration curves 35 6 Approach for Phase Goals and methodologies of Phase WP 1. Developing and conducting a focussed data collection and monitoring programme WP 2. Developing and testing robust methods and models for assessing asset deterioration WP 3. Developing and testing robust methods and models for assessing whole-life costs under different maintenance regimes for selected asset types WP 4: Developing improved guidance on determining asset deterioration and assessing the effects of different maintenance regimes 45 7 Conclusions 46 References 48 Appendix: Reports of interviews 50 Science Report Assessment and measurement of asset deterioration including whole life costing vii

8 Tables Table 2.1 Condition grades in general assessment 4 Table 2.2 Main components of the anchored sheet pile wall 7 Table 2.3 Deterioration modes and indicators of deterioration 9 Table 2.4 Cracking/fissuring: Excitation, ancillary, affected features, the character of the deterioration process and the affected failure modes (after Buijs, 2007) 10 Table 2.5 Excitation, ancillary and affected features for the toe accretion/erosion asset time-dependant process 11 Table 3.1 Identification of key flood defence assets and availability of data for WLC 26 Table 4.1 Details of interviews 29 Table 5.1 Deterioration times (years) to different condition grades for different asset types and exposures 36 Figures Figure 2.1 Example of fragility curve 3 Figure 2.2 Deterioration curve for gabions with normal maintenance. Asset type 1: Fluvial vertical narrow walls 4 Figure 2.3 Uncertainty classification, from Vrijling&Van Gelder (2005) 4 Figure 2.4 List of asset types used in the National Flood Risk Assessment (NaFRA Figure 2.5 Process modelling for anchor corrosion 13 Figure 2.6 Levels of whole life costs data structure (El-Haram, 2002, et al.) 15 Figure 2.7 Influence of construction strategies on deterioration curve (Arya et al. 2003) 16 Figure 2.8 Influence of maintenance strategies on deterioration curve (Hong et al., 2007) 17 Figure 3.1 Conceptual Framework 18 Figure 3.2 Theoretical deterioration curve with no maintenance (above) and deterioration curve modified by a specific maintenance regime (below) 19 Figure 3.3 Conceptual framework for deterioration curves 20 Figure 3.4 Phases in the modelling methodology of statistical models for time-dependent processes (Buijs, 2007) 21 Figure 3.5 Relation between deterioration rates and whole life costing 25 Figure 3.6 Conceptual links with ongoing projects as PAMS 27 Figure 4.1 Location of interviews 28 viii Science Report Assessment and measurement of asset deterioration including whole life costing

9 1 Introduction To optimise the risks, performance, cost of construction, maintenance, upgrading or replacement of Environment Agency s flood defence assets, a more rational whole-life approach to asset management will be implemented through the asset management strategy in Flood Risk Management (FRM). Basic input data for this whole life cycle assessment of investment and management are: cost of asset ownership (design/construct/operate/maintain/remove or replace) rates of deterioration of the overall asset and critical elements This requirement drives the main objective of the Assessment and measurement of asset deterioration including whole life costing research project. One of the main goals of the project is to understand the process of deterioration of common materials and components in FRM assets and the way that deterioration interacts with asset condition and performance under service conditions. Understanding and quantifying deterioration rates is important for estimating and planning programmes of maintenance which contribute to an asset s WLCs, and for the day-to-day maintenance and renewal intervention activities. Deterioration processes are closely related to the costs incurred over the life of assets, and that it is why whole life costing is another main goal of the research project. The availability of data about maintenance costs is poor compared with that of the costs of design and construction. In addition, methods to determine WLCs for flood defence assets are not yet part of current practice. The main link between deterioration processes and WLCs is the relationship between periodic investments in managing the asset (maintenance, refurbishments) and the resulting performance of the asset, which includes the effect of reducing the deterioration. The objectives of the project are: To collate from Environment Agency areas, coast protection authorities and Internal Drainage Boards (IDBs) existing data and available site information on the deterioration and WLCs of the principal asset types and materials in flood and coastal defence construction. To develop a fundamental understanding of failure and deterioration processes together with a framework for maintenance costs related to assets and asset systems based on sound engineering experience and judgement with a strong practical approach. Understanding of deterioration should be performancebased and take account of: the asset s overall ability to resist structural failure leading to breaching and consequential flooding or erosion; the asset s ability to resist overtopping; and the ability to perform non-flood defence/erosion protection related functions, e.g. access, health and safety, and environmental. Prepare reliable generic deterioration curves, expressing deterioration rates and residual life estimates, for structures of different types and different condition grades. These curves will be useful for guiding Environment Agency areas and local authorities in preparation of asset management plans, if supplemented by guidance on how they could be adapted to local conditions. They will also be directly usable for carrying out national and regional assessments of investment need. Science Report Assessment and measurement of asset deterioration including whole life costing 1

10 Develop guidance and tools for determining WLCs for flood defence assets. These have to fit in seamlessly with the wide range of ongoing developments in improving asset management, particularly within the Environment Agency on the basis of the asset management strategy. To set up a long-term monitoring programme for deterioration and maintenance costs to facilitate the gathering of information on asset deterioration and the cost-effectiveness of different maintenance strategies. The Assessment and measurement of asset deterioration including whole life costing research project does not aim to develop a tool to make a direct decision about maintenance, but aims to provide asset managers with the ability to assess the different possibilities considered. The outputs of this project will provide end users with guidance and criteria to improve the management of coastal and fluvial defences. The project was conceived as being likely to require at least three phases: Phase 1 delivers the project scoping together with some interim deliverables. It includes the detailed assessment and collation of existing data and initiation of theories to tackle the process of developing deterioration and fragility curves and basics for whole life costing. It will also identify and draft out the scope of work required in Phases 2 and 3. The scoping in Phase 1 involves summarising available knowledge on deterioration processes, setting up a framework linking WLCs/performance/deterioration; and identifying current approaches for wholelife costing for flood defences in England and Wales. Specific outputs from Phase 1 include a practical guide that presents the improved set of deterioration curves and details how to use them to quantify the residual life of different types of assets (vertical walls, embankments, culverts, dunes and shingle beaches), an identification of candidate pilot sites for Phase 2, and a scoping of the pilot sites and monitoring programme for the next phases. Phase 2 commences the implementation of the programme set down in Phase 1 and could include setting up the monitoring framework and potentially the carrying out of baseline surveys at monitoring or pilot sites. This phase could be progressed by a combination of Environment Agency Science (R & D) funds and funds from local Environment Agency Areas and coast protection authorities in which assets will be monitored. Phase 3 could mainly comprise of ongoing monitoring of Environment Agency or local authority assets at pilot sites and be largely funded outside the Environment Agency science programme. This report presents the results achieved in Phase 1 of the project. Chapter 2 presents some of the basic concepts related to deterioration and whole-life costing. Frameworks to develop a fundamental understanding of deterioration and maintenance costs concepts are described in Chapter 3. Chapter 4 presents data collated from Environment Agency areas, coast protection authorities and IDBs. Deterioration curves for different types of assets are presented in Chapter 5. Chapter 6 drafts out the scope of work required in next phases. Conclusions are presented in Chapter 7. 2 Science Report Assessment and measurement of asset deterioration including whole life costing

11 2 Background information 2.1 General overview A flood defence structure is designed to fulfil several functions during its lifetime. Different chains of events can lead to the situation that a flood defence fails to perform its functions. Such a chain of events is referred to as a failure mechanism. The reliability of a flood defence is represented by a combination between the strength of the defence and the loading of the defence structure in the form of the following limit state equation: Z = R S (eq. 1) S expresses the loading which can, for example, be a function of the hydraulic loading conditions or the ground pressures behind a vertical wall. R represents the strength of the flood defence structure and can be a function of the thickness of the revetment blocks, or the crest level for example. Z is 0 when loading exceeds the ability of the structure to resist it and defines failure according to the limit state equation. The limit state equation can either represent a full failure mechanism or one step in a larger chain of events. WLC analysis needs information on the loss of reliability typified by fragility curves (Figure 2.1). Figure 2.1 Example of fragility curve At present, performance-based modelling links each of the five Environment Agency visually assessed condition grades (Table 2.1) with a unique fragility curve. The time for a structure to move from one condition grade to the next could be estimated from deterioration curves. For different types of assets, these curves had previously been developed by HR Wallingford (Figure 2.2) based on expert judgement. But a fundamental understanding of failure and deterioration processes in the assets and their component elements and materials is needed to ensure that these curves are realistic and representative of actual asset types and performances. Science Report Assessment and measurement of asset deterioration including whole life costing 3

12 Table 2.1 Condition grades in general assessment Condition grade Rating 1 Very good 2 Good 3 Fair 4 Poor 5 Very poor 1 Condition grade Time (years) Figure 2.2 Deterioration curve for gabions with normal maintenance. Asset type 1: Fluvial vertical narrow walls Implicit within any risk analysis are many different types of uncertainty, the majority of which can be conveniently categorised under two headings (HR Wallingford, 2002): natural variability or inherent uncertainty knowledge uncertainty or epistemic uncertainty Different uncertainty types applicable to flood risk analysis are specified in Figure 2.3. They are related to inherent uncertainties in time and space, and epistemic uncertainties about probability statistics and modelling. Inherent uncertainty Inherent uncertainty in time Inherent uncertainty in space Epistemic uncertainty Statistical uncertainty Parameter uncertainty Distribution type uncertainty Model uncertainty Figure 2.3 Uncertainty classification, from Vrijling & Van Gelder (2005) 4 Science Report Assessment and measurement of asset deterioration including whole life costing

13 Uncertainty and sensitivity analysis methods in the literature (Saltelli et al., 2000) are available to investigate the sensitivity of flood defence fragility to the different random variables in a reliability-based design of flood defences. 2.2 Types of assets and maintenance regimes River and coastal flood defence assets protect lives and properties against flooding. An extensive list of different types of asset can be obtained for example from the Environment Agency s National Flood and Coastal Defence Database. Science Report Assessment and measurement of asset deterioration including whole life costing 5

14 Table 2.2 List of asset types used in the National Flood Risk Assessment Table 2.2 shows the list of possible asset types as formulated for the purposes of the National Flood Risk Assessment (NaFRA). 6 Science Report Assessment and measurement of asset deterioration including whole life costing

15 The description of the main asset components helps to identify the deterioration processes and failure mechanisms involved. The main components and descriptors for the analysis of deterioration processes may be derived from the Environment Agency (2004). As an example, the main components and properties of a sheet pile wall are listed in Table 2.3. Table 2.3 Main components of the anchored sheet pile wall Anchored sheet pile wall component Steel sheet pile Anchor Retained ground Toe of sheet pile structure Description Steel pile retaining the ground to provide a vertical frontage. Providing horizontal support to sheet pile, to limit the depth of the sheet pile wall. Ground retained to form a private frontage that was previously used as a dock. The bottom of either the outward or inward embankment faces. Examples of main properties moment of inertia area of sheet pile cross section distance of centre of gravity from the outer fibre of the pile cross section sheet pile length yield stress anchor spacing anchor inclination anchor area anchor length anchor connection ground water level presence of old frontage structures stratification of the soil layers soil properties (i.e. cohesion, angle of internal friction, compressibility, grain size, pore size, density, horizontal grain stress coefficients) water pressure distribution in the pores, presence and size of voids or cracks. toe protection in the form of e.g. grass, geotextile, sheet or timber piling, concrete slab, rock armour or rubble. Depending on the periodicity of maintenance operations, SAMPs provide the following classification: Frequent maintenance: recurring planned activities undertaken every 5 years or less Intermittent maintenance: infrequent activities that are undertaken at greater than 5-yearly intervals Science Report Assessment and measurement of asset deterioration including whole life costing 7

16 Refurbishment: maintenance operations that return the asset to its original design performance Change: works that alter the standard of service of the asset Depending on the objective of the maintenance actions, they also are classified in this project as: Reactive maintenance: includes the replacement of damaged elements. For example, the replacement of broken planks in a groyne. Proactive maintenance: associated with the improvement of the conditions that can deteriorate the asset. For example, plank levels at groynes are raised to maintain their ability to accumulate beach material. 2.3 General concepts about Deterioration Deterioration processes Damage, deterioration and degradation are important notions in life cycle management. According to the Construction Industry Research and Information Association (CIRIA, 2007), damage is the result of an ongoing deterioration or degradation process usually involving a decline in the state of structural properties. Deterioration or degradation is used to refer to the process causing the damage or to the damage itself. All three notions are therefore associated with a negative effect on the overall performance of the structure. Deterioration is a time-dependant process that affects one or more variables in the process model defined for the strength R, and the loading, S. Changes in timedependant variables introduce changes in the probability of failure. Deterioration processes affect the properties of flood and coastal defences and therefore affect their failure modes. Deterioration can trigger failure modes not necessarily related to a storm event. Process-based models for deterioration processes are less organised and not as developed in comparison to those for failure modes. The relevant failure mechanism, deterioration processes and indicators for different types of asset were summarised by the Environment Agency (2004). For example, 8 Science Report Assessment and measurement of asset deterioration including whole life costing

17 Table 2.4 identifies deterioration processes for earth embankments: Science Report Assessment and measurement of asset deterioration including whole life costing 9

18 Table 2.4 Deterioration modes and indicators of deterioration Flood and coastal defence type Embankment / sloping seawall (defence could include crest wall to embankment, particularly for coastal areas) Deterioration Loss of beach or berm in front of defence Settlement Long term erosion (e.g. foreshore erosion, gradual slope erosion) Vermin infestation Cracking/microfissuring Seepage and softening Vandalism Crest and slope erosion due to heavy trafficking (vehicles, animals, pedestrian) Deterioration of vegetation Shallow slips Indicators of deterioration Reduction in beach/berm level Reduction in crest level Reduction in foreshore level, gradual loss of slope material Noticeable presence of vermin, holes within embankment Noticeable cracking (may only be apparent during dry weather) Soft/saturated areas of defence or ground nearby during high water levels Obvious vandalism damage to defence Heavy use by vehicles/ pedestrians/ animals, tyre ruts, vegetation and bank damage, worn surface and access points Loss/ increase in extent and quality of vegetation, infestation by invasive plants Movement of sections of embankment The key variables influencing deterioration relate to a particular deterioration process. Buijs (2007) classifies them into excitation, ancillary and affected features. The excitation features are the flood or coastal defence properties that initiate and drive the asset time-dependent process. These change the asset s intended design conditions. These could include: - Environment-induced factors such as hydraulic loading, wind, temperature changes, sunshine, dying/wetting cycles, and chemical reactions. - Third-party interference associated with vandalism, animal infestation etc. Without these features no asset time-dependency takes place. The excitation features are often the main thrust behind the with-time variability in the time-dependent process. Excitation features can be direct loadings (e.g. wave action causing damage) or other kinds of feature not involving loading (e.g. moisture, oxygen, etc. causing corrosion) Depending on environmental conditions, excitation features promoting deterioration processes could be different for the same type of asset. The ancillary features are the flood defence properties that transform the excitation features into the asset time-dependent process. For example, the damage of the revetment caused by wave impact is also a function of slope, shape and weight of the revetment. The affected features are the flood defence properties that are subject to the asset time-dependent process. In case of revetments, damage affected features may be the revetment density or grading in different limit state functions. The affected features appear in one or more failure mechanisms and thus influence the fragility. 10 Science Report Assessment and measurement of asset deterioration including whole life costing

19 Table 2.4 presents the excitation, ancillary and affected features of the cracking/fissuring deterioration process for earth embankments Table 2.5 Cracking/fissuring: Excitation, ancillary, affected features, the character of the deterioration process and the affected failure modes (after Buijs, 2007) Cracking / fissuring Excitation features Desiccation and dry / wet cycles Slope instability Uneven settlements Ancillary features Embankment body - soil properties Same as above Same as above Affected features Crack / fissure width, crack / fissure depth, structure of cracking / fissuring. Embankment body, crest, outward face, inward face, berm Same as above Same as above Character deterioration process Continuous during a dry cycle, history-dependent Continuous depending on nature instability failure process Depending on nature settlement process Scientific understand ing Poor Poor Poor Affected failure modes Overtopping / overflow, Slope instability, Piping through embankment, Revetment failure and erosion embankment. The character of the asset time-dependant process depends on how the variability in time is introduced and transformed in the deterioration equations. Whilst deterioration often happens gradually (whether in a linear or non-linear fashion), sudden deterioration is also possible during a single (extreme) loading event. In Table 2.4, the fourth column describes the time character of the deterioration process. The uncertainty that different features introduce in a deterioration process has always to be taken into account. As an example, Table 2.5 tabulates the types of uncertainties that different features introduce in toe accretion/erosion asset time-dependent process. Science Report Assessment and measurement of asset deterioration including whole life costing 11

20 Table 2.6 Excitation, ancillary and affected features for the toe accretion/erosion asset time-dependant process Excitation features Tidal currents River discharge Local wave climate Dredging activities Naval activity, it is noted that the frontage is not in use as a dock facility anymore Ancillary features Bathymetry of the river as a system The shape and type of frontage in relation to the tidal currents or wave reflection, e.g. vertical sheet pile wall structure, or sloped foreshore Local sheltering due to the alignment of the frontage Affected features Uncertainty Inherent uncertainty in space and time Inherent uncertainty in space and time Inherent uncertainty in space and time Inherent uncertainty in space and time Inherent uncertainty in space and time Uncertainty Inherent uncertainty in space and time Inherent uncertainty in space Inherent uncertainty in space Uncertainty Toe level Stochastic process introduces inherent uncertainty in time and space Deterioration processes are initiated by environmental excitation features such as: waves, water level difference, turbulence, currents, freeze/thaw cycles, sun, rainfall, salts, drought, etc. Initiation of deterioration can also be caused by third party excitation features such as: vandalism, trafficking, animal burrowing, animal infestation, industrial activity, superimposed loading, boat collision, etc. The excitation features interact with the ancillary features of the flood or coastal defence structure thereby affecting defence features in the form of deterioration. Ancillary and affected features are formed by the flood defence components and their structural properties. Analysis of the excitation and ancillary features leads to a qualitative evaluation of the character of the deterioration process. The character of the deterioration process together with the input of knowledge forms the basis of the choice for the statistical model for the deterioration. The input of knowledge depends on the level of scientific understanding of the deterioration and the required detail of decision-making. The deterioration processes for each defence type can be discussed according to the following structure (CIRIA/CUR/CETMEF, 2007): flood or coastal structure primary functions and components deterioration processes at a system level deterioration processes at a defence (cross) sectional level (defence geometry) 12 Science Report Assessment and measurement of asset deterioration including whole life costing

21 deterioration processes at an structural integrity level (defence composition) deterioration processes at an element composition level Deterioration equations Deterioration equations are a mathematical representation of some of the deterioration processes explained above. They basically depend on the loading (excitation features) history or, in mathematical language, on the excitation of the process. Different categories of deterioration process can be identified related to the timedependency of the excitation: constant, for example the compaction of the soil continuous, such as corrosion recurrent, such as toe accretion in front of an anchored sheet pile wall structure due to wave climate during storms As mentioned above, the excitation sometimes involves direct loadings (which may also be involved in failure processes). On other occasions the excitation may come from the environment, but should not be conceptualised as a loading. For example deterioration equations may be described as: settlement in crest level of embankments, described by the pore pressure variation equation settlement in crest level of embankments due to trafficking events, described by a compaction factor using a Poisson distribution to model events loss of effective cross-section area in anchors by corrosion toe accretion/erosion As an example, Figure 2.5 displays a number of simulations with the deterioration model for anchor corrosion where it can be observed how the anchor cross-section diminishes with time. Science Report Assessment and measurement of asset deterioration including whole life costing 13

22 Figure 2.4 Process modelling for anchor corrosion The current knowledge on asset deterioration has been extensively developed in a PhD thesis, Reliability analysis of the long-term behaviour of flood defences by Foekje Buijs (2007). This work contributes to a better understanding of the theoretical background of the scope of the project and avoids duplication. It aims to investigate how the time-dependent behaviour of flood defence properties can be appropriately characterised and incorporated in a reliability-based approach. Some of the main conclusions of the thesis are presented below. The asset time-dependent processes can either be modelled with a Bayesian or a Markovian approach. This choice affects the type of statistical models that are developed for the asset time-dependent processes. It also determines how observations from monitoring are incorporated in the maintenance optimisation model and therefore affects the structure of the model. The Bayesian approach is favoured over the Markovian approach as it internalises rather than discards historical time series observations. The statistical models for asset time-dependent processes are categorised according to three main compositions: a stochastic process for an overall asset time-dependent quantity; a hierarchical process model consisting of a function of random variables and one or more stochastic processes; a parametric process model consisting of a function of random variables and a deterministic time. A stochastic process for an overall asset time-dependent quantity is in some cases favoured over a more detailed hierarchical process model. Lack of scientific understanding or the unavailability of field information are 14 Science Report Assessment and measurement of asset deterioration including whole life costing

23 examples of reasons to choose for such an approximation. It is also possible that due to financial and time constraints in some situations a detailed approach is not feasible. There is a collection of existing suitable stochastic process models to model different types of asset time-dependent processes, for example, recurrent shock damages or gradual continuously developing processes. Suitable multivariate distributions to represent correlations between asset time dependent processes also exist. 2.4 General concepts about whole life costing This section draws together the outcomes of the literature review on flood defence whole life costing. The general concepts, including the relationship between whole life costing and deterioration, aim to form a basis for the overall conceptual framework as described in Section 3, and will feed into the development of practical guidance and tools that is proposed for phase 2 of the project. WLC is defined as the cost of acquiring it (including consultancy, design and construction costs, and equipment), the costs of operating it and costs of maintaining over its whole life through to its disposal that is, the total ownership costs (Office of Government Commerce, 2003). The terms whole life costing, life cycle costing and through life costing are terms used interchangeably. The objectives of WLC are identified as (Flanagan, et al.,1983): To enable investment options to be more effectively evaluated To consider the impact of all costs rather than only initial capital costs To assist in the effective management of completed buildings and projects To facilitate choice between competing alternatives Whole life cost analysis Although the methodology in formulating the WLC varies with the nature of the project, the purpose, required depth and breadth of the WLC study, in general, the WLC can be calculated using the following equation: n m n m k WLC = Cc + ( Co j ) = ( Cm j ) + Cri + Cd + ( Cl) i= 1 j= 1 i= 1 j= 1 i= 1 Where Cc is capital cost (design/build cost), Co is operating cost, which is composed of maintenance (including inspection) cost Cm and refurbishment cost Cr, Cd is decommissioning cost, Cl is the cost from loss, n denotes the duration of the expected life of the project (year), m is the number of cost elements and k is the number of refurbishments. The capital cost, operating cost and decommissioning cost are included in most WLC studies. Whether the cost from loss is taken into account depends on the nature of the asset and the purpose of the WLC model. For example, in a cost model to asses the alternative design strategies for a bridge, it is appropriate to take the loss from interruption in future intervention into account (Arya et al., 2004). In another case, the Science Report Assessment and measurement of asset deterioration including whole life costing 15

24 risk loss may also need to be considered when a strategy is to be determined for a coastal defence (Simm et al., 2003). Also, benefit analysis is carried out in some cases to make WLC more meaningful. It is worthy to mention that, as one of the components of Performance-based Asset Management System (PAMS), the WLC model in this project will not consider loss or benefit, since evaluating the ratio of cost (including the cost from loss) to benefit is carried out elsewhere in PAMS Cost Breakdown Structure The first step in WLC analysis is to identify a cost breakdown structure that identifies all cost elements in all appropriate life cycle phases (British Standards Institution [BSI], 1997). A generic framework for collecting WLC data for the building industry has been proposed by El-Haram, et al. (2002). The proposed top-down hierarchy of the data structure for a building project is shown in Figure 2.6. The procedure suggests to break down the categories identified in each level until the cost can be easily calculated from the quantity and unit price. The difficulty in WLC is forecasting the operating activities, which can be estimated from historical data and expert opinion. Figure 2.5 Levels of WLCs data structure (El-Haram et al., 2002) Cost estimating relationship The cost of an identified item can be estimated as a function of one or more variables. Normally the cost is estimated by utilising mathematical formulation on the basis of the historically collected costs. The mathematical formulation can be straightforward, in which the individual activity incurred costs are accumulated, or complicated, in which a probabilistic method is used to take the uncertainty into account in order to obtain an optimum strategy (Val et al., 2003; Noortwijk et al., 2004). In estimating the future cost (operating cost, decommissioning cost and cost from loss), the economic trend needs to be forecast to evaluate the influence from the discount rate and/or the inflation. Taking into account the influence from the discount rate, the future cost can be expressed as Present Value (PV), which is defined as the value of a stream of benefits or costs when discounted back to the present time (Ministry of 16 Science Report Assessment and measurement of asset deterioration including whole life costing

25 Agriculture, Fisheries and Food [MAFF], 1999). PV can be calculated using the following equation: PV t = = N Ct r t= 0 (1 + ) 100 t Where N is time horizon in years, discount rate. Ct is the total monetary costs in year t, and r is the Application of deterioration rate in WLC The traditional method for WLC analysis has been mentioned above, which is widely used in optimal decision making. In recent years, a more rational method for WLC, in which the cost is linked with the deterioration of the asset, has been adopted for assets that have a long expected life duration and important role, such as the Ministry of Defence (MoD) defence and infrastructure. This approach would also be appropriate for flood defence assets. A WLC linked with deterioration rate in evaluating concrete bridge tenders is proposed by Arya et al. (2004), in which two strategies with different initial cost are evaluated for a bridge with 75-years expected life. The assumptions adopted in the WLC are (1) the maintenance will only be carried out when the threshold condition of the structure is reached and (2) the maintenance will restore the structure to its original condition. The influence of the construction cost on the deterioration curve is presented in Figure 2.7, based on which the maintenance cost can be estimated. Arya et al. suggested that the deterioration rate for individual key elements of the structure should be integrated into the total condition of the structure. Figure 2.6 Influence of construction strategies on deterioration curve (Arya et al. 2003) Another example of integrating the deterioration rate into WLC is given by Hong et al. (2007). Hong et al. classified the operating activities into time-controlled action, which is carried out at specified intervals and sub-categorised into preventive maintenance and corrective maintenance, and performance controlled action, which is carried out when a threshold condition is reached. The influence of the operating strategies on deterioration curves is presented in Figure 2.8, based on which the operating cost can be estimated. Science Report Assessment and measurement of asset deterioration including whole life costing 17

26 Figure 2.7 Influence of maintenance strategies on deterioration curve (Hong et al., 2007) The two examples given above indicate the extent of the influence from operating strategies on the deterioration curve of the structure; this indicates that the calculation of WLC is dependent on the method for establishing deterioration rates. The implication of this concept for flood defence asset management, and for the envisaged deliverables of the project are described as part of the conceptual framework in Section Science Report Assessment and measurement of asset deterioration including whole life costing

27 3 Conceptual framework Stage 1 of the project has developed a conceptual framework of the two elements of the project (deterioration and whole life costing) and of how these two elements interact. This general framework of the project is presented in Section 3.1. More detailed explanations about deterioration and whole life costing are developed in following sections. The project has also identified the role of deteriotion and whole life costing in the overall approach to the optimisation of asset management strategies, which is being developed in line with the asset management strategy through initiatives such as SAMPs and PAMS; this is discussed in Section General Framework Theoretical and empirical deterioration curves for different types of assets describe how the asset performance varies over time. This can also be expressed as the time for a structure to move from one condition grade to another or the increase of the probability of failure along time. On the other hand, asset managers have to define a maintenance philosophy, for example, whether or not any maintenance is going to be carried out. Applying these inputs to a particular asset, different maintenance regimes and deterioration curves are obtained (Figure 3.1). Maintenance regime 1 Particular deterioration curve 1 WLC cost 1 Maintenance philosophy Maintenance regime 2 Particular deterioration curve 2 WLC cost 2 Theoretical and empirical deterioration curves Particular asset... Maintenance regime n Particular deterioration curve n WLC... cost n MAINTENANCE ASSESSMENT Other inputs Figure 3.1 Conceptual framework The new deterioration curves, for a particular asset, will take into account changes introduced by maintenance as shown in Figure 3.2. Therefore, knowledge on deterioration and maintenance management regimes is linked to the development and improvement of whole life costing methods/models. The main link between the WLC method/model and the deterioration processes is in the relationship between the periodic investments in managing the asset (maintenance, renewals etc) and the resulting performance of the asset, including the effect of reducing or arresting deterioration. Science Report Assessment and measurement of asset deterioration including whole life costing 19

28 Figure 3.2 Theoretical deterioration curve with no maintenance (above) and deterioration curve modified by a specific maintenance regime (below) As every maintenance regime has a cost, the application of a whole life costing model will identify the costs throughout the whole life of the asset. Management options considered by asset managers should take into account, amongst any other inputs (time series of performance, benefits, etc), costs carried out under different maintenance regimes. 3.2 Deterioration framework An overall conceptual framework for developing deterioration curves is given in Figure 3.3. This shows that the two main approaches available to estimate the curves are the use of statistical models and the use of empirical data. The selection of which of these approaches should be adopted in a particular case by an asset manager will depend on the models and data already available. If the structure is low risk, then it may be perfectly acceptable to make estimates of future deterioration from the generic deterioration curves. In other higher risk cases, if long time series empirical data are already available this can be a reasonable predictor of future deterioration at least over the time interval to the next inspection. Brier skill score approaches as adopted in recent work on beach level monitoring and prediction can be used to assess how far into the future trends suggested by historical data can be projected. Where time series data are not available, statistical models may be an alternative way forward, especially where good process models have been framed and calibration coefficients are available from other sites or structures. In an ideal world, especially where the risk arising from failure due to deterioration is significant, both statistical models and databased approaches would be available and used. In the case of the pilot site based research envisaged for Phase 2 of this project (see Chapter 6), the approach will be to use both data and models in combination to maximise the learning. 20 Science Report Assessment and measurement of asset deterioration including whole life costing

29 Statistical models Identify DETERIORATION PROCESSES Expected DESIGN LIFE Empirical data Description of time-dependency Identify KEY VARIABLES Changes in CONDITION GRADE DETERIORATION EQUATIONS CONDITION FEATURE monitoring DETERIORATION CURVES Figure 3.3 Conceptual framework for deterioration curves Statistical models for time-dependent processes are useful tools to perform the deterioration analysis. An analytical approach developed by Buijs (2007) is presented in Figure 3.4 which illustrates the processes involved. Science Report Assessment and measurement of asset deterioration including whole life costing 21

30 Problem formulation phase (internal model properties) Task specification of asset time-dependent process at a process level and overarching maintenance optimisation model level Conceptualisation phase (internal model properties) Identify existing knowledge Site specific information Scientific understanding Qualitative analysis of asset time-dependent process Identify flood defence properties involved and their uncrtainties Excitation features e = e1,..,en Ancillary features a = a1,..,an Inherent uncertainty in space - statistical distribution or random field Inherent uncertainty in time - stochastic process (sections 5.3 & 5.4) Qualify character of asset time-dependent process (conditional on excitation) X i(a,t e) Affected features X = X 1,..,X n Dependencies with other processes Quantitative analysis of asset time-dependent process Statistica model for X (t): Specify statistical model for e1,..,en Quantify X (a,t e), as a function of ancillary and conditional on excitation Specify statistical model for a1,..,an Establish dependencies Parameter estimation, calibration and corroboration phase (external model properties) Individual asset time-dependent process Evaluate performance against task specification External properties (quantitative) Fitting to data (often low availability) Evaluate performance against task specification Sensitivity analysis Process as part of reliability or life cycle costing Fitting to data (usually not present) Figure 3.4 Phases in the modelling methodology of statistical models for timedependent processes (Buijs, 2007) 22 Science Report Assessment and measurement of asset deterioration including whole life costing

31 The modelling methodology for a time-dependent process consists of the following steps: 1. Establish the (functional) requirements against which to corroborate the final statistical model. These requirements are not straightforward and partly depend on the decision-making context. The requirements are not straightforward because even if historical data or scientific understanding are available, that does not necessarily justify extrapolation over the lifetime of a structure. 2. Evaluate the input of knowledge, consisting of the aspects below. a. The availability of site specific quantitative information b. The level of scientific understanding, i.e. in the form of a physical process-based model. Such a process-based model is a function of the excitation and ancillary features. 3. Identify excitation features. Specify the feature and the type of uncertainties associated with the feature, i.e. uncertainty in time and / or space. 4. Identify ancillary features. Specify the feature and the uncertainties associated with the feature, i.e. uncertainty in time and / or space. 5. Qualify the character of the time-dependent process, e.g.: Linear, polynomial, logarithmic, exponential; incremental cyclical incremental; history-dependent; continuous, cyclical continuous. 6. Establish the dependencies between time-dependent processes. For example: time-dependent processes caused by the same excitation; time-dependent processes that are a function of the same ancillary features, therefore leading to similar behaviour; time-dependent processes affecting or interacting with the same flood defence features; time-dependent processes affecting or interacting with the same failure modes; a time-dependent process affecting the ancillary features of another; one time-dependent process which (partly) forms the excitation of the other; a time-dependent process that initiates a failure mode, which in turn forms the excitation of another time-dependent process; several interacting time-dependent processes, e.g. seepage, animal burrowing and settlements. 7. Develop one or more alternative statistical models for the time-dependent process. The uncertainty in the time-dependent process is appropriately represented by capturing those introduced by the excitation and ancillary features. Three types of approaches that might be taken: a. Model the overall quantity Xi(t) b. Develop a process-based model consisting of one or more timedependent variables and a number of constant random variables. c. Develop a process-based model as a function of random variables and a deterministic time variable (conventional engineering approach with random variables). 8. Parameter estimation and calibration. 9. Corroborate the statistical model against the initial requirements. For a particular flood or coastal defence asset it is possible to identify deterioration processes and their key variables, including time-dependence. Deterioration equations can then be used to model the deterioration processes required. Science Report Assessment and measurement of asset deterioration including whole life costing 23

32 The deterioration equations will support an improved elaboration of deterioration curves to be based not just on expert judgement but also on the conceptual framework and the collation of available knowledge. A complementary and supporting approach to deriving deterioration curves is based on empirical data. Different levels of knowledge of real deterioration data can be collated. The first level is related to the design life of the asset. Designers should consider deterioration processes in their design of an asset and these data can be collected. A second level of knowledge concerns the changes in condition grade of assets based on real observations. These data could supply a real approximation to the deterioration curves. The third level is related to the monitoring of condition features. These records could provide a better understanding of deterioration processes and detailed information for empirical equations. Collation of real data from asset deterioration processes could be performed at different levels. visual inspection: trained inspectors collect useful data about deterioration rates, main excitation features, etc measurement level: a second level of expert judgement allows evaluation of the importance of deterioration processes scientific appraisal: e.g. an intrusive investigation collects detailed data about deterioration mechanics Such empirical data will contribute to a better development of statistical models and of course, to the main goal of the project, the deterioration curves. The collation of data from pilot sites listed in the Phase 1 of this Research Project will be required in subsequent phases to improve, complete and calibrate deterioration curves. Data obtained from pilot sites will also contribute to a better development of the statistical approach, and of course, is closed related to the empirical approach. 3.3 Whole life costing framework Some useful background information for the conceptual framework of an integrated WLC model is summarised below. Rogers et al. (2004) have used a WLC approach to estimate the minimum WLC of maintenance to deliver the optimum service life of coastal and estuarine defences. Their model is based upon a Bayesian statistical model allowing efficient use of available data. The modelling process involves developing rules that are applied to every asset on the network to determine a preliminary estimate of the costs (the preliminary cost model). Specific assets are examined in greater detail (using the structures asset management process). The outcomes are combined within the statistical model (engineering cost model) to provide the final profile. Thorstensen and Rasmussen (1999) recognized the need to focus on a continuous time deterioration process, and use trend curves to describe the deterioration process. The finite number of condition levels (or grades) of the system are transformed to a condition transition probability matrix (CTPM). All input data are modelled as a function of time or system status, which allows the flexibility to include cyclic variation. Within this setup there is a need to consider the finite horizon problem which implies that costs accumulate over a finite number of stages. The finite horizon problem is solved using stochastic dynamic programming as follows. At the beginning of a discrete time period, an asset is said to be in a set of possible conditions (states). The states are determined from condition assessment procedures at each decision point. At each stage it is 24 Science Report Assessment and measurement of asset deterioration including whole life costing

33 possible to observe the condition and make a decision on, for example, maintenance (repair, refurbishment etc.). From the state at the time horizon the decision is made (and hence an action is chosen), costs (or profit) can be calculated. The objective of their dynamic model is to determine which policy maximises the expected profits over the planning horizon (sum of all costs), which is achieved through recursive relations based on the principle of optimality (Kall and Wallace, 1994). Blanchard and Fabrycky (1990) propose an iterative whole life costing method which embeds feedback as a vital part of the method. In their approach, feedback from previous iterations informs each stage of the process (which could be the condition of the asset) to successively refine the considered feasible design or maintenance options and associated life cycle implications and costs. Although the authors have used this approach predominantly to evaluate conceptual designs that result in higher costs, the method of iterative feedback proves to be useful in the context of addressing uncertainty though explicit inclusion of sensitivity and risk analyses. The first need is to focus on acquiring data that address whole life costing in conjunction with deterioration rates in a consistent manner. Specifically, a number of existing data sets that relate to the costs associated with fixed flood defence assets need to be collated. Each data set has to include: A description of the base information from which the costs were derived; Key assumptions about the base information; A list of the factors included in and / or excluded from the costs; A description of the factors that are both the most susceptible to change and have the most potential to affect costs. The overall method should enable the user to determine WLC for assets or asset systems based on minimum input. One of the links between deterioration rates and whole life costing is the relation between the maintenance regime and the condition grade of the asset. Figure 3.5 presents the idea behind the integrated WLC model. The maintenance regime forms part of the asset management strategy and it creates a direct link to the SAMPs (further information in chapter 3.4). Science Report Assessment and measurement of asset deterioration including whole life costing 25

34 Integrated WLC model Model Input Output Asset Costs Type Capital costs (Cc) Expected life duration (n) Deterioration rates WLC Cc + ( Co ) = ( Cm ) + Cr + Cd ( Cl ) Condition (k) Operating costs (Co) Maintenance regime Maintenance (Cm) Refurbishment (Cr) Decommission costs (Cd) Whole life costs for flood defence assets n m n m k j j i + i = 1 j = 1 i = 1 j = 1 i = 1 = Figure 3.5 Relation between deterioration rates and whole life costing At a high level, there should be a distinction between ongoing maintenance and refurbishment. In order to develop a bottom-up approach to the costing of ongoing maintenance, we need to produce an overview of the activities that constitute flood defence asset management: Inspection Vegetation management Repair Operations Incident management/emergency response Regulatory activities (permits) General administration The guidance being developed by SAMPs will be very helpful in the process of determining how these activities relate to the maintenance regime, which is at the same time a function of the deterioration rate. Each maintenance regime will have a typical set of maintenance activities with typical characteristics (frequency, staffing, thoroughness, etc), with associated costs. These sets of activities and their characteristics will also be a function of other characteristics and parameters: asset type, accessibility, possibly type of operating authority, etc. The resulting methodology could have a default set of maintenance regimes with associated activities (such as the set of Maintenance Standards that have recently been developed for two asset types for EA s National Process team), with related costs that vary as a function of a limited number of characteristics. But it could also be optional for the users to put together their own maintenance regime, either as a set of maintenance activities, or even more basic as: number of staff 26 Science Report Assessment and measurement of asset deterioration including whole life costing

35 annual maintenance costs Some of these options are likely to be developments for later stages (beyond Phase I). The integrated WLC model needs asset management strategies as an input. It seems likely that these will be developed by users, making use of the deterioration tool that provides information to support decisions. The optimum asset management strategy will strongly depend on the costs and on the benefits; the WLC model only looks at costs, while the benefits (which is expressed as avoided flood risks) can be calculated with other methodologies developed. Based on a first set of interviews the following flood defence assets could be identified as key assets for which an integrated WLC model can be set up (Table 3.1). Note that this list reflects the shortage of costing information from the Environment Agency, which means that other sources of information may have to be used for flood defences. Table 3.1 Identification of key flood defence assets and availability of data for WLC Key Assets Responsible party Data for WLC Concrete Seawalls Worthing Borough Council Yes Canterbury Borough Council? North Norfolk District Council Waveney District Council Response Yes Yes Timber Groynes Worthing Borough Council Yes Rock structures/revetment Canterbury Borough Council North Norfolk District Council Waveney District Council Response Worthing Borough Council Canterbury Borough Council North Norfolk District Council Waveney District Council Response Yes Yes Yes Yes Yes Yes Yes Steel sheet piles Waveney District Council Response Yes Broadland Environmental Services Ltd Yes 3.4 Coordination between ongoing programmes The Assessment and measurement of asset deterioration including whole life costing project is closely related to a larger research project, the PAMS project, which aims to provide a management framework, a methodology, guidance, tools and support for the development of an operational infrastructure for more effective and reliable flood risk management, assessment and decision-making in the UK. This large project will feed into the development of the AMIT. Science Report Assessment and measurement of asset deterioration including whole life costing 27

36 A similar conceptual framework to that in Figure 3.1 is used in Figure 3.6 to show the link between this research project and the thinking process required for optimisation of asset management intervention described in Goulby et al (2008). Maintenance philosophy Theoretical deterioration curves Particular asset Maintenance regime 1 Particular deterioration curve 1 Maintenance regime 2 Particular deterioration curve 2... Maintenance regime n Particular deterioration curve n WLC WLC WLC cost 1 cost 2... cost n DECISION SUPPORT VIA COMPARISON OF MANAGEMENT OPTIONS System Analysis benefit 1 benefit 2... benefit n DETERIORATION AND WHOLE LIFE COSTS PROJECT PERFORMANCE-BASED ASSET MANAGEMENT Figure 3.6 Conceptual links with ongoing projects as PAMS Also important is the link between the Assessment and measurement of asset deterioration including whole life costing project and the SAMP process recently rolled out across the Environment Agency which considers the WLC of the current assets. SAMPs for FRM systems have been developed for around 150 pilot systems across England and Wales. SAMPs will be used as the framework with which to apply asset management principles across flood risk management. Major assets will have their own asset management plans that will focus on the operation of the site as well as the financial investment requirements. Asset managers for the Environmental Agency are required to develop a whole life financial profile to cover a 100-year period identifying: the short term intervention that is unavoidable (minimum needs) the optimal solution required to manage the system to achieve its desired target condition (identified needs) For that purpose insight into whole life costing of each individual asset becomes essential. The results generated from SAMPs could directly feed into the Deterioration and Whole Life Cost project. For instance, cost categories from the SAMPs guidance, or costs resulting from actual SAMPs. On the other hand, results generated by the Deterioration and Whole Life Cost project could also be relevant to or feed directly into SAMPs. 28 Science Report Assessment and measurement of asset deterioration including whole life costing

37 4 Collation of available knowledge 4.1 Interviews The collation of available knowledge from current asset managers underpins the empirical approach to deterioration curves and whole life costing models. The project team interviewed staff from a variety of Maritime Local Authorities and Environment Agency staff in a number of areas and from both Asset Systems Management and Operations Delivery (Figure 4.1) North East Region Sefton Rhuddlan Warrington Nottingham Brampton North Norfolk BESL Waveney Hatfield Ipswich South Wales Feering Frimley Canterbury Blandford Exeter Arun Worthing Bournemouth Chichester Figure 4.1 Location of interviews Science Report Assessment and measurement of asset deterioration including whole life costing 29

38 Table 4.1 Details of interviews Contact Names Titles (if known) Region/Area 1 Andrew Newton 2 Georgina Nichols David Van Beesten Matt Graham Asset System Mgmt. Team Leader ASM - Technical Specialist Wey&Loddon Team Leader Asset Management FRM Asset Mgmt. Team Leader 3 Asset manager 4 John Claydon 5 Andrew Woodhead Asset Sys Mgmt. Team Member Ops. Del. Tech Support TL Anglian (Central) Brampton Thames Frimley Chichester District Council Feering South West Exeter Who interview RH HRW RH RH HRW 6 Morgan Wray Asset System Mgmt. TL 7 Russell Smith Richard Kershaw Jim Wreglesworth Peter Winter ASM Team member 8 Clive Jones Ops Del TL Mike Jones James West Ops Del Tech Support TL Tech Support Team member Midlands Nottingham Blandford, Dorset South Wessex Area Rhuddlan Depot North Wales RH HRW HRW 9 Brian Izzard Ops Del Tech Support TL Darsha Gill Graham Lymbery Paul Wisse Broadland Environmental Services Limited Ops Del Tech Support TL 13 Alan Cadas Asset Systems Manager Hatfield North/East area Sefton Borough Council North Norfolk District Council North East Region 14 Jim Warner Ipswich RH 15 Steven McFarland Canterbury HRW HRW PL PL PL HRW 16 Bryan Curtis Local Authorities Worthing HRW 30 Science Report Assessment and measurement of asset deterioration including whole life costing

39 Contact Names Titles (if known) Region/Area Who interview Glenn Longley 17 Roger Spencer Local Authorities Tony Davison Environment Agency Arun West Sussex- Hampshire HRW 18 David Harlow Bournemouth HRW 19 Waveney PL 20 Martin Cadogan Asset System Mgmt. Team Leader Wales, South Tim Rob Wilkins Asset System Mgmt. East Area Hopkins Technical specialist The list of questions developed and used to guide the semi-structured interviews with asset managers is presented below: Envisaged deliverables of the project {Ask about asset managers needs re. deterioration and whole life costing and talk about our envisaged outputs} 1. Would the suggested approach be useful for your work? Does it cover your most important needs re. deterioration and whole life costing? Do you have any other suggestions at this level? 2. Do you think the proposed approach of the research team is realistic and achievable? 3. Do you have a preference regarding the format of the R & D output (methods, guidance, tools, training)? General asset management 4. Describe and categorise the main assets types under your responsibility {having NFCDD/RASP categorisation available would be helpful} 5. Describe and categorise the maintenance regimes associated with these asset types (frequent, intermittent, asset replacement} Deterioration {Explain the difference between failure and deterioration, making use of the attached tables if helpful} 6. Describe typical ranges of loading conditions (waves, currents, etc.) that your assets experience. 7. To what extent do you think these loadings contribute to deterioration? 8. How frequent/regular were any loadings associated with deterioration? 9. What other environmental conditions also affect deterioration of your assets (e.g. saline conditions) 10. What deterioration processes have you observed? 11. What rates have you observed? {if no monitoring data, prompt consideration of periodic assessments or appraisals for replacement assets} 12. How do you measure / characterise condition? 13. Do you typically estimate residual life, and if so, how? How does this influence your maintenance decisions? 14. What failure processes have you observed dominating your different asset Science Report Assessment and measurement of asset deterioration including whole life costing 31

40 types? 15. Were you aware of any unusual or exceptional loadings or other processes which may have contributed to any failures? 16. Can you think of any specific structures which might typify the main deterioration and failure processes you know about? 17. Describe the maintenance regimes currently adopted. How do you think these affect deterioration, including rates of deterioration? Whole life costing {Explain the main cost categories that we distinguish, using conceptual framework} Ongoing maintenance 18. Describe typical maintenance activities; use conceptual framework as checklist, see if we are missing anything at this level 19. Looking at the maintenance regimes identified under Q5, are there different lists of activities for different regimes, or is there only a different frequency / intensity? 20. What are the main factors that determine the activities and the scale / thoroughness of activities (given a certain maintenance regime), e.g. asset type, accessibility. 21. What are typical frequencies of activities? (if possible as a function of the maintenance regime) 22. Staffing is likely to be a major cost element. Do you see a relation between staffing levels (e.g. per unit length) and asset types, maintenance regimes or any other factors? 23. Could you provide any actual cost information? Refurbishment: 24. Describe typical cases of asset refurbishment: asset type, intervention, frequency 25. Could you provide any actual cost information? Decommissioning: 26. Describe typical cases of decommissioning: asset type, lifespan 27. Could you provide any actual cost information? Participation in research 28. Would you be able/willing to provide further information from your records to assist the research? 29. Can you think of any structures which might be worthwhile monitoring in a structured visual manner on a regular basis? (PAMS visual inspection framework can be provided.) 30. Can you think of any existing or planned structures which could be monitored in detail with instrumentation/measurements? 31. Would you be willing for these structures/sites to become long-term monitoring sites for the research programme? 32 Science Report Assessment and measurement of asset deterioration including whole life costing

41 4.2 Summary of interviews Information collated from asset managers provides the following list of relevant asset types discussed in interviews: TYPE OF PROTECTION Brampton Frimley Chichester Fearing Exeter Nottingham FLUVIAL Earth embankments x x x x x x x x Flood /tidal gates x x x x Concrete walls x x x x Steel walls x Maintained channels (concrete) x x x Pumping stations x x x x x Storage reservoir x Culverts x x x x Brick&Masonry walls x x x x Steel sheet piles x x x COASTAL Earth embankments x x x x Shingle ridges x x Concrete sea walls x x x x x x x x x x x x x Steel sheet piles x x x Tidal barriers x x Watermills x Rock revetments x x x x x Rock groynes x x Timber revetments with steel pile toe protection x x Timber groynes x x x x x x x x Composite groynes x Concrete groynes x Sand and rubble beaches x Blandford The main loading actions described are: - wave action - tidal currents - scouring - abrasion from waterborne sand/stone - impacts of sand/gravel - corrosion - retained soil and water - surcharges Rhuddlan - uncontrolled surface groundwater flows - human activities - geotechnical instabilities Significant variations in the life of particular assets have been found around the country. This arises because of variations in the quality of the original Hatfield Sefton BESL North Norfolk North East Ipswich Caterbury Worthing Arun Bournemouth Waveney South East Wales Science Report Assessment and measurement of asset deterioration including whole life costing 33

42 construction, in the aggressiveness of the loadings and deterioration agents, and the degree and quality of maintenance carried out. The main deterioration problems described can be divided into two categories. First, factors related to the deterioration of the asset material: corrosion, impact, abrasion, etc. Secondly, factors related to instabilities of asset foundations: cliff failures, beach level degradation, etc. The deterioration work was originally focussed on identifying changes of condition grade with time, but processes not necessarily associated with structural condition change, such as global settlement, also need to be covered. The original focus of the work was on linear defences. However, whilst there is significant investment on the maintenance of embankments, the maintenance of culverts and channels is equally important. Maintenance of hard defences is generally restricted to minor repairs for example, repointing of brickwork, resealing of joints. Structural repairs of such defences could be improved by greater use of specialist contractors. Significant expenditure is incurred on assets to meet non-frm requirements, such as health and safety. These can be onerous and expensive, for example maintaining handrailing. There is significant scope to take advantage of and extend existing monitoring, such as the coastal regional monitoring programmes. Whilst national databases such as the National Flood Coastal Defence Database (NFCDD) may suggest that little data are held on many assets, interviews suggested that considerable information on assets is held in paper and electronic form in local offices. We hope to take advantage of back-analysis of this information in the next phase of work as well as setting up pilots A complete transcription of asset interviews can be found at the appendix. 4.3 Consultation with Environment Agency National Process team There were two meetings between the project team and the Environment Agency National Process team. Meeting Attendees from the Environment Agency National Process Team 25/01/08 Lyn Hensman, Richard John, Dave Dennes 20/10/08 Lyn Hensman, Richard John The following key conclusions emerged during these meetings: The characterising of performance (including deterioration) and costs of assets sometimes needs to be carried out at the system level rather than at the level of individual assets. 34 Science Report Assessment and measurement of asset deterioration including whole life costing

43 In the development of SAMPs some problems have arisen, particularly in establishing the actual condition grade of the assets, because surface condition/visual inspections cannot provide the complete story about change in asset condition. A clear interest for a guidance on deterioration curves with the requirement that it should help support decisions on selecting the most appropriate curve. There is a clear interest on whole life costing and how it relates to different maintenance options. As a short term deliverable, the work on whole-life costing needs to focus on supporting the existing SAMPs work. There are various ongoing developments related to the SAMPs initiative and whole life costing (finalisation of guidance, development of maintenance standards and unit costing). For the purpose of efficiency it was decided to await the outcomes of these developments within the Environment Agency, and develop a plan for Phase 2 based on the situation at the end of Phase 1. This plan is described in Section 6.3. Whilst all research commissioned by the Environment Agency must have a clear link to its business objectives for now or for the future, when defining the required outcomes of future research some of these will be focussed on meeting short-term business objectives and others aimed at longer term goals. The short-term business objectives can be met by coarser detail plus interim guidance. In contrast, the longer term goals will require focussed and detailed study based on long-term monitoring and research. A link with the actual guidance for SAMPs was desirable. From the point of view of maintenance works, the levels of maintenance (high, low, medium) depend on the deterioration processes. So, it is important to ensure the links between the ongoing projects on that topic. 4.4 Refined end user requirements The Environment Agency and local authority staff interviewed were identified as the end users of the outputs of the project. During the interviews, they showed a lot of interest in the overarching objective of the project. The main requirement of the end users was a guidance and/or a decision support tool to assess how different types of assets deteriorate over time and how that relates to maintenance management and costs. Training was another requirement of the end users. In general, there was considerable enthusiasm to support the initiative of setting up pilot sites for monitoring purposes in order to provide a reliable evidence base. Science Report Assessment and measurement of asset deterioration including whole life costing 35

44 5 Deterioration curves A separate document (SC060078/TR2) Guidance on deterioration curves has been developed. This details the deterioration curves and how to use them to quantify the residual life of different types of assets, including a step-by-step guide. It covers vertical walls, embankments, culverts, dunes and shingle beaches. The curves are based on the condition grades defined on the Condition Assessment Manual (Environment Agency, 2006). The deterioration curves consider the type of environment (fluvial or coastal), type of material, width of the asset, whether there is maintenance and whether there is rear protection. A table with deterioration rates for different conditions grades and different asset types that summarises information from deterioration curves is presented below. The time (in years) to move between different condition grades is obtained as the difference between figures corresponding to those conditions grades. The sources of uncertainty in detailing deterioration rates are multiple, as has been described in section 2.1. Uncertainty is related to the lack of knowledge of the physical processes (epistemic uncertainty), its natural variability (inherent variability) and to the complexity of the final objectives (decision uncertainty). It is currently difficult to identify these different sources and uncertainty has been considered in the deterioration rates of Table 5.1 based on the professional judgment and experience gathered through interviews. 36 Science Report Assessment and measurement of asset deterioration including whole life costing

45 Table 5.1 Deterioration times (years) to different condition grades for different asset types and exposures Science Report Assessment and measurement of asset deterioration including whole life costing 37

46 6 Approach for phase 2 This chapter describes the main goals and tasks identified for Phase 2 by the research team. This Phase is intended to set up the long-term programme of data gathering from pilot sites in Phase 3, and in addition, to produce a number of user-focused deliverables. The main overall goals, methodologies and work-packages are identified in section 6.1. In the subsequent sections, they are described in more detail. 6.1 Goals and methodologies of phase 2 The main goals of Phase 2 are: To improve the fundamental understanding of the processes of deterioration of common material and components in key flood and coastal risk management (FCRM) assets through a targeted data collection and monitoring programme. Wherever possible this will involve developing quantified or quantifiable models of deterioration processes; To investigate, further develop and test robust methods and models for assessing the deterioration of high risk FCRM asset types in order to provide improved estimates of deterioration and residual life; To develop and test robust cost models to establish criteria for determining the optimum and/or most cost-effective time and type of intervention in the deterioration process and their effects on WLCs; To provide practical updated guidance to ongoing flood and coastal defence asset management in relation to deterioration and WLC. To achieve these goals, it is recommended that a long-term programme of research based on monitoring and evaluation of selected pilot sites will be developed. A monitoring and data collation will commence in Phase 2 and it is recommended that this is extended into a long-term data collection programme (Phase 3), largely carried out by the Environment Agency and other operating authorities (or their contractors). The R & D project will also provide support to the further development of other projects such as: the follow-on to the PAMS Phase 2 project; the PAMS Phase 2 will soon recommend a programme of follow-up research and the R & D recommended in this chapter should be seen as an integral part of that programme; the forthcoming new Environment Agency AMIT, by suggesting which data on asset deterioration should be included in the AMIT; ongoing projects related to maintenance works; the R & D project could provide more practical support on developing the decision-making criteria for operations delivery managers when choosing a particular maintenance standard and estimating the whole-life ownership costs. The proposed work packages for Phase 2 are: WP 1. Developing and conducting a focussed data collection and monitoring programme (0-36 months) 38 Science Report Assessment and measurement of asset deterioration including whole life costing

47 WP 2. Developing and testing robust methods and models for assessing asset deterioration of key FCRM asset types (12-24 months) WP 3. Developing and testing robust methods and models for assessing WLCs under different maintenance regimes for selected asset types (12-24 months) WP 4. Developing improved guidance on determining asset deterioration and assessing the effects of different maintenance regimes (24-36 months) 6.2 WP 1. Developing and conducting a focussed data collection and monitoring programme Task 1.1: (0-6 months) To develop a targeted data collection monitoring programme for key FCRM assets which benefit from an improved understanding of asset deterioration under different maintenance regimes, forcing conditions, condition grades, etc. The outputs of this sub-task will be: (a) Detailed list of asset types requiring improved understanding of deterioration (and under which forcing conditions, maintenance regimes, condition grades ) A list of possible pilot sites has been assembled from data collated during interviews with asset managers and should form a useful starting point. Type of asset Coastal/ timber and rock groynes Coastal/timb er groynes Fluvial/Emb ankment and Sheet piling Fluvial/Emb ankment Condition Grade New structures Construct ed in the 70s and 80s Old and new New schemes Area/Region Littlehampton/ Arun DC/ West Sussex - Herne Bay Groynes - Studhill (between Herne Bay and Tankerton) - Whistable / Canterbury Exeter/ South West Region Downton, Fordingbridge& Harnham(EA Bridport) Contact Details Name Phone Roger Spencer Steven McFarland Andrew Woodhead Fluvial Hatfield Brian Izzard Fluvial/Emb ankment Old Worthenbury/ Rhuddlan Steve Morris Roger.Spence [email protected] k Steve.McFarla nd@canterbur y.gov.uk andrew.woodh ead@environ mentagency.gov.uk brian.izzard@ environmentagency.gov.uk steve.morris@ environmentagency.wales. gov.uk (dir. line) Observations Arun DC scheme for 4 new groynes Contact has changed. Possible to apply different maintenance options to some embankments Back analysis of existing records It s no longer required and could be allowed to deteriorate without Science Report Assessment and measurement of asset deterioration including whole life costing 39

48 Type of asset Fluvial/Emb ankment Demountabl e barriers Condition Grade Construct ed in 2006 Area/Region Almere (near Wrexham)/ Rhuddlan Rhuddlan Contact Details Name Phone Steve Morris Steve Morris Sand dunes Sefton BC Graham Lymbery Coastal/Con crete seawalls Coastal/Con crete seawalls Coastal/Con crete seawalls Fluvial/Earth Banks Fluvial/Earth Banks Clay Walls/Earth Embankmen ts Steel sheet piles Embankmen ts steve.morris@ environmentagency.wales. gov.uk steve.morris@ environmentagency.wales. gov.uk Graham.Lymb [email protected] 2 Sheringham Peter Frew [email protected] 4 Sheringham Peter Frew [email protected] 3 Declining Overstrand Peter Frew [email protected] 2-5 Norfolk Broads Bob Lancastle 2-5 EA North East Alan Cadas Old Essex John Claydon Construct Essex ed between the 60 s and 80 s Old Midlands Fenland St Neots and St Ives Jim Warner Georgina Nichols Andrew Newton besl@edmund -nuttall.co.uk alan.cadas@e nvironmentagency.gov.uk - agency.gov.uk Jim.warner@e nvironmentagency.gov.uk georgina.nich ols@environm entagency.gov.uk Andrew.newto n@environme ntagency.gov.uk Observations maintenance 3m high, 2.5m wide made in aluminium 13 year old walls 113 year old walls 55 year old failing beach New to failing banks. Very willing to cooperate Enthusiastic to cooperate Very willing to cooperate Very willing to cooperate Maintained channels (concrete) Various Leicester Morgan Wray - agency.gov.uk EA Nottingham happy to provide more details on locations and willing to discuss sitespecific details + approach to monitoring of these pilot sites. 40 Science Report Assessment and measurement of asset deterioration including whole life costing

49 Type of asset Embankmen t/walls/shee t piles/revet ments/groyn es Walls Condition Grade completed in 1970s Area/Region Treforest, Taff Valley Contact Details 30 years old Usk Town Martin Cadogan Flood walls Rhonda Valley Martin Cadogan Embankmen ts of polders Complete d 2005 Martin Cadogan Rumney Great Wharf Wentloog foresheo Name Phone Martin Cadogan Observations Polders used for foreshore replenishment (b) Collection (and analysis) of historic data and information on asset-specific deterioration The project should not just consider new pilot sites but also take advantage of and leverage on existing monitoring such as regional coastal monitoring programs and the Environment Agency s embankment vegetation trial sites. The project should also consider carrying out a desk study to back-analyse existing information held in paper and electronic form in local offices on different assets. (c) Detailed monitoring programme using a risk-based approach (which assets, where, how, by whom, how often, to what level of detail, which data/parameters to be collected, how to be stored ). This programme should: - Seek synergies and efficiencies with on going monitoring through Environment Agency, local planning authorities and Coastal Observatories where possible; - Should make use of historic monitoring information which may already be available; - Propose different levels of monitoring depending on risk and need for more information; - Reflect, where relevant, the approaches adopted by other national asset managers such as Network Rail, Highways Agency, British Waterways, Rivers Agency and relevant Dutch authorities The monitoring plan for the potential sites is envisaged to comprise of three tiers: 1. All sites would have regular visual condition assessments using the new PAMS methodology. The PAMS methodology involves inspection and condition grading of individual asset features (rather than the asset as a whole) with an agreed algorithm for combining the results. The regularity, detail and recording of the condition assessments required for the research monitoring would be in excess of that which is currently carried out as part of routine inspections by Operations Delivery teams. Science Report Assessment and measurement of asset deterioration including whole life costing 41

50 2. Some of the sites will be selected to have physical surveys of levels of structures and surrounding ground (e.g. beaches). It will be based on a thorough baseline survey of all selected sites and re-surveys at agreed tiers at regular intervals. Some extra surveys will be carried out after an extreme flood/storm event. In coastal surveys of beaches, surveys are recommended to be at 100-m intervals to match the approach of the Coastal Regional Monitoring Programmes. Some data on excitation parameters, including loadings (e.g. water levels waves etc) and deterioration drivers (erosion, temperature, salinity etc.) should also be collected. 3. At a few of the selected sites there should also be monitoring of physical parameters (involving installation of appropriate equipment as necessary) plus a more spatially dense survey. At these detailed monitoring sites (as a minimum) a more detailed process understanding of deterioration will be developed. All relevant data on excitation parameters, including loadings (e.g. water levels waves etc) and deterioration drivers (erosion, temperature, salinity etc.) will also need to be collected. The monitoring scheme will be designed specifically to each type of structure. The rich range of resources of potential monitoring approaches available, such as sensors, electromagnetic techniques, etc, should be considered. If an automated monitoring scheme is planned, the use of predictive models will also be needed. Monitoring programmes at pilot sites may be able to make use of data on excitation parameters already being collected, for example in regional coastal monitoring surveys or from local asset managers. Equally these survey programmes could potentially be extended to include some of the physical measurements required at the pilot sites. (d) Web portal to make the data collection/baseline survey/monitoring results available to selected stakeholders and to aid communication and dissemination activities. The website of the Channel Coastal Observatory ( see screenshot below is presented as example of how to disseminate and store information. 42 Science Report Assessment and measurement of asset deterioration including whole life costing

51 Task 1.2 (6-36 months): To set up/manage the data collection and targeted monitoring programme. This activity will support data collection and review the needs of ongoing developments on deterioration and whole life costing in relation to data. The research team strongly advocates the principle of long-term monitoring because changes in asset conditions will be slow and the real gains from monitoring will take several years to emerge. They recognise that periodic reviews will be required because of the need to reassess the needs of the business against the emerging data and thus there may be a need, for example, to increase or decrease certain types of monitoring, to add new monitoring types and/or to add or delete structures from the monitoring programme. On this basis it is recommended that the monitoring programme is re-assessed at the end of Phase 2 (i.e after 2-3 years of work). Thereafter, following the principle accepted for the South Coast Monitoring Programme, it is recommended that the monitoring is reassessed every 5 years. The outputs of this sub-task will be: Archived and quality checked monitoring data (or meta data depending where stored) Yearly summary reports outlining the key findings and progress made 6.3 WP 2. Developing and testing robust methods and models for assessing asset deterioration The main objective of Phase 2 related to deterioration activities is to investigate the processes of deterioration and to develop and test robust (i.e. operationally implementable) methods and models for assessing the deterioration of important FCRM assets (likely to be high risk) in order to provide improved estimates of deterioration and residual life. Science Report Assessment and measurement of asset deterioration including whole life costing 43

52 Improved understanding of deterioration of key assets could also benefit broadscale tools used for asset investment and planning (e.g. NaFRA). Task 2.1: Analysis of field data Analysis of field data about deterioration processes and factors from pilot sites and information already held in local offices. The aim is to support the understanding of deterioration and the development of new models based on real data. Task 2.2: Evaluation of available existing results from physical tests on the deterioration of FCRM assets This activity is to determine whether there is a need for more physical testing of actual asset conditions under certain forcing conditions. This activity could be seen as being analogous to the approach being adopted in The Netherlands with the full scale testing of smart dikes ( Ijkdijks ) and thus the outcome of this sub-task may be to suggest future international collaboration. Task 2.3: Development of models for deterioration processes and their impact on asset performance Further development of quantified or quantifiable models (process based, empirical or neural networks) for deterioration processes for a range of asset types and components representative of those held by the Environment Agency and other operating authorities. They will be based on the understanding of the deterioration processes and linked to an understanding of their impact on the likelihood of structural failure in relevant failure modes and the resulting overall reliability of these structures. The goal of the task is to develop an improved and more transparent methodology to determine rates of deterioration, and to predict the amount of deterioration that might take place over a fixed timescale and allow this to be used in assessments of the relevant future defence reliability. The tools developed need to be robust enough to be user operationally and to offer clear benefits compared to approaches currently used. Operational software implementation will not be required at this stage as the models are likely to be fed into the forthcoming Environment Agency asset management systems. Task 2.4: Develop deterioration curves related to different maintenance standards The existing deterioration curves only take into account whether maintenance exists or not. The goal of the task is to cover the needs expressed by the Environment Agency s National Asset Management team to link deterioration and cost to a range of maintenance standards. More evidence is needed on when to intervene (e.g. through capital projects) in the maintenance regime for FCRM asset types to achieve or maintain a certain condition. The goal is to increase consistency between the maintenance regimes and their costs for comparable assets in different regimes/areas. This activity is closely related to WP3 related to whole life costing. The main outputs of the work package are: Tested methods/models for selected asset types Technical report outlining the developed methods and models considering their practical benefits, data requirements and limitations 44 Science Report Assessment and measurement of asset deterioration including whole life costing

53 6.4 WP 3. Developing and testing robust methods and models for assessing whole life costs under different maintenance regimes for selected asset types The tasks envisaged in this work package will support asset management in the short and medium term: Task 3.1 Test those existing WLC methods/models developed under other research programmes (e.g. FRMRC) which are robust enough to be used operationally and have clear benefits for operational asset management. Task 3.2 Development of whole life cost profiles for a set of maintenance standards In the short term, the Environment Agency s National Asset Management team has expressed a need to develop WLC profiles related to the maintenance standards that have recently been developed for the Environment Agency s Operations Delivery teams (see Maintenance Standards and Unit Costs booklets for embankments and for channels Environment Agency internal document, November 2008). WLC information for asset management is required input for SAMPs; the SAMPs guidance and also the Maintenance Standards booklets currently provide cost information, but these require validation and also extension to other asset types. The project will require a clear definition of the maintenance standards for selected FCRM asset types. In principle, this can be based on the descriptions in the booklets (for embankments and channels), but it may need further development for other asset types, in conjunction with related ongoing work from the Environment Agency s National Asset Management team. The project will make use of the information collated in the development of the booklets (mainly from the Environment Agency s 1B1S system), but it is expected that the main source of the costing will be an engineering assessment of time, plant and material requirements. In addition, the information identified in the interviews in this project will be used. Finally, an important source of costing information could be provided by the cost capture pilot that will be started by the Environment Agency s National Process team in April Task 3.3 Development of further methods/models for assessing the WLCs of managing key FCRM assets The aim of this task is to develop a simple tool to allow asset managers to calculate WLCs as a function of asset type and maintenance regime (as incorporated in the maintenance standards in task 1), but also as a function of asset-specific characteristics and local knowledge. This tool can make use of standard whole life costing tools, for example, elements of the tools at the basis of the Environment Agency s AMIT programme, or possibly some of the in-house tools that Royal Haskoning have developed in The Netherlands (ACCRES ). Whilst this task is not required to produce operational software, these existing tools may need to be customised to a proof-of-concept stage to demonstrate their application to the specific purpose of flood defence asset management. The work in Phase 2 will build on the consultation with users during hase 1, and include further consultation to determine user requirements. Given that the National Capital Programme Management Service (NCPMS) is using risk budgets in its assessment of capital costs, it may be appropriate to include some risk components in the WLC assessments. The aim of the tool will be to support assessments, not to make decisions. It should be pragmatic and leave room for local judgement. The main outputs of the work package are: Tested methods/models for a number of high risk asset types Science Report Assessment and measurement of asset deterioration including whole life costing 45

54 Technical report outlining the developed methods/models and their practical benefits, data requirements and limitations 6.5 WP 4: Developing improved guidance on determining asset deterioration and assessing the effects of different maintenance regimes The main goal of this work package is to improve and expand on the existing Guidance on determining asset deterioration and the use of condition grade deterioration curves. This activity will Include FCRM assets which are currently not covered or where improved information on deterioration became available Include improved deterioration curves for specific forcing conditions, maintenance regimes etc. Include guidance on how to perform a more detailed assessment of deterioration for high risk asset types based on WP1 and 2 outputs Include key findings from WP3 whole life costing To support these works and in relation with WP1 a data collection and monitoring protocol will be developed to allow for the extension of the monitoring activities. The protocol should outline which core FCRM assets would benefit from further monitoring and what information should be collected at what intervals. The outputs will be: Revised Practical Guidance on Asset Deterioration and WLC for FCRM Assets (Draft and final) Data collection and monitoring protocol 46 Science Report Assessment and measurement of asset deterioration including whole life costing

55 7 Conclusions 1. The research project has developed a general framework to improve the fundamental understanding of deterioration processes and maintenance costs. 2. Available knowledge was collated from Environment Agency areas, from both Asset System Management and Operations Delivery, and from a variety of Maritime Local Authorities about asset deterioration at the individual asset level and at a system level (for example, systems of groynes). Availability of information about costs has also been identified from interviews. 3. The project has produced Guidance on determining asset deterioration and the use of condition grade deterioration curves. This guide details the deterioration curves and how to use them to quantify the residual life of different types of assets. Vertical walls, embankments, culverts, dunes and shingle beaches are covered. These curves will be useful for guiding Environment Agency areas and local authorities in preparation of asset management plans. Changes in the dominant deterioration processes at different times in composite assets imply the use of limiting values of deterioration curves. Examples of how this should be considered are provided in the guide. 4. The understanding of deterioration processes has been improved by the use of data collated both during the interviews and from the literature review, identifying the relevant failure mechanisms and key variables for a number of assets. 5. As far as whole life costing is concerned the project has identified ongoing developments in state of the art science, which confirm the strong interrelationship with the asset deterioration element. The project has concluded that methods and tools for flood defence asset whole life costing can be developed in the next stage: the general concepts from asset management science can be applied, and examples from other fields of asset management can be used as examples. This will require translation to the specific concepts and language of flood defence asset management. The project has found that there are various highly relevant ongoing developments in the Environment Agency s asset systems management (such as SAMPs and AMIT), where some of these specific concepts are being developed. Hence, any work on whole life costing needs to be fully embedded in those developments. The project has developed a plan for the development of whole life costing methods and tools in Phase 2 which would support flood defence asset management in the short, medium and long term 6. The project has developed the conceptual framework of how deterioration and whole life costing relate and feed into asset management decisions, as part of the wider asset management framework (which also includes the benefits of asset management). 7. The project has identified the main goals and activities required during Phase 2. These include activities related to deterioration, whole-life costing and the setting-up of a long-term monitoring programme to facilitate the gathering of information on asset deterioration and the cost-effectiveness of different maintenance strategies. Monitoring activities (which will need to carry on beyond the end of Phase 2) will provide the required evidence base on Science Report Assessment and measurement of asset deterioration including whole life costing 47

56 deterioration and its relation to maintenance approaches. Phase 2 of the project will include measurement and understanding of deterioration processes allowing the creation of more reliable and site specific deterioration and fragility curves. It also could provide a vehicle within ongoing projects such as PAMS for understanding the impact on flood risk of adopting various maintenance scenarios. 48 Science Report Assessment and measurement of asset deterioration including whole life costing

57 References ARYA, C., VASSIE, PR. (2004) Whole life cost analysis in concrete bridge tender evaluation, Proceeding of the Institution of Civil Engineers, Bridge Engineering 156, Issue BEI, pp BLANCHARD, B.S., and FABRYCKY, W.J. (1990). Systems engineering and analysis. 2nd edition Prentice-Hall, Incl., Englewood Cliffs, N.J. BSI, (1997), Reliability of systems, Equipment and Components Guide to Life Cycle Costing, BS 5760 Part 23, BSI, London. BUIJS, F.A. (2007) Time-dependent Reliability Analysis of Flood Defences, PhD Thesis, Newcastle University, UK CIRIA/CUR/CETMEF (2007) The Rock Manual. C683, CIRIA, London El-HARAM, M.A., MARENJAK, S. and HORNER, M.W. (2002), Development of a generic framework for collecting whole life cost data for the building industry, Journal of Quality in Maintenance Engineering, Vol. 8 No. 2, pp ENVIRONMENT AGENCY (2004) Reducing the risk of embankment failure under extreme conditions: Report 1-Good practice review, Defra/EA project FD2411 ENVIRONMENT AGENCY (2006) Managing Flood Risk. Condition Assessment Manual. Document Reference 166_03_SD01 FLANAGAN, R., and NORMAN, G. (with FURBUR, JD.) (1983), Life cycle costing for construction. Quantity surveyors Division of the Royal Institution of Chartered Surveyors, London, HONG, T., HAN, S. and LEE, S.(2007) Simulation-based determination of optional lifecycle cost for FRP bridge deck panels, Automation in Construction Vol. 16, pp HR WALLINGFORD (2002) Risk, Performance and Uncertainty in Flood and Coastal Defence A Review. DEFRA/EA Report no. FD2302/TR1 GOULDBY, B., SAYERS, P., MULET-MARIT, J., HASSAN, M. and BENWELL, D. (2008). A methodology for regional-scale flood risk assessment. Proceedings of the institution of Civil Engineers Water Management, Vol. 161 (3), pp NOORTWIJK, JMV. and FRANGOPOL, DM.(2004) Two probabilistic life-cycle maintenance models for deteriorating civil infrastructures, Probabilistic Engineering Mechanics, Vol. 19, pp OFFICE OF GOVERNMENT COMMERCE (2003) Whole-Life Costing and Cost Management, Procurement Guide Number 07: Achieving Excellence in Construction, Office of Government Commerce, London ROGERS, J., DORA, J., AND HUTCHISON, Z. (2004). Whole life costing for flood and coastal defence. 39th Flood and Coastal Management DEFRA conference, pp SALTELLI, A., CHAN, K., MARIAN SCOT, E. (2000) Sensitivity analysis, John Wiley & Sons Inc SIMM, J. and MASTERS, N, (2003) Whole Life Costs and Project Procurement in Port, Coastal and Fluvial Engineering How to escape the cost boxes, HR Wallingford Science Report Assessment and measurement of asset deterioration including whole life costing 49

58 THORSTENSEN, T.A., AND RASMUSSEN, M. (1999). A cost model for condition based overhaul/replacement. Journal of Quality in Maintenance Engineering, Vol. 5 (2), pp VAL, DV. And STEWART, MG. (2003), Life-cycle cost analysis of reinforced concrete structures in marine environments, Structural Safety, Vol. 25, pp Science Report Assessment and measurement of asset deterioration including whole life costing

59 Appendix: Reports of interviews Date: 19 / 5 / 2008 Venue: Brampton Participants: Georgina Nicholls (Environment Agency) Andrew Newton (Environment Agency) Marit Brommer (Royal Haskoning) Output R & D project When discussing the overarching objective of the R & D project a lot of interest has been shown, specifically on providing guidance and/or a decision-support tool aimed at assessing how different types of assets deteriorate over time and how that relates to maintenance and costs. 1. What is an asset All natural and physical structures protecting land from flooding. 2. Asset types The dominant asset types are: Fluvial 1. Earth embankments 2. Flood gates (and associated hard structures) 3. Concrete walls Coastal (Tidal defences) 1. Earth embankments 2. Shingle ridges 3. Concrete sea walls 3. Deterioration The main problem with the fluvial earth embankments is the susceptibility to erosion of the (sub-) surface of the embankment due to the river (scour/floods) and human impacts (tractors). Animals have also been noted as a cause of erosion due to grazing (sheep and cows). The level of erosion is not always visible. Deterioration rates of the embankments are not uniform, and may therefore be hard to predict, due to differences in a. shape of the embankment; b. age of the embankment (range in age from 20 years to >100 years); c. content (clay/peat); d. site specific factors (vegetation, animals, etc.) The floodgates and associated concrete structures deteriorate because of their age. Most of the gates are 40 to 50 years old. Some of the floodgates are newer (Kings Lynn, 20 years). Painting of the gates helps mitigate deterioration, however, in general these assets are considered as stable, and no active maintenance regime is provided for. Science Report Assessment and measurement of asset deterioration including whole life costing 51

60 The coastal earth embankments (seawalls) may be impacted by severe weather such as storms, but are nonetheless classified as stable. Most of the embankments have associated concrete revetments, and do not require significant maintenance. The shingle ridges are semi-natural : they provide a first line of defence. Some of the shingle ridges have been replaced by hard structures due to changes in the standard of protection. Deterioration of the shingle ridges mostly due to severe weather (increased water levels). 4. Maintenance Fluvial earth embankments Annual inspections of the earth embankments 1. Grass cutting 2. Vermin control 3. Surveys (visual inspection) Flood gates Annual inspection takes place. Painting of the gates helps mitigate deterioration, however, in general these assets are considered as stable, and no active maintenance regime is provided for. Coastal earth embankments Annual inspections of the earth embankments 1. Grass cutting 2. Visual inspection Responsive inspections 1. After a storm the embankment is checked. 5. Whole Life Costing There are hardly any data on the costing aspect of maintenance. We have been referred to Mr Greg Murphy (Ely office) who might be able to deliver some financial details. Historic information on costs should be stored somewhere, most likely hard copies in Environment Agency libraries. 6. Pilot Sites There is considerable enthusiasm to support the initiative of setting up pilot sites for monitoring purposes. Embankments in the Fenland have been suggested as suitable monitoring sites, for example St Neots and St Ives. 52 Science Report Assessment and measurement of asset deterioration including whole life costing

61 Date: 21 / 5 / 2008 Venue: Swift house, Frimley (Environment Agency Thames South West Area) Participants: David Van Beesten (Mole, Wey & Loddon Asset Manager) Matt Graham Jonathan Simm (HRWallingford) Marta Roca (HRWallingford) Asset types The main assets of the area are culverts, trash screens (224), pumping stations, drainage systems (dikes/ditches), weirs, marsh dikes. Key rivers in this eastern part of this Environment Agency area are the Ravensbourne and the Quaggy Eighty-five percent of reaches are urbanised, channels are concrete-lined or culverted. In general there are not many raised defences in the area. Flood defence is provided by channel conveyance. For example, the Marsh Dikes catchment is dominated by a series of drainage dikes and ditches, collected at pumping stations. On the Quaggy River there are raised defence walls and three flood storage areas. The raised defences on the tidal Thames are managed by a separate unit of the Environment Agency. The non-tidal stretch of the Thames from Teddington to Reading is dealt with by this area office. There is some stone pitching on the tidal Thames. Suggested to talk to Andy Batchelor about this. Deterioration and lifetime Weirs are in pretty good condition. There are 47 weir sites between Teddington and St John s Oxford (Lechlade). Under the Thames weirs strategy, there is a rolling (?) 100- year plan in which all weirs are replaced once. Culverts do not need maintenance as a structure. Some vegetation management is performed in the banks and ramps of the entrances. The life of concrete ring or brick culverts could be 100 years. There is an example of an 800-m long culvert built in 1924 that is still in excellent condition. Other examples are 100-year-old culverts under track rails. There are not any culverts with condition grade 4 or 5. Newer culverts are typically constructed using concrete rings. There are some reaches of the Wey River with very acidic waters that can reduce the life of structures by around 10 years. In this area, sheet piling has a lifetime of years. The only place where this does not hold is in parts of the River Wey where the high level of acidity causes more rapid corrosion. Different structures, gates, mechanical instrumentation, etc from the1930 s are still working. Some gates have been replaced or blast cleaned, repaired and reinstalled. For example between 1990 and 2000 all the weir gates on the River Wey were replaced Science Report Assessment and measurement of asset deterioration including whole life costing 53

62 There could be some problems of cracking in concrete open-channels that become worse with frosts. Estimating residual life is a challenge. Maintenance works Typical maintenance works are channel clearance or dredging (with navigation purposes). Dredging has now become very expensive because of the costs of landing and disposing waste and as a result is rarely done annually. For example 40,000 tonnes of sand were recently taken out of the River Wey at a cost of 30 to 120 per tonne. Dredging is now carried out strategically (for example to remove shoals on bends of the River Thames) Annual basis oiling and greasing of moving parts dredging (just in 1-2 watercourses) grazing removing of debris and rubbish accumulated in the channel (scavenging) grass-cutting vegetation clearance grass cutting cleaning and repairing more critical screens 10 years tree management on either side of channels of main river or on raised defences There are no painting activities Pointing has been done after 30 years of life of structures. 54 Science Report Assessment and measurement of asset deterioration including whole life costing

63 Date: February / 2008 Venue: Participants: Chichester Disctric Council Royal Haskoning Introduction The answers given to the questions below are based upon Royal Haskoning s long term experience of this frontage, their current partnering role with the operating authority (Chichester District Council) and a review by the Council s asset manager. Royal Haskoning have also drawn upon some of their general experience in coastal engineering. The answers mainly focus on matters of structural integrity. Issues relating to hydraulic performance are usually addressed when undertaking periodic refurbishment or replacement. Non-structural, non-hydraulic issues such as health and safety are often dealt with as part of structural integrity. Envisaged deliverables of the project 1. Would the suggested approach be useful for your work? Does it cover your most important needs re. deterioration and whole life costing? Do you have any other suggestions at this level? A manual (see Question 3) on deterioration would be helpful and could join a suite of similar manuals (e.g. Manual on the use of timber in coastal and river engineering). A manual on whole life costing already exists and therefore any new work in connection with deterioration should be linked to this publication. Also the TE2100 project developed some useful whole life costing methods for flood and coastal defence structures. Rates of deterioration are dependent upon a number of factors (see Question 11) many of which (e.g. maintenance) are under the control of the operating authority. Therefore there is no entirely natural deterioration rate but instead it is more a matter of policy as to how long an asset lasts based upon cost and risk management decisions. The project should address all of the above factors. In connection with maintenance there needs to be clear definitions in respect of structural integrity, hydraulic performance and ancillary matters, and frequency, scale and procurement (see Introduction and Question 17). 2. Do you think the proposed approach of the research team is realistic and achievable? More emphasis should be given to researching archive records. A potential data source is a project undertaken by The Local Government Technical Advisers Group (TAG) in June 2000 on Coastal Defence Best Value. The project was led by Terry Oakes who at that time worked for Waveney District Council and who now works as an independent consultant. Operating authorities should be asked specifically about assets with particularly long or short life spans, and the reasons for this. The NFCDD/RASP categorisation system with 61 No. classes is very basic and needs to be used with caution. The TE2100 project looked at potential developments of this system. 3. Do you have a preference regarding the format of the R & D output (methods, guidance, tools, training)? Science Report Assessment and measurement of asset deterioration including whole life costing 55

64 A manual style would be helpful that provides general principles, case histories and useful data sets. It should not be too prescriptive and should be aimed at practising engineers. General asset management 4. Describe and categorise the main assets types under your responsibility {having NFCDD/RASP categorisation available would be helpful} The coast protection assets generally comprise shingle beaches controlled by timber groynes with a seawall back defence. The majority of the seawalls are constructed in concrete with a steel sheet piled toe. However, there are also significant lengths of timber seawalls (breastworks). There also exists some rock revetments that are combined with seawalls, some lengths of gabion wall and some sand dunes but these are not considered further in any detail. The NFCDD/RASP categorisations are as follows:- - Shingle beaches 38 - Timber groynes None - Concrete seawalls 26 - Timber seawalls None 5. Describe and categorise the maintenance regimes associated with these asset types (frequent, intermittent, asset replacement} Shingle beaches intermittent. Mainly involves the adjustment of groyne planking heights to influence beach movements. There is also a need for beach recharge but this is not taking place at present due to a lack of funding and uncertainty over the long-term coastal defence strategy. Timber groynes frequent. Mainly involves the replacement of missing elements such as planking and sheet piles. Concrete seawalls intermittent. Mainly involves periodic encasement works and/or the addition of toe rock revetments. Timber seawalls intermittent. Mainly involves the replacement of missing elements such as planking. Deterioration 6. Describe typical ranges of loading conditions (waves, currents, etc.) that your assets experience. Typical 1 in 1 year near shore wave heights range between 1.2 and 2.2 m. Tidal currents of up to 2.5 knots can occur at Selsey Bill. Dominant wind direction from south west. Isle of Wight provides some shelter to western side of the Peninsula. Shingle beaches are highly mobile resulting in significant abrasion of structures and significant differential beach levels across groynes. Typical maximum tidal range of 4 m. 7. To what extent do you think these loadings contribute to deterioration? Wave action and beach movement (ref: abrasion and loading) probably give rise to the majority of the asset deterioration. 8. How frequent/regular were any loadings associated with deterioration? Significant wave action and/or beach movement probably occurs more than 50% of the time. 9. What other environmental conditions also affect deterioration of your assets (e.g. saline conditions) 56 Science Report Assessment and measurement of asset deterioration including whole life costing

65 Marine borers (e.g. gribble) are believed to be present but problems generally avoided by use of resistant timbers (e.g. Greenheart). There is a potential for accelerated low water corrosion of steel sheet piles but no evidence has been found as yet. The regular wetting and drying of timbers as a result of tidal action adds to their general ageing / wear and tear. The saline conditions promote the corrosion of steel elements and components. Fungal decay (e.g. wet rot) can occur in some timbers, particularly on the buried face of timber seawalls, but the use of resistant timbers (e.g. Greenheart) helps to reduce the problem. Vandalism (e.g. barbecue fires against a timber structure). 10. What deterioration processes have you observed? Undermining of structures due to beach and/or sub-strate lowering. Reduced penetration of piles, etc. into substrate due to lowering of substrate level. Rotation of seawall and groyne toe piling. Leaning / partial overturning of structures (e.g. groynes). Lowering of beach levels. Loss of beach material. Loss of fill material from behind and/or under seawall. Outflanking of groyne inner ends. Loss of components (e.g. groyne planking). Loose components (e.g. groyne planking). Reduced section sizes due to abrasion. Exposed reinforcement due to abrasion of concrete. Reduced size of components / fixings due to abrasion. Rounding of armourstone due to abrasion. Settlement of armourstone. Broken gabion wires due to abrasion. Loss of gabion stone filling. Perforations in structures due to abrasion and fires. Gaps at interfaces due to abrasion (e.g. between groyne planks). Corrosion of steel components. Fungal decay of timber components. Warping / twisting of timber components (e.g. groyne planks). 11. What rates have you observed? {if no monitoring data, prompt consideration of periodic assessments or appraisals for replacement assets} Rates of deterioration depend upon:- - Degree of exposure to imposed loadings - Vulnerability to environmental conditions - Macro-design of asset (size, materials, configuration) - Micro-design of asset (detailing) - Amount and timing of maintenance - Interaction between complementary assets For best case situations the rate of deterioration is very low with potential life spans in excess of 100 years For worst case situations the rate of deterioration is relatively high with effective life spans of less than a few decades Shingle beach - 1 to 100+ years Hardwood timber groyne - 10 to 50 years Concrete seawall - 20 to 100+ years Hardwood timber seawall - 20 to 50 years 12. How do you measure / characterise condition? Condition Assessment Manual published by the Environment Agency. Science Report Assessment and measurement of asset deterioration including whole life costing 57

66 13. Do you typically estimate residual life, and if so, how? How does this influence your maintenance decisions? An estimate of the residual life of each structure is made based upon best judgement (see Introduction). No assessment is made of the residual lives of the individual elements that make up the whole structure. Maintenance decisions are mainly driven by the condition of individual elements rather than the overall residual life of the structure. 14. What failure processes have you observed dominating your different asset types? Shingle beaches lowering of beach levels Timber groynes - reduced penetration of piles - partial overturning - loss of components - reduced section sizes - gaps at component interfaces Concrete seawalls - reduced section sizes - exposed reinforcement Timber seawalls - loss of components - fungal decay 15. Were you aware of any unusual or exceptional loadings or other processes which may have contributed to any failures? As a general rule, no. However, a recent seawall breach at Selsey West Beach was attributable to an unusual connection detail between the apron and steel sheet piled toe. 16. Can you think of any specific structures that might typify the main deterioration and failure processes you know about? Shingle beach - Selsey West Beach Timber groynes - East Wittering Concrete seawall - Selsey West Beach Timber seawall - East Wittering 17. Describe the maintenance regimes currently adopted. How do you think these affect deterioration, including rates of deterioration? Maintenance falls under four categories:- 1. Routine maintenance involving relatively lightweight works; 2. Occasional major repairs involving heavier works usually in response to a significant failure or urgent problem; 3. Periodic refurbishment involving significant works to the existing structure; 4. Periodic replacement involving the construction of a like-for-like or similar structure alongside the existing structure (and the removal of the existing). Annual asset inspection by framework consultant identifies needs. Regular site visits by council staff confirms needs and identifies new issues or previously obscured issues. Council have annual budget set aside for routine maintenance. For occasional major repairs the council uses additional contingency funding and may seek, if appropriate, grant aid under the Coast Protection Act. For periodic refurbishment and replacement the council seeks grant aid under Coast Protection Act. 58 Science Report Assessment and measurement of asset deterioration including whole life costing

67 Framework contractor employed by the council for routine maintenance and possibly major repairs. Within budget and grant aid constraints the council seeks to deal with all maintenance issues as soon as reasonably practicable. The maintenance works have a significant impact on reducing the overall rates of deterioration. However, current cut-backs in funding are having an increasing detrimental effect on keeping pace with rates of deterioration. Whole life costing Ongoing maintenance: 18. Describe typical maintenance activities; use conceptual framework as checklist, see if we are missing anything at this level Routine maintenance - Adjustment of groyne planking levels - Replacement of missing components - Re-fixing of loose components - Re-wiring gabion boxes - Making good perforations Occasional major repairs - Timber works - Concrete works - Armourstone works - Earthworks (filling) Periodic refurbishment - New piling - Beach recharge - Partial reconstruction of groynes - Encasement of concrete seawalls - Partial reconstruction of timber seawalls - New rock toe revetments Periodic replacement - New groyne - Removal of existing redundant groyne 19. Looking at the maintenance regimes identified under Q5, are there different lists of activities for different regimes, or is there only a different frequency / intensity? See Question 17 and Question 18. Routine maintenance on a regular basis throughout the year Occasional major repairs typically once every few years Periodic refurbishment in the past, every few years, but now little or no work being undertaken. Periodic replacement in the past, every 5 or 10 years but now little or no work being undertaken. 20. What are the main factors that determine the activities and the scale/thoroughness of activities (given a certain maintenance regime), e.g. asset type, accessibility. The works undertaken and their degree of thoroughness are determined by the following: - Need (structural, hydraulic performance, health & safety) - Funding - Condition of parent structure - Residual life of parent structure - Importance of parent structure Science Report Assessment and measurement of asset deterioration including whole life costing 59

68 21. What are typical frequencies of activities (if possible as a function of the maintenance regime) See Question Staffing is likely to be a major cost element. Do you see a relation between staffing levels (e.g. per unit length) and asset types, maintenance regimes or any other factors? Staffing presumably relates to administration, obtaining funding, obtaining consents, design, CDM co-ordination and site supervision. Routine maintenance requires the least staffing levels, occasional major repairs usually require a relatively higher level of staffing, and periodic refurbish / replacement require full staffing levels. 23. Could you provide any actual cost information? Some cost records are reasonably accessible but these would still need some processing. Refurbishment: 24. Describe typical cases of asset refurbishment: asset type, intervention, frequency See Question 17, Question 18 and Question Could you provide any actual cost information? See Question 23. Decommissioning: 26. Describe typical cases of decommissioning: asset type, lifespan None exist. 27. Could you provide any actual cost information? No. Participation in research 28. Would you be able/willing to provide further information from your records to assist the research? Yes. 29. Can you think of any structures which might be worthwhile monitoring in a structured visual manner on a regular basis? (PAMS visual inspection framework can be provided.) Yes, there are plenty of classic structures that have easy access for monitoring purposes. 30. Can you think of any existing or planned structures which could be monitored in detail with instrumentation/measurements? See Question 29. Also there are a number of new works in various stages of planning that are awaiting funding. 31. Would you be willing for these structures/sites to become long-term monitoring sites for the research programme? Yes. 60 Science Report Assessment and measurement of asset deterioration including whole life costing

69 Date: 15 / 5 / 2008 Venue: Feering Participants: John Claydon (Environment Agency) Marit Brommer (RH) Output R & D project When discussing the overarching objective of the R & D project some worries have been expressed, specifically on the intention of modelling deterioration rates. Assets are considered to be too diverse to be captured in a model. 1. What is an asset All natural and physical structures protecting land from flooding. 2. Asset types Coastal (Tidal defences) 1. Earth embankments (Clay walls with 2:1 slopes (2 m crest width)); 2. Concrete sea walls (power concrete); 3. Steel sheet piles; 4. Tidal barriers (not discussed further for this project); Colne barrier Thames barrier 3. Deterioration Clay walls these walls do not deteriorate, they become stronger over the years. Steel sheet piles time is prime cause for deterioration, most of them were designed for 50 years (in the 1950s) and are still standing. Concrete sea walls these do not show any evidence of deterioration. 4. Maintenance Annual maintenance regime Clay walls Annual inspections of the clay walls 4. Grass cutting 5. Surveys (visual inspection) Responsive maintenance After heavy storms, clay walls will be checked, especially in areas where weaknesses are known. Annual maintenance regime Steel sheet piles Annual inspection takes place. Paint helps mitigate deterioration, however, in general these assets are considered as stable, and no active maintenance regime is provided for. 5. Whole Life Costing There are hardly any data on the costing aspect of maintenance. 6. Pilot Sites Science Report Assessment and measurement of asset deterioration including whole life costing 61

70 There is considerable enthusiasm to support the initiative of setting up pilot sites for monitoring purposes. Clay walls (with different ages and locations) have been suggested as good examples, however, care should be taken since these walls are strong and do not seem to deteriorate over time but have the tendency to become stronger over time (hence relevance to the R & D project on deterioration rates and WLC is questionable). 62 Science Report Assessment and measurement of asset deterioration including whole life costing

71 Data: 17 / 06 / 2008 Venue: South West Regional Office, Exeter Participants: Andrew Woodhead (Environment Agency) Jonathan Simm (HRW) Foekje Buijs (HRW) The interview took place on in Exeter with Andrew Woodhead, who works for the Operational Delivery team in the Environment Agency for Devon and Cornwall. In terms of recording the maintenance costs. all costs above 10,000 are booked to separate job numbers, while jobs smaller than 10,000 are booked to one job number. The AMIT is used, in combination with SWIPS. The area covers a large range of different flood defence structures. The River Ex has a reaction time of 6 to 8 hours to rainfall events. A large number of rivers in this area have flash flooding, with high debris loading, and there are tidal rivers as well. Along the River Taw in the west there are embankments with concrete walls. The Exmouth is protected from erosion by groynes. There are estuarine walls, manually operated gates, fixed defences, groundwater rivers and flash floods. For most of the rivers no measures have been taken to raise defences, except when there has been a flooding event. In Exeter in , radial gates were put into place, a side weir and broad banks more downstream. In Tiverton there is a 50-year piled defence in a fluvial environment. In Barnstaple there is ALWC, managed by David Turner, and ALWC occurs on sheet piles in Padstow. Nail gunning, painting and repair is carried out in fresh water in Exeter. Somerset levels is a separate system consisting of a drainage system and man-made defences, to the east of the River Parrett. The drainage system fills up with increasing water levels and the water levels stay high for months after filling up. High water levels are associated with some seepage and occasional defence failures. At Barnstaple, the reinforced concrete wall is on top of a stone and cement clad facing, at some locations there is slip occurring due to erosion. Repairs are carried out to support the base of the slip circle. Other deterioration occurs due to vermin infestation, and the growth of vegetation leading to a waterway and derooting causes damage to the surface. At Hale steelwork slag is applied (in which context?), the banks have not been tested in hydraulic boundary conditions. In Cornwall quite sandy banks with no cohesive core occur, there is no real erosion as the flows are not strong enough. The groynes at Exmouth are applied to control beach movement, consist of timber planks (60 / 70 years old, as opposed to years at Bournemouth??), and failure is due to wood splitting rather than to wood rot. Most of the 4.5 million of maintenance and intermittent maintenance finances and attention go to grass cutting on embankments, and, a smaller amount on channel clearing. Science Report Assessment and measurement of asset deterioration including whole life costing 63

72 Other activities are: Filling in gaps in revetments on banks or revetments, usually pitched stone revetment. Of those revetments gaps there are quite a few runholes due to vermin, as there is agricultural land behind those defences. There is damage due to recreational activity, footpath over the crest of the embankment. Reinforced Concrete walls regularly need replacement of joints. Culverts are monitored by CCTV inspection, 10% maximum siltation is allowed before removing silt. In Portreath a flood relief culvert is hewn through the rock and discharges 8 to 10 feet above the sea. Palporrow nearly had a culvert collapse due to a scaffold underneath a shop. Material was falling from a brick ceiling. Flap valves that are not working can introduce significant flooding. Frequent maintenance takes place 3 to 4 times per year. Intermittent maintenance takes place over longer and irregular intervals. It is possible to apply different maintenance regimes to some embankments with similar hydraulic boundary conditions and monitor the influence on the condition of the embankments. It is also possible to monitor some stretches of sheet piling. 64 Science Report Assessment and measurement of asset deterioration including whole life costing

73 Date: 28 / 5 / 2008 Venue: Nottingham Participants: Morgan Wray (Environment Agency) Richard Kershaw (Environment Agency) Russell Smith (Environment Agency) Marit Brommer (RH) Output R & D project A formal assessment of how different types of assets deteriorate over time combined to a WLC approach could be useful, since such evaluations are not known to have been consistently undertaken. Output of the R & D project should be focused on providing training and guidelines. 1. What is an asset All natural and physical structures protecting land from flooding. 2. Asset types Fluvial 1. Earth embankments; 2. Steel walls; 3. Storage reservoirs; 4. Maintained channels (concrete); 5. Pumping stations (not discussed further for this project). 3. Deterioration Earth Embankments Causes of deterioration- small animals such as rabbits and badgers (burrowing holes) and people (damage); Loading of water; Material of the bank (clay and peat), variable from site to site. Walls Weathering of the walls (~ 30 to 40 yrs old). Steel thickness and quality mainly determine weathering rates. Most of the walls, however, are in a good condition. Maintained channels Weathering due to damage. Storage reservoirs (dry reservoir) Reservoir act requires strict inspections by official reservoir engineers (1-2 times a year) and every 10 years an official panel reservoir engineer inspects the reservoirs. 4. Maintenance Annual maintenance regime Earth embankments Annual inspections of the earth embankments 6. Grass cutting Science Report Assessment and measurement of asset deterioration including whole life costing 65

74 7. Surveys (visual inspection) of damage and condition assessment. Steel walls Annual inspection of the steel walls to report their condition and potential damage. Concrete channels Not an active maintenance regime. Would be useful though. Responsive maintenance After heavy storms, embankments are checked, especially in areas where weaknesses/failure mechanisms are known. 5. Whole Life Costing There are hardly any data on the costing aspect of maintenance. Costs are not recorded against activities, however, if costs are stored (e.g. refurbishment costs) in a database we can certainly have access to those data. 6. Pilot Sites There is considerable enthusiasm to support the initiative of setting up pilot sites for monitoring purposes. Especially with regards to monitoring the maintained channels (concrete) and related maintenance regime (e.g. in Leicester), since hardly any knowledge/data exist on these assets. Embankments are also interesting as pilot sites, especially when different parts of one large bank can be monitored under different maintenance regimes such as: Do nothing; Maintain (normal); Maintain (super). Environment Agency Nottingham happy to provide more details on locations and willing to discuss site-specific details and approach to monitoring of these pilot sites. 66 Science Report Assessment and measurement of asset deterioration including whole life costing

75 Data: 17 / 06 / 2008 Venue: South Wessex Area Office, Blandford Participants: Jim Wreglesworth (Environment Agency) Peter (Environment Agency) Jonathan Simm (HRW) Foekje Buijs (HRW) The deterioration interview at Blandford took place with Jim Wreglesworth and Peter on Jim is responsible for the maintenance along the river Avon, Peter is responsible for the maintenance along the River Stour and has been active in the past with design of hydraulic structures. The River Stour takes about 2 days to build up a high water, while the River Avon takes longer to build up a high water. From the River Avon down at Salisbury there is a variety of different structures: new scheme embankments, concrete core and brick clad, sheet pile walls at Harnham at the outskirts of Salisbury. At the swimming pool hatch (?) there is a radial gate with a new flood channel with low flood walls along to control the amount of water coming through. Further down there are pumping stations and walls, flood embankments, and water is taken round Fawding bridge town. Further down there is a pumping sump collecting surface water when the river level is high and discharge into the river is not possible. A range of different structures occur along the River Stour: flood walls, banks and channels. Down towards Blandford there is a large wall (what sort of), pumping station, banks, embankments, but no major river control structures. At Winbourne there are walls through town (no control structures). Weymouth scheme with groynes and a pumping station, Preston beach consists of a beach in front of a vertical wall structure. At Bourne at the top of the River Avon, there is a groundwater flooding issue, takes weeks throughout winter to build and lower. Along River Stour in summary: embankments, brick cladding, sheet pile walls, some maintained channels and some compound channels. Brick cladding looks nice but does not provide any strength, expansion joints in the concrete wall structure are designed for 50 years, but usually deteriorate quicker than that. At Harnham a sheet pile wall is being renewed. At Christchurch there is a sheet pile through the embankment. At Blandford there is a sheet pile wall underneath the ground. Maintenance activities consist of cutting grass (three times a year), some slips, some low settlement locations. Badger burrows have been limited by applying a sheet pile wall through the embankment (near West Bay), some locations with moles. Some embankments are built of a type of clay / gravel that when compacted becomes like a mortar / concrete that seals the infestation off. Paul Harbour, where the Frome discharges, has extensive collapse and cracking, is replaced with material from a ditch nearby (decent material). Science Report Assessment and measurement of asset deterioration including whole life costing 67

76 Channel maintenance is not really a core activity. There are stone structures from a Victorian age, some 250 years ago. Flood relief channel at Swanage with a box culvert, a big culvert with seaweed. 68 Science Report Assessment and measurement of asset deterioration including whole life costing

77 Area/Organization Rhuddlan, Environment Agency Operations Delivery Depot Date: 24 / 6 / 2008 Venue: Environment Agency Operations Delivery Depot, Rhuddlan Participants: Clive Jones, Field Team Leader Mike Jones, Technical Support Team Leader, plus staff Craig, James and John Jonathan Simm (HRW) Asset types in area and their maintenance The area is responsible for a mixture of hard and soft defences. Most maintenance activity is focussed on the soft structures (embankments). In addition they have seven major pumping stations to maintain for the MEICA team, where the main activities are things like upgrading switchgear, replacing pumps etc, with about 10,000 per year spent on maintaining these. The area is responsible for urban, rural and coastal assets. Most of the coastal defences are maintained by other authorities with the exception of some coastal embankments at the Point of Ayr. In land the major catchments are those of the Dee and Clwyd. These rivers have major tidal embankments in their lower reaches, supplemented by pumping stations. Inland the rivers are only embanked at specific locations through towns (e.g. St Asaph) A significant issue for the area is the shift to flood risk management (FRM). The team commented that when the emphasis changed from land drainage to flood defence, there were very few changes in practice. With the shift to FRM, more significant changes are required in practice with landowners needing to take a much larger share of the responsibility. Third-party responsibilities can also be significant in other areas, for example in maintenance of culverts (e.g. Network Rail culverts through embankments) Willow bundling is used to reduce erosion on some watercourses. The point structures (flapped outfalls and intakes) require regular maintenance of the flap valves and trash screens. The supporting hard structures are mainly concrete or steel sheet piled headwalls. Many of the RC structures were built in the 1950s and 1960s and are therefore now in the order of 50 years old. The RC is still typically in Condition Grade 3, helped by the substantial nature of the structures. This indicates quite slow deterioration. They also have a number of brick structures supporting screens, many of which were inherited from when the Environment Agency took over responsibility for the Critical Ordinary Watercourses (COWS) In terms of the linear defences, some hard structures can also be found, in particular: Stone pitching in revetments on the tidal Clwyd. Stone pitching is also used in tributaries to stabilise/prevent erosion of embankments made of rather sandy material. Stone pitching is also used to protect the coastal embankment at the Point of Ayr; here the stones are held in position using a cementitious grout/concrete. Recently some 65,000 was spent to top-up this embankment after settlement. Pre-cast concrete flood retaining walls, clad in stone Science Report Assessment and measurement of asset deterioration including whole life costing 69

78 Very little maintenance is carried out on any of the above hard structures, although it is recognised that some of the 1950s and 1960s structures are at the point where they will need some attention soon. Embankments are made of a mixture of sand and clay. Of the tidal embankments, those on the River Clwyd need to be raised to maintain their standard of protection. On the River Dee tidal embankments cycle tracks have been added recently which have raised/smoothed crests of embankments protecting them, but increasing health and safety risks. Embankment maintenance mainly comprises grass cutting and vermin control. Grass cutting takes places 1-2 times per year, although they would like to do it 3-4 times per year. Purpose of grass cutting apart from maintaining grass sward is to find slumps and vermin holes. There are a few culverted watercourses, a good example being that at Adder near Bangor, where recently a new 3.5 million scheme was constructed. They also have a large coastal outfall, two No 3.5-feet diameter pipes up to 3 m high of the beach. Here the flap valves are regularly damaged and need repair/replacement Pilots Suggestions for pilots included: 1. Old embankment(s) at Worthenbury which are no longer really required and could be allowed to deteriorate without maintenance. 2. Embankment constructed in 2006 at Almere (near Wrexham) which will be maintained. 3. Demountable barriers (3 m high, 2.5 m wide) made in aluminium (Location?) 70 Science Report Assessment and measurement of asset deterioration including whole life costing

79 Data: 12 / 06 / 2008 Venue: North East Area Office, Hatfield Participants: Brian Izzard (Environment Agency) Darsha Gill (Environment Agency) Jonathan Simm (HRW) Foekje Buijs (HRW) Managed areas Environment Agency Hatfield manages assets along: the River Lee (related to that a flood relief channel), there is quite a lot of sediment, though it is not particularly abrasive. the River Roding, similar to the River Lee. Brent / tidal Crane. a number of statutory flood storage areas, and about 8-9 non-statutory flood storage areas. channels. There are over 300 assets along these watercourses and reservoirs (see database with asset types): 10 tidal outfalls on the Thames, flap valves fixed crest weirs siphons radial gates vertical lift gates tilting gates two pumping stations now, two more to be constructed steel sheet piles gates / vertical lift gates some old / Victorian 1800s earth embankments or brick spillways along the reservoirs flood barrier, that needs closure of the highway in order to put into place trash screens weirs / sluices lock gates culverts bridges Flood defence structures are: embankments brick and masonry walls with flap gates sheet pile walls Maintenance activities or deterioration processes Reservoirs: the Environment Agency carries out grass cutting on reservoir defences about eight times a year in order to enable proper inspections. vermin control in embankments: moles (trapped or gassed), rabbits (gassed), programme of pest control brings the number down, but it always remains a problem. The holes are filled up with a mixture of sand and soil (clay? ). Crack/fissures occur regularly along the reservoir embankments (there are notes on Charville Lane Flood Storage Reservoir). Science Report Assessment and measurement of asset deterioration including whole life costing 71

80 Currently there are two locations with an active slip circle occurring Channels or rivers: Occasional repairs on the masonry wall, though not often, once in 2 years, need general inspection, once in 5-6 years specialist inspection. Nothing needs to be done on the sheet pile walls as long as there is active pressure on it, react as it happens. Some locations with ALWC, whereby once a repair was carried out. However, the costs to repair are too high, moreover the cover on the sheet pile was deemed to protect the pile more than when it was removed (it is noted though that ALWC has similar rates underneath the rust cover). There are plans to implement cathodic protection. Scour undermining the structures occurs along channels. Scour upstream and downstream structures, diving surveys carried out to assess the damage. Some embankments are built upon a peat layer, the embankments are supported by piles. Contiguous piles. Costs of maintenance activities The following costs were mentioned: 800,000 is spent on dredging, most of the costs are due to contaminated soil / usually the costs amount to 350,000. The dredging activities are carried out for improving the flood capacity. 2.5 million on the rest of the maintenance activities, whereby the channel maintenance has most costs: grass cutting, weeds in middle river, banks maintenance, debris clearance (i.e. trash screens, shopping trolleys, tree trunks, furniture, garden trimmings, tyres, cars / mopeds). 72 Science Report Assessment and measurement of asset deterioration including whole life costing

81 Area/Organization Date: 25 / 6 / 2008 Venue: Ainsdale Discovery Centre, Sefton Borough Council Participants: Graham Lymbery Paul Wisse Jonathan (HRW) Simm Monitoring programme The Sefton monitoring programme was described. Details, including full reports, available at: Annual comprehensive monitoring had been in place since 2001, which included defence inspections. It was hoped that this monitoring would eventually be absorbed within the North West regional monitoring programme. A bid for this was planned for 2009 with the intention of the first funded period being 2011 to Costs/spend Maintenance spend was described as being minimal. A total of 73,500 was spent last year of revenue funds (secured via Revenue Support Grant). Of this, 30k to 40k was spent on maintaining structures of which at least half was spent on maintaining hand-railing (replacing that which had either been stolen or damaged by the sea or by plant and equipment). 3k to 4k was spent on maintaining dunes 5k on data collection and management. Of the remainder there was some grass cutting on the embankments to the north of the Sefton frontage but most was spent on other activities. These included: training, equipment, conferences, subscriptions, the coastal magazine (to keep residents informed) and funding a PhD on salt marsh areas. The overall philosophy was to move from minimal maintenance to maintenance funded under capital programmes. Assets Water levels dominated flood defence concerns rather than waves which tended to be modest and just assist the water level dominated overtopping. A problem of a lack of maintenance on the Southport frontage had been overcome by a capital scheme, involving three phases from 1997 to Previously on this frontage the defences had comprised or a low lying causeway built to support the coastal road. At North Crosby the defences were dominated by the rubble beach created partly pre- 2 nd World War (WW) and partly as a result of the demolition of properties destroyed in the 2 nd WW blitz. Progressively gabions had been added to the defences (making use of the rubble). Then timber breastworks had been added and finally a concrete wall in the 1970s. At South Crosby the reinforced concrete wall defences had been built as part of the Seaforth Dock development (late 1960s early 1970s). The defences were generally in good condition other than some spalling of concrete off the capping beams. Abrasion of concrete was not severe (sandy beach), but the concrete had been weathered down to an exposed aggregate finish. At Hightown there are no structures. A combination of brick and rubble has formed a shingle beach. Science Report Assessment and measurement of asset deterioration including whole life costing 73

82 Deterioration and maintenance The Sefton team was keen to better understand deterioration rates, in order to better understand the optimum time for interventions. Some repairs had been carried out on concrete structures but the Sefton team was concerned about their quality. Local contractors had been used but the team was unsure of their skills. In future, (e.g. on the Southport frontage) it was envisaged to poll maintenance moneys over several years and then make use of the resulting funds to employ a specialist concrete repair company. Sand dunes dominate the Sefton frontage, but although monitoring of rates of accretion and regression was regular, maintenance was minimal. Such maintenance as was carried out comprised of installation of wind traps to hold sand. As well as conventional sand fences, an approach that had been used successfully was to encourage residents to bring their old Christmas Trees for installation in the sand and to trap more. Management of the sand dunes was complex because there was also involvement from Natural England and the National Trust as land managers. Dune fronts were periodically eroded by the action of the sea during very high tides. For example, gains achieved during the 1980s were lost during the 1990s, but were probably still worthwhile overall. Justifying significant expenditure on sand dunes was difficult because of the low impact on properties behind them. Maintenance was not required on the salt marsh areas to the north of the frontage because these were generally accreting. Accretion was a result of a combination of general accretion of this section of coast combined with changes to the training of the Ribble estuary. PhD work was examining the potential sensitivity of this area to sea level rise. Piloting Sefton team suggested that fixed point photography could be a useful adjunct to other forms of monitoring, even if some computer manipulation were required for comparisons to be effected. Sefton team interested in being involved in piloting, including monitoring of things like sheet pile wall thicknesses. It was suggested that part of the necessary piloting could be embraced within the Regional Monitoring programmes and that the research team should talk to Prof. Andrew Bradbury about this. 74 Science Report Assessment and measurement of asset deterioration including whole life costing

83 Area/Organization Date: Venue: Participants: Broadland Environmental Services Limited (BESL) Peter Lawton Answers to the questionnaire (in Chapter 4.1) 1. BESL needs are covered by its own systems. Deterioration and whole life costing are part of its asset management system. 2. Approach realistic. Pessimism about usefulness. 3. Prefer tools and training in respect of R & D output. 4. Asset types include; Earth banks mainly without protection Steel sheet piles Sluice gates Pumping station Flood boards Outfall structures 5. Large scale asset replacement/renewal. Regular maintenance includes; Grass cutting Clearing watercourses Minor cyclic Reactive Emergency 24 hour response. Ongoing asset inspection which drives maintenance programmes relative to budget. 6. Hydraulic; Wave action Boat wash Flow velocities Tidal Overtopping Weather Geotechnical Earth banks founded on very poor ground, banks made of locally excavated material of poor quality Saturated earth banks Livestock Pedestrians Vehicles (usually maintenance) 7 & 8. Loadings frequently contributing to deterioration are; Vehicles on banks, particularly joy riding and four-wheel drives. Livestock Pedestrians Overtopping damaging the rear face of banks Shrinkage of banks often allowing leaks Science Report Assessment and measurement of asset deterioration including whole life costing 75

84 9. Excessive vegetation particularly scrub resulting in increased wildlife activity. Root growth not perceived to be a problem. Conflict here with conservation bodies that wish to reduce cutting of vegetation. 10. Deterioration processes observed as listed above. 11. The starting point for BESL managed earth banks is condition grade 4/5. Monitoring data include bank crest levels BESL contract is driven by this feature. BESL view is this is essential. 12. Visual inspection and survey data used to categorise assets into three grades: Poor, average and good. Initially only visual inspection used. BESL have the view that the Environment Agency condition grades are too fine for normal use. No BESL bank in grade one condition. Also stated that BESL banks will never achieve high condition grades as Environment Agency require bank crests to be 3.5 m wide compared with the 2 m crests in the Broads area. 13. Residual life not estimated. 14. Failure processes observed include; Overtopping leading to scour of rear face Shrinkage leading to leaks Bank erosion causing slumping 15. Not aware of any exceptional loading 16. Specific structures include the many km of earth bank. 17. Maintenance regimes above are designed to ensure compliance with the performance requirements of the BESL contract. Little doubt that regular maintenance reduces deterioration. 18 & 19. Not used 20. Driver for all activities is budget versus risk. This is a risk-based contract BESL decide all activities. For example, BESL grass cut more frequently than originally budgeted to reduce risk and pre-empt later works Staffing an important cost centre. Not seen as an issue in the context of the contract. 23. Any cost information requested will be provided. 24. Entire contract is refurbishment and maintain. 25. Any refurbishment cost information requested will be provided. 26. Typical cases of decommissioning include the removal of earth banks as part of set back works. 27. Any decommissioning cost information requested will be provided BESL will participate fully in the research project. 76 Science Report Assessment and measurement of asset deterioration including whole life costing

85 Area/Organization Date: Venue: Participants: North Norfolk District Council (NNDC) Peter Lawton Answers to the questionnaire (in Chapter 4.1) 1. Project is useful particularly with respect to ageing assets. 2. Project is realistic only if output is linked to the budgetary process. Otherwise it is not worth having. 3. R & D output should be in the form of guidance notes and training. 4. Coastal defences: Concrete seawalls with and without steel pile toe protection Rock revetments Rock groynes Timber revetments with steel pile toe protection Timber groynes Groynes of composite construction Concrete and timber access ramps 5. Maintenance is now mostly reactive. Refurbishment of defences in hold the line policy areas are planned over the next 10 years. Works in areas where the policy has recently changed from hold the line to no active intervention are limited to ensuring a further 10 years of residual life. In other areas work is limited to health and safety related activity. 6. Loading conditions on this exposed coast include; Wave and tide Abrasion Impact of flint and gravel Corrosion Geotechnical: Unstable cliffs Human activity 7. The direct action of the sea on say beach level degradation is hugely important. The impact of flint and gravel on walls and timber structures is significant in terms of abrasion and physical damage. Cliff failures have severely damaged defences. Human activity has little impact. 8. Daily. 10. Abrasion of concrete, timber and rock, corrosion of piles, change in concrete strength, degradation of beaches. 11. The only measurement of deterioration is the survey of beach levels. Post SMP strategy studies have added to the knowledge of deterioration in that the physical condition of structures are being assessed with a view to determining residual life. Included in this process is, if possible, a review of original designs and geotechnical investigations. e.g. Window sampling in beaches and trial holes at the base of walls. Science Report Assessment and measurement of asset deterioration including whole life costing 77

86 12. The condition of the defences is normally assessed following a visual inspection and categorised using the Environment Agency condition grades. (NNDC are not aware of the revisions to the Condition Grade Manual) 13. Residual life is not typically estimated. 14. Failures have included; Beach degradation leading to wall instability and collapse. Overtopping and penetration, through joints, undermining wall foundations. Structural components of timber revetments abrading to the point of failure. Timber revetments physically destroyed by wave action. Beach degradation leading to steel pile instability. Cliff slumps destroying defences. Inadequate design. Inappropriate design. 15. Exceptional loading is not normally a problem with the exception of cliff failures. 16. Typical deterioration processes include; The abrasion of timber structures such as groynes or revetments where necking of piles is common. Pile corrosion. 17. The reactive maintenance regime has little impact on the rate of deterioration. However, concrete apron renewal works are attenuating the deterioration of walls. 18. There are no typical maintenance activities. Reactive maintenance includes the replanking of revetments and groynes (not systematic) and concrete repairs. 20. The priority given to reactive maintenance is principally health and safety driven. Accessibility is an issue with the seaward ends of groynes, for example, being rarely repaired. 22. Low staffing militates against monitoring and planned maintenance. 23. Cost data can be provided. 24. A typical asset refurbishment is the renewal of the 2-3 m wide concrete apron to a long length of sea wall. This is a one-off project designed to attenuate deterioration and improve health and safety. 25. Actual cost data can be provided. 26. Two recent decommissioning cases have both involved the removal or reduction in length of redundant steel piles left after failure or abandonment of defences. 27. Cost can be provided. 28 to 31. NNDC is willing to provide further information and access to structures for monitoring. The 10-year-old seawalls in Sheringham may be a good site to monitor. There are many older defences which, if monitored properly, could yield valuable data related to deterioration as they tend towards condition grades 4 and Science Report Assessment and measurement of asset deterioration including whole life costing

87 Area/Organization Environment Agency; North East (NE) Region Date: 3 July 2008 Venue: Participants: Alan Cadas, Asset Systems Manager Peter Lawton Background Other than the condition grade returns, there are no monitoring data. The region has collated the annual inspection returns, based on the condition grade manual, and local knowledge. This has enabled the development of a structural assessment programme which, in turn, has informed an asset replacement programme. The region has only recently begun a programme of surveying crest levels. The limitations of the condition grade manual, particularly as the region uses inspectors rather than engineering technicians or engineers, came out strongly in the interview. Deliverables Project has a synergy with the region s asset management goals. This R & D viewed as extremely important in respect of its potential to improve the asset management process. The output, as a package, is regarded as an extremely important aid to forward planning. The approach is regarded as ambitious but achievable. Deterioration curves will be particularly welcomed alongside whole life costing models. Assets 500 km of earth bank, almost all built of locally won material. Concrete and brick walls: Few Sluices and tidal gates: Few Pumping stations: Few Culverts As the principal assets are fluvial earth banks, the interview centred on these rather than the relatively small number of other asset types. Maintenance Annual: Grass cutting Vermin control. NE region has a vermin control programme. Asset replacement: The output of the collation of condition grade assessments and local knowledge has enabled a programme of asset replacement to be implemented. Earth banks are completely replaced following the demolition of the original bank. The local patching of a potential breach is no longer practised. Example: Cor Bridge. Breached in Again in 2000 after overtopping. Intrinsically weak despite having a condition grade of 2. The bank was completely reconstructed. Asset refurbishment: Crest levels are topped up when necessary. Reactive repairs Deterioration Science Report Assessment and measurement of asset deterioration including whole life costing 79

88 Daily flows have little impact. High flow events have a relatively short duration. Typically, the incident control room is open for less than 48 hours. Exceptionally high flows damage by overtopping. High water levels are thought to increase deterioration because of flows through and under the banks. Deterioration processes observed include slumping, slope instability, seepage, vermin infestation, base erosion and farm animal erosion. Rates of deterioration have not been monitored except through the medium of the annual condition grade inspection. The NE region has only recently started a programme of surveying crest levels. Only intuitive assessments of residual life are made but these assessments are influenced if an asset has been the subject of a structural assessment. Loadings contributing to failures have included: Vermin, cattle, four-wheel drive erosion, pedestrian erosion and over vegetation. The dominating cause of failure is overtopping. Costing Annual maintenance is planned around assessed need now resulting from the collation of condition assessments and local knowledge. In the period 1994 to 1997, the annual maintenance demand was balanced by redistributing effort between over-maintained and under-maintained systems. SAMPS is used to assess/manage maintenance. The works programme now includes defences where reduced intervention or decommissioning may be appropriate. The staffing levels of management or operational staff are not perceived to be an issue. The NE region is willing to provide cost information, data and monitoring sites. The NE region is very concerned about culverts. An inspection programme has started recently looking at the condition of culverts on main rivers irrespective of culvert ownership. The Environment Agency guidance on this topic is regarded as very poor. 80 Science Report Assessment and measurement of asset deterioration including whole life costing

89 Date: 15 / 5 / 2008 Venue: Ipswich Participants: Jim Warner (Environment Agency) Marit Brommer (RH) Output R & D project Provision of guidelines and decision-support tools (yet simple) on deterioration rates and WLC analysis of certain assets throughout various sites in the UK could be very useful. Output could be focused on the question when to stop maintaining the asset and decide to replace the old asset with a new one. 1. What is an asset All natural and physical structures protecting land from flooding. 2. Asset types Coastal (Tidal defences) 1. Earth embankments; 2. Concrete sea walls (power concrete); 3. Steel sheet piles; 4. Tidal barriers (not discussed further for this project); 5. Watermills (not discussed further for this project). 3. Deterioration Earth Embankments a general note: embankments do not deteriorate because they are maintained. The armouring (concrete blocks) for instance are replaced when the entire asset is not fit for purpose anymore. There is one exception the East Mirsie Wall. This wall has known weaknesses and is more easily damaged during storms. Most likely reason is the higher percentage of silts in the core of the embankment, which hampers the compaction of the embankment. Yet, differences in deterioration rates of earth embankments have been noted due to site-specific variabilities in the strength of the currents, their exposure to the open sea and differences in tidal regimes. Steel sheet piles designed for a life cycle of 50 years (in the 1950s) and are still OK. They do not seem to deteriorate and little to no maintenance is being undertaken. Concrete sea walls these do not show any evidence of deterioration. However, saline intrusion is thought to be a considerable cause of deterioration. 4. Maintenance Annual maintenance regime Earth embankments Annual inspections of the earth embankments 8. Grass cutting 9. Surveys (visual inspection) of damage and condition assessment. Sea walls Annual inspection of the sea walls to report their condition and potential damage. Science Report Assessment and measurement of asset deterioration including whole life costing 81

90 Responsive maintenance After heavy storms, embankments and sea walls are checked, especially in areas where weaknesses are known (see above). Annual maintenance regime Steel sheet piles Annual inspection takes place. Paint helps mitigate deterioration, however, in general these assets are considered as stable, and no active maintenance regime is provided for. 5. Whole Life Costing There are hardly any data on the costing aspect of maintenance. Costs are not recorded against activities, however, if costs are stored (eg, refurbishment costs) in a database we can certainly have access to those data. 6. Pilot Sites There is considerable enthusiasm to support the initiative of setting up pilot sites for monitoring purposes. Especially with regards to monitoring techniques such ultrasonic devices to evaluate steel sheet piles and loss of residual life would be interesting. There is a preference to monitor hard defences such as concrete structures (saline intrusion) and sheet piles, but the Environment Agency is open to discussion. 82 Science Report Assessment and measurement of asset deterioration including whole life costing

91 Date: 16 / 1 / 2008 Venue: Canterbury Borough Council (BC) Participants: Steven McFarland (Canterbury BC) Jonathan Simm (HRW) Marta Roca (HRW) 1. Definition of asset The first question was the same as is in the Worthing meeting: what is an asset? Answer: all structures and beaches 2. Characteristics of coastal geomorphology: the sea walls and beaches are founded on London clay; the beach material is mixed with 20-25% sand, cobbles of 100 mm and an overall D 50 of mm. The storms have not very high wave levels (for return period 10 years, H s =1.2 m, T z = 5-6 seconds; for return period 100 years, H s =2 m). The foreshore varies between -2m to +2 m OD. 3. Asset types for coastal protection Timber groynes, with different kinds of timber and slopes about 1:8. Concrete seawalls Rock structures, using stone of 1 to 3 tonnes 4. Deterioration and failure processes identified: Seawalls 1) The main problem for the seawalls is the drop of beach levels. This arises from a combination of: Long-term lowering of beach levels due to longshore sediment processes (groyne fields designed to reduce this). Short-term lowering of beaches during storms followed by direct impact of waves of structures during a storm Subsequent failures of seawalls are by overturning or sliding. Overtopping of seawalls has never been a problem. Deterioration curves (obtained from the Thames embayment studies) have been examined, but they realized that material deterioration was not the most important problem but the beach levels dropping. A useful risk analysis of defence failures on this frontage can be found in: McFarland, S., Edwards, T. and Lombardo, S., A risk-based approach to seawall failure, North Kent. Proc. 40 th Defra Flood and Coastal Management Conference, Paper 10.3 Examinations of walls have taken place on 13 occasions when scour holes formed in front of them. In two of these cases failure of the wall itself actually occurred Science Report Assessment and measurement of asset deterioration including whole life costing 83

92 Beaches Problems with failure of beaches (i.e. significant loss of cross section) are usually related to the failure of one key groyne at the end of a system. From surveys of beach cross section area, the failure can be observed in a graph such as this: Cross section area Failure of the key groyne t A further factor in loss of beach profile arises from erosion (downcutting) of the underlying London Clay pavement. Rates of erosion of 0.5 m per 20 years have been observed from surveys of the clay foreshore levels Rock structures Have been very stable and durable, with the only exceptions being: on the Herne Bay Breakwater, where there is some movement of stone on the breakwater outer slope in 10-year return period storm there has been some theft of bedding stone by local residents Groynes are rarely damaged by wave action or the action of the beaches, but do occasionally suffer damage from impact by debris or boats 5. Maintenance tasks Groynes The annual maintenance costs are per year to maintain about 430 groynes, without costs of timber. These works are basically to check and replace planks. Unlike on the South coast, Canterbury BC is not adjusting groyne heights on a routine basis. However, when re-profiling is required, Canterbury BCdo pile extensions on a routine basis (unlike South coast and probably possible because of the less aggressive environment). Life of timber groynes is about 60 years but replanking of groynes is likely to be necessary after about 30 years. This is in contrast to the South coast where abrasion of the timber reduces the life significantly. The figures quoted are for hardwood groynes some such groynes built in the 1950s and 1960s are still serviceable. Hardwoods used have included Ekki (schemes at Tankerton in 1996, 2000) and Greenheart (schemes at Whitstable in 1989 and 2006). Much shorter lives have been experienced with softwood groynes. Some groynes constructed in Douglas Fir only lasted 10 years, being destroyed by the marine borer Teredo. Groynes constructed in oak have a very variable expected life, in some cases being affected by soft rot About 50% of the Canterbury BC groynes are under 20 years old and no maintenance is necessary. When reconstruction takes place, 50% to 60% of the timber sheeters used at low level are from recycled groyne planks. Seawalls Seawalls do not have big abrasion problems and little maintenance is done. Another maintenance task is the displacement of material along the beach There are no changes between beach levels at every side of the groyne. Beaches 84 Science Report Assessment and measurement of asset deterioration including whole life costing

93 Beaches are much less mobile than on the South coast. 6. Pilot sites Canterbury BC is willing to consider one or more pilot sites being located on their frontage. They suggested groyne schemes constructed in the 70s and 80s, such as: Herne Bay Groynes and walks. Studhill (between Herne Bay and Tankerton) strategy study scheme Whitstable groyne scheme. Science Report Assessment and measurement of asset deterioration including whole life costing 85

94 Data: 16 / 1 / 2008 Place: Worthing Borough Council Participants: Bryan Curtis (Worthing BC) Glenn Longley (Worthing BC) Roger Spencer (Arun district council (DC) Tony Davison (Environment Agency West Sussex- Hampshire) Jonathan Simm (HRW) Marta Roca (HRW) 1. Definition of an asset Assets include not only those structures such as seawalls and beaches which directly prevent erosion and flooding but also the wave and sediment control structures which keep beaches in place. 2. Worthing and Arun are involved in, and make use of the South Coast Monitoring Programme which is coordinated from Southampton Oceanography Centre. This includes: waves, tidal levels and bathymetry measured every 5 years to up to 1 km offshore overlapping with nearshore beach survey/profiles,. If significant storm event occurs (assessed from wave monitoring and whether waves exceed trigger levels) a post-storm beach survey programme. This enables checks on whether there is enough beach for coastal protection. 3. Key asset types for the two authorities used for coastal protection Timber groynes constructed from Canadian Douglas Fir (Worthing BC) or Greenheart piles with either greenheart or Douglas Fir planking (Arun DC). Douglas Fir groynes last a shorter period of time than greenheart. Some differences between hardwood types were discussed Timber groynes are sometimes reinforced at their root (end adjoining seawall) using gabion baskets, rock, or more recently geotextile bags filled with beach material Worthing BC are starting to make more use of rock groynes rather than timber Seawalls: whilst a number of seawalls exist very little maintenance is carried out on these. (None for Worthing BC). Arun DC have some exposed concrete and other walls which require maintenance; they also maintain breastworks (shore parallel groyne-like structures) 4. Life of timber groynes Worthing BC report softwood timber groynes have a life of between 15 and 25 years, with an average between 20 and 25 years. Arun DC hardwood groynes typically last between and years (Need to check) 5. Maintenance regimes Seawalls Worthing do not maintain seawalls as structures but do spend some money removing graffiti. Arun maintain their timber breastworks and reface their exposed concrete seawalls to take account of abrasion. Groynes The Environment Agency in West Sussex does not maintain any of its groynes they have adopted a policy of allowing natural coastal reshaping (unclear whether this compromises flood defences in places) 86 Science Report Assessment and measurement of asset deterioration including whole life costing

95 Local authorities do maintenance, partly because they want to maintain defences in good condition but also because they have to take care about the visual state of the beach and also, to avoid heath and safety problems with beach levels being too different between the two sides of the groyne. Two kinds of groyne maintenance were identified: reactive and proactive. Reactive maintenance involves replacing missing planks. o For Worthing BC this is done in the low season between middle of September one year and July of the next). o For Arun DC the work schedule is defined by the two main inspections per year, but supplemented when specific problems are identified). The maintenance gang is on the beach most of the time. Proactive maintenance is associated with the planned re-profiling of beaches, by gradually raising planking levels as beach material accumulates. The life of a timber groyne may be limited by its ability to continue to accumulate beach material (to grow the beach). Pile extensions were not found to be robust by either Worthing or Arun) Beach management Another maintenance job is the removal of beach sediment from some areas to refill other ones. In the Selsey Bill to Beachy Head coastal bay, this is presently done at localised areas associated with the Rivers Arun and Adur (Shoreham), at Brighton Marina and at Newhaven Harbour The long-term aim is to integrate this work all over the entire bay. 6. Deterioration and failure processes Timber groynes The main deterioration process with timber groynes is erosion of the timber piles just above the prevailing beach level. If this becomes too severe, this is mitigated by placing some sacrificial timber elements. Care has to be taken to avoid loss of section to such a point that the piles may fail in bending as beach levels grow. Along the length of the groyne, the main erosion problems are concentrated in the middle of the shingle area Worst conditions area Shingle area Sand area Slope 1:7 Slope 1:100 Softwood rot of groynes is worst at the root near the adjoining seawall Rock groynes require very little maintenance Seawalls A few seawalls managed by Arun DC which are exposed to active beach erosion suffer very bad conditions with losses of surface material (concrete) at rates of 25 mm per year. Otherwise the life of exposed seawalls can be about 50 years or just 30 years without maintenance. Unexposed seawalls (i.e. walls with the front face completely covered by stable beaches) can last 100 years if maintained 7. Costs The cost of new timber groynes 10 years ago (late 1990s) was typically 30,000 compared with per groyne. Worthing BC reported that costs were now something like 100,000 for a 70-m long softwood timber groyne. Science Report Assessment and measurement of asset deterioration including whole life costing 87

96 Maintenance costs had been very well documented by Worthing BC and they had a spreadsheet showing the amount of work that had been carried out on their timber groynes by groyne and by year. (Need to check what is included in the figures.) Worthing BC offered to provide this spreadsheet, along with some supporting drawings of the groynes themselves 8. Participation in long-term monitoring programmes Both Worthing and Arun were willing to be involved in long-term monitoring programmes. Some candidate sites where new works were anticipated were discussed, including an Arun DC scheme for four new timber/rock groynes on the Littlehampton frontage and a Worthing BC scheme for the frontage of the centre of Worthing. 88 Science Report Assessment and measurement of asset deterioration including whole life costing

97 Data: 18 / 06 / 2008 Venue: Bournemouth Participants: David Harlow Jonathan Simm (HRW) Foekje Buijs (HRW) The deterioration interview took place in Bournemouth with David Harlow on In 1907 concrete groynes were built to retain beach along approximately 8.5 km of beach. Costs of replacement are about 12,000/stretching meter beach. In 1987 breach of the seawall occurred (no description about the details, probably the beach was lowered and the vertical wall undermined). From the 1940s until 1970 the beach levels were increased in Beach Improvement Schemes (BIS) 1, 2, 3 and 4. BIS 1 in m 3 of sand BIS 2 in m 3 BIS 3 in m 3 BIS 4 in m 3 And there are two more top ups to come. The first generation of groynes were concrete groynes built in the period The concrete groynes were short, low and poorly designed by current day standards. BIS 2 introduced a beach replenishment and timber groynes, which were higher, longer and more closely spaced. With BIS 2 the beach levels increased significantly. Bournemouth has a nearly diurnal cycle, with two high water level peaks following quite quickly and two low water levels (6 am and 6 pm roughly), resulting in about 8 hours of high water and 4 hours of low water. The tidal range is about 1.8 me. The concrete groynes have been subject to abrasion which exposed the reinforcement, these groynes have been replaced by the second-generation timber groynes. The sand is 200 μm and there is some shingle among the sand as well. Between the groyne planks there are 100 μm (?) gaps, due to temperature differences the gaps can increase and shingle can get caught in between. The shingle grains subsequently leave a larger gap behind, through which sand particles can transport. The groyne is subsequently not impermeable anymore and the beach retention is reduced. In the third-generation groynes, plywood planks are applied to cover the gaps between the greenheart timber planks. Later upon removal of the planks gribble was found behind the planks. To tackle this problem several other solutions were applied, for example: - Neoprene filling between planks - Hygroscopic joint sealing The main problem is that the work can only be carried out in winter at 6am or 6pm during the low water periods. That does not leave enough time to seal gaps appropriately. Following a strategy study in 2004, 25 new groynes were implemented. Science Report Assessment and measurement of asset deterioration including whole life costing 89

98 An example of timber deterioration is demonstrated by a site where in 1986 three different types of timber planks were put into place. Two planks were greenheart timber, two planks were ekki and two of another type of timber. Fifteen years later upon inspection the greenheart timber piles were in a much worse condition than the ekki timber piles. Twenty purple heart timber planks were applied at groyne 43 and were found to be heavily gribble infested after a very short period of time, these planks were replaced by greenheart planks. Rock groynes built in 1987 (to the east of groyne 50 there are three rock groynes and four timber groynes), showed no rock movement since Subsequently, quite some detail is covered in the sketches (which I am not able to retrieve unfortunately). Some other details are that a seawall nearly failed in 2006, from abrasion that is the result of shingle more than sand. There has been piping through the seawall. The costs are about 250,000 per groyne in 2004 and about 500/groyne/year maintenance costs, about 20,000 yearly in total. Another cost indication is about 580/m 3 dredged material for replenishment in total about 14,000 m 3 of sand. 90 Science Report Assessment and measurement of asset deterioration including whole life costing

99 Area/Organization Date: Venue: Participants: Waveney District Council (WDC) Peter Lawton Answers to the questionnaire (in Chapter 4.1) 1. Yes. Tools to forecast / monitor asset element deterioration / processes and rates would aid asset management planning. WDC would like the study to include asset removal upon life expiry as an issue. WDC believe this requires closer scrutiny to inform economic appraisals especially as SMPs appear likely to recommend more retreated defence lines. 2. The approach appears appropriate. Achieving the target deadline is a matter for resource planning by the PM. 3. Difficult to comment at this stage. Training or promotion launch upon completion is always helpful. In terms of application a link with PAG process would also give weight to usage. 4. Main asset types are: Concrete seawalls with steel sheet piled scour protection. Timber wave screens with steel sheet piled scour protection. Rock revetments in front of concrete walls. Timber framed steel sheet piled groynes. Concrete and steel piled harbour piers - rock lined in places. 5. Maintenance regimes are strongly influenced by management policy and vary from hold the line to abandonment. Under hold the line maintenance nature / frequency varies from no intervention to high input maintenance and renewal, according to exposure and environmental pressure. Under abandon (north of Corton Village) a do minimum approach is taken to maintain public safety. 6. Loading conditions. Wave action - typically depth-limited with ds up to 2.5 m under spring range - creating direct loadings, scouring water velocities and also section loss by abrasion from waterborne sand/stone. Tidal currents creating scouring water velocities and also abrasion damage by waterborne sand/stone. Corrosion - ALWC. Retained soil and water. Surcharges - soil and vehicles. Uncontrolled surface groundwater flows. 7. They are key drivers. Most significant issues are beach scour and abrasion by mainly wave action. 8. Constant although variable in severity. 9. n/a Science Report Assessment and measurement of asset deterioration including whole life costing 91

100 10. All under 6 and Most severe steel section loss on groyne piling is 12-mm thick over years on high-energy, shingle-rich beach. Other data are available for less severe exposure conditions and other material types. 12. Routine / reactive inspection using a range of thickness profile change measurement techniques/devices including callipers, sonar, tape measure, scanning survey. 13. Yes. Based upon previous experience of exposure in similar circumstances. This is key part of all project appraisals and also option selection process for maintenance actions. Life expectancy feeds into whole life costing appraisals and will affect preferred option selection. 14. See WDC aware of the nature and variable extent of processes that contribute to failures. 16. Yes. Corton wave screen. Includes pile rotation failure by scour at toe and section loss from abrasion leading to wash out of retained material. This combination also led to failure of now derelict defences over north Lowestoft frontage. 17. Abandonment. Allows unchecked deterioration. 18. Typical maintenance activities: Annual Fixing and timber component check tighten / renewal. Reactive groyne steel pile hole patching (with timber planks). Reactive concrete / masonry surface / joint repairs. Recycling of material to manage scour risk over prone frontages. Public safety works include management of hazards from abandoned defences, handrail upkeep, cutting out worn groyne, trip hazards. Replacing displaced rock armour from revetments. Plant access creation improvement works. Refurbishment projects inc partial groyne re-piling, seawall roe pile provision, rock armour rebuild, fender component renewal. 19. Tasks for structures on a hold the line frontage vary in terms of specification and frequency from say abandon. 20. Management policy, exposure to damage, nature of public use. 21. Regular inspections and experience guide intervention frequencies. Some tasks require twice yearly actions others have intervals of more than5 years. Timber/steel frame/bolted structures tends to require at least annual attention. Rock structure intervention tends to be reactive after storms and varies with exposure levels. More detailed data could be given with more time. 92 Science Report Assessment and measurement of asset deterioration including whole life costing

101 22. Staffing resource levels could limit maintenance planning and implementation but not most significant issue. 23. Yes. 24. Refurbishment. Groyne, wave screen and seawall toe sheet steel re-piling. Frequency variable with exposure from 15 to 50 years. Groyne timber post renewal. Frequency variable with exposure from 25 to 50 years. Concrete wall / slab recasting. Frequency variable with exposure from 40 to 100 years. Promenade slab replacement. Frequency variable with exposure and public usage from 30 to 70 years. 25. Yes. 26. Decommissioning. This is an area that has not been developed but requires greater consideration in policy planning and costing in my view. Examples of `decommissioned' defence sites over WDC. Lowestoft North Denes or Mobbs seawall failed during WW2 and was abandoned. No effort made to remove debris. Private Colman seawall at Corton failed by 1900 and concrete ruins strewn beach. Small percentage recovered for use in recent wall repair. WDC has removed five c.1960 timber and steel groynes from north Lowestoft frontage since Recent Southwold defence scheme removed 8 c.1978 long timber and steel groynes plus around15 shorter groynes c Process was very lengthy and more expensive than estimated. WDC has a groyne field at Gunton that is probably at effective life expiry and requires management. Cost of removal is very high relative to the management of health and safety and amenity. 27. Yes. 28. Yes. 29. Yes. The rock breakwater at Children s Corner and other rock slopes at north Lowestoft. 30. Yes. The new timber piled and rock groynes at Southwold could be used to monitor timber section loss under aggressive high-energy, shingle-rich beach. 31. Yes. Science Report Assessment and measurement of asset deterioration including whole life costing 93

102 Date: 07 / 07 / 2008 Venue: Environment Agency Wales, South East Area Office, St Mellons, nr Cardiff Participants: Martin Cadogan Asset System Management Team Leader Rob Wilkins Asset System Management Technical Specialist Tim Hopkins Environment Agency Asset types in Area In the South East Wales area there are quite variable types of catchment. The South Wales valleys have fairly small catchment areas but steep gradients and therefore very flashy rivers. The catchments of the River Usk & River Wye further east are large with sources in mid-wales so river levels tend to be high for much longer periods. There is also a large area at risk of tidal and coastal flooding, i.e. the Wentloog and Caldicot Levels to the west and east of Newport. These areas are not highly populated but carry much important infrastructure. The area is responsible for a mixture of hard and soft defences probably in the ratio of 15% to 85%. The hard defences, flood walls, are mainly in the flashy catchments of the Rhondda and Eastern valleys. Flood Defences on the Wye, Usk, Taff, Ely, Rhymney and Cynon rivers are mainly embankments. Some masonry walls built in the Rhondda, particularly at Ferndale and Llwynypia have been functioning for 120 years and not replaced with no noticeable movement. Walls at Hopkinstown, Porth, Gelli and Trehafod were replaced after the 1979 floods. Not many walls now remain that were built before Generally it has been found that pre-cast concrete walls need replacement after a maximum of 60 years. A reinforced concrete section of the Usk town wall however now needs replacement after 30 years but mainly as a result of poor workmanship, for example, lack of cover to steel reinforcement. Problems have been encountered with embankments on the Wye being constructed with over porous materials, for example, river gravels. These are not a problem when used in embankments in the more flashy South Wales valley catchments but not on the Wye. The Hampton Bishop stank near Hereford was constructed in 1962 but failed by 2000 with considerable seepage and had to be replaced. In coastal areas it has become evident in practice and by research that the foreshore or saltings are much more important than the defences themselves. Where there is a lack of saltings then wavewalls are generally required. On the Wentloog levels polders have recently been installed on the foreshore to encourage its replenishment as additional protection to the coastal embankment. Maintenance Activities Maintenance activities in general are designed to maintain assets at a Category 3 standard, i.e. fit for purpose, although at locations where the consequence or likelihood of failure is high then Category 2 becomes the required minimum standard. 94 Science Report Assessment and measurement of asset deterioration including whole life costing

103 On masonry walls, cleaning and repointing is carried out as necessary. For embankments the main activities are cutting, mowing, spraying, trashing, raising crests and repairing damage from animals, rabbit holes etc. Maintenance of the grass sward is important and some embankments need top surface repair after damage from animals, people, walkers, bikers and off-road vehicles. Embankments can be functional for long periods with a maintenance regime properly imposed and prioritised appropriately. Maintaining revetments and the toe of embankments is important. Blockstone is the most reliable material particularly in the flashy, fast flowing catchments. Gabion type revetments and others need attention after years so are only good for short periods in less flashy catchments. Sheet piles when used for toe protection appear to have a 40-year life. Catastrophic failure is unlikely in these cases and the asset can still be fit for purpose even if some failure has occurred. Potential Pilot Sites Treforest (Taff valley): Project completed in 1970s comprising embankments, walls, sheet piles, dytap revetment, blockstone groynes. Usk Town walls: 30 year old RC walls Wentloog foreshore: Use of polders for foreshore replenishment in Rumney Great Wharf project completed in Flood walls in Rhondda Valley. Science Report Assessment and measurement of asset deterioration including whole life costing 95

104 Area/Organization Warrington, Environment Agency North West Regional Office Date: 25 / 6 / 2008 Venue: Environment Agency North West Regional Office, Warrington Participants: Ian Hale, Asset Systems Management Jonathan Simm Colin Liptrot, Operations Delivery (HRW) Asset types in area and their maintenance South Area team is responsible for the urban catchments of the Irwell and Mersey and the rural catchment of the Weaver. On the River Irwell all channels are canalised, mostly since the time of the industrial revolution. However, there have been some new schemes for example: A staged scheme commenced on the Irwell some 13 years ago on the Salford Quays, finally finished in 2006 A scheme in Rochdale completed in 2005 In addition, there have been a lot of mill-related works adjacent to rivers. Culverts A major issue for the team is the presence of structures in watercourses. In culverts, services are often found to cross the culvert and act as debris traps. There are many issues with managing landowners along these rivers; but if a blockage of any kind is found in a culvert, then the policy adopted is to fix the problem if it is likely to have a significant impact on the local community. If, however, it represents a low risk then enforcement action on the landowner will be taken if possible. The structural form of the culverts is highly varied. The majority are brick or brick arch. Some are masonry. Roofs to culverts can be as crude as timber floors supported by steel joists. The smaller culverts associated with COWS tend to have a much more varied structural form. Inspection of small culverts will be carried out using robotic CCTV. However many of the more major culverts are inspected manually by the Environment Agency s sevenman breathing apparatus (BA) teams. One member of this team is dedicated to racking defects along the culverts. Major problems identified would then lead to a more detailed inspection and then subsequent interventions depending on the associated flood risk. Minor repairs are carrred out by the BA teams as they progress through culverts. These can include repointing of brickwork, de-silting and solving problems of washouts of paved beds. Collapses can be another story. The life of culverts is enormously varied, between 50 to 200 years it was thought. For example, there is a major culvert under Manchester Piccadilly railway station built in the Victorian era which is almost as good as new. Badly built culverts might only last 50 years. An additional problem in the Merseyside area is that there has been a lot of mining subsidence since many of the culverts were built. As a result many of the culverts now act as siphons and tend to silt up or block. Often such culverts can pass beneath railways of canals and it can be difficult to find a relevant landowner to talk to. As a result of all of the above, maintenance spend in the Irwell catchment is dominated by the attention given to culverts. Some figures cited were an overall spend of 6 million per annum of which 3 million was staff costs. Of the staff costs, 25-30% were 96 Science Report Assessment and measurement of asset deterioration including whole life costing

105 associated with the seven-man BA teams. These staff were all well-paid and part of the costs included retraining every 6 months. A separate.0.25 million contract was also let for robotic CCTV monitoring. Open canalised channels An unusual feature of the catchments is the existence of dry stone masonry walls, typically gritstone/sandstone, found on the upper Irwell in places such as Rottenstall and Bacup. Brick walls are also prevalent and both could well have been there since the early 19 th Century. There are many sections of open canalised watercourses, for example parts of the Medlock River through central Manchester. In many cases the Environment Agency have no access to these rivers. When developments take place, there is a major effort to work with developers to arrange good access points; such accesses are much more useful than the standard 8-m wide easement alongside the rivers. A major issue for these canalised rivers is collapse into the water courses of the walls of old mill buildings and warehouses which form the channel sides (but often rise 10 m high). Generally engineered channels are over-sized and therefore collapses of walls into the channels may not be a major problem. However, it can be a problem if the collapsed material subsequently gets into a culvert and starts to block it. Some sheet piled walls have been used to form channel sides in Salford. Environment Agency Development Control have a policy to work with developers to open up canalised water courses where possible. Rural assets Examples of embankments and maintained channels occur on the River Birket. Debris is now often left in rural channels where it does not cause an immediate flood risk. Maintenance of embankments is a significant issue. The quality of many of the embankments is poor (silty/sandy materials) and vermin are a major problem. In 2000 floods, water came straight through some embankments via foxholes. There has been a policy, where possible, to replace the core of embankments with clay; this provides a better seal and seems to be much harder for vermin to penetrate Major Assets Major assets in the area include pumping stations; major repairs of some of these had recently been completed. Another significant asset is the Mersey Flood Storage Reservoir (comes under Reservoirs Act). Whole life costings had recently been completed on gates, pumping stations and reservoirs for input to the SAMPs process. Piloting Reviewing the files of information held by BA teams on culverts could be a major source for appraising deterioration problems. The Environmental Agency area team agreed to provide a typical culvert record and associated repair record. The area team can also supply some cost information for various types of structures. They could also assess the life expectancy of various structures. Science Report Assessment and measurement of asset deterioration including whole life costing 97

106

107