Ecosystem Services Assessment at Steart Peninsula, Somerset, UK

A systemic valuation was undertaken of marginal changes in ecosystem services assessed as likely to result from the Steart Coastal Management project, some in monetary terms and others semi-quantified. The Steart Coastal Management project entails allowing seawater once again to inundate formerly defended farmland, including modifications to the landform of to assist the re-creation of a range of wetland habitats on the Steart Peninsula. Primary drivers for this project include habitat creation and management of coastal flooding, although implications for a range of other connected services need also to be taken into account. Ecosystem services for which a market exists (typically traded goods with associated use values) were valued using market prices. For non-traded services, this study relied substantially on the economic valuation technique of ‘value transfer’. Despite having to rely on some wide but transparently stated assumptions and uncertainties, a conservative, yet considerable, net annual benefit range of £491,155 to £913,752 was deduced. Research gaps that limited our ability to quantify and/or value several ecosystem services were identified.


Description of the project
The SCMP involves recreating intertidal habitat by managed realignment of the existing coastal defences (Figure 1-2, Area D), the creation of additional saline-influenced habitat through regulated tidal exchange in Area E, and freshwater habitat in Area B. Figure  1-shows the Steart Peninsula prior to and following establishment of the planned new wetland features. This future management option has been assessed and adopted as the preferred one prior to this study. In this way, the present research only looks at this single project versus the present baseline.

Study aims and objectives
The study comprised an assessment of marginal changes to ecosystem services likely to arise through managed realignment at the Steart peninsula. This was based on the Millennium Ecosystem Assessment (2005) classification of ecosystem services, seeking as far as possible to identify economic and non-monetary value of the SCMP. Study aims included: 1) Assessment of the environmental benefits/costs from habitat creation at Steart to provide evidence to guide optimal outcomes in related projects; 2) Learning for the future by identifying ecosystem service assessment best practices, identify knowledge gaps and further research needed and lessons learnt; 3) Contributing to a growing, though still limited, portfolio of ecosystem service assessment case studies of large-scale managed realignment sites in the UK.

Ecosystem services valuation techniques
This study makes use of the benefits transfer (or 'value transfer') technique, as shown in Figure 1-. Value transfer methods estimate the value of an ecosystem (or services and goods from an ecosystem) by applying conservative and stated assumptions values derived from relevant pre-existing valuation studies. Principal advantages of the benefits transfer approach is avoidance of resource-intensive primary valuation studies with their associated time delays, including rapid assessment of uncertainties about deduction of the value of potentially significant services. Principal disadvantages are that values have to be transferred with a high degree of caution, applying conservative assumptions to address the context sensitivity of both the study and source value site, and that only a limited subset of relevant primary values may be available. For services where monetary valuation is not possible, it is essential to ensure that the systemic focus is retained by undertaking non-monetary valuation based on a qualitative assessment of implications for the full spectrum of services, and by implication benefits or detriment to all service beneficiaries. Such non-monetary valuation needs to proportionate to the scale, risk and degree of contention of the scheme (Dunn, 2012) but not be onerous, for example using the 'likelihood of impact' scoring system promoted by Defra (2007).
However, systemic context is essential to ensure that externalities are not unwittingly perpetuated in decision-making, supporting decisions that are equitable, resilient and optimal in terms of public value. Non-systemic valuation, in which only a predetermined service or subset of services are 'cherry picked', not only risks partial valuation but can lead to unintended negative consequences for non-focal services.

The use of value transfer
The robustness of value transfer depends on the availability of suitable pre-existing studies, success of the 'matching' of the policy site to an appropriate study site, appropriate and conservative assumptions used to adjust valuation between studies, and the quality of the original economic valuation study and its relevance to the present day. In this study, two different approaches to value transfer were used: 1) Unit value transfer: estimates the value of a good or service at the policy site by multiplying a mean unit value estimated at the study site (e.g. £/ha).
2) Meta-analytic function transfer: uses a value function estimated from a collection of studies. Specific parameter values for the policy site are introduced into the value function to calculate a transferred value which better characterizes the policy site. Eftec (2010b) identifies and provides background information on studies which are potentially relevant to the valuation of environmental effects associated with FCERM (flood and coastal erosion risk management) capital schemes. Error! Reference source not found. highlights the studies that have investigated the economic value of wetlands and wetland ecosystem services and that have been used for the purpose of value transfer. A limited set of economic valuation studies of ecosystem service outcomes at management realignment sites, based on value transfer methods, has been produced in the UK context. The ones used in this study which represent the most similar sites are Alkborough Flats (Everard, 2009) and the Blackwater Estuary Management Realignments (Luisetti et al., 2011). Brander et al. (2008) Saltmarsh Luisetti et al. (2011) 2 Methods This study used market values where available (such as goods with established use values) and value transfer techniques to estimate the costs and benefits of other marginal changes stemming from the SCMP.

Methodological framework
Figure 2-1 illustrates the methodological framework used in this study, identifying and briefly describing the main steps in the process.

Establish the Environmental Baseline
The environmental baseline aims to identify and categorise ecosystems and their services in the current state, providing a baseline from which to identify the location and type of ecosystem services impacted by the scheme.

Qualitative Assessment
This step seeks to identify and assess the scale of the potential impacts of the scheme on the full range of ES. The purpose here is to ensure that no ES has been missed or overlooked by taking a comprehensive approach from the start. Only ES with significant or unknown impacts are taken forward for economic valuation.

Quantification of Impacts
The purpose of this step is to analyse the extent to which the ecosystem provides the service and how the scheme may impact on that provision. It is the change that takes place in the different ES that is of interest, rather than the absolute level of service provided. In general terms, the level of service provision may be proxied by the extent of habitat (e.g. hectares of habitat).

Economic Valuation of Changes in Ecosystem Services
Economic valuation is estimated using market prices for services where a market exists 1 and through value transfer techniques for all the other services. Careful selection of valuation studies from which values can be transferred is essential to enable a good match between the existing valuation evidence and the case study.

Steart ecosystem services assessment in the context of previous research studies
Most of the research studies on valuation of ecosystem services, such as TEEB (2010a), UK NEA (2011), Luisetti et al. (2011), Eftec (2010 and Brander et al. (2008), make a clear distinction between ecosystem services and benefits derived by people. They then suggest that, due to the complexities and inter-relationships between supporting, provisioning, regulating and cultural services, placing a value on each individual service may not be appropriate. The approach taken in these studies is therefore to value end points of direct relevance to market values or non-use values, to avoid the problem of double-counting.
This study acknowledged and considered the complementarities as well as conflicts between services and the problem of double-counting as well as the complexities of being able to fully account for all changes to ecosystem services. However, this study followed the guidance of Everard and Waters (2013) by seeking firstly to value all services, and only then explicitly excluding values where double-counting would occur. The rationale for this is that it is unsafe to assume that societal values stemming from 'intermediate' ecosystem services are substantially internalised into their contribution to 'final services' and the production of goods. Our approach to averting the risk of continuing to externalise important services is to attempt to value all services, acknowledging that methods are currently imperfect and there is some reliance on prior meta-study values, then explicitly removing those where double counting is likely to occur.

Current main ecosystem types
Key habitats present on the Steart Peninsula (and at the project area) prior to MR are summarised in Table 3-1, with a brief outline of the main characteristics (adapted from Environment Agency, 2011).

Ecosystem type Description
Intertidal These include mudflat, saltmarsh, shingle and sand habitats and saline lagoons. These areas are predominantly within the Severn Estuary designated sites and all are UK BAP habitats.

Ditches and Pools
The extensive ditch networks are integral to the grazing marsh and support an important invertebrate assemblage as well as the nationally-threatened water vole.

Grazing Marsh (improved and semi improved permanent pasture)
There is approximately 500ha of grassland within the peninsula. Grazing marsh is a priority habitat under the UK BAP (Coastal Floodplain and Grazing Marsh) Most of the value of the grazing marsh is associated with wetland features (e.g. ditches and ephemeral standing water).

Cultivated Land (arable and grass leys)
Extensive areas of arable farmland, sown predominantly with cereal crops, may provide breeding habitat for ground-nesting birds such as skylark. Fallow land is also an important food source for insects such as bees, butterflies and moths.

Trees and hedgerows
Hedgerows are associated with many of the field boundaries but are largely absent from most seaward fields. An assessment of hedgerow ecological quality concluded that they are generally species poor. Very few mature trees are present, mainly associated with hedgerows on the higher ground on the perimeter of the site.

Current significant ecosystem services
The ES present at Steart Peninsula flow from these current habitat types and are summarized in Appendix A.

Qualitative assessment
The qualitative assessment of the potential impacts of the project on ecosystem services undertaken for this study is described and presented in Table 3-3 to Table 3-6 for provisioning, regulating, cultural and supporting services respectively.
This initial screening assessment was undertaken for each service using the semi-quantitative scoring system proposed in Defra (2007) (see Table 3-2), including specific additional services relevant for this scheme. The impacts scores for each ecosystem service were attributed based on a thorough review of all documents submitted for the project's planning application. In addition, expert and stakeholder consultation was undertaken to ensure that all significant impacts and evidence gaps were captured and that the scores were appropriate. A total of 20 people from 8 different organizations were consulted, including local and national experts from planning, wildlife, environmental regulator (Environment Agency), fisheries, hydrology, navigation, and academic organisations, seeking to represent a broad spectrum of stakeholder interests. Cobbles and other beach aggregates are not actively collected, and such practices are likely to be discouraged due to the coastal defence benefits to which they contribute. No net change anticipated unless new markets are created. Creation of freshwater and intertidal habitats will significantly improve regulation of water flows on site, and modelling demonstrates that it will reduce the tidal flood risk. There will be no adverse impact on the fluvial flood risk.

++
Natural hazard protection (e.g. storm, flood, landslides) The removal of existing defences to create new habitat will require some replacement of defence further inland to protect Steart village and the access road. Although these will be constructed to the existing standard of protection, by doing so the EA will be able to continue the maintenance of the defences for the next 20 years (15 years longer than if there was no MR). It is also anticipated that there may be enhanced protection from storm surges on the River Parrett.

Pest regulation
Azolla has been discovered at the site, and therefore there is potential for expansion of population with the creation of new wetlands. Potential for seaweed (Enteromorpha) growth.
Not seen as having a significant impact with proper management practices in place. ?

Disease regulation
Potential creation of favourable conditions for waterrelated diseases, particularly in the light of climate change. Intertidal habitats are however good in preventing microbial diseases. Not seen as having a significant impact with proper management practices in place.
? Erosion regulation HR Wallingford (2011) concluded that any changes in physical processes are confined mainly to the area surrounding the entrance to the realignment with negligible changes elsewhere. Over time the estuary morphology will reach a new equilibrium. 0

Water purification and waste treatment
Recreated wetlands will provide a potential significant improvement in natural water treatment (wetland purification processes). Specific treatment wetlands will also be created to treat flows from visitor facilities. In addition, there will be a reduction in current negative agricultural impacts to water quality as fertilizer spreading will cease, potentially leading to improved water quality on the river Parrett and therefore contributing to EA's obligations under the Water Framework Directive (WFD).

++ Pollination
There may be a change in species pollinating/pollinated. Not seen as having a significant impact.  (2011) indicates that, with an appropriate mitigation strategy in place, the residual effects on the archaeological receptors potentially affected by the scheme will have no significant effect.
The scheme will bring additional cultural interpretation of the site providing education to visitors and celebrating the site's recent development and international importance.
Overall considered that there is a potential net positive effect.

Recreation and tourism
The site is used by local residents and visitors for recreational activities, including dog walking, cycling, bird watching, and horse riding. The development, subject to sensitive management, will enhance these activities bringing more visitors into the site. Planning Solutions Consulting Limited & DT Transport Planning (2011) indicate a 3-fold increase in the number of visitors.

Aesthetic value
Environment Agency (2011) assessed the effects on landscape character and visual receptors during operation. The overall assessment of the development concluded that they would have a beneficial longterm impact on the area's landscape character and visual amenity, through the replacement of a managed, agricultural landscape with a more diverse, sustainable and natural one.

Spiritual and religious value
Potentially enhanced by the spiritual value of water and increased sense of wilderness. + Inspiration for art, folklore, architecture, etc.
Potentially enhanced by a more diverse and native landscape with new interpretation signs and viewing points attracting more artists, photographers, etc.

Soil formation and retention
The majority of the managed realignment site will accrete due to sediment deposition thus creating the environments sought for the habitat creation scheme. Accretion will continue until the intertidal area reaches the elevation of existing saltmarshes in the Parrett.

Primary production
The creation of a more complex/varied habitat may lead to enhanced assimilation and accumulation of energy and nutrients by organisms, potentially improving primary productivity. ?

Nutrient cycling
The creation of new intertidal habitat will enhance nutrient processing while significantly reducing the input of fertilizers formerly applied to converted agriculture land. Equally applied to seasonally inundated grassland in freshwater area.

Water cycling
Wetland habitats are efficient in recycling water at local scale and it is therefore anticipated a significant net change from current land use.

++ Photosynthesis
(production of atmospheric oxygen) Oxygen generation has not been quantified. There is possibly an increase from a more complex/varied habitat, however, some tree and hedgerows will be lost. ?

Provision of habitat
Intertidal habitat creation is the main driver for this project aiming at increasing local biodiversity. Mitigation strategies will be put in place to protect legally-protected species such as water voles, badgers, newts, etc. that may die or get injured during the saline inundation of Area D.

++
The above tables show that the project will yield significant benefits across all of the four ecosystem service categories. It should be noted that, for each ecosystem service, there may be some positive and negative impacts within them, but a judgement has been made by the authors in collaboration with a stakeholder group as to the overall semi-quantified impact.
The ES with significant or unknown impacts to take for further evaluation are: Provisioning services • Food Regulating services • Climate regulation • Water regulation • Water purification and waste treatment Cultural services • Recreation and tourism • Education All supporting services The services of pest and disease regulation, although with unknown impacts, are not taken forward as their impacts were not deemed significant with appropriate management practices in place.

Food
This service considers agriculture and fishery goods. Agriculture goods are quantified based on the changes in available area of arable and grassland fields and their consequent productivity in terms of arable and livestock output. Valuation is estimated based on annual farming income using market prices. Fishery goods quantification is associated with enhanced fish biological productivity and recruitment in intertidal habitats and consequent improved catch for commercial purposes. Quantification and valuation were based on the value transfer method.

Agriculture
Environment Agency (2011) quantification and valuation estimates have been used as shown in Table 3-7 . Detailed information on how these figures have been derived are provided in Annex B.

Summary of impacts on farm income within the project area
At the current time, the project area has the ability to generate agricultural net income of the order of £152,500 (gross income -costs + subsidies). Considering the future scenario with less agriculture output income but higher income from agri-environmental schemes the project area will generate income from £79,750 to £190,250 annually. This means that overall the project area may incur from an annual loss of £37,750 to an annual benefit of £72,750. It should be noted that the consideration of impacts on farming incomes does not factor in the added value of future specialist/niche farming opportunities (e.g. saltmarsh lamb, salicornia harvesting, shellfisheries). These are growth markets in the UK and have the potential to make a significant contribution in the overall annual income.

Fisheries
A few studies have been conducted on intertidal fish communities in North West Europe, to which a likely contributory factor are inherent sampling difficulties in large tidal ranges.
Fonseca (2009) studied fish utilisation of newly-created habitats and adjacent mature saltmarshes in the Blackwater Estuary in East England. Quantitative sampling was undertaken at three locations and a total of 18 samples were collected. Bass was selected for further analysis, since this was the only significant commercial species present in the catches. Table 3-9 appears in Fonseca (2009) and is reproduced in Luisetti et al. (2011) as part of an overall economic analysis. The table shows the value of bass per hectare contributing to the inshore fishery after five years (the age at which bass will most likely have attained the minimum length eligible for harvesting) and for every year thereafter. This work remains the only reported quantitative study of its kind in Europe and has therefore been used for value transfer to our policy site despite a high degree of uncertainty and assumptions. In this study only sea bass is considered for economic valuation (ignoring therefore recruitment of other important commercial fish species and shellfish). It was deemed therefore that this could provide at least a minimal figure to the potential of the policy site for commercial fishing.
In this way, for 232 ha of new saltmarsh area to be created at Steart and considering the upper and lower values per hectare at average wholesale price of sea bass estimated in Luisetti et al. (2011) the project area will be able to yield a benefit of around £450 to £11,100.

Greenhouse gas sequestration
The biggest net change in habitat conversion, with significance for changes in greenhouse gases emissions (GHG), is the creation of 232 ha of saltmarsh from both arable (50%) and improved grassland (50%). The greenhouse gases considered most significant at this site are: carbon dioxide (CO2), methane (CH4) and nitrous oxide (N2O).
Following Eftec (2010b) guidance, to determine the change in emissions associated with a change in habitat, an estimate of net (or equivalent) carbon sequestered per hectare (e.g. t CO 2 e/ha) has been estimated. This is to allow for the interactions between the three greenhouse gases referred above.
According to Adams et al. (2011), Blackwater MR saltmarshes are currently sequestering C at a rate of 2.23 t CO2e/ha/yr (medium value).
According to Alonso et al. (2012), agricultural practices based around grasslands are predicted to be a net carbon sink while arable land is a net source. Freibauer et al. (2004) cite mean carbon flux measurements of 0.60 t C/ha/yr and -0.83 t C/ha/yr respectively in a review of European soils. These values however only refer to soil C stocks and have been converted into (CO 2 e) 2 equivalence (grasslands: 2.20 t CO 2 e/ha/yr; arable: -3.04 t CO 2 e/ha/yr). A net C sequestration change of about 615 t CO 2 e/yr is then estimated.
Emissions arising from FCERM schemes that map to the non-traded carbon sector, which is the case for emission from land use change (e.g. habitat gains and losses), should be valued using the 'non-traded price of carbon' (NTPC). The NTPC increases over time, so the values applied here were selected as relevant to the timescale of the study. DECC (2010) guidance specifies a lower-upper bound range of £25 -75/t CO2e. In this way, the scheme will be able to yield a benefit of about £15,375 to £46,125 per year.

Microclimate
It is likely that when compared to arable fields, mudflat, saltmarsh and reedbed habitats will transpire different quantities of moisture and will form a more differentiated microclimate. However, there are no apparent methods at present to quantify this change unless local measurements in temperature are undertaken at the current site and in a saltmarsh with similar characteristics to Steart (e.g. proximity to the sea).

Water regulation
This service focuses on the environmental benefits/costs resulting from changes in the water regulation function of water systems when we create space for water by realigning the defences.
The creation of a more diverse habitat composed by saline and freshwater lagoons, reedbeds, saltmarshes, creeks, etc. is likely to provide a more natural hydrology which will benefit fish and other aquatic wildlife. The benefits of improved hydrology for fish and other estuarine wildlife are however going to be assessed under 'recreation and tourism' and 'provision of habitat' services respectively. In this way, and to avoid double-counting, this service is not assessed discretely.

Water purification and waste treatment
This ecosystem service considers two aspects: 1) The benefits resulting from costs saving of using wetland habitat for wastewater treatment in lieu of any existent conventional treatment; and 2) The environmental benefits stemming from improved water quality.

Wastewater treatment
Currently the run-off flows from agriculture fields and the manure collection of the chicken farm do not pass through any kind of conventional treatment, but are instead washed away and ultimately discharged into the River Parrett. Two treatment wetlands are planned as part of changes on the Steart site. Despite the improved water quality both by treatment wetlands and reduced chemicals used in agriculture, no costs savings are achieved since no conventional water treatment are being currently undertaken in the area. In this way, there are deemed to be no benefits to be assessed on wastewater treatment.

Environmental benefits from water quality improvement
The environmental benefits stemming from improved water quality, such as enhanced fish productivity and other aquatic wildlife, are captured by the estimation of their human welfare impact via amenity/recreational and habitat values, and therefore are not considered in this section to avoid double-counting.

Recreation and tourism
Existing recreational uses of the Steart Peninsula were investigated through a dedicated visitor assessment survey undertaken by Planning Solutions Consulting Limited & DT Transport Planning (2011). The findings indicate that there are up to 11,200 visitors to the Steart Peninsula every year mainly due to the Bridgwater Bay NNR.
The creation of new wetland habitat combined with the extensive development of the multiuse path network and a series of new observation points and hides offers significant opportunities for recreational uses. The visitor assessment survey estimate that visitors to the project site will increase to about 33,000 per annum once the site is fully established and that from these, 8% of the visitors will be holiday makers, 12% will be locals and the remaining 80% will be day visitors for which the Steart wetlands reserve is the main reason for the visit.
Values were transferred from 3 RSPB case studies provided by RSPB (2011) which assessed the main drivers of economic activity and their impacts on 10 of their reserves (Table 3-11). These reserves were chosen based on their similarity with Steart's particular characteristics in terms of remoteness/landscape, habitat and geography/socio economic conditions. It was assumed that local visitors bring additional economic activity through their visit. On this basis, the Steart visitor projection will yield a benefit in the order of £300,840 to £469,310 per annum. Whilst recognising that this figure reflects total expenditure, it is assumed that displaced expenditure (i.e. not additional but due to visitors coming to Steart rather than other local sites) will be balanced by value-add to other economic activities (shops, catering, etc.) occurring locally to the Steart site. It would have been possible to draw upon the extensive literature on the value of recreation, but at this point a detailed breakdown of likely recreational activities was not foreseeable (though angling and birdwatching are recognised as significant contributors).
In previous Sections, it has been considered that the improved hydrology and water quality leading to enhanced fish biological productivity would be estimated through their human welfare impact via recreational values (e.g. enhanced angling activities). Even though the visitor assessment study, as shown in Appendix I, does not explicitly refer to angling visits, it has been assumed that their increased numbers is expressed in the overall assessment. It has also been assumed that the average spend by anglers is similar to the average spend of other types of visitors.

Education
The Steart project will bring enhanced opportunities for public engagement and formal learning of the site, to include schools, colleges, research and other interested groups. Here we consider the benefits that enhanced primary, secondary and tertiary education from school trips and potential 'citizen science' projects bring to society.
According to the demographic profile data provided on the Steart visitor assessment survey, there are 112,275 school aged children (5 to 14 year olds) living with the 60 minute drivetime catchment. There are 180 state primary schools and 53 secondary schools in Somerset. Mourato et al. (2010) considered case studies of both school trips to UK nature reserves and a national 'citizen science' project. These case studies provide just the 'cost of investment' involved in these visits and not the true economic valuations of educational benefits concerned. Nevertheless, assuming that these undertakings are good value for money, such costs should provide a lower bound minimum of the values concerned (UK NEA, 2011).
Transferring the values from Mourato et al. (2010): • £850,000 to £1.3 million of expenditure value on 2,000 school trips per annum (i.e. £425 to £650 per school trip per annum) • £188 expenditures per participating school on citizen science project And adjusting these into the local number of schools and pupils involved and assuming that: • Each trip involves 28 pupils; • 5% of school children aged 5 to 14 years old will undertake school visits to Steart (i.e. 5,600 students and 200 schools trips); and • 5% of schools (i.e. 10 schools) will participate in one annual citizen science project at Steart.
The project will be able to yield a benefit per annum of £87,000 to £132,000. Such an assessment is likely to provide only a very lower bound investigation of such values. As for 'Recreation and tourism', costs are taken as a proxy for benefits as some of this value is not wholly additional (for example displacing values from other sites) yet the contribution of ecosystem services to the tertiary education sector is unaccounted for (though occurring already, this would require further analysis).

Soil Formation
Long-term rates of saltmarsh accretion across the wider Severn Estuary, which take account of compaction, range between 0.46mm/year and 10.5mm/year. No valuation case studies were found to be able to transfer a value into the policy site and therefore no value has been attributed to this service. Nevertheless, the site's potential to develop similar habitat to the saltmarshes elsewhere on the Parrett has been confirmed through accretion rate estimates. It should be noted that soil formation has a strong inter-relationship with carbon sequestration and other supporting services and its economic evaluation could potentially have lead to double-counting of benefits.

Primary Production
According to the UK NEA (2011) 'primary production is the fixation by photosynthesis of either atmospheric or aquatic carbon dioxide, or its assimilation as organic compounds by plants and algae'. As such, primary production both influences, and is influenced by, the supporting services of nutrient cycling, water cycling and soil formation.
Assessment of status and trends of primary production across UK habitats is limited (UK NEA, 2011). There are currently many knowledge gaps which limit full quantification of primary production in both terrestrial and marine habitats. For this reason, it has not been possible to quantify net changes in primary production due to habitat conversion. In addition, even if quantification was possible, there is a strong likelihood that at least a proportion of it might double-count with other ecosystem services as primary production underpins the delivery of all other ecosystem services (i.e. regulating, provisioning and cultural services).

Nutrient cycling
Wetlands are a major provider of water quality improvement benefits through their ability to recycle nutrients. Adams et al. (2011) estimated that potential MR areas within the Blackwater Estuary (29.5 km 2 saltmarsh and 23.7 km 2 intertidal mudflat) could bury 695.5 t N/yr , with a further 476 t N/yr denitrified. The MR saltmarsh was also able sequester 139.4 t P/yr. These figures show the potential of MR intertidal habitats to reduce estuarine nutrient loads which can be particularly relevant in achieving the WFD goals.
The net change in nutrient loads in Steart was not possible to quantify due to unknown current geochemical data. Nevertheless, this service will be captured under 'provision of habitat' through its impact on water quality improvement, avoiding therefore a potential element of double-counting.

Water cycling
According to the UK NEA (2011), the water cycle considers the major water fluxes (rainfall, evapotranspiration, river flow) and the major water storages (soil, groundwater, lakes) that combine to determine the availability of water in time and space.
Our ability to predict future variability in the water cycle in space and time relies on the hydrological models being coupled to climate models capable of producing rainfall fields at an appropriate scale and time resolution. Further research in terms of local hydrological modelling would be required in order to be able to quantify this service.

Photosynthesis
This service is tied closely to primary production and there are currently no pragmatic means for quantifying or valuating it.

Provision of habitat
Although there is substantial anecdotal evidence of non-use (existence and bequest) values associated with maintaining biodiversity, the estimation of such values is problematic (UK NEA, 2011). Most commentators argue that this can be better achieved by the application of estimates of individual preference, with the most common approach to assessing the non-use value of biodiversity being via Stated Preference (SP) studies. Others would argue that a lower boundary estimate of values could be depicted by the payments provided by policies designed to promote biodiversity (e.g. opportunity costs or biodiversity offsets). The later approach however needs to be treated with caution, with the potential circularity of the valuation process being recognised (UK NEA 2011). Given this, the SP approach is adopted and the valuation of habitat creation is estimated by using the meta-analytic function transfer method used in Eftec (2010b) previously described in section 1.4.
According to Eftec (2010b) and Morris & Camino (2010), the recent meta-analyses of wetland valuation provided by Brander et al. (2008) provide the most appropriate value transfer function for valuation of UK wetlands, also in the context of FCERM schemes. It permits control of factors such as habitat type, ecosystem service provision, wetland size, availability of substitutes and affected population that should be controlled for in value transfer analysis.  These figures only consider the provision of the following services: biodiversity, water quality improvement, non-consumptive recreation and aesthetic amenity (only for the lower bound estimates). The services 'omitted' from the specified meta-analysis function are: flood control and storm buffering, surface and groundwater supply, commercial fishing and hunting, recreational hunting, recreational fishing, harvesting of natural materials and fuel wood. This will avoid double-counting with services which have already been estimated separately or that are not relevant for this study. Note that there is potential for doublecounting in terms of the non-consumptive recreation and aesthetic amenity elements, as some of this is addressed through assessment of 'Recreation and tourism' benefits. Although Brander et al. (2008) do not make it clear precisely how questions behind their stated preference surveys leading to meta-analysis figures were framed, there are also nonlinearities of these services with respect to habitat size. Furthermore, service values are also non-additive. In the absence of more detail about the finer breakdown of likely nonconsumptive recreation and aesthetic amenity uses and consequent benefits, the cumulate figure provided by Brander et al. (2008) covering the three elements of biodiversity, water quality improvement, and non-consumptive recreation/amenity is reduced by 1/3 to minimise the potential double-counting and to ensure that estimates are conservative. This approach may be somewhat arbitrary, but is considered indicative of the scale of benefit; to attempt to be more precise about a figure derived from a meta-study transferred to a different site would be to ignore the substantial uncertainties introduced by multiple assumptions in the initial studies, during aggregation and on transfer.
To account for the 'marginal' values of the services, the upper bound estimates of Table 3-14 are used. Considering that the project will create 232 ha of saltmarsh and 30 ha of intertidal mudflat and assuming that the area of substitute wetland available is less than 100 ha, a net annual benefit of £125,240 to £182,467 is estimated. Table 3-15 summarises the results of the economic valuation for the ES assessed.

Efficacy of the Ecosystem Approach
This study highlights the significance of a broad range of values beyond the traditional costs and benefits of near-market goods and services (usually the provisioning services). Ecosystem services provide the analysis with a basis for expressing the broader societal benefits or costs consequent from ecosystem-based interventions, offsetting the often implicit political perception that biodiversity conservation and environmental protection measures are necessarily a net cost and constraint upon economic development (Everard, 2009). This in turn supports taking an Ecosystem Approach (as defined by the Convention on Biological Diversity: www.cbd.int/ecosystem) which takes account of wider impacts on ecosystems, their services, all affected stakeholders and their inclusion in decision-making and economic contexts.
Many of these benefits are assessed on the basis of some wide assumptions, and are subject to further uncertainties where surrogates have to be applied to derive values. However, despite all uncertainties, when those values are assessed, the scale of public benefits arising from improved ecosystem functioning can be significant. This provides a clear argument in favour of ecosystem-based interventions, as opposed to more narrowly-framed engineered solutions, and for assessments of likely scheme outcomes based on ecosystem services.

Economic benefits of the Steart Costal Management Project
This study clearly evidences the economic benefits from a range of ES that can be provided by coastal wetland habitats/MR schemes, which exceed estimated values stemming from bird conservation for which much more evidence has been collected.
The view of this project as favouring regulatory services (i.e. water regulation), supporting services (i.e. biodiversity) and cultural services (i.e. recreation) to the detriment of provisioning services (i.e. crops) is no longer supported. In fact, were we able to (i) properly quantify the contribution of intertidal habitat to fish and shellfish biological productivity (which has both significant food and recreational values), (ii) value all commercially important fish species present within the area, and (iii) estimate the contribution of added value agriculture outputs such as saltmarsh lamb and Salicornia, it is likely that significant additional benefits could be estimated. The management of habitat for wider public services does not necessarily imply a 'trade-off' with other private benefits, even if it may imply a trade-off between beneficiaries and a shift in the benefits provided. This may help inform how ecosystem services can be used as a framework for the beneficial redirection of agricultural subsidies to deliver wider public benefits.
We recognise that some of the values deduced are more robustly supported than others, reflecting their closeness to market. However, to fail to assign sometimes illustrative values to as many services as possible, emphasising only reductive analysis of services more robustly monetised, is to fall into the trap of failing to consider systemic outcomes: the key purpose of this paper. We also recognise three aspects of 'costs': (1) expenditure entailed in scheme implementation; (2) the negative consequences of failing to implement the scheme; and (3) any ecosystem service disbenefit outcomes from scheme implementation. The latter costs are addressed directly. The former two 'costs' are confounded by the primary driver for scheme approval being as a cheap option to address EU Habitats Directive obligations to mitigate habitat loss elsewhere in the wider Severn system, in addition to baselines costs associated with existing defences reaching end-of-life and so risking collapse. As cost savings elsewhere in the Severn and in terms of failing defences are omitted, we perceive assessment only of direct costs versus cumulative net present value of service benefits as a highly conservative approach to assessing cost-benefit.

'Winners and losers' and spatial scale of ES provision
Whilst the economic benefits provided by this scheme are recognized, one other important issue to consider in ES assessment is their distribution: the determination of the 'winners and losers' within the scheme based on change in the balance of ecosystem services and associated beneficiaries. Although the loss of some services will be (to varying degrees) compensated in economic terms by gains in other services, these services and the benefits that flow from them may not be received by the same individuals or groups. In this case, there is a shift from services largely benefiting local farmers across to those benefitting the wider public. Although the main beneficiaries of the scheme are the wider public, the implications for current farmers are not clear. Future grazing opportunities at the site are likely to benefit a different group of farmers. Failure to evaluate the overall impacts of the scheme on the different groups of winners and losers is a limitation of this present study, but is significant in supporting the acceptability of future development schemes.
In addition to the distributional factor, spatial scale is also an important aspect for the valuation of ES considered in this study. The spatial scale at which ES are supplied and demanded contribute to the complexity of ES valuation (Brander et al., 2008). From a supply point of view, the coastal wetlands provide, for example, recreation opportunities (on-site), flood protection at Steart village (local off-site), education (local and national off-site) and climate regulation (global off-site). From a demand perspective, beneficiaries of ES also vary in terms of their spatial distribution, and of course recognition of the value of the service through markets. For many services, as already observed, there are not effective markets to link the supply and enjoyment of these benefits across scales.

ES and economic benefits in the context of previous studies
Many research gaps and complexities found with ES valuation at Steart have already been reported by Everard (2009) and other UK studies. This demonstrates a clear need for further research and long-term monitoring at MR sites once they are established.
Temporal scales also add complexity to ES evaluation. Most ES assessment studies in the context of FCERM schemes, such as Luisetti et al. (2011) and the MR case studies provided at Eftec (2010b), estimate the Net Present Value (NPV) of ES benefits which are calculated for a certain time horizon (typically for 50 and 100 years) based on a specified discount process. This study did not attempt to estimate a NPV as the annual average contributions are a more useful means to assess benefits and disbenefits between services and their various beneficiaries. Were we to need to calculate a NPV, this should take account of the different time horizons associated with the many constituent ecosystem services (e.g. agriculture outputs were assessed for a 20 year time horizon whilst climate regulation estimates underlie a longer time horizon of about 100 years).
The complexities and inter-relationships between the supporting services and the provisioning and regulating services are evident throughout this study and explain why recent major studies such as UK NEA (2011) and TEEB (2010a) have moved away from the traditional MEA services categorization to explore ES economic valuation and rather focus on 'final services' or 'benefits' with direct impacts on human welfare.

Recommendations for improved sustainability and public value
In order to maximize the benefits of such schemes, a full ES assessment should ideally be undertaken earlier in the design/development stage to enable the conclusions and interplay/trade-offs between potential services to be properly factored into decision making and design. One potential element of this scheme which could have lead to enhanced environmental benefits was a proper consideration at the design stage of improved conditions for eel recruitment at the site.

Stakeholder engagement
Engagement with people is essential in undertaking ES assessment. Key stakeholders should be identified and engaged early in the ES assessment appraisal. In this study, only a focussed group almost entirely supportive of or with a vested interest in the scheme was consulted. While all efforts have been made to avoid the introduction of bias as a result, future studies should engage a broader range of stakeholders with diverse interests.

Conclusions
This study clearly demonstrates the economic benefits from a range of ecosystem services that can be provided by coastal wetland habitats. Despite having to rely on some wide assumptions and uncertainties a conservative, yet considerable, net annual benefit range of £491,155 to £913,752 has been deduced. It became apparent throughout the study that many research and knowledge gaps exist with respect to several ecosystem services, particularly the supporting services for which market values are clearly elusive. This is of concern as it affects our current ability to quantify and/or value them, and hence to include these important aspects of ecosystem integrity, functioning and resilience into decision-making. These gaps therefore demonstrate a clear need for further research, both theoretical and through longterm monitoring of these schemes once established.

Straw.
Fresh water* Surface water used for local agriculture.
Two ponds supplied by shallow fresh groundwater.
Two ponds supplied by shallow fresh groundwater.

Main Ecosystem Types
Types of service Intertidal Ditches and Pools Grazing Marsh Cultivated Land Trees and hedgerows

Ornamental resources
Possible cobbles extraction. Flotsam and jetsam (e.g. driftwood, objects from other countries).

Air quality regulation
Fine particulates from sea-salt aerosol and beach sand. Creation of wind breaks.

Pest regulation
Azolla noted in the north of the peninsula.

Disease regulation
'Black leg' reported to be an issue with cattle in the area where freshwater habitat will be created.

Water purification and waste treatment
Natural purification/filtration. Natural purification/filtration. Natural purification/filtration.

Pollination
Habitat for pollinating species.
Food source for pollinating species.
Habitat for pollinating species.

Cultural heritage
(archaeology, historic structures, historic landscapes, maritime archaeology and palaeoenvironment) Historic landscape with archaeological features and palaeochannels.
Historic landscape with archaeological features and palaeochannels.
Historic landscape with archaeological features and palaeochannels.
Historic landscape with archaeological features and palaeochannels.
Historic landscape (parish boundaries) with archaeological features. Public footpaths providing visual amenity.

Main Ecosystem Types
Traditional landscape.
Public footpaths providing visual amenity.

Spiritual and religious value
Contact with nature. Contact with nature. Contact with nature. Contact with nature.
Potential opportunities for public engagement and education (formal and informal).

ADDENDUM: Employment
Tourism/recreation related, agricultural employment.
Agricultural employment.

Agricultural
Employment.

Main Ecosystem Types
Types of service Intertidal Ditches and Pools Grazing Marsh Cultivated Land Trees and hedgerows

Primary production
Saltmarsh plants.
Fresh/brackish water and vegetation cover.

Nutrient cycling
High in nutrientsimportant for natural food chain.
Part of nutrient transfer to intertidal habitats.

Water cycling
Part of water cycling. Part of water cycling. Part of water cycling. Flood storage. Vegetation complexity can enhance recapture of evaporation.

Photosynthesis
(production of atmospheric oxygen) Production of oxygen. Production of oxygen. Production of oxygen. Production of oxygen. Production of oxygen.