Biogeochemical Value of Managed Realignment, Humber Estuary, UK ⁎ J.E

Science of the Total Environment 371 (2006) 19–30 www.elsevier.com/locate/scitotenv

Biogeochemical value of managed realignment, Humber estuary, UK ⁎ J.E. Andrews a, , D. Burgess 1, R.R. Cave 2, E.G. Coombes a, T.D. Jickells a, D.J. Parkes a, R.K. Turner b

a School of Environmental Sciences, University of East Anglia, Norwich NR4 7TJ, UK b CSERGE, School of Environmental Sciences, University of East Anglia Norwich NR4 7TJ, UK

Received 9 May 2006; received in revised form 10 August 2006; accepted 12 August 2006 Available online 25 September 2006

Abstract

We outline a plausible, albeit extreme, managed realignment scenario (‘Extended Deep Green’ scenario) for a large UK estuary to demonstrate the maximum possible biogeochemical effects and economic outcomes of estuarine management decisions. Our interdisciplinary approach aims to better inform the policy process, by combining biogeochemical and socioeconomic components of managed realignment schemes. Adding 7494 ha of new intertidal area to the UK Humber estuary through managed realignment leads to the annual accumulation of a 1.2×105 tof‘new’ sediment and increases the current annual sink of organic C and N, and particle reactive P in the estuary by 150%, 83% and 50%, respectively. The increase in intertidal area should also increase denitrification. However, this positive outcome is offset by the negative effect of enhanced greenhouse gas emissions in new marshes in the low salinity region of the estuary. Short-term microbial reactions decrease the potential benefits of CO2 sequestration through gross organic carbon burial by at least 50%. Net carbon storage is thus most effective where oxidation and denitrification reactions are reduced. In the Humber this translates to wet, saline marshes at the seaward end of estuaries. Cost–benefit analysis (CBA) was used to determine the economic efficiency of the Extended Deep Green managed realignment. When compared to a ‘Hold-the-Line’ future scenario, i.e. the present state/extent of sea defences in the estuary, the CBA shows that managed realignment is cost effective when viewed on N25 year timescales. This is because capital costs are incurred in the first years, whereas the benefits from habitat creation, carbon sequestration and reduced maintenance costs build up over time. Over 50- and 100-year timescales, the Extended Deep Green managed realignment scenario is superior in efficiency terms. The increased sediment accumulation is also likely to enhance storage of contaminant metals. In the case of Cu, a metal that currently causes significant water quality issues, Cu removal due to burial of suspended sediment in realigned areas translates to a value of approximately £1000 a−1 (avoided clean up costs). Although this is not formally included in the CBA it illustrates another likely positive economic outcome of managed realignment. Although we focus on the Humber, the history of reclamation and its biogeochemistry is common to many estuaries in northern Europe. © 2006 Elsevier B.V. All rights reserved.

Keywords: Managed realignment; Cost–benefit analysis; Biogeochemistry; Humber estuary; Carbon; Nutrients; Metals

⁎ Corresponding author. Fax: +44 1603 591327. E-mail address: [email protected] (J.E. Andrews). 1 Current Address: Agricultural and Food Economics Division, Agriculture and Food Science Centre, Department of Agriculture and Rural Development (Northern Ireland), Newforge Lane, Belfast BT9 5PX, UK. 2 Current Address: Department of Earth and Ocean Sciences, National University of Ireland, Galway, University Road, Galway, Co. Galway, Rep. of Ireland.

1. Introduction Habitats Directive to follow a no-net-loss policy within large designated areas (Crooks and Turner, 1999). For over 300 years European management of coastal Our recent geological/geochemical cycles based lowlands and estuaries has been dominated by land research on the Humber estuary on the east coast of reclamation and flood protection, principally through the England (Fig. 1; Andrews et al., 2000; Jickells et al., construction of ‘hard’ sea walls and the drainage of 2000) has highlighted potentially important additional wetlands. However, set against a modern backdrop of issues related to sediment storage, and carbon and nutrient globally rising sea levels and potential for increased cycling in the intertidal and salt marsh zones that are also storminess due to climatic changes (Nicholls and Klein, relevant to managed realignment. In this paper we outline 2005), the cost of maintaining and upgrading existing sea a plausible, albeit extreme, managed realignment scenario defences, many of which have reached the end of their for the UK Humber estuary, to demonstrate the maximum design life, has prompted policy makers to reconsider possible biogeochemical effects and economic outcomes their long-term cost effectiveness. The recent flooding in of estuarine management decisions. Our interdisciplinary New Orleans, for example, has served to highlight the approach aims to better inform the policy process, by vulnerability of large agglomerations of people and combining the biogeochemical and socioeconomic com- economic assets which have continued to expand behind ponents of managed realignment schemes. Although we engineered coastal defences. It has been estimated that focus here on the Humber, the history of reclamation and 23% of the world's population lives near the coast, with a its biogeochemistry is common to many estuaries in density three times higher than the global average (Small northern Europe. Moreover, a number of the issues we and Nicholls, 2003). As we better understand coastal discuss have global relevance and emphasise the virtue of dynamics, hard defences are increasingly considered adopting an ‘ecosystem services’ approach to analysis and unsustainable both from an environmental and economic policy (Daily, 1997). There is concern, for example, that perspective (Crooks et al., 2001). Not only do hard globally increased nutrient fluxes are damaging coastal defences provide a false sense of security and encourage ecosystems (Jickells, 1998). It has been estimated that development immediately behind defences, they prevent 18% of the nitrogen inputs to the Mississippi are lost a natural geomorphic response to rising sea levels, within river catchments (Donner et al., 2002) and the whereby the intertidal zone migrates landward: preven- figure may be as high as 50% for the Rhine (Billen et al., tion of this by sea walls and flood embankments results in 1991). Most of this loss occurs by denitrification in ‘coastal squeeze’ (Pethick, 2001).Forexample,onthe wetlands. In coastal seas in general, and more inshore Essex coastline of the UK, the presence of medieval to waters in particular, wetland nutrient cycling has been 19th century embankments has caused the loss of estimated to represent their most valuable environmental 40,000 ha of saltmarsh (Dixon et al., 1998). service (Constanza et al., 1997). Similarly, wetlands Emphasis is now moving toward a mixed approach to represent the largest component of the global terrestrial coastal lowland management, protecting areas of high organic carbon inventory, with tidal saline wetlands value, whilst allowing coastal processes to proceed rela- storing in excess to 45 Tg C a−1 (Chmura et al., 2003): tively unhindered elsewhere. More flexible “soft engineer- carbon burial in saline wetlands is thus potentially an ing” measures such as managed realignment help effect important sink for atmospheric CO2 (Chmura et al., 2003; this. The term ‘managed realignment’,alsoreferredtoas Choi and Wang, 2004). Decadal-scale carbon burial rates ‘managed retreat’ or ‘coastal setback’ (Reed et al., 1999), in some coastal saline wetlands are higher than millennial- involves deliberately breaching engineered defences to scale rates (Choi and Wang, 2004), possibly a function of allow the coastline to migrate to a new line of defence increased productivity, which may be linked to increased landward of the old one. Managed realignment schemes (anthropogenic) nitrogen fluxes among other factors aim to re-site defences in a manner that not only reduces the (Choi et al., 2001; Choi and Wang, 2004), illustrating a length of defence required, but also increases the area of clear link between nutrient flux and carbon sequestration. intertidal habitat. The driver for managed realignment has Evaluation of managed realignment at the scale of an in most cases been flood defence: the renewed intertidal/ estuary or administrative region using methods such as cost saltmarsh zone acting as a natural sea defence by atten- effectiveness, cost–benefit analysis and/or multi-criteria uating wind wave height and tidal amplitude (Möller et al., assessment is an urgent policy requirement. The appropri- 1999, 2001; Pethick, 2002), providing a sustainable first- ate scale for the analysis is an important factor and em- line of coastal defence. The recreation of intertidal/ phasises the need to move away from a project or scheme- saltmarsh areas also has a biodiversity value and allows based appraisal scale (Bower and Turner, 1998). The task is government compliance with the European Union (EU) not trivial requiring an appreciation of the full functioning ..Adese l cec fteTtlEvrnet31(06 19 (2006) 371 Environment Total the of Science / al. et Andrews J.E. – 30

Fig. 1. Map of the Humber Estuary and Humberside showing localities mentioned in the text, present intertidal area, area below mean high water spring (MHWS) tides and at risk from tidal flooding, and areas of proposed ‘Extended Deep Green’ managed realignment sites. Note that the area below MHWS has all been claimed from the estuary over the last four centuries. 21 22 J.E. Andrews et al. / Science of the Total Environment 371 (2006) 19–30 of the sedimentological, hydrological, ecological and should also allow more extensive microbial processing of geochemical systems together with the societal value of nutrients such as nitrate (Jickells et al., 2000): improving the goods and services provided by such functioning. water quality by removing these compounds from water and storing or degrading them in the sediments. 2. Managed realignment in the Humber From a flood defence, habitats and biogeochemical/ water quality perspective, it thus follows that extensive The macro-tidal Humber estuary (Fig. 1) is one of the managed realignment is desirable. Our research was de- largest in the UK, with a maximum tidal length of 147 km signed to identify, by feasibility study, the .maximum and maximum width of 15 km, comparable with the UK possible area of Humberside that could realistically be Thames and Severn Estuaries (Andrews et al., 2000). The returned to the intertidal zone over the next few decades, area surrounding the Humber estuary (Humberside) is what we term an ‘Extended Deep Green’ future Scenario mainly high quality agricultural land, with many (an extension of the Deep Green Scenario described by thousands of hectares reclaimed from the estuary over Cave et al., 2003; Andrews et al., 2005). A number of the last few centuries (Fig. 1). The banks are extensively criteria were used to decide the feasibility of retreating sea urbanised in parts with over a third of a million people defences and flooding areas of land (Coombes, 2003; living on land below high spring tide level. Humberside is Table 1). These included avoiding areas where there was protected by flooding from rivers and the sea by 405 km of existing infrastructure such as roads and railways, coastal defences: approximately 870 km2 of land would avoiding built-up areas with the exception of isolated be flooded if the present coastal defences were removed buildings and where possible choosing areas of farmland (Coombes, 2003; EA, 2000; Winn et al., 2003). that had most recently been reclaimed from the sea. Fifty- In order to reduce flood risk and to minimise the cost seven areas where defences .could be set back were of maintaining flood defences around the estuary as sea identified (Fig. 1), with the potential to create a maximum level rises over the next century, the Environment of 7494 ha of new intertidal area (increasing the current Agency for England and Wales (EA) is currently in intertidal area by 69%), and leading to a reduction in the negotiation with local landowners for the use of areas of overall length of flood defences in the estuary of 121 km, land bordering the estuary, in order to realign some of the 30% of their current length (Table 2). sea defences. One scheme has already been successfully implemented at Paull Holme Strays on the north shore 3. Biogeochemical effects of managed realignment in (Fig. 1), where 80 ha of intertidal area has been recreated, the Humber after moving the sea defences landward and shortening them by taking advantage of an area of natural high Although sedimentation in realignment sites can be ground (EA, 2002). The EA intends to return some dramatically fast, for example, rates of up to 50 mm per 1000 ha of coastal land to intertidal area around the estuary, to improve overall the sea defences in the estuary Table 1 and to create intertidal habitat to compensate for losses Criteria used to decide feasibility of realigning sea defences/flooding land due to port developments (EA, 2003). Area below the high spring tide level to maximise area of potential While the EA managed realignment policy in the intertidal habitat. Present land use – undeveloped land most suitable for conversion due Humber is primarily to improve flood defence and main- to physical ease and relatively low economic value (Reed et al., 1999). tain commitment to the EU Habitats Directive, a spin-off SSSIs, SACs and other similarly protected areas may not be suitable. benefit of increasing the intertidal area is that it also Infrastructure – presence of urban areas, roads, railway, canals etc. increases the ‘accommodation space’ in the estuary for Historical context – land reclaimed recently, therefore easier to revert sediment accumulation, which allows increased burial of to mudflat or saltmarsh; presence of archaeological sites that need protecting. organic matter, particle reactive phosphate, and some Spatial context of the areas: contaminant metals (Andrews et al., 2000; Cave et al., . Size: realignment not cost-effective for areas under 5 ha (Pilcher et al., 2005). It has been demonstrated that sediment accumu- 2002). lation rate and biomass production is a key control on . Shape: trade-off between wide intertidal area to maximise benefits intertidal mudflat and saltmarsh carbon burial rates and length of realigned defences to protect surrounding land (Pilcher et al., 2002). (Chmura et al., 2003; Brevick and Homburg, 2004), . Elevation: higher ground used as natural defence to absorb wave arguably making them more desirable ‘climax realignment energy minimising length of defences and maintenance costs of the habitats’ than freshwater wetlands/peatlands, where short- realigned defences (O'Riordan et al., 2000). term rates of sedimentation are much slower. The . Proximity to existing intertidal habitats: to facilitate movement of increased intertidal surface area created by realignment species between habitats (Begon et al., 1996). J.E. Andrews et al. / Science of the Total Environment 371 (2006) 19–30 23

Table 2 Future scenario Length of hard Length of realigned Length of Intertidal habitat Total intertidal % Increase in defences after defences (km) unsatisfactory defences created by area (km2) intertidal area realignment (km) after realignment (km) realignment a (ha) Hold-the-line 405a 0 65 0 108 b 0 Extended 284 103 34 7494 183 69 Deep Green a Due to uncertainty over loss of intertidal habitat due to coastal squeeze in the next 50 years, it is assumed that no further coastal squeeze takes place under a ‘Hold-the-line’ scenario. This baseline ‘Hold-the-line’ scenario assumes for simplicity no changes in hard defences or loss of intertidal area. b Measured in this study and essentially the same as 111 km2 given in Andrews et al. (2000). month at Paull Holme Strays (EA, 2004), these are short rate of 1 mm per year. This mass of sediment is less than term ‘catch-up’ rates as low lying areas build up sediment half the present day estimated annual deposition of sedi- to regain equilibrium profiles. Long-term steady state ment in the intertidal and subtidal zones (cf. Townend and accumulation of sediment over the Humber intertidal and Whitehead, 2003) and so should not have a profound subtidal mudflats tracks mean sea level rise at about effect on estuary-wide net sediment budgets as long as 1mma−1 (Andrews et al., 2000) resulting in 1600 tonnes sediment inputs remain more or less as they are today. The (t) of sediment accumulating per square km of deposi- setback areas would be managed in order to facilitate the −3 tional area (using a bulk density of 1.6 g cm ,basedon creation of saltmarsh which is effective at burying Corg best estimates in Parkes (2003) and corroborated by and Norg (Andrews et al., 2000), primarily as below- values in Townend and Whitehead (2003), or a total of ground biomass (roots). Based on measured.net accumu- 2.4×105 t of sediment per year in the present-day estuary lation rates (i.e. accounting for microbial decomposition (Table 3). Using the known average concentrations of on decadal timescales – see below) of Corg in Welwick particulate carbon (C), nitrogen (N) and net particle re- saltmarsh (in the outer estuary, Fig. 1), this additional area active phosphorous (P, i.e. in excess of natural back- would act as a sink for about 3597 t Corg,180tNorg,and ground sediment P) in Humber mudflat sediments 72 t of particle reactive P, increasing the current annual (Andrews et al., 2000; Jickells et al., 2003), then about sink of these compounds in the estuary by 150%, 83% and 2400 t of organic C (Corg), 216 t of organic N (Norg)and 50%, respectively (Table 3). Clearly we cannot be sure 144 t of particle reactive P are buried in Humber sedi- that saltmarsh will establish quickly after realignment, so ments annually (Table 3). these figures, based on a highly simplified approach, must If the strategy of managed realignment could be taken be taken as maximum values that demonstrate the poten- to its maximum (see above), then significant additional tial of such a management scenario. − 1 removal of these compounds from water to sediments will An estuary total burial of 396 t a Norg (i.e. the occur. Adding 7494 ha (74.94 km2) of new intertidal area 216 t a− 1 buried annually today plus the additional through managed realignment would lead to the accumu- 180 t a− 1 buried under the Extended Deep Green Future lation of a further 1.2×105 t of sediment per year, as- scenario) represents b1% of the 44, 000 t entering the suming no change in the average long-term sedimentation estuary from the rivers annually (mid-late 1990s figures

Table 3 Storage terms for sediment, organic matter, nutrient elements and contaminant metals in the present day and realigned Humber estuary

Sediment Corg Norg Net P Net Zn Net Pb Net As Net Cu (t a−1) (t a− 1) (t a− 1) (t a− 1) (t a− 1) (t a−1) (t a− 1) (t a−1) Present-day total estuary area (muds only) a 2.4×105 2400 216 144 31 16 4 6 EDG b new intertidal area (7494 ha) 1.2×105 3597 180 72 21 10 4 4 Total area post-EDG realignment 3.6×105 5997 396 216 52 26 8 10 % Increase of current annual storage 50 150 83 50 68 62 100 67 a Total present-day mud (intertidal+subtidal) area=150 km2 (Andrews et al., 2000; Table 5). Sediment accumulation is based on an estimated long- term sea-level rise of 1 mm a−1 and bulk density of 1.6 g cm− 3 (see text). The metals' values are based on means of core top data from around the estuary (Cave et al., 2005), and cross-compared to other data (Ridgway and Rees, 2001; Cox, 1999) to check values are representative. (Note that some hotspots in the estuary have much higher concentrations than these.) b EDG=Extended Deep Green Scenario. Net element data for EDG realignment are based on our unpublished 1984–1994 data measured from Welwick Marsh (Fig. 1), corrected for pre-industrial (Holocene) background concentrations in Middleton and Grant (1990) and Ridgway and Rees (2001).These values are not directly comparable to those in Cave et al. (2005), due to a change in the assumed bulk density values (1.6 g cm−3 in this paper). 24 J.E. Andrews et al. / Science of the Total Environment 371 (2006) 19–30 from Neal and Davies, 2003). However, the increase in .capital cost from the physical realignment of the sea intertidal area would also increase the amount of walls and an.opportunity cost from the loss of economic denitrification in the estuary (Jickells et al., 2003). value of the former land-use of the land lying seaward of Denitrification rate is a strongly non-linear function of the newly aligned defences (Table 4). The capital costs nitrate concentration, the rate increasing rapidly with of realigning the Humber defences were based on concentration before plateauing at about 1000 μmol l− 1 contemporary realignment schemes (Halcrow, 2000; M. concentration (Nedwell et al., 1999). Thus denitrifica- Dixon, personal communication to E. Coombes, 2003). tion rates should be much higher in the low salinity As most of the land identified for conversion is un- region of estuaries like the Humber with high nitrate developed and primarily used for agriculture, the value concentrations, suggesting that creation of an area of of the agricultural land was used as a proxy for the wetland in the inner estuary or tidal rivers will provide a opportunity cost. The average sale prices of agricultural more effective method of nitrate removal than an land were obtained from DEFRA (2004) for each of the equivalent area at the seaward end. Our recent mod- Agricultural Land Grades (http://www.defra.gov.uk/ elling (as yet unpublished) indicates that about 40% of environ/landuse/alcleaflet.pdf). However to account ammonia species entering the tidal Ouse (Fig. 1), and for the impact of agricultural subsidies inflating the 55% entering the tidal Trent (Fig. 1), is converted to sale price of agricultural land, the opportunity cost was nitrite or nitrate before reaching the estuary. The derived by multiplying the sale price by 0.3 (DEFRA, remaining ammonia species, augmented by direct inputs 2001a; Dickie and Pilcher, 2001). to the estuary (some 5000–8000 t of ammonia species), In addition to the capital cost of realignment, both the together with any nitrite, are converted to nitrate within existing and newly aligned defences must be maintained the estuary. This net DIN source to the estuary exceeds over their 100-year design life (DEFRA, 2003)toa nitrate removed by denitrification. An increase of standard satisfactory for the protection of infrastructure 7494 ha of intertidal area in the tidal rivers and estuary and urban areas. Just over a quarter of the existing would convert all ammonia species in the tidal rivers to defences in the Humber are in an unsatisfactory con- nitrate, and reduce the input of total DIN to the estuary dition (EA, 2000) and require replacement. Clearly the by some 200 t a− 1. Some 600 t a− 1 of nitrate within the standard of protection required varies with land-use, to estuary itself would be removed by denitrification. reflect the value of damage caused by flooding. (The About 57% of the area identified for recreation of intertidal area was in the tidal rivers (Fig. 1); clearly if further suitable areas could be identified, then the deni- Table 4 trification potential would be further enhanced. Values used to estimate the costs and benefits of realignment before and after standardising to the financial year 2004–2005 by using GDP deflators 4. Costs and benefits of Extended Deep Green recommendedbyHMTreasury(http://www.hm-treasury.gov.uk/) realignment Item Value of item at Year Value of item time of reference of after adjustment study to the financial To help reduce uncertainty and aid decision-making, year 2004–2005 cost–benefit analysis (CBA) was used to determine the − − Capital costs of £811, 893 km 1 2001– £878, 159 km 1 economic efficiency of managed realignment in the realignment 2002 Humber. Five possible future scenarios (Turner, 2005; (realigning defences) Andrews et al., 2005) were constructed but of these we Opportunity costs £2, 110 ha− 1 2001– £2, 282 ha−1 compare here only a ‘Hold-the-Line’ scenario, i.e. the (Grades 1 and 2 land) 2002 −1 – −1 present state/extent of sea defences in the estuary with the Opportunity costs £2, 382 ha 2001 £2, 576 ha (Grade 3 land) 2002 Extended Deep Green scenario. The latter in this context Maintenance costs of £3560 km− 1 a− 1 2005 Years 0–4 translates to the absolute maximum theoretically achiev- defences (years 0–4) £3560 km− 1 a− 1 able amount of managed realignment in the estuary and £3170 km− 1 a− 1 Year 5 − − the tidal reaches of the Rivers Ouse and Trent, without (year 5 onwards) £3170 km 1 a 1 − 1 −1 adversely affecting infrastructure (Table 2). Replacement costs £618, 000 km 2001 £668, 441 km General habitat creation US$ 2003/ £132– In CBA, the costs and benefits of a project are benefits 211 ha−1 a−1 1990 621 ha− 1 a− 1 compared over a fixed time horizon and subject to a – US$ discounting procedure, to determine whether the project 306 acre− 1 a− 1 will produce a net gain or loss in economic welfare for Carbon sequestration £7 per tonne 2000 £7.77 per tonne society as a whole. Realigning sea defences incurs both a benefits CO2e CO2e J.E. Andrews et al. / Science of the Total Environment 371 (2006) 19–30 25 annual probability of failure ranges from 0.5% for tion can be treated independently from the other amenity/ intensively developed urban areas to 20% for low-grade recreation functional outcomes. Various approaches agricultural land.) Maintenance costs in this CBA were exist for estimating the monetary value of carbon stor- based on those in use by the EA (2005) and included age. In this study we made a modest monetary estimate costs of replacing unsatisfactory defences not affected (£7 t− 1 C Table 4; based on Pearce, 2003; Tol et al., by realignment, the costs being derived from contem- 2000) of the global environmental damage done per porary defence schemes (DEFRA, 2001b). tonne of carbon equivalent emitted into the atmosphere – Quantification of realignment benefits should, in the “damage cost avoided” by storing rather than re- principle, encompass all relevant social welfare gains leasing a given quantity of carbon equivalent units. Some that could be assigned to the new investment: e.g. the damage cost avoidance estimates are as high as benefits associated with the new landscape/amenity, £70 t− 1 C, but corroborative evidence from the carbon recreation, biodiversity/habitat and carbon storage oppor- credit trading market (related to the Kyoto Protocol and tunity. These values are difficult to quantify because of the greenhouse gas emissions taxation regimes) and from interrelated and hierarchical nature of ecosystems and emissions abatement costs information, suggests that a their functioning outcomes. CBA also requires that all value per tonne of C between £7 and £15 is reasonable. benefits (and costs) be expressed in monetary terms. To enable a comparison of costs and benefits on Published valuation studies have to be interpreted with equal terms, all the values in the CBA were standardised caution and screened for a number of potential limitations and discounted to the financial year 2004–2005. The (see Balmford et al., 2002; Turner et al., 2003; Shepherd et effects of price changes due to inflation were removed al. in press; for such analysis). In this study a conservative using UK Government GDP deflators (http://www.hm- figure (£574 ha−1 a−1 Table 4) was used based on a meta- treasury.gov.uk/; Table 4). As costs and benefits accrue analysis by Woodward and Wui (2001),representinga at different time periods over the life span of the de- composite recreation and amenity environmental value fences, for example maintenance costs; present values for wetlands. were determined using a declining discount rate: years Extension of intertidal/wetland habitat will also 1–30, 0.035; years 31–75, 0.03 and years 75–100, increase storage and microbial degradation of nutrient 0.025 (HM Treasury, 2003). elements and carbon (see above).. However, a subtle but The results of the CBA (Table 5) show that managed important issue in CBAwhen aggregating benefits is that realignment schemes are cost effective when viewed on of double-counting, i.e. where a benefit (or a cost) is N25 year timescales, principally because most of the included twice within the evaluation process and there- capital costs, including replacement of unsatisfactory fore gains more importance within the final decision than defences, are incurred in the first year, whereas the it deserves. Double-counting is an issue in CBA for benefits from habitat creation, carbon sequestration and realignment sites because wetland ecosystems are multi- reduced maintenance costs (through realignment reduc- functional, with some benefits (and costs) dependent on ing the length of defences), build up over the lifespan of a combination of functions. If nutrient (N and P) re- the project. Thus, viewed over a 25-year period the tention is integral to the maintenance of biodiversity Extended Deep Green scenario is not as economically (Barbier, 1994) then assigning separate monetary values, efficient as Holding the Line. However, over 50- and that are then aggregated, would double count the nutrient 100-year timescales, the Extended Deep Green scenario retention value which is already ‘captured’ in the bio- is superior in efficiency terms, an outcome that supports diversity value (Shepherd et al., in press). To circumvent other CBA results in the UK (Shepherd et al. in press). this problem our CBA included only one composite environmental benefit value (.habitat creation benefit): 5. Microbial reactions and greenhouse emissions? nutrient storage functional benefit has not been valued independently (Shepherd et al., in press). We infer that From the above discussions it might seem that the the improved water quality impact resulting from N and ‘biogeochemistry of managed realignment’ leads to wholly P storage will be valued as part of the general envi- positive outcomes. However, this is a simplification. ronmental amenity gain. This approach probably under- During carbon burial, some of the near surface sediment estimates the total economic value of the environmental carbon is oxidised by microbial-mediation to other species, benefits of managed realignment schemes (see also including, among others, carbon dioxide (CO2), methane Shepherd et al., in press). The carbon burial (.carbon (CH4) and nitrous oxide (N2O). Production of these species sequestration) function has not been included in the is potentially important as they are all greenhouse gases, all composite environmental value estimates as this func- with different global warming potential (GWP). GWP 26 J.E. Andrews et al. / Science of the Total Environment 371 (2006) 19–30

Table 5 organic matter in intertidal mudflats and saltmarsh below ‘ Net present values (NPV) of providing flood defence for the Extended 50 cm is dominated by sulphate reduction. Deep Green’ scenario as compared to the ‘Hold-the-Line’ scenario for the Humber Estuary Because our measure of carbon burial is essentially the .net un-reactive carbon that is buried in the sediment long- Scheme 25 years 50 years 100 years term (Andrews et al., 2000), rapid microbial respiration by .Hold-the-Line scenario (HTL) oxidation, denitrification and sulphate reduction, which Costs . occurs on monthly–annual timescales before deep burial, Capital (£) 0 0 0 Maintenance (£) 23,199,000 34,813,000 43,632,000 is mainly accounted for in the net burial term. Once Replacement (£) 43,181,000 43,182,000 43,181,000 sulphate is exhausted, methanogenesis, which occurs over .Benefits much longer timescales (decades–millennia) may take Habitat creation (£) 0 0 0 place, producing not only CO2 but also the more powerful Carbon sequestration 000 greenhouse gas CH . Methane emissions are not likely to (£) 4 .Total value (HTL)(£) −66,380,000 −77,995,000 −86,813,000 be significant in marine mudflats where dissolved sulphate supply from seawater, which inhibits methanogens .Extended Deep Green scenario (EDG) (Stumm and Morgan, 1981), is essentially infinite. .Costs Although methane emissions from marine saltmarsh are Capital (£) 108,909,000 108,909,000 108,909,000 measurable, they vary by seven orders of magnitude Maintenance (£) 22,161,000 33,255,000 41,678,000 Replacement (£) 22,727,000 22,727,000 22,727,000 (Table 6): in northern latitudes they appear to be very low, .Benefits either essentially zero (Giani et al., 1996)oraround −2 −1 Habitat creation (£) 81,351,000 118,764,000 150,703,000 0.06–0.5gCH4 m a , although at lower latitudes Carbon sequestration 489,000 713,000 905,000 −2 −1 emissions could rise to about 5 g CH4 m a .Clearly (£) we need.in situ data measured on the Humber saltmarsh to Total value (EDG) (£) −71,957,000 −45,414,000 −21,706,000 .NPV (EDG vs. HTL) .−5,577,000 .32,581,000 .65,107,000 be confident about methane flux, but using these figures (£) as a guide and applied to the area of temperate saltmarsh All values are in Great Britain Pounds and rounded for clarity. created by managed retreat, they yield likely values of −1 between 0 and 0.45–3.75 t CH4 a .Thisrange,basedon the GWP figures given above translates to between zero accounts for the infrared absorption property of a gas, the elapsed time before that gas is purged from the atmosphere Table 6 and any indirect effects caused by feedbacks. In a 100-year Published values for methane flux from saltmarshes time-frame CH has a 21 times more powerful GWP than 4 Degrees CH4 flux Site Reference (country) − 2 − 1 CO2, while N2O has a 310 times more powerful GWP than (gCH4 m a ) description CO2 (Houghton et al., 1995). Production of these 30° 4.3 Saltmarsh DeLaune et al., 1983 (east greenhouse gases in recreated mudflat or marshland 97.0 Brackish coast USA) could thus offset the benefits of carbon burial as a means 37° 1.3 Saltmarsh Bartlett et al., 1985 (east coast USA) of reducing greenhouse atmospheric CO2. The amount of CO returned to the atmosphere (i.e. not 37° 5.6 Saline Bartlett et al. (1987) (east 2 22.4 Brackish coast USA) buried permanently) by aerobic oxidation depends on a 18.2 Brackish/ number of factors including the amount of organic matter fresh sedimented, the sediment grain size, amount of root tubes 41° 1.6 Saline Howes et al. (1985) (east and burrows and degree of water-logging. Fine-grained marsh coast USA) 45° 0.07–0.09⁎ Saltmarsh Magenheimer et al. (1996) intertidal sediments and pioneer saltmarsh in the Humber ⁎ ( based on 150 days of CH4 are mostly anoxic below the surficial 2 mm, such that emission) (east coast USA) aerobic reactions only occur in a small volume of sedi- 54° 0–0.07 Saltmarsh Giani et al. (1996) (German ment. By contrast, in more mature and better drained N Sea) saltmarsh in the outer Humber at both Welwick and 52° 0.26–0.48 Saltmarsh Senior et al. (1982) (eastern Tetney (Fig. 1) the upper 50–60 cm are essentially oxic, (pan) England) 0.06–0.11 Saltmarsh becoming more sulphidic below 50 cm. Once oxygen is (creek) consumed the next significant alternative electron accep- − In this study the data for sites N of 45° latitude are taken as indicative tors utilised by microbes are nitrate (NO3 ; see discussion 2− of likely fluxes for the Humber. Data for sites S of this show that below) and sulphate (SO4 ). Sulphate is a major ion in methane fluxes vary significantly with latitude, probably responding to seawater and this means that microbial destruction of temperature. J.E. Andrews et al. / Science of the Total Environment 371 (2006) 19–30 27

−1 −1 and 216 equivalent tonnes of CO2 a or 0 to 59 t C a . come suggests that short-term microbial reactions de- When compared to the maximum burial estimate of crease the potential benefits of CO2 sequestration through −1 3595 t Corg a (see above), the likely maximum methane gross organic carbon burial by at least 50%. It is thus fluxes account for b2% of the carbon burial term, worth emphasising that net carbon storage will be most suggesting that methanogenesis in saltmarshes has little effective where both oxidation and denitrification reac- effect on the decadal–centennial burial term. tions are limited or reduced. In the Humber context this If methane flux from minerogenic temperate saltmarsh translates to relatively wet (high water table), saline is proven to be zero or very low (Giani et al., 1996; Chmura marshes at the seaward end of estuaries where nitrate et al., 2003), this suggests that managed realignment at the concentrations should be lowest. This outcome suggests seaward end of estuaries will be a better option for carbon that management of, for example, ammonia species in the burial than at the landward end or in adjacent river valleys. Humber tidal rivers, is better addressed by focussing on This is simply because in freshwater marshes and peat- reducing inputs rather than decreasing concentrations by lands, where sulphate (and hence sulphate-reduction) is biogeochemical reactions in sediments. In a wider con- limited, methanogenesis is a much more important path- text, for example in smaller, less agriculturally dominated − way for microbial degradation of organic carbon and thus a catchments, where NO3 inputs are lower, the effects of contribution to the greenhouse gas inventory (Bartlett and denitrification will be smaller (see above). Harris, 1993; Whiting and Chanton, 2001). Overall, the result corroborates the conclusions Although emissions of the greenhouse gas N2Ofrom reached above for methane, that managed realignment saltmash/mangrove are generally considered negligible at the seaward end of estuaries will be a better option for (Smith et al., 1983; DeLaune et al., 1990)wehave net carbon burial. Clearly, in some estuaries the optimum suggested above that for nitrate reduction (denitrifica- location of managed realignment for biogeochemical − tion), the riverine end of estuaries with high NO3 inputs value could conflict with optimum locations for flood will be the most efficient sites, the reverse of the outcome defence or habitat creation which are the usual ‘drivers’ for high carbon storage/low methane production. Hence for realignment. creation of wetlands in low-salinity high-nitrate areas will maximise denitrification, but also maximise N2Opro- 6. Added value of managed realignment: storage of duction. Of the three possible products of bacterial nitrate contaminants reduction processes – nitrate ammonification produces ammonium as a product and hence no loss of fixed The Humber estuary has a history of industrialisation nitrogen from waters, while denitrification produces di- and resultant contamination, beginning in Roman times nitrogen (N2)andN2O gases. The controls on the relative when lead mining began in the upper parts of the catch- proportions of these three possible products are not well ment, and peaked during the 18th century. Such conta- known, but it appears that as nitrate concentrations mination has been negligible since the 19th century and increase, the relative proportions of ammonium and then iron ore mining ceased in the 1980s. The catchment is N2 declines. N2O is thus always a minor product, now in a ‘post-industrial’ phase, with mining of coal probably accounting for b2% of the N species produced having fallen greatly and textile industry in decline. Iron but increasingly important at higher nitrate concentrations and steel production, chemicals and food processing are (Nedwell et al., 1999; Seitzinger et al., 2000). still important manufacturing industries but subject to Under our Extended Deep Green scenario about rigorous pollution control. Direct contaminant dis- 43 km2 (57% of the total recreated intertidal area) of the charges into the estuary have decreased greatly since realigned area is in the tidal rivers where denitrification is the 1980s as a result of the provision of treatment for expected to be high. We estimate that this new intertidal industrial and sewage effluents, and factory closure. area will account for about 140 t a−1 of nitrate removal by Sediment cores from the estuary document the record denitrification. Assuming a value of 2% N2Oproduction of metal inputs and indicate maximum burial in the 1970s during denitrification this equates to about 0.63 molar (e.g. Lee and Cundy, 2001, Cave et al., 2005). Despite this −1 tonnes of N2O–Na . Based on the GWP figures (see evidence of anthropogenic trace metal storage in the above) this translates to 8620 equivalent tonnes of sediments, the reduction of sediment storage capacity −1 −1 CO2 a or the equivalent of 2350 t of buried C a . through 300 years of reclamation (Andrews et al., 2000), This figure is identical to the maximum net burial estimate accompanied by increasing inputs, have prevented the −1 of 2350 t Corg a in the total realignment (see above), and estuary from acting as a natural large-scale sink for trace would be higher if any significant denitrification occurs in metals, which over decadal timescales largely by-pass the the newly created outer estuary intertidal area. This out- estuary to the North Sea. 28 J.E. Andrews et al. / Science of the Total Environment 371 (2006) 19–30

Clearly if managed realignment results in increased effects of such remobilisation are difficult to judge, sediment storage capacity a more effective sink for although when compared to the ongoing effects of contaminant metals and other particle reactive contami- dredging in the estuary, the overall deleterious effect is nants should result. Using the maximum areas calculat- likely to be small. Even if realignment causes resuspen- ed for the Extended Deep Green scenario, and assuming sion of muddy sediments that are more metal-contam- the space created fills with sediment with metal concen- inated than the sandier dredged material, the effect on trations similar to those measured in mudflat core tops water quality is likely to small compared to that caused from around the estuary in the mid 1990s (described in by dredging. For example, in the present-day Humber, Cave et al., 2005), the maximum possible managed dredging redistributes ca. 1×107 t of wet sediment realignment will allow annual storage of about 21 t of annually (Binnie et al., 1999). Much of this material is Zn, 10 t of Pb, and about 4 t of As and Cu (Table 3). metal-contaminated, some of it severely so (Murray These values are .net contaminant values (Table 3), i.e. et al., 1980). Some of the dredged material is deposited over and above the natural background concentrations in sites in the outer estuary, where it is likely to be mixed of these metals (discussed further in Cave et al., 2005). with cleaner sediment entering from the North Sea Storage of As under this scenario is highly significant, (Cave et al., 2005), potentially reducing the overall representing about 85% of the modern particulate As concentrations of contaminants, which may be benefi- flux to the estuary (Cave et al., 2005, Table 8). Storage cial in the longer term. Moreover, the silt and clay of Cu and Pb is less significant, representing about 9% component disturbed by dredging may be buried longer- of the modern particulate fluxes of these metals (Cave term in the new realigned areas. On balance, managed et al., 2005, Table 8), while Zn storage is trivial, realignment appears to offer an ‘added value’ function in representing b3% of the modern particulate Zn flux helping control contaminant metal outputs from the (Cave et al., 2005, Table 8). Creation of 69% more estuary to the North Sea. intertidal mudflats and saltmarsh environments under our Extended Deep Green realignment appears to in- 7. Concluding remarks crease non-linearly the storage of As in particular (Table 3), although this outcome is heavily dependent This interdisciplinary analysis focused on ‘ecosystem on the local concentration of As in Welwick Marsh services’ has demonstrated that a more flexible and (Fig. 1; the model for the Extended Deep Green geo- adaptable approach to coastal protection and sea defence chemical values) versus the estuary muds as a whole can play a significant role in future coastal management (see Table 3 footnotes). strategies. While not representing a panacea, managed In the case of Cu, one of the few Humber metals that realignment in appropriate contexts does represent an currently causes significant water quality issues, it is economically efficient solution. The relevant context possible to assign an economic value to the storage term parameters for such an approach include a minimum because Beaumont (2000) has undertaken a Humber- spatial scale at the level of an estuary or sub-catchment, specific abatement cost analysis for this metal. Based on a an appraisal time horizon of at least 50 years and a range of possible abatement options and avoided costs declining discount rate procedure; as well as the scien- (reported in Andrews et al., 2005, Fig. 6), 4 t a−1 of Cu tific understanding of the range of ecosystem functions reduction due to sediment burial in realigned areas trans- and services present in coastal areas. lates to a value of approximately £1000 a−1 (avoided clean up costs). Although this value has not been formally Acknowledgements included in the CBA it is illustrative of the likely positive economic effect of metal storage by burial in realigned This work was carried out as part of the EU funded sediments. To include such calculations directly into the EUROCAT project (EKV1/2000/00510) although some CBA, a further study of the impacts that contaminants of the data was collected during work on a Natural such as copper have on biodiversity and/or human health Environment Research Council (NERC) funded Land would be required. These impacts would then have to be Ocean Interaction Study Special Topic (GST/02/736). assigned an economic value commensurate with the Duncan Parkes was supported by an NERC Ph.D. social welfare losses involved. studentship (NER/S/A/2000/03328). Stephen Bennett Construction of extensive realigned areas will change and Greg Samways helped collect some of the metals the hydrodynamics of the estuary, and it is possible that data. We are grateful for the comments of three ano- long-buried (legacy) contaminants stored in sediments nymous reviewers who improved the impact of our will be remobilised by changed patterns of erosion. The paper. J.E. Andrews et al. / Science of the Total Environment 371 (2006) 19–30 29

References Cox, J. Coastal–estuary sediment transfers: a geochemical and mineralogical study of sources. Ph.D. thesis, 1999. University of Andrews JE, Samways G, Dennis PF, Maher BA. Origin, abundance and East Anglia, 321 pp. storage of organic carbon and sulphur in the Holocene Humber Crooks S, Turner RK. Integrated coastal management: sustaining estuary: emphasising human impact on storage changes. In: Shennan estuarine natural resources. Adv Ecol Res 1999;29:241–89. I, Andrews JE, editors. Holocene land–ocean interaction and Crooks S., Turner RK, Pethick JS, Parry ML. Managing catchment– environmental change around the North Sea, vol. 166. Geological coastal floodplains: the need for a UK water and wetlands policy. society of London, special publication; 2000. p. 145–70. Policy Analysis (PA) Working Paper 01-01, Centre for Social and Andrews JE, Beaumont N, Brouwer R, Cave RR, Jickells TD, Ledoux L, Economic Research on the Global Environment (CSERGE), et al. Integrated assessment for catchment and coastal zone University of East Anglia and University College London, 2001. management: the case of the Humber. In: Vermat JE, Bouwer L, http://www.uea.ac.uk/env/cserge/pub/wp/pa/pa_2001_01.pdf. Turner K, Salomons W, editors. Managing European coasts, past, Daily GC, editor. Nature's services: societal dependence on natural present and future. Berlin: Springer; 2005. p. 323–53. ecosystems. Washington, DC: Island Press; 1997. Balmford A, Bruner A, Cooper P, Costanza R, Farber S, Green RE, et DeLaune RD, Smith CJ, Patrick WH. Methane release from Gulf Coast al. Economic reasons for conserving wild nature. Science wetlands. Tellus 1983;35B:8-15. 2002;297:950–3. DeLaune RD, Patrick WH, Lindau CW, Smith CJ. Nitrous oxide and Barbier EB. Valuing environmental functions: tropical wetlands. Land methane emission from Gulf Coast wetlands. In: Bouwman AF, Econ. 1994;70:155–73. editor. Soils and the greenhouse effect. New York: Wiley; 1990. Bartlett KH, Harris RC. Review and assessment of methane emissions p. 497–502. from wetlands. Chemosphere 1993;26:261–320. DEFRA (Department for Environment, Food and Rural Affairs). Bartlett KB, Harriss RC, Sebacher DI. Methane flux from coastal salt FCDPAG3 flood and coastal defence project appraisal guidance: marshes. J Geophys Res 1985;90(D3):5710–20. Economic appraisal; 2001a. http://www.defra.gov.uk/environ/fcd/ Bartlett KB, Bartlett DS, Harriss RC, Sebacher DI. Methane emissions pubs/pagn/fcdpag3/default.htm. along a salt marsh salinity gradient. Biogeochemistry 1987;4:183–202. DEFRA. National Appraisal of Assets at Risk from Flooding and Beaumont, N.J. 2000. The assimilative capacity of the Humber estuary Coastal Erosion, including the potential impact of climate change. for copper: an interdisciplinary study. Ph.D. thesis, 2000. Report produced for the department by: Halcrow Group Ltd., H.R. University of East Anglia, Norwich. Wallingford, John Chatterton Associates. 2001b. http://www.defra. Begon M, Harper JL, Townsend CR. Ecology. Oxford: Blackwell gov.uk/environ/fcd/policy/NAAR1101.pdf. Science; 1996. DEFRA. Flood and coastal defence project appraisal guidance: Billen G, Lancelot C, Maybeck MNP. Si retention along the aquatic Economic appraisal: Supplementary note to operating authorities. continuum from the land to the ocean. In: Mantoura RFC, Martin J- Revisions to economic appraisal procedures arising from the new M, Wollast R, editors. Ocean margin processes in global change. HM Treasury “Green Book”; 2003. http://www.defra.gov.uk/ Wiley; 1991. environ/fcd/pubs/pagn/FCDPAG3supplementv3.pdf. Binnie, Black, Veatch (Consultants). Humber estuary environmental DEFRA. Agricultural land sales and prices in England quarterly; 2004. baseline study. The Environment Agency 1999. On-line edition, (16/06/04), http://statistics.defra.gov.uk/esg/ Bower B, Turner RK. Characterising and analysing benefits from integrated statnot/alp.xls. coastal zone management. Ocean Coast Manag 1998;38:41–66. Dickie I, Pilcher R. Coastal flood defence: recent economic evidence. Brevick EC, Homburg JA. A 5000 year record of carbon sequestration RSPB discussion paper; 2001. from a coastal lagoon and wetland complex, Southern California, Dixon AM, Leggett DJ, Weight RC. Habitat creation opportunities for USA, vol. 57. Catena; 2004. p. 221–32. landward coastal re-alignment: Essex case study. J Chart Inst Water Cave RR, Ledoux L, Turner K, Jickells T, Andrews JE, Davies H. The Environ Manag 1998;12:107–12. Humber catchment and its coastal area: from UK to European Donner S, Coe M, Lenters J, Twine T, Foley J. Modelling the impact of perspectives. Sci Total Environ 2003;314/316:31–52. hydrological change on nitrate transport in the Mississippi Basin Cave RR, Andrews JE, Jickells T, Coombes EG. A review of sediment from 1955 to 1994. Glob Biogeochem Cycles 2002;16, doi:10.1029/ contamination by trace metals in the Humber catchment and 2001GB001396. estuary, and the implications for future estuary water quality. EA (Environment Agency). The Humber estuary – shoreline manage- Estuar Coast Shelf Sci 2005;62:547–57. ment plan; 2000. p. 59. Chmura DGL, Anisfeld SC, Cahoon DR, Lynch JC. Global carbon EA. Humber tides news. Environment Agency; 2002. p. 6. December. sequestration in tidal, saline wetland soils. Glob Biogeochem EA. The Humber estuary: update on managed realignment. Planning for Cycles 2003;17(4):1111, doi:10.1029/2002GB001917. the rising tides report. Environment Agency; 2003. Report No. 9. Choi Y, Wang Y. Dynamics of carbon sequestration in a coastal EA. Humber tides news. Environment Agency; 2004. p. 6. February. wetland using radiocarbon measurements. Glob Biogeochem EA. Draft technical report of strategy development study of the Cycles 2004;18:GB4016, doi:10.1029/2004GB002261. Humber estuary flood defence strategy; 2005. Version No 3.0, May Choi Y,Wang Y,Hsieh YP,Robinson L. Vegetation succession and carbon 2005. sequestration in a coastal wetland in northwest Florida: evidence from Giani L, Dittrich K, Martsfeld-Hartmann A, Peters G. Methonogenesis carbon isotopes. Glob Biogeochem Cycles 2001;15:311–9. in saltmarsh soils of the North Sea coast of Germany. Eur J Soil Sci Constanza R, d'Ardge R, De Groot R, Farber S, Grasso M, Hannon B, 1996;47:175–82. et al. The value of the world's ecosystem services and natural Halcrow UK (Consultants). Humber estuary tidal defences, urgent capital. Nature 1997;387:253–60. works: 1. Little Humber to Thorngumbald Clough, Engineer's Coombes, EG. Habitat creation and loss within the Humber estuary, Report. Prepared for the Environment Agency. 2000. and the associated environmental and economic costs and benefits. HM Treasury. The Green Book: appraisal and evaluation in central M.Sc. thesis, 2003 University of East Anglia. government; 2003. 30 J.E. Andrews et al. / Science of the Total Environment 371 (2006) 19–30

Houghton JT, Meiro Filho LG, Callander BA, Harris N, Kattenburg A, Parkes, D.J. Storage and cycling of organic carbon and nutrients in Maskell K. The science of climate change: contribution of Working Holocene coastal sediments. Ph.D. thesis 2003, University of East Group I to the Second Assessment Report of the Intergovernmental Anglia, 257 pp. Panel on Climate Change. Cambridge: Cambridge University Pearce DW. The social cost of carbon and its policy implications. Oxf Press; 1995. 584 pp. Rev Econ Policy 2003;19:362–84. Howes BL, Dacey JWH, Teal JM. Annual carbon mineralization and Pethick JS. Coastal management and sea-level rise. Catena belowground production of .Spartina alterniflora. Ecology 2001;42:307–22. 1985;66:595–605. Pethick JS. Estuarine and tidal wetland restoration in the United Jickells TD. Nutrient biogeochemistry of the coastal zone. Science Kingdom: policy versus practice. Restor Ecol 2002;10:431–7. 1998;281:217–22. Pilcher R, Burston P, Davis R. Seas of change. Royal Society for the Jickells TJ, Andrews JE, Samways G, Sanders R, Malcolm S, Sivyer protection of birds report; 2002. D, et al. Nutrient fluxes through the Humber estuary – past, present Reed DJ, Spencer T, Murray AL, French JR, Leonard L. Marsh surface and future. Ambio 2000;29:130–5. sediment deposition and the role of tidal creeks: implications for Jickells TD, Andrews JE, Cave RR, Parkes DJ. Biogeochemical value created and managed coastal marshes. J Coast Conserv 1999;5:81–90. of intertidal areas – a case study of the Humber estuary. In: Ledoux Ridgway J, Rees JG. LOEPS geochemical studies: 2. The Humber L.(Editor), Wetland valuation: state of the art and opportunities for region. British geological survey technical report WB/98/54, further development. CSERGE working paper PA 04-01. Work- Nottingham; 2001.

shop organised for the Environment Agency by Environmental Seitzinger S, Kroeze C, Styles R. Global distribution of N2O emissions Futures Ltd and CSERGE, Norwich, UK, 2003, pp. 60-71. from aquatic systems: natural emissions and anthropogenic effects. Lee SV, Cundy AB. Heavy metal contamination and mixing processes Chemosphere Glob Chang Sci 2000;2:267–79. in sediments from the Humber estuary, Eastern England. Estuar Senior E, Lindstrom EB, Banat IM, Nedwell DB. Sulphate reduction Coast Shelf Sci 2001;53:619–36. and methanogenesis in the sediment of a saltmarsh on the east coast Magenheimer F, Moore TR, Chmura GL, Daoust RJ. Methane and of the United Kingdom. Appl Environ Microbiol 1982;43:987–96. carbon dioxide flux from a macrotidal salt marsh, Bay of Fundy, Shepherd D, Burgess D, Jickells TD, Andrews JE, Cave RR, Turner New Brunswick. Estuaries 1996;19:139–45. RK, et al. Modeling the effects and economics of managed Middleton R, Grant A. Heavy metals in the Humber estuary: .Scrobi- realignment on cycling and storage of nutrients, carbon and cularia clay as a pre-industrial datum. Proc Yorks Geol Soc sediments in the Blackwater estuary, UK. Estuar Coast Shelf Sci, in 1990;48:75–80. press. Möller I, Spencer T, French JR, Leggett DJ, Dixon M. Wave Smith CJ, DeLaune RD, Patrick WH. Nitrous oxide emission from Gulf transformation over salt marshes: a field and numerical modelling Coast wetlands. Geochim Cosmochim Acta 1983;47:1805–14. study from North Norfolk, England. Estuar Coast Shelf Sci Small C, Nicholls RJ. A global analysis of human settlement in coastal 1999;49:411–26. zones. J Coast Res 2003;19:584–99. Möller I, Spencer T, French JR, Leggett DJ, Dixon M. The sea-defence Stumm W, Morgan JJ. Aquatic chemistry. Hoboken, NJ: John Wiley; values of salt marshes: field evidence from north Norfolk. J Chart 1981. Inst Water Environ Manag 2001;15:109–16. Tol RSJ, Fankhauser S, Richels RG, Smith JB. How much damage will Murray L, Norton M, Nunny R, Rolfe M. The field assessment of climate change do? recent estimates. World Econ 2000;1:179–206. effects of dumping wastes at sea: 6. The disposal of sewage sludge Townend I, Whitehead P. A preliminary net sediment budget for the and industrial waste off the River Humber. MAFF Directorate of Humber estuary. The Sci Total Environ 2003;314/316:755–67. Fisheries Research; 1980. p. 35. Turner RK. Integrated environmental assessment and coastal futures. In: Neal C, Davies H. Water quality fluxes for eastern UK rivers entering the Vermat JE, Bouwer L, Turner K, Salomons W, editors. Managing North Sea: a summary of information from the Land–ocean European coasts, past, present and future. Berlin: Springer; 2005. Interaction Study (LOIS). Sci Total Environ 2003;314/316:821–82. p. 255–70. Nedwell D, Jickells T, Trimmer M, Sanders R. Nutrients in estuaries. Turner RK, Paavola J, Cooper P, Farber S, Jessamy V, Georgious S. Adv Ecol Res – Estuaries 1999;29:43–92. Valuing nature: lessons learned and future research directions. Ecol Nicholls RJ, Klein RJT. Climate change and coastal management on Econ 2003;46:493–510. Europe's coast. In: Vermat JE, Bouwer L, Turner K, Salomons W, Whiting GJ, Chanton JP. Greenhouse carbon balance of wetlands: methane editors. Managing European coasts past, present and future. Berlin: emission versus carbon sequestration. Tellus 2001;53B:521–8. Springer; 2005. p. 199–225. Winn PJS, Young RM, Edwards AMC. Planning for the rising tides: O'Riordan T, Andrews JE, Samways G, Clayton K. Coastal processes the Humber estuary shoreline management plan. The Sci Total and management. In: O'Riordan T, editor. Environmental science Environ 2003;314/316:13–30. for environmental management. 2nd edition. Harlow, UK: Prentice Woodward RT, Wui Y-S. The economic value of wetland services: a Hall; 2000. p. 243–66. meta-analysis. Ecol Econ 2001;37:257–70.