Ecological Economics 35 (200) 25 – 33

www.elsevier.com/locate/ecolecon

SPECIAL ISSUE

THE VALUES OF WETLANDS: LANDSCAPES AND INSTITUTIONAL

PERSPECTIVES
The value of wetlands: importance of scale and landscape
setting
William J. Mitsch a,*, James G. Gosselink b
a
School of Natural Resources, The Ohio State Uni6ersity, 2021 Coffey Road, Columbus, OH 43210 -1085, USA
b
Rt. 1, Box 496A, Rock Island, TN 38581, USA

Abstract

Wetlands have value because their functions have proved to be useful to humans. The unit value for some wetlands
also increases with human development (agriculture and urban) because of increased use and/or increased scarcity.
Yet, paradoxically, its functions can easily be overwhelmed in areas of heavy human development, thus lessening
those values. Thus wetlands appear to work best in the landscape as spatially distributed systems. Also, the value is
partially dependent on where they are found in the landscape, e.g., the degree to which a wetland is open to
hydrologic and biological fluxes with other systems, including urban and agricultural landscapes. A paradox of
assigning values to wetlands and other ecosystems is that it can argue for the replacement of one system with another
if a landscape view is not taken. Estimates of percent of landscape for various functions, e.g. water quality or flood
control, are presented. It is suggested that a range of 3 – 7% of temperate-zone watersheds should be in wetlands to
provide adequate flood control and water quality values for the landscape. © 2000 Elsevier Science B.V. All rights
reserved.

Keywords: Wetland value; Marginal value; Watershed management; Landscape ecology; Wetland economics

1. Introduction sons that wetlands are often legally protected

have to do with their value to society, not with
Wetlands function as part of the landscape with the abstruse ecological processes that occur in
or without the presence of humans. They have wetlands… Perceived values arise out of the func-
value because many of their functions have tional ecological processes… but are also deter-
proved to be useful to humans. We (Mitsch and mined by human perceptions, the location of a
Gosselink, 1993) summarized this point: ‘The rea- particular wetland, the human population pres-
sures on it, and the extent of the resource.’ In that
* Corresponding author. book (Mitsch and Gosselink, 1993, 2000) we went
E-mail address: mitsch.1@osu.edu (W.J. Mitsch). on to suggest a few general principles that should

0921-8009/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved.
PII: S 0 9 2 1 - 8 0 0 9 ( 0 0 ) 0 0 1 6 5 - 8
26 W.J. Mitsch, J.G. Gosselink / Ecological Economics 35 (2000) 25–33

be remembered when attempting to estimate the 1.4. Commercial 6alues are finite, whereas
value of wetlands. wetlands pro6ide 6alues in perpetuity

1.1. Wetlands are multiple-6alue systems Wetland functions and thus values have the
potential to last for a very long time. Modern
Wetlands do not just do one thing. They per- agriculture or industrial/commercial activity are
form many processes simultaneously and there- generally unsustainable and resource-depleting
fore they provide a suite of values to humans. (soil loss; use of fossil fuels) so the lifetime of
Optimizing for one is usually at the expense of these human-based alternatives is short-lived.
another. Even large public works projects have time spans
of 50–100 years. A corollary of this point is that
1.2. Most 6aluable ‘products’ of wetlands are once wetlands are lost through development, the
public amenities loss of their functions and values is often
irreversible.
While wetlands do provide economic ‘payoff’ to
individuals, mainly through the population values 1.5. Comparison of short-term economic gain
described above, most of the services of wetlands from a non-wetland commercial use of a piece of
accrue to the public at large. Thus wetland protec- landscape, with long-term wetland 6alues is often
tion is, properly, the domain of a representative not appropriate
government working in concert with private
landowners. Compensation for these values has The social trap of assigning value to wetlands,
become a controversial topic. in addition to the danger of comparing ecosys-
tems, is that it will always be possible to find a
1.3. The relationship between wetland area and more profitable use for the land if short-term
marginal 6alue is complex economic analyses are made. For example, wet-
land value estimates cannot compete very well
Traditional economics supports the general with the economic return of corn and soybeans in
concept that the less there is of some commodity, the Midwestern USA so drainage will always be
the more valuable it is. This is the general eco- economically favored over wetland conservation if
nomic law of scarcity. The concept sometimes traditional cost-benefit analysis is carried out.
applies to wetlands. When there are extensive
wetlands in a given area (e.g. the Louisiana delta, 1.6. Values are influenced by cultural bias and
the Everglades area in Florida) the conversion of economic system
wetlands to other uses is often viewed as a social
necessity because there are many more available There are good reasons why we wish to protect
wetlands elsewhere. In areas where almost all nature; developed countries, having taken care of
natural wetlands have been drained (e.g. Europe, the basic needs of their citizens, are particularly
Midwestern USA, etc.), the desire to conserve involved in protecting ecosystems, including wet-
what few wetlands are left is often very strong. lands, for their aesthetic as well as more func-
However, the idea of wetland conservation does tional attributes, not all of which translate into
not always work even in areas where almost all direct economic benefit. Other cultures, where the
natural wetlands have been drained, and where basic needs of food and shelter cannot be taken
critical thresholds exist that define the minimum for granted, have a different view of the econom-
ecologically acceptable amount of wetlands in a ics of wetlands. Many cultures do live in and
given region. If wetlands are too small, functions among wetlands and use them for daily subsis-
such as support of certain mammals or storage of tence—the production of food and fiber. Yet they
floodwater (for mitigation of floods) no longer generally leave the normal wetland functions in-
exist. tact. The values that we ascribe to wetlands are
 W.J. Mitsch, J.G. Gosselink / Ecological Economics 35 (2000) 25–33 27

not separate from the institutions and culture

from which we come.
Ascribing economic value to wetlands has been
an exercise frequently practiced in the past 25
years. Starting with the publication of ‘The
Southern River Swamp — A Multiple-use Envi-
ronment’ (Wharton, 1970) and ‘The Value of the
Tidal Marsh’ (Gosselink et al., 1974), a significant
number of papers have discussed attributed values
to wetlands for the services that they provide (e.g.
Mitsch, 1977; Farber and Costanza, 1987;
Costanza et al., 1989; Turner, 1991; Barbier, 1994;
Fig. 1. Ease of calculation, accrual of benefits, and probable
Gren et al., 1994; Gren, 1994; Bell, 1997).
importance of values of wetlands at different ecological scales.
Costanza et al. (1997) took these types of calcula-
tions one step further by estimating the public
service functions of all the earth’s ecosystems, tions where human populations have over-
including wetlands. whelmed the functions of the last remaining
This paper considers a number of landscape wetlands.
and scale phenomena that make generalizations 3. The hydrogeomorphic principle — wetland val-
about wetland values particularly difficult. These ues depend on the hydrogeomorphic location
include the following: in which they are found.
1. The scale principle — wetland values are dif- 4. The ecosystem substitution paradox — if dif-
ferent, accrue to different ‘stake holders’, and ferent ecosystems are ascribed different values
probably have different importance depending in a given landscape, recommending the sub-
on the spatial scale on which we base our stitution of more valuable types for less valu-
estimations. able ones would be a logical extension of
2. The marginal 6alue paradox — fewer wetlands economic analysis.
do not necessarily imply greater value in situa- Given these phenomena, the paper then consid-
ers the question of how much area of a watershed
Table 1
General categories of wetland values at three different ecolog-
needs to be wetland for certain watershed-scale
ical scales (from Mitsch and Gosselink, 2000) ecosystem values to occur.

Ecological scale Value

2. The scale principle
Population Animals harvested for pelts
Waterfowl and other birds
Fish and shellfish The values of wetlands occur at three levels of
Timber and other vegetation harvest ecological hierarchy — population, ecosystem,
Endangered/threatened species and biosphere (Table 1). The ease by which we
Ecosystem Flood mitigation can quantify these values are in the order: popula-
Storm abatement tion\ ecosystem\ biosphere, yet the importance
Aquifer recharge of the values may be in the opposite order (Fig.
Water quality improvement
1).
Aesthetics
Subsistence use
Ecological populations, generally harvested for
food or fiber, are the easiest values to estimate
Biosphere Nitrogen cycle
and agree on. The wetland environment provides
Sulfur cycle
Carbon cycle pelts from muskrats, mink and beaver. Waterfowl
Phosphorus cycle can be hunted or simply observed in wetlands.
Fish and shellfish use wetlands as spawning and
28 W.J. Mitsch, J.G. Gosselink / Ecological Economics 35 (2000) 25–33

conditions exist in the sediments. Under these

conditions wetlands may remove high concentra-
tions of nitrates through denitrification and hence
improve water quality. Even flood control, princi-
pally a physical factor that depends on basin
morphometry and location in the watershed, can
be enhanced or altered by the presence of plants
to slow down currents, increase transpiration and
shade water. These ecosystem values are real, but
their quantification is difficult and the benefits are
generally regional and less specific to individual
land owners.
At the highest level, the biosphere, we know the
Fig. 2. Overall value of additional wetland area to a given least about values, and benefits accrue to the
region as a function of human population nearby (based on entire world. Wetlands are estimated to cover
figure by King, 1997). Overall value is a product of population about 4 to 6% of the world’s land (Mitsch and
times functional value per capita. Functional marginal value of
additional wetland per capita does initiallly increase as popula-
Gosselink, 1993), yet they are \ 20% of the land-
tion increases as wetlands are becoming rare. At some point of scape in boreal regions where great expanses of
population density, however, these functions become taxed peatlands are found (Gorham, 1991). It is likely
with pollution, lost corridors, etc. and marginal functional that one of the most important roles of wetlands
value drops precipitiously for additional population increase. may be as linchpins of global climate change by
sequestering and releasing a major proportion of
feeding locations and we, in turn, harvest the fixed carbon in the biosphere (Mitsch and Wu,
result. Timber such as cypress (Taxodium) wood 1995).
and plants such as Spartina (salt cord grass),
Juncus (rush), Phragmites (reed), and Zizania
(wild rice) are harvested for a variety of wetland 3. The marginal value paradox
goods. Although the link between landowner and
timber harvest is direct, many wetland-dependent The term ‘value’ such as the value of pollution
resources (fish, shellfish, ducks, geese) are consid- control or nature appreciation depends on human
ered public, not private, resources. The land- perceptions. Thus, either in the absence of human
owner may profit from leasing his land for population or in situations where population is so
hunting or fishing, but neither he nor the owner dense as to make wetlands nonfunctional, wetland
of duck nesting grounds thousands of kilometers value is low or zero (Fig. 2). As human popula-
away gains directly by harvesting the wetland-de- tions increase from low to high, the marginal per
pendent species grown on their land. Similarly, in capita value of wetlands first increases for two
the case of a migratory aquatic resource such as reasons, both having to do with increased wetland
fish or shellfish, a coastal wetland owner has no scarcity per capita: first, human population; and
control over the harvest, nor does he profit from second, the conversion of wetlands to other non-
it because of his ownership. wetland uses such as housing and highways. Thus
At the ecosystem scale, wetlands provide flood a higher human population implies that there is
control, drought prevention, and water quality less land per capita available and therefore, on
protection. These are referred to as ecosystem average, fewer but more precious wetlands (Fig.
services because they are provided most effectively 2). But the marginal value of wetlands increases
when the abiotic and biotic parts of the ecosystem with human development (agricultural and urban)
are synchronized. Wetland plants under hot, eu- only to a point as wetland functions begin to be
trophied conditions produce excess carbon, mi- lost. For example, far ranging mammals are elim-
crobial communities proliferate, and anaerobic inated as wetlands become smaller sized and frag-
 W.J. Mitsch, J.G. Gosselink / Ecological Economics 35 (2000) 25–33 29

mented; aquatic species are lost as wetlands be- number of legitimate human enterprises, the sys-
come isolated from streams. Sediments and pollu- tem can be overwhelmed and no longer perform
tants stress other wetlands. When human the services the community has grown to expect.
population increases to the point where the land is
an urban-suburban sprawl, a wetland’s functions
can easily be overwhelmed with too much pollu- 4. The hydrogeomorphic principle
tion or even too many bird watchers. In this
circumstance wetlands are no longer effective in The value of wetlands depends on both their
reducing floods, sequestering pollutants, or even hydrogeomorphic position in the landscape and
supporting a diverse biota that is of interest to the positions of human settlements, near and far,
hunters, anglers, or bird watchers. Thus wetland who find value in these ecosystems. Their hydro-
value appears to be maximum when distributed geomorphic position means the degree to which a
spatially across a landscape that is not dominated wetland is open to hydrologic and biological
either by cities or agriculture, but one that bal- fluxes with other systems, including urban and
ances nature and human enterprises. agricultural landscapes. As stated by us (Mitsch
To illustrate this point, a wetland in a natural and Gosselink, 1993): ‘Regional wetlands are inte-
riparian landscape may transform small quantities gral parts of larger landscapes—drainage basins,
of naturally produced stream nitrate-nitrogen to estuaries. Their function and their values to peo-
nitrogen gas and support a range of fish, terres- ple in these landscapes depend on both their
trial animals and birds. In the absence of either an extent and their location. Thus, the value to man
upstream or downstream human population (not of a forested wetland varies. If it lies along a river
likely except in the most remote locations on it probably has a greater functional role in stream
earth) this wetland is considered of little value to water quality and downstream flooding than if it
humans. When a town develops upstream of the is isolated from the stream. If situated at the
wetland and large quantities of nutrients and sed- headwaters of a stream, a wetland would function
iments enter the wetland, generated by land clear- in ways different from those of a wetland located
ing, agricultural fertilization and urbanization, its near the stream’s mouth. The fauna it supports
value to humans rapidly increases. As a second depend on the size of the wetland relative to the
town develops downstream, the value increases home range of the animal. Thus to some extent
even further. But eventually, when the riparian each wetland is ecologically unique. This compli-
system loses its natural functions because it is cates the measurement of its ‘value’.’
leveed off from the river or destroyed for any
4.1. Geomorphic position
Table 2
Examples of wetland position in the landscape and related In a geomorphic sense, wetlands can be
probable values
classified as in-stream systems, riparian systems,
Position in Enhanced value isolated basins, and coastal (fringe) systems
landscape (Table 2). The different types of wetlands can
provide different values. In-stream wetlands pro-
In-stream Fisheries, organic export cess large amounts of water and inflows approxi-
wetland
Riparian Detrital production; sediment retention;
mately equal outflows. But high productivity of
wetland wildlife corridor; flood control; nitrogen these systems translates to enhanced aquatic food
and phosphorus retention; migratory chains and export of detrital material. This loca-
song-birds tion is particularly vulnerable during flooding and
Isolated basin Groundwater recharge; flood control; might be unpredictable in its ultimate stability. It
waterfowl; amphibians
Coastal (fringe) Fisheries; offshore productivity;
has the advantage of potentially ‘treating’ a sig-
wetland waterfowl; storm buffer nificant portion of the water that passes that point
in the stream. A riparian wetland fed primarily by
30 W.J. Mitsch, J.G. Gosselink / Ecological Economics 35 (2000) 25–33

Table 3 steep terrain are quite susceptible to activities

Estimated unit values of ecosystems (from Costanza et al.,
upstream and are less likely to provide some
1997)
ecosystem functions. However, groundwater-fed
Ecosystem Unit value $ ha−1 yr−1 steep wetlands adjacent to hilly terrain can
provide a valuable and unique habitat in an oth-
Estuaries 22 832 erwise upland landscape.
Wetlands 14 785
Lakes/rivers 8498
Forest 969
Grasslands 232 5. The ecosystem substitution paradox

A paradox of assigning values to ecosystems is

a flooding stream allows flood events of a river to that, unless we take a landscape view, it can be
deposit sediments and chemicals on a seasonal argued that we should replace a less valuable
basis in the wetland. The wetland captures flood- system, e.g. a grassland, for another more valu-
ing water and sediments and slowly releases the able one, e.g. a wetland. Costanza et al. (1997),
water back to the river after the flood passes. when estimating the value of the world’s ecosys-
Riparian systems provide corridors for animal tem services, estimated that wetlands are 75%
movement along the river, and also a zone of more valuable than lakes and rivers, 15 times
transition between uplands and aquatic systems in more valuable than forests, and 64 times more
the transverse direction (Mitsch and Gosselink, valuable than grasslands and rangelands (Table
2000). Coastal or fringe systems, as generally 3). A straightforward economic analysis would
found along coastlines, are important to produc- thus argue for the replacement of forests and
tivity in the off-shore waters. prairies with wetlands. While this physical substi-
tution is, of course, not possible in most instances
4.2. Upstream 6ersus downstream because climatic and hydrologic variables deter-
mine what ecosystem occurs in a particular land-
The relative flood control advantages of several scape, on a micro-scale it is not only possible to
small wetlands in the upper reaches of a water- substitute wetlands for grasslands and upland
shed (but not in the streams themselves) as op- forests, but it is frequently done to meet regula-
posed to fewer larger wetlands in the lower tory requirements of wetland mitigation in the
reaches has been discussed elsewhere. Loucks USA. Many question whether the created wetland
(1989) argued that to maintain the pulse control can achieve the same functional and hence ‘eco-
function of wetlands a greater number in the nomic’ value as did the original ecosystem at that
upper reaches of a watershed is preferable to site. Some argue that these created ecosystems are
fewer larger wetlands in the lower reaches. A doomed to failure (Roberts, 1993; Malakoff,
modelling effort on flood control by Ogawa and 1998) while others are more optimistic that these
Male (1986) suggested the opposite: the usefulness systems do indeed provide real measurable value
of wetlands in decreasing flooding increases with that might even exceed what was at the site previ-
the distance the wetland is downstream. ously (Young, 1996).

4.3. Steep 6ersus flat terrain

6. How much of a watershed should be wetland?
Wetlands are a phenomenon of naturally flat
terrain. However, wetlands frequently develop in A basic question remains: how much of a given
steeper terrain as groundwater discharge points. watershed should be wetland? This is a particu-
As steep slopes are susceptible to high erosion larly good question when considering landscape
rates that pollute wetlands with suspended sedi- restoration. Table 4 gives estimates of optimal
ments and leached soil chemicals, wetlands in percent of wetland in landscapes for a few case
 W.J. Mitsch, J.G. Gosselink / Ecological Economics 35 (2000) 25–33 31

Table 4
Estimated area of wetlands required in watersheds for specific values

Value/location Watershed area, km2 Percent of watershed Reference

recommended as wetland

Small scale
General water quality improvement (IL) 378 1–5 Hey et al. (1994)
Phosphorus retention, Great Lakes basin 208 15 Wang and Mitsch
(MI) (1998)
Nitrogen control
Southeastern Sweden 882 5 Arheimer and
Wittgren (1994)
Large basins
Flood control, Upper Mississippi Basin 1.9×106 7a Hey and Philippi
(1995)
Nitrate-nitrogen retention, Mississippi 3.0×106 3.4–8.8b Mitsch et al. (1999)
River Basin

a
Includes 4% of present area as wetlands and an additional 3% restored.
b
Includes both wetland and riparian zone restoration.

studies. Using the results from the 10 ha of exper- hypoxia in the coastal waters of Scandinavia,
imental wetlands at the Des Plaines River Wet- especially the Baltic Sea; and in the Gulf of
land Demonstration Project, Hey et al. (1994) Mexico, which is fed by the 3 million km2 Missis-
suggested that 1–5% of a watershed would be sippi River basin. Arheimer and Wittgren (1994)
necessary to accomplish the water quality func- found that about 5% of a watershed in southeast-
tion on a landscape scale that these small wet- ern Sweden is needed to be converted into wet-
lands were performing. This estimate was based lands to reduce nitrogen transport by 50% to a
on the wetland’s ability to remove nutrients. In a bay on the Baltic Sea. Mitsch et al. (1999) esti-
related calculation, Hey and Philippi (1995) sug- mated that 3.4–8.8% of the Mississippi River
gested that the restoration of approximately 13 Basin would have to be converted to wetland and
million acres (5.3 million ha) in the Upper Missis- riparian forest to reduce nitrogen loads to the
sippi and Missouri Basins would provide enough
Gulf of Mexico by 20–40%. The percentages
floodwater storage ( :1 m deep) to accommodate
the excess river flow from the disastrous flood in
Table 5
Midwestern USA in 1993 (Table 5). If that 5.3 Wetland flood storage potential in Upper Mississippi and
million hectares were added to the existing 7.7 Missouri River Basins, USA (from Hey and Philippi, 1995)
million hectares in the region, an estimated 7% of
the watershed would be sufficient to deal with Water surface% area % of watershed
(million ha)
even extreme event floods on a large scale. To put
these numbers in perspective, it was estimated Original wetland area estimates
that 9–11% of the landscape was in wetland-type Hydric soils 17 8.9
environments prior to European settlement (Hey Wetland (1780) 18 9.8
and Philippi, 1995). Beaver ponds 21 11
(1600)
A number of studies have addressed the
restoration of wetlands on a large scale to mini- Wetlands
mize the impacts of agricultural and urban runoff Existing (1980) 7.7 4
Recommended 5.3 3
on coastal bodies of water. Hypoxic zones in restoration
coastal waters are prevalent around the industrial- Total 13 7
ized world. Two well-studied examples are the
32 W.J. Mitsch, J.G. Gosselink / Ecological Economics 35 (2000) 25–33

would be lower if wetlands alone were used, and Farber, S.J., Costanza, R., 1987. The economic value of
higher if riparian forests alone were used, as wetland systems. J. Environ. Manag. 25, 149 – 156.
Gorham, E., 1991. Northern peatlands: role in the carbon
wetlands are generally more efficient per unit area cycle and probable responses to climatic warming. Ecol.
in nutrient removal. Appl. 1, 182 – 195.
Gosselink, J.G., Odum, E.P., Pope, R.M., 1974. The Value of
the Tidal Marsh.’ Center for Wetland Resources Publ.
7. Conclusions LSU-SG-74-03, Louisiana State University, Baton Rouge,
30 pp.
Several landscape-scale considerations should Gren, I.-M., 1994. Costs and benefits of restoring wetlands:
be taken into account when ascribing value to two Swedish case studies. Ecol. Eng. 4, 153 – 162.
Gren, I.-M., Folke, C., Turner, K., Batemen, I., 1994. Primary
wetlands. All things being equal, a wetland in a
and secondary values of wetland ecosystems. Environ. Res.
region with moderate but not excessive urban Econ. 4, 55 – 74.
development will have the greatest value because Hey, D.L., Barrett, K.R., Biegen, C., 1994. The hydrology of
an adequate human population is present to four experimental constructed marshes. Ecol. Eng. 3, 319 –
benefit from those values, but the population is 343.
not so large as to overwhelm the wetland func- Hey, D.L., Philippi, N.S., 1995. Flood reduction through
wetland restoration: the Upper Mississippi River Basin as
tions. In a hydrogeomorphic setting, flow-through
a case history. Rest. Ecol. 3, 4 – 17.
and riparian wetlands have a better chance of King, D.M., 1997. The fungibility of wetlands. Nat. Wetland
having high values than isolated basin wetlands, Newslett. 19 (5), 10 – 13.
although caution has to be taken with this gener- Loucks, O.L., 1989. Restoration of the pulse control function
alization. Finally, in answer to the question of of wetlands and its relationship to water quality objectives.
how much of a watershed should be wetland, In: Kusler, J.A., Kentula, M.E. (Eds.), Wetland Creation
and Restoration: The Status of the Science. USEPA, Cor-
several examples in Midwestern USA and Scandi-
vallis, OR, pp. 55 – 74.
navia suggest that an optimum amount of wet- Malakoff, D., 1998. Restored wetlands flunk real-world test.
lands in a landscape might be around 3 – 7% Science 280, 371 – 372.
(average  5%) in temperate-zone watersheds to Mitsch, W.J., 1977. Energy conservation through interface
optimize the landscape for their ecosystem values, ecosystems. In: Rocco, A., Fazzolare, R.A., Smith, C.B.
e.g. flood control and water quality enhancement. (Eds.), Energy Use Management: Proceedings of the Inter-
national Conference. Pergamon Press, Oxford, pp. 875 –
In all cases, one must consider whether the values
881.
of wetlands are based on biological populations Mitsch, W.J., Gosselink, J.G., 1993. Wetlands, 2nd ed. John
living next to and within the wetlands, the wet- Wiley, New York.
land ecosystem itself, or the entire biosphere of Mitsch, W.J., Wu, X., 1995. Wetlands and global change. In:
which wetlands are a small part. Lal, R., Kimble, J., Levine, E., Stewart, B.A. (Eds.),
Advances in Soil Science, Soil Management and Green-
house Effect. CRC Lewis Publishers, Boca Raton, FL, pp.
References 205 – 230.
Mitsch, W.J., Day, J.W. Jr, Gilliam, J.W., Groffman, P.M.,
Arheimer, B., Wittgren, H.B., 1994. Modelling the effects of Hey, D.L., Randall, G.W., and Wang, N., 1999. Reducing
wetlands on regional nitrogen transport. Ambio 23, 378– nutrient loads, especially nitrate-nitrogen, to surface water,
386. groundwater and the Gulf of Mexico. Decision analysis
Barbier, E.B., 1994. Valuing environmental functions: tropical series No. 19, Coastal Oceans Program, Silver Spring, MD,
wetlands. Land Econ. 70, 155–173. 111 pp.
Bell, F.W., 1997. The economic value of saltwater marsh Mitsch, W.J., Gosselink, J.G., 2000. Wetlands, 3rd ed. John
supporting marine recreational fishing in the southeastern Wiley, New York.
United States. Ecol. Econ. 21, 243–254. Ogawa, H., Male, J.W., 1986. Simulating the flood mitigation
Costanza, R., Farber, S.C., Maxwell, J., 1989. Valuation and role of wetlands. ASCE J. Water Resour. Plan. Manag.
management of wetland ecosystems. Ecol. Econ. 1, 335– 112, 114 – 127.
361. Roberts, L., 1993. Wetlands trading is a loser’s game say
Costanza, R., d’Arge, R., deGroot, R., et al., 1997. The value ecologists. Science 260, 1890 – 1892.
of the world’s ecosystem services and natural capital. Na- Turner, K., 1991. Economics and wetland management. Am-
ture 387, 253 – 260. bio 20, 59 – 63.
 W.J. Mitsch, J.G. Gosselink / Ecological Economics 35 (2000) 25–33 33

Wang, N., Mitsch, W.J., 1998. Estimating phosphorus reten- Wharton, C.H., 1970. The Southern River Swamp, A Multi-
tion of existing and restored wetlands in a tributary water- ple-Use Environment. Bureau of Business and Economic
shed of the Laurentian Great Lakes in Michigan, USA. Research, Georgia State University, Atlanta, GA.
Wetlands Ecol. Manag. 6, 69–82. Young, P., 1996. The ‘new science’ of wetland restoration.
Environ. Sci. Technol. 30, 292 – 296.