Landscape Change and the Ecological Effects of Mitigation Banks
                             Curtis C. Bohlen
  To appear in: Deborah Hayes, ed. Wetland Mitigation Banking Workshop
                Proceedings.  Argonne National Laboratory.    .
University of Maryland, Center for Environmental And Estuarine Research,
    Chesapeake Biological Laboratory, P.O. Box 38, Solomons, MD 20688.

    This work  'was supported-under_^ Cooperative Agreement CR-818227-CI from the U.S..
Environmental Protection Agency, with the University of.Maryland Center for Environmental
                            and Estuarine Studies.   .   •

                          EPA 230-R-96-008


     Wetland mitigation and mitigation banking markets are complex legal,
economic and political phenomena played out against a background of
dynamic ecological systems.  The ultimate environmental effects of increased
reliance on mitigation banking therefore will be co-determined by economic,
political, and ecological processes.  In this study, I focus on landscape-level
ecological consequences of what might be called a successful mitigation
banking program, one in which mitigation sites are "successful", in that they
support ecological processes typical of wetland systems.

     The ecological processes that occur in wetlands are determined as much.
by the landscape context of the wetland as by characteristics of the wetland
itself. Moreover, the social and economic values derived from those
ecological processes also vary from site to site. Even a "successful" mitigation
banking  program engenders a gradual reconfiguration of the wetland
inventory of a region including (1) geographic changes in the distribution of
wetlands, (2) alterations in the size distribution of wetlands, and (3) alteration
of'the hydrogeomorphic context of wetlands. Such the
characteristics of the wetland inventory may have profound ecological and
economic consequences.
     A framework for analyzing landscape-level consequences of mitigation
banking is introduced, and ways in which negative landscape-level effects
may be reduced and positive effects enhanced are discussed.


     It is difficult, as an ecologist, to identify the ecological consequences of
 expanded reliance on wetland mitigation banks.  The long term
 environmental consequences of mitigation banks will be determined at least '•
 as strongly by legal, institutional, and political forces as by the ecology of
 wetlands. The concept of mitigation banks is sufficiently broad that it
 comfortably encompasses programs that would be widely regarded as an
 environmental improvement' over current practice, and also programs likely
 to accelerate deterioration of wetlands and other aquatic resources. In that
 respect, mitigation banking has a great deal in common with all
 compensatory mitigation, the consequences of which depend the details of   •
 how it Is implemented.         •

     A fragile (and  brief) consensus on an appropriate role for, mitigation was
 hammered out  in 1988, when the National Wetland Policy Forum developed
 the national goal of no net loss and an eventual net gain of wetland resources
 by acreage and function (Conservation Foundation 1988). This goal was
 adopted by the Chesapeake Bay Program, included as an explicit goal of    •
 Maryland's Nontidal Wetlands Protection Act, referenced in Maryland's state
 mitigation banking  legislation, and adopted by federal regulators  (Chesapeake
 Bay Program 1988, Maryland Code 1989, EPA and DOA 1990). With a clear
 vision of the role of wetland mitigation in the larger context of wetland
 policy, regulators have developed a variety of rules and policies that have
 improved compensatory mitigation practice.

     Mitigation at the programmatic level, however, has of ten-not performed
 as well as anticipated.  Site-by-site reviews of the  environmental  performance
 of mitigation programs  have often found poor regulatory compliance, Iqw
 frequencies of environmentally successful mitigation, and widespread .
 replacement of wetlands by structurally and biologically simpler wetland
 types (e.g. forested wetlands replaced by emergent wetlands, or salt marshes
 replaced by mud flat;  Race 1985, Harvey and Josselyn 1986, Bernstein and
 Zepp 1990, Crewz and Lewis 1991> Erwin 1991, Florida Department of
 Environmental Regulation 1991,. Race and Fonseca, In Press)/ Forty six   • •
percent of wetland consultants recently surveyed by the Chesapeake Bay
Foundation thought that "existing regulatory requirements lead to
compensatory mitigation that has  little environmental benefit" either
frequently or always (Bohlen et al 1994). And frustration with the regulatory
complexity engendered by mitigation requirements has  also figured
prominently in development-oriented criticisms of wetland policy (Albrecht
and Goode 1994).

Landscapes and Mitigation Banks       .-:...

                                              i              •            , '
Mitigation  Banks

     Recently, it has been suggested that the poor performance of mitigation
programs is due, at least in part, to the perverse incentives faced by purchasers
of mitigation services, who have  no reason to be quality conscious, and every
reason to be price-conscious as they select ways to satisfy their mitigation
obligations (King and Bohlen 1994, Shabman et al. 1994). With few
exceptions, the people and institutions paying for mitigation are not
interested in producing  functional wetlands, but in negotiating regulatory
hurdles required to gain approval for other projects.  Suppliers of low cost—
and generally low quality—wetland creation and restoration services
therefore dominate the market, and drive suppliers of higher quality—and
higher cost—services to take refuge in niche markets providing wetland
creation and restoration for customers (environmental groups, some
government agencies, a few private landowners) that are more quality
conscious. Mitigation banks may provide an opportunity to change the
incentives and disincentives perceived either by purchasers of mitigation
services, or by suppliers of those services, and thus improve the performance
of mitigation programs (Shabman et al. 1994).

     More generally, mitigation failure has been blamed on everything  from
poor enforcement, to regulatory  preference for on-site mitigation, from  the
ineptitude of regulatory personnel to inadequate identification of the goals of
wetland protection.  Mitigation banking has been promoted as a solution to
many of the problems plaguing compensatory mitigation.  It has been argued
that the availability of banks will (1) increase the chance that mitigation
obligations will be met, (2) allow agencies to deploy their limited enforcement
and oversight resources to greater effect, (3) reduce reliance on poorly sited
mitigation, (4) improve average  performance of mitigation, and (5)  enable
mitigation of previously unmitigated small wetland impacts  (e.g. Shabman et
al. 1994).

     While these benefits  may follow from establishment of  mitigation
•banks, such a beneficent outcome is far from assured. The rules
underpinning mitigation banks can be seen as reflecting a three-way tug of
war between project sponsors, mitigation suppliers, and the general public (as
represented by the regulatory agencies) over who will bear the risks and
uncertainties of mitigation1.  The more the risks of mitigation can be  shifted
to the general  public, the more profitable  the venture will be, either for the
project sponsor, or for the- mitigation supplier.  Simultaneously, however,
 1Mitigation risk includes both risk of project failure, and also the risk of a "successful" project
 that provides lower than anticipated environmental benefits, or benefits different from those
 anticipated. Uncertainty is more inclusive, and reflects the limits of our current understanding
 of wetlands and aquatic ecosystems.

 Landscapes'and Mitigation Banks

 such a shift increases the risk that effective mitigation for impacts to wetlands
 will not occur, thus denying the general public wetland benefits that they had
 enjoyed prior to the wetland impacts that engendered a mitigation obligation.

      The devil, as so often the case, is in the details, the trading rules under
 which a mitigation banking system is established, (compensation ratios,
.financial and legal assurances, systems of evaluating credits, timing of credit
 withdrawal, frequency of impacts deemed mitigatable through banks and so
 on) will largely determine whether banks will improve upon existing
 mitigation programs/or just magnify their shortcomings.  Banks, per se,  are a
 dangerous opportunity:  neither better nor worse than existing practice.  As
 we shape the rules under which mitigation banks develop, we
 simultaneously determine whether or not experiments with banks will be
 seen as the start of a better era in managing aquatic resources.

      Of almost equal significance as the initial rules will be the stability of
 rules in the face of political and legal challenges.  Agencies responsible for
 wetland regulation will  before long find themselves in the uncomfortable
 position of adjudicating  a market in which people will be playing for big
 stakes., and occasionally  loosing2. Undoubtedly, regulatory intransigence will
 be blamed for financial difficulties,  real or imagined;  among mitigation banks.
 How the agencies respond to the inevitable legal challenges and inquiries
 from congressional offices may well determine'the long term environmental
 effectiveness of mitigation banking programs.

     At this point, therefore, the long term environmental effects of more
 widespread reliance on mitigation banks are difficult tef predict.  The rules
 that will govern mitigation banks are still in flux, and our experience with
 existing banks is sufficiently small that the indirect economic, legal, and  -
 political consequences, can not be predicted. We as yet have too little
 experience with commercial mitigation banks to know how well the
 environmental purposes of mitigation will be integrated with the financial
 demands for commercial success. And a variety of subtle consequences may
 yet crop up. .Widespread banking may change the  incentives and
 disincentives faced by individuals, deciding whether or not to disturb existing
 wetlands (Bohlen et al. 1994). Increased reliance on and publicity.about banks
 2The market for wetland mitigation credits is an artificial market, in which demand for
 credits, and the supply of credits are both determined by agency decisions.' Demand is shaped
 by such decisions as whether to require mitigation for small-scale wetland impacts, whether a
 bias for on site or in kind mitigation exists, and whether credits can be applied to distant
 wetland impacts. The supply of credits will also largely be determined by the conditions the
 agency places on mitigation projects and the number of credits assigned to projects. Thus agency
 decisions largely determine whether it is possible to make a profit as a mitigation supplier,
 and even whether an individual supplier will be  profitable. The likelihood of political and
 legal challenge when companies go belly up should be obvious.

Landscapes and Mitigation Banks
                                              i .                         '
for mitigation may affect people's willingness to take part in cooperative
wetland restoration efforts3 as they realize that similar activities with other
sponsors could generate income.  Whether such changes in incentives will be
of sufficient magnitude to alter the frequency or characteristics of wetland
impacts remains unclear.           '                                •    '

A Landscape Analysis of  Mitigation Banking

     While it is probably impossible to be certain what the overall
environmental impact of reliance on mitigation banks will be, one can begin'
to understand the kinds of ecological changes banking implies.  In order to do
that, I will compare the ecological, implications of what might be called a
"successful" mitigation banking scheme—one in which individual
mitigation banks provide the environmental-services for which they were
designed—with a "successful" project-by-projed: mitigation system..  Both
these mitigation systems, it should be pointed out, are mythical in that they
do not suffer from ariy of the vices we have  come to expect from mitigation,
But by examining the ecological consequences of these mythological beasts, I
hope that to be able to isolate the ecological issues from the other
uncertainties (and the politics) that would otherwise arise.
                                       '       I           '  •
Values  and  Functions
     The physical and biological phenomena that occur in wetlands are
conventionally termed "functions"; the subjective experience of those
functions are called "values" (Richardson 1994).  Functions are a property of a
wetland and its geographical context.  Values are a property of the interaction
of human social, political, economic and ethical systems with  the wetland,
and with other ecological systems to which it is functionally related.

      Values, as the term is used here, include not only market values (e.g.
timber production, hunting rights), but also  unrnarketted direct (flood
control) and indirect (support of anadromous fishes) values, including often
obscure values mediated through ecosystem services. These values are
emergent properties that arise neither from the wetland itself, nor from
human society, but from the interaction of society with the wetland within a
specific geographic (and therefore ecological, economic, and social) context.
3 A variety of cooperative programs have been designed to encourage wetland restoration,
especially on agricultural lands. The most important of these are the the U.S. Fish and
Wildlife Service's Private Lands Program and, U.S. FWS's Wildlife Extension Agreements,
the U.S. Department of Agriculture's Wetland Reserve Program, and the USD A Water Bank
Program, and the Department of the Interior's Small Wetlands Acquisition Program.

 Landscapes and Mitigation Banks         ' -•        '•.,"'    >

 Values emerge out of , but are not the same as, the underlying functions that
, occur in the wetland.

      Functions and values are linked both by the ecological, chemical, and
 physical processes occurring in the wetland and by the social and economic
 phenomena that generate human perceptions of value.  Tracing a particular
 function (e.g. sedimentation rate) to its many derivative values may be
 exceptionally difficult (Bohlen and King 1995).  However, categories of values
 can be linked on a qualitative basis' with broad ecosystem functions
 (Richardson 1994), to provide an approximate basis for comparative analysis
 of wetland values.  '                               ,

 The Wetland  Resource Portfolio

      Any form of wetland compensation represents a decision by society to
 invest in wetlands and other aquatic resources.  It seams natural to speak (in
 analogy with other types of investments)  of the wetland resource portfolio,
 A region's  wetland portfolio represents the sum total of wetland resources in
 the region. The wetland resource portfolio is a component of the larger set of
 environmental resources in the  region which together represent: the region's
 natural capital4 (Jansson et al. 1994, King 1995). The  value of any'
 particular wetland is determined jointly by its characteristics (hydrology, '
 vegetation, biota, etc.), location  with respect to other .environmental
 resources, proximity to  human activities,  and so on.

      Wetland regulatory programs are among the most important tools we
 have for managing the wetland  portfolio.  We use them primarily to
 minimize conversion of wetlands to environmentally less beneficial • uses, but
. regulatory programs also steer development more strongly away from certain.
 wetlands than from others.  Compensatory mitigation represents an effort to
 reduce the long term negative consequences of impacts to  wetlands by
 reinvesting in wetlands following unavoidable impacts.  It represents an
 investment in natural capital to  make up for a loss of natural capital
 elsewhere.  Like more familiar  (financial) investments, any investment in
 natural capital can be analyzed ,in terms of its risks and expected return, with
 risks and returns appropriately  measured in terms  of environmental benefits
 (King et al. 1994). A strategy for carrying out such investments (e.g.
 mitigation banking) can be analyzed in terms of (1) the kinds of investments
 it is likely to generate, (2) the changes it will produce in the resource portfolio.
 Wetland compensation  through mitigation banks and compensation- through
  A. variety of natural systems including wetlands provide human society with environmental
 goods and services that improve the quality of human life. Natural capital, just like built
 capital, is valuable because of its ability to produce goods and services now and in the future.

Landscapes and Mitigation Banks

project-by-project mitigation are contrasting strategies for reinvestments in
natural capital following wetland impacts.

     If wetland impacts are rare, or if impacts are seldom mitigated, then the
overall effects of mitigation banking versus project-by-project mitigation on
the wetland resource portfolio will be small.  With little impact to, and hence
reinvestment in wetlands, the particular reinvestment strategy we follow
would change the wetland portfolio only slowly, We would have time .to
recognize nascent problems, and potentially respond to them in an adaptive
manner (Walters 1986). If recent history is any indication, however, we can
not assume that future wetland impacts will be insignificant. Between 1982
and 1989, there was approximately a 2.5% net loss of wetland area in the
Chesapeake Bay Watershed, totaling about 37,000 acres (Tiner et al. 1994).
Furthermore, those losses were unevenly distributed, with substantially
higher losses in Virginia, especially-in the  Norfolk—Virginia Beach area.
                                             i   .  .       •          •     •
     The extent to which wetland losses will be mitigated remains to be seen.
Currently, mitigation is required by federal regulators for only a small
proportion of wetland losses (many wetland impacts are  partially or entirely
exempt from federal regulations, and most wetliand impacts are routinely
approved through general permits that require no mitigation).  State agencies
differ widely in the degree to which they regulate small wetland impacts, and
the extent to which they require mitigation for those impacts (Bohlen et al
1994).  Few regulatory programs are currently re-quiring mitigation for a
sufficient proportion of wetland impacts to approach the r>o-net-loss goal,
even on paper.

     One  of the benefits that has been suggested for mitigation banks is that
many of currently unmitigated wetland impacts can be mitigated more
practically and less expensively 'through a mitigation bank than through
individual projects.  Whether regulatory programs will acquire the authority
to  require mitigation more widely (either with banks or on a project-by-
project basis) remains  to be seen.  Given the current anti-regulatory
enthusiasms of lawmakers, especially at the federal level, the effective
frequency of mitigation for wetland conversions appears more likely to
decline than to rise, regardless of the development of mitigation banking

Landscape Properties of Mitigation Banks

     An increased reliance on mitigation banks as a way  to accomplish
mitigation is likely to affect the wetland  resource portfolio  in different ways
than would continued reliance on project-by-project mitigation.  Some of
these differences are inherent in the concept of mitigation banks.  Others
reflect the economics of wetland creation and restoration, and could,
presumably be altered or managed with carefully constructed regulatory
incentives and disincentives. Costs of wetland creation and restoration are

 Landscapes .and Mitigation Banks

 strongly dependent on site-specific and design-specific factors, from local
 hydrologic conditions, to the cost of land, to the amount of earth that must be
 moved (King and Bohlen 1994). Mitigation Banks are intended as general, not
 specific mitigation, and thus will often have fewer regulatory constraints on
. location and design than will project-by-project mitigation.

      The consequences of mitigation banks  on the wetland resource portfolio
 can be divided into three broad areas, none of which are 'new or surprising:

      (1)  Mitigation banking systems typically are based on relatively few,
          large creation or restoration projects.

      (2)  Mitigation banking systems are likely to rely on construction or
          restoration of wetland types that are least expensive 'to produce.  -

      (3)  Mitigation banks involve expanded use of off-site mitigation.

      These effects may also occur to some extent with project-by-project
 mitigation; but the trends will be more pronounced as mitigation decisions
 are increasingly divorced from characteristics of the sites at which wetland
 impacts occur, and are driven instead by the costs of providing mitigation.

    Wetland Size

      Large wetlands, it is often said, are better than small ones.

      From a purely ecological point of view, this statement is untenable.
 Naturally occurring  large and small wetlands are distributed non-randomly
 across the landscape, generally differ in hydrologic conditions, and  interact
 with the surrounding- landscape in contrasting ways. To generalize  by saying
 that one is "better" than  the other is to ignore the importance of the
 interrelationship among  ecosystems in the landscape, and thus do  violence to
 principles of landscape ecology.  Large wetlands and small wetlands are
 different, and play distinctly different roles in .the landscape. Which one is
 "better" will depend both on local landscape condition, and on local and
 regional environmental  goals.                   ,-

      mdeed, it is not especially difficult to think of specific wetland values or
 functions for which small wetlands are likely to be better than large ones.
 Many amphibians, for example, breed preferentially in small pools  that lack
 fish, and are located in or near the upland habitats in which the adults are
 found.  For such amphibians, small wetlands are an important reproductive
 resource precisely because they are small, isolated from other surface waters,
 and closely associated with uplands. For a wide variety of birds and
 mammals, wetlands  are not primary habitat, but provide valuable food, cover
 or other resources under certain conditions or at certain times of year.  For
 such species, the role of a wetland as a ecotone—a transition between upland

Landscapes and Mitigation Banks

and aquatic systems—is of central importance, and the area of wetland may be
far less important than the length of the wetland's perimeter.  For a given
area of wetland, small wetlands will generally have a longer wetland-upland
transition zone than would a single, large wetland.

     Even within a superficially similar class of wetlands, small and large
wetlands need not supply identical environmental.benefits. For example, a
majority of the water in many stream systems enters the surface water system
by traveling through the narrow (often wetland)  riparian areas along the
smallest streams in the watershed.  The larger riparian wetlands further
downstream are less abundant (there are many small streams for every  large
one), and a relatively  small amount of water enters the stream by traversing
them. They thus may be less important than' the -upstream riparian wetlands
for regulating water quality.  The larger downstream wetlands, however may
play an important role in river and floodplain ecology as components of the
regularly flooded "aquatic-terrestrial transition zone" (Bayley 1995, Junk et al.
1989). Which ecological role (water quality or flood ecology) is more
important depends on local and regional environmental goals, the presence
or absence of other ecosystem stressors, and the relative importance of clean
water and annual floods to the ecology of downstream ecosystems and the
economy of downstream communities.,

     I am therefore skeptical that large wetlands always represent the best
investment in wetland resources.  Whether they are, however, is not solely
determined by the ecological services large and small wetlands provide, but
also by the relative cost of constructing large and. small wetlands.  Large
wetlands are typically less expensive than small ones to construct, with  a
doubling in project size typically resulting in about a 20% decline in per acre
costs (King and Bohlen 1994). Some of this difference in cost is attributable to
a tendency for more intensive methods of restoration to be used  on  smaller
projects, but costs associated with designing projects, transaction costs
incurred while gaining regulatory approval and  costs of monitoring  are also
important.                         .

     Unless strong regulatory incentives and disincentives are established
otherwise, these economies of scale will lead mitigation suppliers primarily
to produce large wetlands for mitigation banking purposes. In some  regions,
such large banks will represent the best possible investment in wetland
resources for achieving water quality and environmental goals. In other
regions, investment in several smaller wetlands may be a better approach to,
take. Where  that is true, regulatory agencies may want to encourage
mitigation banks that are assembled from several separate mitigation projects,
Such an approach is likely to be intermediate in cost between project-by-
project mitigation and conventional mitigation banks, But can allow a  more
careful tailoring of mitigation characteristics to local and regional
environmental goals.

 Landscapes and Mitigation Banks               ,              .    •

    Changes in Wetland Type

      Certain types of wetlands are more difficult to create or restore than
 others.  All mitigation, therefore, whether via banks or via project-by-project
 mitigation, results in a shift in the wetland resource portfolio.  Certain types
 .of wetlands, especially those with'peat substrates or those! reliant on ground
 water or direct rainfall as a source of water, are difficult to create or restore.
 Bogs, nutrient-poor fens, and certain sedge meadows, if 'impacted, are thus
 likely to be  replaced inefficiently, if at all by any form of mitigation. Even in
 wetlands for which restoration technologies are well advanced, a variety of
 ecological properties, from plant species diversity, to soil characteristics, to
 canopy height and abundance of snags may take decades to return to levels
 typical of undisturbed Wetlands. On the whole/mitigation leads to a.shift in
 the wetland portfolio toward structurally simple, early successional wetland
 systems dominated by wetland plants with wide ecological  amplitude like
 Typha and Peltandra  (e.g., Benstein and Zepp 1990).

      Switching toward the increased use of.mitigation banks may exacerbate
 the problems this can cause.  Banks are intended as general mitigation. The
 ecological characteristics of banks, and thus the environmental services
 provided by them are not as closely linked to specific impacts as are
 characteristics of project-by-project mitigation.  While banks may be
 constrained in a variety of ways to ensure compensation of'wetland losses
 "in-kind", that term, is notoriously imprecise.  The term is frequently used to
 refer only to the relatively coarse characteristics of vegetation described by the
 Cowardin et al. (1978)  class designations (e.g., forested, scrub-shrub, emergent
 and so on for palustrine wetlands).  Jn-kind constraints on mitigation>anks
 are less often aimed at soil or hydrologic conditions that are important in
 determining biodiversity at both the ecosystem and species  levels.

     Project-by-project mitigation retains a closer link between the
 characteristics of the wetland lost and the characteristics of the compensatory
 wetland than do mitigation banks.  While  that constraint may  sometimes be a
 source of inefficiency and higher mitigation costs, it also increases the
 likelihood that successful  mitigation efforts will provide a close functional
• replacement for the original wetland.  Project-by-project mitigation may also
 have another, more subtle consequence.  By requiring that the functions
 provided by compensatory mitigation be closely matched to the losses
 incurred, project-by-project mitigation may, in effect, discourage impacts to
 those wetlands for which  compensation  is most difficult.

     In a highly developed or disturbed landscape, close functional
 replacements for existing wetlands may be undesirable; specific wetland
 functions may be considered to be more beneficial than were the functions
 that were lost.  In less severely modified landscapes, however, such a
 judgment is  difficult to make. In less modified landscapes/ long term

Landscapes and Mitigation Banks

evolutionary and ecological processes have linked the characteristics of
wetland systems with those.of adjacent terrestrial and aquatic systems as part
of a larger landscape mosaic.  Existing landscape linkages have been little
modified, and landscape-level ecological processes are more or less intact.
Any changes humans make to the local wetland portfolio therefore will
disrupt to a greater or lesser degree ecological processes that support the
physical, chemical, and biological integrity of surface waters. We would be
displaying  considerable arrogance to imagine that we can design or manage
landscapes to support that integrity as successfully as millions of years of
evolution and thousands of years of ecological change (Ehrenfeld 1978, Karr
1990, Angemeier and Karr 1994).

     The changes in the spectrum of wetland characteristics that are likely to
accompany expanded mitigation banking will be most acceptable within
moderately to highly modified landscapes.  In such landscapes, the
information embodied the characteristics of existing wetlands may be of
reduced value to present-day landscape managers.  Even in altered
landscapes, however, considerable information from watershed planning or
other in-depth analysis is needed in order to make informed judgments about
whether out of kind mitigation is, in fact, appropriate.

     Rules for mitigation banks should be constructed so as to ensure that any
changes the banks induce in the spectrum of wetland types in the wetland
portfolio will be beneficial. Even in the context of mitigation banking, out-of-
kind mitigation should be avoided to the extent possible unless an explicit
rationale for replacing one  wetland type with another (preferably based on
detailed analysis). The bias toward in-kind replacement should be strongest
for banks that are intended to ameliorate the negative consequences of
wetland impacts in largely intact landscapes. Legislation, regulations, and
agreements establishing a bank or banking program should explicitly define
the relationship between banking and impacts to difficult-to-restore wetlands,
high value wetlands, and other wetlands of special  concern.

   Locations of Wetlands

      The decision to allow mitigation via a bank represents a choice to invest
in wetlands at the site of the bank in favor of investing in wetlands at the site
of wetland impacts.  Expanded reliance on banks thus will lead  to an altered
spatial distribution of wetlands across the landscape. Location is central to
both the functions of individual wetlands, and also to the values derived
from those functions.  Wetlands, like all ecosystems, are not isolated from
their surroundings, but experience flows of organisms, energy, nutrients,
water, and pollutants across their boundaries. Thus the ecology of a wetland
reflects the nature of its surroundings. A wetland in an urban area will be
qualitatively different from a similar wetland in a rural landscape; the two
wetlands will differ in terms of the  functions occurring within  them.

 Landscapes and Mitigation Banks

      It is perhaps less obvious that the value of specific wetland functions will
 also differ with geographical context. The .value of nutrient trapping and
 transformation functions depends  critically on what is upstream and
 downstream of a wetland.  If no source of pollutants is found upstream of the
 wetland, than the wetland, while capable of providing water quality benefits,
 Will not have the opportunity. If downstream water bodies are unlikely to be
 harmed by additional nutrient influxes (perhaps because they have
 substantial assimilative capacity, or are so badly polluted that a little more
 nutrient will make little difference to ecosystem condition) then the ability of
 the wetland to remove nutrients will have little value.  Location-dependence
 of the value derived from wetland functions may also arise because-of
 geographic variation in the existing  availability of environmental services.
. Wetland functions may be in short supply in some landscapes (from an
 ecological or societal perspective),  and abundant elsewhere (Bohlen and King
 1995).                     -               ..                  •

      On-site biases embedded in current project-by-project mitigation systems
 are often reviled as irrational and unjust.  While in practice, these'biases have
 sometimes resulted in poor or inefficient mitigation  decisions, they arise
 naturally from the acknowledgment  of the context-specific nature of wetland
 functions'and  values. In the absence of adequate information about the
 values derived from a specific wetland, and knowledge of the need for specific
 environmental services in a region, an on-site bias (but not an absolute
 requirement) makes eminent sense.

      In the context of mitigation banking on-site biases will be weakened or ,.-
 eliminated all  together.  Thus regulatory agencies must better understand the
 consequences of moving wetlands about on the landscape, lest they risk
 creating successful mitigation, yet jeopardizing water quality and
 environmental goals that are the real purposes behind wetlands protection.
 The relative importance of different wetland functions can not be accurately
 determined outside of an ecosystem context. Watershed and ecosystem"  •. . -
 planning and analyses thus would  be valuable tools for regulatory agencies
 assessing either the credits to be assigned to  a mitigation bank or mitigation
 ratios to be required for mitigation of specific impacts.  Watershed planning
 efforts, however, are expensive, and in the current budget climate with
 respect to regulatory programs, are unlikely to be widely developed. Rules of
 thumb, and understanding of basic principles must often substitute for
 detailed analysis. Many wetland mitigation transactions  in mitigation banks
 will he carried out with incomplete scientific understanding of the
 environmental trade-offs involved. Even in the absence of watershed  or
 ecosystem-based planning, decisions  must be made.  The  following discussion
 may help regulators  better understand the consequences of moving wetlands
 about on the landscape (these ideas are developed somewhat more fully in
 Bohlen and King 1995),

Landscapes and Mitigation Banks                       :

   Wetland Values and Associated Spatial Scales

     Considerable insight into wetland ecology, management, and politics can
be gained by consideration of the scales over which wetland values and
functions occur and are managed. Understanding the scales important for
generating and managing wetland functions will be especially important in
the context of mitigation banking. Only by understanding which functions
develop over large scales, and which over small can regulatory agencies
understand which wetland functions can be protected by off-site mitigation,
and which can not.

     Figure 1 presents a framework for understanding wetland policy and
management in terms of two scales. The. first is the largest scale over which
management efforts are likely to be necessary and effective. The second scale
reflects the distances over which the values derived from wetland functions
are perceived.   These two scales need not be, and in general are not, the same.
Mismatches between the two scales lead to practical and political difficulties
managing wetland resources (Bohlen 1993, Bohlen and Friday 1995)

     Wetland functions depend on the  physical and biological environment
of the wetland. However, the scale over which that dependence is harnessed
for management purposes varies. For any wetland function, one can identify
a range of spatial scales over which actions may be necessary for effective
management.  For most wetland functions, the smallest scale of management
is on the order of a few meters or tens of meters, but the largest scale that
needs to be considered varies.  Production of timber in forested wetlands
depends only loosely on conditions outside of the wetland. Thus timber
production can be managed with little attention to conditions outside  of the
wetland itself.  Anadromous fish populations, on the other hand, can seldom
be effectively managed on a site-by-site basis because of their migratory habits.

     Values derived from  wetland functions sometimes accrue locally, but at
other times may accrue to individuals some distance away., The value of a
tidal wetland for producing salt marsh hay, for example, accrues directly to
the landowner or farmer.  The benefits  of waterfowl hunting accrue primarily
to those with physical access to the wetland, or to nearby wetlands where
resident birds  may also forage. Benefits of flood attenuation by wetlands
generally accrue to those living inland from coastal wetlands, or downstream
from riparian wetlands. Water quality benefits of wetlands may be realized
primarily in lakes, impoundments and  estuaries hundreds of miles away
from the wetland itself. And the values derived from some wetland
functions (e.g., carbon sequestration) accrue over global scales.

Landscapes and Mitigation Banks
                        Scales At Which Wetland Values
                           Are Received and Managed
          1000 Km
     *j (5  1 00 Km

           1 0 Km
            1  Km
     x fc   100m  ..
            10m •
                       Habitat for Forest
                        Interior Birds
Wild Trout
               Timber      '
              Production /
                        ;._: Production of
                       Anadromous Fishes
               Nutrient Removal
              and Transformation
                                                         Global ,
                   10m    100m  . ,.1 Km  '  10 Km  100 Km 1000 Km  10,000 Km

                             Max.  Distance   From  Wetland
                           At  Which  Benefits  Are  Received
Figure 1:  Scales and wetland values. Management to affect the degree to
which a wetland provides certain values is only possible at certain scales.
Those scales do not necessarily match the scales at which the benefits of those
wetland values are distributed.                     '

     For purposes of developing wetland mitigation banking policy, one can
identify three simplified cases. If the distance between the site of wetland
impact and the site of mitigation is:

     (1)   Larger than both characteristic scales associated with a particular
          wetland value then mitigation redistributes wetland benefits.

     (2)   Similar to the characteristic scales of a particular wetland value,
          then mitigation may or may not replace wetland benefits.

     (3)   Smaller than both characteristic scales then mitigation replaces
          wetland benefits.

Landscapes and Mitigation Banks

     For wetland benefits with characteristic scales smaller than the distance
between the site of impact and the site of compensation (case 1), mitigation
represents a redistribution of environmental benefits from one location (and
thus often group of people) to another. A successful mitigation site may
provide opportunities for education of local school children similar to the
opportunities that would have been available at the site of wetland loss.  If
the two sites are more than a few miles apart, however, chances are good that
a different group of school children will enjoy thos;e benefits. Similar
arguments can be offered for most local wetland benefits, from timber
production to hunting rights. Thus the analysis of the effects of mitigation
banks and off-site mitigation on functions with small characteristic scales
should focus on the degree to which a change in the location and primary
beneficiaries of functions is desirable (Bohlen and King 1995). Local policy
makers may haye strong opinions about the wisdom of such intra-regional
transfers of environmental benefits.             ;

     For wetland benefits with characteristic scales similar to the distance  .
between the two sites (case 2), the effects of mitigation depend on the details
of the ecology of the region, and on characteristics of the impact site and
mitigation site. Since benefits accrue or are managed on a scale similar to the
distance between the two sites, the benefits at one site can not be treated as
independent of benefits at the other. Benefits may be dynamically linked. For
example,  the net effect on anadromous fish populations in the Chesapeake
Bay of off-site mitigation for impacts to a freshwater wetland will depend on a
number of features of the regional landscape: Are water quality in, both areas
adequate to support reproduction of anadromous fish? Are waterways below
the sites of impact and compensation both free of blockages  to fish
migration?. Similarly, the effects of off-site wetland compensation on water
quality or biological integrity below the confluence of the sub-watersheds
containing the site of wetland impact and a mitiga.tion bank will be
influenced by a variety of ecological details. Typically, it will require careful
biological analysis, and sometimes economic and social analyses to determine
whether case 2 wetland benefits will be effectively maintained.

     For wetland benefits with characteristic scales substantially larger than
the distance between the site of wetland impact and the site of mitigation
(case 3), a successful mitigation project that provides appropriate functions at
the mitigation site replaces the benefits lost at the other wetland.  This is most
easily seen in consideration of the benefits of the sequestration of carbon, and
its effects on global atmospheric change. For all practical purposes, if the
mitigation site sequesters carbon, it successfully replaces the carbon
sequestering benefits lost at another wetland site, regardless of the distance
between them. Thus wetland benefits with very large characteristic scales are
the most likely to be replaced with off-site mitigation.

Landscapes and Mitigation Banks

     The debate that often accompanies discussion about mitigation banking
over what the appropriate distance.,bet ween, the site of wetland impact and
the site of mitigation thus has no simple answer. The greater the distance
from the site of impact, the fewer wetland benefits will be completely and
effectively replaced, and the more redistribution of benefits will occur. The
distance between sites of impacts to wetland and mitigation should reflect a
judgment of which wetland functions and values are intended to be replaced
by the mitigation bank.  In some cases, a form of split mitigation, in which
wetland values,relatively insensitive to location (e.g. habitat values for"
migratory birds) are compensated for through a mitigation bank, while site-
specific benefits such as water quality and flood attenuation benefits are
addressed via on-site mitigation may provide a superior method of managing
aquatic resources than would either conventional mitigation approaches or -
simple use of mitigation banks.

   Wetlands Across a Modern Landscape Gradient

     Landscapes change in systematic ways as one moves from an urban
center,! through suburbs, to high intensity agriculture, through low-intensity
agriculture, to non-agriculturalforestand wildlands. This urban-rural
landscape gradient (McDonnell and Pickett 1990), will play an important role
in the shaping the consequences of mitigation banks.

     Wetland mitigation costs vary along the urban-rural landscape gradient,
with higher costs in urban settings and lower costs further from urban
centers, for several reasons. The price of land is higher in urban areas, and
the costs of some inputs to wetland mitigation projects (e.g. costs for disposal
of fill material) also decline away from cities. Urban and suburban
watersheds are often degraded by hydrolqgic modification, chemical
contamination, and thermal changes; restoration and creation projects must
be engineered to handle the stresses imposed by the surrounding urban
setting.  Urban and suburban mitigation must often be located in second or
third best sites because of conflicting high value land uses. Poorer sites drive
up project costs and reduce project performance.

     The expanded use of mitigation banks is likely to lead to a shift in the
geographical distribution of mitigation efforts away from the more expensive
urban and suburban sites and toward rural areas. Mitigation banks are not as
closely tied to impact locations as project-by-project mitigation has been.  This
gives those supplying mitigation through banks more freedom to select cost-
effective sites for mitigation. Wetland mitigation banks are therefore likely to
be located in more rural areas; where costs are lower, and (often) the
probability of producing a high quality mitigation project is higher.  Yet most
wetland impacts that trigger mitigation requirements occur within suburban
landscapes.  As a result,  expanded use of mitigation banks is likely to result in
a gradual shift in the location of wetlands out of the suburban and urban

Landscapes and Mitigation Banks.

landscapes in which most impacts are occurring, towards more rural
landscapes, in which wetland creation and restoration efforts are least

     Regulators may well perceive that they have little reason to resist
replacing urban or suburban wetlands with wetlands in rural areas. Both the •
probability of successful wetland creation or restoration and the overall
quality of the resulting wetland, are generally thought to increase as one
moves from urban to more rural settings (NRC 1992). It is far from clear,
however, whether a gradual shift in the wetland portfolio away from urban
and suburban areas to predominately rural areas \yould, in fact, be beneficial
from ecological or socioeconomic perspectives.                          ,

     As one moves from urban centers through suburban areas to rural
landscapes, the functions of an acre of wetland change.  In general the overall
quality of a wetland constructed for mitigation purposes is likely to increase
(in terms of biodiversity, biological integrity, and ability to support fish or
waterfowl populations, for example). Specific wetland functions, however,
may increase, decrease, or change in some more complex way. Some wetland
functions (e.g. flood storage capacity) are determined primarily by properties
of the wetland itself, and show little change with landscape context. Other
functions are controlled by dynamic relationships between the wetland and
its surroundings, and thus are strongly influenced by the quality of
surrounding lands.

     Habitat functions and sediment trapping functions demonstrate
contrasting ways that wetland functions may vary according to the condition
of the surrounding landscape (Figure 2a). Most habitat functions of a created
or restored wetland can be expected to increase as one moves further from
urban centers/and the proximity of the wetland to and average quality of
adjacent habitat patches improves. Thus habitat functions typically will be
inversely related to the level of urbanization of the surrounding landscape.

     The degree to which a wetland traps sediment, on the other hand,
depends on both wetland hydrology and the influx of. sediments from the
surrounding landscape. Without sediment influx induced by nearby land use
practices, the wetland may have the capacity, but not the opportunity, to trap
sediment. Agriculture and the construction of roads and suburban
subdivisions are major sources of sediments to aquatic environments. Thus
the opportunity  to trap sediments is likely to be greatest in agricultural and
suburban landscapes, intermediate'in urban landscapes, and lowest in
landscapes dominated by forest.

 Landscapes and Mitigation Banks
           Wetland Functions and Values
           Accross a Landscape Gradient
                                        Ecological Function
Societal Values
         Urban    Suburban  Agricultural Other Rural
                  Location of Wetland
 Figure 2: Wetland functions and values across the urban-rural landscape
 gradient (From Bohlen and King, 1995).

     More than just functions change as one moves from urban to rural
.landscapes.  The significance of specific functions also changes, both from an
 ecological and from a socioeconomic perspective. Similar wetlands
 embedded in different landscape contexts may exhibit similar functions, but
 generate substantially different values. The marginal value of a particular
 wetland function (the value of a small change in the level of the function)"
 may be expected to decline as one moves from landscapes  where wetlands
 and healthy environmental assets are relatively scarce (urban centers) to
 landscapes where they .are relatively abundant (rural areas).  Construction of ,a
 forested flood plain wetland for mitigation along a stream  that already has
 substantial forested flood plain, for example, may be of little ecological or
 socioeconomic value; if the stream in question is already healthy, water
 quality and ecosystem support services provided by the extra wetland area

Landscapes and Mitigation Banks

may be of little consequence. Construction of a similar wetland in a suburban
area may offer recreational opportunities otherwise unavailable in a
developed landscape (socio-economic value), and also make a significant
difference in the health of adjacent stream systems (ecological value).

    The value of certain wetland functions may also be affected by the
wetland's proximity and accessibility to human populations. Values
associated with consumptive and non-consumptive uses of wetlands such as
hunting or bird watching drop off quickly as the distance between the wetland
and human populations increases, if only because costs of travel to such sites
increases. Non-use values associated with the continued existence of a rare
animal or plant species, on the other hand, are less dependent on access  or
proximity and decline only slightly with distance.

     These relationships can be depicted as shown in Figure 2b.  The per
capita marginal value of a (hypothetical) wetland benefit declines as you
move from urban to rural landscapes, and related environmental services
becomes less scarce. The number of people who gain from the benefit
simultaneously declines with distance from the urban center. The result is
that the net value of an additional unit of a wetland benefit provided in  an
urban area may be substantially greater than the value of an additional unit
supplied in a rural area.                        ;

     The challenge for those trying to understand the consequences of
wetland mitigation banks is to combine the thinking of figure 2b with the
thinking of figure 2b. Whether wetland mitigation banks close to urban
centers is more valuable than mitigation far from urban centers rests, again,
on an understanding of environmental goals.  Is it improved local hunting,
increased regional waterfowl production, protection of biodiversity,
establishment of clean water, reduction in flood hazard, or creation of
environmental education opportunities that drives wetland policy?


     Wetland  mitigation banking will change wetland resource portfolios in a
variety of ways. The effects of banking will include changes in the size,
characteristics, and location of wetlands in the regions in which it is used.
The significance of those effects, however, remains to be seen. While great
uncertainty revolves around political, legal, and institutional features of
mitigation banking, the success or failure of banking as an environmental
management tool will rest in part on the sophistication with which
regulators are able to assess and understand the consequences of banks. That,
in turn, will require consideration of the underlying goals of wetland

Landscapes and Mitigation Banks    ,     •

     The goals of wetland protection programs have often been poorly
articulated. Legislative underpinnings have often been framed with little or,
no regard for the larger environmental goals' of wetland protection (clean
water, waterfowl, biodiversity, aquatic ecosystem support), and even less
regard for how to make trade-offs among disparate program goals.
Management of mitigation banks, however, can not be carried out in the
same seat-of-the-pants way that has sometimes characterized project-by-
project mitigation.  Clear goals will be essential. The shift to mitigation banks
represents a shift to more active intervention in landscapes, and a retreat
from the belief that the best way to  manage wetland resources is to follow
natural example. As we rely increasingly on mitigation banks, we increasingly
substitute human decisions (based, one hopes, on sound science) for the zero-
order assumption that existing wetland resources represent the best available
mix of resources for a watershed or region. Yet we will find it impossible to  '
make the necessary decisions without clear goals, and sophisticated landscape
or watershed planning. Only after the goals are established and analysis of
regional environmental resources carried out can we ever hope to make
sound decisions with respect to what changes in regional wetland portfolios
may be desirable.                         .       .

     It is ironic that even as we have moved toward policy for mitigation
banks, overall wetland regulatory policy is still tied to a project-by-project, or
impact-by-impact mind set fundamentally unable to address cumulative
impacts. Mitigation decision making, whether for mitigation via banks or
not> remains focused on on-site wetland functions. Little consideration is
given in most regulatory programs to the role of wetlands in larger ecological
and social contexts. Mitigation banking will have made a fundamental
positive contribution to environmental policy in this country if the lessons
we learn from mitigation banking help us to develop the tools for the
analysis of watersheds to enable us to address cumulative impacts in a more
rational way.                                                        ''   -

Landscapes and Mitigation Banks
     Bibliography         .                                      .      -

Albrect, V.S. and B.N. Goode. 1994. Wetland Regulation in the Real World.
      Beveridge and Diamond, P.C.  Washington,; D.C.

Angermeier, P.L. and J.R. Karr: 1994.  Biological integrity versus biological
      diversity as policy directives. Bioscience 44 (10):690-

Bayley, P.B. 1995. Understanding large river-floodplain ecosystems. Bioscience

Junk, W.J., P.B. Bayley and R.E. Sparks. 1989. The flood pulse concept in
      river-floodplain systems. Can. Spec. Publ. Fish. Aquat. Sci. 106:110-127.

National Research Council (NRC).. 1992. Restoration of Aquatic Ecosystems:
      Science, Technology, and Public Policy. National Academy Press.
      Washington, DC.

Race, M. S. 1985. Critique of present wetlands mitigation policies in the
      United States based on an analysis of past restoration projects in San
      Francisco Bay. Environmental Management 9(1): 71-82.    '

Bernstein, G., and R. L. Zepp, Jr. 1990. Evaluation of selected Wetland
      Creation Projects Authorized Through the Corps of Engineers Section
      404 Program. U.S. Fish and Wildlife Service, Annapolis Field Office,
      Annapolis, MD.

Bohlen, C.C. and D.M. King. In Press. Location and wetland values: some  .
      pitfalls of offsite wetland mitigation in the Chesapeake watershed. In S.
      Nelson and P. Hill, eds., Towards a Sustainable Coastal Watershed:
      The Chesapeake Experiment. Proceedings of a Conference. Chesapeake
      Research Consortium, Edgewater, Maryland.

Bohlen, C.C., T.V. Grasso, M. Hirshfield, B. Kooser, A. Powers, J. Watts, E.
      Zucker, and J. Goldman-Carter. 1994. Wetlands Permitting Programs
      in the Chesapeake Bay Area.  Chesapeake Bay Foundation, Annapolis

Bohlen, C.C. 1993. Wetlands politics from a landscape perspective. Maryland
      Journal of Contemporary Legal Issues 4:(1)1-11.

Chesapeake Bay Program. 1988.  Chesapeake Bay Wetlands Policy.
      Chesapeake Bay Program, Annapolis, MD.                   -

Conservation Foundation. 1988.  Protecting America's Wetlands: An Action
      Agenda. The Final Report of the National Wetlands Policy Forum. The
      Conservation Foundation. Washington DC.

Landscapes and Mitigation Banks
                                                       '   : >         •
Cowardin, Lewis M., V. Carter, F, C,Golet,and E. 1\ LaRoe. 1979.
      Classification of Wetlands and Deepwater Habitats of the United States.
      FWS/OBS-79/31 Office of Biological Services, Fish and Wildlife
      Service. U.S. Government Printing Office./Washington DC.

Crewz, D. W., and R. R. Lewis III. 1991. An Evaluation of Historical Attempts
      to Establish Emergent Vegetation in Marine Wetlands in Florida.
      Florida Sea Grant Technical Paper TP-60. Florida Sea Grant College,
      Univ. of Florida, Gainesville, FL.                                ,

Environmental Protection Agency and the Department of the Army (EPA and
      DO A). 1990. Memorandum of Agreement Between the Environmental
      Protection Agency and the Department of the Army Concerning the
      •Determination of Mitigation Under the Clean Water Act'§ 404(b)(l)
      Guidelines. June 6,1990,

Ehrenfeld, D. 1978. The Arrogance of Humanism.  Oxford University Press.

Erwin, K. L. 1991. An Evaluation of Wetland Mitigation within the South
      Florida Water Management District. South Florida'Water
      Management District, West Palm Beach, FL.         ,

Florida Department of Environmental Regulation. 1991. Report on the
      Effectiveness of Permitted Mitigation. Department of Environmental  .
      Regulation, Tallahassee, FL.

Harvey, H.T., and M. N. Josselyn. 1986. Wetlands restoration and mitigation
      policies: comment. Environmental Management 10(5): 567-9.

Jansson, A.M., M. Hammer, C. Folke, and R. Costanza. 1994. Investing in
      Natural Capital: The Ecological Approach to Sustainability.  Island
      Press.  Washington, D.C.

Karr, Jarries R. 1990. Biological integrity and the goal of environmental
      legislation: Lessons for conservation biology. Conservation Biology
      4(3):244-250. •

King, D.M. and Curtis C. Bohlen. 1994. Estimating the Costs of .Wetland
     . Restoration. National Wetlands Newsletter 16(3):3-8.

King, D.M., C.C. Bohlen and K. Adler.  Watershed management and wetlands
      mitigation: a framework for determining compensation ratios. Report
      prepared for the EPA Office of Policy, Planning and Evaluation.^
      Washington DC. University of Maryland Center for Environmental
      and Estuarine Studies Ref. No. [UMCEES] CBL-93-098.

Landscapes and Mitigation Banks

King, D.M., C.C. Bohlen, and P. Crosson. 1995.  Natural Resource Accounting
      and Sustainable Watershed Management: With Illustrations for the
      Upper Mississippi River Watershed. Report prepared for the
      President's Council on Sustainable Development.  University of
      Maryland, Center for Environmental And Estuarine Studies.
      Reference Number UMCEES CBL-95-037.

Maryland Code. 1989. Nontidal Wetland Protection Act. MD Nat. Res. Code
      § 8-1201 et. Seq.                         ;

McDonnell, MJ. and S.T.A. Pickett. 1990. Ecosystem structure and function
      along urban-rural landscape gradients: an unexploited opportunity for
      ecology. Ecology 71(4):1232-1237
Race, Margaret S., and Mark S. Fonseca. In Press. Overhauling compensatory
      mitigation: what will it take. Ecological Applications.

Richardson, C.J. 1994. Ecological functions and human values  in wetlands:  a
      framework for assessing forestry impacts., Wetlands 14(1): 1-9.

Shabman, L., P. Scodari and D. King. 1994. Expanding Opportunities for
      Successful Mitigation: The Private Credit Market Alternative.
      National Wetland Mitigation Banking Study. 1WR Report 94-WMB-3.
      U.S. Army Corps of Engineers Institute for Water Resources,
      Alexandria, Va.

Tiner, R.W., I. Kensenski, T. Nuerminger, J. Eaton, D.B. Foulis, G.S. Smith,
      and W.E. Prayer. 1994.  Recent Wetland Status and Trends in the
      Chesapeake Watershed (1982 To 1989): Technical Report.  Chesapeake  >
      Bay Program, Annapois, Maryland.

Walters, C.J. 1986.. Adaptive Managmement of Renewable Resources.
      Macmillan, New York.                                .  .    '