Content uploaded by Mark D Bertness
Author content
All content in this area was uploaded by Mark D Bertness on May 06, 2014
Content may be subject to copyright.
Coastal and Estuarine Research Federation
Reconstructing New England Salt Marsh Losses Using Historical Maps
Author(s): Keryn D. Bromberg and Mark D. Bertness
Source:
Estuaries,
Vol. 28, No. 6 (Dec., 2005), pp. 823-832
Published by: Coastal and Estuarine Research Federation
Stable URL: http://www.jstor.org/stable/3526949 .
Accessed: 11/08/2011 12:49
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
Coastal and Estuarine Research Federation is collaborating with JSTOR to digitize, preserve and extend
access to Estuaries.
http://www.jstor.org
Estuaries
Vol.
28,
No.
6,
p.
823-832 December 2005
Reconstructing
New
England
Salt Marsh Losses
Using
Historical
Maps
KERYN
D. BROMBERG*
and MARK D. BERTNESS
Department
of
Ecology
and
Evolutionary
Biology,
Brown
University,
Providence,
Rhode Island
02912
ABSTRACT:
Analyses
of habitat
loss are
often restricted to the
past
75
years
by
the relative
youth
of aerial
photography
and
remote
sensing
technologies.
Although
photographic
techniques
are
highly
accurate,
they
are unable to
provide
measurements
of
habitat loss
prior
to the 1950s.
In this
study,
historical
maps
from the late 1700s and
early
1800s
covering
portions
of Rhode
Island,
Massachusetts,
New
Hampshire,
and Maine were used to
approximate
naturally
occurring
salt
marsh
cover
in New
England.
Historical
data
was
compared
to current salt marsh
coverage
available in
public geographic
information
system
(GIS)
data sets.
The
average
loss
in New
England
is estimated
at 37%
using
this
technique.
Rhode Island
has
lost the
largest
proportion
of
salt
marshes
by
state,
a
staggering
53%
loss since
1832.
Massachusetts has also
experienced
large
losses,
amounting
to
a
41%
loss
of
salt
marsh since 1777. The Boston area alone has lost 81% of its salt marshes. Salt
marsh
loss was
highly
correlated
with urban
growth.
Restoration
and
preservation
efforts have resulted in the retention of salt
marsh
in less
populated
areas
of
New
England.
Although
historical
maps
are difficult
to
verify,
they
represent
an
extremely
valuable
and underused
data
repository.
Using
historical
maps
to trace land
use
practices
is an effective
way
to overcome the
short-term nature
of
many ecological
studies.
This
technique
could be
applied
to
other
habitats and other
regions,
wherever
accurate
historical
maps
are
available.
Analysis
of historic conditions
of habitats can
help
conservation
managers
determine
appropriate goals
for restoration
and
management.
Introduction
Habitat
destruction
has been
recognized
as
a
universal threat to
biodiversity
(Soule
1991).
Analyzing
trends
of habitat loss
on a
regional
scale
has become
more feasible
since
the
advent
of
remote
sensing
and
geographic
information
system
(GIS)
technologies,
and rates of habitat
loss have
been
tabulated more
frequently
in
the last few
decades. Habitat
loss is not a new
phenomenon.
Recent
studies have revealed that humans
have
been
significantly altering
the
landscape
since
prehistoric
times
(Flenley
et al.
1991;
Willis et al.
2004),
and
in
New
England,
that effect has
dramatically
reduced salt marsh
coverage.
Limited data
availability
has
curtailed efforts
to
document earlier
periods
of salt
marsh loss.
Aerial
photography
dates
back less than 75
yr
for
most
areas, and,
consequently,
the last 75
yr
are the
only
years
represented
in
most wetlands trends
analyses.
This
paper, using
historical
maps
of
portions
of
Rhode Island
(RI),
Massachusetts
(MA),
New
Hampshire
(NH),
and Maine
(ME),
extends
beyond
the realm of aerial
photography
to
examine
the
last
200
yr
of human
effects on New
England
salt
marshes.
Coastal habitats in the
densely populated region
of New
England
have
long experienced particularly
deleterious
anthropogenic
effects. New
England's
population
has increased
nearly continuously
in
the
last
200
yr.
Increased
population
densities
and
suburban
sprawl
resulted
in the conversion of
substantial areas
of natural land to urban
and
industrial use.
Expansion
of the coastal cities of
New
York,
New York
(NY),
New
Haven,
Connecticut
(CT),
Providence, RI,
and
Boston, MA,
has
formed
a
nearly
continuous corridor
of
developed
land.
Although
humans
have discovered
many
benefits
of
converting
salt
marshes,
there
are
countless
benefits of
maintaining
these habitats
(Costanza
et
al.
1997).
Salt
marshes
buffer
inland areas from
erosion
and
flooding
during
the severe storms that
are
characteristic of
the
region. They
are home to
filter
feeding organisms
that cleanse
polluted
waters
and
commercially
viable
species
that
thousands of
people
in
the
northeastern United States
depend
upon
for their
livelihood. Salt
marshes,
with
their
dense
intertidal
vegetation,
serve as sheltered
nurseries
for
many
species
of
young
fish, lobster,
and
shrimp
(Turner
1977;
Boesch and Turner
1984;
Bertness
1999).
Salt
marshes also are
among
the
most
productive ecosystems
on
earth,
with
primary
productivity
rates
in some
areas
comparable
to
coral
reefs and
tropical
forests
(Reidenbaugh
1983;
Mitsch and
Gosselink
1993;
Silliman and
Bortolus
2003).
All
evidence
suggests
that
salt marsh
loss
in
New
England
has been severe.
Salt
marshes once
covered
much of
the coastal
northeastern
U.S.
(Nixon
1982;
Stilgoe
1994).
Anecdotal
estimates
place
total
loss at
*
Corresponding
author;
tele:
401/863-2619;
fax:
401/863-
2166;
e-mail:
KerynBromberg@brown.edu
?
2005 Estuarine
Research Federation
823
824
K.
D.
Bromberg
and
M.
D.
Bertness
around
50%
(Agardy
1997).
Although
numerous
estimates have
documented
the rate of salt
marsh loss
over the last 50 or 100
yr,
an
appreciation
of
salt marsh
loss since
European
colonization has
proved
elusive.
HUMAN EXPLOITATION OF
NEW
ENGLAND
SALT MARSHES
New
England
salt marshes have a
history
of
exploitation dating
back
to
the arrival
of
Europeans
in
New
England.
Dutch
and
English
settlers
took to
the salt marshes as
familiar
landscapes
and
founded
towns with access to marshes in
mind
(Russell
1976).
Salt marsh
plants
were central to
early
colonial
life.
Salt
hay,
Spartina patens,
was farmed for
animal
bedding
and
used as animal feed
with
high
marsh
black
grass,
Juncus
gerardi,
mixed
in
for better
nutrition
(Nixon
1982).
Thatch
grass,
Spartina
alterniflora,
was used for
roofing
houses
(Russell
1976).
So valuable were salt marshes
in
the
1700s
that
there
are accounts of farmers
attempting
to
convert land into salt marsh
by
extending
creeks,
although
little marsh was
probably
created
in
this
way
(Nixon
1982).
In
the
mid
1800s,
salt
hay farming
fell
out of favor
as
freshwater
hay
species
became
more
commonly
used
for animal feed. Most
agriculture
moved west-
ward
following
the
promise
of
ample
and
inexpensive
fertile land
in the
Mississippi
River basin
(Pavelis
1987).
The
U. S.
Federal
Swampland
Acts
of
1849,
1850,
and
1860
passed
authority
over
large
areas
of
wetlands
to the
states,
which,
in
turn,
sold
the land to
farmers
for revenue
(Gosselink
and Baumann
1980).
Reclamation
of salt marshes
became
widespread,
as
farmers were
encouraged
to drain marshland
by
ditching
or
installing
of
tidal
gates
in
order to
cultivate freshwater
crops
(Stilgoe
1994).
Upon
the
discovery
in 1897
that
mosquitoes
are
disease
vectors,
attempts
were made
at
mosquito
eradication.
Ditching
of marshes for
mosquito
control became
common
during
the
depression,
when the Civilian
Conservation
Corps
and the
Works
Progress
Administration
ditched
over
95%
of the
northeastern
marshlands,
primarily
to offer
employ-
ment
opportunities
(Buchsbaum
2001).
Immigration
in
the
late 1800s
and
early
1900s
necessitated
housing
and
construction
projects
of
a scale that
had not been
previously
known
in the cities
of New
England.
Over
2,000
ha
of salt
marsh and mudflat in the Boston
area
were
filled
in
for
various
industrial
and urban
growth
projects,
most
of which
took
place
between 1830
and
1930
(Seasholes
2003).
Maltreatment
of New
England
salt marshes
continued
until the
1970s,
at
which
point
the
U.S.
general
public
and federal
government
began
to
recognize
the
ecological
services that
salt marshes
provide
as marine
nurseries,
shorebird
habitat,
and
coastal stabilizers.
The Federal
Water Pollution
Control
Act,
later called
the Clean
Water
Act,
was
enacted
in
1972.
Under Section 404 of
the Clean
Water
Act,
salt
marsh,
both
public
and
private,
became
protected
from
dredging
or
filling
except
by
permit
issued
by
the
Corps
of
Army
Engineers.
In
1988,
President
George
H.
W. Bush set
a
national
goal
of "no
net loss" of
wetlands and
began
more
rigorous
enforcement of
Section 404
(USGPO
1990).
Losses of wetlands
nationally
have since
slowed
(Heimlich
et
al.
1998).
New
England
salt
marshes are still
plagued by
a
number
of
problems.
Nutrient
runoff,
Phragmites
invasion,
overfishing,
and
sea level rise
continue
to
threaten
remaining
salt
marshes
(Donnelly
and
Bertness
2001;
Bertness
et al.
2004).
WETLAND Loss
ESTIMATES
Estimates of wetland
loss,
which
incorporate
salt
marsh
loss,
have
been used to assess the risk to
different
types
or locations
of
wetlands,
to
pinpoint
the
causes
of
loss,
and
to
develop
effective
pre-
vention
methods.
Surveys
of wetlands
in
the
U.S.
have
been commissioned
by
numerous federal
agencies,
including
the
National Resources Inven-
tory,
Bureau of
Agricultural
Economics,
Soil
Con-
servation Service
(all
within the U.S.
Department
of
Agriculture),
and
the National
Wetland
Inventory
(within
the U.S. Fish and Wildlife
Service),
each
agency
with its own
survey
methods. Since
2000,
federally
commissioned wetland
inventories have
been handled
exclusively by
the
National Resources
Inventory
(National
Wetland Newsletter
1998).
At least three
major
drawbacks
have limited the
utility
of
past
wetland
loss
estimates
in
interpreting
long-term
trends.
Definitions of wetlands
have
shifted over the
years,
making
a consistent
analysis
over time
difficult.
In
some
studies,
swamps
refers to
salt marshes
(Shaler 1886),
yet
in others
swamps
refers
exclusively
to freshwater marshes
(Wright
1907).
The treatment
of subtidal
vegetation
differs
between
surveys
as
well
(Gosselink
and
Baumann
1980).
The creation of a standardized
classification
system by
Cowardin
et al.
(1979)
has
fixed this
problem
in
current data sets
(Tiner
1996).
National
long-term
loss estimates
often
group
all
wetland
types
together
(e.g.,
Dahl
1990;
Heimlich
et
al.
1998).
These estimates
are useful
in
correlating
general
wetland
loss with
national economic
and
cultural
trends,
but
many
details
are lost
in this
type
of
analysis,
such as the
particular
risk
to coastal wetlands.
Estimates
of loss
require
baseline
data,
which
ideally
would
predate
human effects.
Percentages
of
loss
using
baseline data
from
the 1950s
or later
disregard
any
losses that
occurred
earlier,
which,
considering
the
high
level
of historic
human
exploitation
of salt
marshes
in New
England,
is
likely
to be substantial.
Salt
Marsh
Loss in New
England
825
Before the U.S.
Geological Survey
was formed
in
1879,
the
quality
and
availability
of historic data
sets
was
unreliable. One
exception
is the
early
maps
published by
the U.S.
Coast
Survey,
founded
in
1834. Accurate
maps containing
land
use data
from
before
then
are
rare.
Only
starting
in 1879
have
scientists had
access to
consistent,
highly
detailed,
and
accurate
maps.
Most
existing
wetland loss
estimates are
either
anecdotal
(Teal
and Teal
1969;
Watzin
1992;
Agardy
1997)
or
cover
only
the
past
50
yr (Frayer
et
al.
1983;
Heimlich
and
Melanson
1995;
Dahl
2000).
Older
baseline data of
wetlands in
the U.S.
comes
from
U.S.
Department
of
Agriculture
surveys
of
uncultivable land. While
comprehensive,
the
De-
partment
of
Agriculture's
baseline
data are
impre-
cise;
the 1906
survey
was
done
piecemeal,
with
each
county
providing
its own
data,
and the
1922
survey
of
wetlands was
based on
soil
surveys
and
drainage
reports,
which
are
capable
of
identifying
only
about
85%
of
wetlands
(Heimlich
et
al.
1998).
Only
three
prior
studies have
attempted
to
capture
wetland loss
since
European
settlement
in
North
America
by
using
baseline
historical
data from
before
1900. Dahl
(1990)
used
a
variety
of
state
park
documents
and soil
surveys
to
estimate
wetland
loss
from
the 1780s
to 1980s.
He
did not
separate
wetlands
by
type,
limiting
the
utility
of
the
data.
In
a
review
of
wetland
inventories,
Gosselink
and
Baumann
(1980)
used data
collected
by
Shaler
(1886)
and
various
government
surveys
to
reconstruct
salt
marsh
loss
in
New
England
and
New York
from
1886 to
1976.
Gosselink
and
Baumann
found
that the
period
of
most
rapid
salt marsh
loss
was
between
1922
and
1954,
with
losses
slowing
in
the
second half
of
the
20th
century.
Marsh
loss
rates
(1954-1974)
were
closely
correlated
with
population
densities of
coastal
counties.
While
thorough,
that
research
still
ignored
a
long
period
of
human
effects.
The
Connecticut
Department
of
Environmental
Protection
(CTDEP)
did a
study
comparing
CT
salt
marsh
coverage
in
the
Coast and
Geodetic
Map
Series
from
the
1880s to
1970s
using
methods
similar
to
this
study
(Dreyer
and
Niering
1995).
Like
Gosselink
and
Baumann
(1980),
the
CTDEP
study
had
no
data
from
before
the
1880s,
a
period
of
intense
population
growth
and
urban
development
in
New
England.
This
study
attempts
to
set a
baseline
earlier in
time
using
salt
marsh
coverage
data
from
historical
maps
to
provide
a
more
comprehensive
picture
of
coastal
marsh
conversion in
New
England
over
the
years
since
European
settlement.
We
expected
to
find
that
salt
marsh
loss in
New
England
over
the
last
few
centuries
is
slightly
higher
than
the
30%
loss
of
salt
marsh
found
in
CT
by
the
CTDEP
from
the
1880s-1970s
(Dreyer
and
Niering
1995).
Shared
geography
and common
history
would
likely
have
resulted
in
similar
loss rates
in other
New
England
marshes
but
the
longer
time
period
examined
by
this
study
would
account
for
wetland
losses
that
occurred
even
earlier
in
history
than
1880.
Methods
We
calculated an
estimate of
salt
marsh
and
urban
land
cover
change
in New
England by
comparing
historical
maps
with current
land
use
data.
For
portions
of New
England
where
both historical
and
current
data
were
available,
we
selected
historical
maps
and current
GIS
data,
which
delineated
marshes,
and we measured
salt marsh
and
urban
coverage
areas
within
each
map.
These
areas
were
used
to
develop
an
estimate
of
salt marsh
loss,
from
which
total
salt marsh
loss
was
extrapolated.
The Cowardin
classification
of
"estuarine
emer-
gent"
was
used
to
identify
salt marshes
in
current
GIS
data
(Cowardin
1979).
Fresh
and brackish
tidal
marshes
are
included
within
this
classification.
Fresh,
brackish,
and salt marshes
were
undifferen-
tiated
on historical
maps
and,
on most
historical
maps,
were indicated
by
a
mottled
pattern
defined
as marsh
on
the
legend.
Areas
were classified
as
urban
if
they
had four
or more residences
per
acre.
Commercial
and
industrial
areas were
also
included
as
urban
areas.
Percent
change
in salt
marsh
and
urban
area
was
calculated
as
follows,
where
current
and
historical
areas
refer
to land
use
areas,
and
a
negative
percent
implies
a
loss:
Percent
change
(current
area
-
historical
area)
x
100
historical
area
We were
able
to
compare
change
in
salt marsh
and
urban
areas
over
approximately
a
200-yr
time
interval
for
portions
of
the
coastal
states
of
RI,
MA,
NH,
and
ME
(Fig.
1).
The
actual
area
of
salt
marsh
lost
by
each
state
was
back-calculated
using
the
percent
change
and
current
area
of
salt
marsh
in the
entire
state
(from
National
Wetlands In-
ventory
[NWI]
and
MassGIS
Land
Use
data
layers,
see Table
1),
using
the
following
equation,
wherein
the
first term
represents
the
calculated
historical
area
of
salt
marsh in
the
entire
state:
Area
lost
current
statewide
area
1
-
current
statewide
area
(1
+
percent
change)
To
find
adequate
historical
maps,
we
combed
the
archives
of
the
John
Carter
Brown
Library
at
Brown
826
K.
D.
Bromberg
and M. D. Bertness
-t2.
--E
.'..
..~l..-~
?
..?.
,
.
ME
.
.
fl gg?
MR..
~y:?:?.
? .
We!'1
:F
pi:r?w
"fi-i-
wl"4`
AD=:. "
tag
. ~ ? ?-
;0
?W&N
s
wile.~?
AW7,;
?'` ?''.~
74A'~ .d
4w.
Fig.
1.
Map
areas
used to estimate
land use
changes
in
New
England
(41-44'N,
68-72?W).
Map
areas
are
oddly
shaped
because most
charts detailed
only
the coastal
land area.
Overlaps
in the York and Portsmouth
map
areas and Boston
and
Ipswich
map
areas were accounted
for in the
percent
change
calculations.
University,
the
Sterling
Memorial
Library
at Yale
University,
and the Rhode
Island Historical
Society
in a search
for the earliest
maps
containing
accurate
delineation
of salt
marshes. We also searched
the
collections of the
U.S.
Library
of
Congress,
U.S.
Naval
Archives,
and
U.S. Office of
Coast
Survey
for
relevant
material.
Maps
were included
only
if
they
were constructed
by trigonometric
survey,
depicted
land use
types
within distinct
borders,
and
had
accurately
represented
geographic
formations.
Coastline
accuracy
was
used as an indicator
of
accurate
surveying
practices
that would extend
to
land use data
as well. Loss estimates
by
state
were
calculated
using
only
maps
that covered
a
regional
area of
at least
7,500
ha,
to decrease
sampling
bias
by
maps
that
depicted
only developed
or undevel-
oped
areas.
Historical
maps
covering
an area
smaller than
7,500
ha were used
in the
development
of a
regression
model
to examine
the
relationship
between
salt marsh loss
and urban
development.
The
scope
of this
study
was limited to
RI, MA,
NH,
and
southern
ME. No
adequate
regional
historical
maps
of CT could be
found,
and the
NWI,
where
most of the
present
day
data on
salt
marsh
coverage
was
obtained,
has not
yet
been
completed
for
the CT coastline. CT
was excluded
in
the
development
of our salt marsh loss
statistic,
but
smaller
maps
from areas of the CT
coast where
the
NWI data
is
complete
were
included
in
develop-
ment of
the
regression
model. Salt marshes
in NY
and New
Jersey
are often
grouped
with
other
Northern Atlantic salt
marshes,
but were
excluded
in this
study.
Consequently,
New
England
averages
in
this
paper
are based
on six
large maps
of
portions
of
RI,
MA, NH,
and
ME
(see
Table
2
for
map
information).
The areas covered
by
the historic
maps
used
in
this
study
are shown
in
Fig.
1.
All
maps
from the 1700s came
from the
Atlantic
Neptune,
an atlas of the east coast of
the U.S.
and
TABLE 1. Historical
maps
with salt marsh
coverage
used
in
this
study.
An
asterisk
after the
map
title
denotes use
in
calculating average
salt
marsh loss
estimate for New
England.
All
maps
were used
in the
development
of the
regression
model.
Map
title
Scale
Date of
publication
Surveyor
Source
The Atlantic
Neptune: Ipswich,
1:50,000
1777
DesBarres,
J.
F.
W.
John
Carter
Brown
Library,
MA*
Brown
University
The
Atlantic
Neptune:
Boston
Har-
1:50,000
1777
DesBarres,
J.
F. W.
John
Carter Brown
Library,
bor,
MA*
Brown
University
The Atlantic
Neptune:
Plymouth,
1:25,000
1777
DesBarres,
J.
F.
W.
John
Carter
Brown
Library,
MA*
Brown
University
The Harbor of
Hyannis
1:30,000
1850
Bache,
A. D. NOAA Office
of Coast
Survey
Nantucket Harbor
1:20,000
1848
Bache,
A.
D.
NOAA
Office of Coast
Survey
The
Atlantic
Neptune:
Portsmith, 1:25,000
1779
DesBarres,
J.
F. W.
Sterling
Memorial
Library,
NH*
Yale
University
A
Chart
of
Narragansett
Bay*
1:24,000
1832
Wadsworth,
Capt.
M. S. Rhode Island Historical
Soci-
ety Library
Coast of
Maine
in the
Vicinity
of Unavailable
1851
Unknown NOAA
Office
of Coast
Survey
Kenneybunk
Port*
York River
Harbor,
Maine
1:20,000
1854
Bache,
A. D. NOAA Office of Coast
Survey
Harbors
of
Blackport
Rock and
1:20,000
1848
Hassler,
F. R. NOAA Office of Coast
Survey
Bridgeport
New Haven
Harbor
1:30,000
1846
Hassler,
F. R. NOAA Office
of
Coast
Survey
The Harbor
of New
London
1:20,000
1846
Hassler,
F. R. NOAA Office of
Coast
Survey
Salt Marsh Loss
in New
England
827
TABLE
2.
GIS
data
layers
used
in
this
study.
Source
acronyms
are GRANIT:
Geographically
Referenced
Analysis
and
Information
Transfer,
Complex
Research
Center,
University
of New
Hampshire;
MAGIC:
Map
and
Geographic
Information
Center,
University
of
Connecticut;
MassGIS: Office
of
Geographic
and Environmental
Information,
Commonwealth
of Massachusetts
Executive
Office
of
Environmental
Affairs;
MEGIS: Maine Office
of
Geographic
Information
Systems,
State of
Maine;
NWI: National
Wetlands
Inventory,
U.S.
Fish and
Wildlife
Service;
RIGIS: Rhode Island
Geographic
Information
Systems,
University
of Rhode
Island.
State
Data
layer
title
Used to
classify
Scale Date of data collection
Source
Connecticut Land use Salt marsh
1:24,000
1990
MAGIC
Land use Urban
1:24,000
1990
MAGIC
State
boundary
State
boundary
1:24,000
1995
MAGIC
Rhode Island Statewide wetlands Salt marsh
1:24,000
1988
RIGIS
Land use Urban
1:24,000
1995
RIGIS
State of Rhode
Island
State
boundary
1:24,000
1976-1983
RIGIS
Massachusetts Land use
Salt marsh
1:25,000
1999
MassGIS
Land use Urban
1:25,000
1999
MassGIS
Community
boundaries
State
boundary
1:25,000
2002
MassGIS
New
Hampshire
National wetlands Salt marsh
1:24,000
1986
NWI
inventory
Land use
Urban
1:12,000
1998
GRANIT
New
Hampshire political
State
boundary
1:24,000
1986
GRANIT
boundaries
Maine
National
wetlands Salt marsh
1:24,000
1983-1986
MEGIS
inventory
data
Digital
raster
graphics
Urban
1:24,000
1995
MEGIS
(DRGCLIP)
Maine
townships
State
boundary
1:24,000
1971
MEGIS
(METWP24)
Canada
surveyed
and
published by
J.
F.
W.
DesBarres
(Henry
Stevens,
Son
and Stiles
1937).
DesBarres
was
among
the
first to use
triangulation
for
surveying
coastlines;
the
results are
maps
that
are
uncommonly
accurate
for their
time
(Evans
1969).
Surveying
for the
atlas was
done between
1763 and
1773.
The
other historical
maps
used were
published
between
1832
and 1854
by
the U.S.
Coast
Survey,
with
the
exception
of
the
map
of
Narragan-
sett
Bay,
RI,
published
by
Capt.
M.
S.
Wadsworth
in
1832.
Throughout
the
text,
all
maps
published
between 1777
and
1854,
are
grouped
as
historical,
the
intent
being
to
establish an
early
baseline
to
which
present
day
data can
be
compared.
To
analyze
the
Atlantic
Neptune
and
Wadsworth
historical
maps using
GIS,
we
digitally photo-
graphed
the historic
maps
and
georeferenced
them
using
Blue
Marble
Graphics'
Geographic
Trans-
former
(version
4.2).
This
software
was used
to
project
the
maps
into a
NAD83
Massachusetts
State
Plane
projection,
in
which
form
they
could
be
compared
to
current
data
sets.
This
procedure
effectively
reduced
any
error
or
skew
within the
historical
maps
to the
level of
state
boundary
GIS
data
(Table
1).
Coastlines were
intentionally
lined
up
during
georeferencing,
assuming
no
change
along
the
coast
due to
sea
level rise or
erosion
(major
anthropogenic
changes
to
the
shoreline
were
avoided
and
only
natural
features
were
matched).
Although
the
rubber-sheeting
transfor-
mation
technique
is
imperfect
(see
Petry
and
Somodevilla
2000
for
discussion),
it
was the
only
option
for
transforming
unprojected
historical
maps
in order
to
compare
them with
projected
GIS data
layers
(most
of the historical
maps
were
made before
standardized
map projections
were
being
used).
The
root-mean-square
(RMS)
error
associated
with
the transformation
models for
these
five
maps
was between
160 and
440
m,
with
a
mean
value
of
245
m.
RMS error
is a
measure
of
the
distance
the historical
map points
were
refitted
to
match
the current
map
projection.
Although
these
RMS errors
are
high
by
current
standards,
this
error
was
effectively
removed from
the
analysis
during
the
transformation
process.
Maps published
by
the
U.S.
Coast
Survey
were
obtained from
National
Oceanic
and
Atmospheric
Administration's
Office of
Coast
Survey
archives
in
a
digital,
georeferenced
form
and
needed no
transformation.
The
transformation
process
did
not line
up
the
historical and
present
day
maps
seamlessly.
In
some
places,
land in
the
historical
map
covered sea in
the
present
day
map
and
vice
versa,
despite
the
assumption
of
no
coastline
change.
For this
reason,
only
changes
in
area
were
analyzed.
Conversion
of
tidal
marshes
to
different
land
use
types
could
not
be
addressed.
Land
coverage
data
was
analyzed
in
ESRI's
mapping
program
Arcview
3.3.
Salt
marsh
and
urban
land
use
features of
historical
maps
were
hand-outlined in
GIS and
converted
into
digital
shapefiles.
Hand-outlined
historical
salt
marsh
features
excluded
tidal
creeks,
because
water
bodies
are also
excluded in
current GIS
wetland
data
sets.
All
current
wetland
and
urban
data
used
were
available
through
public
GIS
catalogs
(NH
GRANIT,
828
K.
D.
Bromberg
and
M. D.
Bertness
60-
53
50-
41
S40-
37
I-
30-
Cu
220-
18
10
-
0
RI MA
NH
ME
All
NE
Fig.
2.
Percentage
of salt marsh
lost
in
Rhode Island
(RI),
Massachusetts
(MA),
New
Hampshire
(NH),
and Maine
(ME)
over the last
200
yr.
At
right,
a
weighted average
of
states losses are
used to estimate salt marsh
loss over
all
of
New
England
(NE).
Numbers above bars
are the
percentage
values.
MAGIC, MassGIS, MEGIS, NWI,
and
RIGIS,
Table
1).
Current GIS
data
layers
were
clipped
to
the
areas detailed
in
the historic
maps
to make the
data sets
comparable.
When the historical and current
data
sets
were
both
in
digital
forms,
they
were divided
by
state
and
by
watershed for
comparisons
between
culturally
and
geographically
relevant land areas. Classifica-
tion scheme of watersheds varied state
to state. In
all
states,
major
basin or the
equivalent
was
used.
A
polynomial regression
model
was used to
relate
the area of urban
growth
to the area of salt
marsh
loss,
with a
square
root
power
transformation used
to normalize the urban
coverage
data. The
12
map
areas
(6
regional
and
6
smaller
maps, Fig.
1)
covered
portions
of
22
different watersheds
(n
=
22
for the
regression
model).
Results
Based
on
the
sampled portions
of New
England
examined
in
this
study,
37%
of the
original
salt
marsh of New
England
has been lost. RI
has lost the
highest percentage
of salt
marsh,
a
53%
loss
(Fig.
2).
MA
has lost the
second
largest percentage,
41%.
Most of the loss
in MA
occurred around
Boston;
the
greater
Boston area has lost
81%
of its marshes
since 1777
(Fig.
3).
NH has lost a lower
proportion
of salt
marsh, 18%,
and
ME
has lost
only
1
ha of salt
marsh or
<1%
since 1851.
It should be
noted that
in
the
Kennebunk, ME,
map
area,
57%
of the
remaining
salt marsh is
protected
within the
Rachel Carson National Wild-
life
Refuge.
Only
a small
proportion
of Maine's
coast was
adequately
detailed
by
historic
maps,
and
the inclusion of the
refuge
in
the
sample
may
have
made the state's
average
percentage
of loss lower
than it would
have been
if
data from the entire
coastline had been
considered.
From the
percent
loss
estimate,
we
calculated
area
lost
by
each state.
According
to this
calculation,
MA
has lost the
largest
area of salt marsh at
13,352
ha.
RI
has lost
1,831
ha. NH has lost 500
ha,
and
ME
1777
1999
,
mSab
Marsh
Bosto
mlrba
R
?P
~qBbolp
cps
CP
0 5 10 15 20 25
KIometars
, ....
..
Fig.
3. Salt
marsh
and
urban
land
cover
in
Greater Boston
(42'N, 710W)
in
1777 and 1999. The coastlines in
both
maps
are the 1999
coastline and include some land built on
fill
that did not
exist
in 1777
(identifiable
from the
unnaturally
shaped
coastline made
up
almost
entirely
of wharves
surrounding
the star
denoting
Boston).
Salt
Marsh Loss
in
New
England
829
4000
3000-
8
2000-
J
S000-
c,
?
-1000
0 20
40 60 80
100
120
Square
root
transformation
of urban
growth
(ha)
Fig.
4.
The
percentage
of salt marsh
loss increased
signifi-
cantly
with the
extent
of
urbanization
(p
<
0.001,
R2 =
0.8889,
f(x)
=
0.356x2
-
7.847x
+
62.328).
Each
point
represents
growth
of urban area
and loss
of
salt
marsh area estimated
within one
watershed
over a
period
of about
200
yr.
Area
of
urban
growth
was
square
root
normalized
to increase
accuracy
of the
regression,
and
2
ha was added to
every
urban
growth
value to allow
square
root transformation of one
negative
value
(f(x)
=
(y
+ 2)).
has lost 569
ha. For
reference,
according
to
NWI
data,
there
currently
remains
30,679
ha
of coastal
marsh
in those four states combined.
There was a
relationship
between
area of salt
marsh lost
and area of urban
land
gained.
Salt
marsh loss was
significantly
correlated
with urban
growth (Fig.
4,
R2
=
0.8889,
p
<
0.001).
At low levels
of urban
growth,
little
to no salt
marsh
was
lost,
and
7
watersheds of
low urban
growth
(<500 ha)
showed
a
slight
increase
in
salt marsh
coverage
over
the last two centuries.
Discussion
SALT MARSH CONVERSION
As
expected,
the
greater
amount of time ac-
counted
for in this
estimate resulted
in
the
finding
of a
greater
amount
of salt
marsh loss than
past
short-term
estimates,
in
some cases
by
an order
of
magnitude. Frayer
et al.
(1983)
estimated an
8%
loss of estuarine
emergent
wetlands
(salt
marsh and
mangroves,
as
classified
by
the Cowardin
system)
in
the
U.S. between 1954 and 1974. Dahl
(2000)
found
a
less
than
1%
loss of
estuarine
emergent
wetlands
in
the
U.S. between 1986 and
1997. These short-
term
estimates are useful
in
analyzing
broad trends
of
wetland
loss,
but the
absolute losses of
estuarine
emergent
wetland in
New
England
states has been
much
greater
over the
long
term.
Our
estimate of
37%
loss of
New
England's
salt
marshes over the
last
200
yr
is consistent with
CTDEP's
finding
of a
30%
loss of
salt marshes in
CT
from 1880
to 1970.
Gosselink
and Baumann
(1980)
found
a
54%
loss of coastal
marsh in
New
England
and
Atlantic
NY from 1886 to 1976. Our
estimate
is
lower,
despite
the
longer
time
interval,
perhaps
due
to the exclusion
of
NY,
where
explosive
urban
growth
has no
doubt resulted in
considerable
marsh
losses.
The
direct
cause
of loss is
difficult
to ascertain.
Sea
level
rise,
hydrologic
alterations
(by damming,
ditching,
or
filling),
and
development
of urban
or
agricultural
land
are all common causes of salt
marsh
conversion
(Roman
et
al.
1984;
Dahl
1990)
and are
probably
all,
in
part,
responsible
for the
losses
in
New
England.
In
the time
span
covered
by
this
study,
multiple
conversions
may
have taken
place.
Many
northeastern salt
marshes were
con-
verted to
cropland
in the 1800s and later
converted
to urban land.
With
only
two
snapshots
in time
of
land
use,
we could
incorrectly
conclude
that these
marshes
were
converted
directly
to urban land.
There
is evidence that
urban
growth
was
a direct
cause of salt
marsh loss. Salt
marshes were valued as
sources
of natural
resources,
such as for salt
hay,
even without
conversion to
agricultural
land.
In
the
mid 1800s
when conversion
techniques,
such as
damming
and
filling,
became
more efficient
and
commercially
available,
the
country's agricultural
center
had
already
moved west of New
England.
The
positive
correlation between
area of salt
marsh
lost and urban
growth
suggests
that urban
development
has been a
large
cause of coastal
habitat destruction
in New
England.
In
greater
Boston,
70%
of the
original
salt marsh is now urban
land
(Fig.
3),
and much was converted
directly
from
salt
marsh to residential and industrial land
(Sea-
sholes
2003).
LIMITATIONS
OF
USING HISTORICAL MAPS
FOR
COMPARATIVE MAPPING
Historical
maps
and
literature
represent
a
rich
data source
and
a valuable tool
in
overcoming
the
short-term nature of
many
ecological
studies.
Working
with historical data also has its
limitations.
Assessing
the
accuracy
of
old
maps
is difficult.
Ideally,
several
maps
of each area
from the same
time
could be
analyzed
and
averaged
to correct for
inaccuracies,
but a
scarcity
of
accurate
historical
maps depicting
salt marshes
made
repetition
un-
feasible in
this
study.
RI's
estimate
is
based
on
only
one
map,
and ME's and NH's
estimates are
each
based
on
two.
Historical
maps
were not
available for the
entire
coastline. The data
presented
here
provides only
percentage
estimates of
loss,
based on a
subset of
the coastline of each
state. The
historical
maps
used
here as a
representative
sample
of
New
England
covered
2,220
km
of
coastline or
20%
of
the
coastlines of
RI, MA, NH,
and ME.
Published
maps
830
K. D.
Bromberg
and
M. D.
Bertness
are also
inherently
biased
by
their intended
purpose.
The historical
maps
used
in this
study
were all
made for coastal
navigation
and had
detailed coastal
land use and were not
particularly
biased towards
urban centers. Urban
(e.g.,
Boston,
MA,
and
Portsmouth,
NH),
suburban
(e.g.,
Barring-
ton, RI,
and
Kennebunk,
ME),
and rural areas
(e.g.,
Plum
Island,
MA)
were
included
in
these
maps.
EFFICACY
OF
SALT
MARSH CONSERVATION
EFFORTS
The data
in
this
paper
can
alternatively
be
approached
as case studies.
Where
have salt marshes
been
conserved,
where
not,
and
why?
Within
the
Ipswich,
MA,
map
area,
little salt marsh
area
(8%)
has been lost. Low
levels
of
urbanization
(1.4%)
and
restoration
efforts
are
probably responsible
for
the
preservation
of
these marshes.
The towns
around
Ipswich
have worked
with
state environmental
agencies
and
scientific
institutions
to undertake
56
restoration
projects
on the
north shore
of
MA,
at
least
35
of which have
been
completed.
Though
the
equivalency
of
restored
marshes
to
pristine
marshes
is under debate
(Zedler
and
Lindig-Cisneros
[2000]
report
restored
salt
marshes
to be
<60%
function-
ally equivalent
to
natural
salt
marsh),
restored
marshland
is
better
habitat
than
asphalt
for salt
marsh
flora and
fauna.
The effectiveness
of
conservation
is demonstrated
in
the
Kennebunk,
ME area.
The area
of salt
marsh
around
Kennebunk
has
actually
increased
by
3%.
This
3%
could
represent
growth
of salt
marsh
by
natural
processes
or the
extent
of error
in the
estimation
techniques.
The fact
that
salt
marsh
coverage
has not
decreased
is
likely
the
result
of
explicit
protection
and
management
of wetlands
within
the
Rachel
Carson
National
Wildlife
Refuge
since
1966.
The
Ipswich
and Kennebunk
areas
provide
evidence
that
conservation
and
restoration
are
effective
tools
in
preserving
salt
marshes.
THE FUTURE
OF
SALT
MARSHES
In the
U.S.,
recent
declines
in
rates
of salt
marsh
loss
are
encouraging,
although
the future
of
the
remaining
salt
marshes
in New
England
is
un-
certain.
Loss
estimates
describe
only
the
presence
or
absence
of
marshes;
they
communicate
nothing
of
marsh
health.
No GIS
data
are
currently
available
on
the health
of
wetlands.
In
some
states,
environ-
mental
agencies
are
working
to create
GIS
data
layers
that
will assess
salt
marsh health
(Tiner
2003;
Pesch
personal
communication).
The
NWI
is
updating
its
digital
database
of wetlands
(data
currently
available
is
from
digitized
photographs
that were
taken
in
the
early
1980s),
with
a focus
on
heavily
populated
coastal
areas
(U.S.
Fish and
Wildlife Service
2002).
Descriptive
and
timely
data
on
wetlands
will increase our
understanding
of the
formidable threats to their
existence and
help
focus
conservation efforts.
APPLICATIONS
OF LONG-TERM HABITAT
LOSS
ESTIMATES
Comparative mapping techniques
using
historical
maps
should be
applied
to other coastal
ecosystems,
particularly
in
regions
with well-documented
histo-
ries
of land use.
Already
comparative mapping
techniques
have been
successfully
used to estimate
changes
in
eelgrass,
Zostera
marina,
cover
in south-
eastern
MA
(Costa 1988),
salt marsh
in
central
California
(Grossinger
2001;
Van
Dyke
and Wasson
2005),
and wetlands
in the fenland
region
of
southeastern
England
(Butlin 1995).
Louis
Agassiz
visited coral reefs while
employed
by
the U.S. Coast
Survey
in
the
1800s,
and
good
historical
data
may
exist for
a
change
analysis
of that valuable
habitat
(Shalowitz
1964).
It
is
only
with
an
historical
perspective
that
current
monitoring
programs
will succeed.
Rem-
nants
of
past
land uses
are
often seen
in the
landscape
today
and can
be mistaken
for a
natural
state.
Understanding
the historical
alterations to the
natural state
of a
habitat
can
help
resource
managers
answer
the fundamental
question
of how
best
to
reverse
decades
of human effects
and
restore
a habitat
to its natural
state.
ACKNOWLEDGMENTS
This
manuscript
has benefited
from
the comments
of
R.
Grossinger
and an
anonymous
reviewer.
We
would
like
to
thank
Brown
University
for
providing
resources
and
funding
for
this
study.
L.
Carlson
was
a
great
help
with
GIS
techniques.
We thank
J. Hogan
for statistical
advice.
S. Danforth
of the
John
Carter
Brown
Library,
F. Musto of
Yale
University
Library,
and
D.
Munroe of Rhode
Island
Historical
Society helped
in our search
for
historical
maps.
This work
would
not have been
possible
without
the
GIS
data
sets
provided
by
University
of
Connecticut's
MAGIC,
University
of
Rhode Island's
RIGIS,
Massachusetts'
MassGIS,
University
of New
Hampshire's
GRANIT,
and
the State
of
Maine's
MEGIS.
LITERATURE
CITED
AGARDY,
T.
S.
1997.
Marine
Protected
Areas
and
Ocean
Conservation.
R.
G.
Landes
Company
and
Academic
Press,
Austin,
Texas.
BERTNESS,
M.
D. 1999.
Ecology
of Atlantic
Shorelines.
Sinauer
Associates,
Sunderland,
Massachusetts.
BERTNESS,
M.,
B. R.
SILLIMAN,
AND
R.
JEFFERIES.
2004.
Salt
marshes
under
siege.
American
Scientist
92:54-62.
BOESCH,
D. F. AND
R.
E. TURNER.
1984.
Dependence
of
fishery
species
on salt
marshes:
The
role of
food
and
refuge.
Estuaries
7:460-468.
BUCHSBAUM,
R
2001.
Coastal
marsh
management,
p.
323-350.
In
D. M.
Kent
(ed.),
Applied
Wetlands
Science
and
Technology,
2nd
edition.
Lewis
Publishers,
New
York.
BUTLIN,
R. 1995.
Social,
economic and
environmental
change
in
the fen
and
fen-edge
region
of east
Cambridgeshire
in the
Salt Marsh
Loss
in New
England
831
nineteenth
century,
p.
146-168.
In R. A.
Butlin
and
N.
Roberts
(eds.),
Ecological
Relations
in
Historical
Times.
Blackwell
Publishers,
Cambridge,
Massachusetts.
COSTA,J.
E. 1988.
Distribution,
production,
and historical
changes
in
abundance of
eelgrass
(Zostera
marina
L.)
in
southeastern
M.A. Ph.D.
Dissertation,
Boston
University,
Boston,
Massachu-
setts.
COSTANZA, R.,
R.
D'ARGE,
R.
DE
GROOT,
S.
FARBER,
M.
GRASSo,
B.
HANNON,
K.
LIMBURG,
S.
NAEEM, R.
V.
O'NEILL,J.
PARUELO,
R.
G.
RASKIN,
P.
SUTTON,
AND
M. VAN DEN BELT.
1997. The
value
of
the
world's
ecosystem
services
and natural
capital.
Nature
387:
253-280.
COWARDIN,
L.
M.,
V.
CARTER,
F.
C.
GOLET,
AND E. T.
LAROE.
1979.
Classification
of
Wetlands and
Deepwater
Habitats
of
the
United States.
Department
of the
Interior,
U.S. Fish
and
Wildlife
Service,
Washington,
D.C.
DAHL,
T. E. 1990.
Wetland Losses in
the United
States 1780s
to
1980s. U.S.
Department
of the
Interior,
Fish
and
Wildlife
Service,
Washington,
D.C.
DAHL,
T. E.
2000.
Status
and Trends
of
Wetlands
in
the
Conterminous United
States 1986 to
1997.
U.S.
Depart-
ment of
the
Interior,
Fish
and
Wildlife
Service,
Washington,
D.C.
DONNELLY,
J.
P. AND
M. D.
BERTNESS.
2001.
Rapid
shoreward
encroachment of salt
marsh
cordgrass
in
response
to
acceler-
ated
sea-level
rise.
Proceedings
of
the
National
Academy of
Sciences
98:14218-14223.
DREYER,
G. D. AND
W. A.
NIERING.
1995. Tidal
marshes of
Long
Island Sound:
Ecology,
History,
and
Restoration.
Connecticut
College
Arboretum
Bulletin
No. 34. Connecticut
College
Arboretum,
New
London,
Connecticut.
EVANs,
G. N.
D. 1969.
Uncommon
Obdurate: The
Several
Public
Careers
of
J.
F. W.
DesBarres.
Peabody
Museum of
Salem,
Massachusetts,
U.S.A. and
University
of
Toronto
Press,
Toronto,
Canada.
FLENLEY,
J.
R.,
S. M.
KING,
J. JACKSON,
C.
CHEW,
J.
T.
TELLER,
AND
M.
E.
PRENTICE.
1991. The
late
quaternary
vegetational
and
climatic
history
of
Easter
Island.
Journal
of
Quaternary
Science
6:85-115.
FRAYER,
W.
E.,
M.
J.
MONAHAN,
D.
C.
BOWDEN,
AND
F. A.
GRAYBILL.
1983.
Status
and
Trends of
Wetlands
and
Deepwater
Habitats in
the
Conterminous
United
States,
1950's to
1970's.
Colorado
State
University,
Fort
Collins,
Colorado.
GOSSELINK,
J.
G.
AND
R.
H.
BAUMANN.
1980.
Wetland
inventories:
Wetland
loss
along
the United
States
coast.
Zeitschrift
fir
Geomorphologie
Supplementband
34:137-187.
GROSSINGER,
R.
2001.
Documenting
local
landscape
change:
The
bay
area
historical
ecology
project,
p.
425-442.
In
D.
Egan
and
E.
Howell
(eds.),
The
Historical
Ecology
Handbook: A
Restorationist's
Guide
to
Reference
Ecosystems.
Island
Press,
Washington,
D.C.
HEIMLICH,
R. AND
J.
MELANSON.
1995.
Wetlands
lost,
wetlands
gained.
National
Wetlands
Newsletter
17:22-25.
HEIMLICH,
R.
E.,
K.
D.
WIEBE,
R.
CLAASSEN,
D.
GADSBY,
AND
R.
M.
HOUSE.
1998.
Wetlands
and
Agriculture:
Private
Interests
and
Public
Benefits.
U.S.
Department
of
Agricul-
ture,
Agricultural
Economic
Report
No. 765.
Washington,
D.C.
HENRY
STEVENS,
SON
AND
STILES.
1937.
Notes
on
Biographical
and
Bibliographical
on
The
Atlantic
Neptune.
Henry
Stevens,
Son
and
Stiles,
London,
England.
MITSCH,
W. J.
AND J. G.
GOSSELINK. 1993.
Wetlands, 3rd
edition.
Van
Nostrand Reinhold,
New York.
NATIONAL
WETLAND
NEWSLETTER.
1998. NRI
to be
national
standard
for
wetland
inventory.
National
Wetland
Newsletter
3:27.
NIXON,
S.
W.
1982.
The
Ecology
of
New
England
High
Salt
Marshes:
A
Community
Profile.
U.S.
Department
of
the
Interior,
Fish
and
Wildlife
Service,
Washington,
D.C.
PAVELIS,
G.
A. 1987. Farm
Drainage
in the United
States:
History,
Status,
and
Prospects.
U.S.
Department
of
Agriculture.
Mis-
cellaneous
Publication
No.
1455.
Washington,
D.C.
PETRY,
F. E. AND
M.
J.
SOMODEVILLA.
2000.
Modeling
issues
for
rubber-sheeting process
in an
object
oriented,
distributed
and
parallel
environment,
In
R.
Loganantharaj,
G. Palm,
and
A.
Moonis
(eds.),
Proceedings
of the
13th
International
Confer-
ence
on
Industrial
and
Engineering
Applications
of
Artificial
Intelligence
and
Expert Systems:
Intelligent
Problem
Solving:
Methodologies
and
Approaches.
Springer-Verlag,
Berlin,
Ger-
many.
REIDENBAUGH,
T.
G. 1983.
Productivity
of
cordgrass,
Spartina
alterniflora,
estimated
from
live
standing
crops, mortality,
and
leaf
shedding
in
a
Virginia
salt marsh.
Estuaries
6:57-65.
ROMAN,
C.
T.,
W.
A.
NIERING,
AND
R. S.
WARREN.
1984.
Salt
marsh
vegetation change
in
response
to tidal restriction.
Environmental
Management
8:141-150.
RUSSELL,
H.
S.
1976.
A
Long,
Deep
Furrow:
Three Centuries
of
Farming
in New
England.
University
Press
of New
England,
Hanover,
New
Hampshire.
SEASHOLES,
N. 2003.
Gaining
Ground:
A
History
of
Landmaking
in
Boston.
Massachusetts
Institute
of
Technology
Press,
Cam-
bridge,
Massachusetts.
SHALER,
N.
S.
1886. Sea-coast
swamps
of the eastern
United
States.
U.S.
Geological Survey,
6th Annual
Report,
Washington,
D.C.
SHALOWITZ,
A.
L. 1964. Shore
and
Sea Boundaries.
U.S.
De-
partment
of
Commerce
and
Coast
and Geodetic
Survey,
Publication
10-1.
Washington,
D.C.
SILLIMAN,
B. R.
AND A.
BORTOLUS. 2003.
Underestimation
of
Spartina
productivity
in western
Atlantic
marshes:
Marsh
invertebrates
eat more
than
just
detritus.
Oikos
101:549-
554.
SOULE,
M. E. 1991. Conservation:
Tactics
for
a
constant
crisis.
Science
253:744-750.
STILGOE,J.
R. 1994.
Alongshore.
Yale
University
Press,
New
Haven,
Connecticut.
TEAL,
J.
AND M. TEAL.
1969.
Life
and Death
of
the Salt
Marsh.
Atlantic
Monthly
Press,
Boston,
Massachusetts.
TINER,
R.
1996. Wetland
definitions
and
classifications
in
the
United
States,
In
J.
D.
Fretwell,
J.
S.
Wiliams,
and
P.
J.
Redman
(eds.),
National
Water
Summary
on
Wetland
Resources.
Water
Supply
Paper
2425. U.S.
Geological
Survey,
Washington,
D.C.
TINER,
R. W.
2003.
Correlating
Enhanced
National
Wetlands
Inventory
Data
with
Wetland
Functions for
Watershed
Assess-
ments:
A Rationale
for
Northeastern
U.S.
Wetlands.
U.S.
Fish
and
Wildlife
Service,
Hadley,
Massachusetts.
TURNER,
R. E. 1977.
Intertidal
vegetation
and
commercial
yields
of
penaeid
shrimp.
Transactions
of
the
American Fisheries
Society
106:411-416.
U.S.
FISH
AND
WILDLIFE
SERVICE,
NATIONAL
WETLAND
INVENTORY.
2002.
A
Strategy
for
the 21st
Century,
U.S. Fish
and
Wildlife
Service,
National
Wetland
Inventory,
Washington,
D.C.
U.S.
GOVERNMENT
PRINTING
OFFICE
(USGPO).
1990.
Public
Papers
of
the
Presidents
of
the United
States:
George
Bush.
1989:
Book
I.
Washington,
D.C.
VAN
DYKE,
E. AND
K.
WASSON.
2005. Historical
ecology
of
a
central
California
estuary:
150
years
of
habitat
change.
Estuaries
28:173-189.
WATZIN,
M.
C.
1992.
The
Fragile
Fringe:
Coastal
Wetlands
of
the
Continental
United
States.
Sea
Grant,
U.S. Fish
and
Wildlife
Service,
and National
Oceanic
and
Atmospheric
Administra-
tion,
Washington,
D.C.
WILLIS,
K.J.,
L.
GILLSON,
AND
T.
M.
BRNCIC.
2004.
How
"virgin"
is
virgin
rainforest?
Science
304:402-403.
WRIGHT,
J.
O.
1907.
Swamp
and
Overflowed
Lands
in
the
United
States.
U.S.
Department
of
Agriculture,
Circular
No.
76.
Washington,
D.C.
ZEDLER,
J.
B. AND
R.
LINDIG-CISNEROS.
2000.
Functional
equivalen-
cy
of
restored
and
natural
salt
marshes,
p.
565-582.
In
M. P.
Weinstein
and
D. A.
Kreeger
(eds.),
Concepts
and
Controver-
832
K.
D.
Bromberg
and
M. D.
Bertness
sies
in
Tidal
Marsh. Kluwer Academic
Publishers, Boston,
Massachusetts.
SOURCE OF
UNPUBLISHED
MATERIALS
PESCH,
C.
personal
communication. U.S.
Environmental Pro-
tection
Agency,
Environmental Effects
Research
Laboratory,
Atlantic
Ecology
Division,
27
Tarzwell
Drive,
Narragansett,
Rhode Island
02882.
Received,
March
1,
2005
Revised,
August
4,
2005
Accepted, August
16,
2005
