Passive retreat of Massachusetts coastal upland due to relative sea-level rise executive summary

[From the U.S. Government Printing Office, www.gpo.gov]

Massachusetts Coastal Zone Management
MASSACHUSETTS COASTAL SUBMERGENCE'PROGRAM

Passive Retreat of Massachusetts Coastal Upland
Due to Relative Sea-Level Rise

Executive Summary
CZIC FILE COPY

monwealth of Massachusetts
@hael S. Dukakis, Governor
GB cutive Office of Environmental Affdirs
459.4 1; S. Hoyte, Secretary
.G53 issachusetts Coastal Zone Management
1987 ard F. Delaney, Director
c.2
71

Executive Summary

PASSIVE RETREAT OF -MASSACHUSETTS COASTAL UPLAND
DUE TO RELATIVE SEA-LEVEL RISE

by G.S. Giese, D.G. Aubrey and P. Zeeb
Woods Hole Oceanographic Institution
Woods Hole, MA 02543

U IS - DEPARTMENT OF COMMERCE NOAA.
COASTAL SERVICES CENTER
2"19_34 SOUTH HOBSON AVENUE
CHARLESTON , SC 29405-2413

PrOPOrtY Of CSC Library

The preparation of this document was funded by the Office of Ocean and Coastal
Resources Management, National Oceanic and Atmospheric Administration, U.S.
Deparlment of Commerce, under a program implementation grant to the
Commonwealth 'of Massachusetts.

Executive S-amary
FIE

PASSIVE RETREAT OF MASSACHUSETTS COASTAL UPLAND
DUE TO RELATIVE SEA-LEVEL RISE

INTRODUCTION

Shoreline recession is recognized widely as a major environmental management issuein
Massachusetts as well as in many other parts of the United States and throughout the world (Bird,
1976). In considering this issue, it is essential to separate the retreat of coastal upland areas fr&m
the retreat of wetlands because of the differences between the processes involved. The retreat of a
barrier beach, for example, may involve the landward translation of an entire. feature without
diminution in its size, but upland retreat always results in the loss of upland area. Although upland
loss usually is accompanied by wetland gain, the upland lost is an irreversible loss of that area from
those land uses for which wetlands are considered unfit. In Massachusetts these uses include, for
example human habitation, transportation and commerce.
Coastal upland retreat takes two distinct forms: active wave-produced erosion and passive
loss resulting from relative sea-level rise. While a rise in relative se a level contributes to active
wave-produced erosion, it is not possible at present to quantify the contribution to erosion made by
sea-level rise. On the other hand, the recession of a passive shoreline as sea level rises can be
estimated with reasonable accuracy.
Unfortunately, estimates of passive shoreline recession are seldom available, probably
because upland loss due to this cause generally is considered to be small compared to that due to
erosion. Relative sea-level rise along the Massachusetts coast over the past 40 years ranges between
2 and 3 mm'. per year (Aubrey and Emery, 1983). Within recent years, however, a rapidly
increasing body of data has appeared in support of the hypothesis that global climatic warming
within the next century will cause increasing global sea level rises that can not be ignored. Hoffman
(1984), for example, has projected global sea-level rises by the year 2100 ranging from 1.8 ft.
("low scenario") to 11.3 ft. ("high scenario").
Some emphasis in this report is placed on relative sea-level rise rather than absolute sea-level
rise. Coastal submergence results not only from rise of ocean levels, but also from sinking of the
land. In Massachusetts, nearly two-thirds of the submergence during the past century (documented
by tide-gauge data) results from subsidence of the lancL Only one-third of the submergence appears

to be due to ocean rise (Aubrey and Emery, 1983; Braatz and Aubrey, in press). In quantitative
terms, over the past sixty years Massachusetts has been sinking at a rate of 1.9 mrn/yr (0.0062
ft/yr) while the ocean has been rising at 1 mm/yr (0.003 ft/yr) on average.
The estimates of magnitude of sea-level rise provided by Hoffman (1984) do not include the
effect of land sinking. If the higher rate of rise scenarios (averaging 28 mm/yr) prove correct, the
impact of land submergence is important only in the short term (the next 20-30 years). If the lowest
rate-of-rise scenarios prove correct, however, then land subsidence will be a large fraction of the
magnitude of sea-level rise (2 mrn/yr versus 4-6 mm/yr).
'Me calculations based on hypsometric curves in the present study include the effect of land
submergence but th.e color-coded maps do not. The reader should be aware of this continental
submergence, however, particularly if the low sea-level rise scenario is assumed to be true. Since
the management implications for a lower rate of relative sea-level rise are less stringent than for the
higher scenarios, the explicit neglect of land motions on the color-coded maps is justified.
The study reported here was designed to quandfy the passive retreat of upland within-the
coastal communities of Massachusetts due to relative sea-level rise. The losses that presently occur
annually, and those that will occur by the year 2025 given three specified projections of future
relative sea-level rise, are presented for each community. Also presented are data that provide the
means for predicting the rates and cumulative amounts of land area losses due to passive retreat that
these communities will suffer in the future given any specified future relative sea-level rise or tidal
range change scenario. Finally, color-coded maps -are presented for the harbors of Hyannis,
Westport and Gloucester that display the areas that would be lost by the year 2 100 given any one of
four different sea-levei rise scenari os. An appendix contains tables, graphs and figures that present
the results of the study. A detailed description of the data analysis methodology also is included in
the appendix.
ne three sea-level rise scenarios presented here illustrate the potential magnitude of coastal
flooding from global climate warming. These scenarios are based on predictions containing
significant uncertainties, given the lack of precise understanding of complex atmosheric chemical
exchanges, ocean/atm osphere interactions, and effects of landalbedo, for example. Consequently,
the three scenarios presented are not necessarily the most probable sea-level rise scenarios; however
they are commonly cited scenarios widely*thought to approximate the range of possible impacts.
One advantage of the hypsometric data is the ease with which updated scenarios can be applied, to a
coastal town to obtain a first-order quantification of impac ts, on that town.
Application of the results of this study to coastal zone management and policy is fraught with
theoretical and practical questions. How might the Commonwealth, a coastal county, or a coastal
town respond to these results, and incorporate them into a managment scheme? With the

.1 Color-coded maps are included in the full report, available at the
Main Library and Planning Office. of each coastal comunity. Additional
copies may be viewed at the main czm office in Boston; regional CZM-
offices in Gloucester, Norwell, N-Dartmouth and Barnstable; and the'
Martha's Vinyard Commission in Oak Bluffs.
2
2Color-coded maps and appendices can be found in the full report.

considerable uncertainty in the scientific basis for predicting the details of global warming, how can
these uncertainties be translated into an equitable planning or zoning process? That -a global
warming is in process and will continue is incontrovertible. What are not known precisely are the
magnitude and timing of this global warming, and its exact impact on sea levels.
The appropriate response to these issues and results on local and state-wide levels is one of
increasing awareness. Legislation and re-zoning may-be premature. However, awareness by t own
planners, politicians, and Conservation Commissions, for instance, must be increased.
Long-range planning could take these shoreline retreat data into account when making major land
use decisions. Conservation Commissions could err on the side of caution in a coastal construction
issue, mandating pile foundations in areas of critical concern. Public works could incorporate these
data in siting wells or new sewer systems. In summary, some rational response to these sea-level
rise issues are appropriate at this time. Major legislation and drastic changes in regulations,
however, may be premature and might bet ter await a clearer consensus from the scientific
community before enactment.
Users of the data presented in this report must be aware that passive shoreline retreat via
inundation is not the sole-effect arising from global warming to which coastal communities 'must
respond. Although the present study considers only the effect of passive retreat due to inundation,
other impacts may be equally important. For example, rising sea levels will change the base level
for river drainage and groundwater flow. Water quality deterioration may result from this impact.
In addition, global warming will raise the ocean surface temperature, increasing the size of the
"warm-pool" of water that is responsible for generating tropical cyclones. Although difficult to
predict in detail because of the complexities of non-linear atmospheric physics, this ocean warming
is certain to alter storm chmates along the eastern seaboard and elsewhere. If the net product is an
increase in tropical cyclones reaching the northeast, this could result in more severe short-term
(order of decades) economic impact than that due to simple passive retreat. While the present study
investigates an issue of fundamental importance, the user should be aware of these other significant
impacts, and p Ian their rational response to global warming accordingly.

METHODS
Quantification of the passive retreat of coastal upland presents special problems due to the
peculiar "fractal" nature of the passive shoreline (Mandelbrot, 1977). Simply stated, the problem is
that the complex form of the passive shoreline does not simplify as smaller and smaller segments
are examined, and thus the "tangible" shoreline always remains just out of reach of the investigator
who would measure it. In order to skirt this problem, the present study deals not with the linear
retreat of the shoreline, but rather with the areas that are lost as the shoreline recedes. Two separate
approaches are used, each having special advantages and disadvantages'. In the first, which treats
entire coastal communities, the distribution of the area of the community with respect to its
elevation is presented in the form of "hypsometric" curves, or cumulative frequency diagrams.

While this is a powerful tool for the analysis of such geographical units as a whole, the results give
no information about the change at a specific point within that unit. The second approach makes use
of color-coded maps of areas that are of special concern for the management of ports and harbors.
For this purpose the harbors of Hyannis, Westport and Gloucester were chosen. While it is
difficult to quantify the effects of small changes from these color-coded maps, the areas that will
(and will not) be affected are displayed clearly.
Hypsornetry 3
As a tool for calculating the retre.-t of coastal upland resulting from relative sea-level rise,
hypsometry has been discussed by Giese ez al. (1985). Unlike previous work, however, the
present study makes use of digital elevation data that permits the application of the hypsometric
method to large areas. A separate hypsometric curve was calculated for each of 72 Massachusetts
coastal communities.
The initial data for the upland hypsometric calculations were obtained from the U.S.
Geological Survey's (USGS) National Cartographic Information Center (NCIC). They consist of
two separate types of digital information, both of which are stored on magnetic tapes. The first type
is elevation data that consists of land surface elevations to the nearest meter arranged in
south-to-north profiles for entire one-degree latitude by one-degree longitude areas. The data points
within the profiles, as well as the profiles themselves, are separated by intervals of three arc
seconds, which is equivalent to a distance of about 92 m in a north-south direction and about 69 m
in an east-west direction at a latitude of 42 degrees (the approximate mid-point of the study area).
The second type of digital data consists of land use and land cover codes arranged in
west-to-east rows aligned along a Universal Transverse Mercator (UTM) grid and covering entire
one-degree latitude by two-degree longitude areas. The UTM coordinate system is rotated slightly
counterclockwise with respect to the geographic latitude-longitude coordinate system. The land use
and land cover code data points, and the rows containing them are separated by intervals of 200 m.
The land use codes include the U.S. Bureau of the Census designation for each 200 m square;
these data permit the assignment of each square to a specific town or city. The land cover
classification codes are sufficient to permit exclusion of wetland and inland water areas.
A large part of the effort for this study consisted of the programming required to combine the
raw digital data described above to produce a single data set consisting of elevation, census code
and land-cover code for each 3-second box within one-degree blocks. A description of the
programs and their use is included in Appendix B.
During the study, the accuracy of the census and land-cover codes was checked by reference
to the appropriate U.S.G.S. 7.5-minute series topographic maps, as well as by comparing the total
calculated upland area of individual communities with the known value of their total land area. No
problems were encountered with either type of code. Unfortunately, the same was not true of the
elevation data. Initial tests of these dam were performed by comparing profiles derived ftom the
digital data to profiles based on the 7.5-minute series maps. The results of these tests generally

3 4
Hypsometric tables and graphs are included in Appendix A of the
complete report. Appendix B covers the data base processing
methodologies.

were satisfactory, particularly considering the fact that the entire elevation data set for each
--ommunity was to be combined. However, when the cumulative distributions of elevation data
were completed, it was evident that the USGS data were biased toward maxima in the vicinity of 3,
15, 30, 45 and higher multiples of 15 m. A program was written to smooth the distributions by
redistributing the excessive values linearly to the depleted elevation categories between the maxima.
A description of this procedure is included in Appendix B. The hypsometric curve thus calculated'
for one community (Bamstable@ then was compared to the curve derived by a graphical method,
and found to be acceptable. Nevertheless, it must be noted that cumulative hypsometric data
presented in this report are less accurate than those that could be obtained using unbiased elevation
data.
4
Color-coded Ma 12S
. The three maps that accompany this report were prepared to illustrate the effect upon three
harbors of the relative sea-level ri se predicted by four different scenarios for the year 2100. The
harbors of Hyannis,. Westport and Gloucester were chosen for this purpose because of their
contrasting geological settings and because of their distribution along the Massachusetts coast.
rn
These maps were generated using data derived from the digitization of selected portions of
the 7.5-minute series topographic maps for the three harbors. These maps have a contour interval
of 10 feet, which is too great to resolve the flooding that was to be shown. Therefore, a surface
was modelled to fit the digitized contours using a modified form of existing software. The levels of
flooding characterizing the four scenarios then were applied to this modelled surface. Using
color-plotting software and equipment, the flooded areas were displayed on color-coded maps. A
detailed description of the methodology employed is included in Appendix B.
The four sea-level rise scenarios illustrated on the maps were presented by Hoffman(1984),
and produce flooding of 1.8, 4.7, 7.1 and 11.3 feet by the year 2100. These values were added to
the NGVD elevations of the mean high water shorelines shown on the 7.5 minute series maps. The
shoreline elevations were assumed equal to the half-tidal range at each particular harbor plus 0.5 ft
to account for relative sea-level rise since 1929, the date of the NGVD datum. Local variations
between NGVD and mean sea level were ignored, although these data are available.
Two important differences between the hypsometric calculations and the color-coded maps
should be noted. First, while the hyposometric calculations refer only to coastal uplands and
wetland areas are entirely excluded, the color maps use as their basic reference level the present
mean high water -shoreline that, in many areas, borders on coastal wetlands. Therefore, the areas
shown as being flooded according to the lowest rise scenario include wedand areas, many of which
are salt marshes. The second difference, discussed in more detail below, is that the maps include a
consideration of the ground water table rise that accompanies a rising relative sea level. This effect
is excluded from the calculations based on hypsometry.

4Color-coded-.maps are included in the complete report.

Upland 100 yr.
flood

MHWS
V/ 'VI
Marsh MHW.

CIO, MSL

0
Sand
Flats 7 MLW
'*@Up @an,

Figure I Schematic of datum planes selected for sea-level rise scenarios.

RESLLTS AM DISCUSSION
The hypsometric curves for each community, together with tables iving the cumulative
91 IZ,
distribution of upland area with respect to elevation for each, are presented in Appendix A. The first
area value presented in each table and graph (that for 3 m ) represents the upland area that ties
between 2.5 and 3.5 m. This interval was chosen because at lower elevations it is impossible to
distinguish between upland and wetland in the source data, and does not imply that there is no
upland below 2.5 m. in the community. The assumption is made throughout this study that the
areal frequency of upland below 2.5 m. Iis equal to that at 3 m. No assumption is made, however,
about the elevations of the wetland/upland boundaries within the community, other than that these
boundaries, whatever their elevations, rise at the same rate as relative sea level (figure 1). It also
should be noted that the data terminate at an elevation of 60 m., even w hen higher land exists
within a community, in order to limit the size of the figures.,
There is a sniking variation between communities in the shape of their hypsornetric curves,
reflecting variation in the geological processes that formedthern. For example, communities on
glacial outwash plai ns, such as Yarmouth, have curves with flatter slopes at low,elevations as
compared to those, such as Brewster, that lie on glacial moraines. Certain well-known local
topographic features, such as the "Wellfleet Plains", also show up clearly on the figures.,
Making use of these hypsometric data, calculations have been made of the upland areas that
each community would lose given particular changes in relative sea level. The results of these
calculations are presented in Table 1. The first column in Table 1 lists the names of the coastal
communities of Massachusetts, and the second column gives the upland area, in acres, of each
community. The third column lists the percentage of upland area - and the fourth column the actual
area measured in acres - that each community looses in response to a relative sea-level rise of 0.01
ft. Q mm), considered here to be the historical mean annual rate of rise (Aubrey and Emery, 1983).
The following three pairs of columns give the amount of retreat, first in percent of total upland area
and then in acres, that will occur between 1980 and 2025 given three different sea-level rise
scenarios. The first scenario, case 1, calls for a continuation of the historical mean annual relative
sea-level rise rate of 0.01 ft/yr, giving a total rise of 0.45 ft over the 45 year period. Case 2
assumes that global sea level will rise 0.86 ft over the 45 year period (as given by Hoffman's
"mid-range low" scenario) and that the local coastal subsidence rate will remain at 0.0062 ft/yr,
giving a total relative rise of 1. 14 ft by 2025. Case 3 is based on the same assumption about local
subsidence, but uses Hoffman's "mid-range high" global sea-level rise estimate of 1.29 ft by 2025,
yielding a total relative rise of 1.57 ft.
The total Massachusetts upland loss at the historical relative sea-level rise rate is 65.4 acres
per year. Averaged among the 72 communities, this works out to be 0.9 acres per year per
community. However, the variation between communities is great, covering two orders of
magnitude: Nantucket loses 6.1 acres per year, while Winthrop loses only 0.06 acres. After
Nantucket, other communities having large annual losses are: Wareham, 4.7 acres; Falmouth, 3.8

TABLE 1

CALCULATED UPLAND RETREAT
(Areas are in acres, % represents percent of upland submerged)

HISTORICAL TOTAL RETREAT: 1980-2025-
UPLAND ANNUAL RETREAT Case I Case 2 Case 3
AREA 0.01 ft/yr 0.45 ft 1.14 ft 1.57 ft
RISE RISE RISE RISE
TOWN
NAME L&CRES) AREA AREA .5 AREA Zc AREA

ACUSHNET 11520 0.002 0.22 0.09 9.8 0.22 25.0 0.30 34.4
AMESBURY 8052 0.002 0.13 0.07 5.8. 0.18 14.7, 0.25 20.2
BARNSTABLE 30709 0.012 3.72 0.54 167.2 1.38 423.6 1.90 583.4
BERKLEY 9582 0.005 0,43 0.20 19.4 0.51 49.2 0.71 67.7
BEVERLY 9748 0.003 0.28 0.13 12.7 0.33 32.2 0.46 44.4
BOSTON 24264 0.009 2.16 0.40 97.2 1.01 246.2 1.40 339.0
BOURNE 23935 0.006 1.53 0.29 68.9 0.73 174.6 1.00 240.5
BREWSTER 14110 0.005 0.72 013 32.4 0.58 82.0 0.80 113.0
CHATHAM 5250 0.020 1.0.4 0.89 46.8 2.26 118.5 3.11 163.2
CHELSEA 1217 0.010 0.12 0.44 5.4 1.12 13.6 1.54 18.7
CH[LMARK 7196 0.007 0.50 0.32 22.7 0.80 57.4 1.10 79.1
COHASSET 3505 0.003 0.11- 0.14 4.7 0.34 12.0 0.47 16.5
DANVERS 7866 0.003 0.25 0.14 11.3 0.36 28.7 0.50 39.5
DARTMOUTH 34785 0.006 2.05 0.27 92.4 0.67 234.0 0.93 322.2
DENNIS 10622 0.024 2.51 1.06 112.8 2.69 285* 8 3.71 393.6
DIGHTON 13208 0.006 0.77 0.26 34.6 0.66 87.3 0.91 120.3
DUXBURY 12725 0.002 0.25 0.09 11.5 0.23 29.0 0.31 40.0
EASTHAM 6628 0.014 0.91 0.62 41.2 1.57 104.3 2.17 143.6
EDGARTOWN 9964 0.025 2.44 1.10 109.9 2.79 278.3 3.85 383.3
ESSEX 6227 0.003 0.22 0.16 9.8 0.40 24.8 0.55 34.2
EVERETT 1696 0.008 0.14 0.37 6.3 0.93 15.9 1.29 21.8
FAIRHAVEN 6765 0.020- 1.35 0.90 60.9 2.28 154.2 3.14 212.4
FALL RIVER 20708 0.001 0.19 0.04 8A 0.10 21.2 0.14 29.3
FALMOUTH 24340 0.016 3.82 0.71 172.0 1.79 435.6 2.46 600.0
FREETO)hN 19962 0.002 0.34 0.08 15.2 0.19 38.5 0.27 53.0
GAYHEAD 1933 0.012 0.24 0.55 10.7 1.40 27.1 1.93 37.3
GLOUCESTER 15009 0.003 0.48 0.14 21.6 0.36 54.8 0.50 75A
GOSNOLD 4327 0.013 0.58 0.60 26.1 1.53 66.1 2.10 91.0
HARWICH 11825 0.016 1.92 0.73 86.2 1.85 219.4 2.54 300.8
HINGHAM 8772 0.002 0.17 0.09 7.5 0.22 19.0 0.30 26.2
HULL 624 0.026 0.16 1.18 7.4 3.00 18.7 4.13 25.8
IPSWICH 14516 0.006 0.83 0.26 37.2 0.65 94.3 0.89 129.9
KINGSTON 11415 0.003 0.35 0.14 15.9 0.35 40.3 0.49 55.6
LYNN 6336 0.004 0.26 0.18 11.7 0.47 29.6 0.64 40.8
MANCHESTER 4793 0.002 0.12 0.11 5.2 0.27 13.1 0.38 18.1
MARBLEHEAD 2353 0.007 0.16 0.30 7.1 0.76 18.0 1.05 24.8
MARION 6883 0.031 2.13 1.39 96.0 3.53 243.2 4.87 335.0,
MARSHFIELD 14332 0.004 0.60 0.19 27.1 OA8 68.6 0.66 94.5
MASHPEE 13386 0.010 1.35 0.45 60.8 1.15 154.1 1.59 212.3
MATTAPOISETT 5647 0.012 0.69 0.55 3.1.3 1.40 79.2 1.93 109.0

TABLE I (conti'nued)

CALCULATED UPLAND RETREAT
(Areas are in acres, % represents percent of upland submerged)

HISTORICAL TOTAL RETREAT: 1980-2025
UPLAND ANNUAL RETREAT Case I Case 2 Case 3
AREA 0.01 ft/yr 0-.45 ft 1. 14 ft 1.57 ft
RISE RISE RISE RISE
TOWN
N'A M E (ACRLS) fg AREA _01g AREA Lo AREA _T2 AREA

NAHANT 461 0,111 0,01 0.83 3,9 2.11 9.8 2.90 13.5
NANTUCKET 232@5 0.027 6.15 1.19 277.0 3.02 701.6 4.16 966.3
NEW BEDFORD 10410 0.006 0.60 0.26 27.2 0.66 68.8 0.91 94.8
NEWBURY 9442 0.009 0.81 0.39 36.5 0.98 92.6 1.35 127.5
NEWBURYPORT 4705 0.005 0.22 0.21 9.7 0.52 24.7 0.72 34.0
OAK BLUFFS 4288 0.014 0.59 0.62 26.@ 1.56 67.0 2.15 92.2
ORLEANS 6211 0.017 1.07 0.78 48.4 1.97 122.7 2.72 168.7
PLYMOUTH 59264 0.001 0.77 0.06 34.7 0.15 87.8 0.20 121.0
PROVINCETOWN 1173 0.018 0.21 0.81 9.5 2.05 24.1 2.83 33.1
QUINCY 8062 0.010 0.94 0.47 37.7 1.19 95.6 1.63 131.6
REHOBOTH 2770, 0*003 0*71 0*13 34*9 0,32 18A 0*44 121,8
REVERE 2595 0.009 0.24 0.41 10.7 1.05 27.2 1.44 37.5
ROCKPORT 3715. 0.004 0.14 0.18 6.5 0.44 16.5 0.61 22.7
ROWLEY 9184 0.002 0.17 0.08 7.4 0.21 18.8 0.28 26.0
SALEM 3956 0.007 0.29 0.33 13.0 0.83 32.9 1.15 45.3
SALISBURY 6167 0.013 0.82 0.60 36.9 1.52 93.5 2.09 128.8
SANDWICH1 24469 0.005 1.20 0.22 54.0 0.56 136.7 0.77 188.21
SAUGUS 5859 0.002 0.13 0.10 6.1 0.26 15.4 0.36 21.2
SCITUATE 8745 0.004 0.38 0.20 17.3 0.50 43.9 0.69 60.4
SEEKONK 11433 0.001 0.09 0.04 4.1 0.09 10.4 0.13 14.4
1.25 515 1.73 72.3
SOMERSET 4184 0.011 0.46 0.50 20.7
SWAMPSCOTT 1931 0.006 0.11 0.25 4.8 0.63 12.1 0.86 16.7
SWANSEA 12599 0.007 0.86 0.31 38.6 0.78 97.7 1.07 134.5
TISBURY 3539 0.012 0.41 0.52 18.5 1.32 46.8 1.82 64.5
TRURO 10734 0.006 0.61 0.26 27.5 0.65 69.7 0.89 96.1
WAREHAM 19822 0.024 4.70 1.07 211.4 2.70 535.6 3.72 737.6
WELLFLEET 9127 0.011 1.01 0.50 45.6 1.27 115.5 1.74 159.1
WESTPORT 27340 0.004 1.12 0.18 50.4 0.47 127.8 0.64 176.0
WEST TISBURY 14466 0.006 0.90 0.28 40.4 0.71 102.2 0.97 140.8
WEYMOLTIH 9944 Or.00r- (Y. 14 0.06 6.3 0.16 15.9 0.22 21.9
WINTHROP 300 0.021 0.06 0.94 2.8 2.37 7.1 3.27 9.8
YARMOUTH 1= 0.026 1.15 144.6 2.92 10@4' 4.02 5-Q4_7

TOTALS 804246 65.4 2945. 7459.1 10273.

The following coastal towns look less than 0.001 % of their total upland area annually as the result the historical mean
sea-level rise rate of 0.01. ft/yr, and therefore were omitted from this table: Braintree, Hanover, Milton, Norwell,
Peabody and Pembroke.

acres; @arnstable, 3.7 acres; and Yarmouth, 3.2 acres. In terms of annual percentage of total upland
lost per year, the communities most affected are: Marion, which loses 0.031 % per year, followed
by Nantucket which looses 0.027 % per year, and Hull and Yarmouth, which loose 0.026% per
year.
Looking forward to the year 2025, if the historical rate of relative sea-level rise were to
remain unchanged (case 1), the total Massachusetts upland loss would be 2,945 acres. A relative
sea-level rise of 1. 14 ft, as projected in case 2, would be accompanied by an upland loss of 7,459
acres, and a rise of 1.57 ft (case 3) would cost the commonwealth 10,273 acres of upland.
When considering these figures, it is important to realize that they do not include the upland
losses that would result from the response of ground water levels to sea-level rise. In those
communities where bedrock is absent and the terrain consists of unconsolidated sediments, the
water table level over. geological time periods is controlled by relative sea level. As sea level rises,
the water table level rises with it, increasing the size of existing streams, ponds and bogs, and
creating new ones. This effect has not been included in the hypsometric an alysis discussed above,
although it was taken into account in the construction of the coor-coded maps.
The reader also should bear in mind that the calculated upiand retreat rates are based on the
assumption that the coastal uplands have a natural form and are not protected by engineering
structures. Particularly in urban coastal areas where seawalls, riprap and fill are prevalent, the
actual losses will be less than those predicted here. As the color-coded maps indicate, however,
when large values of sea-level rise are considered, these structures are overwhelmed.
It is of interest that the presently existing rate of upland retreat due to the passive effects of
relative sea-level rise is much greater than the upland retreat rate due to active wave-produced
erosion. This may be illustrated by a consideration of the Cape Cod coast, which is well-known as
a region of rapid erosion. While detailed estimates for cliff retreat do not exist for the entire region,
the rate of erosion of the outer coast is well-known (e.g., Zeigler et al., 1964), and reasonable
estim,ates can be made for the remaining and more slowly retreating cliff areas. Using such existing
information and reasonable estimates, the annual upland loss experienced by Cape Cod as the
result of active wave-produced erosion is about 9 acres per year. On the other hand, the annual
loss due to the passive effects of relative sea-level rise, calculated from the figures for each Cape
Cod town listed in Table 1, is'about 24 acres per year. Thus it is seen that even considering a
region of rapid erosion, and excluding the effects ground water table rise, passive retreat accounts
for 73% of coastal upland loss under present conditions.
5
Figures 2, 3, and 4 present the color-coded maps depicting the submergence patterns of
Hyannis, Gloucester, and Westport harbors that would accompany each of the four Hoffman
(1984) sea-level rise scenarios for the year 2100. The maps show in red the land areas that would
be lost given the low scenario rise of 1.8 ft, in yellow the submerged areas given themid-range low
scenario rise of 4.7 ft, and in green and blue the areas submerged by the mid-range high scenario
rise of 7.1 ft and the high range scenario rise of 11.3 ft respectively. The low scenario changes are

5 10 1
included in the complete report. See footuote

extensive only in wetiand areas, such as the salt marshes northwest of Gloucester Harbor, the sand
spit southwest of Hyannis Harbor, and fringing marshes in Westport Harbor. While the upland lost
given this scenario is not extensive, the increased potential for storm wave and flooding damage
should be of concem.
The submergence that would accompany the other scenarios is extensive and would impact
severely operations of harbor facilities. In addition, the maps show locally significant flooding of
inland areas for these scenarios resulting from elevated ground water levels. As has been discussed
above, it should be kept in mind that the levels used in applying these scenarios do not include the
effects of coastal subsidence, and that for the lower rise rates the increases would be significant
were they to be included.

CONCLUSIONS
Major conclusions of the present study are:

1 . Relative sea-level rise is the major process responsible for upland loss in Massachusetts.
Neglecting coastal erosion and fresh water table changes, Massachusetts presently looses about
65 acres of upland each year due to passive submergence.
2. The rate of upland loss due to passive submergence varies widely from town to town, and
depends upon the geology of the region in which the town lies.
3. The hypsometric curves of the towns provide important basic information that permits the
calculation of the upland areas which those towns will lose to passive submergence as the
result of any given increase in relative sea-level.
4. The total land loss by the year 2025 has been calculated for several'relative sea-level rise
scenarios. At the present rate of rise, Massachusetts will have lost about 3,000 acres of upland
between 1980 and 2025. This is the same upland loss that occurred between 1935 and 1980,
an equal length of time. For a rise of 1.14 ft, about 7,500 acres would be lost; and for a rise of
1.57 ft, the maximum likely, over 10,000 acres would be lost. Given a nominal value of
ocean-front property of $1,000,000 per acre, the economic impact of this retreat is substantial.
5. Color-coded maps are a useful device for depicting the specific areas that will be submerged as
the result of specified increases in relative sea level. These maps could be developed for each
coastal town in the future, to provide guidance for land use, public works, and conservation
decisions.
6. These data can be used immediately to help provide a rational basis for local response to global
climate warming. Dam from this report, although representing hypothetical scenarios, remove
the quantification of the impacts of passive retreat from the realm of speculation, placing them
on a firmer basis. Although enactment of legislation and major revision of regulations may-be
premature, local communities must increase their awareness of these impacts, and begin to
incorporate these data in planning, design, and conservation issues.

7. Although the present study has shown that passive retreat is an important element of the
shoreline response to anticipated global climate change, this inundation is certainly not the sole
impact. Future research is mandated for other impacts on the coast of Massachusetts, including
but not limited to:

- Effects of relative sea-level rise on groundwater resources.
- Effects of relative sea-level rise on marshes and other biotopes.
- Possible global climate change impact on storm climatology of Massachusetts wate rs.

ACKNOWLEDGEMENTS
T"his study has benefited from the contributions of many of our colleagues at the Woods Hole
Oceanographic Institution. We are grateful for the assistance of Chris Pelloni of the Woods ficle
office of the U.S. Geological Survey who provided coordinate translation of the numerical data.
Jeff Benoit, of the Massachusetts Office of Coastal Zone Management, provided the foresight and
management ability which made the study a reality.

REFERENCES
Aubrey, D.G. and K.O. Emery, 1983. Eigenanalysis of recent United States sea levels.
Continental Shelf Research, v. 2, n. 1, p. 21-33.
Bird, E.C.F., 1976. Shoreline changes during the past century. In Proc. 23rd International -
Geographical Conacss, Moscow, 54 p.
Braatz, B.V. and D.G. Aubrey, 1987. Recent relative sea-level change in eastern North America.
Journal QL Sediment= Petrology, in press.
Giese, G.S., H. Bokuniewicz, G. Zarillo, M.Zimmerman, J. Hennessy, G. Smith and S.
Tangren, 1984. Hypsometry as a tool for calculating coastal submergence rates. Proc. 4th
5=. g_n Coastal Md Ocean Manag=ent, v. 2, p. 1971-1978.
Hoffman, J., D. Keyes and J. Titus, 1983. Projecting future sea-level rise: methodology,
estimates to the ym 2100 and research needs. U.S. Envirommental Protection Agency, EPA
230-09-007, Washington, DC, 121 p.
Mandelbrot, B.B., 1977. Fractala: Form. Chance. and DiMmsion. W.H. Freeman and Co.,
San Francisco, 365 p.
Zeigler, J.M., S.D. Tuttle, G.S. Giese and H.J. Tasha, 1964. Residence time of sand composing
the beaches and bars of Outer Cape Cod. B=. a.C=.Qa Coastal Engineering. p. 403-416.

DATE DUE

GAYLORDINo. 2333 1

1101111111111111
366681
4107