ABSTRACT 
Title of Thesis: Island Land Loss in the Chesapeake 
Bay: A Quantitative and Process 
Analysis 
Name of degree candidate: Rachel Donham Wray 
Degree and Year: Master of Science, 1992 
Thesis directed by: Stephen P. Leatherman, Professor, 
Department of Geography 
The rates and processes of land loss were studied 
for seven islands in the Chesapeake Bay: Barren, 
Bloodsworth, Hooper, James, Poplar, Smith and South Marsh 
Islands. Rates and patterns of land loss were quantified 
for the years 1848 to 1987 with the Metric Mapping 
technique which utilizes digitized data from historical 
maps and vertical aerial photographs. Processes of land 
loss were determined through field surveys and correlated 
with environmental factors. 
Two distinct island types were identified which 
exhibited different, long-term patterns of land loss. 
Small, upland islands, termed the Northern Group, showed 
rapid land loss along the main stem of the Bay primarily 
due to wave action driven by the predominant westerly 
winds. Land loss appeared to accelerate during periods 
of high storm frequency. The long-term averaged land 
loss rate for Northern Group islands is 1.9 ha/yr. The 
averaged erosion rate on the western side of the islands 
is 4.9 m/yr, compared to 0.68 m/yr on the eastern side of 
the islands. 
In contrast, the large, marshy islands of the 
Southern Group experienced uniform marsh edge erosion and 
interior marsh degradation. The Southern Group islands 
lost land at an averaged rate of 5. 6 ha/yr, with an 
averaged rate of marsh edge erosion of 1.2 m/yr. Land 
loss appeared to be weakly correlated to storm frequency. 
Interior marsh loss was not quantified for this study, 
however, so this study provides an underestimation of 
total land loss of coastal wetlands. 
ISLAND LAND LOSS IN THE CHESAPEAKE BAY: 
A QUANTITATIVE AND PROCESS ANALYSIS 
by 
Rachel Donham Wray 
Thesis submitted to the Faculty of the Graduate School 
of the University of Maryland in partial fulfillment 
of the requirements for the degree of 
Master of Science 
1992 
C. ' 
MQ 
D..t I Q; } , 
Advisory Committee: 
Dr. Stephen P. Leatherman, Chairman/Advisor 
Dr. Robert J. Nicholls 
Dr. Stephen D. Prince 
Dr. Edward C. Pendleton 
t'\o r \ 
't 
..) , 
ACKNOWLEDGEMENTS 
This thesis would not have been possible without the 
generous support of the U.S. Fish and Wildlife Service. 
In addition, there are many people who I would like to 
acknowledge due to their assistance, advice and time, 
including C. Wray, B. and P. Donham, R. Nicholls, L. 
Downs, D. Forsell, S. Funderburk, J. Gill, S. Leatherman, 
and E. Pendleton, to name only a few. 
TABLE OF CONTENTS 
Page 
List of Tables. . . ? . ? . . ? . . . . . . . . ? v 
List of Figures ??... vi 
List of Photograph Plates. X 
CHAPTER 1: INTRODUCTION. . ? . . . . . 1 
a. Introduction ?.?.?..???...?? 1 
b. Study Objectives ....????..??. 5 
CHAPTER 2 : STUDY AREA . ? ? ? ? ? ? ? ? ? . . . 7 
a. Environme ntal Characteri stics ..??... 7 
1. Geomorphology . . . . ? . 7 
2. Climate. . . . . ? . . . 9 
3. Storms. . . . . . . . . . . . . . . . 12 
4 Waves ? . . ? ? ? . . . ? ? ? . ? ? ? 14 
5. Vegetative Communities .?..... 16 
b. Geologic History of Chesapeake Bay .... 16 
c. Sea-Level Rise ?.....?...?... 21 
1. Marsh Response to Sea-Level Rise ... 23 
2. Future Sea-Level Rise .???...? 24 
d. Islands in the Study Area ..??.... 29 
1. Introduction ???....?.... 29 
2. Barren Island .........?.. 32 
3. Bloodsworth Island ......... 33 
4. Hooper Island .........?.. 35 
5. James Island ....?.?..... 39 
6. Poplar Island ......??...? 42 
7. Smith Island. ? ? ? . . ? . 43 
8. South Marsh Island. . . . . 48 
CHAPTER 3: METHODS . . . ? ? ? . . ? ? ? . . . . ? 50 
a. Introduction ...??.....?.?.. 50 
b. Historical Shoreline Mapping ..??... 51 
1. Data Selection .?.?..???.. 51 
2. Data Preparation ?.....?.?. 54 
3. Digitizing ?.?...??.?... 59 
4. Data Analysis ?....?...... 60 
5. Accuracy Assessment ..?...?.. 63 
c. Field Surveys .??...??...?.. 66 
d. Data Analysis .?...?....??.. 69 
1. Sea-Level Change Analysis ?... 69 
2. Aerial Analysis ..?.?...... 72 
3. Land Loss v. Sea-Level Change ...? 74 
e. Forecast Modeling ? ? . . . . ? ? . . . . 7 6 
1. Introduction. . ..?.?? 76 
2. Bruun Rule. . . . . . ...?. 76 
iii 
3. Inundation Model ????? . . 77 
4. Historic Trends Analysis. 78 
CHAPTER 4: RESULTS ? ? ? ? ? ? . . ? ? ? ? ? ? ? ? ? 80 
a. Historical Shoreline Mapping ??....? 80 
1. Introduction ...?...??... 80 
2. Northern Group ...?...?... 88 
A. Patterns of Land Loss ??.?? 88 
B. Rates of Land LOSS ??????? 95 
C. Rates of Erosion ? ? . ? ? . . ? 95 
D. Sediment Analysis .????.. 99 
3. Southern Group ?.???????.? 103 
A. Patterns of Land Loss ??.?? 103 
B. Rates of Land Loss ????.. 107 
c. Rates of Erosion ?..?... 111 
D. Sediment Analysis ?...? 111 
b. Shoreline Response Modeling ?..??.? 116 
1. Inundation Model . ? ? ? ? ? ? ? ? ? 116 
A. Northern Group ???????? 116 
B. Southern Group . ? ? ? ? ? ? . 119 
c. Hooper Island .???????. 119 
2. Historic Trends Analysis ????.? 123 
A. Northern Group .??????? 123 
B. Southern Group . ? ? ? ? . 126 
CHAPTER 5: DISCUSSION .?  ..  ..  ..  ..  . . . . . . . . 131 a. Introduction . . 
b. Northern Group . . . . . . .  . . ..  ... .   ..  . 131 . 131 
1. Wave characteristics . . 
2. Storm frequency.  ..  . . . . 
. . . . . . . . . . 131 . 132 
3. Sediment type . . 
4. Tidal range . . . ..  . . ..  ..  . .. . 135 .  . 138 
c. Southern Group. . . . . . . . . . . . . 141 
1. Marsh fringe erosion . . . . . . 
2. Interior marsh loss . . . . . . . . . . . . 
. . . 142 
A. Sediment source.  . . . . . ..  . 
147 
148 
B. Tidal Range . . . . . . . . . . 149 c. Marsh t.y p.e . . 
. . 150 
d. Hooper Island ?  ? . . . . . . ? . 152 
CHAPTER 6: CONCLUSIONS. . . . . . . . . . . ? 155 
Appendix A: Historical Storm Data ?. 166 
Appendix B: Historical Shoreline Data 169 
References .. . . . . . . . . . . . . . . . 172 
iv 
LIST OF TABLES 
Table Caption 
2.1 Fourty-year average wind data 13 
from Baltimore, Maryland. 
2.2 Major vegetation communities. 17 
4.1 Historic Island Land Loss in 89 
the Chesapeake Bay: Northern 
Group. 
4.2 Historic Island Land Loss in 90 
the Chesapeake Bay: Southern 
Group. 
4.3 Elevational Characteristics 94 
of the Northern Group Islands. 
4.4 Fetch Analysis for 7 Islands 96 
in the Chesapeake Bay. 
4.5 Erosion Rates for the Northern 97 
Group. 
4.6 Erosion Rates for the Southern 113 
Group. 
4.7 Historic and Future Rates of 124 
Land Loss for the Northern 
Group 
4.8 Historic and Future Rates of 128 
Land Loss for the Southern 
Group. 
4.9 Future Projections of Island 129 
Size for the Southern Group and 
Hooper Island. 
5.1 A Sample of Marsh Edge Erosion 143 
Rates from Several Geographic 
Locations. 
V 
LIST OF FIGURES 
Figure Caption 
1.1 Map of the Chesapeake Bay 6 
with the location of seven 
islands. 
2.1 Vicinity map of the Chesapeake 8 
Bay. 
2.2 Wave energy distribution along 15 
the northern Chesapeake Bay 
shoreline (from Wang et al., 
1982). 
2.3 Location of the three major 18 
paleochannels in the Chesapeake 
Bay (from Coleman et al., 1990). 
2.4 Global sea-level rise, 1900- 28 
2100, for Policy scenario 
Business-as-usual (adapted 
from IPCC, 1990). 
2.5 Enlargement of 1848 T-sheet of 36 
Bloodsworth Island, showing 
the location of an orchard and 
several diked areas (from GEO-
RECON, 1980). 
2.6 Enlargement of 1848 T-sheet of 40 
James Island, showing the 
island attached to the mainland. 
2.7 Enlargement of 1901 T-sheet of 41 
James Island, showing the island 
separated from the mainland. 
2.8 Schematic drawing of Poplar 45 
Island habitat enhancement 
proposal (from U.S. Fish and 
Wildlife Service, undated draft). 
2.9 Map of Smith Island with town 47 
locations. 
3.1 Flow chart of Metric Mapping 52 
Procedure. 
3.2 Example of primary control 55 
points. 
vi 
3.3 Example of secondary control 57 
points (circled). 
3.4 Example of transects along 62 
western side of Barren Island . 
3.5 Baltimore, Maryland, tide gauge 70 
r ecord, with five-year running 
mean. 
3.6 A comparison between Baltimore 71 
tide record and four other tide 
gauge stations in the Chesapeake 
Bay, showing station locations. 
3.7 Regression analysis between the 73 
Baltimore, Maryland, and New 
York, New York, tide gauge 
stations. 
3.8 Baltimore, Maryland, tide gauge 75 
record, with five-year running 
mean. Vertical lines represent 
dates of available shoreline 
data from maps and photographs. 
3.9 Conceptual diagram of the 78 
inundation model (from 
Leatherman 1991). 
4.1 Historical shoreline change, 81 
Barren Island. 
4.2 Historical shoreline change, 82 
James Island. 
4.3 Historical shoreline change, 83 
Poplar Island. 
4.4 Historical shoreline change, 84 
Bloodsworth Island. 
4.5 Historical shoreline change, 85 
Smith Island. 
4.6 Historical shoreline change, 86 
South Marsh Island. 
4.7 Historical shoreline change, 87 
Hooper Island. 
vii 
4.8 Island land loss in the 91 
Chesapeake Bay (Northern 
Group). 
4.9 Island land loss in the 92 
Chesape ake Bay (Southern 
Group ) . 
4.10 A comparison between the rate 98 
of sea-level rise and the rate 
of land loss for the Northern 
Group. 
4.11 Sample locations for Barren 100 
Island - 5/17/91. 
4.12 Sample locations for James 101 
Island - 5/15/91. 
4.13 Sample locations for Poplar 102 
Island - 5/1/91. 
4.14 Historical change in the 108 
percent of total open water 
in four quadrants of Smith 
Island: Terrapin Sand Point, 
Kedges Straits, Great Fox Island, 
and Ewell (from Davison 1990). 
4.15 A comparison between the rate of 110 
sea-level rise and the rate of 
land loss for the Southern Group. 
4.16 Location of transects for 112 
determining erosion rates on 
Bloodsworth Island. 
4.17 Location of transects for 114 
determining erosion rates on 
Smith Island. 
4.18 Location of transects for 115 
determining erosion rates on 
South marsh Island. 
4.19 Geologic section through 117 
northern Bloodsworth Island 
showing development of marsh 
over clay layer (from GEO-
RECON 1980). 
viii 
4.20 Geologic section of Lower 121 
Hooper Island. 
4.21 Location of survey sites for 122 
Hooper Island - 11/23/91. 
4.22 Trends of land loss. Northern 125 
Group. 
4.23 Trends of land loss. Southern 127 
Group. 
4.24 Future Projections of Southern 130 
Group Islands. 
5.1 Land loss vs. storm frequency. 134 
The Northern Group. 
5.2 Schematic diagram of erosional 140 
processes of a Northern Group 
Island 
5.3 Land loss vs. storm frequency. 145 
The Southern Group. 
5.4 Schematic diagram of erosional 146 
processes of a Southern Group 
Island 
6.1 Schematic diagram of Bodkin 161 
Island habitat enhancement 
proposal (from U.S. Fish and 
Wildlife Service, undated 
draft). 
ix 
LIST OF PHOTOGRAPHIC PLATES 
Plate Caption Page 
2.1 View of an eroding marsh edge on 10 
Bloodswor th Island . 
2. 2 View of an eroding clay bluff on 11 
Poplar Island. 
2.3 View of a pocket beach on Coaches 11 
Island (Poplar Island), with a 
marsh 11 headland 11 ? 
2.4 View of trees dying at edge of an 25 
upland area on Hooper Island, an 
example of the upland conversion 
process. 
2.5 View of a marsh peat layer over the 26 
clay layer on Lower Hooper Island. 
2.6 View of the hunting lodge on Barren 34 
Island. 
2.7 View of an eroding graveyard on 34 
Lower Hooper Island. 
2.8 View of cement encased graves on 38 
Middle Hooper Island. 
2.9 View of a flooded lawn on Middle 38 
Hooper Island. 
2.10 View of Poplar Island islets. 44 
3.1 View of a pit dug on Hooper 67 
Island with clay layer showing 
beneath a Phragmites marsh. 
4.1 Mixed upland and wetland habitat 93 
on Coaches Island (Poplar Island). 
4.2 Clay bluff and dead trees on Poplar 93 
Island which is typical of the Northern 
Group Islands. 
4.3 Sand bridge on James Island, 104 
looking south. 
X 
4.4 View of an upland ridge on Smith 104 
Island . 
4.5 View of Rhodes Poi nt on a ridge in 106 
the distance, one of the towns an 
Smith Island. 
4 .6 Interior marsh ponding on Smith 106 
Island. 
5.1 Eastern side of Barren Island. 133 
5.2 Bluff toe weathering on Poplar 137 
Island. 
5.3 Example of erosion control measures 153 
along the western shore of Smith 
Island. 
5.4 View of an upland ridge on Hooper 153 
Island, surrounded by the encroaching 
marsh . 
xi 
CHAPTER 1: INTRODUCTION 
Introduction 
During the last two million years, coastal areas 
worldwide have evolved dramatically with the oscil lations 
of sea level due to Ice Ages, altering the physiography 
as well as the climate of coastal regions. Climate 
change during the last ten thousand years has caused the 
transformation of the Susquehanna River valley to form 
the Chesapeake Bay estuary that exists today ( Coleman and 
Mixon, 1988). Widespread saltmarsh development 
throughout the northeast United States was associated 
with decreased rates of sea-level rise around 4,000 years 
ago (Redfield and Rubin, 1962; Rampino and Sanders, 1981; 
Orson, et al., 1987). Marshes in the Chesapeake Bay also 
began to develop around this time. The Bay has continued 
to evolve geomorphologically during the last few 
centuries, through shore erosion, marsh degradation and 
accretion, and gradual submergence of low-lying upland 
areas. Erosion and marsh loss are collectively called 
land loss. 
Coastal erosion is the most obvious means of land 
loss. Erosion results in a loss of valuable shorefront 
land and wetland habitat, damage to buildings and other 
structures, diminished beach capacity at recreational 
1 
areas, and adverse impacts to cultural and historic 
resource s (Leat herman, 1984). In the past century, it is 
esti mated that over 18,000 hectares of coa stal a reas of 
Chesapeake Bay have eroded, providing about 3.6 million 
cubic meters of sediment to the Bay each year (US Army 
Corps of Engineers, 1991). From his torical records of 
the Bay, it is clear that land loss has been occurring 
since at least the mid-19th century ( Singewald and 
Slaughter, 1949; Mowbray, 1981; Kearney and Stevenson, 
1991). Due to the record of eustatic sea-level rise 
during the last 15,000 years, however, it is clear that 
land loss has been occurring since long before the 19th 
century. 
Shore erosion has previously been shown to be an 
important process in the Bay (Singewald and Slaughter, 
1949; Wang, et al., 1982). Coastal erosion, however, has 
only recently been recognized as the major input of 
sediment into the Bay (Marcus and Kearney, 1991), 
increasing toxins and nutrient loads in the water. 
Sediment loading from increased runoff due to land 
clearing is also responsible for many problems in the 
Bay, principally subsidence. Such a discovery may be a 
first step towards focusing on land loss as a problem in 
the Bay and treating it on a Bay-wide basis. For 
example, sediment input to the Bay will be reduced by 
curbing erosion. 
2 
The extent of land loss in the Chesapeake Bay has 
been significant, and its importance and impact is 
perhaps easiest to comprehend in terms of the response of 
the islands in the Bay. The Bay islands provide 
excellent case studies of land loss because they have 
been so reduced in size that most have become 
uninhabitable; others have even been reduced to shoals. 
In addition, most have essentially unprotected 
shorelines, whereas much of the mainland has been 
protected by bulkheads and revetments. Without such 
structures, the natural processes of land loss are 
unimpeded, and can be studied more easily. In addition, 
anecdotal and historical records exist for many of the 
islands, and provide examples of relatively large island 
communities which no longer exist. This is good indirect 
evidence of the extent of land loss in terms of both 
erosion and the conversion of uplands to marsh. Many 
islands, which once provided homes and ample farm land, 
are no longer habitable and some are barely large enough 
to stand on. Today, only a few of the islands are 
inhabited, including Hooper Island, Smith Island, and 
Tangier Island. 
The human exodus from the islands can be attributed, 
in part, to three mechanisms: submergence, erosion, and 
the impact of large storms. However, the specific causes 
have never been thoroughly investigated. A combination 
3 
of factors, including the harsh island environment, 
erosion, waterlogged soils, flooding from hurricanes, and 
a more desirable lifestyle on the mainland, presumably 
provided incentives to leave. For some islands, such as 
Bloodsworth Island, the frequency of flood events due to 
submergence increased to the point where living there 
became impractical ( GEO-RECON, 1980). For other islands, 
such as Poplar Island, erosion continually encroached 
upon established communities until there was no longer 
room to continue living and farming (Meyer, 1986). There 
are many other examples of island land loss in the 
Chesapeake Bay, and some of these will be discussed in 
detail in this thesis. 
Waterfowl in the Bay are also affected by island 
land loss. For example, most black ducks rely on remote 
areas, such as uninhabited islands for breeding and 
nesting presumably because of the species' aversion to 
human disturbance. The loss of isolated islands and the 
increasing development in other areas are thought to be 
primary causes of the black duck population decline in 
the Chesapeake Bay ( Krementz, et al. , 19 91) . The 
distribution of other waterfowl species such as bald 
eagles, ospreys, herons, egrets, various duck species, 
and swans is being impacted as available space for 
breeding and nesting is becoming increasingly limited due 
to land loss (Stotts, 1985). The mainland is becoming a 
4 
less viable option for inhabitation for many wildlife 
species because of development and cultivation. As a 
result, species distribution and diversity is being 
affected by the reduction of available habitat. 
Study objectives 
The processes and rates of historic land loss in the 
Chesapeake Bay were studied for seven islands (Figure 
1.1): 
Barren Island 
Bloodsworth Island 
Hooper Island 
James Island 
Poplar Island 
Smith Island 
South Marsh Island 
The most important goal of this study was to understand 
how and why this land loss is occurring. Therefore, the 
specific objectives were to: 
(i) quantify the rates and patterns of island 
land loss; 
(ii) determine and quantify the causes of land 
loss; 
(iii) project the future evolution of these 
islands with and without accelerated sea-
level rise; 
(iv) correlate these findings with field data. 
5 
Figure 1. 1 Map of the Chesapeake Bay with the location of 
seven Islands 
1. Poplar Island 
2. James Island 
3. Barren Island 
4. Hooper Island 
5. BIOOdSWOr1h Island 
6. South Marsh Island 
7. Smith Island 
ATLANTIC OCEAN 
6 
CHAPTER 2: STUDY AREA 
Environmental Characteristics 
Geomorphology 
The Chesapeake Bay, located in the middle Atlantic 
Coastal Plain Province, is a classic coastal plain 
estuary formed by the post-Wisconsin rise in sea level 
which drowned the lower valley of the Susquehanna River 
(Ryan, 1953). The Bay is about 300 km long from the 
mouth of the Susquehanna River to the Cape Charles-Cape 
Henry entrance to the Bay (Figure 2.1). It ranges in 
width from 5 to 56 km, the widest point being in Tangier 
Sound in the southern Bay, with an average width of about 
40 km. The shoreline of the Bay is extremely irregular, 
totalling 12,900 km in length. With an average depth of 
only 8 to 10 m, the Bay is very shallow compared to its 
width. The deepest part of the Bay is the incised main 
channel of the former Susquehanna River which runs the 
entire length of the Bay, with depths over 50 m (Kehrin, 
etal., 1988). 
Much of the western shore consists of high relief, 
clay/sand cliffs and narrow sandy beaches, especially in 
Calvert County, Maryland. The eastern shore of the 
Chesapeake Bay is characterized by low elevation and a 
scarcity of sandy deposits with few exceptions. All the 
islands in the study area generally lie less than 2.5 m 
7 
DEL 'lt4~V4 
?ENINSUL.l 
~CH?SAP?AK? t. 
8AY ~ 
~ 
,""  
~ 
34?~-------...... ---------' 
Figure 2.1 Vicinity map of the Chesapeake Bay 
8 
above mean sea-level according to recent topographic 
surveys. The majority of the island shorelines are 
eroding marsh edge ( Pl ate 2 .1) and eroding silt/clay 
bluff (Plate 2.2). The bluffs are general ly between 1 
and 2 m above mean sea level. In addition, several 
small, sandy, pocket beaches have developed between 
resistant marsh headlands in some places on the island 
shores (Plate 2.3). These beaches are thin veneers of 
sand which overlay marsh peat or clay. Rosen ( 1980) 
classifies these as "impermeable beaches", which overlie 
impermeable sediments such as silt/clay. They are highly 
erodible because they have low swash filtration and low 
beach elevation. 
Climate 
The Chesapeake Bay is in the northern Temperate 
Zone, with mild winters and hot, humid summers. Average 
annual rainfall for the region is 106 cm, with the most 
rainfall occurring between June and August. During the 
winter, the Appalachian Mountains and the waters of the 
Bay have a moderating effect on the cold air from the 
northwest ( US Department of Agriculture, 19 6 6 ) . The 
predominant wind direction in the Bay on an annual basis 
is west-northwest at an average speed of 9.2 mph. The 
only exception is during September when the predominant 
9 
Plate 2.1 View of an eroding marsh edge on Bloodsworth 
Island 
10 
Plate -2.2 View of an eroding clay cliff on Poplar Island 
Plate 2. 3 View of a pocket beach on Coaches Island 
(Poplar Island), with a marsh "headland" 
11 
wind direction switches to south. Higher wind speeds are 
generally experienced during the winter months, with more 
gentle winds during the summer (Table 2. 1) (US Department 
of Commerce, 1990). The highest wind speeds are 
experienced during periodic storms such as northeasters 
which usually occur during the winter months, and 
hurricanes and tropical storms which usually occur in 
late summer. 
Storms 
Seventy-nine major storms, both tropical and 
extratropical, have occurred in the Bay vicinity between 
1871 and 1986 (Appendix A) (Neuman, et al., 1987). 
Hurricanes and tropical storms generally occur in late 
summer and early fall months, but can occur as early as 
June and into December. In the winter months, 
extratropical or "northeasterly" storms, which originate 
over land, bring the highest winds and worst weather to 
the Bay area. On an annual basis, northeasters occur 
more frequently than hurricanes or tropical storms. 
However, due to elevated water levels (storm surge) and 
high wind-driven waves, hurricanes and tropical storms 
can be highly destructive forces on coastal areas. 
Wang et al. ( 19 8 2) performed a wave hindcast for the 
Chesapeake Bay to simulate storm-wave conditions. The 
12 
.???,  
Table 2.1 Fourty-year average wind data fro? Baltl?ore, Maryland 
(fro? u.s. Dept. of co-erce, NOM, 1990) ri 
JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC 
.... Rean Speed 9.7 10.3 10.9 10.6 9.2 8.5 8.0 7.8 8.0 B.7 9.3 9.3 
w (MPH) Ii 
Prevailing lfN1I "" lfN1I WNW ? WNW Direction ? ? s "" """ WNW 
distribution of zones of "high" and "medium" wave energy 
which resulted from the model are presented in Figure 
2.2. More "high" wave energy areas are found along the 
western shore of the Bay than the eastern shore. 
Al though most of the high wave energy zones are 
located in areas known to have high erosion rates, the 
reverse is not true. There are high wave energy zones in 
places with lower erosion rates, such as Calvert County, 
Maryland along the western shore and some of the island 
shorelines along the eastern shore (Wang, et al., 1982; 
Downs, in prep.). Clearly, there are several factors 
which determine the potential erosion rate of a 
particular area. One factor alone, such as wave energy, 
cannot explain the entire process. 
Waves 
Wave conditions near the shore and the directions of 
wave energy flux are probably the most important factors 
which are needed to assess erosion potential. Wang et 
al. (1982) used a wave-hindcast numerical model to 
calculate wave statistics for the Chesapeake Bay, 
accounting for bottom friction, irregular fetch areas, 
and wave breaking. The results indicate that "annual" 
average wave climate is composed principally of waves 
whose heights are 0.15 to 0.3 m. These wave heights are 
fairly small due to limited fetch and shallow water 
14 
WAVE ~NERGY 
CISTRi3UTION 
ALONG 7HE 
NO Rir-iE~ N 
CHE:S~PE:AKE 3AY 
SHORE~INE 
ST'A~UTE Mll.[S 
l 10 , ' 0 2: 
I 
Figure 2.2 Wave energy distribution along the northern 
Chesapeake Bay shoreline (from Wang, et al., 
1982) 
15 
depths in the Bay which preclude the formation of large 
wind-driven waves (US Army Corps of Engineers, 1984)). 
Because these average heights are fairly low, storm-wave 
conditions are almost certainly more important in 
assessing shore erosion. 
Vegetation Communities 
The two major ecosystems of all the islands in the 
study area consist of open coastal marshes and upland 
forested areas. The vegetation found on the islands is 
typical of the ecosystem in the Chesapeake Bay region. 
Table 2.2 identifies the major vegetation communities on 
the islands in the study area. 
Geologic History of Chesapeake Bay 
The Chesapeake Bay was formed as sea level rose 
during the past 15,000 years, and the Susquehanna River 
valley was flooded to form the present estuary (Coleman 
and Mixon, 1988). The modern Bay is the most recent of 
at least three generations of estuaries, which have 
formed in a similar fashion during interglacials. Three 
paleochannels of the former Susquehanna River valleys 
have been located and dated (Figure 2. 3} . They are known 
as the Cape Charles, Eastville and Exmore paleochannels, 
16 
TADLB 2.2 
MAJOR VEGETATIVE COMMUNITIES 
Environment Latin Nawe co-on ,,.e 
Salt?arsh ? soartina alterniflora Cordgrass 
? Spartina oatene Saltmeadov Hay 
Pistichlis soicata Salt Gr11BB 
Juncue roemerianus Black Needlerush 
Atriplex patula orach/Spe11rsc11le 
salicornia Saltwort/Gl11sswort 
Borrichia frutescens sea oxeye 
Margin Iva frutes~ene Marsh Elder 
a,accharis hali?ifolia Groundsel Tree 
~ 
..,J 
Beach PaniCU11 viraatua switch Gra1111 
Upland Pinus taeda Loblolly Pine 
nvrica pennsylvanica Bay Berry 
Mvrica sop, Wax Myrtle 
sassafras albldu? Sasl!lafras 
Prunus serotina Wild Black Cherry 
Juniperus virqiniana Red Cedar/Juniper 
Ilex opaca Alllerican Holly 
Lonicera japonica Honeysuckle 
Phvtolacea a?ericana Poke weed 
Rhus radicans Poison Ivy 
Celtis occidentalis Hackberry (l!lpp.) 
EXPLANATION 
39-00'N D Cape Charles 
- Eastville 
m:il Exmore 
......_ Mocsern T idol 
38?20' 
38?00' 
37?401 
37?20' 
1 
37?00
Figure 2.3 Location of the three major paleochannels in the 
Chesapeake Bay (from Coleman, et al., 1990) 
18 
in order of increasing age (Coleman, et al., 1990). 
These channels were formed during glacial low sea-level 
stands. Each paleochannel exhibits the same 
sedimentology, with lower fluvial channel-fill deposits 
consisting of sand and fine gravel. These f 1 uvial 
deposits are covered by river-estuarine sediments, 
consisting of interbedded muddy sand, silt and peat 
(Coleman and Mixon, 1988). 
The oldest channel, the Exmore channel, is not 
clearly dated, but appears to be 200 to 400 thousand 
years old. It extends from the mouth of Eastern Bay, 
through the Poplar Island area, into the Taylor Island 
area, and down into the southern Bay. This channel runs 
essentially parallel with the chain of islands in this 
study. The Eastville channel appears to be late 
Illinoian in age, or about 150 ka. The youngest 
paleochannel, the Cape Charles channel, is clearly of 
late Wisconsin age, about 18 ka (Coleman, et al., 1990). 
This channel was formed when sea level was about -85 m on 
the mid-Atlantic continental shelf during the most recent 
low sea-level stand. During this time the area occupied 
by the Chesapeake Bay was subaerially exposed and a 
narrow, steep-walled valley was incised into the coastal 
plain strata by the Susquehanna River and its major 
tributary the Potomac River (Coleman, et al., 1990). Sea 
level began to rise around 15 thousand years ago and the 
19 
Cape Charles channel was flooded, eventually forming the 
modern Chesapeake Bay. 
All the islands in the study area appear to be 
composed of fine-grained clay deposits which are either 
exposed in areas of high elevation or buried under marsh 
peats in low-elevation marshy areas. This deposit is 
known as the Kent Island Formation, which is thought to 
be estuarine in origin and likely represents the "old 
Chesapeake Bay bottom which preceded the formation of the 
modern Chesapeake Bay" (Owens and Denney, 1979). 
Therefore, the sediments which comprise the islands were 
probably deposited during the most recent Pleistocene 
high sea-level stand. The Kent Island Formation has 
never been dated, however, so its precise age and origin 
remains a subject of research. 
While the Cape Charles paleochannel was filling in 
at the beginning of the Holocene, the deposits which 
formed the Kent Island Formation were submerged and 
reworked by the rising water, and areas of high elevation 
were surrounded by water to become islands. Therefore, 
the deposits that form the islands and parts of the 
southern Delmarva Peninsula are all geologically young, 
younger than the Eastville or Exmore paleochannels 
(Coleman, et al., 1990). 
20 
Sea-Level Rise 
An underlying cause of land loss in the Chesapeake 
Bay is rising sea level. Sea-level rise affects coastal 
areas in several ways, including erosion, inundation of 
low-lying areas, saltwater intrusion into aquifers, 
higher water tables, and increased flooding and storm 
damage (NRC, 1987). Erosion and inundation account for 
the loss of land which has been occurring in the Bay. 
Rising water tables and saltwater intrusion have altered 
the vegetation on large areas of some of the islands and 
along the margins of the eastern shore. Increased 
flooding and storm damage have reduced the amount of 
inhabitable land on the islands. 
The rate of local sea-level rise in the Chesapeake 
Bay appears to be accelerating (Kearney and Stevenson, 
1991) at the rate of about 3.0 mm per year for the last 
few centuries (Froomer, 1980), as compared to the slower 
rate of 1. 2 to 1. 5 mm per year for the last several 
millennium (Newman, et al., 1980). A recent rise in 
global sea level is consistent with the termination of 
the Little Ice Age around 1850 (Grove, 1988). It is 
assumed that the recently accelerated rate of sea-level 
rise in the Chesapeake Bay accounts for the increased 
rate of island erosion, interior marsh loss, and vertical 
marsh accretion, since the mid-19th century as 
demonstrated by area estimates of islands and marsh core 
21 
samples (Kearney and Stevenson, 1991). 
The rate of local sea-level rise in the Chesapeake 
Bay is also well above the eustatic (global) rate of sea-
level rise during the last century for which best 
estimates are 1 to 2 mm/yr (IPCC, 1990). This has been 
attributed to downwarping of the earth's crust underneath 
the Chesapeake due to sediment loading of approximately 
8 trillion kilograms of silt during the last century, as 
a result of human land-use practices (Donoghue, 1991). 
Davis ( 1987) has also suggested that high rates of 
relative sea-level rise in the Chesapeake Bay are caused 
by regional subsidence due to withdrawal of underground 
water sources. In the Chesapeake Bay, the rate of 
subsidence appears to increase towards the south and 
culminates in the lower Virginia portion of the Bay 
(Holdahl and Morrison, 1974) 
In a detailed examination of tide gauge records, 
Douglas (1991) estimates that the rate of sea-level rise 
at the Baltimore, Maryland, tide gauge station during the 
period 1880 to 1980 is 2.1 mm/yr? 0.1, when the effects 
of post-glacial rebound (PGR) are removed from the 
calculations. Douglas I finding is the only instance 
where the process of post-glacial rebound is ascribed to 
the Chesapeake Bay; most studies of the Bay suggest that 
the Bay is sinking rather than rebounding (e.g., Kearney 
and Stevenson, 1991). 
22 
We are presently in an interglacial sea-level 
period. It is unclear whether the observed eustatic 
trend of increased sea-level rise during the last century 
is simply the natural variability in the long-term 
climatic record, or whether it is an indication of 
anthropogenic global warming due to the greenhouse 
effect. 
Marsh Response to Sea-Level Rise 
Marsh stratigraphic records and pollen dating 
analysis show that marshes can develop and keep pace with 
sea-level rise. This is accomplished by building upward 
and outward with additions of dead biomass (detritus) and 
inorganic sediment settling on the marsh surface 
(Redfield, 1972; Stevenson et al. 1986). By reporting 
basal peat dates in the Chesapeake Bay as old as 4510 BP, 
Pardi et al. ( 1984) demonstrated that Chesapeake Bay 
marshes have generally been keeping up with rising sea-
levels for at least this long. However, in the last few 
centuries the reduction of a sediment source and the 
increased pace of sea-level rise in the Chesapeake Bay 
has created a sediment deficit in relation to sea-level 
rise (Stevenson, et al., 1985). As a result, marshes 
such as Blackwater National Wildlife Refuge are 
deteriorating (Pendleton and Stevenson, 1983). The large 
23 
marshy islands, such as Bloodsworth, South Marsh and 
Smith, are also experiencing interior marsh degradation, 
which is evident by examining sequential aerial 
photographs of the same location. It is likely that a 
combination of a sediment deficit, local subsidence, and 
rising sea levels are causing the marsh loss. 
Marshes also respond to sea-level rise by migrating 
inland, encroaching on upland areas and subsequently 
converting them to marsh. This process of upland 
conversion is evident on many of the islands today, where 
trees are dying at the edge of upland forests ( Plate 
2.4), and where marsh peats are developing over the clay 
layer of the Kent Island Formation (Plate 2.5). The peat 
layer varies in thickness and therefore in age. GEO-
RECON ( 1980) estimates that some of the marshes on 
Bloodsworth Island first began to develop around 400 
years ago, based on depths of the peat layer. 
Future Sea-Level Rise 
Another question to consider is the effects of a 
continued and/or accelerated future sea-level rise. The 
possible impacts of an accelerated sea-level rise 
include: (1) coastal inundation, (2) increased erosion, 
(3) change in the circulation and salinity of estuaries 
and lagoons, (4) increased storm damage, (5) loss of 
24 
Plate 2.4 View of tress dying at the edge of an upland 
area on Lower Hooper Island, an example of the 
upland conversion process. 
25 
Plate 2.5 View of a marsh peat layer over the silt/clay 
layer on Lower Hooper Island 
26 ? 
wetlands, (6) changes in the ecotomes and habitats, (7) 
loss of turtle and bird nesting areas, and (8) increased 
saltwater intrusion into groundwater (Emery and Aubrey, 
1991). The Intergovernmental Panel on Climate Change 
(1990) estimates that eustatic sea level will rise 
between 8 and 29 cm by the year 2030, with a best 
estimate of 18 cm (Figure 2.4). This rate of sea-level 
change is from 3 to 6 times faster than the last 100 
years. If sea-level is rising in the Bay at a rate two 
to three times fast'er than the global average and this 
trend continues, then the best estimate for the Bay would 
be a rise in sea level of about 24 cm by the year 2030, 
56 cm by the year 2070, and 82 cm by 2100. The 
Chesapeake Bay figures are obtained by calculating the 
rate difference between the global sea-level trend (1.8 
mm/yr) and the Baltimore trend ( 3 ? 3 mm/yr) ? This 
difference (1.5 mm/yr) is multiplied by the number of 
years in a given time period, and then is added to the 
IPCC best estimate calculation. 
The IPCC estimates are based on scientific theories 
and careful modeling of climate warming due to the 
increased presence of radiative gases in the atmosphere. 
Radiative or II greenhouse II gases, including CO2, NO, water 
vapor, methane and chloroflourocarbons, have the ability 
to trap outgoing radiation or heat emanating from the 
Earth, s surface, thereby trapping heat in the atmosphere. 
27 
Scenario BA U 
100 Chesapeake Bay 
best estimate 
171 
" I Q...J'   
QJ 
E 
C 
(J so 
estimate 
t,.J 
(X) 
n~rrrr,, ,, ,,,, I 
2000 2025 2050 2075 2100 
Year 
Figure 2.4 Global sea-level rise, 1990-2100, for Policy Scenario Business-as-Usual. 
(Adapted f~om IPCC, 1990). 
The theory of global warming predicts that the higher 
surface temperatures on earth will cause substantial 
climate warming which will result in sea-level change due 
to thermal expansion of the surface layer of the ocean, 
continental ice melting and retreat, and changes in ocean 
circulation and wind patterns. Due to large 
uncertainties about the extent of future temperature 
change due to global warming, there are many questions 
about how important each of these effects might be. 
However, even with substantial reductions in the 
emissions of the major radiative gases, future increases 
in temperature and consequently sea level are unavoidable 
due to the lags in the climate system. In other words, 
there is a "commitment" to a rise in sea level which is 
estimated to be 18 cm by 2030 and 41 cm by 2100 (IPCC, 
1990). Future rates of sea-level rise are an important 
consideration for coastal areas such as the Chesapeake 
Bay. An acceleration in the rate of sea-level rise can 
only exacerbate the rapid coastal land loss already 
occurring in the Bay. 
Islands in the Study Area 
Introduction 
The study area consists of a sample of seven islands 
along the eastern shore of Chesapeake Bay (Figure 1.1): 
29 
Barren Island 
Bloodsworth Island 
Hooper Island 
.James Island 
Poplar Island 
Smith Island 
South Marsh Island 
The islands are very low-lying, with elevations less than 
about 2.5 meters above sea level, based on USGS 
topographic charts. The highest elevation measured 
during fieldwork was 2.4 m above mean sea level (MSL) on 
Poplar Island. Tidal range in the Bay is about 1 mat 
the mouth of the Bay and decreases to about O. 3 m at 
Baltimore. The tidal range at all the island sites is 
about 0.5 m (Wang, et al., 1982). Much of the eastern 
shore is used for agriculture and farming, and the towns 
are the homes and ports for watermen. 
The islands can generally be divided into two 
morphologically distinct types: large marshy islands and 
small upland/marsh islands. Bloodsworth, Hooper, Smith 
and South Marsh Islands are large, marshy islands? 
Barren, .James and Poplar are small, upland/marsh islands. 
The island shores mainly consist of eroding clay bluffs, 
eroding marsh, and a few small pocket beaches. 
No detailed study of the origin of the islands has 
been undertaken. One possible mechanism is related to 
the antecedent topography of the Chesapeake Bay, as 
mentioned earlier (Kehrin, et al., 1988). As sea level 
rose and flooded the Susquehanna River valley, areas with 
30 
elevations high enough to remain above the rising water 
eventually became cut off from other areas and became 
islands or peninsulas. Since sea level has continued to 
rise, these islands have been reduced in size by 
submergence and erosion. These processes are still 
occurring in the Bay today, as existing islands and the 
mainland shore are experiencing rapid land loss. 
Human populations have historically used the islands 
for living, farming and fishing. Watermen and their 
families from nearby areas settled on the islands because 
they provided easy access to the Bay. In addition, in 
the 18th and 19th centuries, many of the islands offered 
ample space for settlement and farming which was an 
attractive proposition for many Bay-area pioneers (Meyer, 
1986). The populations of most of the islands peaked 
around the end of the 19th century. 
However, the processes of land loss since the mid-
19th century have reduced the availability of arable, 
habitable land. As a result, most of the islands which 
were once inhabited have been abandoned. Hooper and 
Smith Islands still have permanent towns which exist 
barely above the water level. South Marsh Island is the 
only island which was never inhabited by European 
settlers. Poplar, James, Barren and Bloodsworth Islands 
each have a history of settlement and subsequent 
abandonment of human communities. 
31 
Barren Island 
Barren Island, located in Dorchester County, 
Maryland, is about 1.6 km west of Hooper Island. It is 
currently about 7 5 ha which is dominated by upland 
forested areas, with some fringe marshes. Barren Island 
has seen the rise and disappearance of a prosperous 
community. Families settled on Barren Island because of 
its proximity to the Bay for fishing and oystering, and 
for available farm land. By 1877, the community of 
Barren Island reached a maximum of 13 farms and a 
schoolhouse (Cronin, 1988). Soon thereafter, families 
began moving their houses to the mainland where the 
living conditions were preferable. By 1916, the last 
family had left the island. 
There is still a hunting lodge on Barren Island 
which was built in the 1920 's by William Siskind who 
owned the island until recently. originally, it was more 
than 300 m from the western shore of the island. Twenty-
three years later, in 1952, the lodge had to be protected 
by a wooden bulkhead built about 30 m to the west of the 
building, a clear sign that erosion was rapidly 
encroaching on the lodge. In 1964 the breakwall was 
still intact, but was seriously undermined on either end. 
By 1987, the breakwall had failed, the house had 
partially fallen in the Bay and the site was abandoned. 
32 
A site visit in 1991 revealed that the property is 
completely abandoned, except for a family of peregrine 
falcons nesting on the roof of the dilapidated structure 
(Plate 2.6). The erosion continues to cut away at the 
western side of the island at a rapid rate with no 
likelihood of stopping. At one time, Siskind appealed to 
Dorchester County and the u. s. Army Corps of Engineers to 
help stabilize his property, but his proposal was denied 
(Cronin, 1988). 
The U.S. Fish and Wildlife Service has recently 
acquired Barren Island due to its important habitat 
resources for ducks and other waterfowl. In addition, 
the island hosts a large heron and egret rookery and a 
bald eagle nest. Exact plans for the island are not yet 
known, and feasibility studies would be required for any 
type of habitat restoration project (Walter Quist, U.S. 
Fish and Wildlife Service, personal communication, 
October, 1991). 
Bloodsworth Island 
Bloodsworth Island, in Dorchester County, Maryland, 
is located about 5.6 km west of Deal Island. The island 
today is about 1,909 ha of marsh with one linear upland 
ridge of about 1. 2 ha. The ridge, called Fin Creek 
Ridge, is sparsely covered with Virginia pine and black 
cherry trees. Surveys on the ridge revealed building 
33 
Plate 2.6 View of the hunting lodge on Barren Island 
Plate 2.7 View of an eroding graveyard on Lower Hooper 
Island 
34 
foundations, brick rubble, and other artifacts beneath 
about 15 cm of loam, and overlaying a layer of sterile 
tan clay ( GEO-RECON, 1980). A 1849 NOS chart shows seven 
buildings on several small upland areas, but the majority 
of the island is denoted as salt marsh. Some of the 
upland areas appear to be cleared and diked, and an 
orchard can be seen in one area (Figure 2.5). In 1877, 
land records indicate that 14 landowners or residents 
occupied the island (GEO-RECON, 1980). Now, Bloodsworth 
Island is completely uninhabited by humans. 
In 1948 the U.S. Navy bought the island which has 
since been used as a bombing range and testing area. 
Despite the regular bombing of the island, it is an 
important overwintering and stop-over area for waterfowl, 
including geese, ducks, herons, egrets, songbirds, 
ospreys, and a Bald Eagle (U.S. Navy, 1981). The u.s. 
Fish and Wildlife Service and the U.S. Navy are 
formulating a cooperative waterfowl/wetland management 
program for Bloodsworth Island. Part of their study will 
determine the effects of bombing craters on waterfowl and 
the health of the marsh. 
Hooper Island 
Hooper Island, in Dorchester County, Maryland, has 
been occupied since at least the mid-19th century. It is 
a combination of upland and marsh areas. The island is 
35 
1" 
N 
AREA 
,. 
DIKED 
.-? .  -.. .. . .. ~ ~ ..;,,,;,.. .?.. .. ? ' . 
...... . . .... ... . ?: -. .-.  . 
--?- ?- ---?---.. - --
--.- ?? ..  --? -? ........?. .., --?- -
 --.. -... -~-?--.. -
.. --? . . . . . .. .- ' . ?-
-- ,,. :. - .. . .. . - . 
?- .. .. ? 1 .; ? ? . ? ,. 
0 1000 2000 FHI 
SCALE 
Figure 2.5 Enlargement of the 1848 T-sheet of Bloodsworth 
Island, showing the location of an orchard and 
several diked areas (from GEO-AEGON, 1980) 
36 
attached to the southern part of Taylor's Island by a 
bridge, so access to Hooper Island is relatively easy. 
Running along the western side of the Honga River, the 
island is about 13 km long (North-South) and 1.6 km wide 
(East-West) at the widest point. Three towns on the 
island are Honga, Fishing Creek, and Hoopersville. 
Erosion is an evident problem for the residents of 
Hooper Island. Wooden breakwalls and revetments have 
protected much of the island from the erosion on both the 
western and eastern sides since the mid 1900, s. This has 
provided the island with physical stability which, 
together with vehicular access to the island, has allowed 
the inhabitants to remain. on the southern end of Hooper 
Island there is a small graveyard on the edge of a marsh 
which is being eroded to the point where gravestones are 
falling in the water and wooden coffins are protruding 
from beneath the surface layer of the marsh (Plate 2.7). 
This part of the island does not have any shore 
protection structures and appears to be rapidly eroding. 
Flooding is also a problem for the residents as many 
of the houses and buildings are elevated above the ground 
by about o.s m or more. In addition, coffins must be 
encased in cement to prevent the wooden coffins from 
becoming afloat with the high water table (Plate 2.8), 
and many lawns are level with the watertable (Plate 2. 9). 
The entire island is less than a meter above sea-level, 
37 
Plate 2.a View of cement encas/d graves on Middle Hooper 
Island 
Plate 2.9 View of a flooded la~ on Middle Hooper Island 
38 
-
and in many places only about 100 m wide. A major 
Northeaster on October 31, 1991 flooded the entire island 
by about 0.3 m. 
James Island 
James Island, in Dorchester County, Maryland, is 
located about 1 mile north of the northernmost point of 
Taylor's Island. From observations of a 1848 NOS chart, 
it is clear that James Island was formerly a peninsula 
which was attached to the northern point of Taylors 
Island (Figure 2.6). A single road from Taylors Island 
extended north along the length of James Island with a 
few small side roads. There were eleven buildings and 
about 70\ of the island appears to be cleared and 
cultivated. Because the Island was close to the Bay?s 
fishery resources and readily accessible by Taylor's 
Island, it was probably an attractive place to settle. 
By 1901 James Island had become a true island as the 
connecting neck of lowland was totally eroded (Figure 
2.7). There was no road and only five buildings at this 
time. About half of the island was cultivated, including 
one tree farm. Clearly, around the time the island was 
separated from the mainland to become a true island, the 
inhabitants began to move to the mainland rather than 
remain on a rapidly eroding island. 
39 
. i?.".?   
?4, ._,.
?
..
 J...; 
? 
.?.  
?, \. 
- . ~ 
. . . . . . . . ': ~- .. ~. . ~, 
?' I ? [; ,; ?? 
..? . ? ? '! 
? I 
.... 
; 
? I 
.;;..N. ;-. '.'  .. . 
.,.._ . 
. ... 
.:  .,.?  :- . ~. . r  .:- .. ' . 
. ... a 
. .l- !' ....  ..  ' 
?.: ' 
Figure 2.6 Enlargement of 1848 T-Sheet of James Island 
showing the island attached to the mainland 
40 
+ + 
,. , , 
: 
~ 
-.----~+ .,,,,,, . . ... .-..L ?? ?- ,..  1 ? ? ..;.. .. ??.  ?- I . . 
"' 
?. 
0 ??? o , ? h ? 
.. .... . .- . .. . ??~-- . -.. ... ... .... , ?? ~ I? l? .... ... ??:? ~.. . .? ? . ..,.. 
 
... . "  
?, . ': ?.-? 
?- -..
.
h   
?, +. . 
? t ? . - ?":-,. 
Figure 2.7 Enlargement of 1901 T-sheet of James Island 
showing the island separated from the mainland 
4 1 
By 1941, there were no farms, and the island had 
separated into two pieces. No one inhabited the island 
after this point; it was eroding so rapidly from the west 
that living on the island had probably become very 
unattractive. Today the island is only about 45 ha in 
size. 
Poplar Island 
Poplar Island, in Talbot County, Maryland, is 
located 1.6 km west of Tilghman Island, about 8 km north 
of the mouth of the Choptank River. In the late l600's 
Poplar Island was a single island. Less than 200 years 
later, in 1846, it had broken into three islands, known 
as Coaches, Jefferson, and Poplar Island. Together these 
three islands are known as the "Poplar Complex". The 
Poplar Complex now consists of two large islands, 
Jefferson and Coaches, and seven small islets. Today, 
the total size of the Poplar Complex is about 43 ha. 
Poplar Island has received a great deal of attention 
in the press presumably because a large community 
persisted on the island for nearly 50 years, and the 
Jefferson Island Club which entertained Presidents 
Roosevelt and Truman was located on Jefferson Island. 
From the 1aao?s to 1920's, as many as 20 families lived 
on the island and the community included a general store, 
post office, one-room schoolhouse, church, lumberyard, 
42 
and 6 farms which grew tomatoes, tobacco, watermelons 
I 
cantaloupes, corn, wheat and trees. Delores Reese, a 
former resident of Poplar Island, who now lives in st. 
Michaels, Maryland, described Poplar as a nbeautiful 
island with oyster shell walks and little white picket 
fences" ( Cronin, 1985). By 1918, the schoolhouse was 
closed, which indicates that the population had begun to 
decline and as soon as 1929 the island was uninhabited. 
The island is currently owned by the Poplar 
Investment Group who use the island during the hunting 
season. There is one building on the southern point of 
Jefferson Island and a trailer on Coaches Island for 
visitors. The small islets range in size from about 1 m2 
to under o. s hectare, but they are completely 
uninhabitable (Plate 2.10). 
The State of Maryland and the U.S. Army Corps of 
Engineers are presently considering a proposal by the 
U.S. Fish and Wildlife Service to use Poplar Island as a 
Waterfowl habitat restoration project. This project 
proposes to use dredge material to recreate valuable 
waterfowl habitat including tidal marsh and upland areas 
(Figure 2.8) (John Gill, us Fish and Wildlife Service, 
Personal communication, June 16, 1991). 
Smith Island 
Smith Island, in somerset County, Maryland, is 
43 
Plate 2.10 View of Poplar Island islets 
44 
Figure 2.8 Schematic drawing of Poplar Island habitat enhancement 
proposal (from U.S. Fish and Wildlife Service, undated 
draft) 
USFWS Proposed 
~. .. , .... lfuture 
?L-9<~\ -._. ~ 
..t\')'~~i; J? ~ Low Marsh 
? I ligh Marsh 
~ D Low Dunelands 
~~ ~ Shrub & Forest 
~ ~ 
l1l JJ 
-~<A, 
Coaches 
-
located 13 km west of the town of Crisfield. It is about 
2,800 ha, dominated by wetlands with a few linear upland 
ridges. Three of the l a rgest ridges are occupied by the 
towns of Ewell, Tylerton, and Rhodes Point, which make up 
the entire population of the island of about 530 {Figure 
2.9). The northern portion of the island comprises the 
Glenn L. Martin National Wildlife Refuge. 
Smith Island was colonized in 1657 by the Tyler, 
Bradshaw and Evans families who settled on the island in 
search of available land for farming. Although farming 
is no longer an industry on the island because of 
frequent flooding, the towns have evolved into important 
fishing communities for the entire Chesapeake Bay. Most 
buildings are slightly elevated to help prevent damage 
from frequent flooding. 
The effects of floodi ng and inundation are more 
important to the island's residents than is erosion 
because flooding affects them more directly. The 
majority of the island is very low-lying marsh, less than 
about 0.5 m above mean sea-level {msl). The ridges are 
a little higher, being only about 1 m above msl, 
according to recent USGS topographic surveys. Over 95% 
of Smith Island is mapped within the 100-year flood zone 
(FEMA, 1980; U.S. Army Corps of Engineers, 1984). As a 
result, flooding from storms causes frequent and 
recurring damage. 
46 
Historical Shoreline Change, Smith Island: 184B to 19B7 
1jJ8(J00() .. ---~ ~ - -illQQPD~--~ 
Glenn L. Martin National Wildlife Refuge 
,., . 
t 
~: 
n ~--~~ ... I i 
a 
\ llhn,IP<: l>ninL ~~ I .1 
"..'."J'  
NOTE: The shaded land 
existed in 1987. 
.. ? ? SM9 + f,Of 
? SM7 
fscala I I I I --1--t 10 S.P. Zona 50: Maryland 
b:J!.OP.9()_ o F'aet  25000 Fr?!!HLIL~!'!~r.:_\_j)_at;!!~_g_9/ru9,,,,1.____ _ 
Figure 2.9 Map of Smith Island with town locations 
The eroding edges of the island are not directly 
impacting the towns which are slightly inland. Rhodes 
Point is the most threatened town. Erosion cannot be 
overlooked as a component of land loss for this island, 
however, since over 1,200 hectares have been eroded from 
the perimeter of the island since the mid-1800 1 s. 
Because Smith Island has been so isolated from the 
mainland, it has managed to retain some of its 
traditional culture from when it was colonized in the 
17th century. The main lifestyle of most Smith Islanders 
is still that of the watermen. Some residents of the 
island speak with a unique dialect which originates from 
colonial English. Many of the residents are descendants 
from the original families who settled on the island. 
However, the ferry service from Crisfield, Maryland and 
Reedville, Virginia carries tourists regularly to the 
island, bringing along modern-day ideas of development 
and tourism. Although the culture is changing with the 
introduction of modern conveniences, the island and its 
residents still retain some of their original charm and 
uniqueness. 
South Marsh Island 
south Marsh Island, in Somerset County, Maryland, is 
a Maryland State Wildlife Management Area. European 
48 
- ' 
settlers have never occupied South Marsh Island, which is 
a 1,200-hectare, marshy island and an important breeding 
and nesting area for waterfowl in the Bay. At the 
present time, the island is entirely salt marsh, with no 
upland ridges. Despite the lack of ridges, this island 
likely formed in a similar manner as Bloodsworth and 
Smith Islands (GEO-RECON, 1980). The lack of ridges can 
be explained if the island has a very flat or lower pre-
Holocene clay layer. Hovever, no cores or detailed 
geologic survey have been undertaken to confirm this 
conclusion. 
49 
CHAPTER 3: METHODS 
Introduction 
The processes and rates of shoreline change were 
investigated for seven islands in the Chesapeake Bay. 
The study consisted of several phases, each of which 
contributed to the understanding of the processes of land 
loss for the study area and predictions of the islands' 
future. The phases were: 
I: Historical Shoreline Mapping 
II: Field Surveys 
III: Data Analysis 
IV: Forecast Modeling 
Historical shoreline change maps were generated for 
each island using a computer mapping procedure, showing 
the land loss for each island between the period of about 
1848 to 1987. Therefore, the historical data for each 
island covered nearly 140 years, enabling long-term 
trends of shoreline behavior to be identified. Modeling 
future shoreline response was based on the long-term 
historic erosion rates and the predictions of future sea-
level rise (see Figure 2.4). 
50 
Historical Shoreline Mapping 
The historic rate of land loss for each island was 
quantified using a computer mapping technique termed 
Metric Mapping (Leatherman and Clow, 1983), which: 
1. utilizes different historical shoreline data 
from NOS Topographic Surveys ("T-sheets") and 
vertical aerial photographs; 
2. corrects errors inherent in these sources; and 
3. displays each shoreline on a common grid 
system to allow for quantitative comparisons. 
Shoreline change maps generated by this system meet and 
generally exceed National Map Accuracy Standards ( Crowell 
et al., 1991). Metric Mapping proceeds in three general 
steps: 1) data selection and preparation; 2) shoreline 
digitization; and 3) data processing and analysis (Figure 
3.1). The Metric Mapping Users Guide (Laboratory for 
Coastal Research, 1990) provides a detailed explanation 
of the entire procedure. 
Data selection 
The data for each island includes a combination of 
NOS T-sheets and vertical aerial photographs. Twenty-two 
NOS T-sheets and 48 aerial photographs were used for this 
study (Appendix B). 
51 
Figure 3.1 Flow chart of Metric Mapping procedure 1 
---- ------------
NOS CHARTS AIR PHOTOS COMMON POIHJ5 
- 11lect 11cond1,, conhol poinll 
1elect conhol poinll on annolal1 Hrial pholog11ph1 cornmofl on UOS <.h11t1 
tlOS ch1r11 and ??1111 photo, 
2 3 I 
' 
~~- 
~ 
Ul 
ts) 1 
DIOIIIZI AIR PHOTOS CONVERT PROGRAM DIGIHZE CIIARlS 
d1g1hH 1w photo?. including ? obt1in 11111 rl?ne coo1dinalea ol .-.(-- d,gili1? puma,, ?nd eecnnd11y 
??co11de1, conltol pomle aecond11y conlculs hom puma,, conhol r,011,11 on tlOS 
and hducial m11li11 conhol pornta wilh 1 aheal1 
\& . !I conve1l progran, 4 
SPACE RESlCllON 
PROGRAM MESH PROGRAM PROGRAM PLOI 
u111pac1 11 ~ echon p109ram to 
11an?lo1m ??? pholos lo ??mow? - 1djn1I junction? belween plol m1p1 wrlh compule1 
ud,al and lrll d11lmtion and adj1cenl photo? wrlh 
uale d,ll11ence1 in MESII p1og1am 
7 each pholo 8 II 
~?~ ~~~-~--- ~$a;-:=-;~~;~~ 
Nos T-Sheets 
NOST-sheets were produced about every 40 years, 
beginning in the mid-1800 's. These maps are an excellent 
source of shoreline data for several reasons: (1) they 
are the most accurate historic shoreline data commonly 
available ( Leatherman and Clow, 1983); ( 2) they have 
large scales of 1:10,000 or 1:20,000, and therefore 
Provide a high level of detail; (3) the surveying program 
covered the entire coastline in the Chesapeake Bay, 
including each of the islands in the study area; and (4) 
the surveying program extends back to the mid-1800's, ,, 
., . 
Providing about 140 years of data. ' ?:,  
I 
I? 
'? 
However, there are three major problems with using 
these maps as a data source: (1) there are no recent T-
Sheets available in the study area due to a reduction in 
the surveying program beginning in the mid-1900's; (2) 
some of the older NOST-sheets are distorted because less 
accurate surveying techniques were used in the past; and 
( 3 ) in some cases the triangulation stations are not 
Updated to the 1927 datum (Shalowitz, 1964). These 
respective problems are overcome by: (a) using recent 
aerial photographs to update the map shoreline 
information; (b) quality control which identifies and 
hence, eliminates distorted or inaccurate maps; and (c) 
UP<iating triangulation stations to North American Datum 
of 1927 (NAD 27 ) using coordinate data from the National 
53 
Geodetic Survey in Rockville, Maryland. 
Vertical Aerial Photographs 
The photographs used in this study were obtained 
from the U.S. Department of Agriculture. These black and 
white vertical aerial photographs, dated from 1952, 1964 
and 1987, are at a scale of either 1:7,920 or 1:12,000. 
For the smaller islands - Barren, James and Poplar - one 
photograph provided total coverage for each island. For 
the larger islands, such as Bloodsworth, Hooper, Smith 
and South Marsh, a mosaic of overlapping photographs was 
used for complete coverage of each island. 
Data Preparation 
Several steps are required to prepare the maps and 
the photographs for digitizing by the Metric Mapping 
procedure. First, primary control points were carefully 
chosen on the maps, digitized, and then checked for 
accuracy. Primary control points are points with known 
latitude-longitude coordinates which have been updated to 
NAD 27. These are either latitude-longitude tick marks 
or specific triangulation stations, for which the exact 
location were obtained from the National Geodetic Survey 
(Figure 3 ? 2 ) . To check for accuracy, the computer 
compared the digitized coordinates of the primary control 
points -with the known coordinate system. In all cases, 
54 
Figure 3.2 Example of Primary Control Points 
() 
1-, , 
? .. ? Ul .. 
Ul p 
Primary Control Points ? ~ ... .. ' 
/~ 
Q BE:J'l(ON ?3. BLACK l':'2'} 
Cho..rit.~ Poi.n.t ~ 
. 1l-~ 
f "st'-1, 
~ 
f::. 0 B~coN ?4, Rr:o 1323 
the accura cy o f the pri,m ary control po,i nts was within o. 2 
mm of the exact location, meaning that it is within 4 m 
of the exact location for a 1:20,000 map and within 2 m 
for a l:10,000 map (Crowell et al., 1991). Since 0.2 mm 
is Within the accuracy acceptance limits, no T-sheets 
Were discarded. Shoreline segments were then identified 
on the maps at intervals around the island and numbered j/ 
for digitizing. 
For the aerial photographs, secondary control 
Points, Which are locationally stable points common to 
both the maps and the photographs, were identified on the 
Photographs and the base map. The base map is used to 
transform the photographs to the latitude-longitude 
coordinate system of the maps. Some of the most commonly 
Used secondary control points are structures such as 
corners at the base of buildings, road intersections, 
Piers and jetties (Figure 3.3). In some cases, where 
roads or buildings were not in the photograph, geomorphic 
features such as stream intersections, stream openings 
and small, erosion-resistant promontories served as 
secondary control points. 
The paucity of both geomorphic and structural 
secondary control points was the major constraint on data 
accuracy in this study. For smith Island and South Marsh 
Island, there was an insufficient number of viable 
geomorphic control points. This is due to the dramatic 
56 
-"C 
a, 
.,, ?
IC '- -~ o 
0 :a. 
~ 0 
~ -Cl) C 
a0..  
-.0.. . C 
0 
(.) 
c." 
,_ '? ro 
,...  1J C 
0 
(.) 
Q) 
(f) 
I ... , ,. 
I , 
I ~ 
I , 
/ , 
?? ...., ,  
,,, ,
, , 
57 
. ~ .. ~ -~ -~ ..__ 
Physical changes which had occurred between the date of 
the most recent map (the "base map") and the photographs. 
For this study, the base maps were dated around 1942 or 
1943 ? The lack of stable infrastructure also caused 
difficulties in locating accurate structural control 
points. The 1952-series photographs for these two 
islands could not be tied accurately to the latitude-
longitude coordinate system of the maps in the same 
manner as the other islands. Thus, it was necessary to 
discard them. 
In addition, an alternative method was developed to 
,,,,?. 
tie the 1987 photographs for Hooper, South Marsh and .:; , .'.   
I, 
Smith Islands to the map coordinate system. This method ii 
,l~f, 
Used recent 7.5 minute topographic maps ("USGS quads"), .!.J , 
from 1972 and 1985, to identify secondary control points f ?<'' ,:1' 
1/ 11 
for the 198 7 photographs. The USGS quads were used li 
merely to identify secondary control points; their I 
shorelines were not digitized. Significantly less 
shoreline change had occurred between the time the USGS 
quads were surveyed and the photographs were taken. As 
a result, finding viable geomorphic secondary control 
Points was much easier. In general, using the USGS quads 
to identify secondary control points proved to be an 
accurate methodology. 
The approximate shoreline location was identified on 
the photographs and marked with a fine red pencil. Most 
58 
of the shoreline was composed of a 1 to 2 m cliff face or 
an eroding marsh s c arp. In these areas, the shoreline 
was i dentified as the marsh/water or cli ff/water 
inte rface. Where a small pocket be ach was present, the 
mean h igh water line was identified as the line of dark 
sand. Shoreline segments were then numbered for 
digitizing. 
Digitizing 
Digitizing was accomplished with the Atlas Draw 
digitizing program, an integral component of Metric 
Mapping. Shoreline segments were identified on both the 
maps and photographs at lengths appropriate for accurate 
digitizing. For the smaller islands, Barren, James, and 
Poplar, each map and photograph covered an entire island. 
The larger islands required a mosaic of photographs and 
maps for complete coverage of the island. As a result, 
line segments had to be connected on adjoining 
photographs and maps, which made digitizing more 
complicated and added a potentially significant error 
factor due to photograph distortion and overlap. The 
1987 photographs of Hooper Island and South Marsh Island 
exclude a small fraction of the island. However, this 
omission is insignificant and the overall pattern of 
shoreline change is still discernable. 
59 
Data Analysis 
The digitized data were then run through the Space 
aesection program, another integral component of Metric 
Mapping (Leatherman and Clow, 1983). This program 
corrects for scaling differences among the maps and 
photographs, and determines if the data are distorted. 
space Resection is also used to overlay the various data 
sources onto a common grid system, which allows for 
comparison between data sources and years, and for the 
calculation of erosion rates. 
For the maps, Space Resection computes the scale ,, 
differences between maps and adjusts the digitized 
information accordingly? For the photographs, space 
Resection is much more complicated as it must adjust for ,, 
several potential sources of error, such as 1) scale I '!,1,  
I j 
differences in the photographs and the maps, and 2) ''  
distortion due to photograph angle, relief displacement i 
and flying height of the camera. The Space Resection 
program uses the secondary control points to 11tie 11 the 
photograph to the map, thereby pulling the photograph 
into place on the latitude-longitude grid system of the 
maps. 
After the data were digitized and Space Resected, 
the maps were merged together to make a complete file for 
each island which included all historical shoreline data. 
The ends of the digitized line segments were joined by 
60 
another program called the Tie Program. This program 
also joined line segments on adjacent photographs and 
maps, thereby melding each map or photo-mosaic. 
The complete file for each island was then plotted 
using the Plot Map Program. The resulting maps had 
several shorelines, each representing a particular time 
period. The final maps graphically demonstrated the 
spatial and temporal shoreline change for each island 
from about 1848 to 1987. 
All the data for each island were compiled, 
annotated, digitized, Space Resected, merged, tied 
together, and plotted. The erosion rates and net amounts 
of erosion were then calculated for each island using the 
Transect Program of Metric Mapping. The Transect Program 
is designed to calculate the rate of change and total 
amount of change between two historical shorelines at any 
desired location along the study shoreline. This is done 
by projecting transects across the shorelines (Figure 
3.4), from which the computer calculates the 
distance/time (i.e., erosion rate), as well as the net 
distance between any two shorelines. The program is 
designed for straight or gently curving shorelines, and 
problems arose in this study because the shorelines and 
island shapes are quite irregular. Additional problems 
arose where an island had broken into sections. 
Ordinarily, one spine is used which parallels the entire 
61 
Q) 
- - ?-u05  
:: :- C.....  
?, Q) 
1n 
:.. - Q) 
?~-- ::_ 3: 
0) 
C 
0 
ca 
0 Cl) 
f' -u Q) 
Cl) 
C 
0-.
c..a.  
u 
?1J  
LL Q) 
0. 
E 
ca 
X 
UJ 
,...  -"" .-...- -- ' 
62 
shoreline, from which perpendicular transects are 
projected across the historical shorelines. For the 
islands, it was necessary to run several small spines 
along relatively straight areas which were representative 
of the rest of the island in terms of shoreline change. 
This was necessary to prevent the transect lines from 
crossing one another and producing spurious results. 
The transect data were subjected to careful quality 
control before erosion rates were calculated. Some of 
the transects had to be omitted from the final analysis 
because they appeared to measure both the near and far 
shorelines and produced spurious results. Other 
transects were omitted because they were oriented at an 
angle to the parallel shorelines, and therefore 
overestimated the erosion rates and net amount of erosion 
(for example, see transect number 22 on Figure 3.4). 
Accuracy Assessment 
Because of the several data sources and the many 
steps involved in the Metric Mapping program, there are 
a number of potential sources of error which can be 
quantified to give confidence limits to the erosion rates 
and shoreline change maps developed from the program. 
Past shoreline mapping studies have successfully used the 
Metric Mapping program in Massachusetts, New York, New 
Jersey, Delaware, and Calvert County, Maryland (Crowell, 
63 
et al., 1991). These studies have shown that if care is 
taken to screen and correct the various data sources for 
error and distortion, and if raw data are computer 
corrected, then an accurate and reliable map product and 
erosion rate analysis can be obtained. In addition, in 
areas where the shoreline change is large, such as these 
islands, the associated measurement error will be small 
in comparison with this change, and erosion rates will be 
highly reliable. The actual magnitude of error is much 
less than the worst-case error estimates calculated by 
Crowell et al. (1991). 
In general there are two sources of error: error 
associated with the raw data and error associated with 
digitizing the raw data. The original raw data is prone 
to error due to distortion, and surveying or cartographic 
error. Digitizing errors are due to such things as 
digitizing the inner or outer margin of the mean high 
water line on the maps, digitizer error, and digitizer-
operator error. Crowell et al. (1991), quantified the 
worst-case error estimates for all types of data which 
can be applied to most historic mapping studies which use 
similar data sources. These estimates represent the root 
mean square of all possible sources of error. The worst-
case error estimate for a shoreline digitized from an NOS 
T-sheet dated prior to 1880 was calculated as ? 8. 9 
meters pl us sketching error (cartographer' s error in 
64 
creating the map). For a T-sheet dated between 1880 and 
1930, the worst-case error estimate is estimated at? 8.4 
meters plus sketching error. A recent T-sheet, dated 
after 1930, has a calculated error estimate of ? 6 .1 
meters plus inaccurate interpretation of the high water 
line. For aerial photographs, the worst-case error 
estimate using structural control is calculated to be? 
7.5 meters and? 7.7 meters if using geomorphic control, 
plus misinterpretation of the high water line. 
Using the same methodology and error estimates 
calculated by Crowell et al ( 1991), worst-case error 
estimates of the annual average erosion rate for each 
island in this study were calculated. The error estimate 
for each island was calculated as the sum of the error 
estimate for the oldest map and the most recent aerial 
photograph, divided by the number of years between the 
data. Because each island had a unique set of data, this 
calculation was done separately for each island. 
However, the maximum possible error was calculated to be 
+ 0.12 m for all long term erosion rates from the 1848 
data and 1987 data, for all islands. Other error 
estimates resulted from calculations derived from varying 
time spans. It is important to note that the error 
estimates for this study are considered to be 
conservative. 
65 
Field Surveys 
Field surveys were conducted to determine the 
composition and geomorphic characteristics of the 
islands. All islands were visited except for South Marsh 
Island. Near surface sediment samples were taken in 
appropriate locations on each island. Samples were taken 
to a depth of about 0.15 musing a shovel to obtain the 
sample after having removed the surface layer. Samples 
were taken of eroding cliff, eroding marsh, heal thy 
marsh, sandy pocket beaches, and sand spits. 
Subsurface samples were taken in marshes, marsh-
upland margins, and ridges on Hooper Island to the depth 
of the clay layer using a shovel ( Plate 3 .1). A 
transect at 7.62 m intervals and about 45.7 min total 
length was conducted from the crest of an upland ridge to 
the center of an adjacent marsh to determine the 
stratigraphic relationships at the marsh-upland border 
and the slope of the surface of the clay layer underlying 
the area. 
At Poplar Island, offshore samples were taken with 
a Van Veen grab sampler in a transect at about 100 m 
intervals offshore. The transect was extended to a point 
just outside the offshore limit of the island as it was 
mapped in 1848. This position was determined by locating 
the latitude-longitude coordinates of an offshore point 
from a map and then relocating the exact position using 
66 
Plate 3 .1 View of a pit dug on Hooper Island with 
silt/clay layer showing beneath a Phragmites 
marsh 
67 
a hand-held Global Positioning system (GPS). 
sediment samples were analyzed in the laboratory to 
determine the percent sand by weight. The samples were 
dried in the oven at 140 degrees celsius, and then 
weighed to obtain the total dry weight. The dried sample 
was then defloculated with Calgon to break up the silt 
and clay particles and rinsed through a 4 phi sieve to 
retain the entire sand fraction. The sediment retained 
in the sieve was then dried and weighed to determine the 
percentage sand in the sample. 
A transit and rod were used to take beach profiles 
and to determine the present day elevational 
characteristics of the islands relative to the water 
level. The time of day was recorded and used later to 
determine the approximate tidal elevation at the time the 
measurements were taken. The measurements were then 
reduced to the common datum, National Geodetic Vertical 
oatwn (NGVD), by extrapolating from the Baltimore tide 
tables. Subsequent elevations were then recorded where 
it was determined to be useful: marsh edge, upland 
margin, upland, top and base of an eroding edge, storm 
wrack lines on the marsh. 
The geomorphological character of the island was 
noted on each field visit, including the presence and 
location of eroding scarps, eroding marsh, stable marsh, 
pocket beaches, sediment composition, and the condition 
68 
and composition of vegetative and wildlife communities. 
Data Analysis 
Sea-Level Change Analysis 
The Baltimore, Maryland tidal record was used as a 
measure of the change in sea level because it has the 
longest record of sea-level change for the Chesapeake 
Bay, dating back to 1903. As-year running mean was used 
to smooth the Baltimore record and to reduce the large 
interannual variation which is typical of mean sea-level 
records. This variation can be caused by storms and 
other climatic and astronomical factors (Figure 3.5). 
Comparisons of the rate of sea-level change at 
Baltimore were made with other tide stations around the 
Bay including Annapolis, Solomons, Washington, D.C., and 
Kiptopeke (Figure 3.6). The rate of sea-level rise has 
been slightly higher at the four other stations, but the 
Baltimore record is much longer than the other stations 
so it was used as a more conservative estimate of sea-
level change in the Bay during most of the 20th Century. 
The Hampton Roads, Virginia station also has a 
fairly long record dating to 1927, but the area has been 
experiencing a relatively large amount of subsidence 
(Holdahl and Morrison, 1974; Davis, 1987). Therefore, 
69 
Sea-Level Rise 
Baltimore, Maryland 
Mea?n Sea Level (ft) 
Sr------------------------~~ 
4.51 ? ...... ?-?? ?-
s1ntllesized uata 
..J 
0 41 ;:;;J-~~-:????? uat3 
p.ct~al . 
3.5 1--- ------ -???---??? ..... ??-??-??-???- .. -------
3 111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111111 
1850 1870 1890 1910 1930 1950 1970 1990 
Year 
-El-- Annual Mean - 5-Year Mean 
Figure 3.5 Baltimore, Maryland tide gauge record, with five-year 
running mean 
......... ,.1.011 ? .... rl ? tJ 
'f ? 1 Ob ? OL r l ? ,a 
I 
? 4 1 4 l 4 l 4 ? 4 S 4, ,. 7 ,. I , t 
. ~' ~'r 
: ;I ~ ::-~. 
i:; 'r -~ .&;:.' frl 
,, 
U?i---....;.'-"'- ?"-??. ;.;.;!l.__;_el_.~ IS----, . :i'  ., 
j ? I j 
J 9 
Kiptopeke 
Figure 3.6 A comparison between Baltimore tide record and 
four other tide gauge stations in the Chesapeake 
Bay, showing station locations 
11 
the Hampton Roads tidal record is unrepresentative of the 
Bay. 
It was also observed that a strong correlation 
exists between the rates of sea-level change at Baltimore 
and New York (r=+.98) from 1903 to 1986 (Figure 3.7). 
This is useful as the New York record extends back to 
1856. As a result, the regression equation from the New 
York-Baltimore analysis was used to predict sea level at 
Baltimore between 1856 and 1902. 
It is important to note that different processes are 
affecting sea level at the New York and Baltimore 
stations. The Baltimore station, near the head of a 
large estuary, is affected by various processes such as 
subsidence. The New York station, in a more open ocean 
environment, is being affected by neotectonic activity 
(Emery and Aubrey, 1991). However, there is a strong 
agreement between the rates of change between the two 
stations which gives confidence to hindcasting the record 
at Baltimore from the New York data. 
Areal Analysis 
A planimeter was used to determine the size of the 
islands in hectares for every shoreline year for all 
islands. Each year-interval was measured three times and 
the average of the measurements taken. The planimeter 
72 
Figure 3. 7 Regression analysis between the Baltimore, 
Maryland and New York, New York tide gauge 
stations 
y ? 1.01x - 1.04, r 2 ... 97 
5 
4.8 
4.6 
Iii 
.'s5 - 4.4 ii 
00 
4.2 
4 
3.8 
4.8 5 5.2 5.4 5.6 5.8 6 
New York 
0 Annual mean sea level 
73 
error was calculated to be between? 3.29 ha and? 19.74 
ha, based on the averages of three measurements. Because 
the island shorelines are so irregular and the historic 
shoreline change on these islands exhibited such spatial 
variability, the rate of land loss in ha/yr is a more 
meaningful assessment of historical land loss than is an 
erosion rate in m/yr. The sizes of the islands were 
then plotted against time to assess the trend of land 
loss for each island. These data were also used to 
determine land loss rates and percentages, and to 
correlate sea-level rise with land loss. 
Land Loss vs. Sea-Level Rise Analysis 
The rates of island land loss in the Bay were 
correlated with the rates of sea-level change at 
Baltimore during concurrent time periods to determine if 
any relationship existed between the rate of sea-level 
rise and the rate of perimeter land loss. The time spans 
which were used for comparisons were determined by the 
map and photograph data used for digitizing (Figure 3. 8). 
The New York tide gauge data was used to hindcast the 
Baltimore record to 1856, so that the earlier land loss 
data could be used. There is a gap between the earliest 
land data of 1848 and the earliest sea-level data of 1858 
synthesized from the New York data. This synthesized 
74 
Figure J.B Sea-Level Rise 
Baltimore, Maryland 
Mean Sea Level (ft) 
5r-----------------------------.. 
4.5 1----???----- ???? -- . -- .. 
s"Jl'tl\esiseel oata 
..J 
U1 41 -- ~ ::_:? 
3.5 I ? ?? ??????????~ ? ?? ?- ??? ? - ? ?? 
3 111111J,111U" "I JII m I , J mI I I" I 
1850 1870 1890 1910 1930 1950 1970 1990 
Year 
--a- Annual Mean - 5-Year Mean 
record is the best data available and it is assumed that 
the general trend did not change during the 10 years of 
missing data. 
Forecast Modeling 
Introduction 
In order to predict how these islands will respond 
in the future with or without an acceleration in the rate 
of sea-level rise, three models were examined which 
predict shoreline response to sea-level rise. The three 
models include the Bruun Rule, Inundation Model, and 
Historic Trends Analysis. 
The Bruun Rule 
The Bruun Rule was omitted from consideration for 
several reasons. The Bruun Rule which applies to sandy 
beaches and nearshore areas (Bruun, 1962), does not fit 
the eastern shore Chesapeake Bay environment. Indeed, 
the island shorelines generally fall into two categories, 
neither of which are appropriate for Bruun Rule 
calculations: marsh edge and eroding clay cliff. In 
addition, the Bruun Rule is invalid because the island 
shorelines have been erosional features since the 
Holocene and have not been in an equilibrium state for 
76 
thousands of years, if ever . The Bruun Rule loses 
physical meaning along marsh shores because the flora 
controls both vertical and horizontal shoreline moveme nt 
of the marsh (Rosen , 1978). According to Hands (1983), 
the Bruun Rule predicts rapid and permanent erosion for 
shorelines comprised of fine sediment such as silt and 
clay because the sediment is suspended in the water 
column rather than placed off shore, and is therefore lost 
from the equilibrium profile. 
The Inundation Model 
The inundation model, also called the "drowned 
valley concept", uses the existing topography and 
bathymetry of a coastal area to model shoreline response 
to future sea-level rise. For this model, shore slope is 
the most important variable, because it will determine 
the amount of horizontal displacement that an area will 
experience (Figure 3.9) (Leatherman, 1991). Shore 
profiles and slopes were determined by field surveys. 
Topographic charts were not used because the resolution 
of these maps is too low to be useful. 
Historic Trends Analysis 
The historic trends analysis was used to calibrate 
historic erosion trends with respect to sea-level rise. 
This model accounts for the natural variability of 
77 
.... .... -+-
FUTURE SEA LEVEL SHORE LINE MOVEMENT 
PRESENT SEA LEVEL "v C?NTL? SLOP?S 
-..l 
CD 
;I 
Figure 3.9 Conceptual diagram of the inundation model (from 
Leatherman, 1990a) 
shorelines to respond to sea-level changes due to coastal 
processes, sediment types, and energy conditions. The 
underlying assumption behind the historic trends analysis 
is that shorelines will respond in similar ways in the 
future as they have in the past, since sea-level rise is 
the main variable and all other parameters remain 
essentially the same (Leatherman, 1984). The historic 
rate of shoreline change was determined using the Metric 
Mapping procedure. The Balt irnore tide record was used to 
establish a rate of historic sea-level change for the 
study area, as previously described. Future rates of 
shoreline change were based on these two variables. 
Future rates of sea-level rise were taken f rorn the 
Intergovernmental Panel on Climate Change ( 1990) 
scenarios, which were calibrated to the higher rate of 
sea-level rise which has been occurring in the Chesapeake 
Bay (Figure 2.4). 
79 
CHAP~ER 4: RESULTS 
Historical Shoreline Mapping 
Introduction 
The results of the historical shoreline mapping are 
presented in Figures 4.1 through 4.7. Two very distinct 
patterns of land loss are immediately apparent from the 
analysis. The islands were thus divided into the 
"Northern Group" and the "Southern Group", according to 
geographic location, geomorphic conditions and patterns 
of shoreline change. The Northern Group consisted of 
Barren Island, James Island and Poplar Island; the 
Southern Group consisted of Bloodsworth Island, Smith 
Island and South Marsh Island. 
Although the geomorphology and general pattern of 
land loss on Hooper Island fits that of the Southern 
Group, it is excluded from either Group because much of 
the shoreline has been protected with engineering 
structures so the natural processes of land loss are 
obscured. Despite shore protection along most of the 
island, Hooper Island has been reduced in size by 25% 
since 1848, at an average rate of 2.9 ha/yr (Figure 4.7). 
Much of this loss occurred between 1848 and 1952. The 
rate of land loss has slowed since 1952, presumably 
because shoreline protection structures were built around 
this time. Hooper and South Marsh Islands had similar 
80 
Figure 4.i HISTORICAL SHORELINE CHANGE. BARREN ISLAND 
g1 
01 + 
~ 
CP 
I-' 
I I : I 
I I 
j 
1 
I ~
o! -- 19 
gl -- 1943 + 1-- 191M 
~ ' :_ _ 19117 
MD 
eg_t_ 1 Pate: 01/30/92 
.... ~-~ ?--------------~-~- 
Figure 4 . 2 HISTORICAL SHORELINE CHANGE, JAMES ISLAND 
sr?--?- - ----- - - - - ~QlL_ A __ __ _!l92QQO _ _ __ - ----
i i + + le; -7 
I 
I 
CD 
I\) 
O ' 
o
.~...-' ,  ... + 
m; 
ft== : I 1941 I 
,....s__.c_a_l_e ____l - -,--,-----1---1_  __._, __I _.,..._;;--_s_ __p _ __zo _n_e_ _5_ 0_: -M?------~ 
1: 2400.Q_ _ O Feet 7000 ~ Frame o'-1.!!!t!er: 1. Date:_ _ __Q_WQ/ ~.? __ _____ J 
, 
r~- Figure 4.3 Historical Shoreline Change, Poplar Island li4ll.QO. 
+ 
1 
00 
l,J 
i 
c l 
gl 
~1 + 
111-48 
1899 
1941 
1984 
1~, 1987 l~cale r---- I f I I I I I S.P. Zone 50: MD I 
000 0 Feet 7000~ Frame numpec.;__ 1 Date: Q.1LW92 __ ---- - - ~ . 
r----?- _____ .. __ _ I ! 
Cl I 
!I ~ 
en ? 
H c=, 
I 
l-
a: ~ 
a ~I 
e3n:  I I 
Cl I 
g ? ~ ~I 
fi1 OIO .. _J t.1 
CD ~ 1111 
I cu .Q 
C: E 
u.i 0 :J 
t N z!J  
< 
I 
u 
UJ 
z 
H 
_J 
UJ 
a: 
0 
I 
en 
_J 
4 
u 
H 
a: 
0 
I-
en 
H 
I 
"q' 
,q-' 
Q,J 
c.. 
:, 
Cl 
?r1 
u. 
I 0 
L_ __ ,_,-- ------ cu 0 .... 0 - 000"'03,i!i.--- -- _ ~~ 
84 
Figure 4.5 HISTORICAL SHORELINE CHANGE, SMITH ISLAND 
r-- ----- --10ZtillD.Q. ____ _ ~--- - ____ _ _ .110.ll.00.12 . - -- - --?-----, ~~ . 
I 
; 
- + j I I 
\ I 
I ! 
\ 
'I  
ex, 
Ul'. \ 
i 
! 
I 
o? I 
+ 
~ I I 
I I 
\ \ 
I 
\ 
i ..fa 9' \ ; 
I ' 
l i 
I 
I 
\ 
I l t1----84!1 1 I - !~1 I L r::= 
~le t --+------i- - -t------i l, ~ s . p . Zone 50: MO -- - j 
li;~ _ _,_. _____ __ Fe@.t. 25000 \""ET Er~!lYm!fil'~. _1 _ ...Q!t_e.: 01/30/92~-- -- -- --
Figure 4.6 HISTORICAL SHORELINE CHANGE. SOUTH MARSH ISLAND 
ipz2000 -----?------ 1oasaoo_ ___ ------~ 
( + + 
1 
I 
I 
I 
OJ 
CJ'\ I 
I 
J + 
~ 
I 11"49 
1901 
~_ 19-42 
- - 1987 
!scale 1--t ~ S.P . Zane 50: MO 
~~_q _ ___ o__ Feet 15000 ~ Frame nulllber: ___ j _ Dat~:_ 01L~9a ___ --
Figure 4.7 HISTORICAL SHORELINE CHANGE, HOOPER ISLAND 
,---- - -- ---- .?1 02QQQQ._ __ - - - -? --- ?? - - ?? ?- - -- _ .1QSQQ.QQ. ?- --?- --?-1 
I I 
I I i 
\ 
g\ 
!oI,  + 
I 
l 
CD \ \ 
....J 
\ 
i 
\ \ 
\ 
\ 
I \ 
I 
+ \ 
1\ I t_ ifMB 
\ t"_ 19(11 
19'12 
1952 
\scale !----+----+ --:-ii ~ _ 1987 s.P. zone 50: Maryland ; 
i 1: 100000 ___ _ Q_ _ -- _ _f..ttt aoooo l'.::'.I" Frame number: __ 1 L_DM_g_;_ ___0 ..1.L;l.Q..:~2---- - - ?- -
rates of land loss until 1952, when the rate of land loss 
on Hooper Island slowed slightly (Figure 4.9). 
All the islands in the study area are losing l and 
rapidly. Tables 4.1 and 4.2 present the historic land 
loss for the Northern Group and Southern Group and the 
rates of land loss. Figures 4.8 and 4.9 show the r a tes 
of land loss during the study period. 
Northern Group 
Patterns of Land Loss 
The Northern Group of isl ands are all similar 
geologically and geomorphologically. They are dominated 
by thick, upland Loblolly and Virginia Pine forests with 
fringing Spartina patens and~- alterni flora marshes in 
some areas ( Plate 4. 1). The islands are fairly low 
lying; the highest elevation measured during field 
surveys was about 2 m above mean sea level. The marsh 
areas are less than about o. s m above mean sea level 
(Table 4.3). 
The Northern Group showed dramatic loss of land from 
the north and west, with very little change on the 
protected, eastern side of each island (Figures 4.1, 4.2, 
and 4.3). Each island has a steep (45 to 90 degree), 
eroding clay bank which varies in height from 1 to 2 m on 
88 
Table 4.1. 
HISTORIC ISLAND LAND LOSS IN THE CHESAPEAKE BAY: 
NORTHERN GROUP 
BARREN ISLAND 
Average 
Year Hectares % Reduction Rate (ha/yr) 
1848 306 
1901 217 30 1. 7 
1929 176 20 1.5 
1943 153 13 1.6 
1964 107 30 2.1 
1987 75 30 1.4 
Total Lost: 231 76% 1.7 
JAMES ISLAND 
Average 
Year Hectares % Reduction Rate (ha/yr) 
1848 398 
1901 230 43 3.2 
1941 137 40 2.3 
1952 133 3 0.4 
1964 95 29 2.4 
1987 45 53 2.2 
Total lost : 353 89% 2.1 
POPLAR ISLAND 
Average 
Year Hectares % Reduction Rate (ha/yr) 
1848 343 
1899 209 39 2.6 
1941 127 39 1.9 
1952 103 19 2.2 
1964 79 24 2.0 
1987 43 46 1.6 
Total Lost: 300 88% 2.0 
89 
Table 4.2 
HISTORIC ISLAND LAND LOSS IN THE CHESAPEAKE BAY: 
SOUTHERN GROUP 
BLOODSWORTH 
Average 
Year Hectares % Reduction Rate (ha/yr) 
1848 2,280 
1901 2,111 8 3.2 
1942 2,066 2 1.1 
1952 2,003 4 6.3 
1987 1,909 5 2.7 
Total Lost: 371 16% 3.3 
SMITH 
Average 
Year Hectares % Reduction Rate (ha/yr) 
1849 4,467 
1901 3,737 16 13.8 
1987 3,168 15 6.6 
Total Lost: 1,299 29% 10.2 
SOUTH MARSH 
Average 
Year Hectares % Reduction Rate (ha/yr) 
1849 1,538 
1901 1,336 13 3.9 
1942 1,285 4 1.2 
1952 1,238 4 4.7 
1987 1,113 10 3.6 
Total Lost: 425 28% 3.3 
90 
Figure 4.8 Island Land Loss 
in the Chesapeake Bay 
(Northern Group) 
Hectares 
500 ~-- - - -?- - - --- ? . ~ ?---~ - - - "" 
400 
IC 
~ 300 -
200\ - . - . ~~ '~  
--~~I \ 
100 
0 
1850 1900 1950 1987 
Vear 
I- ?- James + Bar~en * ~oplar__\ 
Figure 4.9 Island Land Loss 
in the Chesapeake Bay 
(Southern Group) 
------------------------------------
Hectares (Thousands) 
5.-------------------------, 
4 
10 
I:\) 3 
??---------1?1---- -
2 ?-----?-~--,--------- ?,.  
1 
0 ..___.____,__..._~~~--------'----'---'----'----L---L--'---J 
1850 1900 1950 1987 
Year 
. --- Bloodsworth * Hooper -?- Smith * South Marsh 
Plate 4 .1 Mixed upland and wetland habitat on Coaches 
Island (Poplar Island) 
Plate 4. 2 Clay bluff and dead trees on Poplar Island 
which is typical of the Northern Group 
islands. 
93 
Table 4.3. Elevational Characteristics of the 
Northern Group Islands* 
ELEVATION (m) ** 
Island Marsh Margin Upland 
Barren .25, .39, .52, .30, .53, .37, 
.15 .39 1.0 
James .11 .s0 .75, 1.3, 
.86, .57 
Poplar .48, .so *** 2.3, 1.3 
* Elevation above p resent day  mean water IeveT. ** Each elevation represents a measurement taken in 
the field and is referenced to present day mean 
sea level, as extrapolated from the Baltimore 
tide tables. 
*** No measurements taken in upland/marsh margin 
areas. 
the western shore of the island (Plate 4.2). The rate 
of shoreline recession on the eastern side of the 
islands is considerably lower with only a small amount 
of shoreline recession. A fetch analysis for each 
island (Table 4.4), presents the variations in fetch 
length which can alter wind and wave patterns along a 
shoreline resulting in variable shoreline response. 
Rates of Land Loss 
All the islands in the Northern Group have been 
reduced in size by more than 76% since 1848 ( Table 
94 
4.1), all currently being less than 100 ha. Barren, 
James and Poplar Islands have lost 76%, 89% and 88%, 
respectively. The mean rates of loss between 1848 and 
1987 have been 1.7 ha/yr, 2.1 ha/yr, and 1.9 ha/yr, 
respectively. The rates of land loss tend to vary 
during different periods, but the long-term rate has 
remained relatively constant ( Figure 4. 8) ? The rate of 
land loss for the Northern Islands does not appear to 
be directly correlated to the rate of sea-level rise 
during the same periods (r= +.04) (Figure 4.10). This 
result is not unexpected as sea-level rise per se does 
not cause erosion. Sea-level rise exacerbates the 
effects of waves and storms by allowing larger, higher 
energy waves to reach the shore. Thus, sea-level rise 
is the underlying driver of shoreline change caused by 
wave action. 
Rates of Erosion 
The overall averaged annual rates of erosion of 
the island shorelines for each time span are presented 
in Table 4.5. Annual average rates of erosion for the 
western side of Barren, James and Poplar Islands are 
4.38 ? 0.12 m/yr, 6.52 ? 0.12 m/yr, and 3.99 ? 0.12 
m/yr, respectively. These figures are in sharp 
contrast to the annual erosion rates on the islands' 
95 
Table 4.4: Fetch Analysis 
for 7 Islands in the Chesapeake Bay 
Approximate Distance (km) and Direction 
I Island I N NE E SE s SW w NW 
Barren 2.7 1.8 1.8 1.8 143.2 15. 7 . 12.9 20.3 
-I) 
0\ 
Bloodsworth 1.8- 8.3 6.4 6.4 2.7 24.0 22.2 6.0-
7.4 36.1 
Hooper 2.7- 2.7- 1.8 - 1.8 - 131. 9 36.1 13.8 - 1.8 -
7.4 5.5 6.0 7.4 18.5 7.4 
James 17.5 3.7 - 6.4 3.7 1.8 12 14.8 20.3 
12.0 
Poplar 7.4 4.0 2.7 2.7 - 41.6 - 18.5 15.7 16.6 
3.7 60 .1 
Smith 4.0 12.0 9.2 9.2 12.9 20.3 30.0 33.3 
South Marsh 2.7 6.0 6.0 - 13.8 4.0 30.0 25.0 5.0 
12.0 
Table 4.5 
&roaion rat?? for the Rorthern Group 
(in m/yr) 
BARRER 1849 - 1901 - 1929 - 1943 - 1964 - TOTAL 
1901 1929 1943 1964 1981 AVERAGE 
E~st * * * * * * 
West 3.86 4.21 3.16 4.91 4.91 4.38 
Error range ? 0.33 :t 0.60 ? 0.81 :!: 0. 64 :!: 0. 60 ? 0 . 12 
JAMES 1848 - 1901 - 1941 - 1952 - 1964- TOTAL 
?1901 1941 1952 1964 1981 AVERAGE 
'.?.J  
East 0.15 0.42 0.36 1.43 0.13 0.56 
West 6.50 6.15 4.03 8.13 1.81 6.52 
North 5.82 6.04 * * 19.15 9.59 
Error range :t 0.32 :t 0.36 :t 1. 25 :!: 1.25 :t 0.66 ? 0.12 
POPLN\ 1848 - 1899 - 1941 - 1964 - TOTAL 
1899 1941 1964 1981 AVERAGE 
East 0.99 0.41 1.03 * 0.81 
West 3.80 4.13 6.51 2.36 3.99 
Error range ? 0.33 ? 0.34 :!: 0 .59 ? 0.66 :t 0.12 
* Insufficient data to calculate erosion rates 
Y ? -.Z1x + 2.66, r Z ? .04 
3.4 ,-_..__  __,_ _~  _ __._ _ ___. _______  
3.2 
3 
......_ 2.8 
~
-"' 
 2.6 0 
.r:. 
~ 2.4 
0 )( 
~ 2. 2 ...._;___ 
~ 2 --------.ll,_,LI 
0 
1.8 i ? 
1.6 
1.4----?- -----......----,.--~------J. 
2 2.2 2.4 2.6 2.8 3 3.2 3.4 ?, 
Sea-level rise (mm/yr) 
O Poplar- )( J::1r:ies ? Bat"t"en 
Figure 4.10 A comparison between the rate of sea-level 
rise and the rate of land loss for the Northern 
Group 
98 
eastern shores, which are o. 5 6 ? o. 12 m/yr for James 
Island and 0.81 ? 0.12 m/yr for Poplar Island. These 
numbers are the average of all the transects included in 
the analysis. The highest rates of erosion are on the 
north shore of James Island. 
Sediment Analysis 
The location of sediment samples and surveys 
performed during fieldwork are presented in Figures 4.11, 
4 ? l 2 , and 4 ? 13 ? Sediment analysis revealed that the 
islands are composed of silt and clay and contain very 
little sand. Grain size analysis of a sediment sample 
from Poplar Island demonstrated that all the sand in the 
samples was greater than 2 phi, meaning that it is fine 
to very fine according to the Wentworth Classification of 
grain sizes. In fact, 95% of the sand was greater than 
3 phi, meaning it is very fine. 
Poplar Island had 13.6% and 17.6% sand by weight in 
two samples analyzed. Barren Island had 7.9% and 3.7% 
sand in two samples. James Island had 3.3% and 2.3% sand 
in two samples analyzed. Clearly, there is some 
variability in the percent sand found among islands, 
although there is a relatively strong agreement within 
each island. Other samples collected during fieldwork 
were examined visually and texturally and were determined 
to be similar to the 6 analyzed samples. 
99 
i..  
~ 
.... 
'M e"'  0 
C1) -? 
"' 
r--- -"0  Gi ., +J ,rl .., ID 0 
tr')  ...,:  -e..  -0.".' ...II   C ... ~ ti) :c X 
'O 
C 0 0 ro . ~ 'G-I 
r--i .Q 
(/) ~ CJ Q) E 
H ... C :, 
+ 0 C: N 
C 
Q.) + i 
c.. a. Cl t. 
c.. d U1 ... 
l1J 
::c < ri i L 
..0._  ,, 
0 
(/) 0 0 
C I'-
0 
?r"i 
,? 
n:, 
u 
,.0... . 
-r., ..., a. 
E Ill 
l1J u.. 
en 
,rl 
,rl 1 
... e 
(l) 
c.. 0 
:::, 
.C.....l  
Li.. 
0 
0009L.> 
o ovet 
100 
Figure 4.12 Sample locations for James Island - 5/15,.91 
~ Hanh 
~ Sand 
(Y - ~ ? Uplaod 
Ci) Sa?ple locations 
/;"\ '1:?_:- \ 
.... ~ ""\;:.;J 0 
0 
.0.. . .0,  ? ? 
(?\I  L 
Al 
~ \'/?.J'.,.J.?   ID 
- - . 1987 
I~ S.P. Zono 50: MD 
?g ?OOQ ~ fram11_n11,nher: f QatP.: 01 / :i0 -' !=12 
I 
Figure 4.13 Sample locations for Poplar Island - 5/1/91 
~ 
~ + ~ Harsh 
t( 
.'Q ?. Upland ? )1 1 ----0) . 
? , ?~--::. (- ? (1) Sample locations 
. :?\ 
.. ;\ 
0 \) 
I-' 
0 
N 
~ 
C 
C~  CD- - -----.. i .,--.:?~:-. + ~:-.-? .:'_-..- ::-- + 
0 ~?? ? ?? ?? ???? 
: :~_?:-?:-> 
. 1987 
Scale I I I I I I I I 1??~- S.fJ. Zone 50: MU 
\ i: 24000 0 F~et 7000 I_ f Frame number: 1 Date: 01,'30/9? 
Sometime between 1964 and 1987 a sand bridge 
developed between the northern and southern sections of 
James Island (Plate 4.3). A ver:y small spit is visible 
on the northern section of the island in the 1964 
photograph, but a complete bridge between the two island 
halves is clearly visible in the 1987 photograph. The 
spit is currently about 35 m wide and 2,100 m long. 
Sediment analysis revealed that the composition of the 
spit is more than 96% sand. The middle section is 
dominated by a Spartina marsh with no peat development 
beneath the marsh plants. This spit possibly developed 
over the years as a lag deposit of sand which remained in 
the nearshore area as the island eroded and the fine-
grained silt/clay was carried into suspension by waves 
and currents, away from the island. The sand deposit has 
been subsequently shaped by longshore currents to form 
the spit which exists today. 
Southern Group 
Patterns of Land Loss 
The Southern Group, consisting of Bloodsworth, Smith 
and South Marsh Islands, were grouped together based on 
their similar geomorphology, shoreline response pattern 
and relative geographic location in the southern section 
of the study area. Geomorphologically distinct from the 
103 
Plate 4.3 Sand bridge on James Island, looking south 
- - ??? 
Plate 4.4 View of an upland ridge on Smith Island . 
104 
Northern Group, these are large, marshy islands with 
general elevations less than about 0.5 m above msl. All 
of these islands are over 1,000 ha (Table 4.2). 
Smith Island has several linear ridges running 
approximately North-South with upland vegetation. 
Bloodsworth has one ridge, known as Fin Creek Ridge. 
Maps and photographs of South Marsh Island do not show 
any upland ridges. According to measurements for this 
study and topographic surveys, the ridges lie between 0.5 
and 1.5 m above MSL. The subtle elevational differences 
between marsh and ridges define the landscape on these 
low-lying islands. Even small increases in elevation are 
enough to support upland vegetation ( Plate 4. 4) ? on 
Smith Island, the largest ridges host the island's three 
towns: Ewell, Rhodes Point and Tylerton (Plate 4.5). The 
Ewell, Maryland-Virginia USGS topographic quadrangle, 
dated 1968, indicates that small areas of these ridges 
reach 1. 6 m above MSL. This survey uses NAD 27 data, 
however, which would overestimate the present elevation 
of the islands since sea level in the Chesapeake has 
risen about 0.2 m since 1927. However, the majority of 
these ridges are below the 5 foot (1.6 m) contour and lie 
almost imperceptibly above the surrounding marsh. The 
distinct vegetation on the ridges causes them to stand 
out above the flat marsh surface. 
The Southern Group demonstrated a very different 
105 
Plate 4.5 View of Rhodes Point on a ridge in the 
distance, one of the towns on Smith Island 
Plate 4.6 Interior marsh ponding on Smith Island 
106 
pattern of land loss than the Northern Group. Since 
1848, these islands have had a more uniform pattern of 
land loss around their perimeters (Figures 4.4, 4.5, and 
4.6). The Southern Group also experienced land loss in 
terms of interior ponding of open marsh areas and apical 
erosion of tidal creeks (Plate 4.6). This process was 
clearly visible from observations of the maps and aerial 
photographs, al though it was not quantified in this 
study. Because the methodology in the present study does 
not account for internal ponding and marsh loss, the land 
losses reported for the Southern Group are under-
estimated. A detailed examination of internal marsh loss 
shows a dramatic increase in open water area since the 
turn of the century on Smith Island (Davison, 1990) 
(Figure 4.14). Similar analyses are unavailable for the 
other islands. 
Rates of Land Loss 
The Southern Group has been losing land at higher 
rates than the Northern Group (Table 4.2). However, they 
have lost smaller percentages of land since they are all 
larger than the Northern Group. Bloodsworth, Smith, and 
South Marsh Islands have lost 16%, 29%, and 28% of their 
land area since 1848, at rates of 3.3 ha/yr, 10.2 ha/yr, 
and 3. 3 ha/yr, respectively. As with the Northern Group, 
the trend in the rate of loss is fairly constant over 
107 
Figure 4.14 Historical change in the percent of total open water 
in four quadrants of Smith Island: Terrapin Sand 
Point, Kedges Straits, great Fox Island, and Ewell 
(from Davison, 1990) 
Percentage of To tol Open Water, % 
35?. -
a 
30?. ;,,, 
25?. _.,.~ ....... ???????????? ??* 
~ 
0 
CX> 20?. 
.. -? ,.,,~ ., 
15?. 
r _.. ?:::? ?.:.:??.::? --- -* "' / 1 O?. 
.. --
5~ 
O?. 'I [ [ [ [ I [ I I I f-1--LL I I I I I I 1--L.l....Ll..-f ..l. I I I I I I I I I I I I I I I I I ? I I I I I I 11 
1849 1864 1879 1894 1906 1921 19.36 1951 1966 1975 1982 
Time 
- Terrapin -+- Kedges ?????? Great fox ? -0- Ewell 
time (Figure 4.9). 
The r ate of land loss on Smith Island is much higher 
than the other islands. Since 1849, there has b een a 
significant amount of perimeter erosion along the western 
shore and from Terrapin Sand Point in the northeast 
corner of the island (Figure 4.5). Thus, the pattern of 
erosion resembles that of the Northern Group. However, 
due to the geomorphic similarities between Smith and the 
two other southern islands, Smith remains in the Southern 
Group. Smith has also been experiencing interior marsh 
loss, which is characteristic of the Southern Group. 
The rate of land loss for Bloodsworth and South 
Marsh Islands appears to be weakly correlated to the rate 
of sea-level rise during the same time periods (r= +.84) 
(Figure 4. 15) ? The rate of sea-level rise was calculated 
as the difference in sea level divided by the number of 
years between measurements, using the actual and 
synthesized tide gauge data at Baltimore. The trend for 
Smith Island does not fit into either the Northern or 
Southern Group, due to the anomalous rates of land loss. 
Although this correlation is not very strong due to a 
small data set (p = .04), there does appear to be a 
relationship between the rate of sea-level rise and land 
loss for the Southern Group but not for the Northern 
Group. Clearly, more land loss data from other Southern 
Group type of islands is needed to strengthen this 
109 
Figure 4.15 A comparison between the rate of sea-level 
rise and the rate of land loss for the Southern 
Group 
y ? 1.7x - 1.76, r 2 - .7 
4~-~----__,..___.._____,..______._ ________ __ 
3.5 
.; 
3 ? So?u th Macsh 
I-' ~ 
I-' 
0 "rti 2.5 
...r_:., , 
U) 0 Blood:;wocth 
~ 2 
~ 
C: 
~ 
-1 1 .s 
1 ~ 
1 .8 2 2.2 2.4 i.6 2.8 3 3.2 3.4 3.6 
Sea-level rise (mm/yr) 
relationship. 
Rates of Erosion 
Erosion rates for the Southern Group were more 
difficult to obtain due to the irregularity of the 
shorelines. Therefore, erosion rates were not calculated 
for the entire shoreline of these islands. Instead, 
erosion rates were determined for relatively straight 
segments of the shorelines in order to get a 
representative idea of the rate of erosion for each 
island. For example, three transects were run for 
Bloodsworth Island from which erosion rates were 
calculated (Figure 4.16). Average annual erosion rates 
from these transects were 1.19 ? 0.12 m/yr, 1.67 ? 0.12 
m/yr and 1.24 ? 0.44 m/yr (Table 4.6). For Smith Island, 
two transects produced erosion rates of 2.64 ? 0.12 m/yr, 
and 0.47 ? 0.12 m/yr (Figure 4.17). Two transects were 
run on segments of the South Marsh Island shoreline, 
producing erosion rates of 1.18 ? 0.12 m/yr and 0.55 ? 
0.12 m/yr (Figure 4.18). 
Sediment Analysis 
Because the Southern Group islands are geomor-
phologically similar, it is likely that the clay layer 
which was identified under Bloodsworth Island extends 
111 
r- -.F igure 4.16 Location of transects for determining erosion rates - .... - . . - .. - . -- . . . - . - -.. . --on Bloodsworth Island 
! Spinet2 
+ ~ Spine #1 
........  
"' 
~ 
Spine#J 
!Scala 
?9. 
Table 4.6 
Erosion rates for the Southern Group 
( in m/yr) 
BLOODSWORTH 1848 - 1901 - 1942 - 1952 - TOTAL 
1901 1942 1952 1987 AVERAGE 
Transect 1 * * * 1.19 1.19 
Transect 2 0.88 1.58 1. 61 2.62 1. 67 
Transect 3 0.63 0.38 3.65 0.83 1.24 
Error rates ? 0.32 ? 0.35 ? 1. 38 ? 0.44 ? 0.12 
SMITH 1849 - 1901 - TOTAL 
1901 1987 AVERAGE 
Transect 1 2.83 2. ,15 2.64 
I-' 
I-' 
w Transect 2 0.31 .64 .47 
Error rates ? 0.33 ? 0.18 ? 0.12 
SOUTH MARSH 1849 - 1901 - 1942 - TOTAL 
1901 1942 1987 AVERAGE 
Transect 1 1.00 0.72 1. 83 1.18 
Transect 2 0.65 0.21 0.79 0.55 
Error rates ? 0.33 ? 0.35 ? 0.30 ? 0.12 
* Insufficient data to calculate erosion rates 
Figure 4.17 Location of transects for determining erosion rates 
.. -?-- -- . --- -- -?- ?---?-? ? ?- ? - -- . - ?---? --- - --? .. . - ?-
on Smith Island 
Spine #1 
? 
\ 
I 
+ 
Spine #2 
........  ? I ~ l 
\ 
I 
+ + 
IQ _.  . ? ? i I I, -~ I ---i 50: MD 
l'Hj;. 
-F-igure 4.18 ?- Location of transects for determining erosion rates -. . --- - -- ---- - --- ------ -- ?? ?-?--? - ---- --------~ 
on South Marsh Island 
+ 
........  Spine #2 
U1 
Spine #1 
+ + 
'.Scelo i ~ -=-?+=--t-- ?-+----f-----=.i??1??~; S.P. Zone 50: Maryland 
it: 50000_ __0 _ ___ __ .. f@@t_ ? _ _. ____ 1~Q90 _l ~ F~--~~ _nJM!l12tr_; _ J, __ _gt_t_~__;__ QUQ4/ tl2 
south and underl i es both South Marsh and Smith I slands 
(Figure 4.19) (GEO-RECON, 1980). Although cores have not 
been taken through the marsh on Smith Island, the soil 
type beneath the ridges has been classified general ly as 
silt and silt-clay loam (USDA, 1966). 
No sediment samples were taken on Smith or South 
Marsh Islands. Two sediment samples were collected from 
Bloodsworth Island along the channel edge of Fin Creek 
Ridge. These samples were not analyzed in the 
laboratory, however they were examined visually and 
texturally and determined to be high silt/clay content 
with little to no sand. The samples appeared similar to 
the samples analyzed for the Northern Group, suggesting 
that the basement composition of the Southern Group 
islands is possibly the same as the Northern Group. 
Shoreline Response Modeling 
The Inundation Model 
Northern Group 
Slopes and heights were determined from field 
surveys since the resolution of topographic maps is too 
coarse, with contour intervals of 1.6 m (5 feet). Most 
of the shorelines of the Northern Group islands are steep 
scarps, either clay or marsh. Therefore, the most 
116 
NW BLOODSWORTH ISLAND EJ Marsh Peat 
w E 
5~ 16 METERS fZJ Silt/Oay 
/[~ Fu l Loam 
\ o / [ 2 
\ ~~, 
,ecc 
500 METERS cct-/ NE BLOODSWORTH ISLAND 
tJ 
tJ ~~22~ 23 f, .. ,., 25E METE...J - - - o- I  RS 
2 
Figure 4.19 Geologic section through northern Bloodsworth 
Island showing development of marsh over the 
clay layer (from GEO-RECON, 1980) 
i mportant variable is the height of the scarp, or the 
i s land. The heights measured for the three islands are 
presented in Table 4.3. 
It is clear from Table 4. 3 that there are height 
variations throughout the islands. The upland areas lie 
between 0.37 and 2.3 m above MSL; the marsh areas range 
from 0. 11 m to 0. 5 m above MSL; and the marsh/upland 
marginal areas are between 0.3 and 0.58 m above MSL. The 
upland area which measured 0.37 m was probably an 
upland/wetland margin area rather than an upland area. 
If sea level rises according to the IPCC scenarios for 
the Bay, by 2030 the water level will rise 21 cm (Figure 
2.4). This will have large impacts on the marsh areas 
which will be inundated unless they can keep pace by 
increased sedimentation rates. In addition, as the 
marsh/upland marginal areas are inundated they will be 
converted to marsh; and upland areas will become margin 
areas, and eventually marsh. Thus, vegetation zones will 
migrate landward, where possible. By the year 2100, 
virtually everything except the highest upland areas will 
be below the water level, which is predicted to be 82 cm 
higher than it is today (Figure 2.4). Greater than 
seventy percent of the areas surveyed during this study 
will be completely inundated. 
118 
Southern Group 
If the i s lands in the Southern Group were static 
sys tems , then the inundation model would predict that a 
0. 5 m rise in sea-level would completely inundate the 
islands except for the few remaining upland ridges. 
However, these marshy systems are not static over time. 
The depth of the peat layer on these islands suggests 
that they have been vertically accreting in response to 
sea-level rise for at least the last few centuries. 
Without such accretion the islands would have been mostly 
inundated one or more centuries ago. 
In order to understand how successfully these 
islands are responding to sea-level rise, it is essential 
to know both the vertical accretion rates of the islands 
and the rate of sea-level rise in the Bay. Island 
vertical accretion rates are not available, so such an 
analysis was impossible. However, the apparent 
degradation of the interior marshes indicates that the 
rate of sea-level rise is currently exceeding the rate of 
vertical accretion. 
Hooper Island 
Hooper Island is a good area for examining the 
process of upland conversion to marsh in response to sea-
level rise. There are many examples of upland ridges 
119 
Whi' ch  are being submerged and are becoming marsh (Plate 
2 ? 4 ) ? Figure 4.20 presents the results of a transect 
from the center of a ridge to the middle of the adjacent 
marsh, showing the depth of the clay layer beneath the 
marsh surface and the upland ridge. Figure 4. 20 also 
shows the intruding wedge of marsh. 
The slope of the surface of the basement clay layer 
is only o. 2 degrees, which translates to 2. 86 m of 
horizontal displacement of the wetland/upland border for 
every l cm rise in sea level. This model suggests that 
the entire transect in Figure 4.20 was upland during the 
last century since sea level has risen approximately 30 
cm in the last 100 years to cause nearly 86 m of 
horizontal displacement at a 0. 2 degree slope. A rise in 
sea level of about 65 cm would inundate this ridge, which 
may occur by 2090. 
The topography of Hooper Island will ultimately 
determine the extent of inundation due to sea-level rise. 
All the marsh areas lie less than about 0.5 m above MSL, 
but the ridges vary in height throughout the island. A 
small ridge measured only about 7.6 cm above the adjacent 
marsh; a larger ridge measured 1.4 m above the adjacent 
marsh (Figure 4.21). Even if some of the ridges are high 
enough to remain above the rising water level, they are 
small in area and could not support the island 
population. More importantly, the lack of fresh water 
120 
- . 1 2 0 4 -e 5 6 7 
~ 
,c 
tlO 
~ 
QI 
:x: , . ?' ? ltfil  . r~,I' 1U:::1?? .? ? . ?77!.I: : :? ? :: ?t:;1 ??? l:::J ? .:? 
? ? I , ? ? ? .. ? ? ? ? ?? '_: =? ??: 
-.5 
.... , . . . . . ..:  . .. ? . -:;? :? :: .:  "...'. . .. ' ' . . ... . ?.::: . ?,,_ . . . 
0 5 10 15 20 25 30 35 40 45 
Distance (m) 
illillill Upland soil II Harsh till..   Clay 
Figure 4.20 Geologic section of Lower Hooper Island. 
Co.,, 
----survey?~-.. 
large r1'a 
,-
' 
Location of_ ___a ...~ 
?action in 
F11ure 4,21 
1 
Figure 4.21 Location of survey sites for Hooper 
Island- 11/23/91. 
122 
due to salt water intrusion into the groundwater would 
render the small ridges uninhabitable long before they 
are submerged. 
Historic Trends Analysis 
Northern Group 
The trend of land loss was extended into the future 
to predict when the islands will disappear, given the 
current rate of erosion and sea-level rise. The trends 
in Figure 4.22 suggest that James and Poplar Islands may 
disappear around the year 2000, and that Barren Island 
will disappear by the year 2040. This prediction is 
based on the existing conditions and does not account for 
an accelerated rise in sea level. The historic trend of 
rapid erosion of the Northern Group islands will likely 
continue regardless of any change in the rate of sea-
level rise. 
Future rates of sea-level rise were calculated for 
the Northern Group, based on the IPCC scenarios as 
calibrated to the higher rate of sea-level rise in the 
Chesapeake Bay. Table 4. 7 suggests that by 2030 the 
current rates of land loss will double; by 2070 they will 
nearly triple; and by 2100 they will more than triple. 
At these rates, Barren Island will be gone in 20 years; 
James and Poplar Islands will be gone in less than 10 
123 
y ears . 
Table 4.7 Historic and Future Rates of Land loss 
for the Northern Group 
Historic Land Future Land Loss 
Loss Rate (ha/yr)** 
Erosion SLR 1990 2030 2070 
island (ha/yr) Rate to to to 
(mm/yr)* 2030 2070 2100 
Barren 1.7 2.7 3.7 5.0 5.4 
James 2.1 2.7 4.6 6.2 6.6 
Poplar 2.0 2.7 4.4 5.9 6.3 
Based on the Baltimore tl.Cle g aug  e station I 1903 to 
1986. 
** Based on scenarios of sea-level rise for the 
Chesapeake Bay as calibrated from the Best Estimate 
IPCC (1990) scenarios: 6 mm/yr by 2030, 8 mm/yr by 
2070, and 8.6 mm/yr by 2100 (see Figure 2.4). 
124 
Figure 4.22 
Trends of Land Loss 
Northern Group 
500r---------------__.. 
? ? ? p. ? ?? '. ' 
400 ? ? 
300 
.e.nCl)-. 
 
 
n, 
(J 
Cl) 
%: 
200 ' 
100 
oL.---------------_.::, 
1900 1950 2000 1850 
Year 
-- James I. - Barren I. ~ Poplar I. 
125 
::.t;,?-?. .. 
Southern Group 
The historic trends analysis for the Southern Group 
is conservative because it only accounts for perimeter 
erosion and does not include the effects of interior 
ponding, stream widening, and marsh loss which are 
important land loss processes for the Southern Group 
(DeLaune, et al., 1983). When the trend line for the 
Southern Group is extrapolated beyond the end of the 
graph, it suggests the disappearance of the islands 
sometime in the 23rd Century (Figure 4.23). However, 
when the effects of interior marsh degradation were 
considered for Smith Island, the actual loss of the 
island is more likely to be sometime in the middle part 
of the 21st Century (Davison, 1990). 
Tables 4. a and 4. 9, and Figure 4. 24 present the 
results of the historic trends analysis for the Southern 
Group. Future calculations were based on the IPCC 
scenarios of sea-level rise to the years 2030, 2070 and 
2100, which were then calibrated to the higher rate of 
sea-level rise in Chesapeake Bay. Table 4. 8 suggests 
that by 2030, the current rates of land loss will more 
than double; by 2070, they will nearly triple, and by 
2100 they will more than triple. Table 4.9 predicts that 
Smith Island and South Marsh Island will disappear 
sometime between the years 2070 and 2100. Bloodsworth 
126 
Figure 4.23 
Trends of Land Loss 
Southern Group ? 
Hectares (Thousands) 
' 
I 
4 f-
i 
I 
2- . 
1 - 7 
0I '----------------------' 
1850 1900 1950 1987 2050 
Year 
---- Bloodsworth -+-- Hooper -+- Smith ~ South Marsh 
127 
IIU 
Island will be reduced to less than half of its 1987 
size; Hooper island will be reduced to less than one-
fourth of its 1987 size. However, it is important to 
note that if calculations of interior marsh loss are 
considered, the life of these islands will surely be much 
shorter. 
Table 4.8 Historic and Future Rates of Land Loss 
for the Southern Group and Hooper Island 
Historic Land I Future Land Loss 
Loss Rate (ha/yr)** 
I 
Island Erosion SLR 1990 2030 I 2070 
(ha/yr) Rate to to to 
(MMfyr) 2030 2070 2100 
I 
Bloodsworth 3.3 2.7 7.3 9.8 10.5 
Hooper 2.9 2.7 6.4 8.5 9.2 
Smith 10.2 2.7 22.7 30.2 32.5 
South Marsh j 3.3 2.7 7.3 I 9.8 10.5 J I 
* Based on the Baltimore tide gauge station, 1903 to 
1986. 
** Based on scenarios of sea-level rise for the 
Chesapeake Bay as calibrated from the Best Estimate 
IPCC scenarios (1990): 6 mm/yr by 2030, 8 mm/yr by 
2070, and 8.6 mm/yr by 2100 (See Figure 2.4). 
128 
Table 4.9 Future Projections of Island Size 
for the Southern Group and Hooper Island 
Island Size (ha)* 
Island 1987 2030 2070 2100 
Bloodsworth 1909 1595 1203 783 
Hooper 1285 1010 670 302 
Smith 3168 2192 984 - 0 -
South Marsh 1113 799 407 - 0 -
* Based on scenarios of sea-level rise for the 
Chesapeake Bay as calibrated from the Best Estimate 
IPCC scenarios (1990): 6 mm/yr by 2030, 8 mm/yr by 
2070, and 8.6 mm/yr by 2100 (See Figure 2.4). Also 
based on rates of land loss presented in Table 4.7. 
129 
Figure 4 . 24 Future Projections 
of Southern Group Islands 
Hectares 
3500------ - ---- -------, 
3000 
2600 ..... ????????????????? ? ? ?? ?? 
2000 
1600 
1000 
600 
oL.-....1....----L-L-_L.__L____;L...-......L...~__.J~...L.....--'-__.,. 
1987 2030 2070 2100 
Year 
- Bloodsworth --+- Hooper 
_.. Smith -G- South Marah 
130 
CHAPTER 5: DISCUSSION 
Introduction 
The patterns of land loss of the Northern and 
Southern Groups suggest that very different processes are 
causing the land loss for each group. The Northern Group 
is eroding from the west with more limited change on the 
protected, eastern shore of the islands. The Southern 
Group has a different pattern, consisting of (a) more 
uniform erosion around the perimeter of the island, and 
(b) interior marsh degradation by means of ponding and 
apical erosion of tidal marsh creeks. The differences in 
the processes of land loss can be attributed to the 
geomorphological characteristics of the two groups. 
The Northern Group 
The pattern of land loss for the Northern Group is 
related to several factors, including ( 1) wave 
characteristics, (2) storm frequency, (3) sediment type, 
and ( 4) tidal range. These interrelated factors are 
driving the erosion which has reduced the aerial extent 
of the islands by more than 76% in 138 years. 
Wave Characteristics 
Waves are the primary agent of coastal erosion 
(Komar, 1983). The pattern of erosion of the Northern 
131 
Group suggests that wind-driven waves from the west and 
northwest are the driving force. Indeed, the predominant 
wind direction in the Bay is from the west and northwest 
(Table 2.1) and the longest fetch distances are from the 
western quadrants ( Table 4. 4) . As a result, larger waves 
with more energy reach the western shore of the islands. 
The eastern shore has a shorter fetch and is protected 
from the predominant winds. Significantly less shore 
erosion is occurring on the eastern shores of the islands 
(Table 4.5). During all field visits, the eastern shores 
were noticeably calmer than the western shores (Plate 
5. 1) ? 
The rates of erosion for the Northern Group islands 
are very high, much higher than the Atlantic coast 
average of 0.8 m/yr (NRC, 1987) or the Chesapeake Bay 
average of 0.6 to 0.9 m/yr (Wang, et al., 1982). The 
erosion of the western facing clay bluffs is a continuous 
process which is occurring at all tide levels and wave 
conditions, and is exacerbated during storm conditions, 
thereby increasing the rates of erosion. 
Storm Frequency 
Erosion rates appeared to increase during periods of 
high storm frequency (Figure 5.1), suggesting that the 
erosion is linked to storm frequency. James Island did 
not clearly respond to the high storm period between 1942 
132 
Plate 5.1 Eastern side of Barren Island 
133 
Figure 5.1 Land loss v. Storm frequency 
The Northern Group 
Land Loss: ha/yr Storm Frequency: #/yr 
2.4 
/\. 0_9 
2.21 ~ / / ' ' 
/ 
.... 2 
w - 0.8 I 
~ ~ 0 - - Storms/yr 
1.8 / + Barren 
*James 
1.6 I /-? 
--- Poplar 
- 0.6 
/ 
/ 
1.41 / 
/ 
/ 
<I 
I 0.5 
1.2 
. - ------- ----- -~-? 0.4 
1901-42 1942-64 1964-87 
Year Group 
and 1964 (see Appendix A). However, the land loss rate 
in Figure 5. 1 represents an average of two periods , 1942-
l952 and 1952-1964, which experi enced very different 
rates of loss. Between 1942 and 1952, James Island 
changed very little, at a rate of 0.4 ha/yr. However, 
between 1952 and 1964 James Island lost an average of 2.4 
ha/yr. Both were periods of high storm frequency, but 
the period 1952-64 affected the island more severely. 
The orientation and strength of the storms, relative to 
James Island, could possibly account for the different 
erosion rates during these two periods. The 1962 
northeaster which affected the Bay area could have had a 
major impact on the Bay islands. James Island, unlike 
all the other islands, has a very short fetch from the 
south and is thus relatively well protected from storms 
approaching from this direction. However, the 
orientation of James Island is such that it is prone to 
wave attack from the northeast. 
Sediment Type 
The sediments and the environmental characteristics 
for the Northern Group islands are very similar to those 
of Lake Erie and the processes of bluff toe erosion in 
the two areas also appear to be much the same. The 
cliffs along the western shores of the islands are 
composed of cohesive silt/clay, which are eroding 
135 
rapidly. In studies along a Lake Erie coast, Carter and 
Guy (1988) documented that erosion of clay cliffs which 
are comprised of cohesive sediments occurs by two main 
processes: (1) abrasion, the gradual wearing away of a 
deposi t by sediment transported by the waves and wave 
uprush, and (2) quarrying, the tearing away or dislodging 
of discrete pieces of the deposit by hydraulic and/or 
pneumatic forces. Quarrying was determined to be the 
principle erosion process. An important difference is 
that during the winter, the Lake Erie bluffs are 
prot ected from erosion by ice packs along the shore 
(Carter and Guy, 1988). The Chesapeake Bay rarely has 
ice and certainly not for prolonged periods, so erosion 
occurs throughout the year. 
Examples of bluff toe weathering and erosion can be 
seen along the western edge of Barren, James and Poplar 
Islands ( Plate s. 2). Blocks of silt/clay which have 
broken off the bluff are more easily eroded by wave 
action. Wave erosion is the crucial erosion process; 
removal of material from the toe prevents the development 
of a stable slope, thereby allowing the process of 
erosion to continue indefinitely (Carter and Guy, 1988). 
Because the eroded sediment is nearly all fine 
silt/clay, it remains in suspension or is carried 
offshore, rather than depositing in the nearshore (Komar, 
1976). As a result, it is not available for beach 
136 
Plate 5.2 Bluff toe weathering on Poplar Island 
137 
recovery following a storm, as in a sandy, coastal 
environment. Therefore, once the fine material has been 
quarried and weathered, it is easily and permanently 
eroded. 
This mode of sediment behavior is very different 
than on a sandy coastline, where periods of accretion 
usually follow erosional events. The shore position 
along a sandy coast is essentially an average of the 
location of the shoreline over time. For areas where the 
eroded sediment goes directly into suspension, there is 
no accretionary period and the shoreline only retreats. 
As a result, the erosion rates are high because they 
reflect only retreat and no accretion. 
Tidal Range 
During all observations in the field, waves impacted 
against the bluff toe, thus expending their energy 
directly against the toe during most, if not all, of the 
tidal cycle. The tidal range in the area of the Northern 
Group is low, only about o.s m. In areas with a large 
tidal range the energy is spread over a wide area and a 
beach face can form (Marcel Stive, pers. comm.). In 
microtidal environments such as the Chesapeake Bay and 
where there is no beach face, the energy is focused more 
locally at the bluff toe through all tidal cycles. 
Without the protection of sandy beaches, the energy is 
138 
not dissipated so even small waves provide an erosive 
force. Under these circumstances, the pattern of erosion 
can be expected to continue indefinitely as long as 
erodible sediments are exposed to wave attack (Phillips, 
1986). 
For these low-lying islands, virtually all wave 
conditions are causing erosion and erodible sediments are 
continually available, so this pattern will persist until 
the islands have disappeared or shore protection 
structures are built. Marcel Sti ve, ( personal 
discussions, 1991) described a similar situation in 
Holland, where there is a small tidal range and wave 
action is producing a wave-cut scarp with a level 
platform offshore. 
Summary 
The process of bluff toe erosion for the Northern 
Group is illustrated in Figure 5. 2. Non-storm waves 
quarry and undercut the vertical bluff surface on a daily 
basis, eventually creating an unstable bluff (Plate 5.2) ? 
Storm waves provide enough energy to break off unstable 
blocks of bluff substrate. Significant and rapid 
horizontal retreat of the shore can therefore result from 
storm wave energy. Erosion of the Northern Islands has 
occurred in pulses of erosion during high storm spells, 
followed by slower erosion rates during more quiescent 
139 
E >W 
a. 
Ill 
b. 
- .......-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-... . .... . 
?---?-?-?-----------------------? 
~ :::::::::::::::::::~-:-:-:-:-:-:-:-:-:-:-:-:-:-: ______________________ _ ~] 
Ill 
I-' 
"0" ' 
Q SILT-aAY 
c ? 
. ------
....... -?----------------------------------------------
2 
.........r..-..-....-...-._____._.._.._._.._.._.._._.._.._._.._.._.._..
.-
_..
.-
._..
..-..-.-.-..-.-.-..-.-.-..-.-.-..-.-.-..-.-.-..-.-.-..-.._. ..............................................................................
-.
...
.-...-..-...-..-..-..-.-.-............................
.-..-..  ..... .  
....................................................................... \-]. 
Figure 5.2 Schematic diagram of erosional processes of a Northern Group 
Island; (a. and b.) Unstable bluff created by erosion of bluff 
toe by non-storm wave activity; ( C. ) Energy of storm waves quarry 
bluff and cause shore recession. 
periods. 
This pattern of erosion will continue in the future 
r egardless of a rise in sea level. However, due to 
rising sea-level, erosion rates during storms may 
increase. Therefore, rising sea level will likely 
exacerbate the existing effects of storms on the Northern 
Group islands by allowing higher storm surge and storm 
waves to reach the shore. 
Southern Group 
Land loss for the Southern Group is occurring 
through marsh edge erosion and interior marsh 
degradation. Both of these processes have been 
attributed to sea-level rise (Orson, et al., 1985). 
Interior marsh loss was not quantified for this study, so 
land loss estimates refer only to marsh edge erosion, 
except where indicated otherwise. The land loss 
estimates in Table 4.2, therefore, are much lower than 
true rates of land loss since interior marsh loss 
represents an important process of marsh response to sea-
level rise (DeLaune, et al., 1983). Interior marsh loss 
estimates at Blackwater Wildlife Refuge in the Chesapeake 
Bay approximates 56 ha/yr (Pendleton and Stevenson, 
1983), and 49.6 ha/yr in the Nanticoke estuary (Kearney 
et al., 1988). These rates are an order of magnitude 
141 
higher than marsh edge erosion rates estimated in this 
study (from 3.3 to 10.3 ha/yr). Therefore, estimates of 
total land loss for the Southern Group would probably be 
much higher if interior marsh loss is factored in. This 
is not to compare interior marsh loss and marsh edge 
erosion; those processes are quite different. However, 
it is important to note that both processes are occurring 
and therefore both should be considered when estimating 
true rates of land loss. 
The results of this study indicate that the observed 
marsh edge erosion, at least for Bloodsworth and South 
Marsh Islands, is weakly correlated to increasing sea 
level in the Chesapeake Bay (Figure 4.15). As long as 
sea level continues to rise, these processes will 
continue; if sea level rise accelerates, the rate of land 
loss will increase accordingly. 
Marsh Fringe Erosion 
Marsh fringe erosion is causing high erosion rates 
around the United States, from over 4 m/yr in the St. 
Lawrence River estuarine marshes (Dionne, 1986), 3.2 m/yr 
in Delaware Bay marshes (Phillips, 1986; French, 1990), 
and 5 m/yr in Louisiana marshes (Penland, et al., 1985). 
Estimates of island marsh fringe erosion from this study 
are lower than these trends, between about 0.5 m to 1.6 
m/yr. (Table 4.6). 
142 
Table 5.1 
A Sample of Marsh Fringe Erosion 
Rates from Several Geographic Regions 
Regions Erosion Rate lm/yr) Study 
St. Lawrence 4.0 Dionne 1986 
River estuary 
Delaware Bay 3.2 Philliips 1986; 
French 1990 
Louisiana 5.0 Penland, et al, 
1985 
Certain areas of all the islands, notably 
promontories, are eroding at rates comparable to the 
areas noted in Table 5.1. However, due to difficulties 
calculating erosion rates for irregular shorelines as 
previously described, erosion rates were not computed for 
these areas from the Metric Mapping program. It is 
possible to visually estimate the erosion rates of some 
rapidly eroding areas such as promontories. Some of the 
island promontories and most of the western edge of Smith 
Island were estimated to be eroding at rates exceeding 3 
m/yr. The truncation of marshy peninsular points due to 
erosion was also described for Delaware Bay (Phillips, 
1986). In addition, dramatic shoreline change was often 
associated with the opening of streams, both in Delaware 
Bay (Phillips, 1986) and in this study. 
French ( 1990) found that marsh edge erosion in 
143 
Delaware Bay is constant through time and does not 
increase during high storm periods. In contrast, 
Finkelstein and Hardaway (1988) found that marsh edge 
erosion was particularly impacted by storm wave activity. 
The results of this study show an increase in erosion 
during a particularly stormy period (Figure 5. 3), thereby 
suggesting that marsh edge erosion is influenced by wave 
activity. Exposure to high wave energy has been found to 
contribute to seaward edge erosion of marshes in 
Chesapeake Bay (Froomer, 1980). Marshes in Delaware Bay 
are somewhat protected from high storm energy approaching 
from the Bay mouth, which may explain their lack of 
response to storms (French, 1990). 
Smith Island is not included in Figure 5.3 because 
there were too few year groups for comparison. Hooper 
Island was included because of the rapid erosion of the 
southern end of the island, which is almost entirely 
marsh. 
Figure 5. 4 depicts the process of marsh edge erosion 
for the Southern Islands. This process occurs when 
chunks of marsh peat are undermined by normal, daily wave 
energy and loosened by biogenic activity of crabs and 
worms (Plate 2.1). During smaller storms, waves which 
impact the marsh edge break off any loosened blocks of 
peat, causing horizontal recession of the marsh edge. 
Under this scenario, erosion rates may increase. 
144 
Figure s. J Land loss v. Storm frequency 
The Southern Group 
----------------- - -- ----??? 
Land Loss: ha/yr Storm frequency : #/yr 
8 ~------------~----------------~1 
,I'-.. 
7 .... 
; / 1. ...,_ , .... 0 .8 
61 , , 
.... , , 
,e,. 
U1 sl , ,, 
, 
, 0.6 
.... . Storms/yr 
, -t- Bloodsworth 
41 ? a' 
* South Marsh 
0.4 
3 --- Hooper 
2 
0.2 
- - -------- ----Jo 
0 1901-42 1942-52 1952-87 
Year Group 
E -----------------------> W 
a. 
~----:~:'!""P:::!:?:;::?:;:?::?:i::?:::::z:-:i;~!!'i;:'.:~~?:~- 0. 5m 
b. 
~--..J! '.'=::::::..:!:~;!(!K?=: --: --:-=?:~?~?:?:?~1???1??~~---- ] 0.5m 
(:3 MARSH PF.AI 
I-' 
~ 
?' ____ ,,. __ e 1 ] 0.5m __.._ r 1:::;:t1~Z!?>-:--:,. ~ --Z~?:!.w ......fM. 
Figure 5.4 Schematic diagram of erosional processes of a Southern Group 
island; (a. and b.) Unstable marsh edge created by non-storm 
wave activity and loosened by biogenic activity; (c.) During 
major storm surges, marsh is submerged and waves overtop the 
marsh. 
However, during larger storms the surge is high 
enough to overtop the low-lying islands, thereby largely 
dissipating wave energy on the marsh surface rather than 
at the marsh edge (Figure 5.4). When this occurs, a 
storm may have little effect on the marsh edge in 
proportion to the storm? s size. In this regard, the 
response of the Southern Group to storms differs from 
that of the Northern Group. The Northern Group bluffs 
are higher and are therefore affected at all times by 
wave activity, both during major storm surges and on a 
daily basis. The marsh edge of the Southern Group 
islands is low enough that large waves accompanied by 
storm surge may have a smaller effect on the marsh edge. 
Interior Marsh Loss 
Several factors appear to contribute to the 
degradation of the interior island marshes, including 
lack of a sediment source, land subsidence, small tidal 
range, and the vulnerability of submerged upland type of 
marsh. Although not quantified for this study, it is 
clear from other studies that this process is 
significantly affecting marshes around the Chesapeake 
Bay. 
Where submergence is occurring at a rate that is 
higher than the marsh's ability to keep up with relative 
sea-level rise, marsh loss occurs through interior pond 
147 
formation, coalescence and enlargement. This process has 
been documented previously in marshes in the Chesapeake 
Bay at Blackwater Wildlife Refuge (Pendleton and 
Stevenson, 1983), the Nanticoke Estuary (Kearney, et al., 
1988), and Smith Island (Davison, 1990). This has also 
been shown to be important in Louisiana marshes ( DeLaune, 
et al., 1983) Although the process of marsh loss was not 
quantified for this study, it was observed in sequential 
aerial photographs of the three Southern Group islands 
that ponds are enlarging, channels are widening, and the 
incidence of interior open water is increasing. 
Sediment source 
Marsh loss occurs by submergence resulting from 
subsidence of the land and an inadequate sediment supply 
in the face of rising sea level. These islands are prone 
to submergence because there is no upland inorganic 
sediment supply; the most important source of inorganic 
sediment is probably erosion of the marsh edge (cf. Reed, 
1988) and storm flooding (Baumann, 1980). Organic 
production through plant death and culm debris accounts 
for a small percent of vertical accretion, but cannot 
amount to much without sufficient inorganic input 
(personal communication, Pendleton, E.C., 1992). The 
percent of organic vs. inorganic material in the island 
marshes has not yet been studied. 
148 
The material being deposited on the marsh surf ace by 
overwash probably does not reach the interior of the 
marsh and is more likely placed near the edge of the 
marsh, creating a streambank ef feet. However, because of 
rapid marsh edge erosion this material may never benefit 
the island marsh system. It is ironic that storms, which 
are so detrimental to the marsh edge, are critical to the 
survival of the marsh by sediment input. 
Tidal range 
Stevenson et al (1986) suggest that tidal range is 
important to a marsh's ability to respond to sea-level 
rise, and that a marsh with a low tidal range is more 
vulnerable to sea-level changes. This would imply that 
marshes around the Chesapeake, including the islands, are 
particularly vulnerable to the rapid rate of sea-level 
rise in the Bay. A marsh in an area with low tidal 
amplitude has a small vertical range, thus is more 
sensitive to elevational changes in water level. The 
interior of the expansive island marshes do not flood 
regularly and therefore have limited inorganic sediment 
input. Accretion of the interior marsh surface is 
therefore limited to organic input from culm debris and 
organic material from root production al though this 
source is poorly (or not) quantified. Once this material 
149 
is c ompressed  and degraded, it probably adds little to 
the surface elevation (personal communication, Pendleton, 
E.c., 1992). Inorganic sediment input is critical for 
marsh accretion and, therefore, survival of the islands. 
Marsh Type 
The Southern Group islands appear to be submerged 
Upland marshes. Kearney et al. ( 1988) reported that this 
type of marsh experienced the most rapid rate of loss of 
the principal marsh systems in the Chesapeake Bay. As a 
first response to rising sea level, interior ponds form 
in apparently random locations due to anoxia and plant 
death (Mendelssohn et al., 1981). The ponds enlarge and 
coalesce after they form, eventually becoming large areas 
of open water. Davison (1990) established that interior 
pond formation is occurring very rapidly on Smith Island. 
It is likely that all the large marshy islands are 
experiencing similar modes of interior pond formation and 
enlargement. 
The processes of interior pond enlargement have 
not been identified for this study area. However, a 
discernible west-northwest to east-southeast axis of many 
ponds suggests that wave erosion, driven by the 
predominant winds, may be an important factor for pond 
growth (Stevenson, et al., 1985b; Davison, 1990). After 
150 
ponds reach a critical size, wind-generated waves begin 
to erode the marsh edge, which in turn expands the pond. 
Figure 5.3 explains the various methods of open water 
formation in an island marsh system. 
The process of land loss becomes nonlinear as it 
progresses and begins to accelerate. With more ponds, 
larger tidal channels, and an eroding marsh edge, the 
incidence of coalescence is higher and the percent of 
open water increases. Thus, the rates of marsh 
deterioration in advanced stages are higher than in early 
stages. 
Summary 
Land loss for the Southern Group is occurring by two 
processes: shore erosion and interior marsh loss. Both 
processes are significantly affecting the integrity of 
the large marshy Southern Islands. Interior marsh loss 
is probably accounting for a higher percentage of land 
loss than perimeter erosion. It is likely that if rates 
of interior marsh loss were quantified for the islands, 
they would be an order of magnitude higher than the 
perimeter erosion rates. Thus, the land loss estimates 
for the Southern Group are very low and the ultimate 
demise of the islands is more imminent that predicted in 
Figure 4.25. 
151 
Hooper Island 
Hooper Island is being affected by the processes of 
er ? 
osion and submergence. Erosion has been slowed where 
engineering structures have been built (Plate 5.3), but 
some unprotected shores have been eroding at an average 
annual rate of about o. 7 m/yr since 1848. Submergence is 
increasingly becoming a problem as the island surface 
becomes closer to mean water level. Hooper Island is 
slowly submerging by land subsidence and sea-level rise. 
There is ubiquitous physical evidence for this process, 
including the conversion of upland to marsh, elevated 
groundwater, recurrent flooding, and erosion. 
These effects are important because they indicate 
that the real problem is not being solved, despite 
attempts to prevent erosion. The most serious problem 
for the residents of Hooper Island is the encroaching sea 
which is turning upland to marsh, causing frequent 
flooding, saturating lawns and basements, and generally 
decreasing the quality of life on the island. These 
Problems cannot be solved with seawalls, revetments or 
bulkheads; the only options are retreating from the area, 
raising the height of the land, or possibly using dike 
and water control systems as in Holland. 
The timing of these processes is an important 
variable. According to Figure 4.21, the marsh has most 
recently extended from location #3 to #2, a distance of 
152 
Plate 5.3 Example of erosion control measures along the 
wetsern shore of Hooper Island 
Plate 5. 4 View of an upland ridge on Hooper Island, 
surrounded by the encroaching marsh 
153 
approximately 7.62 m. If the slope of the clay layer 
beneath the marsh is .22 degrees, then mean sea level 
must have risen 3 cm for a horizontal displacement of 
7. 81 m. The refo re, using the Baltimore tide gauge record 
with an average rise of 2. 7 mm/yr, the process of 
inundation from location #3 to #2 has occurred in about 
the last 11 years. 
Hooper Island has transformed since 1901 from an 
island with a high proportion of upland to an island 
dominated by wetlands. The shrinking upland ridges are 
the only upland areas remaining (Plate 5. 4). As the 
island is submerging and the process of upland conversion 
continues, the island will progressively become less 
habitable. Table 4. 8 predicts that Hooper will be 
reduced to less than one-fourth of its current size by 
the year 2100. However, this does not include marsh loss 
processes which will become more important as the island 
submerges and sea-level rises in the Chesapeake Bay. 
154 
CHAPTER 6: CONCLUSIONS 
This s tudy is the first instance where an accurate 
historical mapping procedure has been used to quantify 
the spatial and temporal processes of l and loss for 
islands in the Chesapeake Bay. Previous studies have 
used visual comparisons of historic maps and photographs 
to identify island shoreline changes, which have been 
enhanced by anecdotal descriptions of the islands 
(Singewald and Slaughter, 1949; Kearney and Stevenson, 
1991). Field surveys were performed to identify the 
geomorphological processes at the shoreline. The 
processes of land loss were then analyzed from this data. 
This is also the first example where Metric Mapping has 
been used for irregular, island shorelines. 
This study identifies two geomorphologically 
distinct island types, termed the Northern Group and the 
Southern Group, which have exhibited very different 
patterns of shoreline behavior. A comparison of the two 
island groups has resulted in a detailed examination of 
the mechanisms of land loss. Erosion is the dominant 
land loss mechanism for the Northern Group; erosion and 
submergence are the dominant processes of land loss in 
the Southern Group. For the Northern Group, the trend 
will continue with or without an accelerated rise in sea 
155 
level . The processes of erosion are b e ing controlled by 
wind and storm-driven waves which will continue 
regardless of further sea-level change, but will increase 
with accelerate d sea-level rise predicted for the coming 
decades. The processes of land loss for the Southern 
Group and Hooper Island can be attributed to sea-level 
rise and wave-induced erosion of the marsh edge. 
Submergence of these islands, which is causing marsh 
deterioration and conversion of uplands to marsh, will be 
accelerated with an increased rate of sea-level rise. 
The prognosis for either group of islands is not 
good. The Northern Group is eroding rapidly and the 
islands are small; the Southern Group islands are larger, 
but submergence is reducing any available upland and both 
salt water intrusion and inundation is rendering these 
islands uninhabitable. 
Land loss for the Southern Group has been grossly 
underestimated in this study because it did not account 
for interior marsh loss. Further analysis should include 
quantifying internal marsh loss, and identifying the 
processes and causes of ponding and pond enlargement on 
the islands. The islands' future response to sea-level 
rise can be more accurately predicted with knowledge of 
the timing, magnitude and mechanisms of marsh loss. 
To fully understand the process of island 
submergence, a study of island marsh accretion rates is 
156 
essential. Rates of accretion could be compared to 
subsidence rates in the Bay to determine how the islands 
are responding to sea-level rise. In addition, island 
accretion rates could be compared to other marsh 
accretion rates from previous studies. In addition, it 
would be critical to know the organic/inorganic 
composition of the island marsh peat, as a means to 
determine whether the islands are sediment starved and 
are therefore vulnerable to interior marsh loss. 
Additional field research should include transects 
to determine the depth of the clay layer beneath 
Bloodsworth, Smith and South Marsh Islands. This 
information would be helpful for the definitive 
identification of the Kent Island Formation beneath the 
Southern Group islands. In addition, identification of 
the clay layer beneath the small marshes on the Northern 
Group islands would support the theory that the Kent 
Island Formation extends from Kent Island to Tangier 
Island. This research could also clarify the present 
understanding of the evolution of these islands during 
the Holocene. 
The U.S. Fish and Wildlife Service has a vested 
interest in Barren Island, Smith Island, and Bloodsworth 
Island. Other islands are also important to the Service 
because they provide isolated sanctuaries for breeding 
and nesting waterfowl. The reduction of these islands is 
157 
already reducing potential nesting habitat. However, 
with the exception of black ducks ( Krementz, et al. , 
19 91) , a connection between reduced habitat and the 
nwnber of waterfowl in the Bay has not yet been made. 
Management implications of reduced nesting area on 
waterfowl is dependent on delineating the islands 
evolution and what is likely to happen in the future. 
Because of their importance as waterfowl habitat 
along the Atlantic flyway, there is a growing interest in 
developing management alternatives for the islands. 
There are several options for island protection and 
restoration, including (1) hard stabilization such as 
bulkheads and revetments, (2) soft stabilization using 
dredge material, or (3) a combination of hard and soft 
stabilization alternatives. 
Hard stabilization includes structures such as stone 
revetments, and metal or wooden bulkheads. Bulkheads and 
revetments are common around the mainland shores of the 
Bay. However, except for Hooper Island, they are 
generally absent from the island shores. A small 
revetment was built in the last few years on Coaches 
Island (Poplar Island). The Federal government has on 
occasion shown interest in protecting valuable habitat by 
using hard stabilization if necessary. A segmented 
offshore breakwater and beach fill project is currently 
being constructed to protect Eastern Neck National 
158 
Wildlife Refuge, in the northern part of Chesapeake Bay. 
For the marshy islands, coastal erosion is not an 
immediate threat to the upland areas which are surrounded 
by marsh, so alternative hard stabilization suggestions 
have been developed. The U.S. Army Corps of Engineers 
( 1984) has proposed building a floodwall around the 
ridges on Tangier Island, which is geomorphologically 
similar to Smith Island. The floodwall would be designed 
to the 100-year flood level plus 1 m of freeboard. An 
alternative is to build a 100-year floodwall around a 
community center or school to provide a sanctuary against 
severe floods. 
Hard stabilization will not be the best recourse in 
all situations. As Hooper Island demonstrates, erosion 
control structures will not ensure the integrity of a 
marshy island, since submergence is the most important 
problem for these islands. It will be necessary to raise 
the island's surface in response to sea-level rise. One 
option which has been tested in Louisiana marshes and has 
great potential, is to apply a thin layer of dredge 
material onto a marsh surface using a high pressure 
spray. This technique was developed to avoid creating a 
spoil bank as a result of channel dredging for oil 
operations (Cahoon and Cowan, 1988). Spoil banks 
restrict overbank flooding which contributes to marsh 
submergence and deterioration. Cahoon and Cowan (1988) 
159 
found that in two sites monitored, marsh vegetation 
recolonizes after two growing seasons or after one 
growing season in areas where the sprayed layer is very 
thin. This method appears to have great potential for 
maintaining subsiding marshes, and could be very 
effective in the Chesapeake Bay. A critical parameter is 
the height or thickness of the spoil layer on the marsh 
Which allows marsh species to establish. The grain size 
and suitability of the spoil material must also be 
considered. 
Another potential option for habitat protection or 
restoration is the use of clean dredge material to expand 
the island size in combination with hard stabilization to 
impound the dredge material. Such a proposal is being 
considered by the State of Maryland, the U.S. Fish and 
Wildlife Service, and the U.S. Army Corps of Engineers, 
for Bodkin Island, a small one-acre island in the 
northern part of the Bay (Figure 6.1). Preliminary 
designs have also been made for Poplar Island (Figure 
2.9). Parameters such as dredge spoil transportation 
costs, dredge spoil suitability, and the cost of 
additional protection must be considered for these 
projects and may not prove to be cost effective. 
Islands made of dredge material have proven to be 
good nesting sites for waterfowl, so recreating islands 
with dredge material may be very effective. Sundown 
160 
Schematic drawing of Bodkin Island habitat 
Figure 6.1 enhancement proposal (from U.S. Fish and Wildlife 
Service, undated draft) 
_/ 
/ 
$~111!eY 
AGP,S~E-e~~~: 
/ 
~CJPUf,/t:) 
~~~~-~ITIO/w' AA?J.. 
eML:p----MldfM MUI.~ 
IMT"EAT'IPAL. MAASH 
E,XlST'ltJ<il 
IJP&-ANO l SL.A~P s 
1,td.E 
FlilePOS&O a 
~ff.P I SL>Nt' ,i: 
S~&S 
~MffP UPL-'IJ~ ? 
I '?"'?. 
161 
Island in Port O'Conner, Texas is a dredge spoil island 
that has been a breeding site for brown pelicans and 
other species since 198 7. This island is eroding 
rapidly. Concerned for the nesting birds, the National 
Audubon Society is hoping to enlarge the island with 
dredge material (Associated Press, 1991). 
Queen Bess Island in Barataria Bay, Louisiana is 
another small, eroding island which is an important 
breeding ground for brown pelicans. This island is 
eroding at a rate of about 0.25 ha/yr, a much lower rate 
than the islands in this study. The state of Louisiana 
and the Army Corps of Engineers are now working to 
rebuild Queen Bess Island. A dike is being built and 
dredge material is being placed behind the dike. 
Preliminary surveys show that pelicans are already taking 
advantage of the newly created areas (Marcus, F. F., 
1990). 
The last alternative is to encourage retreat from 
the inhabited islands. Many residents of flood-prone 
communities view flooding as a temporary inconvenience 
which is balanced by the aesthetic and cultural 
attractions of living on an island (U.S. Army Corps of 
Engineers, 1984), so retreat may not be the preferred 
alternative for island residents. A time will come, 
however, when flooding on Smith and Hooper Islands will 
be more than an inconvenience; it will not be feasible to 
162 
inhabit the islands once submergence has completely 
altered the physiography of the area. This may not come 
for another generation, but it is inevitable as long as 
sea-level continues to rise. 
Summary 
1. Two distinct types of islands were identified in the 
Chesapeake Bay, small upland islands (the Northern Group) 
and large marshy islands (the Southern Group). 
2. The two island groups are losing land in different 
manners: bluff erosion by wave action is the mechanism of 
land loss for the Northern Group; and marsh edge erosion 
by wave action as well as marsh deterioration caused by 
submergence are causing land loss for the Southern Group. 
3. The Northern Group is losing land at an averaged 
long-term rate of 1.9 ha/yr. The western side of the 
islands are eroding at an averaged rate of 4.9 m/yr; the 
eastern side is eroding at an averaged rate of 0.6 m/yr. 
4. The Sothern Group is losing land at an averaged, 
long-term rate of 5.6 ha/yr, with an averaged erosion 
rate of 1. 2 m/yr. Interior marsh loss was not quantified 
for this study, however, so these land loss rates are 
163 
considered to be low. 
5 . Land loss appears to be weakly correlated to sea-
level rise for the Southern Group. Land loss appears to 
be related, in part, to storm frequency for both island 
types. 
6. The prognosis for the islands is poor. At the 
current rates of land loss, the Northern Group will be 
gone between the years 2000 and 2040. It is difficult to 
predict the demise of the Southern Group islands without 
a quantitative understanding of the rate and processes of 
interior marsh loss. Marsh edge erosion rates for 
Bloodsworth, Hooper, Smith and South Marsh Islands 
grossly underestimate the life expectancy of these 
islands. With an accelerated sea-level rise, marsh edge 
erosion calculations alone predict that Smith and South 
Marsh Islands will be gone by the year 2100. With the 
progression of interior marsh loss, however, these 
islands would become uninhabitable long before 2100. 
7 ? Management alternatives for the islands include 
shoreline control structures, beneficial use of dredge 
spoil, a combination of hard and soft techniques of shore 
stabilization, and retreat. All of these options have 
important benefits and costs which must be weighed 
164 
carefully on a site-specific basis. Management decisions 
should consider the beneficial use of dredge material a s 
a feasible and positive solution to island deterioration. 
165 
APPENDIX A 
HISTORICAL STORM DATA 
Major Storm Tracks in the Chesapeake Bay Region, 
1871 - 1986* 
XE.AR DATES STAGE IN IN OR NEAR BAY NAME 
CHESAPEAKE 
1872 October 25-26 Hurricane Near 
1874 September 29-30 Hurricane In 
1876 September 17-18 Hurricane Near 
1877 October 4-5 Hurricane In 
f--'j 1878 October 23 Hurricane Near 
CJ'\ 
CJ'\ 1879 August 25 Hurricane In 
1881 September 10 Hurricane In 
1882 September 23 Hurricane In 
September 11-12 Hurricane In 
1883 September 12-13 Hurricane Near 
1885 October 13 Hurricane Near 
1886 June 22-23 Hurricane In 
July 2 Hurricane In 
1888 September 10-11 Tropical Storm In 
1889 September 24-25 Hurricane In 
1893 October 23 Tropical Storm In 
June 16-17 Hurricane Near 
1894 October 9-10 Hurricane In 
September 29 Hurricane Near 
1897 September 23-24 Tropical Storm In 
October 25 Tropical Storm Near 
1899 August 18-19 Hurricane Near 
October 31 - November 1 Extrat ropical In 
1900 October 13-14 Extrat ropical Near 
1901 July 11 Hurricane Near 
1902 June 16 Extra tropical In 
October 10-11 Extrat ropical In 
1904 September 14-15 Extrat ropical In 
1907 September 23 Extra tropical In 
June 29 Extrat ropical Near 
1912 June 15 Tropical Storm Near 
1915 August 3-4 Extrat ropical In 
1918 August 15 Tropical Storm Near 
1923 October 23-24 Extratropical In 
1924 September 30 Extrat ropical In 
August 25-26 Hurricane Near 
1925 December 2-3 Tropical Storm Near 
f-J 1927 October 3 Extratropical Near 
?-...:'i ' 1928 August 11-12 Extrat ropical In September 19 Tropical Storm Near 
1929 October 2 Extrat ropical Near 
1933 August 23 Tropical Storm Near 
1934 June 18-19 Extratropical Near 
1935 August 6 Tropical Storm Near 
1936 September 18 Hurricane Near 
1938 October 25 Tropical Storm Near 
1943 September 30 Tropical Storm Near 
1944 August 2-3 Tropical Storm In 
September 14 Hurricane Near 
October 20-21 Tropical Storm In 
194.5 September 18 Extrat ropical Near 
1947 September 25 Extrat ropical Near 
1949 August 28-29 Tropical Storm Near 
1952 September 1 Tropical Storm Near Able 
1953 August 14 Hurricane Near Barbara 
1952 September 1 Tropical Storm Near Able 
1953 August 14 Hurricane Near Barbara 
1954 August 30-31 Hurricane Near Carol 
1955 August 12-13 Tropical Storm In Connie 
September 19-20 Tropical Storm Near Ione 
1956 June 27 Extratropical Near 
1958 August 28-29 Hurricane Near Daisy 
1959 July 10-11 Tropical Storm Near Cindy 
1960 July 29-30 Tropical Storm In Brenda 
September 12 Hurricane Near Donna 
1961 September 14-15 Tropical Storm In 
1962 August 28 Hurricane Near Alma 
1965 June 16 Extratropical Near 
1966 September 16-17 Tropical Storm Near Doria 
1969 August 20 Tropical Storm Near Camille 
1970 May 27 Extratropical Near 
1-1 1971 August 27-28 Tropical Storm Near Doria 
(J'\ 
(X) 1972 June 21-22 Tropical Storm In Agnes 
1976 August 9-10 Hurricane Near Belle 
1979 September 5-6 Tropical Storm Near David 
1981 July 1 Tropical Storm Near Bret 
1983 September 30 Tropical Storm In Dean 
1984 September 14 Tropical Storm Near Diana 
1985 August 19 Extratropical In Danny 
September 26-27 Tropical Storm Near Gloria 
1986 August 17-18 Tropical Storm Near Charley 
* (Neuman et al., 1987) 
APPENDIX B 
HISTORICAL SHORELINE DATA SOURCES 
l. National Ocean Survey Topographic Charts 
ISLAND YEAR MAP NUMBER MAP SCALE 
Barren 1848 T 255 1:20,000 
1901-2 T 2564 1:20,000 
1929 T 4445 1:10,000 
1943 T 8117 1:20,000 
Bloodsworth 1848 T 269 1:20,000 
1901 T 2558 1:20,000 
1942 T 8135 1:20,000 
Hooper 1848 T 265 1:20,000 
1848 T 255 1 :20 ,000 
1901 T 2564 1:20,000 
1942 T 8136 1:20,000 
1942 T 8118 1:20,000 
James 1848 T 250 1:20,000 
1901 T 2561 1:20,000 
1941 T 5718 1:10,000 
Poplar 1846 T 215 1:20,000 
1899 T 2293 1:20,000 
1941 T 5723 1:10,000 
Smith 1849 T 271 1:20,000 
1901 T 2556 1:20,000 
South Marsh 1849 T 269 1:20,000 
1901 T 2558 1:20,000 
1942 T 8135 1 :20,000 
1942 T 8149 1:20,000 
169 
2. Aerial Photographs 
ISLAND DATE SCALE NUMBER 
Barren 1952 1: 7920 ANJ-6K-20 
1964 1:7920 ANJ-4EE-135 
1987 1:7920 NAPP-B-142B 
Bloodsworth 1952 1:12000 ANJ-lK-63, ANJ-
lK-74, ANJ-lK-
134, ANJ-lK-
136, ANJ-lK-72, 
ANJ-lK-65 
1987 1:12000 NAPP-10-135D, 
NAPP-10-135C, 
NAPP-13-75B, 
NAPP-13-75A 
Hooper 1952 1:12000 ANJ-5K-186, 
ANJ-5K-188, 
ANJ-5K-133, 
ANJ-5K,125, 
ANJ-5K-127, 
ANJ-5K-43 
1987 1:12000 NAPP-13-135-TC, 
NAPP-13-135-BC, 
NAPP-13-81L, 
NAPP-13-80-ECIW 
James 1952 1:7920 ANJ-lK-03 
1964 1:7920 ANJ-4EE-177 
1987 1:12000 NAPP-14-3JC 
Poplar 1952 1:7920 AHY-5K-101 
1964 1:7920 AHY-4DD-277 
1987 1:7920 NAPP-14-64B 
Smith 1952 1:12000 ANL-5K-10, ANL-
5K-28, ANL-5K-
56, ANL-5K-58, 
ANL-5K-26, ANL-
5K-12 
1987 1:12000 NAPP-10-129A, 
NAPP-10-128R, 
NAPP-10-129D, 
NAPP-10-128L, 
NAPP-10-130D, 
NAPP-10-130A 
South Marsh 1952 1:12000 ANL-5K-35, ANL-
5K-04, ANL-5K-
170 
33, ANL-SK-31 , 
ANL- 5K- 06 
1987 1:12000 NAPP- 10-134-EC, 
NAPP-10-133-EC 
3. USGS 1.s Minute Quads 
ISLAND DATE QUAD NAME 
Hooper 1972 Richland Point 
1985 Honga 
Smith 1 972 Ewel l 
19 72 Great Fox Island 
1972 Kedges Straits 
1972 Terrapin Sand Point 
South Marsh 1972 Kedges Straits 
17 1 
REFERENCES 
Associated Press, 1991. Bird island threatened by tides. 
Calhoun County Times, June 6, 1991. 
Baumann, R.H., 1980. Mechanisms of maintaining marsh 
elevation in a subsiding environment. MS Thesis, 
Louisiana State University, Baton Rouge, LA. 91 pp. 
Bruun, P. 1962. Sea-level rise as a cause of shore 
erosion. J. Waterways and Harbors Div. ASCE 88:117-
130. 
Cahoon, D.R. and J.H. Cowan, Jr., 1988. Environmental 
impacts and regulatory policy implications of spray 
disposal of dredged material in Louisiana wetlands. 
Coastal Management, 16:341-362. 
Carter, C.H. and D.E. Guy, 1988. Coastal erosion: 
processes, timing and magnitudes at the bluff toe. 
Marine Geology, 84:1-17. 
Clow, B. and S.P. Leatherman, 1984. Metric Mapping: An 
automated technique of shoreline mapping. 
Proceedings of the American Society of 
Photogrammetry, Washington, D.C., pp. 309-318. 
Coleman, S.M. and R.M. Mixon, 1988. The record of major 
quaternary sea-level changes in a large coastal 
plain estuary, Chesapeake Bay, Eastern United 
States. Paleogeography, Paleoclimatology, 
Paleoecology, 68:99-116. 
Coleman, S.M., J .P.Halka, C.H.Hobbs, R.B. Mixon, and 
D.S. Foster, 1990. Ancient Channels of the 
Susquehanna River beneath the Chesapeake Bay and 
Delmarva Peninsula. Geological Society of America 
Bulletin, 102:1268-1279. 
Cronin, W.B., 1988. Barren Island. Chesapeake Bay 
Magazine, pp. 66-68. 
Crowell, M., S.P. Leatherman, and M. Buckley, 1991. 
Historical shoreline change: Error analysis and 
accuracy. Journal of Coastal Research, 7:839-852. 
Davis, G.H., 1987. Land subsidence and sea-level rise on 
the Atlantic coastal plain. Environmental Geology 
and Water Science, 10:67-80. 
172 
Davis, R.A., jr., and H.E. Clifton, 1987. Sea-level 
change and the preservation potentia l o f wave-
dominated and tide-dominated coastal sequences. in 
Sea-level Fluctuation and Coastal Evolut i on. D. 
Nummedal, o. Pilkey, and J. Howard (eds.), Societ y 
of Paleontol ogists and Mineralogists. p. 167-178. 
Davison, A.T., 1990. Historic and future land loss and 
sea-level rise at Smith Island, Maryland, 
unpublished manuscript. 
DeLaune, R.D., R.H. Baumann and J.G. Gosselink, 1983. 
Relationship among vertical accretion, coastal 
submergence and erosion in a Louisiana Gulf coastal 
marsh. J. Sed. Petrol., 53:147-157. 
Dionne, J.C., 1986. Recent erosion of the intertidal 
marsh in the St. Lawrence estuary, Quebec. 
Geographie physique et Quaternaire, 40:307-323. 
Donoghue, J.F., 1981. Estuarine sediment transport and 
Holocene depositional history, Upper Chesapeake 
Bay, Maryland. Estuaries, 4: 271. 
Douglas, B.C., 1991. Global sea-level rise. Journal 
Geophysical Research, 96:6981-6992. 
Downs, L.L., in prep. Chesapeake Bay coast of Calvert 
County, Maryland: Historic shoreline analysis and 
determination of littoral cell and subcell 
boundaries. MS Thesis, University of Maryland, 
College Park, Maryland. 
Emery, K.O. and D.G. Aubrey, 1991. Sea levels, Land 
Levels, and Tide Gauges. Springer-Verlag, New York. 
Federal Emergency Management Agency, 1980. Flood 
insurance study for Somerset County, Maryland. 
Federal Insurance Administration, Washington, D. C. , 
12 pp. 
Finkelstein, K. and C.S. Hardaway, 1988. Late Holocene 
sedimentation and erosion of estuarine fringing 
marshes, York River, Virginia. Journal of Coastal 
Research, 4:447-456. 
French, G., 1990. Historical shoreline changes in 
response to environmental conditions in west 
Delaware Bay. Masters thesis. University of 
Maryland, College Park. 243 pp. 
173 
Froomer , N., 1980. Sea level changes in Chesapeake Bay 
during historic times. Marine Geology, 36:289-305. 
GEO-RECON International, 1980. Cultural Resource Surve y 
of U.S. Naval Reservation Bloodsworth I s land, 
Dorchester County, Maryland. Report to Maryland 
Historical Trust, Annapolis, Maryland. 110 pp. 
Grove, J.M., 1988. The little Ice Age. New York, 
Methuen. 498 pp. 
Hands, E. G., 1983. Erosion of the Great Lakes. In 
Komar, P.D. (ed.) CRC Handbook of Coastal Processes 
and Erosion. 
Holdahl, s. R. and N. L. Morrison, 1974. Regional 
investigations of vertical crustal movements in the 
U.S., using precise relevelings and mareograph 
data. Tectonophysics, 23: 373-390. 
Intergovernmental Panel on Climate Change, 1990. 
Scientific assessment of climate change. Report 
prepared for IPCC Working Group I, pp. 264-285. 
Kearney, M., R. Grace and J.C. Stevenson, 1988. Marsh 
loss in the Nanticoke Estuary, Chesapeake Bay. 
Geographic Review, 78:205-220. 
Kearney, M. and J.C. Stevenson, 1991. Island land loss 
and marsh vertical accretion rate. Evidence for 
historical sea-level change in the Chesapeake Bay. 
Journal of Coastal Research, 7:403-415. 
Kehrin, R.T., J.P. Halka, D.V. Wells, E.L. Hennessee, 
P. J. Blakeslee, N. Zol tan and R.H. Cuthbertson, 
1987. Surficial sediments of the Chesapeake Bay: 
physical characteristics and sediment budget. 
Department of Natural resources, Investigation No. 
48. 82 pp. 
Komar, P.D. 1976. Beach Processes and Sedimentation. 
Prentice-Hal?l, Inc. , New Jersey. 42 9 pp. 
Komar, P. D. , 19 8 3 . Beach processes and erosion - an 
introduction. in CRC Handbook of Coastal Processes 
and Erosion, pp. 1-20. 
Krementz, D.G., V.D. Stotts, D.B. Stotts, J.E. Hines and 
S.L. Funderburk, 1991. Historical changes in laying 
date, clutch size, and nest success of American 
black ducks. Journal of Wildlife Management, 
55:462-466. 
174 
Laboratory for Coastal Research, 1990. Metric Mapping 
. Users Guide. University of Maryland, College Park. 
70 pp. 
Leatherman, S. P. and B. Clow, 1983. UMD, Shoreline 
mapping project. IEEE Geoscience and Remote Sensing 
Society Newsletter, pp. 5-8. 
Leatherman, s. P., 1984. Coastal geomorphic responses to 
sea-level rise: Galveston Bay, Texas, in Greenhouse 
Effect and Sea-Level Rise, M. C. Barth and J. G. 
Titus, (eds.). Van Nostrand Reinhold, New York, Pp. 
151-178. 
Leatherman, S.P., 1991. Modelling shore response to 
sea-level rise on sedimentary coasts. Progress in 
Physical Geography, 14:447-464. 
Leatherman, S.P., 1992. Global Warming and Sea-Level 
Rise: Implications for the Chesapeake Bay Region. 
A scientific overview submitted to the Chesapeake 
Bay Estuary Program, U.S. Fish and Wildlife 
Service, Annapolis, Maryland, in press. 
Marcus, F.F., 1990. on a ravaged island, the state bird 
hangs in. The New York Times, October 3, 1990. 
Mendelssohn, I.A., K.L. McKee and W.H. Patrick, 1981. 
oxygen deficiency and Spartina alterniflora roots: 
metabolic adaptation to anoxia. Science, 214:439-
441. 
Mendelssohn, I.A. and K.L. McKee, 1988. Journal of 
Ecology, 76:509-521. 
Meyer, E., 1986. The vanishing islands. Maryland lost 
and found, people and places of the Chesapeake? 
Johns Hopkins University Press, Baltimore. P? 41-
50. 
Mowbray, 1981.The Dorchester County Fact Book. Privately 
Published, 29 pp. 
National Research Council, 1990. Managing Coastal 
Erosion. National Academy Press, Washington, D.C. 
182 pp. 
Neuman, C.J., B.R. Jarvinen and A.C. Pike, 1987. 
Tropical cyclones of the North Atlantic ocean, 
18 71-1986. Historical Climatology Series, 6-2. 
National Oceanic and Atmospheric Administration, 
Coral Gables, Florida. 62 pp. 
175 
Newman, w. s., L. J. Cinquemani, R.R. Pardi, and L. F. 
Marcus, 1980. Holocene delevelling of the United 
States? east coast. in Merner, N. A. (ed.), Earth 
Rheology, Isostacy and Eustasy. John Wiley, New 
York. Pp. 449-463. 
Nordstrom, K.F., 1980. Cyclic seasonal beach response: a 
comparison of oceanside and bayside beaches. 
Physical Geography, 1:177-196. 
Orson, R., w. Pan~geotou, ands. P. Leatherman, 1985. 
Response of tidal salt marshes of the U.S. Atlantic 
and Gulf coasts to rising sea levels. J. Coastal 
Research, 1: 29-37. 
Orson, R.A., w.s. Warren and W.A. Niering, 1987. 
Development of a tidal marsh in a New England river 
valley. Estuaries: 10:20-27. 
Owens, J.P. and c.s. Denny, 1979. Upper Cenozoic 
Deposits of the Central Delmarva Peninsula 
Maryland and Delaware. U.S. Geological Survey; 
Professional Paper No. 1067-A. 28 pp. 
Pardi, R.R., L. Tomecek, and w.s. Newman, 1984. Queens 
College Radiocarbon dates IV. Radiocarbon, 26:412-
430. 
Pendleton, E.C. and J.C. Stevenson, 1983. Investigations 
of marsh loss at Blackwater Wildlife Refuge. Final 
Report submitted to U.S. Fish and Wildlife Service 
and Coastal Resources Division, Tidewater 
Administration, 213 pp. 
Penland, s., J .R. Suter and R. Boyd, 1985. Barrier 
island arcs along abandoned Mississippi River 
deltas. Marine Geology, 63:197-233. 
Phillips, J. D., 1986. Coastal submergence and marsh 
fringe erosion. J. Coastal Research, 2: 427-436. 
Rampino, M.E. and J. Sanders, 1981. Episodic growth of 
Holocene tidal marshes in the northeast United 
States: A possible indicator of eustatic sea level 
fluctuations. Geology, 9:63-67. 
Redfield, A.C. and M. Rubin, 1962. The age of salt marsh 
peat and its relation to recent changes in sea 
level at Barnstable, Massachusetts. Proceedings of 
the National Academy, 48:1728-1735. 
176 
Redfield, A.C., 1972. Development of a New England 
salt-marsh. Ecol. Monog., 42:201-237. 
Reed, D.J., 1988. Sedimentation dynamics and deposition 
in a retreating coast al salt marsh. Coastal Shelf 
Estuarine Science, 26:67-79. 
Reed, D.J., 1990. The impact of sea-level rise on 
coastal salt marshes. Progress in Physical 
Geography, 14: 465-481. 
Rosen, P. S., 1978. A regional test of the Bruun Rule on 
shoreline erosion. Marine Geology, 26: M7-Ml6. 
Rosen, P.S., 1980. Erosion Susceptibility of the 
Virginia Chesapeake Bay Shoreline. Marine Geology, 
34:45-59. 
Ryan, J. D ? , 19 5 3 . The Sediments of Chesapeake Bay. 
Maryland Department of Mines and Water Resources 
Bulletin 12, 120 pp. 
Shalowitz, A.L., 1964. Shore and Sea Boundaries, Volume 
2. U.S. Department of Commerce, Publication No. 10-
1, u. S. Government Printing Office, Washington, 
D.C ?. 749 pp. 
Singewald and Slaughter, 1949. Shore erosion and 
measurement in Maryland. Maryland Department of 
Geology, Mines and Mineral Resources Bulletin, 
Volume 6. 128 pp. 
Stevenson, J.C., L. Ward, M. Kearney, and T.E. Jordan, 
1985. Sedimentary processes and sea-level rise in 
tidal marsh systems of Chesapeake Bay. in H. A. 
Gorman, et al. (eds), Wetlands of the Chesapeake. 
Proceedings of the Conference. Environmental Law 
Institute, Washington, D.C. p. 37-62. 
Stevenson, J.C., L. Ward and M. Kearney, 1986. Vertical 
accretion rates in marshes, in D.A. Wolfe (ed.), 
Estuarine Variability. Academic Press, New York, 
pp. 241-259. 
Stotts, V.D., 1985. Values and Functions of Chesapeake 
Bay wetlands for waterfowl, in H.A. Gorman, et al. 
(eds.), Wetlands of the Chesapeake. Proceedings of 
the Conference. Environmental Law Institute, 
Washington, D.C. 
U.S. Army Corps of Engineers, 1984. Chesapeake Bay Tidal 
Flooding Study. Baltimore District. 90 pp. 
177 
u. s. Army Corp s of EngineE3rs, 1984. Shore Protection 
Manual. Coastal Engineering Research Center , 
waterways Experiment Station, Vicksburg, 
Mississippi. 2 Volumes. 
u.s. Army corps of Engineers, 1991. Chesapeake Bay 
Shoreline Erosion Study, Baltimore District. 116 
pp. 
U.S. Department of Agriculture, Soil Conservation 
Service, 1966. Soil survey of Somerset County, 
Maryland. 90 pp. 
U.S. Department of Commerce, National Oceanic and 
Atmospheric Administration, 1990. Local 
climatological Data 1990. Annual Summaries. 
Rockville, Maryland. 
u. s. Fish and Wildlife Service, undated draft. American 
black duck nesting and brooding habitat enhancement 
project briefing statement. 
Wang, H., R. Dean, R. Dalrymple, R. Biggs, M. Perlin, 
and V. Klemas, 1982. An Assessment of Shore 
Erosion in Northern Chesapeake Bay and of the 
Performance of Erosion Control Structures. Final 
Report. Tidewater Administration, Coastal 
Resources Division, Maryland Department of Natural 
Resources, Annapolis, Maryland. 258 pp. 
178