Paleoecology of Calf Island in Boston’s Outer Harbor
WILLIAM A. PATTERSON III1,*, JULIE A. RICHBURG
1, KENNEDY H. CLARK
1,
AND SALLY SHAW
1
Abstract - We used microfossils preserved in salt-marsh peat to understand
the landscape processes (both natural and anthropogenic) that have influenced
the environment. Variations in the abundance of fossil pollen of native species
suggest that the vegetation of this small, exposed island has been dominated
by low, shrubby vegetation since before the arrival of Europeans in the early
1600s. Increases in non-native species since that time may reflect disturbance
of the soil associated with grazing and other activities. Sorrel (Rumex) pollen,
which indicates local grazing, declines by the late 19th century, whereas Chenopodiaceae/
Amaranthaceae pollen, an indicator of disturbed soil, is most abundant
since 1900. Charcoal abundance shows that fires, probably ignited by humans
both before and after 1600 A.D., have burned on the island throughout the last
1000 years. In addition, increases in soot and opaque spherules in sediments
reflect increased air pollution during the last 100 years. Our analyses provide
benchmarks for modern management by documenting pre-European conditions
as well as the extent to which the modern environment differs from that prior to
the settlement of Massachusetts Bay by Europeans.
Introduction
Coastal vegetation changes over time periods ranging from nearly instantaneous
to almost imperceptibly long. Overwash due to storm surges
can obliterate upland vegetation in minutes to initiate new “primary”
successions (Zaremba 1982), whereas changes in sea level can cause
variations in the patterning of high and low marsh vegetation that are
perceptible only after many decades or centuries (Clark and Patterson
1984). Understanding and documenting changes in these environments
requires knowledge not only of the biology of the species involved and
the factors that drive succession, but also the time frames over which
changes are likely to occur. In our broader study of the vegetation history
of the Boston Harbor Islands (Richburg and Patterson 2005), we
used written records, photographs, and maps to describe vegetation of
the islands since about 1600 A.D. In this paper, we examine the history
of the vegetation and environment of Calf Island—a small, windswept
island in the Brewster group (Fig. 1)—in greater detail over a longer
period of nearly 1200 years.
1Department of Natural Resources Conservation, 214 Holdsworth Natural Resources
Center, University of Massachusetts, Amherst, MA 01003-0421. *Corresponding
author - wap@forwild.umass.edu.
Boston Harbor Islands National Park Area: Natural Resources Overview
2005 Northeastern Naturalist 12(Special Issue 3):31–48
32 Northeastern Naturalist Vol. 12, Special Issue 3
The earliest descriptions of the Boston Harbor Islands, including
those by John Smith, Samuel de Champlain, and William Wood (Barbour
1986, Winship 1968, Wood 1993), suggest the nature of vegetation
in the early 17th century. These observers described a landscape that
was primarily forested, but with open areas maintained around Native
American settlement sites, on the larger islands, with the smaller islands
dominated by shrubs (Richburg and Patterson 2005). But given the dynamic
coastal environment (Clark 1986), we know that vegetation was
almost certainly changing rapidly. Even so, Europeans profoundly altered
vegetation (Altpeter 1937, Westveld et al. 1956) and the processes
that influenced it (Brugam 1978). To better understand the pre-European
landscape of one of Boston Harbor’s islands, and the processes that
influenced the vegetation, we studied a peat core from a small marsh
on Calf Island. Fossil analyses, which we report here, provide insights
beyond our study of the recorded history of the Boston Harbor Islands
group as a whole (Richburg and Patterson 2005).
Field Site Description
The Boston Harbor Islands lie in Massachusetts Bay and range in
size from less than 0.4 hectare to 105 hectares. Calf Island, an island of
Figure 1. Map of the Boston Harbor Islands (dark shading) showing the location
of Calf Island in the outer harbor.
2005 W.A. Patterson III, J.A. Richburg, K.H. Clark, and S. Shaw 33
7.5 ha, lies in the outer harbor as part of a group called the Brewsters
(Fig. 1). These islands are more exposed to weather than some of the
larger islands in the inner harbor. There are no good historical descriptions
of the vegetation of Calf Island and no archaeological evidence for
cultivation (Luedtke 1980). Houses were built on the island by at least
1860 (Shurtleff 1891, Stark 1880; see Fig. 2), including a Colonialstyle,
two-story summer estate built in 1902 for Benjamin Cheney and
Julia Arthur. Since the estate was burned by vandals in the 1940s, the
island has been abandoned. Nineteenth-century accounts of the island
describe groves of wild-cherry trees, sumac, and manicured lawns. Once
the island was abandoned during the later part of the 1900s, the vegetation
became weedy and overgrown with sumac and introduced species
of trees, low scrubby weeds, and high grasses (Kales and Kales 1983,
Luedtke 1975, Mikal 1973).
The modern vegetation of Calf Island (Elliman 2005) is comprised
of 90 species of vascular plants. Of these, 47 are native and 53 nonnative.
There are 11 woody species including grape (Vitis sp.), poison ivy
(Toxicodendron radicans L. Kuntze), two willows (Salix spp.), raspberry
(Rubus idaeus L.), blackberry (R. allegheniensis Porter), staghorn sumac
(Rhus hirta L. Sudworth), the non-native Morrow’s honeysuckle (Lonicera
morrowii Gray), and three species of Rosa including one that is nonnative
(nomenclature follows Magee and Ahles 1999).
Figure 2. US Coast and Geodetic
Survey map of Calf
Island, 1860. Structures are
shown near the island’s south
shore. The Calf Island marsh
that we cored bisects the
northern half of the island.
34 Northeastern Naturalist Vol. 12, Special Issue 3
Elliman describes the modern landscape as including thickets, rough
old fields, exposed ledges, rocky beachfront, sandy beach strand, a large
brackish marsh in the central part of the island, and a moist semi-open
field south of the marsh (T. Elliman, Slingerlands, NY, pers. comm.). He
describes the upland vegetation as a whole as maritime shrub (Elliman
2005), with extensive thickets of staghorn sumac (Rhus hirta). The approximately
0.4-ha brackish marsh from which we recovered our core
nearly bisects the island (Fig. 2). It has a mixture of native plants, such
as Olney’s three-square (Scirpus americanus Pers.) and broad-leaved
cattail (Typha latifolia L.), and invasive species like purple loosestrife
(Lythrum salicaria L.) and common reed (Phragmites australis (Cav.)
Trin. ex Steud.). Salt hay (Spartina patens (Ait.) Muhl.) covers much of
the marsh.
Ditching suggests that efforts were made to drain the marsh at
some time in the past, probably for mosquito control. It is separated
from salt water not more than 15 meters to the east by a boulder
beach and berm (Fig. 3).
Methods
A peat core was collected from the Calf Island marsh on 16 October
2001. We used a modified piston corer, 10 centimeters in diameter, with a
saw-tooth barrel edge to cut through the peat. The 2-meter-long core was
brought to the laboratory and sampled at 5-to-10-centimeter intervals
for fossil pollen analysis and the quantification of combustion residues
Figure 3. Coring apparatus at the Calf Island marsh core site, October 2001. The
photograph is taken looking to the northeast, with the shoreline just beyond the
rocky berm in the background.
2005 W.A. Patterson III, J.A. Richburg, K.H. Clark, and S. Shaw 35
(charcoal, soot, and opaque spherules). One-half-cubic-centimeter subsamples
were removed and placed in centrifuge tubes, to which 0.5 ml of
non-native Eucalyptus pollen (86,750 grains/cm3 in a glycerin suspension)
was added as a reference marker. Sediments were treated with 10%
potassium hydroxide to dissolve humic compounds, 10% hydrochloric
acid to dissolve carbonates, 48% hydrofluoric acid to dissolve silicates,
and an acetolysis solution (Faegri and Iversen 1975) to remove organic
residue other than pollen, charcoal, and soot. Residual material was
mounted in silicone oil and mixed thoroughly. Slides were examined at
400x with a Zeiss Standard light microscope. Fossil pollen grains were
identified with the aid of illustrated keys (Kapp 1969, McAndrews et
al. 1973) and reference material in the Paleoecology Laboratory at the
University of Massachusetts. While identifying fossil pollen grains, we
observed that grains of the Chenopodiaceae/Amaranthaceae (Cheno/
Am) families appeared to vary, systematically, in numbers and size
throughout the core, with more abundant, small grains in samples from
the upper 0.5 meter of peat. To quantify this, we measured the diameters
of grains from each sample where Cheno/Am pollen comprised more
than 10% of the total fossil pollen identified. We also measured the
diameters of reference pollen for eight species recently recorded as present
on Calf Island. Average grain sizes for fossil pollen were compared
with the diameters of reference pollen.
We separated charcoal produced by the combustion of plant material
(i.e., dead leaves, grass, and woody material) from soot produced by the
combustion of fossil fuels based on morphological characteristics (Fig.
4). Charcoal and soot concentrations were quantified using the grid-count
method (Patterson et al. 1987), which estimates the surface area of char-
Figure 4. Charcoal (upper
left), soot (middle
right), and opaque spherules
(lower left) shown
at 400X on a slide from
Calf Island peat prepared
for fossil analysis. Soot
is distinguishable from
charcoal by its perforated
nature and lack of any
cellular structure. For
scale, the opaque spherule
is approximately 20 micrometers
in diameter.
36 Northeastern Naturalist Vol. 12, Special Issue 3
coal fragments on microscope slides. Eucalyptus grams were tallied at the
same time that fossil pollen and charcoal were counted. The ratio of charcoal
or soot area divided by the number of grains of Eucalyptus pollen, to
fossil pollen divided by Eucalyptus pollen yields the ratio of charcoal or
soot area to fossil pollen (μm2Ch:P or μm2So:P), which is used as a measure
of past fire or air pollution occurrence on the landscape. The abundance
of opaque spherules (most likely from industrial combustion; see
Huhn [1974] and Nriagu and Bowser [1969]) was determined to identify
the depth in the sediments associated with early-20th-century industrial
activity (Brugam 1975, Clark and Patterson 1984).
A radiocarbon date (standard AMS dating by Beta Analytic Inc.,
Miami, fl) was obtained for a section of peat taken from 90.5 to 93.5
cm below the marsh surface. Sediment accumulation rates were calculated
for the entire section using this date. Chronostratigraphic markers,
including increases in the abundance of pollen indicative of agricultural
activity (Brugam 1978), the decline in chestnut (Castanea) pollen associated
with the early-20th-century chestnut decline (Anderson 1974),
and increases in the abundance of opaque spherules, were also used to
calculate the sediment accumulation rates for sections of the upper portion
of the core. Calculated sediment accumulation rates combined with
fossil concentrations determined with the aid of the added Eucalyptus
grains were used to calculate the yearly accumulation of pollen and
combustion residues.
Results
Sediment dating
We focused our analyses on the upper 1 meter of the core because
we were primarily interested in documenting how the island’s landscape
has changed in the several hundred years spanning the arrival
of European settlers in the Boston area. The upper core section, as described
in the field, was of uniform, humified sedge peat. We observed
no obvious discontinuities or dark bands suggesting discrete layers of
charcoal. Fossil data for 17 samples between 0 and 95 cm in the core
were summarized as percentage (Fig. 5) and fossil accumulation rate
(Fig. 6) values using TILIA software (Grimm 1992). Abundant pollen
of species indicative of agricultural activity (ragweed [Ambrosia],
sorrel [Rumex], and plantain [Plantago]) above 50 cm indicate the
period under influence of European settlers. This “settlement horizon”
has been identified for sediments throughout North America, and for
southern New England it represents the early 17th through 20th centuries
(Brugam 1978). The increase in opaque spherules relative to fossil
pollen at 20 cm and the decline in chestnut (Castanea) pollen above
20 cm indicate the past 100 years of peat accumulation in the marsh.
2005 W.A. Patterson III, J.A. Richburg, K.H. Clark, and S. Shaw 37
A date of 1290 ± 40 radiocarbon years before present (Beta-177807)
yields a minimum (2 sigma calibration) calendar age of 840–860 A.D.
for the 92 cm depth. These ages are indicated on the depth (y) axis of
Figure 5. Fossil pollen percentage diagram, plus the ratios of charcoal and soot
to fossil pollen for the Calf Island marsh (• indicates the presence of a single
sumac grain).
38 Northeastern Naturalist Vol. 12, Special Issue 3
Figure 6. Pollen accumulation rate diagram, plus accumulation rates of charcoal
and opaque spherules, for the Calf Island marsh. Agricultural indicators include
ragweed (Ambrosia), sorrel (Rumex), and plantain (Plantago).
2005 W.A. Patterson III, J.A. Richburg, K.H. Clark, and S. Shaw 39
the percentage diagram (Fig. 5). Assuming a date of 1900 A.D. for 20
cm and 1635 A.D. for 50 cm plus the radiocarbon age for 92 cm yields
approximate age/depth associations for the following sections of the
core: 5.05 years/cm for 0–20 cm; 8.8 years/cm for 20–50 cm; and 18.5
years/cm for 50–92 cm. Apparent increases in rates of peat accumulation
in the upper portions of the core are likely due to increased erosion
from uplands with European settlement activities on the island and the
lack of consolidation compared to that of the lower sediments which
have been subjected to a greater degree of compaction and decomposition.
Ages calculated from the above rates are shown on the y-axis of
the accumulation rate diagram (Fig. 6).
Vegetation history
We included all types plus unidentifiable and unknown grains in
the pollen sum, as we were interested in changes at the local, extralocal,
and regional scales, but poor preservation limited our ability to
count and identify large numbers of grains in some samples. Fossil
pollen sums averaged 312 grains for the 17 samples, although sums
for the 5, 10, 40, and 50 cm samples fell to 150–200, chiefly due to
poor preservation. Unidentifiable grains averaged 30% of the pollen
sum in these four samples.
Although pollen preserved in small, isolated basins like the Calf
Island marsh are more likely to represent local rather than regional
vegetation (Jackson 1990, Jacobson and Bradshaw 1981), pollen in the
Calf Island peat appears to document both regional and local changes in
vegetation, both before and after the arrival of Europeans in the early 17th
century. The abundance of and changes in major tree species are typical of
those seen in cores from other south coastal New England sites (Backman
1984, Brugam 1975, Motzkin et al. 1993, Stevens 1996, Winkler 1985).
Oak (Quercus) percentages of 10% to 20% and percentages for pine (Pinus),
birch (Betula), hickory (Carya), and hemlock (Tsuga) of less than
10% suggest that species represented by these types were unimportant
before Europeans settled the Boston area. Birch, oak, and especially pine
tend to be overrepresented in fossil pollen assemblages, and pine foliage
is very sensitive to salt spray. So it seems unlikely that these species were
common, if present at all, on Calf Island, which is small and as one of the
Brewsters exposed to the harsh maritime climate of Massachusetts Bay.
Declines in these types between 15 and 30 cm suggest the regional deforestation
that occurred during the late 18th and early 19th centuries. Increases
in the abundance of pine and oak at 10 and 5 cm, respectively, reflect
the recent reforestation of southern New England following agricultural
abandonment in the late 19th century.
Pollen accumulation rates calculated from sediment accumulation
rates and pollen concentration (Davis 1965, 1969) support percentage
40 Northeastern Naturalist Vol. 12, Special Issue 3
data which may be difficult to interpret due to their relative nature;
i.e., if one type increases in absolute terms while others remain the
same, the abundances of the others will appear to decline as percentages
are adjusted downward to reflect the increase in the more
abundant type. This can be a particular problem in cores from wetlands
where an abundant pollen producer growing at the core site can
mask changes (or the lack thereof) in the abundances of species that
are common outside the marsh. Our data (Fig. 6) show, for example,
that the rate of oak (Quercus) pollen accumulation averages approximately
2300 grains/cm2/yr at Calf Island, whereas Davis (1969)
reports values for Rogers Lake in southern Connecticut that average
approximately three times that figure. Rogers Lake is surrounded by
oak forests, and the comparison suggests that oak probably was not
present on Calf Island, despite the fact that oak pollen percentages
were as high as 15% to 20% before 1600 A.D.
Of the woody species on the island today, sumac (Rhus hirta) is the
most common. It is a poor pollen producer, and pollen grains of different
species of Rhus are largely indistinguishable (poison ivy [Rhus radicans,
now Toxicodendron radicans] is also present on the island). Single
Rhus grains were identified from peat that dates both to before and after
settlement (at 90, 60, 55, and 10 cm in the peat), but little can be said of
the varying abundances of sumac (and/or perhaps poison ivy) based on
the few grains identified, other than the fact that Rhus has surely been
present on the island for many centuries.
Grass (Gramineae) pollen is abundant, with accumulation rates
more than twice those for oak, both before and after European settlement,
suggesting that grass species dominated the marsh, and perhaps
the upland, throughout the period represented by the core. With few
exceptions, it is not possible to identify grass pollen to species. Pollen
of corn (Zea mays L.) is much larger than pollen of other members of
the grass family. Although grass pollen of several different and probably
distinct size classes was observed, no corn pollen was identified.
Sparse pollen of agriculture indicators below 50 cm suggests that there
was little Native American agricultural activity on Calf Island, not surprising
given its small size and isolation at the outer entrance of Boston
Harbor. Only at 75 cm does Ambrosia pollen rise above 1% (to 1.9%),
and there are no other indicators of disturbed soil (e.g., abandoned
agricultural land) at this level suggesting that the ragweed (Ambrosia)
pollen may be from regional rather than local sources. Sorrel (Rumex),
an indicator of grazing (Brugam 1978), first appears at 45 cm (1%) and
peaks at 4.7% at 40 and 30 cm before falling to 1.7% at the start of the
20th century (20 cm). This probably indicates local grazing of livestock
(perhaps sheep?) in the 18th and 19th centuries. Absence of sorrel pollen
2005 W.A. Patterson III, J.A. Richburg, K.H. Clark, and S. Shaw 41
above 20 cm is consistent with a lack of other evidence for grazing on
the island in the 20th century.
Pollen of Compositae (goldenrods, asters, and the like; but not including
wormwood [Artemisia] and ragweed [Ambrosia] which were
tallied separately) fluctuates throughout the core and probably represents
variations in the abundance of native species like the seaside
and Canada goldenrods (Solidago sempevirens L. and S. canadensis
L.) and Canada hawkweed (Hieracium canadense Michx.), all of
which inhabit the island today and are common in coastal environments.
The facultative wetland shrub salt bush (Baccharis halimifolia
L.) is not present on the island today, but may have been in the past,
as it is common in coastal wetlands elsewhere in the region. Composites
are most abundant in presettlement sediments, with percentages
in excess of 5–10% at 60, 70, and 95 cm. Although 12 non-native
species of the Composites occur on the island today, percentages
exceed 5% only once (at 40 cm) in the past 400 years. In the most
recent sample (1 cm), it is 1.8%. Pollen grains of the many species of
composites now on the island are not distinguishable at the genus or
species level.
The abundances of pollen of Chenopodaceae and Amaranthaceae
(Cheno/Am) fluctuate throughout the core and are consistently abun-
Table 1. Average diameters (of N grains) for pollen of members of the Chenopodiaceae and
Amaranthaceae (C/A) species. Included are values for reference pollen of species found
on Calf Island today, as well as average values for peat samples with > 10% C/A fossil
pollen. Species/samples are ranked by average diameter. In genus/species column NA =
native, E = non-native. * Indicates species currently found in the marsh. Others are found
on the beach strand or in upland thickets (Amaranthus retroflexus L.). Suaeda richii Fern.
occurs, but is rare on the island today.
Average
Depth (cm) (%C/A) diameter (μm) N (#) Genus/species
10 (17.7%) 17.1 16 -
5 (14.0%) 17.7 7 -
15 (41.3%) 17.8 21 -
20 (10.4%) 18.3 9 -
1 (40.5%) 19.2 25 -
75 (29.4%) 19.2 35 -
- 20.4 50 S. richii (NA)
- 21.1 100 Chenopodium ambrosiodes L. (E)
- 22.5 50 Atriplex patula L. (E)
- 22.6 50 Suaeda linearis Ell. Moq. (NA)*
- 23.2 50 Bassia hirsuta (E)
- 23.2 50 Amaranthus retroflexus (E)
50 (10.4%) 23.5 9 -
- 24.0 50 S. maritima L. Dum. (NA)*
- 25.8 92 C. album L. (E)
45 (11.2%) 26.5 20 -
42 Northeastern Naturalist Vol. 12, Special Issue 3
dant in 20th-century sediments, where average diameters are smaller
than in older sediments. The diameters of members of these families
varies among species, and because several native and non-native
species occur on the island today, we attempted to identify patterns
related to the abundance of native and non-native species. Fossil
grains at 1 through 20 cm in the core are smaller than those below
and are smaller than any of our reference types, either native or nonnative
(Table 1). Most members of the two families currently on the
island are found on the beach strand, although the two native Suaeda
species, tall and salt-marsh sea-blite (S. linearis and S. maritima), are
found both on the strand and in the marsh. Bassia hirsuta L. Aschers
is also common on the marsh (T. Elliman, Slingerlands, NY, pers.
comm.). The pollen grains of these three species are larger than those
of the fossil pollen found in 20th-century sediments, however (see Table
1). The smallest fossil grains are in samples from 5 to 5 cm (25 to
75 years ago). It is possible that an unknown but probably non-native
species was present until very recently. A precedent for such a decline
is found in the rise and fall in pollen of sorrel, which we attribute to
changes in land use (an increase and then decline in sheep grazing).
An as yet unidentified Chenopodiaceae/Amaranthaceae species may
have followed a similar pattern in the 20th century. The abundance of
one or more small-grained species at 75 cm (ca. 1175 A.D., see Figs.
5 and 6), suggests, however, that the recent abundance of Cheno/Am
species may not be attributable solely to recently introduced species.
Table 2. Abundances expressed as ratios of charcoal (Ch), soot (So), and opaque spherules
(OS) to fossil pollen (P) in samples from the Calf Island marsh.
Depth (cm) Ch:P (μm2) So:P (μm2) OS:P (#)
1 507 171 0.16
5 727 413 0.23
10 1080 345 0.32
15 6935 2091 1.44
20 3964 942 0.70
30 55 3 0.09
40 1139 13 0.07
45 1199 5 0
50 2367 0 0
55 4088 0 0
65 1646 0 0
60 870 0 0.1
70 6835 0 0
75 3074 0 0
80 4504 0 0
90 284 0 0.07
95 262 0 0
2005 W.A. Patterson III, J.A. Richburg, K.H. Clark, and S. Shaw 43
Fire history
The abundance of charcoal (measured as square micrometers relative
to fossil pollen) varies enormously at different depths in the core
(Fig. 5), with the highest concentrations (6935 μm2 charcoal:pollen at
15 cm) reflecting the local occurrence of fire. Charcoal to fossil pollen
ratios (Ch:P) for sites elsewhere in New England show that values
in excess of 500–1000 reflect fires which burned in local watersheds
or, in the case of wetlands, through the wetland itself (Clark and Patterson
1997, Motzkin et al. 1993, Patterson and Backman 1988, and
others). In the Calf Island core, values exceed 2000 at 80, 75, 70, 55,
50, 20, and 15 cm. The fact that values at other depths are low is not
unusual. Unlike pollen abundance, which reflects annual production
that usually changes gradually, charcoal is the product of individual
fires that are transient on the landscape. Estimates of the importance
of fire on the island from these data are almost certainly conservative,
as the identification of individual fire occurrences would require sampling
the entire core at 1-cm intervals, an extremely time-consuming
and costly exercise which was beyond the scope of our study.
Environmental pollution
Few charcoal reconstructions done in conjunction with fossil pollen
analysis have attempted to identify the by-products of fossil fuel
combustion as distinct from charcoal produced in wildland fires.
We earlier identified what we referred to as “opaque spherules” as
a by-product of industrial pollution. Increases in opaque spherules
in Long Island marsh peat were used to date 20th-century sediments
(Clark and Patterson 1984). In this study, we also separated what we
refer to as “soot” from opaque spherules and charcoal (Table 2). Both
soot and opaque spherules rise sharply above 30 cm, as does charcoal
abundance. All three measures of combustion co-vary for the five
20th-century samples. Correlation coefficients are 0.93 to 0.99 for the
three comparisons.
Discussion
The vegetation of Calf Island has been influenced by disturbances
ranging from changing sea level and temperatures since the retreat
of glaciers to human use and modification of the island’s natural
resources. Although this small island in the outer harbor may have
had less human occupation than other islands within Boston Harbor,
evidence of prehistoric fire is found in the abundant charcoal in the
pre-European peat. Lightning rarely starts fires in the coastal environment
(Patterson 1984), so we assume that fires were started by
Native Americans who visited the island and perhaps established at
44 Northeastern Naturalist Vol. 12, Special Issue 3
least temporary encampments to support hunting or fishing activities.
Although some fires may have spread from fires lit for cooking or
warmth, Native Americans were also observed to ignite fires in natural
fuels for a variety of reasons including reducing brush for ease of
walking, burning around wetlands to attract waterfowl in the spring,
and reducing insect pests in late summer (Day 1953, Patterson and
Sassaman 1988).
Calf Island currently has no forests or woodlands, and the pollen
record suggests that it has been open since before 1600. Tree species
(perhaps oak and birch?) that may have been present were probably
scattered or in low, shrubby thickets rather than forming closedcanopy
forests. Although the charcoal record shows that fires burned
on the island before that time, the open character of the pre-Colonial
landscape was not necessarily the sole consequence of burning, at
whatever frequency it may have occurred. The harsh maritime climate
and exposure of Calf Island to salt spray would have made it difficult
for large trees to establish and persist on the island. The Boston area
is outside the range of the white and red spruce (Picea glauca Moench,
Voss and P. rubens Sarg.) that cloak small islands on the Maine
coast (Davis 1966), and the pollen record provides no evidence for
their presence on Calf Island.
The inner harbor islands are larger and more sheltered, and it is not
surprising that historical accounts speak of them as being wooded (Richburg
and Patterson 2005). The decline of these forests and those of the
mainland are detected in the decline in pine and oak pollen in the Calf
Island peat. Percentages and accumulation rates for these types are low,
indicating they were not locally abundant.
Measures of environmental pollution, in our case the abundances
of soot and opaque spherules, rise sharply in the 20th century, but have
declined by as much as 75% in the top two samples, which correspond
to the latter decades of the 20th century when air pollution regulations
have sought to reduce particulate pollution. Concentrations at the surface
remain well above background levels (of near 0 through at least
the early to mid-19th century; our sampling precision is no better than
50–100 years for the 20–30 cm portion of the core). Although particulate
pollution is often thought to be of only regional environmental and public
health significance, recent research (Hansen and Nazarenko 2003)
suggests that soot can change the albedo (i.e., the ability of a substance
to reflect radiation) of glaciers and continental ice masses and thus is a
factor in global climate forcing.
Paleoecologists often do not report soot separately from charcoal,
and, in fact, it is not clear how often the two are tallied separately. But
there may be benefits in doing so both for interpreting the magnitude
2005 W.A. Patterson III, J.A. Richburg, K.H. Clark, and S. Shaw 45
of environmental pollution (e.g., Davis et al. 1994) and for comparing
modern and prehistoric fire regimes. The strong correlations between
charcoal, soot, and opaque spherules for 20th-century peat and the lack
of any correlation between these parameters for earlier sediments, suggest
that some particles (as much as 20–40% for our data) in 20th-century
sediments may be from industrial pollution rather than from wildland
fires. Comparisons between pre- and post-settlement charcoal abundances
(e.g., Parshall and Foster 2002, Patterson and Backman 1988)
could be influenced by a failure to distinguish between sedimentary
carbon particles derived from wildfires and those from the combustion
of fossil fuels.
Satisfactory coring sites are few on the Boston Harbor Islands, in
part due to their small size and lack of topographic relief, but also because
of extensive alteration of wetlands during the past 200–300 years
by human activity including draining, dredging, and filling. Although
the results of our analysis of the Calf Island core do not necessarily
represent what we might find on other, especially larger, islands, they
do give us a sense of vegetation change on one island over a time period
(i.e., a few hundred years before Europeans first entered the harbor) for
which no other data are available. We do feel, however, that the changes
we document in our fossil analyses of the Calf Island peat are probably
typical of those occurring on at least the other small islands in the outer
harbor. Searches of the larger, currently more wooded islands of the inner
harbor should continue in an effort to identify appropriate sites for
paleoecological investigations like ours at Calf Island. Our studies show
that change has been a constant part of these landscapes, and that only a
lack of change could be considered to be truly unnatural.
Acknowledgments
Our research was funded by the National Park Service on behalf of the
Boston Harbor Islands Partnership and facilitated by Mary Foley and Bruce
Jacobson. We thank Mitch Mulholland and Lucinda McWeeney for assistance
in the recovery of the core. Charles Roman, Bruce Jacobson, Ronald Davis, and
two anonymous reviewers provided helpful suggestions to improve an earlier
version of the manuscript.
Literature Cited
Altpeter, L.S. 1937. A history of the forests of Cape Cod. M.F. Thesis, Harvard
University, Cambridge, MA. 90 pp.
Anderson, T.W. 1974. The chestnut decline as a time horizon in lake sediments
in eastern North America. Canadian Journal of Earth Sciences 11:678–685.
Backman, A.E. 1984. 1000-year record of fire-vegetation interactions in the
northeastern United States: A comparison between coastal and inland regions.
M.S. Thesis, University of Massachusetts, Amherst, MA. 99 pp.
46 Northeastern Naturalist Vol. 12, Special Issue 3
Barbour, P.L. 1986. The Complete Works of Captain John Smith 1580–1631.
University of North Carolina Press, Chapel Hill, NC. 3 volumes.
Brugam, R.B. 1975. The human disturbance history of Linsley Pond, North
Branford, Connecticut. Ph.D. Thesis, Yale University, New Haven, CT.
185 pp.
Brugam, R.B. 1978. Pollen indicators of land-use change in Southern Connecticut.
Quaternary Research 9:349–362.
Clark, J.S. 1986. Dynamism in the barrier-beach vegetation of Great South
Beach, New York. Ecological Monographs 56:97–126.
Clark, J.S., and W.A. Patterson, III. 1984. Pollen, Pb-210, and sedimentation in
the intertidal environment. Journal of Sedimentary Petrology 54:1249–1263.
Clark, J.S., and W.A. Patterson, III. 1997. Background and local charcoal in
sediments: Scales of fire evidence in the paleorecord. Pp. 23–48, In J.S.
Clark, H. Cachier, J.G. Goldammer, and B. Stocks (Eds.). Sediment Records
of Biomass Burning and Global Change. NATO ASI series. Series
I, Global Environmental Change; No. 51. Springer-Verlag, Berlin, Germany.
489 pp.
Davis, M.B. 1965. Determination of absolute pollen frequency. Ecology
47:310–311.
Davis, M.B. 1969. Climatic changes in southern Connecticut recorded by pollen
deposition at Rogers Lake. Ecology 50: 409–422.
Davis, R.B. 1966. Spruce-fir forests of the coast of Maine. Ecological Monographs
36:79–94.
Davis, R.B., D.S. Anderson, S.A. Norton, and M.C. Whiting. 1994. Acidity of
twelve northern New England (USA) lakes in recent centuries. Journal of
Paleolimnology 12:103–154.
Day, G.M. 1953. The Indian as an ecological factor in the northeastern forest.
Ecology 34:329–346.
Elliman, T. 2005. Vascular flora and plant communities of the Boston Harbor
Islands. Northeastern Naturalist 12(Special Issue 3):49–76.
Faegri, K., and J. Iversen. 1975. Textbook of Pollen Analysis. Hafner, New
York, NY. 295 pp.
Grimm, E.C. 1992. Tilia and Tilia-Graph: PC spreadsheet and graphics software
for pollen data. INQUA Commission for the Study of the Holocene Working
Group on Data-Handling Methods Newsletter 4:5–7.
Hansen, J., and L. Nazarenko. 2003. Soot climate forcing via snow and ice albedos.
Proceedings of the National Academies of Science 101:423–428.
Huhn, F.J. 1974. Lake sediment records of industrialization in the Sudbury area of
Ontario. M.S. Thesis, University of Toronto, Toronto, ON, Canada. 84 pp.
Jackson, S.T. 1990. Pollen source area and representation in small lakes of
the northeastern United States. Review of Palaeobotany and Palynology
63:53–76.
Jacobson, G.L., Jr., and R.H.W. Bradshaw. 1981. The selection of sites for paleovegetational
studies. Quaternary Research 16:80–96.
Kales, E. and D. Kales 1983. All About the Boston Harbor Islands. Captain
George’s Inc., Millis, MA. 120 pp.
Kapp, R.O. 1969. How to Know Pollen and Spores. W.C. Brown, Dubuque, IA.
249 pp.
2005 W.A. Patterson III, J.A. Richburg, K.H. Clark, and S. Shaw 47
Luedtke, B. 1975. Final report on the archaeological and paleobotanical resources
of twelve islands in Boston Harbor. Massachusetts Historical Commission
Report No. 25-107.
Luedtke, B. 1980. The Calf Island site. Man in the Northeast 20:25–76.
Magee, D.W., and H.E. Ahles. 1999. Flora of the Northeast. A Manual of the
Vascular Flora of New England and Adjacent New York. University of Massachusetts
Press, Amherst, MA. 1213 pp.
McAndrews, J.H., A.A. Berti, and G. Norris. 1973. Key to the Quaternary pollen
and spores of the Great Lakes region. Royal Ontario Museum Life Sciences
Miscellaneous Publication. Toronto, Ontario, Canada. 61 pp.
Mikal, A. 1973. Exploring Boston Harbor. Christopher Publishing House, North
Quincy, MA. 128 pp.
Motzkin, G., W.A. Patterson III, and N.E.R. Drake. 1993. Fire history and vegetation
dynamics of a Chamaecypris thyoides wetland on Cape Cod, Massachusetts.
Journal of Ecology 81:391–402.
Nriagu, J.O., and C.J. Bowser. 1969. The magnetic spherules in sediments of
Lake Mendota, Wisconsin. Water Research 3:833–842.
Parshall, T., and D.R. Foster. 2002. Fire on the New England landscape: Regional
and temporal variation, cultural and environmental controls. Journal
of Biogeography 29:1305–1317.
Patterson III, W.A. 1984. The research basis for fire management planning in the
Northeast. Pp. 24–32, In M.K. Foley and S.P. Bratton (Eds.). Barrier Islands
Critical Fire Management Problems, Proceedings of a Workshop. Canaveral
National Seashore, February 8–10, 1984. 41 pp.
Patterson III, W.A., K.J. Edwards, and D.J. Maguire. 1987. Microscopic charcoal
as a fossil indicator of fire. Quaternary Science Reviews 6:3–23.
Patterson III, W.A., and A.E. Backman. 1988. Fire and disease history of forests.
Pp. 603–632, In B. Huntley and T. Webb, III (Eds.). Handbook of Vegetation
Science, Volume 7: Vegetation History. Kluwer Academic Publishers,
Dordrecht, The Netherlands. 803 pp.
Patterson III, W.A., and K.E. Sassaman. 1988. Indian fires in the prehistory
of New England. Pp. 107–131, In G. Nicholas (Ed.). Holocene Human
Ecology in Northeastern North America. Plenum Publishing Co., New
York, NY. 319 pp.
Richburg, J.A., and W.A. Patterson III. 2005. Historical description of the vegetation
of the Boston Harbor Islands: 1600–2000. Northeastern Naturalist
12(Special Issue 3):13–30.
Shurtleff, N. 1891. A Topographical and Historical Description of Boston.
Rockwell and Churchill Printers, Boston, MA. 720 pp.
Stark, J.H. 1880. Illustrated History of Boston Harbor: Compiled from the Most
Authentic Sources, Giving a Complete and Reliable History of Every Island
and Headland in the Harbor, from the Earliest Date to the Present Time.
Photo-electrotype Company, Boston. 99 pp.
Stevens, A. 1966. The paleoecology of coastal sand plain grasslands on Martha’s
Vineyard, Massachusetts. Ph.D. Thesis. University of Massachusetts,
Amherst, MA. 232 pp.
Westveld, M., R.I. Ashman, H.I. Baldwin, R.P. Holdsworth, R.S. Johnson, J.H.
Lambert, H.J. Lutz, L. Swain, and M. Standish. 1956. Natural forest vegetation
zones of New England. Journal of Forestry 54:332–338.
48 Northeastern Naturalist Vol. 12, Special Issue 3
Winkler, M.G. 1985. A 12,000-year history of vegetation and climate for Cape
Cod, Massachusetts. Quaternary Research 23:301–312.
Winship, G.P. 1968. Sailors Narratives of Voyages Along the New England
Coast 1524–1624. Burt Franklin, NY. 292 pp.
Wood, W. 1993. New England’s Prospect. Alden T. Vaughan (Ed.). University
of Massachusetts Press, Amherst, MA. 144 pp.
Zaremba, R. 1982. The role of vegetation and overwash in the landward migration
of a northern barrier beach: Nauset Spit—Eastham, Massachusetts.
Ph.D. Thesis, University of Massachusetts, Amherst, MA. 490 pp.