700 yr Sedimentary Record of Intense Hurricane Landfalls
in Southern New England

Donnelly J. P., S. S. Bryant, J. Butler, J. Dowling, L. Fan, N. Hausmann, P. Newby,
B. Shuman, J. Stern, K. Westover and T. Webb III

GSA Bulletin Volume 113 no. 6, pages 714 –727 (June 2001)

Abstract      Figures      Full Text (pdf file, 753K)

Abstract

Five intense (category 3 or greater) hurricanes occurring in 1635, 1638, 1815, 1869, and 1938 have made landfall on the New England coast since European settlement. Historical records indicate that four of these hurricanes (1635, 1638, 1815, and 1938) and hurricane Carol, a strong category 2 storm in 1954, produced significant storm surges (.3 m) in southern Rhode Island. Storm surges of this magnitude can overtop barrier islands, removing sediments from the beach and nearshore environment and depositing overwash fans across back-barrier marshes, lakes, and lagoons. In a regime of rising sea level, accumulation of marsh, lake, or lagoon sediments on top of overwash, deposits will preserve a record of overwash deposition.

We examined the record of overwash deposition at Succotash salt marsh in East Matunuck, Rhode Island, and tested the correlation with historical records of intense storms. Aerial photographs taken after hurricanes in 1954 and 1938 show overwash fans deposited at the site. Analysis of 14 sediment cores from the back-barrier marsh confirmed the presence of these fans and revealed that four additional large-scale overwash fans were deposited within the marsh sediments.

The four overwash fans deposited since the early seventeenth century at Succotash Marsh matches the historical record of significant hurricane-induced storm surge. These fans were most likely deposited by hurricanes in 1954, 1938, 1815, and either 1638 or 1635. Radiocarbon dating of two prehistoric overwash fans indicated that these were deposited between A.D. 1295–1407 and 1404–1446 and probably represent intense hurricane strikes. In the past 700 yr, at least seven intense hurricanes struck the southern Rhode Island coast and produced a storm surge that overtopped the barrier at Succotash Marsh.

Figures

Click below to enlarge this figure.

Figure 1
72dpi (21K), 300dpi (163K)

Figure 1. Map view and cross section of conceptual model of overwash deposition and the landward translocation of the barrier-marsh system in a regime of rising sea level. Overtopping of the barrier beach by storm surge results in overwash fan deposition across back-barrier marshes. Overwash fans are preserved as sea level increases and they are covered with marsh deposits. Transects of Vibracores are used to identify and map overwash deposits. Vertical exaggeration is ~100x to 500x.

Click below to enlarge this figure.

Figure 2
72dpi (51K), 300dpi (736K)

Figure 2. General study site location map of southern New England (A) with storm tracks of historic hurricanes mentioned within the text. Dashed tracks are poorly defined by historic data. Map of Succotash Marsh in East Matunuck, Rhode Island, shows core locations, noted by open circles. Dashed lines indicate the extent of overwash deposition within Succotash Marsh associated with hurricanes in 1954 and 1938 determined from historical aerial photography. The dotted line indicates the 1903 configuration of the inlet that was active during the nineteenth century A.D. (U.S. House of Representatives, 1903).

Click below to enlarge this figure.

Figure 3
72dpi (18K), 300dpi (125K)

Figure 3. Storm surge heights in southern Rhode Island relative to contemporary mean sea level (MSL). The modern barrier height of ~3 m above MSL at Succotash Marsh is noted by a dashed line. The height of storm surges dating back to 1936 has been measured by nearby tide gauges at Newport, Rhode Island, and New London, Connecticut (NOAA/NOS/CO-OPS, 2000; USACE, 1962). Historical written accounts of storm surge heights were used for storms that predated the tide gauges (Ludlum, 1963).

Click below to enlarge this figure.

Figure 4
72dpi (24K), 300dpi (282K)

Figure 4. Photographs of the upper 55 cm of selected cores. Dark units are salt-marsh peat. Light units are very fine to medium sand. Note the abrupt nature of the upper contacts between the peat and the overlying sand. The contact between the peat and the underlying sand units tends to be more gradational. A complex unit of sand interbedded with mud and thin peat intervals interpreted as channel levee or dredged deposits is evident between 19 and 30 cm in core 1.

Click below to enlarge this figure.

Figure 5
72dpi (36K), 300dpi (401K)

Figure 5. Logs of cores from transect A-B. Overwash fans have been correlated between cores (shaded in gray) and labeled fans I-VI. Solid arrows mark the location of radiocarbon-dated samples, the dates of which are presented with 2 s calibrated age ranges (sample numbers noted in parentheses correspond to those presented in Table 1). Large arrows labeled Pb denote the stratigraphic interval where the concentration of Pb increases over background levels associated with the industrial revolution (see Fig. 10). Arrows labeled rx denote the stratigraphic interval where Rumex pollen increases above 3%, indicating widespread European-style clearance of the landscape (ca. A.D. 1700) (see Fig. 11).

Click below to enlarge this figure.

Figure 6
72dpi (37K), 300dpi (482K)

Figure 6. Logs of cores from transect A-D. Overwash fans have been correlated between cores (shaded in gray) and labeled fans I-VI. Solid arrows mark the location of radiocarbon-dated samples, the dates of which are presented with 2 s calibrated age ranges (sample numbers noted in parentheses correspond to those presented in Table 1). Large arrows labeled Pb denote the stratigraphic interval where the concentration of Pb increases over background levels associated with the industrial revolution (see Fig. 10). Arrows labeled rx denote the stratigraphic interval where Rumex pollen increases above 3%, indicating widespread European-style clearance of the landscape (ca. A.D. 1700) (see Fig. 11).

Click below to enlarge this figure.

Figure 7
72dpi (25K), 300dpi (323K)

Figure 7. Logs of cores from transect E-E'. Overwash fans have been correlated between cores (shaded in gray) and labeled fans I-VI. Solid arrows mark the location of radiocarbon-dated samples, the dates of which are presented with 2 s calibrated age ranges (sample numbers noted in parentheses correspond to those presented in Table 1). Large arrows labeled Pb denote the stratigraphic interval where the concentration of Pb increases over background levels associated with the industrial revolution (see Fig. 10). Arrows labeled rx denote the stratigraphic interval where Rumex pollen increases above 3%, indicating widespread European-style clearance of the landscape (ca. A.D. 1700) (see Fig. 11).

Click below to enlarge this figure.

Figure 8
72dpi (21K), 300dpi (289K)

Figure 8. Logs of cores from transect A-C. Overwash fans have been correlated between cores (shaded in gray) and labeled fans I-VI. Solid arrows mark the location of radiocarbon-dated samples, the dates of which are presented with 2 s calibrated age ranges (sample numbers noted in parentheses correspond to those presented in Table 1). Large arrows labeled Pb denote the stratigraphic interval where the concentration of Pb increases over background levels associated with the industrial revolution (see Fig. 10).

Click below to enlarge this figure.

Figure 9
72dpi (9K), 300dpi (78K)

Figure 9. Activity profile of 137Cs and core log from the upper 30 cm of core 7 (see photo in Fig. 4). Note the initial rise in 137Cs attributable to A.D. 1954 occurs in the sample between 12 and 13 cm. The peak in 137Cs activity occurs in the sample between 9 and 10 cm and represents A.D. 1963.

Click below to enlarge this figure.

Figure 10
72dpi (35K), 300dpi (259K)

Figure 10. Pb concentrations from peat intervals within cores 3, 4, 5, 6, 7, 8, 9, 11, and 12. We infer that the increase in Pb concentration between 42 and 20 cm depth indicates that this sediment was most likely deposited in the mid- to late nineteenth century A.D.

Click below to enlarge this figure.

Figure 11
72dpi (8K), 300dpi (55K)

Figure 11. Abundance of Rumex pollen relative to total tree pollen in samples selected from cores 4, 2, 9, 11, and 12. Dashes indicate that no Rumex was observed in samples from that depth. We infer that the increase in Rumex abundance between 60 and 40 cm depth represents the onset of widespread European-style clearance and agriculture (ca. A.D. 1700).

Click below to enlarge this figure.

Figure 12
72dpi (16K), 300dpi (183K)

Figure 12. Plot of calibrated (2 s) radiocarbon age ranges for 11 radiocarbon dates for peat intervals directly below overwash fans at Succotash Marsh. The data are arranged in groups according to the overwash fan that overlies each sample: e.g., diamonds are dated peat sample directly underlying fan VI, and triangles are dated peat samples directly underlying fan V. These symbols represent the intercepts of the radiocarbon date with the calibration curve. Time intervals shaded with diagonal lines denote the mutual overlap of dates constraining the age of that peat interval. Time intervals shaded with solid gray denote the most likely age range based on the calibrated radiocarbon dates and other stratigraphic data. The intervals denoted by agr and Pb represent the time of the onset of widespread European-style clearance and Pb pollution, respectively. Asterisk indicates calibrated radiocarbon date from a peat deposit at the base of a large-scale overwash fan at Wells, Maine (Kelley et al., 1995).