BENJAMIN PETRICKBeloit College

Sponsor: Carol Mankiewicz

Petrick, B. 2006. 19th Annual Keck Symposium; http://keck.wooster.edu/publications

SEdIMENTATIoN ANd INdICAToRS of HoloCENE ClIMATE CHANgE IN KEuKA lAKE, NEw YoRK

109

INTRoduCTIoN

Understanding the causes and impacts of past climate change can help us predict the possible effects of global warming. Lakes in mid-latitude, temperate areas such as Keuka Lake, which is located in upstate New York, are sensitive to climate change because they respond to temperature fluctuations by changing the amount of stratification and amount of primary productivity. These lakes also preserve sediment records that document environmental changes in their drainage basin and that can be used to reconstruct their climate history. Here, I examine the Holocene sediment record of Keuka Lake to investigate climate change during the middle Holocene and late Holocene. The warmest period of time since the last ice age occurred during the middle Holocene (8-4.9 ka) when summer temperatures were 2-4° C warmer than today (COHMAP, 1988). Environmental changes during the middle Holocene could be similar to future changes that might result from global warming.

In June 2005, a core was collected from Keuka Lake to reconstruct Holocene climate change (Fig. 1). Keuka Lake is the third largest (by volume) of the eleven Finger Lakes (Mullins et al., 1996). Prior to the Pleistocene ice ages, rivers carved the landscape of the now Finger Lakes region (Mullins et al., 1996). These river valleys were carved into Devonian sandstone, siltstone, and shale and then deepened by ice and subglacial meltwaters during the last ice age (Mullins and Hinchey, 1989). Seneca,

figure 1. The approximate location of where the core was taken (black diamond). Notice the proximity to the delta, Boyd Point and wagner glen (drainage basin shown by black line). This indicates a strong fluvial sediment source that probably affects the core.

Cayuga and Canandaigua Lakes were incised below sea level, and Keuka and Skaneateles Lakes’ bedrock floors are within 25 meters of sea level. These valleys were then filled by glacial, proglacial and post-glacial sediment as the glaciers retreated northward into Canada. Keuka Lake is unique as it is the only Finger Lake with more than one branch (Fig. 1). The lake is narrow and deep, with an average water depth of ~40 m (Mullins et al., 1996)

METHodS

The core site was a good site for recovering a high-resolution, relatively continuous record of environmental and climate change as shown

110

Petrick, B. 2006. 19th Annual Keck Symposium; http://keck.wooster.edu/publications

by an earlier high-resolution seismic survey. The 290-cm-long core was collected from 50 m water depth in the northwest arm of the lake offshore of Boyd Point and Wagner Glen (Fig. 1). The core was then split, described and photographed. Approximately 1 cm thick subsamples were taken every 2 cm down the length of the core. Each subsample was freeze-dried to determine the water content (wt. %). Each subsample was analyzed by loss-on-ignition (LOI) to determine the weight percent organic matter (500°C) and carbonate (1000°C) content (Dean, 1974). The LOI method overestimates the carbonate content by ~3-4% in clay-rich sediment because of water loss from clay minerals at 1000°C (Dean, 1974). Therefore, samples with <4% carbonate content likely contain no carbonate. The grain size distribution (0.04 -2000 µm) was determined on organic and carbonate-free samples at a 4-cm interval using a laser particle size analyzer (Coulter LS 230).

RESulTS

The laminated sediments were black (N1) and olive gray (5Y 4/1) clayey silt. There was a distinct change in the laminations from the top (Fig. 2) to the bottom of the core (Fig 3). In the middle laminations were nonexistent to spotty and poorly defined in the top of the core. The bottom of the core had the best defined laminations. The contacts between the laminations were diffuse all the way through the core. When laminations were distinct, they were very thinly laminated to finely bedded. The black laminations ranged between <1 mm to 5 mm thick and the olive gray laminations varied from <1 mm to about 1-2 cm thick. The larger olive gray beds appeared massive. There were often twigs or even whole leaves preserved throughout the core. The mineral vivianite (Fe3(PO4)2∙8H2O) was present and was typically associated with terrestrial organic material.

The grain size results divide the core into two sections (Fig. 4). Below 200 cm, the core is distinguished by having very little sand and relatively consistent amounts of clay, silt, and sand. Between 0 and 200 cm, there are nine sandier intervals that range from <4 cm to 14 cm in thickness. Up core, there is a steady decrease in the percent sand content of each of these eight events and the thickness of the sandier intervals increases.

The total weight percent organic matter (average: 4.6%) and carbonate content (average: 6.2%) is constant throughout the core. In general, the organic and carbonate content display an inverse relationship (Fig. 5). During the sandier intervals, the amount of carbonate

figure 2. Core at about 60-90 cm depth. Notice the indistinct laminations.

figure 3. Core at about 250-270 cm depth. Notice the better defined laminations.

111

Petrick, B. 2006. 19th Annual Keck Symposium; http://keck.wooster.edu/publications

figure 4. graph showing percent sand, silt, and clay verses depth in the core. Notice the change in the core that comes at about 200 cm down core.

is higher than average and the amount of organic material is lower than average. One radiocarbon date showed an age of 3750 +/- 40 BP at 157 cm. Assuming a zero time frame for the top of the core, the average sedimentation rate in this core is 0.42 mm/year. Using this sedimentation rate, the base of the core yields an extrapolated age of 6.9 ka. This core provides a high-resolution paleoclimate record that begins in the late middle Holocene and continues through the late Holocene.

figure 5. The percent of organic carbon and CaCo3 determined using loss-on-ignition. Note the negative correlation between the organic carbon and carbonate.

dISCuSSIoNMethods of Sedimentation

The process by which sediment was deposited is important for interpreting changes in climate. As most samples showed both fine silts and clays in addition to sands and coarse silts I infer pelagic, hemipelagic and fluvial deposition. The sediment record for the top 200 cm of the core was punctuated by several sandier events. The sand lenses are characterized by fining upwards sequences, larger amounts of carbonate, and less organic matter. Similar sedimentation patterns

0 10 20 30 40 50 60 70 80

Percent

300

280

260

240

220

200

180

160

140

120

100

80

60

40

20

0D

epth

inco

re(c

m)

Percent Clay

Percent Silt

Percent Sand

2 4 6 8 10

% OrganicCarbon

300

280

260

240

220

200

180

160

140

120

100

80

60

40

20

0

Dep

thin

core

(cm

)

% OrganicCarbon% Carbonate

0 4 8 12 16

% Carbonate

112

Petrick, B. 2006. 19th Annual Keck Symposium; http://keck.wooster.edu/publications

are observed in Cayuga and Seneca Lakes and are interpreted as turbidity flows (Ludlum, 1967; Ludlum 1974; Rogers, 2005). However the core was not analyzed in enough detail to determine if this is the case. Given the proximity of the delta and Wagner Glen to the core site (Fig. 1), a deltaic origin for the sand events is a reasonable explanation for the method of deposition. Sediment Reworking

The distinct and rapid change over 4 cm in the lower part of the core signifies an important boundary in the core. The fact that there is more sand at the top of the core is probably a result of progradation of the nearby delta and increasing storm frequency or intensity during the late Holocene. Based on previous work in the Finger Lakes region, lake levels are thought to have lowered after the middle Holocene, which could facilitate progradation of the delta closer to the core site. However, both of these processes were likely more gradual and not as sudden as what appears in the core. Analysis of the high-resolution seismic reflection profiles at this core site reveals a feature known as the mid-Holocene reflector (Mullins and Halfman, 2001). The large sand layer at 197 cm could match with the seismic reflector. In my core this event dates to about 4.9 ka and therefore represents not only a change in the core but perhaps the upper boundary of the middle Holocene as well. This layer probably represents a time when erosion was occurring due to sediment reworking. Previous studies have shown that large currents can affect the bottoms of these lakes (Halfman and Herrick, 2001). A similar kind of truncated sediment pattern and continued poor laminations up core suggest that these conditions might have continued for a while.

Climate Change

A very basic model for climate change can be

interpreted from these data. First in the middle Holocene lake levels were probably higher but sediment deposition was fairly constant suggesting a wet environment. Given the clearer laminations, there might have been large amount of seasonality. Then about 4.9 ka the climate became more dominated by storm events and lake levels dropped. These storm events might have caused reworking of sediments on the bottom of the lake and wiped out the laminations. Finally, in the late Holocene either the amount of erosion dropped, the amount of moisture increased, or humans clearing forests began to affect the amount of sediment deposited in the lake. ACKNowlEdgMENTS

Thanks to Tara Curtin for her willingness to answer way too many questions by email. I am grateful to John Halfman for spending long days on the boat with us this summer. Thanks to the rest of Team Keuka for their help with this project and the entire Keck Finger Lakes group for making sure the summer wasn’t too boring. Thanks to Carol Mankiewicz for being patient as I struggled with figuring out what was going on and to Carl Mendelson for being a wonderful adviser. Steve Vavrus helped me understand climatology. Finally, thanks to the rest of the staff and students of the Beloit College Geology Department for being there.

REfERENCES CITEd

COHMAP, 1988, Climatic changes of the last 18,000 years: observations and model simulations: Science, v. 241, p. 1043-1052.

Dean, W., 1974, Determination of carbonate and organic matter in calcareous sediments, sedimentary rocks by loss on ignition: comparison with other methods. Journal of Sedimentary Petrology, v. 24, p. 242-248

Halfman, J.D., and Herrick, D.T., 1998, Mass movement and reworking of late glacial and postglacial

113

Petrick, B. 2006. 19th Annual Keck Symposium; http://keck.wooster.edu/publications

sediments in northern Seneca Lake, New York: Northeastern Geology and Environmental Sciences, v. 20, p. 227-241.

Ludlum, S. D., 1967, Sedimentation in Cayuga Lake, New York: Limnology and Oceanography, v. 12, p. 618-632.

Ludlum, S.D., 1974, Fayetteville Green Lake, New York. 6. The role of turbidity currents in lake sedimentation: Limnology and Oceanography, v. 19, p. 656-664.

Mullins, H.T., and Halfman, J.D., 2001, High-resolution seismic reflective evidence for middle Holocene environmental change, Owasco Lake: Journal of Sedimentary Research, v. 68, p. 569-578.

Mullins, Henry T., and Hinchey, E. J., 1989, Erosion and infill of New York Finger Lakes: implications for Laurentide ice sheet deglaciation: Geology, v. 17, p. 622-625.

Mullins, H.T., Hinchey, E.J., Wellner, R.W., Stephens, D.B., Anderson, W.T. Jr., Dwyer, T.R. and Hine, A.C., 1996, Seismic stratigraphy of the Finger Lakes: A continental record of Heinrichof Heinrich event H-1 and Laurentide ice sheet instability: Geologic Society of America, Special Paper 311, p. 1-36

Rogers, C.E. 2005, Got varves? Reconstructing Holocene climate change in Seneca Lake New York: unpublished thesis paper, Hobart and William Smith, Geneva, New York, p. 89.