USGS visual identity and link to main Web site

This report is preliminary and has not been reviewed for conformity with U.S. Geological Survey editorial standards or with the North American Stratigraphic Code. Any use of trade, product, or firm names is for descriptive purposes only and does not imply endorsement by the U.S. Government.

U.S. Geological Survey Open-File Report 00-306: Chapter 10

Trace Metals, Stable Isotopes, and Biogenic Silica from Cores Collected at Marion-Dufresne Site MD99-2209, Chesapeake Bay

by John F. Bratton1, Steven M. Colman1, Pattie C. Baucom1 and Robert R. Seal, II2

Abstract

Authigenic mineral phases and organic matter in Chesapeake Bay sediments record Holocene history of the paleoenvironmental conditions present during deposition and early diagenesis. Solid-phase geochemical studies of Marion-Dufresne cores collected in 1999 consist of analyses of trace metals, stable isotopes, and biogenic silica. Results from core MD99-2209, along with core RD-98 taken at the same site, indicate that fluctuations in bottom-water dissolved oxygen took place in the Chesapeake prior to significant European colonization. There is a clear perturbation of the system, however, that began in the late 18th century resulting in greater variation and overall increase in indicators of oxygen-depleted conditions.

Introduction

Advances in analytical techniques and parallel development of geological applications have made chemical analysis of sediment solids a vital part of paleoenvironmental investigations in lacustrine and marine settings. The 1999 R/V Marion-Dufresne coring operation provided the opportunity to expand the timescale of previous geochemical work in Chesapeake Bay, to allow robust correlation with other Holocene climate records. For this reason, solid-phase geochemical analyses are being performed on the longest cores obtained (MD99-2207 and -2209). The present report describes the methods being used and demonstrates their utility with results from a shorter piston core (RD-98) and Marion-Dufresne core MD99-2209, both taken at the same site. This northernmost Chesapeake coring location is expected to be particularly sensitive to fluctuations in bottom-water dissolved oxygen because it is located in the first part of the axial basin to show depleted oxygen in the spring, and the last part of the basin to be ventilated in the fall. This is based on water-column measurements collected since 1984 (dissolved oxygen data are available from the Chesapeake Bay Program at http://noaa.chesapeakebay.net/data/do.htm). The position in the basin, combined with the relatively high sedimentation rate at the site, produce strong potential for preservation of high-resolution records of paleo-anoxia in these cores.

Sediment Geochemical Methods

Proxy records of paleo-oxygenation, denitrification intensity, plankton productivity, sources of organic carbon, and salinity are critical for reconstructing evolution of the Chesapeake Bay ecosystem over time, and evaluating the impact of human modifications of the watershed. Biological proxies for some of these parameters are available (e.g., Cooper and Brush, 1991; Cooper, 1995; Cronin and others, 2000). Geochemical proxies being used in the current Chesapeake study to detect changes over time in these environmental variables include trace metals (oxygenation), stable isotopes (denitrification, carbon cycling, and salinity), and biogenic silica (diatom productivity). Methods incorporating these analytes have also been applied to cores taken in Chesapeake Bay in 1996 (Kerhin and others, 1998) and 1998 (core RD-98). Sampling procedures for solid-phase analyses of Marion-Dufresne cores taken in 1999 are described by Baucom and others (Chapter 5, this volume).

Trace metals

Certain metals that are naturally present in trace quantities in seawater become concentrated in low-oxygen sediments due to several factors: the development of reducing conditions, enhanced preservation of organic matter (adsorption substrate), lower pH, and favorable conditions for growth of anaerobic microbes. Anaerobic microbes mediate the reactions that convert many of these metals from dissolved species to solid phases. Measuring the concentrations of these redox-sensitive metals in authigenic phases of sediment cores makes it possible to reconstruct the redox history and, in turn, the dissolved oxygen history, of Chesapeake Bay bottom waters and shallow porewaters. Elements that have proven particularly useful in Chesapeake Bay or other low-oxygen settings are rhenium (Bratton and others, 1999; Adelson, 1997; Colodner and others, 1993; see also Morford and Emerson, 1999); vanadium (Bratton and others, 1998), molybdenum (Erickson and Helz, 2000; Adelson, 1997); and uranium (Shaw and others, 1994). Trace metal analyses for this study have been performed on acid-leach and acid-digest sample solutions by isotope dilution using an inductively coupled plasma mass spectrometer at Woods Hole Oceanographic Institution (WHOI) in Woods Hole, Massachusetts. In this paper, we focus on results for rhenium (Re).

Rhenium is among the most highly enriched dissolved species in seawater relative to its crustal abundance. Consequently, it shows authigenic enrichment more clearly than most other trace elements (Crusius and others, 1996). It behaves conservatively in the water column and is not generally affected by biological processes. Dissolved Re+7 (as ReO4-) is reduced to Re+6 (ReO3) and other more-reduced oxides and sulfides and deposited in sub-oxic sediments by processes demonstrated to act at the sediment-water interface and within the sub-oxic zone of shallow pore fluids. Colodner (1991) showed that Chesapeake Bay sediments with thicker sub-oxic zones (no O2 or H2S present) are more effective at sequestration of Re than more anoxic sediments with thin or absent sub-oxic zones. This implies that periods of transition from better oxygenated to more oxygen-depleted bottom waters and vice versa would show the greatest Re enrichments in Chesapeake sediments. Also, Re should be more concentrated in sediments deposited around the edges of the oxygen-depleted axial basins in the Chesapeake than in sediments in the deep centers of the basins. Leachable Re rather than total Re was measured in these sediments because it is more representative of the authigenic Re fraction in the sediment. Concentrations reported are parts per billion (ppb) of dry sediment weight.

Stable isotopes

Relative ratios of stable isotopes of carbon and nitrogen preserved in sediment make it possible to reconstruct changes in cycling of these elements in the estuary through time. Carbon ratios are controlled by changes in terrestrial, marine, and atmospheric carbon cycling (e.g., Hunt, 1966; see also Goni and others, 1997) and by microbial breakdown of organic matter in sediment. Nitrogen isotopes record the balance in the system between nitrogen cycling processes including assimilation, nitrification, denitrification, and remineralization of organic matter (Bratton and others, 1998; Montoya and others, 1990; see also Pride and others, 1999). All measurements have been made by gas-source mass spectrometry at USGS facilities in Reston, Virginia. Results are reported in parts per million (ppm) notation relative to the PeeDee Belemnite standard (PDB) for delta13Corg and relative to atmospheric nitrogen for delta15N.

Measurements of delta15N and delta13Corg in surface sediments in Chesapeake Bay by Hunt (1966) and Spiker (unpublished data) show significant differences along the length of the bay from north to south. Values of delta13Corg increase from around -24 ppm in the northern bay to -20 ppm near the mouth. This is interpreted as the product of mixing of terrestrial and marine organic carbon sources, with the northern bay reflecting mostly river input and the mouth of the bay showing deposition of oceanic organic matter brought in by tidal circulation. Changes in the salinity at any given point in the bay produced by variations in streamflow would be expected to be reflected in the delta13Corg signal (e.g., wetter climate = higher streamflow = lower average delta13Corg).

A north-south transect of delta15N of surface sediments (Spiker, unpublished data) shows a peak of 8.5 ppm 140 km south of the Susquehanna River (about 30 km south of the MD99-2209 site). Values are 1.5 to 2.0 ppm lower to the north and south. The mid-bay delta15N peak is interpreted as resulting from greater denitrification in this portion of the bay caused by depletion of oxygen by decomposition of algal biomass. Increased anthropogenic or climatically-induced eutrophication of the bay would be expected to result in higher average delta15N values due to more intense and areally extensive denitrification in the water column. Wetter or drier periods would also shift the delta15N maximum south or north, respectively.

Biogenic Silica

Diatoms are the dominant members of the Chesapeake Bay phytoplankton community. When they die, their silica skeletons (or frustules, see Cronin and others, 1999) are deposited in the sediment. Diatom assemblages can be studied microscopically to reconstruct pre-existing conditions from known environmental tolerances of extant forms (Cooper and Brush, 1991). Alternatively, total biogenic silica (BSi) can be measured chemically using selective basic digestion methods described by Carter and Colman (1994) and Mortlock and Froelich (1989). Measurements of BSi in samples from Chesapeake sediment cores are thought to reflect changes in overall diatom productivity in response to both natural and anthropogenic changes (such as those related to increased nutrient loads) in the bay. Diatoms are particularly useful for this type of study because their siliceous frustules are well-preserved in the sediments and they represent about 80-90% of the spring algal biomass (Cooper, 1995).

Several assumptions must be made when interpreting BSi data as a measure of diatom productivity: 1) the majority of the amorphous or opaline silica in the sediments is in the form of diatom frustules, 2) the proportion of diatom silica recycled in the water column remains approximately constant, and 3) post-depositional dissolution of silica is minor. In order to interpret biogenic silica as an index of trophic status, it must also be assumed that diatoms constitute most or a representative proportion of the total algal biomass. Dried Chesapeake Bay sediments typically contain 5 to 10 percent biogenic silica by weight. The balance of the sediment consists of organic matter (0.5 to 3 percent), calcareous shells (about 5 percent except in shell beds), and mineral grains (silt, clay, and sand; typically 80 to 95 percent). Under conditions of constant detrital sediment input, increases in percent biogenic silica indicate increased diatom blooms usually associated with greater nutrient supplies (especially nitrate), more spring runoff, and lower salinity (Cornwell and others, 1996; Baucom and others, 1999).

Results

Figure 10.1 shows data from core RD-98 (site of MD99-2209) for delta15N ( ppm air), delta13Corg ( ppm PDB), leachable Re (ppb), and BSi (%) using the age model of Colman and others (Chapter 6, this volume). Curves shown represent three-point moving averages. Analytical values discussed below refer to the actual data points (also shown) rather than the curves.

The delta15N plot in Figure 10.1 shows that values fluctuate around 5.8 ppm between 1400 and 1750 A.D. About 1775, delta15N begins to increase toward maxima of 10.1 ppm in 1952 and 9.8 ppm in 1972, with a slight decrease near the top of the core. The signal appears to have some cyclicity, with an amplitude of about +0.8 ppm and a period of about 60 to 80 years. The frequency of variability appears to decrease in the last 200 years of the record to about 20-year cycles, but this may be an artifact of the increase in sedimentation rate in this same period and the higher resolution of the data.

The delta13Corg values are relatively constant from the base of the core to about 1775. The average value over this interval is -23.1 ppm +0.5. After 1775, the data are more variable (+0.5-1.0 ppm) but show no increasing or decreasing trend. The period of variability is fairly irregular over the full length of the core. As discussed above, delta13Corg would be expected to vary with salinity due to mixing of light carbon from freshwater sources and heavy carbon from saltwater sources.

Data for leachable Re from the RD-98 core show positive excursions from two baseline levels rather than cyclical variation like N and C isotopes. The baseline in the lower part of the core (before about 1775 A.D.) is around 1.1 ppb with excursions of up to 1.0 ppb centered at 1540, 1670, and 1720. After 1775 the baseline drops to 0.9 ppb. This is followed by several small excursions (<0.5 ppb), a large excursion centered at 1936 of up to 1.9 ppb (highest data point off the scale of figure), and a smaller excursion (0.5 ppb) at 1970. The excursions are interpreted as being the result of transitions to lower oxygen conditions in Chesapeake bottom waters. Discussion of long-term trends and magnitude from these concentration data is difficult because they are affected by increased sedimentation rates over the last few centuries. This effect is suggested by the drop in Re baseline around 1775, a time of rapid increase in sedimentation rates in the bay. However, timing of excursions and the inferred transitions in oxygen concentrations derived from the concentration data are valid.

Figure 10.2 (A) shows a comparison of leachable Re concentrations between high-resolution samples from the upper 50 cm of cores RD-98 and MD99-2209. These data show strong correlation between the two cores. The RD-98 samples were collected at regular 3-cm intervals. Samples from MD99-2209, however, were collected from the centers of 10 pairs of alternating black and gray color bands in the sediment. The data show that Re is not strongly affected by the process controlling color banding (i.e., formation of iron monosulfides such as hydrotroilite).

Figure 10.2 (B) shows combined Re data for all of core RD-98 and the portion of core MD99-2209 below 300 cm. Like the duplicated interval shown in the upper plot of Figure 10.2, the region of overlap between data sets from the two cores (3.0 to 4.5m depth, 300 to 700 yr BP) shows their data to be highly correlated, with depth offsets of no more than 25 cm. Radiocarbon data indicate that a hiatus or unconformity is present in the MD99-2209 core at a depth of approximately 820 cm (see Fig. 6.1). The missing gap represents about 3500 years between 2300 and 5800 yr BP (see Fig. 10.2B). Sample resolution between the base of core RD-98 and the hiatus in MD99-2209 (700 to 2300 yr BP) is fairly low because of relatively low sedimentation rates. Re concentrations are similar to those in the bottom part of the RD-98 core (1.1 ppb +0.5), with lows around 750 and 1500 yr BP and a peak at 1200 yr BP. Immediately below the hiatus is an anomalously high Re peak followed by an extreme low. These are interpreted as having been produced by remobilization of Re orginally present in the sediments due to oxygen penetration during the period of probable non-deposition. The part of the core deposited between 6000 and 6800 yr BP shows a plateau around 1.1 ppb with brief (<100 yr) negative excursions of about 0.3 ppb. Between 6800 and 7600 BP there is a gradual decline in Re concentrations to 0.7 ppb except for an anomalous high value around 7500 yr BP. This is consistent with a transition to more restricted estuarine conditions that existed in the bay when sea level was lower.

Results for BSi concentration show a gradual decrease from near the base of the core (10.6% BSi) to about 4.4% in 1882; after this values start to increase again. Because BSi, like Re, is a concentration parameter, it must be corrected for total detrital sediment flux (Baucom and others, 1999) before diatom productivity changes over time can be interpreted. Work is in progress to determine the history of BSi flux; initial indications are that BSi flux has increased at nearly the same pace as sedimentation rate (see Chapter 6).

Discussion

Figure 10.3 shows plots of data from core RD-98 data correlated with tree-ring data (PHDI; negative = dry, positive = wet) and paleo-salinity data from Stahle and others (1998) and Cronin and others (2000), respectively. Ages are based primarily on radiocarbon data for A, B, C, and E and on counting of tree rings for D. 100-year intervals (A.D.) are labeled on A and are marked by ticks on the time axis of each plot. Correlation lines (dotted) are labeled with wet period designations used in Cronin and others (2000), Figure 3, for W1-W14; W15-W18 are extrapolated based on PHDI data. Lines for the weaker excursions, W5, W7-9, and W11, are not shown. Slight offsets of peaks in the data sets are due to uncertainties in sediment age models, irregularities in sedimentation rates, different sampling intervals, and different smoothing factors.

Core RD-98 data for delta13Corg, delta15N, and Re (A, B, and C, respectively) show systematic changes over the length of the core. Isotope measurements of delta13Corg and delta15N are generally inversely correlated, especially prior to about 1890 (W3b, W4, W6, W12, W13, and W16). This is consistent with a scenario where higher streamflow during wet periods brings more terrestrial carbon (isotopically more negative) into the bay, while also enriching the bay in nutrients. This in turn drives increased phytoplankton productivity and more denitrification in the water column, yielding sediments enriched in 15N (isotopically more positive). The two isotopes are more closely correlated after 1890 at the site indicating a change in the nitrogen source to one more closely tied to dry periods. One possibility is an increased contribution to the nitrogen budget from ammonia and nitrate in riverbed and submarine groundwater discharge.

Leachable Re data variations are loosely correlated with delta13Corg and inversely correlated with delta15N. This pattern is best seen at the wet periods W3a/b, W12, W13, and W17, but is not consistent across the transition period between 1850 and 1750. Maximum Re enrichments correlate with transitions from stable climate periods (wet, dry, or average) to more extreme periods (wet or dry) based on PHDI (e.g., peaks at approximately 1970, 1935, 1735, 1540, and 1420). Such transitions would be expected to create greater thicknesses of sub-oxic sediment by either partially oxygenating previously anoxic sediments (wet to dry transition), or removing oxygen from previously oxygenated sediments (dry to wet transition). The increase in sub-oxic sediment would remove Re from bottom waters. Changes in sedimentation rate could have similar effects. The correlation with delta13Corg may indicate a connection between Re and salinity. Additional analysis of Re mass flux data, and comparison with other trace metal data and BSi data will provide more insight on interpretation of Re cycling in this system.

Prior to significant human impact on the bay, the general relationships between the three parameters discussed was as follows. Heavier delta15N from increased denitrification would correlate with lighter delta13Corg derived from increased riverine discharge, and variable Re based on conditions preceding the change and the duration of the shift. This conclusion is supported by correlation with the PHDI and paleo-salinity data in Figure 10.3 D and E. Agricultural development of the watershed and construction of flood-control dams in the 1930s changed these relationships, producing more variable delta13Corg, greater oxygen depletion and denitrification indicated by heavy delta15N, and lower average concentrations of leachable Re with extreme excursions.

Conclusions

These data from cores RD-98 and MD99-2209 suggest that fluctuations in bottom-water dissolved oxygen took place in the Chesapeake prior to European colonization. However, there is a clear perturbation to the system that began in the late 18th century resulting in greater variation and overall increase in indicators of oxygen-depleted conditions and eutrophication. Concentration data for Re and BSi show dilution by increased sedimentation beginning during this same time period. Interpretations are supported by correlation with other paleoclimate data sets based on benthic foraminifera and tree rings.

Acknowledgements

We thank the captain, crew, and members of the scientific party of the R/V Marion-Dufresne, particularly Elisabeth Michel and Yvon Balut, for assistance with core collection and processing. Gregory Wandless and Adam Johnson performed stable isotope analyses at USGS-Reston. Trace metal measurements were made at WHOI with assistance and advice from Dave Schneider, Larry Ball, Jurek Blusztajn, and Greg Ravizza. We also thank David Stahle and Thomas Cronin for supplying PHDI and foraminiferal data.

References

Adelson, J.M., 1997, The evaluation of geochemical indicators of anoxia in the Chesapeake Bay: College Park, University of Maryland, Ph.D. dissertation, 214 p.

Baucom, P., Colman, S., and Bratton, J., 1999, Biogenic silica trends in Chesapeake Bay: Eos, American Geophysical Union Transactions, v. 80, no. 46, p. F46.

Bratton, J.F., Colman, S.M., and Ravizza, G., 1999, Sedimentary rhenium enrichment and hypoxia in Chesapeake Bay [abs.]: Geological Society of America Abstracts with Programs, v. 31, no. 7, p. A460.

Bratton, J.F., Colman, S.M., Seal, R.R., II, and Murray, R.W., 1998, Anoxia history in Chesapeake Bay based on nitrogen isotopes and redox-sensitive metals: Eos, American Geophysical Union Transactions, v. 79, no. 4, p. F496.

Carter, S.J., and Colman, S.M., 1994, Biogenic silica in Lake Baikal sediments - results from 1990-1992 American cores: Journal of Great Lakes Research, v. 20, p. 751-760.

Colodner, D., 1991, The marine geochemistry of rhenium, iridium, and platinum: Cambridge, Massachusettes Institute of Technology/Woods Hole Oceanographic Institute Joint Program in Oceanography, Ph.D. dissertation.

Colodner, D., Sachs, J., Ravizza, G., Turekian, K.K., Edmond, J., and Boyle, E., 1993, The geochemical cycle of rhenium - a reconnaissance: Earth and Planetary Science Letters, v. 117, no. 1-2, p. 205-221.

Cooper, S.R., 1995, Chesapeake Bay watershed historical land use - impact on water quality and diatom communities: Ecological Applications, v. 5, p. 703-723.

Cooper, S.R., and Brush, G.S., 1991, Long-term history of Chesapeake Bay anoxia: Science, v. 254, no. 5034, p. 992-996.

Cornwell, J.C., Conley, D.J., Owens, M., Stevenson, J.C., 1996, A sediment chronology of the eutrophication of Chesapeake Bay: Estuaries, v. 19, no. 2B, p. 488-499.

Cronin, T.M., Wagner, R.S., and Slattery, M., eds., 1999, Microfossils from Chesapeake Bay sediments - illustrations and species database: U.S. Geological Survey Open-File Report 99-45. Also available on WWW: https://pubs.usgs.gov/pdf/of/of99-45/.

Cronin, T., Willard, D., Karlsen, A., Ishman, S., Verardo, S., McGeehin, J., Kerhin, R., Holmes, C., Colman, S., and Zimmerman, A., 2000, Climatic variability in the eastern United States over the past millennium from Chesapeake Bay sediments: Geology, v. 28, no. 1, p. 3-6.

Crusius, J., Calvert, S., Pedersen, T., and Sage, D., 1996, Rhenium and molybdenum enrichments in sediments as indicators of oxic, suboxic, and sulfidic conditions of deposition: Earth and Planetary Science Letters, v. 145, p. 65-78.

Erickson, B.E., and Helz, G.R., 2000, Molybdenum (VI) speciation in sulfidic waters - stability and lability of thiomolybdates: Geochimica et Cosmochimica Acta, v. 64, no. 7, p. 1149-1158.

Goni, M.A., Ruttenberg, K.C., and Eglinton, T.I., 1997, Sources and contributions of terrigenous organic carbon to surface sediments in the Gulf of Mexico: Nature, v. 389, p. 275-278.

Hunt, J.M., 1966, The significance of carbon isotope variations in marine sediment - advances in organic geochemistry, in International Congress on Organic Geochemistry, 2nd , p. 27-36.

Kerhin, R.T., Williams, C., and Cronin, T.M., 1998, Lithologic descriptions of piston cores from Chesapeake Bay, Maryland: U.S. Geological Survey Open-File Report OF98-0787, 141 p.

Montoya, J.P., Horrigan, S.G., and McCarthy, J.J., 1990, Natural abundance of 15N in particulate nitrogen and zooplankton in the Chesapeake Bay: Marine Ecology Progress Series, v. 65, p. 35-61.

Morford, J.L., and Emerson, S., 1999, The geochemistry of redox sensitive trace metals in sediments: Geochimica et Cosmochimica Acta, v. 63, no. 11-12, p. 1735-1750.

Mortlock, R.A., and Froelich, P.N., 1989, A simple method for the rapid determination of biogenic opal in pelagic marine sediments, Deep-Sea Research, Part A: Oceanographic Research Papers, v. 36, no. 9, p. 1415-1426.

Pride, C., Thunell, R., Sigman, D., Keigwin, L., Altabet, M.A., 1999, Nitrogen isotopic variations in the Gulf of California since the last deglaciation - response to global climate change: Paleoceanography, v. 14, no. 3, p. 397-409.

Shaw, T.J., Sholkovitz, E.R., Klinkhammer, G., 1994, Redox dynamics in the Chesapeake Bay - the effect on sediment/water uranium exchange: Geochimica et Cosmochimica Acta, v. 58, no. 14, p. 2985-2995.

Stahle, D.W., Cleaveland, M.K., Blanton, D.B., Therrell, M.D., Gay, D.A., 1998, The lost colony and Jamestown droughts: Science, v. 280, no. 5363, p. 564-567.

U.S. Department of Interior, U.S. Geological Survey
URL of this page: https://pubs.usgs.gov/openfile/of00-306/chapter10/
Maintained by: Eastern Publications Group Web Team
Last modified: 03.30.01 (krw)