Using Chemical Contaminant Profiles to

Determine Sediment Depositional History at

Little Paint Branch CreekDarya Slobodyanik

GEOL394

Advisors: Dr. Karen Prestegaard

Dr. William McDonough Dr. Richard Ash

 

 

Darya Slobodyanik 2

Using Chemical Contaminant Profiles to Determine Sediment Depositional History at Little Paint Branch Creek

Abstract

At Little Paint Branch Creek, a downstream-facing wedge of sediment aggradation has formed due to upstream erosion. It was theorized that a depth v. concentration plot of a trace metal contaminant of the upstream site would reflect a single source of contamination. The downstream site was hypothesized to have a depth v. concentration plot with a broad peak, resulting from a single source of contamination followed by remobilization of upstream sediment. Sediment samples were collected from an upstream and a downstream site, and the <300μm grain size fraction was compressed into pellets. LA-ICP-MS was performed on the pellets to determine trace metal concentrations. The results did not support the hypothesis that the upstream and downstream sites have different trace metal contamination profiles. Rather, both the upstream and the downstream site seem to follow the same trend. It is likely that the downstream site was too far upstream to represent the downstream reach of the stream. However, it was shown that LA-ICP-MS is a viable method for determining trace element concentrations of the coarse-grain sediment found in stream banks.

Introduction

Study Site History

Little Paint Branch Creek is a tributary of the Anacostia River located in the North-East Branch watershed. This watershed feeds into the Chesapeake Bay, a rich natural resource and biologically diverse habitat, and is, therefore, of particular interest. The Little Paint Branch Creek Watershed has undergone significant amount of urbanization in the past 40 years. The Beltsville USGS quadrangle (dated 1964) shows little urban development in the upper watershed Before urbanization, Little Paint Branch Creek was a small stream with a low, flat floodplain in a marshy setting. As urbanization changed the landscape, there was a correlated rise in the frequency and severity of flooding (Vicars-Groening, J. and Williams, H., 2007; Leopold, L, 1968). One of the greatest contributing factors to the change flood behavior is the increase in the percent of impervious surfaces (Figure 1). A comparison between a map of the area from 1964 and a 2006 satellite photo shows a marked increase in the number of impervious surfaces, such as roads and other paved surfaces. Note that the entire upper portion of the watershed between Calverton and Burtonsville was not urbanized in 1964. By 1979, Interstate 95 and the Beltway had been built in the watershed and extensive gravel mining (purple areas) was ongoing to support these construction activities. By 2006, most of the gravel mining areas have become

 

Darya Slobodyanik 3 suburban developments or golf courses. (Figures 2 and 3). Additional suburban developments have increased impervious surfaces in the Paint Branch watershed to 19% (personal communication, Karen Prestegaard, 2009). Floods carry with them sediment, which is deposited as stream banks as the flood gradually loses power. In opposition to this process is erosion, which is responsible for remobilizing the sediment in the watershed, and is the dominant process at work on upstream banks. Downstream banks are primarily shaped by deposition, where, though the stream appears to be an incised channel, the banks are actually increasing in height due to sediment aggradation over time. This creates a downstream-facing wedge of sediment accretion, as downstream banks collect both flood-deposited sediment and remobilized sediment from upstream banks.

Figure 1.The relationship between % impervious surface area and flood frequency (Leopold, 1968).

Darya Slobodyanik 4

Figure 2. 1964 USGS map of the area surrounding Little Paint Branch Creek . 1979 revisions are shown in

purple.

Darya Slobodyanik 5

Figure 3.A satellite photo of the modern-day Little Paint Branch Creek area (Google Earth, image date 2006). Note the suburban development between Burtonsville and Calverton.

Contaminant Behavior

The trace metal contaminants present in this watershed are anthropogenic in origin, and were introduced into the watershed through urbanization. A previous study found that anthropogenic emissions of some of the trace metals investigated in this paper are several factors greater than natural emissions of the same trace metal (Pacyna & Pacyna, 2001). These contaminants have varied over time, in a relatively well-documented fashion.

One might expect that the contaminant concentration in sediment would follow the same trend as contaminant emissions. For instance, leaded gasoline was introduced in the 1920s, and efforts to phase it out began in the 1980s (Thomas, V. M., 1995). It seems as though lead concentration in stream banks, should, therefore, peak at a point corresponding to the peak of lead emission, and then subsequently taper off as the lead source tapered off. This, however, is not necessarily the case. Compare the graph of average US lead emissions (Figure 4) to the

Darya Slobodyanik 6 concentration of lead in a core sample of a small rural lake in Kansas (Figure 5) and a core sample of the Chesapeake Bay (Figure 6).

Figure 4. Total US lead emissions by year (Juracek & Ziegler, 2006).

In the case of the lake core, the contaminant was introduced to the lake from a single source. Therefore, the oldest stratigraphic layers are relatively free of contamination, then a spike in contaminant concentration when the contaminant is introduced. Finally, concentration tapers off, and eventually disappears as the source disappears.

Figure 5. Lead concentration in a rural Kansas lake core (Juracek & Ziegler, 2006).

Figure 6. Lead concentration in a Chesapeake Bay core (Dolor et al., in press).

Darya Slobodyanik 7 The bay core, on the other hand, shows a much broader peak, indicating the presence of more than one source over time. Given the downstream-facing sediment wedge pattern seen in Little Paint Branch Creek, it is likely that the additional source of lead is remobilized sediment, eroded from banks of the upstream watershed.

Laser Ablation Inductively Coupled Plasma Mass Spectrometry (LA-ICP-MS)

ICP-MS has been used in many studies to determine contaminant concentrations in sediment cores or bank samples (e.g. Juracek and Ziegler, 2006; Devereux, 2006). In these studies, however, sediment is usually put into solution using acid digestion techniques. Dolor et al, however, have successfully used laser ablation ICP-MS to derive trace element concentrations in sediment (Dolor et al., in press).

LA-ICP-MS was chosen because it has several advantages over other forms of analysis. Samples require little preparation, and the instrument can analyze the concentrations of any amount of elements simultaneously. LA-ICP-MS is sensitive, with detection limits as low as ng/g to pg/g (Pickhardt, C. and Becker, J. S., 2001). Most of analyses, however, have been done either on individual mineral grains or on fine grained sediments (e.g. Dolor et al., in press). LA-ICP-MS has the added advantage of being safer than acid digestion. One of the challenges in this study is to determine whether high quality LA-ICP-MS analyses can be obtained from coarser-grained sediments found in stream banks.

Hypotheses

1. Laser ablation mass spectrometry (LA-ICP-MS) of compacted sediment samples can be used to determine contaminant profiles in stream bank sediment because they contain significant amounts of fine grained material. .

2. Contaminant profiles of stream bank sediment profiles should fall on a spectrum (depicted in Figure 7) between the behavior exhibited by headwater small ponds and downstream depositional sites, such as the Chesapeake Bay.

a. At headwater sites, the contaminant concentration is the result of a single source of contamination; therefore, the sediment profile should show a sharp peak in concentration, followed by a gradual decline, corresponding to a decline in contaminant source (similar to Figure 5).

b. At the furthest downstream site, the sediment profile will reflect a single source of contamination followed by erosion of upstream banks re-releasing contaminants into the watershed (similar to Figure 6).

Darya Slobodyanik 8

Methods

Site selection procedures

Pre-urbanization, downstream reaches of Little Paint Branch Creek represented a marsh-type environment, with a low, wide floodplain, this is recorded in the basal sediments and this type of floodplain still borders one side of the channel. Urbanization of the Little Paint Branch Creek watershed increased sediment transport and thus deposition in the downstream floodplain environment. In the furthest upstream reaches, the stream in incised into hillslope material, which was mobilized by agricultural activities and suburban development. Between the headwater reaches and the downstream depositional reaches, most of the Little Paint Branch Creek is bounded on both sides by tall, stratified banks. Though the stream appears to be an incised channel, the banks document significant deposition - strata have built up suspended sediment deposition during floods as well as some channel incision. This deposition is caused by floods; as the banks get taller, increasingly larger floods are necessary to deposit more sediment on top of the existing layers. The strata represent sediment deposition over time. The composition of each sediment layer should reflect the environment in which it was deposited. Any contaminants present (both from surface erosion or stream bank erosion) at the time would be deposited with the sediment. This model assumes negligible bioturbation, which can be an important factor in the vertical distribution of trace metals in a stream bank. Animals can burrow and churn the soil, and plant roots can push sediment aside as they extend downward (Gabet, et al., 2003). There was little evidence of bioturbation at the sample sites.

For this study, I selected two types of stream bank sites for sampling and analysis. The

first is an upstream site near the headwaters (Fairland), which should mean that it is not exposed

Figure 7. A graphical representation of the hypothesis, with a theoretical headwater site (Case 1) in red, and the theoretical furthest downstream site

(Case 2) shown in blue.

Darya Slobodyanik 9 to remobilized contaminants like sites farther downstream. Its contaminant concentration v. depth profile can therefore be modeled as similar to the aforementioned lake. The second site is located in the downstream depositional reach (Cherry Hill). At both of these sites, other data are available on other properties of the stream. Bank grain size distributions have been conducted by Dangol (2009) at these same sites. Stream gauges are in place to measure gauge height, conductivity, and turbidity. Long term USGS data are available at the mouth of the North East Branch watershed to determine flood frequency and intensity.

Figure 8. A photograph of Fairland, the upstream site.

Figure 9. A photograph of Cherry Hill, the downstream site.

Darya Slobodyanik 10

Little Paint Branch Cr. Watershed

Sellman

Cherry Hill

0 3,000 meters

USGS

Fairland

Figure 10. Adapted from Devereux, 2006 (right) and Council of Governments (left)

Sample collection procedures

Stream bank samples were from Little Paint Branch Creek. A stainless steel tool was used to first scrape off the first few centimeters of sediment, and then collect the sample. Bank stratigraphy is fairly well defined in the sampling site. Samples were collected every 10cm when the stratigraphy was not well-defined. When the stratigraphic layers were obvious, stratigraphic layers were sampled individually. With this method, it is possible to track changes in concentration of trace metals by stratigraphic layer or by position in the bank, and, therefore, over time.

Sieve procedures

It was necessary to restrict the sample to a small portion of the grain size distribution. This was done for two principle reasons: 1) because trace metals sorb preferentially onto smaller grains (Russell et al., 2001), and, 2) to restrict the sample to grain sizes that can be mobilized and transported by a flood event (He and Walling, 1996). Thus, restricting the sample size creates a sample that is representative of sediment that: 1) has approximately equal potential for trace metal sorption, and, 2) is small enough to be transported through the watershed by fluvial

Darya Slobodyanik 11 processes.

Samples were dried completely before sieving. Sieves were chosen with care to avoid sieving samples through materials that would contaminate the sample. Samples were first sieved to remove sediment material greater than 2mm in size with a stainless steel sieve. The edges of the sieve were crimped to the rim of the sieve rather than soldered, as solder has been shown to be a major source of contamination (Thompson, G. and Bankston, D. C., 1970). A 300μm nylon mesh was then place over the sieve, and samples were sieved through again to remove sediment greater than 300μm in size. During the sieving of small grain sizes, the sieve tends to become clogged, increasing the time the sediment spends in contact with the sieve. A nylon mesh was used for this step rather than a metal sieve to reduce the chance of metal contamination.

Typically, in analyses of this type, the sample size is conventionally restricted to grain sizes less than 63μm. In this case, 300μm was chosen on the basis that this sediment fraction best represents the fine grain fraction throughout the watershed. Not all small grain sizes are well represented at every site in this watershed (see Figure 10), and in such cases, the fine grains are likely biased toward heavy accessory minerals. To get a more accurate average composition, the fine grain fraction less than 300μm in size was chosen.

Figure11. Grain size distribution at Selman Rd., a site midway between farthest upstream and farthest

downstream sites in this study (personal communication, Achyut Dangol, 2009).

Sample preparation for LA-ICP-MS

Sediment samples from Little Paint Branch Creek show local heterogeneity, which could present inconstancies in analysis. It has been shown that by pressing the samples into pellets, the effects of local heterogeneity in the sample are further decreased (Tao, G. et al., 2002). Samples were pressed into pellets using a stainless steel die, relying on organic compounds present in the sediment to act as a natural binder. The tool used to compact the sediment is pictured in the

Darya Slobodyanik 12 appendix (Figure A-2).

Laser ablation ICP-MS procedures

The process of analyzing the fine grain particles of a sample can be simplified into three basic steps.

1. LA: Using a laser, the sample is ablated in an airtight chamber. 2. ICP: A flow of gas then introduces the sample to the inductively coupled plasma (a super hot gas, with more free electrons and ions than neutral molecules), where molecules are further dissociated into atoms, and ionized. 3. MS: In the mass spectrometer, the ions are sorted based on their ratio of mass to charge, and filtered according to their spread in kinetic energy. The electrical signal that results from the ion beam is recorded in counts per second, also known as ions per second, or ion current (Sylvester, 2008).

The next step is to transform the raw data from counts per second to concentrations. This was accomplished with a program called LAMTRACE, which uses a series of computations to calculate the isotope concentrations from raw counts per second. These concentrations are deduced as follows:

where C(AN,SAM) is the concentration of the analyte element of the sample, R(AN,SAM) is the count rate, and S is the normalized sensitivity. When using an internal standard, as we are, S is determined by the expression:

where R(AN,SRM) is the count rate of the analyte in the standard reference material (SRM), C(AN,SRM) is the concentration of the analyte in the SRM, R(IS,SAM) is the count rate of the internal standard in the sample, R(IS,SRM) is the count rate of the internal standard in the SRM,C(IS,SRM) is the concentration of the internal standard in the SRM, and C(IS,SAM) is the concentration of the internal standard in the sample (Longerich et al., 1996). The internal standard chosen in this study was SiO2, because of its abundance, low relative variability, and low interference (Bea et al., 1996). LA-ICP-MS analyses were conducted in the Plasma Mass Spectrometry Laboratory at the University of Maryland, using a New-wave UP-213-nanometer laser ablation unit coupled to a Finnigan Element2 single collector ICP-MS. The mass spectrometer was tuned prior to analyses, using 43Ca, 180Hf and 232Th to maximize signal, and 232Th16O/232Th to minimize oxide production while ablating a 40um-wide line of NIST 610 (Table A-2). The laser was fired for approximately 20 seconds with the shutter closed to obtain a background signal. Then each sample was ablated for approximately 90 seconds. Laser settings, such as spot size, frequency, and power output

Darya Slobodyanik 13 were chosen to ensure that power density remained between the range of 1.5 to 2.5 Jcm-2 without overwhelming the mass spectrometer with the major elements present in the sample. At least two minutes were allowed between each sample ablation in order to ensure that particles from the previous ablation were flushed out, thus minimizing the chance of cross-sample contamination. Table A-1 in the appendix details the LA-ICP-MS method file, and the list of elements analyzed. Standard reference materials with well-characterized concentrations of the desired analytes were chosen: MAG-1, a marine mud powder collected near Boston, MA (Smith, 1995), as well as BCR-2g and NIST 610 (Reed, 1992), both wafers of trace elements in a glass matrix. An illustrated layout of the laser chamber can be seen in the appendix (Figure A-1).

Results and Discussion

Table 1. Element concentrations ± 1σ standard deviation.

Darya Slobodyanik 14

Table 1 (cont). Contaminant concentrations.

Results are presented as concentration versus depth in Table 1. Average element

concentration was calculated from four replicate analyses of a sample. Error is generally in the range of approximately 20% of the value of the concentration.

Enrichment factors (Table 2), calculated as follows:

the ratio of concentration of the analyte to the concentration of iron in the sample, (X/Fe)sample, divided by the ratio of the concentration of the analyte over the iron in the standard reference material, (X/Fe)ref. The purpose of this enrichment factor is to scale the results such that all isotope concentrations may be presented in terms of enrichment factor = 1 (Dolor et al., in press).

Of the elements that were analyzed, V, Ga, As, Nb, Mo, Sn, W, Re, Tl, Pb and Bi were consistently enriched as compared to both SRM MAG-1 and average crustal values (Rudnick & Gao, 2003). For all elements but Co and Re, a trend appears to emerge: the enrichment factor gradually increases with depth. Figures 12 – 18 show this trend graphically. The sites appear, in fact, as though they could both be described by one well-fitting trend line. The base level at the Fairlands site was approximately 70cm below the top of the stream bank, so no data exists at greater depth. At depths greater than 80cm at the Cherry Hill site, the element enrichment factor seems to generally increase with depth.

Darya Slobodyanik 15

ENRICHMENT FACTORS site name Fairlands Cherry Hill depth (cm) 10 20 30 40 50 60 70 13 56 74 112 145

43 CaO 0.08 0.15 0.17 0.10 0.09 0.20 0.16 0.36 0.22 0.54 0.27 0.41 51 V 0.92 1.68 1.21 0.81 0.94 1.33 1.29 0.77 1.19 1.23 1.86 1.46 55 Mn 0.59 0.75 0.69 1.13 1.35 2.21 1.72 1.29 0.69 0.73 0.46 0.96 57 FeO 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 59 Co 3.57 2.25 1.45 3.16 1.20 3.14 1.81 1.94 0.94 3.18 3.03 61 Ni 0.53 1.52 0.99 0.50 0.77 1.03 0.85 1.50 0.78 0.97 1.24 1.80 63 Cu 0.44 0.72 0.48 0.26 0.29 0.56 0.73 1.23 0.88 0.82 0.55 0.78 66 Zn 0.25 0.58 0.38 0.20 0.21 0.39 0.39 0.42 0.38 0.46 0.77 0.99 69 Ga 0.69 1.54 1.35 0.94 1.10 1.55 1.79 1.14 1.61 1.67 2.19 3.07 72 Ge 0.64 0.00 0.00 0.00 0.00 0.00 3.79 1.44 73 Ge 0.59 1.65 0.00 1.95 0.00 2.27 1.15 3.79 75 As 2.74 1.80 1.65 1.59 1.20 1.46 1.21 1.11 1.22 1.27 1.06 1.10 77 Se 0.98 0.00 0.00 4.32 0.00 0.00 2.47 1.57 93 Nb 0.64 1.69 1.33 1.13 1.09 1.77 2.68 1.58 1.75 1.74 3.21 3.75 95 Mo 3.45 3.86 4.18 2.39 2.51 4.20 2.72 3.38 2.80 2.09 4.21 2.31 97 Mo 3.46 4.64 4.70 2.79 3.56 5.42 3.23 3.73 2.91 2.80 4.46 2.24

118 Sn 0.51 1.42 1.01 0.73 0.79 1.26 1.40 1.30 1.41 1.56 1.93 2.44 121 Sb 0.11 0.15 0.12 0.09 0.09 0.12 1.31 0.27 0.20 0.29 0.14 0.19 123 Sb 0.11 0.18 0.15 0.11 0.13 0.13 1.74 0.26 0.14 0.33 0.14 0.19 125 Te 184 W 0.52 1.31 0.97 0.75 0.68 1.26 2.10 1.27 1.46 1.60 2.29 2.96 185 Re 0.17 3.55 0.32 0.00 0.04 0.00 2.76 3.08 0.76 1.02 2.19 1.18 202 Hg 205 Tl 0.65 2.04 1.29 0.82 0.98 1.73 1.98 1.42 1.39 1.73 2.94 2.99 206 Pb 1.87 2.63 1.92 1.51 1.70 2.41 2.46 5.12 2.33 2.25 3.89 3.73 208 Pb 1.62 2.20 1.65 1.45 1.60 2.13 2.22 4.34 2.13 2.26 3.54 3.69 209 Bi 0.67 1.40 0.90 0.60 0.58 1.03 1.04 1.78 1.17 1.12 1.48 1.96 232 Th 0.53 0.97 1.03 1.07 0.67 1.28 1.35 0.70 0.98 0.92 1.06 1.85 238 U 0.19 0.31 0.26 0.17 0.16 0.32 0.30 0.27 0.32 0.35 0.44 0.58

Table 2. Enrichment factors.

Darya Slobodyanik 16

Figure 128. A graph of the vanadium enrichment factor at Fairland and Cherry Hill.

Figure 139. A graph of the gallium enrichment factor at Fairland and Cherry Hill.

Darya Slobodyanik 17

Figure 14. A graph of the niobium enrichment factor at Fairland and Cherry Hill.

Figure 15. A graph of the tin enrichment factor at Fairland and Cherry Hill.

Darya Slobodyanik 18

Figure 16. A graph of the tungsten enrichment factor at Fairland and Cherry Hill.

Figure17. A graph of the lead enrichment factor at Fairland and Cherry Hill.

Darya Slobodyanik 19

Figure 18. A graph of the bismuth enrichment factor at Fairland and Cherry Hill.

Discussion

Laser ablation ICP-MS seems to be an accurate method of gleaning trace element concentration from the fairly coarse-grain material that makes up stream banks. Concentration values are comparable to previous work on trace element concentration in this watershed (Dolor et al., in press). Although much of the bank sediment is sand-sized material, a significant proportion of it is fine grained material. For heterogeneous bank materials, sediment is commonly sieved into grain size fractions less than 63 microns and analyses are conducted on the fine fraction. This would have eliminated most of the sample at some of the sites, therefore sediment was sieved to 300 microns. For this grain size fraction, LA-ICP-MS appeared to give reasonable results, with a standard deviation in the neighborhood of 20%. It is not clear whether standard deviations are due to grain size heterogeneities or due to core intervals (10 cm). In addition, since sediment transport is neither a high-temperature nor a high-pressure process, a degree of heterogeneity is to be expected.

V, Ga, As, Nb, Mo, Sn, W, Re, Tl, Pb and Bi were all enriched in the stream bank due to

anthropogenic pollution. These trace metals are released into the atmosphere during the burning of fossil fuels, production of metals and other industrial goods, and waste disposal. These findings may be a cause for concern. A summary of possible health consequences associated with exposure to the enriched elements is presented in Table 3. The fine-grained sediment studied in this project remains suspended in the upper reaches of the Anacostia River segment of the Chesapeake Bay. Filter-feeding organisms ingest this sediment, because it has a similar size distribution to their food. Fish and other aquatic organisms in the Anacostia River show tumors and lesions that may result from trace metal and other contaminant – especially organisms that are sensitive to trace metals, such as Acartia tonsa (Sunda et al., 1990).

Darya Slobodyanik 20

Element  Effect on health V  ‐‐ Co  heart, lung and skin effects Ga  ‐‐ 

As digestion issues, decreased blood cell production, damage to the circulatory system, death 

Nb  ‐‐ Mo  ‐‐ Sn  digestive, liver and kidney problems W  ‐‐ Re  ‐‐ 

Tl digestion issues, hair loss, effects on the nervous system and major organs 

Pb neurobehavioral effects, anemia, brain and kidney damage, death 

Table 3. A table of enriched elements found in Little Paint Branch Creek, and possible health consequences of prolonged exposure (ATSDR).

It appears, however, that the contaminant concentration sediment profiles of the

headwater site and the downstream depositional site are not significantly different. This may be due to the recent development of roadways in the Paint Branch watershed in the time period between 1970’s and the present. Both sediment cores suggest an accumulation of slightly contaminated sediment for the bulk of the core, which might correspond with agricultural era erosion and sedimentation in the watershed. And, perhaps, it also represents sediment released from gravel mining. Both cores indicate peaks of contamination in the upper portion of the core, perhaps related to the increase in roadways in the mid-to late 1970s.

Both cores appear to respond like headwater sites. It may be that the site chosen to represent the downstream portion of the watershed is not far enough downstream. If this is the case, the core from the two sites may both respond like headwater sites because they both are, indeed, headwater sites.

Another answer may involve the balance and timeline of erosion and deposition. The assumption was made that at the headwater site, erosion would occur shortly after deposition of sediment. If erosion is no longer much greater than deposition, the periodic nature of the enrichment factor increases and decreases may be due to large, episodic floods. These floods may have deposited far more sediment than erosive processes could remobilize before they were covered with a new layer of sediment from the next flood. One way to validate this idea is a dating method, such as 210Pb dating. Ages of core layers could be obtained in this way, and, by constraining the core layers temporally, it could be possible to characterize them not only by depth, but by rates of erosion and deposition.

Bioturbation could also account for the fact that the two trace metal concentration sediment profiles are not significantly different. The model assumes negligible bioturbation, and

Darya Slobodyanik 21 while this may be true today, it may not have been true historically. Further sampling of the two sites may be useful in discovering the extent of bioturbation. If the trace element concentrations of core strata demonstrate considerable lateral variability, this could be an indication that sediment was mixed and churned.

Further continuation of this study would involve the sampling of additional sites farther downstream. By expanding the sample size, it may be possible to obtain a complete trace element contaminant profile for this part of watershed. This would also allow for observation of any changes in the sediment profile as distance downstream increases. Fortunately, as demonstrated in this study, it is possible to take advantage of LA-ICP-MS analysis for coarse-grain stream bank sediment.

Darya Slobodyanik 22

Sources

Bea, F., Montero, P., Stroh, A., & Baasner, J. (1996). Microanalysis of minerals by an Excimer UV-LA-ICP-MS system. Chemical Geology. 133 (1-4), 145.

Agency for Toxic Substances and Disease Registry (ATSDR). (2005). Toxicological Profiles. Washington, DC: U.S. Government Printing Office.

Devereux, O., Prestegaard, K. & Needelman, B. (2006). Quantifying Fine Sediment Sources in the Northeast Branch of the Anacostia River, Maryland, Using Trace Elements and Radionuclides. Thesis. Digital Repository at the University of Maryland.

Dolor, et al. (in press). Sediment Profiles of Less Commonly Determined Elements Measured by Laser Ablation ICP-MS.

Gabet, E. J., Reichman, O. J., & Seabloom, E. W. (2003). The Effects of Bioturbation on Soil Processes and Sediment Transport. ANNUAL REVIEW OF EARTH AND PLANETARY SCIENCES. 31, 249-274.

He, Q., & Walling, D. E. (1996). Interpreting Particle Size Effects in the Adsorption of 137Cs and Unsupported 210Pb by Mineral Soils and Sediments. Journal of Environmental Radioactivity. 30 (2), 117.

Juracek KE, & Ziegler AC. (2006). The legacy of leaded gasoline in bottom sediment of small rural reservoirs. Journal of Environmental Quality. 35 (6). Leopold, L. B. (1968). Hydrology for urban land planning; A guidebook on the hydrologic

effects of urban land use. Longerich, H. P., Jackson, S. E., & Guenther, D. (1996). Laser Ablation Inductively Coupled

Plasma Mass Spectrometric Transient Signal Data Acquisition and Analyte Concentration Calculation. JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY. 11 (9), 899-904.

Pacyna, J. M., & Pacyna, E. G. (2001). An assessment of global and regional emissions of trace metals to the atmosphere from anthropogenic sources worldwide.ENVIRONMENTAL REVIEWS -OTTAWA- NATIONAL RESEARCH COUNCIL-. 9, 269-298.

Pickhardt C, & Becker JS. (2001). Trace analysis of high-purity graphite by LA-ICP- MS.Fresenius' Journal of Analytical Chemistry. 370 (5), 534-40. Reed, W.P. (1992). Certificate of Analysis, Standard Reference Materials: 610, 611. Rudnick, R. & Gao, S. Composition of the Continental Crust. Treatise on geochemistry. (2003). New York, NY: Elsevier. Russell, M. A., Walling, D. E., & Hodgkinson, R. A. (2001). Suspended sediment sources in two

small lowland agricultural catchments in the UK. Journal of Hydrology. 252 (1), 1. Smith, D. B. (1995). U.S. Geological Survey Certificate of Analysis. Sunda, W. G., Tester, P.A., & Huntsman, S.A. (1990). Toxicity of trace metals to Acartia tonsa

in the Elizabeth River and southern Chesapeake Bay. Estuarine, Coastal and Shelf Science, 30, 207-221.

Sylvester, P. J. (2008). Laser ablation-ICP-MS in the earth sciences: Current practices and outstanding issues. Québec, Québec: Mineralogical Association of Canada.

Tao, G., Fujikawa, Y., Mitsui, M., & Yamada, R. (2002). Determination of mercury in sediment samples by laser ablation inductively coupled plasma mass spectrometry. JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY. 17, 560-562. Thompson, G., & Bankston, D. C. (1970). Sample Contamination from Grinding and Sieving

Darya Slobodyanik 23 Determined by Emission Spectrometry. Applied Spectroscopy. 24 (2), 210-219.

Vicars-Groening, J., & Williams, H. F. (2007). Impact of urbanization on storm response of White Rock Creek, Dallas, TX. ENVIRONMENTAL GEOLOGY -BERLIN-. 51 (7), 1263-1269.

Darya Slobodyanik 24

Appendix

Table A-1 . LA-ICP-MS analysis settings and method file

Darya Slobodyanik 25

Figure A-1. Laser chamber sample configuration.

NIST610 58% 40um 7Hz 2.29Jcm‐2 

Ca43  = 100000 cps 

Hf180  = 700000 cps 

232Th  = 2.5E6 cps 232Th16O/232Th   = 0.09 

Table A-2. Tuning results.

Darya Slobodyanik 26

Figure A-2. The steel tool used to compact stream sediment into pellets. It is composed of 316 steel.