A Geochemical Atlas of South Carolina--An Example Using Data from the National
U.S. Geological Survey Open-File Report 2004-1368
Published 2005 - Online Version 1.0
Description of element maps
Figures 1-5 are in PDF format.
Figure 1. National Geochemical Survey streams-sediment sample locations [260-KB PDF]
Figure 2. National Geochemical Survey soil sample locations [185-KB PDF]
Figure 3. Periodic table of selected elements [170-KB PDF]
Figure 4. South Carolina counties [125-KB PDF]
Figure 5. Mineral Resources Data System (MRDS) sites in South Carolina [375-KB PDF]
The atlas maps and graphs are in PDF format, and the average size is about 750 KB. Each element is listed below as a separate PDF file. Alternately, the stream-sediment sample maps have been compressed into one volume for downloading, as have the soil sample maps.
Stream-sediment sample maps in a zipped volume [28.7 MB]
Soil sample maps in a zipped volume [28.4 MB]
|Download Adobe Reader.|
|Figure||Stream-sediment sample maps||Soil sample maps|
|Atlas maps and graphs showing concentrations of major chemical elements in stream-sediment and soil samples|
|6.||Aluminum (Al)||Aluminum (Al)|
|7.||Calcium (Ca)||Calcium (Ca)|
|8.||Iron (Fe)||Iron (Fe)|
|9.||Magnesium (Mg)||Magnesium (Mg)|
|10.||Phosphorus (P)1||Phosphorus (P)1|
|11.||Potassium (K)||Potassium (K)|
|12.||Sodium (Na)||Sodium (Na)|
|13.||Titanium (Ti)1||Titanium (Ti)1|
|Atlas maps and graphs showing concentrations of trace chemical elements in stream-sediment and soil samples|
|14.||Arsenic (As)||Arsenic (As)|
|15.||Barium (Ba)||Barium (Ba)|
|16.||Beryllium (Be)||Beryllium (Be)|
|17.||Bismuth (Bi)||Bismuth (Bi)|
|18.||Cerium (Ce)1||Cerium (Ce)1|
|19.||Chromium (Cr)1||Chromium (Cr)1|
|20.||Cobalt (Co)1||Cobalt (Co)1|
|21.||Copper (Cu)||Copper (Cu)|
|22.||Europium (Eu)1||Europium (Eu)1|
|23.||Gallium (Ga)||Gallium (Ga)|
|24.||Gold (Au)||Gold (Au)|
|25.||Holmium (Ho)1||Holmium (Ho)1|
|26.||Lanthanum (La)1||Lanthanum (La)1|
|27.||Lead (Pb)||Lead (Pb)|
|28.||Lithium (Li)1||Lithium (Li)1|
|29.||Manganese (Mn)1||Manganese (Mn)1|
|30.||Mercury (Hg)||Mercury (Hg)|
|31.||Molybdenum (Mo)||Molybdenum (Mo)|
|32.||Neodymium (Nd)1||Neodymium (Nd)1|
|33.||Nickel (Ni)1||Nickel (Ni)1|
|34.||Niobium (Nb)1||Niobium (Nb)1|
|35.||Scandium (Sc)||Scandium (Sc)|
|36.||Selenium (Se)||Selenium (Se)|
|37.||Silver (Ag)||Silver (Ag)|
|38.||Strontium (Sr)||Strontium (Sr)|
|39.||Thorium (Th)||Thorium (Th)|
|40.||Tin (Sn)1||Tin (Sn)1|
|41.||Vanadium (V)1||Vanadium (V)1|
|42.||Ytterbium (Yb)1||Ytterbium (Yb)1|
|43.||Yttrium (Y)1||Yttrium (Y)1|
|44.||Zinc (Zn)||Zinc (Zn)|
|1 Summary reports on mineral commodities that may be the sources of these elements have been published by the International Strategic Minerals Issues (ISMI) Working Group and are available in the U.S. Geological Survey Circular 930 volume (see appendix).|
National Geochemical Survey data from stream-sediment and soil samples, which have been analyzed using consistent methods, were used to create maps, graphs, and tables that were assembled in a consistent atlas format that characterizes the distribution of major and trace chemical elements in South Carolina. Distribution patterns of the elements in South Carolina may assist mineral exploration, agriculture, waste-disposal-siting issues, health, environmental, and other studies. This atlas is an example of how data from the National Geochemical Survey may be used to identify general or regional patterns of elemental occurrences and to provide a snapshot of element concentration in smaller areas.
A Geochemical Atlas of South Carolina is a set of graphs, maps, and tables that shows analytical values of chemical elements in stream-sediment and soil samples in South Carolina. It was prepared using data from the National Geochemical Survey (NGS), which consists of both reanalyzed National Uranium Resource Evaluation (NURE) stream-sediment and additional soil-sample data collected for the NGS. These data are applicable to mineral exploration, agriculture, waste-disposal-siting issues, health, environmental, and other studies. State and Federal governmental applications include resource surveys to assist mineral exploration by identifying geochemical anomalies and areas of mineralization (Larsen, 1993; Drew and others, 2004). Agriculture seeks to identify areas with favorable (or unfavorable) conditions for plant growth, disease, and crop productivity (Shacklette and others, 1970; Reid, 1991; Reid, 1993; Drew and others, 2002). Trace amounts of elements such as chromium, cobalt, copper, iron, manganese, molybdenum, and zinc must be present within narrow ranges in soils for optimum growth and productivity. Elements, such as arsenic, mercury, and selenium, can be contributing factors to disease, are of concern to health professionals.
The NGS is a collaboration among the U.S. Geological Survey (USGS), other federal and state government agencies, industry, and academia to produce a body of geochemical data for the United States based primarily on the analysis of stream sediments using a consistent set of methods. When fully compiled, these data will constitute a national-scale geochemical coverage of the United States, and will enable the construction of geochemical maps, refine estimates of baseline concentrations of chemical elements in the sampled media, and provide a context for a wide variety of studies in the geological and environmental sciences. The goal of the NGS is to analyze at least one stream-sediment sample in every 289 km2 area across the entire nation using a single set of analytical methods, as well as other solid-sample media, such as soils, that are substituted where necessary. The NGS incorporates geochemical data from a variety of sources, including existing analyses in USGS databases, reanalyses of samples in USGS archives, and analyses of newly collected samples. As of mid-2004, the NGS includes data covering approximately 71 percent of the land area of the United States, including samples in all 50 states (U.S. Geological Survey, 2004).
USGS Open-File Report 2004-1001 provides full documentation of and complete access to NGS data, describes the history of the project, the methodology used, and presents preliminary geochemical maps for all analyzed elements (U.S. Geological Survey, 2004). Future updates of USGS Open-File Report 2004-1001 and other related reports shall include the results of analysis of variance studies, as well as interpretive products related to the NGS data.
The goal of this geochemical atlas was to make available in atlas format the NGS data for South Carolina containing minimal interpretation to encourage prospective users, whatever their purpose, to obtain NGS data for any area and to modify and manipulate it for their end use. This atlas should serve as an introduction to NGS data and to provide an example of what may be done with the data. With this goal in mind, this atlas provides a systematic picture of the geochemistry of South Carolina using samples analyzed by consistent methods under controlled conditions. A principal aim of this atlas is to present patterns of elemental occurrences and provide a snapshot of element concentrations in smaller areas at the time the samples were collected. It provides regional indications of geochemical distribution patterns and should not be used for site-specific studies.
The NGS data for South Carolina consist of analytical values for stream-sediment and soil samples. The stream-sediment samples cover the entire state (figure 1), but the soil samples were collected only within the Atlantic Coastal Plain (figure 2). The dataset consists of analyses for 707 stream-sediment samples and 584 soil samples totaling 1,291 samples; analyses for one additional soil sample were not used because it was analyzed for only arsenic. Analyses of stream-sediment samples represent the relative abundances of chemical elements in stream drainage basins. The soil samples were mainly taken where streams were not available.
Inductively Coupled Plasma (ICP) and Atomic Absorption (AA) are two of the primary geochemical analytical techniques used in the NGS. U.S. Geological Survey Open-File Report 2004-1001 (2004) describes these methods and gives upper and lower limits for each element analyzed. ICP analyses were provided for the major elements aluminum (Al), calcium (Ca), iron (Fe), magnesium (Mg), phosphorous (P), potassium (K), sodium (Na), and titanium (Ti), and the minor (or trace) elements arsenic (As), barium (Ba), beryllium (Be), bismuth (Bi), cadmium (Cd), cerium (Ce), chromium (Cr), cobalt (Co), copper (Cu), europium (Eu), gallium (Ga), gold (Au), holmium (Ho), lanthanum (La), lead (Pb), lithium (Li), manganese (Mn), molybdenum (Mo), neodymium (Nd), nickel (Ni), niobium (Nb), scandium (Sc), silver (Ag), strontium (Sr), tantalum (Ta), thorium (Th), tin (Sn), uranium (U), vanadium (V), ytterbium (Yb), yttrium (Y), and zinc (Zn). AA analyses were provided for arsenic (As), mercury (Hg), and selenium (Se). As arsenic was the only element analyzed by both methods, ICP values for that element were discarded in favor of the more sensitive AA data. Figure 3 shows, in periodic table format, the elements analyzed by these methods. Minimum detection limits for Ag, Cd, Ta, and U were so high that they were not detected in either stream-sediment of soil samples, so they were not considered in this atlas. No Sn was detected in the soil samples, but it was detected in the stream-sediment samples. For several other trace elements, such as Bi, Eu, and Ho, a majority of the samples had values below the detection limits, however, those maps are included in this atlas. NGS data for South Carolina and other areas may be downloaded from the URL http://tin.er.usgs.gov/geochem/.
The maps in this atlas portray the regional occurrences of chemical elements. Regional geochemical studies provide 'background' data of the concentration of elements in rocks, soil, and water, provide a framework to help evaluate geochemical analyses, and have wide application in many disciplines. Coupled with a geologic map--especially in a geographic information system (GIS) environment--the geochemical data may provide useful information for the interpretation of soil, rock, and related physical characteristics, and indicate patterns of chemical concentration or depletion processes.
A geochemical atlas, can be a reference for scientists, interest groups, and public servants with very different interests. A geochemical atlas provides a multi-purpose systematic image of the chemistry of an area or region at the time the samples were taken. The geochemical atlas format has proven popular both for determining background values in unmineralized and uncontaminated areas and as an initial source for locating mineralized or contaminated areas. Numerous geochemical atlases have been published since the late twentieth century. The British Geological Survey (BGS) created a series of 10 geochemical atlases that provide a systematic picture of the geochemistry of Great Britain (see http://www.bgs.ac.uk/gbase/atlas.html). The principal aims of the BGSs Geochemical Baselines Survey of the Environment Project (G-BASE) are to identify new occurrences of metalliferous minerals and to provide quantitative information on natural element levels that may be used to assess contamination, but the data also have potential applications for a range of other disciplines, particularly agriculture, medical geology, land-use planning, and regional geological studies (http://www.bgs.ac.uk/gbase/pdf/history.pdf). In 1998, the USGS published the National Geochemical Atlas for the United States, a predecessor to the NGS, which showed the geochemical landscape of about two-thirds of the conterminous United States derived from stream sediment and other solid sample media analyzed by the NURE Program (Grossman, 1998). Plant and others (2004) published a geochemical atlas showing the distribution in Europe of uranium, and its geological and environmental significance.
Additional maps showing the locations of South Carolinas counties (figure 4), and the known mineral occurrences as reported in the USGS Mineral Resources Data System (MRDS) (figure 5) are included to provide the reader with additional reference information.
The atlas is organized alphabetically by element name with the eight major elements appearing first followed by the trace elements. For details on an individual elements chemical characteristics, many excellent geochemical atlases are available, including Reimann and others (1998), Reid (1991), and numerous online periodic tables of the elements, such as http://www.ktf-split.hr/periodni/en/ and http://environmentalchemistry.com/yogi/periodic/.
Two pages are devoted to each element. One page illustrates element concentrations in the stream-sediment samples and the other element concentrations in the soil samples; the sample type is noted at the center-top of the page. At the top of each page, also, is the name of the element and its chemical abbreviation. On the pages for soil samples, beneath this introductory information are data pertinent to that particular element. At the upper left is the detection minimum, which is the minimum value above which the element could be reliably detected by the method used to analyze the samples. Directly beneath the detection minimum is the default value, which was substituted in the data when the detection minimum was found in the raw data. By convention, the default value is one-half of the detection minimum. It was not necessary to repeat these values on the stream-sediment-sample page.
To the right, beneath the chemical abbreviation, are the crustal abundance of the element and the abundance of the element typically found in granite-granodiorite-type rocks--granite-granodiorite abundance. These values may be useful in comparing the relative values of possible sources of the stream sediments and soils. The granite-granodiorite abundances are from Reimann and others (1998), and the crustal abundances are from Lide (1999) and Reimann and others (1998).
Beneath the information at the top of each element page are statistics for that element and four graphs illustrating those statistics. The column to the left shows statistics for the data analyzed as original concentration units, whereas the column to the right shows the statistics after the data had been transformed to logarithms and the statistics recalculated. Using logarithms is a nonlinear transformation that reduces positive skewness by compressing the upper end (tail) of the distribution while stretching out the lower tail, because the distances between 0.1 and 1, 1 and 10, 10 and 100, and 100 and 1,000 are identical in the logarithmic scale. This is well illustrated by the histograms for strontium in stream-sediment samples. In the original (upper) histogram, using natural numbers, the strontium data are mainly plotted along the left axis and long-tailed to the right, but after the logarithmic transformation is applied (lower histogram), the distribution is much more symmetric.
On the left side of each elements page, the summary statistics for that element list the number of cases analyzed for the sample type, 707 for stream-sediment samples and 584 for soil samples. Minimum and maximum values in the data are the next items listed. For many elements the minimum value is the default value, which shows that in at least one sample that element was not detected and thus the default value was used. At the bottom of the statistic information, samples set to the default value is the number of samples where the element concentration was below the minimum detection limit and the default value was used in the statistics. It provides the reader with a sense as to the sensitivity of the analytical method of the data. In the case of nickel, for example, of the 707 stream-sediment samples analyzed, 356 samples had to be set to the default value, because they had no detectable nickel. Thus, the statistics and maps for nickel cannot be seen as being as sensitive as if a more nickel-sensitive analytical method had been used. The maximum value is the highest value found in the data for that element.
The median and mean of the data are the next values given. The median estimates the central tendency of a distribution. If the data are sorted in increasing order, the median is the value above which half of the values fall (SYSTAT Software Inc., 2002). The mean (or arithmetic mean) of a variable is the sum of the values divided by the number of values (SYSTAT Software Inc., 2002). These values help define and characterize the distribution of the element in the samples analyzed. Two final statistical values are giventhe standard deviation and the variance. The standard deviation is a measure of spread of the data. It is the square root of the sum of the squared deviations of the values from the mean divided by (n-1) (SYSTAT Software Inc., 2002). The variance is the standard deviation squared (SYSTAT Software Inc., 2002). These numbers help further define the distribution of the element in question in the type of sample analyzed.
Four graphs are provided for each sample type for each element. In the left-hand column of graphs are the bar histograms, and in the right-hand column are the cumulative histograms. The graphs in the upper row deal with analysis using the natural original concentration units, and the lower row of graphs are based on the logarithms. The bar histograms show the distributions of the data in real numbers and logarithms with superimposed box plots. In a box plot, the center vertical line marks the median of the sample. The length of each box shows the range within which the central 50% of the values fall, with the box edges (called hinges) at the first and third quartiles. The short vertical lines are called fences and show the distributions beyond the first and third quartiles. Values between the quartiles and the fences and outer fences are plotted with asterisks. Values beyond the fences are plotted with empty circles (SYSTAT Software Inc., 2002).
The cumulative histograms (the graphs on the right-hand side of the page) show the data sorted by value and plotted against a Y-axis that is scaled for a normal distribution. That is, those elements whose data values are normally distributed would plot in a straight line in the upper cumulative histogram, and those elements whose data values are log-normally distributed would plot in a straight line (for example, lithium or neodymium in stream-sediment samples) in the lower cumulative histogram. Bends or breaks in the lines (like those seen in chromium in soil samples) may be evidence that natural geologic processes have affected the distribution of the element in that type of sample.
In several of the cumulative histograms, for example cobalt and gallium soil samples, the values are plotted as a series of vertical bars or separated dots, rather than in a continuous straight or curved line of points, because of the inability of the ICP analytical method to differentiate intermediate values between the individual points.
The relative abundance of each element for each sample type is illustrated on a large-scale South Carolina state map in two ways; 1) by the diameter of the circle representing that site; and 2) by the background color of the map. The maps are small scale; 1 inch = about 53.35 miles, or 1 centimeter = 33.8 kilometers. The maps of the stream-sediment samples show essentially the entire state colored, whereas those for the soil samples show only the Coastal Plain and about 3 km of the adjoining piedmont. The soil-sample maps each have a reminder that those samples were only taken in the Coastal Plain.
The point maps were computer generated in ESRIs ArcGIS commercial software using NGS data. The point data were usually divided into five different sizes of circles with the circle representing the highest values having the largest diameter being shaded light grey for emphasis. These divisions were made at natural breaks in the data.
The colored surface background to the maps was generated as follows: data points for each element were used as input to a spline function, a mathematical calculation capability within the ArcGIS software, which created a smoothed surface representing spatially the concentration of each element fitted to the latitude, longitude, and value of the data. No consideration was given to topography or geology in the creation of the spline surface. Geological and topological features, such as mountains and valleys underlain by particular rock types, and rivers, may have acted as natural barriers or conductors that influenced the shape of the surface or geochemical landscape. With the computer software, the smoothed surface created by the spline function was sliced horizontally, usually into seven equal-area slices, although for some elements the range of the data was not wide enough to allow for seven divisions. The slicing was done using the equal area option so that each colored slice has a similar number of grid cells, that is, each of seven (the number of zones on a majority of the maps) zones represents an equal amount of area with blue and green representing the areas having the lowest element concentrations, blue being the lowest, and pink and red representing those areas with the highest element concentrations, red being the highest. The maps were exported from the ArcGIS software and copied to a page format in Adobe Illustrator 10.
Avery A. Drake, Jr., Lawrence J. Drew, and Suzanne W. Nicholson reviewed this report and offered many suggestions for its improvement. The National Geochemical Survey Team, made up of Jeffrey N. Grossman, Andrew E. Grosz, Peter N. Schweitzer, and Paul G. Schruben, conceived of, built, and maintain the NGS for the USGS and developed data retrieval software and provide website programming.
Drew, L.J., Sutphin, D.M., and Schuenemeyer, J.H., 2002 [extended abstract], Geology and medicine--The recovery and use of a large set of spatial geochemistry data collected in the 1960s and 1970s: International Association of Mathematical Geology, Annual Conference--Creation, Management, Distribution, Access and Exploitation of Digital Spatial Data, Berlin, Germany, September 15-20.
Drew, L.J., Sutphin, D.M., and Gohn, G.S., 2004 [abst.], Associations between bedrock geology and stream and soil geochemical samples in South Carolina: International Association of Mathematical Geology, 2004 annual meeting, Florence, Italy.
Grossman, J.N., 1998, National Geochemical Atlas: The geochemical landscape of the conterminous United States derived from stream sediment and other solid sample media analyzed by the National Uranium Resource Evaluation (NURE) program: U.S. Geological Survey Open File Report 98-622, 1 CD-ROM.
Larsen, C.E., 1993, Heavy minerals at the Fall Zone--A theoretical model of grain size, density, and gradient, in Scott, R.W., Detra, P.S., and Berger, B.R., eds., Advances related to United States and International Mineral Resources--Developing frameworks and exploration technologies: U.S. Geological Survey Bulletin 2039, p. 167-180.
Lide, D.R., editor-in-chief, 1999, CRC Handbook of Chemistry and Physics: CRC Press, Washington, D.C., 84th edition, p. 14-14.
McFaul, E.J., Mason, G.T., Jr., Ferguson, W.B., and Lipin, B.R., 2000, U.S. Geological Survey Mineral Databases--MRDS and MAS/MILS: U.S. Geological Survey Digital Data Series DDS-52, 2 discs.
Plant, J.A., Reeder, Shaun, Salminen, Reijo, Smith, D.B., Tarvainen, Timo, De Vivo, Benedetto, and Petterson, M.G., 2004, The distribution of uranium over Europe--geological and environmental significance: Applied Earth Science, Transactions of the Institute of Mineralogy and Metallurgy Bulletin, v. 112, p. B221-B238.
Reid, J.C., 1991, A geochemical atlas of North Carolina: North Carolina Geological Survey Bulletin 93, 48 p.
Reid, J.C., 1993. A geochemical atlas of North Carolina, U.S.A., in F.W. Dickson and L.C. Hsu (eds.), Geochemical Exploration 1991, Journal of Geochemical Exploration, v. 47, p. 11-27.
Reimann, Clemens; Ayras, Matti; Chekushin, V.A., Bogatyrev, I.V., Boyd, Rognvald; de C.P.; Dutter, Rudolf; Finne, T.E.; Halleraker, J.H.; Jaeger, Oystein; Kashulina, Galina; Lehto, Olli; Niskavaara, Heikki; Pavlov, V.A.; Raisanen, M.L., Strand, Terje; Volden, Tore, 1998, Environmental geochemical atlas of the central Barents region: Geological Survey of Norway, Trondheim, Norway, 745 p.
Shacklette, H.T., Sauer, H.I., and Miesch, A.T., 1970, Geochemical environments and cardiovascular mortality rates in Georgia: U. S. Geological Survey Professional Paper, 574-C, p. C1-C39.
SYSTAT Software Inc. (SSI), 2002, SYSTAT help, in SYSTAT 10.2 for Windows: SYSTAT Software Inc., Richmond, California, licensed software.
U.S. Geological Survey, 2004, The National Geochemical Survey--Database and Documentation: U.S. Geological Survey Open-File Report 2004-1001, version 1.0, http://tin.er.usgs.gov/geochem/doc/home.htm.
Sutphin, D.M., 2005, A geochemical atlas of South Carolina--An example using data from the National Geochemical Survey: U.S. Geological Survey Open-File Report 2004-1368, 97 p.
Send questions or comments about this report to the author, David M. Sutphin at firstname.lastname@example.org.
930A. Manganese (1984)
930B. Chromium (1984)
930C. Phosphate (1984)
930D. Nickel (1985)
930E. Platinum-Group Metals (1986)
930F. Cobalt (1987)
930G. Titanium (1988)
930H. Natural Graphite (1988)
930I. Lithium (1990)
930J. Tin (1990)
930K. Vanadium (1992)
930L. Zirconium (1992)
930M. Niobium (Columbium) and Tantalum (1993)
930N. Rare-Earth Oxides (1993)
930O. Tungsten (1998)
For more information, contact David M. Sutphin
[an error occurred while processing this directive]