Preliminary Report on Mercury 
Geochemistry of Placer Gold Dredge 
Tailings, Sediments, Bedrock, and Waters 
in the Clear Creek Restoration Area, 
Shasta County, California

By Roger P. Ashley1, James J. Rytuba1, Ronald Rogers2, Boris B. Kotlyar1, and David Lawler3

Open-File Report 02-401

2002

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, firm, or 
product names is for descriptive purposes only and does not imply endorsement by the U.S. 
Government.

U.S. DEPARTMENT OF THE INTERIOR
U.S. GEOLOGICAL SURVEY
1 U. S. Geological Survey, Menlo Park, California
2 U.S. Bureau of Land Management, Redding, California 
3 U.S. Bureau of Land Management, Sacramento, California 


TABLE OF CONTENTS

Abstract	Page 1

Introduction	Page 1

Mining history	Page 2

Geology	Page 3

Sample sites	Page 4

Field sampling methods	Page 5
Dry sediments and tailings	Page 5
Wet sediments and tailings	Page 5
Waters	Page 6

Analytical methods	Page 7
Sediments and rocks	Page 7
Waters	Page 7

Results for dry sediments and tailings	Page 8

Results for wet sediments and tailings	Page 9

Effect of sample treatment on analytical results, and implications	Page 10

Results for waters	Page 10

Conclusions	Page 13

References	Page 13


TABLES

1-Locations and descriptions of sample sites in and near the Clear 
Creek Restoration Project area

2-Comparison of preparation and analytical procedures used for solid 
samples by ALS Chemex and Frontier Geosciences laboratories

3-Mercury and methylmercury concentrations in dredge tailings and 
sediment samples

4-Concentrations of major and minor elements in dredge tailings and 
sediment samples, determined by ICP-MS and ICP-AES

5-Field parameters for water sample sites and concentrations of major 
anions, mercury, and methylmercury in waters

6-Concentrations of major and minor cations in waters, determined by 
ICP-MS and ICP-AES


ILLUSTRATIONS

Figure
1-Locations of samples in the Clear Creek study area

2-South Pond, on the flood plain of Clear Creek, showing clouds of fine 
sediment from local tailings used to fill the pond

3-East Pond, a marshy pond occupying an abandoned channel (meander 
scar) on the flood plain of Clear Creek

4-Pond 3, a shallow pond occupying a meander scar on the flood plain of 
Clear Creek

5-Dragline dredge stacker tailings, Spring Creek

6-Dredge sluice tailings (sand, silt, and clay), overlain by stacker tailings 
(pebbles and cobbles), exposed in wall of Schmidt Pit, Shooting Gallery 
area.  Bulk mercury concentrations in the various tailings layers are shown

7-Red Pond, an abandoned dredge pond fed by high-chloride springs

8-Conglomerate of the Tehama Formation, exposed in prospect tunnels 
beneath the Nomlaki Tuff, at Red Pond

9-Mercury concentrations in bedrock, unmined conglomerate, dredge 
stacker tailings, and dredge sluice tailings, as a function of grain size

10-Mercury concentrations in sediments from ponds, hydraulic mine 
drainage tunnel, and Clear Creek

11-Methylmercury concentrations in sediments from ponds, hydraulic 
mine drainage tunnel, and Clear Creek

12-Comparison of total mercury concentrations in sediments and tailings 
determined by two different methods (ALS Chemex method and Frontier 
Geosciences method)

13-Mercury concentrations in waters, showing amounts of dissolved and 
suspended mercury

14-Methylmercury versus total mercury concentrations in waters

15-Sulfate versus total mercury concentrations in waters

16-Iron versus total mercury concentrations in waters

17-Chloride versus total mercury concentrations in waters

18-Chloride versus lithium and boron concentrations in waters


Abstract

Clear Creek, one of the major tributaries of the upper Sacramento River, drains the eastern 
Trinity Mountains.  Alluvial plain and terrace gravels of lower Clear Creek, at the 
northwest edge of the Sacramento Valley, contain placer gold that has been mined since the 
Gold Rush by various methods including dredging.  In addition, from the 1950s to the 
1980s aggregate-mining operations removed gravel from the lower Clear Creek flood plain.

Since Clear Creek is an important stream for salmon production, a habitat restoration 
program is underway to repair damage from mining and improve conditions for spawning.  
This program includes using dredge tailings to fill in gravel pits in the flood plain, raising 
the concern that mercury lost to these tailings in the gold recovery process may be released 
and become available to biota.  The purposes of our study are to determine concentrations 
and speciation of mercury in sediments, tailings, and water in the lower Clear Creek area, 
and to determine its mobility.

Mercury concentrations in bedrock and unmined gravels both within and above the mined 
area are low, and are taken to represent background concentrations.  Bulk mercury values 
in flood-plain sediments and dry tailings are elevated to several times these background 
concentrations.  Mercury in sediments and tailings is associated with fine size fractions.  
Although methylmercury levels are generally low in sediments, shallow ponds in the flood 
plain may have above-normal methylation potential.

Stream waters in the area show low mercury and methylmercury levels.  Ponds with 
elevated methylmercury in sediments have more methylmercury in their waters as well.  
One seep in the area is highly saline, and enriched in mercury, lithium, and boron, similar 
to connate waters that are expelled along thrust faults to the south on the west side of the 
Sacramento Valley.  This occurrence suggests that mercury in waters may at least in part be 
from sources other than placer mining.


Introduction

The Klamath Mountains of northern California and southwestern Oregon have produced 
significant amounts of gold, both from placer and lode deposits.  The most important placer 
deposits occur along the major rivers, including Clear Creek and the Trinity, Klamath, and 
Smith Rivers, and their tributaries (Clark, 1970).  The placers of lower Clear Creek, which 
are the subject of this report, have been mined intermittently by various methods since the 
1850s (Clark, 1970; Averill, 1933), with the result that all the alluvial gravel forming the 
flood plain of Clear Creek and most of the gravel capping adjacent terraces has been 
disturbed.  In addition, in recent decades gravel has been removed from the lower Clear 
Creek alluvial system for aggregate.

Through most of the placer mining period, mercury was used for recovery of gold by 
amalgamation, resulting in contamination of the sediment that was processed.  Elevated 
levels of mercury have been recognized as a water quality problem throughout the 
Sacramento River basin (Domagalski, 1998; Domagalski and others, 2000).  Sources of this 
mercury include mercury mines in the Coast Ranges and mercury losses from gold mining 
operations, both placer and lode, in the Klamath Mountains and Sierra Nevada.  

Habitat restoration for migratory salmon has been underway in the lower Clear Creek basin 
since 1998.  Mercury contents of the sediments impounded behind Saeltzer Dam, in the 
upper part of the lower Clear Creek basin, were examined as part of a floodway 
rehabilitation project involving removal of the dam (Yahnke, 2001).  Sediments that have 
been exposed to mercury have been used for flood plain restoration, and may be used in the 
future.  The purpose of this study is to determine how much mercury has been added to the 
sediments, how it is distributed within them, how it may be released, and whether unusual 
amounts of methylmercury are associated with sediments under any conditions.  In addition 
we have looked at mercury and methylmercury concentrations in springs, ponds, and Clear 
Creek itself within and above the restoration area, to determine whether significant amounts 
of mercury are being leached from sediments into waters, and whether there are conditions 
permitting high rates of methylation.  A study of mercury levels in biota is being carried out 
concurrently to determine whether amounts of methylmercury greater than background 
levels are entering the food chain in the area.  


Mining History

Following the initial discovery of gold in gravels of the Trinity River in 1848, placer 
deposits were the main source of gold in the region until the 1880s (Clark, 1970).  Gold 
was discovered in Clear Creek in 1849.  Several mining camps were established in the 
lower Clear Creek drainage basin, including Whiskeytown, Horsetown, Shasta, and Igo.  
For the first few years, crude methods were used to mine the gravels in the active channel 
and adjacent flood plain; mercury was not available for gold recovery.

In the mid-1850s hydraulic mining was developed.  In this mining method, hydraulic 
monitors were used to excavate off-stream gravels, especially in paleochannels or on 
benches.  Extensive systems of sluices charged with mercury were used to recover the 
gold.  Hydraulic mining of the terrace deposits in the lower Clear Creek area likely took 
place mainly before 1884, at which time most hydraulic mining in tributaries of the 
Sacramento and San Joaquin Rivers was shut down by judicial decree (Sawyer Decision).

From the 1880s until World War I, lode and placer mines produced roughly equal amounts 
of gold in the Klamath Mountains.  Gold production declined in the 1920s, but revived in 
the 1930s as many dredges were put in operation, returning placer deposits to prominence.  
The placer deposits of lower Clear Creek supported numerous dredging operations 
(Averill, 1933).  Large areas in the lower Clear Creek valley and some tributary drainages 
were dredged during this period, using both drag-line and bucket-line dredges.  Most 
mining operations were forced to close in 1942 in response to War Production Board Order 
L-208, and few were able to reopen after World War II.

Bedrock in the area hosts some narrow and shallow gold-quartz veins that yielded small but 
rich pockets.  The most productive of these was the Yankee John mine, located about 3 km 
north of the center of the study area.  The French Gulch district, the largest lode gold 
district in the Klamath Mountains, lies in the Clear Creek drainage basin north of 
Whiskeytown, above the main placer-mining areas.

Aggregate mining began in the lower Clear Creek basin in the 1950s.  Gravel has been 
obtained both from in-stream and off-stream pits. Mining in-stream and in the adjacent 
flood plain ceased in the 1980s, but off-stream mining continues.


Geology

The Klamath Mountains in northern California consist of a series of northwest- to north-
trending terranes, or belts of deformed and metamorphosed sedimentary and volcanic rocks 
ranging in age from Ordovician to Jurassic (Irwin, 1972, 1981).  These belts represent a 
stack of east-dipping thrust plates; from east to west the belts comprise progressively 
younger rocks.  The thrust-fault zones that bound the plates contain ultramafic bodies that 
are mostly serpentinized.  Numerous granitic plutons ranging in age from Devonian 
through Cretaceous intrude the terranes.  Groups of plutons of similar age form belts that 
parallel the trends of their host rocks (Irwin, 1985).

The Clear Creek drainage basin lies entirely within the Eastern Klamath Terrane, the 
easternmost and oldest of the belts recognized in the Klamath Mountains.  The upper part 
of the basin, in the French Gulch area and northward, is dominated by slates of the 
Bragdon Formation of Mississippian age.  The lower part of the basin is underlain by 
metavolcanic rocks of Devonian or older age, including the Copley Greenstone and the 
Balaklala Rhyolite, and granitic rocks including the Mule Mountain stock of Devonian age, 
and the Shasta Bally batholith of Early Cretaceous age (Strand, 1962; Albers, 1964; 
Kinkel, Hall, and Albers, 1956; Fraticelli and others, 1987).

The oldest rocks in the Great Valley province are nearly unmetamorphosed Late Jurassic 
and Cretaceous marine sedimentary rocks of the Great Valley sequence, which crops out 
along the west side of the Sacramento Valley.  Lithologies include mudstone, graywacke, 
conglomerate, and shale.  There are small exposure areas of both Lower and Upper 
Cretaceous on the north side of Clear Creek (Strand, 1962).

Formations of Tertiary and Quaternary age occupy most of the area of the Great Valley 
province, including lower Clear Creek.  Tertiary rocks in the lower Clear Creek area are 
included in the Tehama Formation of Pliocene age (Helley and Harwood, 1986).  It 
consists of sandstone and siltstone with lenses of conglomerate derived from the Coast 
Ranges and Klamath Mountains to the west and north.  The Tehama grades eastward into 
the Tuscan Formation, which consists of volcanic and volcaniclastic rocks erupted and 
transported from volcanic vents in the Cascades volcanic province to the east.

The Nomlaki Tuff Member of the Tehama Formation is locally exposed in the bluffs and 
gulches incised into the terrace on the north side of Clear Creek.  Some of the best 
exposures are in drainage cuts and tunnels related to hydraulic mining.  The Nomlaki is a 
dacitic ash-flow tuff, water-reworked in some areas (Helley and Harwood, 1986).  In the 
vicinity of lower Clear Creek it is typically a white or pale gray massive, non-welded 
pumice lapilli tuff.  Its stratigraphic position is at or near the base of the Tehama Formation, 
and it has been dated at 3.4 million years (Evernden and others, 1964).

The Tehama Formation is overlain unconformably by the Red Bluff Formation of 
Quaternary age, which forms a thin veneer of red, weathered gravels.  Helley and 
Harwood (1986) interpret the Red Bluff as a sedimentary cover on a widespread pediment 
surface that formed in the Sacramento Valley between 450 thousand years and about 1.08 million years.  
It occupies the broad flat divide between lower Clear Creek and Dry Creek to the south, and 
scattered patches remain on the north side of Clear Creek.  It probably covered the terrace 
on the north side of Clear Creek that was extensively mined by hydraulic methods.

The flood plain of Clear Creek, including low terraces adjacent to the active stream channel, 
is underlain by alluvium of Holocene age.  The bulk of this material is probably gravel and 
sand.  As a result of restricted sediment supply in the current hydrologic regime, at many 
places the stream has eroded down to a hard-pan clay layer.


Sample Sites

Samples were collected to characterize mercury species and associated elements in gold 
placer tailings, bedrock, sediments, and stream and pond waters in and adjacent to the 
Clear Creek restoration area.  Samples were collected from August 27 to 29, 2001, during 
late summer base stream-flow conditions. During the time of sampling, the weather was 
stable and no precipitation occurred.  Field sites are shown on Figure 1 and listed and 
described in Table 1 along with all samples collected at each site.  The geographic 
coordinates in Table 1 were obtained in the field using a hand-held Global Positioning 
System (GPS) unit.

Field samples are clustered in four parts of the area (Figure 1 and Table 1).  The eastern 
group of samples (21CC1,2,3,4,5,6,7) includes sediments and waters from three ponds in 
the floodplain.  South Pond (Figure 2), the largest of the three, was an aggregate pit.  The 
other two, designated East Pond (Figure 3)  and Pond 3 (Figure 4), probably represent 
remnants of abandoned channels.  South Pond and Pond 3 were filled in summer 2001 as 
part of the restoration program.

The next group of samples to the west (21CC8,9,10,11) include hydraulic mine tailings, 
water and tailings from a hydraulic mine drainage tunnel, and the Nomlaki tuff, which 
forms the substrate beneath the gravels that were mined by hydraulic methods.  In addition, 
sediments and water from Clear Creek were sampled; this site is immediately above the 
restoration project area, but still within the reach that was mined.

The cluster of samples toward the west end of the area (21CC12,13,14,15,16) comprises 
mainly tailings.  One site yielded gravel from drag-line dredge stacker piles along Spring 
Creek, an intermittent tributary of Clear Creek (Figure 5).  The second site yielded gravel 
from probable bucket-line dredge stacker piles on a low Clear Creek terrace.  The third site 
is a backhoe trench dug in probable bucket-line dredge tailings on the same low terrace, 
both coarse stacker tailings and fine sluice tailings, at the site of a former rifle range known 
as the "Shooting Gallery" (Figure 6).  Also in the Spring Creek drainage is an impoundment 
remaining from dredging known as "Red Pond" (Figure 7), which catches water from saline 
springs.  Adjacent to Red Pond are exposures of Tehama Formation gravels beneath the 
Nomlaki Tuff that contain prospect and short drift mine tunnels (Figure 8).  We sampled Red 
Pond water and sediment, and unmined gravel from a drift tunnel.

The westernmost samples (21CC18,19) are located at the U.S. Geological Survey stream 
gauging station known as the "Igo Gauge."  This site provided water and sediment 
upstream from the mined area.


Field Sampling Methods

Dry sediments and tailings

We separated material coarser than 2 millimeters (the lower size limit for gravel in the standard 
Wentworth size classification scheme) from material finer than 2 millimeters (the upper size limit 
for sand in the Wentworth scheme), using a steel wire-mesh screen.  We did not attempt to 
separate any finer fractions in the field, because dry sieving done by hand is ineffective in 
removing fines from coarser particles.  Also we did not homogenize and split bulk material 
to provide all subsamples; we screened approximately 5 kilograms of material dug from a shallow 
hole or cut from trench walls at each site.  Separate grab samples dug from the same holes 
or cuts supplied material for the bulk subsamples.

All dry tailings subsamples were placed in pre-cleaned and certified borosilicate glass jars 
with Teflon-TM seals (I-CHEM-TM Series 300).  The maximum amount of material retained 
for any subsample was generally 1 to 2 kilograms.


Wet sediments and tailings

Wet sediments and tailings were wet-sieved in the field using stainless steel sieves and 1 to 2 
liters of ambient water.  The sieves used are dedicated to handling of materials expected to 
have low levels of mercury and other metals; however, we have not run blanks to 
determine whether these sieves contribute significant amounts of metals to samples.  
Although several size fractions were obtained at some sites, those analyzed and reported 
here are greater than 2 millimeters, less than 2 millimeters, less than 63 micrometer, and unsieved 
bulk.

Certified Teflon-TM-sealed I-CHEM-TM glass jars (the same as for dry sediments) were used 
for subsamples analyzed for metals.  Polycarbonate jars (100-milliliter capacity) were used for 
mercury subsamples.  These were frozen with dry ice immediately after collection (freezing 
time approximately 20 to 30 minutes), and kept frozen until analysis.


Waters

Spring and stream waters were grab samples collected in 1-liter HDPE bottles; all 
subsamples for analysis of metals and anions were removed from the same 1-liter bottle.

Water sampling protocols used here, including bottle preparation and sample preservation, 
generally follow those of Water Resources Division of the U.S. Geological Survey for 
trace metals (Horowitz and others, 1994). Field filtrations were generally done with 
disposable 25 millimeter-diameter sterile cellulose acetate filters (0.45 micrometer openings) and 
disposable syringes.

For metals determinations, both filtered and unfiltered subsamples were preserved with 
Ultrex-grade HNO3, acidified to pH less than 2, and stored in HDPE (high-density polyethylene) 
bottles pre-rinsed with similar trace-metals grade acid.  Subsamples for anion 
determinations were filtered, stored in HDPE bottles, and chilled to less than 4 degrees Centigrade until analysis.  Subsamples for alkalinity determinations were treated similarly but were not 
filtered.

Sampling for mercury analysis followed ultra-clean sampling and handling protocols 
(Bloom, 1995, Gill and Fitzgerald, 1987) during the collection of field samples and 
analysis to avoid introduction of mercury.  Subsamples for mercury analysis were separate 
grab samples collected in 1-liter borosilicate bottles (I-CHEM-TM Series 300, spot-checked 
for mercury levels by Frontier Geosciences) with excluding Teflon-TM caps (to eliminate 
head space), and kept chilled to less than 4 degrees Centigrade until analysis.

During every sampling event, a field blank was collected by processing de-ionized water 
and collecting the same subsamples (except for alkalinity) by the same procedures as used 
for the field samples.

Water parameters including pH, conductivity, and temperature were measured in the field 
with a battery-powered pH meter (Orion Model 290, with low-maintenance sealed gel 
triode electrode) and a specific conductivity meter (Orion Model 120).  The pH triode, 
which has automatic temperature compensation, was also used for temperature 
measurements.  Dissolved oxygen concentrations were determined with a CHEMets(r) 
colorimetric field test kit (CHEMetrics, Inc.).

Laboratory blanks and acid blanks were processed periodically to determine whether our 
equipment, containers, reagents, and procedures introduced significant contamination.


Analytical Methods

Sediments and rocks

Multi-element analyses for all dry sediments and rocks were performed in the laboratories 
of ALS Chemex.  Bulk samples were ground in a zirconia ring mill and subjected to a near-
total four-acid digestion.  Major elements were determined by inductively coupled plasma-
atomic absorption spectroscopy (ICP-AES).  Minor elements other than mercury were 
determined by inductively coupled plasma-mass spectrometry (ICP-MS).  Mercury was 
determined by cold vapor atomic absorption spectroscopy (CVAAS) following methods 
similar to those described by Crock (1996) and O'Leary and others (1996).

Mercury and methylmercury analyses for all wet sediments were carried out at Frontier 
Geosciences.  Some dry sediments analyzed by Chemex were also analyzed by Frontier.  
For total mercury, the sediment was leached with cold aqua regia, followed by stannous 
chloride (SnCl2) reduction, two-stage gold amalgamation, and cold vapor atomic 
fluorescence spectroscopy (CVAFS) detection.  Methylmercury was obtained by acid 
bromide/methyl chloride extraction followed by aqueous phase ethylation, isothermal gas 
chromatographic (GC) separation, and CVAFS detection (Horvat, Bloom, and Liang, 
1993).  Results were reported on both a wet- and dry-weight basis.

Chemex and Frontier sample preparation steps are shown for comparison in Table 2.  
Chemex preparation utilizes a relatively large split, which is pulverized.  The Chemex 4-
acid digestion dissolves all but a few resistant minerals such as zircon.  It is important to 
note that our samples were treated as rocks rather than soils, except that crushing was not 
needed, so no sieving was performed in the laboratory; bulk sample including clasts as 
large as 1 to 2 centimeters diameter was processed.  Frontier's preparation avoids handling and uses a 
strong leach (aqua regia), which removes elemental mercury, adsorbed mercury, mercury 
in amalgams, and combined mercury including mercury sulfide, but does not affect silicate 
minerals.  Material coarser than sand size, however, is not included in Frontier's leach.


Waters

Alkalinity as CaCO3 was determined in the laboratory by titration with H2SO4 using Gran's 
technique (Orion Research, Inc., 1978), within 2 to 4 days after sample collection.  Sulfate, 
chloride, nitrate, and fluoride concentrations were determined by ion chromatography 
(Fishman and Pyen, 1979).  

Cations were analyzed by inductively coupled plasma-atomic emission spectrometry 
(ICP-AES) and inductively coupled plasma-mass spectrometry (ICP-MS).  Ion 
chromatography and alkalinity analyses were performed by U.S.G.S. laboratories under 
the direction of Paul Lamothe.  The ICP-AES analyses were determined by U.S.G.S. 
laboratories under the direction of Paul Briggs.  Duplicate water samples, blank samples, 
and U.S.G.S. Water Resource Division standard reference waters were analyzed with the data 
set.

At Frontier Geosciences samples were handled in a Class-100 clean air station monitored 
routinely for low levels of total gaseous mercury.  An ultra-clean mercury trace metal 
protocol was followed, including the use of rigorously cleaned and tested Teflon-TM 
equipment and sample bottles and pre-screened and purified reagents.  Laboratory 
atmosphere and water supply are also routinely monitored for low levels of mercury.  
Primary standards used in the laboratory were NIST-certified or traceable to NIST-certified 
materials.  Monomethylmercury (MMHg) standards were made from pure powder and 
calibrated against NBS-3133, and cross verified by daily analysis of Certified Reference 
Material DORM-2 (National Research Council of Canada Institute for National 
Measurement Standards, 1999).  EPA Method 1631 was used.  Total mercury was 
determined by bromine monochloride (BrCl) oxidation followed by SnCl2 reduction, two-
stage gold amalgamation, and detection by cold vapor atomic fluorescence spectroscopy 
(CVAFS) (Bloom, Crecelius, and Fitzgerald, 1988).  Methylmercury was liberated from 
water using an all-Teflon-TM distillation system.  Distilled samples were analyzed using 
aqueous phase ethylation with purging onto Carbotrap, isothermal gas chromatographic 
(GC) separation, and CVAFS detection (Bloom, 1989).  To address accuracy and 
precision, quality assurance measures were employed with the following minimum 
frequency: laboratory duplicates, one per ten samples; method blanks, three per analytical 
batch; filtration blanks, one per ten samples; and spike recovery or standard reference 
material, one per ten samples. 


Results For Dry Sediments and Tailings

Figure 9 shows results for dry tailings, sediments, and bedrock samples; analytical data are 
shown in Tables 3 and 4.  The bedrock category consists of pebble composites for 
metamorphic rocks and plutonic rocks (6 to 10 pebbles each) and Nomlaki Tuff from an 
exposure in a hydraulic mine drainage cut.  The pebbles came from the gravel being used at 
the time of sampling to fill South Pond, which was from the flood plain in the vicinity of 
that pond.  Metamorphic rock types dominate the Clear Creek sediments.  Granitic rock 
clasts are common but subordinate to metamorphic types.  These materials all show 
relatively low mercury values (10 to 30 nanograms per gram).  All subsamples of unmined gravel from 
the Tehama Formation (greater than 2 millimeters, less than 2 millimeters, and bulk) did not have 
detectable levels of mercury, at a 10 nanograms per gram threshold.  These limited data suggest that 
background mercury values in the area are relatively low, and a background range of less than 10 to 30 
nanograms per gram is assumed here for comparison purposes.

The remaining categories in Figure 9 represent results for dredge stacker tailings in the 
Spring Creek area, tailings at the Shooting Gallery, and hydraulic tailings at the hydraulic 
mine drainage cut.  The Clear Creek gravel reported here is from a gravel bar adjacent to the 
active channel.  At all sites where bulk material was analyzed, it is enriched in mercury 
relative to probable background values, but values for the greater than 2 millimeter fraction are 
lower than the bulk values.  In several cases the less than 2 millimeter fraction is higher than the 
bulk value, as expected from the relatively lower values for the greater than 2 millimeter fractions, 
but the two Spring Creek sites show lower values for the less than 2 millimeter fraction than for bulk 
material.  Either sampling or analytical error or both may account for this apparently anomalous 
relationship between values for bulk versus coarse and fine fractions.

Figure 6 shows in more detail the relationship between mercury concentrations and grain 
size.  The photo shows part of the wall of the backhoe trench sampled at the Shooting 
Gallery site.  Here the coarse tailings at the top of the section are very poorly sorted, and 
mercury is clearly enriched in the less than 2 millimeter fraction, which includes sand, silt, and 
clay (values for the fractions from this layer are those shown in Figure 9).  The layer below 
contains the finest material in the section (silt-clay), which shows the highest mercury value 
(160 nanograms per gram), whereas sand interbedded with and underlying the silt-clay shows an 
intermediate value of 90 nanograms per gram.  Bulk values for the greater than 2 millimeter portion of 
the coarse upper layer approach background (40 nanograms per gram).

Various criteria have been proposed for evaluating mercury levels in bulk sediments (U.S. 
Environmental Protection Agency, 1997).  The lowest of these is the U.S. EPA Threshold 
Effects Level (TEL), at 130 nanograms per gram.  This is a proposed lower limit for effects on biota, 
above which there is potential for observable effects such as bioaccumulation and 
biomagnification.  Only the fine sediment fractions at the Shooting Gallery site exceed this 
level.

Methylmercury values for dry tailings and sediments are all very low, ranging from less than 0.015 
nanograms per gram to 0.387 nanograms per gram, indicating that no extraordinary conditions favorable 
for methylation of mercury occur in these materials in the area.


Results For Wet Sediments and Tailings

Figure 10 shows results for wet sediments and tailings.  Mercury values for bulk material 
from the South Pond exceed any bulk values obtained for dry tailings.  This material was 
local tailings being used at the time of sampling to fill the South Pond.  Bulk sediments 
from the other two ponds, which were still undisturbed at the time of sampling, are slightly 
elevated but close to background levels.  Sediment from the mouth of the hydraulic mine 
drainage tunnel, which probably represents hydraulic tailings, is also clearly enriched in 
mercury.  Bulk sediment from the active channel of Clear Creek west of the restoration 
project area showed a mercury concentration similar to those in South Pond and the 
hydraulic drainage tunnel, whereas mercury in Clear Creek sediment at the Igo Gauge is at 
background levels.

At all sites where silt-clay size material (less than 63 micrometers) was separated and analyzed, it 
shows substantially higher mercury concentrations than bulk material, regardless of the bulk 
mercury concentration.  Mud from Red Pond, which is similarly fine-grained, has a 
mercury level similar to fine-grained material at other sites.  Bulk mercury values are 
generally below the EPA TEL of 130 nanograms per gram, whereas fine fractions all exceed it.

Figure 11 shows results of methylmercury analyses for the same sites and subsamples as 
shown in Figure 10.  Methylmercury generally shows the same tendency to be enriched in 
the fine fraction relative to bulk material.  With the exception of Red Pond, methylmercury 
generally accounts for about one percent of total mercury present, which indicates normal 
rates of conversion of mercury to methylmercury.  Conditions in Red Pond, however, 
apparently strongly favor methylation processes.

We have not attempted to evaluate atmospheric deposition of mercury, which could add to 
total mercury in sediments and tailings in the area (Fitzgerald and others, 1998).  Notable 
increases in mercury concentrations in soils and sediments from atmospheric deposition, 
however, are likely to be restricted to within a few centimeters of the surface (see, e.g., 
Rood and others, 1995).  Also we have not yet attempted to evaluate mercury vapor 
emission and redeposition, which could reduce or redistribute mercury.  The observation 
that finer grain-size materials generally show the highest mercury concentrations, which is 
anticipated based on the gold recovery processes used, suggests that gold mining is the 
main source of mercury in the area.


Effect of Sample Treatment on Analytical Results, and Implications

About 60 percent of the sediment and tailings samples, both wet and dry, were analyzed by 
both Chemex and Frontier laboratories using different methods of preparation and analysis 
(see section on analytical methods for sediments and rocks, and Table 2). 

Results are compared in Figure 12.  Identical results fall on the 1:1 line in the figure.  The 
lines on either side of the 1:1 line show a two-standard deviation analytical error envelope.  
The analytical error is not rigorously determined from replicate analysis, but rather is 
estimated from laboratory experience.  The comparison plot demonstrates that samples with 
low mercury concentrations (less than 30 nanograms per gram) show similar results by both methods.  At 
higher mercury concentrations, samples whose median grain size is sand size or smaller 
also yield similar results by both methods.  Results that fall outside the envelope, 
considered likely to be significantly different, are all for samples with large median grain 
size, and in all cases the Frontier analyses show the higher mercury values.  Because the 
Frontier analysis method does not include material from the silicate clasts, which are likely 
low in mercury, this indicates that much of the mercury present is associated with sand or 
smaller grain sizes and localized in aqua regia-leachable mineral grains or coatings on 
silicate grains.  The mercury in leachable grains and coatings is also incorporated in the 
Chemex analyses, but the pulverization and digestion includes low-mercury material from 
clasts.  Although mercury losses from the Chemex drying step cannot be ruled out, the 
similarity between Chemex and Frontier results for other samples, especially those having 
relatively high mercury concentrations and relatively small grain size, suggest that such 
losses are not significant.


Results For Waters

Figure 13 shows results for waters, which were sampled at the same sites as wet sediments 
(Figures 10 and 11; data in Tables 5 and 6).  The stacked bars show the total mercury 
concentrations, as well as proportions of dissolved versus suspended mercury.  
"Dissolved" mercury is here operationally defined as the concentration of mercury 
measured in an aliquot passing a 0.45 micrometer filter.  The filtration was done in the laboratory 
under ultra-clean mercury-free conditions (see section on analytical methods for waters).

Except for water impounded at the outlet of the hydraulic mine drainage tunnel, total 
mercury values are generally low, and similar to many values for waters obtained 
elsewhere in the Sacramento River Basin (Domagalski, 1998; Domagalski and others, 
2000a,b).  In all cases the majority of the mercury present (about 60 to 90 percent) is 
associated with suspended particles, which is also typical of waters elsewhere in the 
Sacramento River basin.  Proposed aquatic life criteria for total mercury in water span a 
wide range. As is the case with the criterion for sediment, this is the level above which 
there is evidence of potential for observable effects in biota, not necessarily including 
bioaccumulation and biomagnification.  The lowest proposed criterion for chronic effects in 
fresh-water aquatic life (12 nanograms per liter total recoverable mercury) is shown on Figure 13 for 
reference.  It was first recommended by EPA in 1986, and is the level recommended by 
BLM's National Applied Resource Sciences Center (Barkow, 1999).  In April 1999 EPA 
issued a guideline of 770 nanograms per liter (dissolved mercury) for chronic effects in fresh-water 
aquatic life.  The 50 nanograms per liter level shown on Figure 13 is the California Water Quality 
Standard.  This number, proposed in 1997, refers to total recoverable mercury and is a 
human health standard.  Only the drainage tunnel sample exceeds the 12 nanograms per liter level, but 
low values are generally anticipated during the low-flow late-summer conditions under which 
we obtained the samples.

With the exception of the Igo Gauge site, where mercury is low in all media, total mercury 
values in waters show no obvious correlation with mercury levels in bulk or fine sediments 
at the same sites.

Figure 14 shows amounts of methylmercury versus total mercury for unfiltered waters.  
Lines of equal proportions of methylmercury are shown for reference (note that the scales 
for both axes and the intervals between the percentage lines are logarithmic).  
Methylmercury values for most of the stream and pond waters fall within the range seen 
elsewhere in the Sacramento River Basin (Domagalski, 1998, 2000b).  In these waters 
between 1 and 10 percent of total mercury is methylmercury, which is within the range of 
values commonly seen in relatively uncontaminated surface waters with normal methylation 
potential (see e.g., Kelly and others, 1995).  Proportions of methylmercury are high in Red 
Pond and Pond 3, producing absolute values higher than those reported by Domagalski 
(1998, 2000b).  At the time of sampling these ponds were shallow, warm, and contained 
decaying organic matter.  The high proportion of methylmercury in Pond 3 is particularly 
significant, as it shows that conditions in shallow ponds on the flood plain may favor 
methylation of mercury.  Further work is needed to explain the extraordinarily low 
methylmercury level seen in water in the hydraulic drainage tunnel.

Sulfate values are in the normal range seen in West Coast rivers (Hem and others, 1990); 
however, values in East Pond and the hydraulic drainage tunnel are comparable to the 
highest values seen in river waters (Figure 15).  Prospect tunnels in the gravels of the 
Tehama Formation exposed above Red Pond in the Spring Creek valley contain fine-
grained pyrite that is now being oxidized, because the tunnels provide drainage and have 
thus lowered the water table.  At the time of sampling, a small area of pyrite-bearing 
conglomerate was also exposed at the north edge of the Clear Creek flood plain near Pond 
3.  This exposure was too small to determine its relations to other units, but it probably 
represents Tehama Formation material covered by thin Recent alluvium.  These exposures 
suggest that the Tehama Formation contains pyrite-bearing reduced zones.  Where exposed 
to oxidation, these zones could be sources of sulfate in waters.

Figure 15 shows that in the Clear Creek area sample suite, waters with relatively high 
sulfate concentrations also have relatively high total mercury values.  Methylmercury, 
however, shows no consistent relationship to sulfate, because methylmercury levels in Red 
Pond and Pond 3 are relatively high, whereas the methylmercury level in the hydraulic 
drainage tunnel water is relatively low (Figure 14).

Iron concentrations vary greatly in the waters sampled (Figure 16), but are high only in the 
Red Pond, as anticipated from the observation that iron oxyhydroxide precipitates on the 
bottom and on pond vegetation during the summer months (Figure 7).  The high iron in 
Red Pond can be explained in light of other chemical characteristics, as discussed below.  
Pond 3 showed surprisingly high iron, although no iron mineral precipitates were 
observed. The source of iron in Pond 3 is not obvious, but oxidizing sulfides in nearby 
bedrock are a possible source for iron, as well as sulfate.  Note that data for filtered 
samples are used in Figure 16, because we do not have unfiltered iron data for all samples.  
In all cases where we have iron analyses for both unfiltered and filtered subsamples, the 
unfiltered values are several times the filtered values.  This indicates that the majority of 
iron in the waters is present in suspension, probably as iron oxyhydroxide particles or 
coatings on other particles, a condition that is commonly seen in natural waters.

Chloride concentrations (Figure 17) in pond and stream waters, which fall between about 
1.5 and 5 ppm, are in the normal range for surface waters near the Pacific Ocean (Hem and 
others, 1990).  The higher concentrations in East Pond and Pond 3 (about 10 to 15 ppm) are 
within the range seen, but since both are shallow small-volume ponds, the values more 
likely reflect effects of evapoconcentration.

Red Pond has an extremely high chloride concentration (13,000 milligrams per liter), approaching that 
of seawater (approximately 30,000 milligrams per liter).  High-chloride waters that represent connate 
fluids from the Great Valley sequence at depth appear in springs to the south along the west side 
of the Sacramento Valley, and contribute to fluids in The Geysers geothermal field 
(Donnelly-Nolan and others, 1993) and to ore-forming fluids and groundwaters at the 
McLaughlin gold deposit (Rytuba and others, 1993).  A contribution from such high-
salinity fluid probably also accounts for the elevated chloride concentration in the hydraulic 
mine drainage tunnel.  Although such fluids may be enriched in mercury, total mercury 
does not correlate with chloride at these two sites.

High boron and lithium values are distinguishing geochemical features of the Great Valley 
connate fluids.  As Figure 18 shows, high concentrations of these elements accompany the 
chloride in the Red Pond water, and confirm that connate water is a major component in the 
seeps that feed Red Pond.  The intermediate boron and lithium values seen in the hydraulic 
drainage tunnel water, in approximately the same ratio to Red Pond water as chloride, 
confirm that it contains a small contribution of connate water.  The contributions of connate 
waters at these sites suggests that waters of deep-seated origin may contribute mercury to 
the surface environment in the Clear Creek area, and that gold mining may not be the only 
source of mercury.


Conclusions

The following preliminary conclusions are drawn from the results of mercury 
determinations reported here.

1) Background mercury values are low, as determined from bedrock, clasts in gravels, and 
unmined auriferous gravels.

2) Bulk mercury values in sediments and tailings are elevated to as much as several times 
background concentrations.

3) Mercury and methylmercury in sediments and tailings are associated with fine grain-size 
fractions.  

4) Mercury and methylmercury values are relatively low in Clear Creek stream waters and 
ponds on the Clear Creek flood plain.

5. High-salinity connate waters enter the surface environment in the area through springs 
and seeps.  Some mercury in waters in the area may be from this source.

These conclusions are considered preliminary, primarily because the number of sample 
sites is relatively small, and the data for waters are derived from only one sampling event.  
Thus for solids, spatial variations cannot be evaluated and the full range of mercury values 
cannot be determined.  For waters, which vary both spatially and temporally, a single 
sampling event is not enough for reliable characterization.

Characterization of mercury concentrations and speciation in inorganic media of the Clear 
Creek system will continue in 2002 and 2003, in coordination with studies of mercury in 
biota.  Detailed studies of mercury mobility are underway, focussed on fine-grained size 
fractions, and colloids that may be released from sediments and tailings.


References

Albers, J.P., 1964, Geology of the French Gulch quadrangle, Shasta and Trinity Counties, 
California: U.S. Geological Survey Bulletin 1141-J, 70 p.

Averill, C.V., 1933, Gold deposits of the Redding and Weaverville quadrangles: California 
Division of Mines Report 29, p. 3-73.

Barkow, Lee, 1999, Site characterization for abandoned mine/mill sites, modification 1: 
U.S. Bureau of Land Management, National Applied Resource Sciences Center, 
Information Bulletin No. RS-2000-021, November 18, 1999, 3 p.

Bloom, N.S., 1989, Determination of picogram levels of methylmercury by aqueous phase 
ethylation, followed by cryogenic gas chromatography with cold vapour atomic 
fluorescence detection: Canadian Journal of Fisheries and Aquatic Sciences, v. 46, p. 
1131-1140.

Bloom, N.S., 1995, Mercury as a case study of ultra-clean sample handling and storage in 
aquatic trace metal research: Environmetal Laboratory, v. 3-4, p. 20-25.

Bloom, N.S., Crecelius, E.A., and Fitzgerald, W.F., 1988, Determination of volatile 
mercury species at the picogram level by low temperature gas chromatography with 
cold vapor atomic fluorescence detection: Analytica Chimica Acta, v. 208, p. 151-161. 

Clark, W.B., 1970, Gold districts of California: California Division of Mines and Geology 
Bulletin 193, 186 p.

Crock, J.G., 1996, Mercury: Chapter 29 in Sparks, D.L., ed., Methods of Soil Analysis, 
Part 3, Chemical Methods: Soil Science Society of America Book Series, Number 5, p. 
769-791.

Domagalski, J., 1998, Occurrence and transport of total mercury and methyl mercury in the 
Sacramento River basin, California: Journal of Geochemical Exploration, v. 64, p. 
277-291.

Domagalski, J.L., Dileanis, P.D., Knifong, D.L., Munday, C.M., May, J.T., Dawson, 
B.J., Shelton, J.L., and Alpers, C.N., 2000a, Water-quality assessment of the 
Sacramento River Basin, California: water-quality, sediment and tissue chemistry, and 
biological data, 1995-1998, U.S. Geological Survey Open-File Report 00-391, 
http://water.wr.usgs.gov/sac%5Fnawqa/waterindex.html.

Domagalski, J.L., Knifong, D.L., Dileanis, P.D., Brown, L.R., May, J.T., Connor, 
Valerie, and Alpers, C.N., 2000b, Water quality in the Sacramento River Basin, 
California, 1994-1998: U.S. Geological Survey Circular 1215, 36 p.

Donnelly-Nolan, J.M., Burns, M.G., Goff, F.E., Peters, E.K., and Thompson, J.M., 
1993, The Geysers-Clear Lake area, California-thermal waters, mineralization, 
volcanism, and geothermal potential: Economic Geology, v. 88, p. 301-316.

Evernden, J.F., Savage, D.E., Curtis, G.H., and James, G.J., 1964, Potassium-argon 
dates and the Cenozoic mammalian chronology of North America: American Journal of 
Science, v. 262, p. 145-198.

Fishman, M.J. and Pyen, G., 1979, Determination of selected anions in water by ion 
chromatography: U.S. Geol. Survey Water Resources Invest. 79-101, 30 p.

Fitzgerald, W.F., Engstrom, D.R., Mason, R.P., and Nater, E.A., 1998, The case for 
atmospheric mercury contamination in remote areas: Environmental Science and Technology, 
v. 32, p. 1-7.

Gill, G.A. and Fitzgerald, W.F., 1987, Picomolar mercury measurements in seawater and 
other materials using stannous chloride reduction and two-stage gold amalgamation 
with gas phase detection: Marine Chemistry, v. 20, p. 227-243.

Fraticelli, L.A., Albers, J.P., Irwin, W.P., and Blake, M.C., Jr., 1987, Geologic map of 
the Redding 1x2 degree quadrangle, Shasta, Tehama, Humboldt, and Trinity Counties, 
California: U.S. Geological Survey Open-File Report 87-257, scale 1:250,000, text 24 
p.

Helley, E.J., and Harwood, D.S., 1985, Geologic map of the Late Cenozoic deposits of 
the Sacramento Valley and Northern Sierran Foothills, California: U.S. Geological 
Survey Miscellaneous Field Investigations Series Map MF-1790, scale 1:62,500 (Sheet 
5 of 5),   text 24 p.

Hem, J.D., Demayo, Adrian, and Smith, R.A., 1990, Hydrochemistry of rivers and lakes, 
Chapter 9 in Wolman, M.G., and Riggs, H.C., eds., Surface Water Hydrology: 
Geological Society of America, The Geology of North America, v. O-1, p. 189-232.

Horvat, Milena, Bloom, N.S., and Liang, Lian, 1993, Comparison of distillation with other 
current isolation methods for the determination of methyl mercury compounds in low level 
environmental samples; Part 1, Sediments: Analytica Chimica Acta, v. 281, p. 135-152.

Irwin, W.P., 1972, Terranes of the western Paleozoic and Triassic belt in the Southern 
Klamath Mountains, California: U.S. Geological Survey Professional Paper 800-C, p. 
C103-C111.

Irwin, W.P., 1981, Tectonic accretion of the Klamath Mountains, in Ernst, W.G., ed., 
The Geotectonic Development of California: Prentice-Hall, New Jersey, p. 29-49.

Irwin, W.P., 1985, Age and tectonics of plutonic belts in accreted terranes of the Klamath 
Mountains, California and Oregon, in Howell, D.G., ed., Tectonostratigraphic 
Terranes of the Circum-Pacific Region: Circum-Pacific Council for Energy and Mineral 
Resources, Earth Science Series, v. 1, p. 187-199.

Kelly, C.A., Rudd, J.W.M., St. Louis, V.L., and Heyes, Andrew, 1995, Is total mercury 
concentration a good predictor of methyl mercury concentration in aquatic systems?: 
Water, Air, and Soil Pollution, v. 80, p. 715-724.

Kinkel, A.R., Jr., Hall, W.E., and Albers, J.P., 1956, Geology and base-metal deposits 
of the West Shasta copper-zinc district, Shasta County, California: U.S. Geological 
Survey Professional Paper 285, 156 p.

National Research Council of Canada Institute for National Measurement Standards, 1999, 
DORM-2, DOLT-2, Dogfish muscle and liver certified reference materials for trace 
metals: INMS Certified Reference Materials Data Sheet, 
http://www.ems.nrc.ca/ems1.htm , 4 p.

O'Leary, R.M., Hageman, P.L., and Crock, J.G., 1996, Determination of mercury in 
water, geologic, and plant materials by continuous flow-cold vapor-atomic absorption 
spectrophotometry, in Arbogast B.F., ed., Quality Assurance Manual for the Branch of 
Geochemistry, U.S. Geological Survey: U.S. Geological Survey Open File Report 96-
525, p. 42-55.

Orion Research, Inc., 1978, Analytical methods guide (9th ed.): Cambridge, MA, 48 p.

Rood, B.E., Gottgens, J.F., Delfino, J.J., Earle, C.D., and Crisman, T.L., 1995, 
Mercury accumulation trends in Florida Everglades and Savannas Marsh flooded soils: 
Water, Air and Soil Pollution, v. 80, p. 981-990.

Rytuba, J.J.,  Enderlin, D.A., McLaughlin, R.J., and Donnelly-Nolan, J.M., 1993, 
McLaughlin gold and Knoxville mercury deposits-road log, in Rytuba, J.J., ed., 
Active Geothermal Systems and Gold-mercury deposits in the Sonoma-Clear Lake 
Volcanic Fields, California: Society of Economic Geologists Guidebook Series 16, p. 
350-361.

Strand, R.G. (compiler), 1962, Geologic map of California, Redding sheet: scale 
1:250,000.

U.S. Environmental Protection Agency, 1997, The incidence and severity of sediment 
contamination in surface waters of the United States, Volume 1-National Sediment 
Quality Survey: Office of Science and Technology, EPA-823-R-97-006, 6 separately-
numbered chapters and 9 appendices.

Yahnke, J.W., 2001, Lower Clear Creek rehabilitation project-potential mercury 
contamination of sediments for use as spawning gravels: Bureau of Land Management, 
Land Suitability and Water Quality Group, Technical Service Center Report, Denver, 
Colorado, April 19, 2001, 5 p.