is the full text of this report, but the accompanying tables are lacking.
You may download the full report with tables in Adobe pdf format
of suction dredging on water quality, benthic habitat, and biota in
the Fortymile River, Resurrection Creek, and Chatanika River, Alaska
US Environmental Protection
Prussian, Todd V. Royer, and G. Wayne Minshall
Department of Biological Sciences
Idaho State University
Table of Contents
Part I - Suction Dredging in the Fortymile
Part II - Recreational Dredging in
Results and Discussion
This report describes the results of our research
during 1997 and 1998 into the effects of commercial suction dredging
on the water quality, habitat, and biota of the Fortymile River and
recreational dredging on Resurrection Creek and the Chatanika River.
On the Fortymile River, water chemistry, heavy metal concentrations,
riverbed morphology, algal (periphyton) standing crop, and aquatic macroinvertebrate
abundance and diversity were measured in relation to commercial suction
dredging for both years. The focus of our work on the Fortymile in 1997
was on an 8-inch suction dredge (Site 1), located on the mainstem and
a 10 inch dredge located on the South Fork (Site 2a). Our research in
1998 included (1) resampling the 1997 sites on the mainstem and SF Fortymile
to determine recovery after one year, (2) sampling a dredge site on
the South Fork to examine for possible spatial variability in the effects
of large-scale suction dredging on benthic communities (3) sampling
a dredge site on the North Fork Fortymile to determine whether impact
and recovery differ from conditions on the South Fork and the mainstem,
and (4) again sampling unmined sites on the NF and SF to better document
suspected background differences between the two forks in terms of macroinvertebrate
communities. In all of the suction-mined sites studied, dredges were
operated by experienced miners. Sampling was performed at fixed transects
above and below the dredge locations. Additional sampling above the
confluence of the North and South Forks revealed differences in background
conditions in these two main tributaries.
At Site 1, dredge operation had no discernable effect
on alkalinity, hardness, or specific conductance of water in the Fortymile.
Of the factors we measured, the primary effects of suction dredging
on water chemistry of the Fortymile River were increased turbidity,
total filterable solids, and copper and zinc concentrations downstream
of the dredge. These variables returned to upstream levels within 80-160
m downstream of the dredge. The results from this sampling revealed
a relatively intense, but localized, decline in water clarity during
the time the dredge was operating. The impact of suction dredging on
water clarity and heavy metal concentrations may be greater or lesser
than we measured, depending on the type of material the dredge is excavating.
The cross-sectional profiles indicate that the impact
of the dredge piles relative to the width of the Fortymile River was
small. After one year, dredge piles at Site 1 had largely disappeared
following the scouring flows that accompany snow-melt in the Fortymile
drainage. However, at Site 2, dredge piles were clearly discernable
after one year. Macroinvertebrate abundance and diversity were greatly
reduced in the first 10 m below the dredge at Site 1 during 1997, relative
to the upstream reference site. For example, macroinvertebrate abundance
was reduced by 97% and the number of taxa by 88% immediately below the
dredge. The abundance and diversity of macroinvertebrates returned to
values seen at the reference site by 80 to 160 m downstream of the dredge.
A similar decline in macroinvertebrate abundance and diversity was observed
at Site 2a. One year after dredging at both Site 1 and Site 2, recovery
of macroinvertebrate diversity appeared to be substantial. The cumulative
effect of suction dredging on the biota of the Fortymile is a function
of the number of dredges operating concurrently, the size of the dredges,
the strategy and effectiveness of their operators, and the rate and
extent of re-colonization on the excavated dredge piles.
We compared conditions in the North Fork versus the
South Fork of the Fortymile under the hypothesis that the greater background
mining activity (of all types) on the SF would result in reduced macroinvertebrate
abundance and diversity. We also expected that suction dredging would
be relatively less harmful at already impacted sites than at sites that
were less disturbed. An increase in macroinvertebrate density was found
in the NF, relative to the SF, and this we attributed to the lower variability
of benthic organic matter and greater amounts of periphyton standing
crop that occurred in the NF. We could discern no natural reason for
this difference and therefore attribute this result to the greater disturbance
in the SF from all forms of mining, historic and current.
The second component of this project is to examine
the effects of recreational suction dredging on smaller streams in Alaska.
In 1997, sampling was conducted on a single site on Resurrection Creek,
a designated recreational mining stream on the Kenai Peninsula. In 1998,
sampling was conducted on the Chatanika River, known to be popular for
recreational dredging. The Chatanika River was sampled at a location
north of Fairbanks. The results from Resurrection Creek indicated that
there was no difference in the macroinvertebrate community between the
mining area and the locations downstream of the mining area, in terms
of macroinvertebrate density, taxa richness, EPT richness, or food resources.
Results from the Chatanika showed slight downstream decreases in macroinvertebrate
density, but all other measures remained similar to those of the reference
area. In general, our results are in agreement with other studies that
have found only localized reductions in macroinvertebrate abundance
in relation to small-scale suction dredging.
Part I - Suction Dredging in the
This report describes the results of research performed
during 1997 and 1998 to determine the possible impacts of commercial
suction dredging on the water quality, benthic habitat, and biota of
the Fortymile River, Alaska (hereafter, Fortymile). Also described in
this report are the impacts by recreational dredging on the Chatanika
River and Resurrection Creek. This is the first study of its kind to
describe the effects of suction dredge mining on river ecosystems in
In stream ecosystems, aquatic macroinvertebrates have
become the primary assessment tool for resource managers (see Barbour
et al. 1996, Cairns and Pratt 1993). Several characteristics of aquatic
macroinvertebrates, as a group, have led to their general acceptance
as reliable indicators of ecological condition: (1) they are generally
immobile (relative to fish), (2) they consist of a relatively large
number of species that, collectively, display a range of sensitivities
and responses to various types of habitat degradation, (3) they tend
to be ubiquitous throughout streams and rivers, and (4) they are relatively
easy to sample and identify. For these reasons, our assessment of the
effect of suction dredging on the Fortymile, Chatanika, and Resurrection
focused on macroinvertebrates. In addition to aquatic macroinvertebrates,
water chemistry, streambed geomorphology, algal (periphyton) standing
crop, and benthic organic matter (BOM) standing crop also were measured
in relation to suction dredging for both years. The latter two components
form the food base for stream herbivores and detritivores and are vital
to the production and recovery of aquatic macroinvertebrates. Variations
in the sampling method between years are described in the Methods section.
Historically, gold mining occurred throughout the
Fortymile basin and several types of operations are still active, including
placer mining, hydraulic mining, and suction dredging. Large scale placer
mining also occurs in some sections of the Chatanika River and historically
in the lower reaches of Resurrection Creek. Our research was limited
to investigations on the effects of suction dredging. We addressed two
general topics: (1) the effect of relatively large (8-10 inch) commercial
suction dredges on ecological conditions in the Fortymile and (2) the
general effect of smaller (2-6 inch) recreational suction dredges on
benthic habitat and biota in the Chatanika River and Resurrection Creek.
Part I of this report presents the results from the Fortymile; Part
II describes results of small-scale mining within the recreational mining
Suction dredging typically involves excavating the
deeper, largely uninhabited sediments and depositing them on top of
the ecologically more important surface substrates. Sorting and re-deposition
of substrata moved through a dredge were expected to alter the streambed
geomorphology and create "dredge piles" downstream of the dredges. Our
effort here was directed toward determining the size (height, width)
of the dredge piles, relative to the cross-sectional width of the river.
This type of physical disturbance of benthic substrata generally reduces
periphyton standing crop, BOM, and macroinvertebrate density. Thus,
substrata moved through the dredge were expected to support less periphyton
than substrata in undisturbed areas of the river (see Peterson 1996).
Abundance and diversity of macroinvertebrates also were expected to
be sharply reduced in dredged areas, as physical tumbling of substrata
is known to kill and/or dislodge associated organisms (see Resh et al.
1988 for review), in addition to reducing the available food base.
The impact of commercial suction dredging on benthic
organisms was evaluated in 1997 on the South Fork and the mainstem Fortymile
River (Fig 1.). One site was also sampled in the North Fork near the
confluence of the North and South Forks. In addition to resampling the
1997 mainstem and South Fork dredge sites in 1998, we expanded our sampling
to include one dredge site on the North Fork and two additional dredge
sites on the South Fork. We also sampled three reference sites unaffected
by mining activity on the North and South Forks, including the 1997
North Fork Confluence site. Overall, our goals for 1998 were (1) to
determine the potential for recolonization of the previous year's dredge
spoils, (2) to expand the spatial scale of our sampling by including
sites that were dredged early (June), and late (September) in the season,
and in different geomorphic settings (inside and outside of a meander
bend), (3) to sample dredged sites in a less-disturbed portion of the
basin (North Fork) than our other sites, and (4) to compare impact and
recovery potentials of dredge mining between more disturbed (South Fork),
and less disturbed (North Fork) streams in the same basin.
The research on recreational dredging was designed
to assess the potential impacts on the aquatic macroinvertebrate community
in streams from geographically diverse locations and streams known to
have annually repeated, relatively, intense mining occur in the same
location. Several potential sites were examined but most proved to be
unsuitable for study because of the absence of discrete areas of concentrated
suction dredging confounded by other disturbances. Resurrection Creek
contains a section of stream designated for recreational mining activity
by the State Department of Fish and Game and the U.S. Forest Service
and is located on the Kenai Peninsula in Southcentral Alaska. The Chatanika
River has no such designation that we know of, however it appears that
mining is restricted to a section of river near Milepost 60 on the Steese
Highway. The Chatanika River site is known to receive a sizeable amount
of suction dredge activity throughout its available mining season.
Sampling Design - The
majority of our work on the Fortymile in 1997 was conducted at a single
site, with an 8-inch suction dredge operated by an experienced miner
(hereafter, Site 1). Site 1 was located approximately 13 kilometers
(8 miles) upstream of the Taylor Highway-Fortymile River Bridge (approximately
141° 30' W, 65° 17' N; Township 7 south, Range 32 east). Sampling was
performed at fixed transects above, within, and below the dredge location
(Fig. 2). Work at this site occurred from 14 through 17 August 1997,
under baseflow conditions. Less intensive sampling also was conducted
above and below a larger (10 inch) dredge located on the South Fork
Fortymile also by a veteran miner (Site 2a), and near the mouth of the
North Fork Fortymile (NF, Site 4). Sampling at Site 2a and in the NF
was performed from 17-18 August 1997 and was restricted to recently
dredged piles and un-dredged reference areas because the dredge was
not active at the time, due to elevated water levels and turbidity following
an intense rainstorm over an extensive part of the basin.
During 1998, we returned to both Site 1 and Site 2a
to determine the degree to which the areas dredged in 1997 had recovered
relative to the reference areas. At Site 1, the previous year's dredge
piles were re-sampled using the same design as in 1997. At Site 2a,
the area that had been dredged in 1997 was re-sampled and another location,
of different mining history and geomorphic setting, was studied for
the first time (2b). During 1998, we also sampled a dredge site located
on the NF Fortymile (Site 3) to increase the spatial extent of the study
and to determine if the NF and SF respond differentially to effects
of suction dredging. Also in 1998 the reference site near the mouth
of the NF was resampled and a comparable unmined site on the SF just
upstream of the confluence was added for better evaluation of potential
SF/NF background differences.
The Before-After-Control-Impact (BACI) approach is
a powerful and generally accepted sampling design for detecting environmental
impacts (e.g., Smith et al. 1993, Stewart-Oaten et al. 1986, Green 1979).
For the present study, a BACI design was used for water chemistry and
turbidity sampling at Site 1. Water samples were collected prior to
and during dredge operation (Before and After) as well as upstream and
downstream of the dredge (Control and Impact). Single measurements'
were made at each of ten transects. It was not possible to employ a
BACI design for periphyton and macroinvertebrate measurements because
of the logistic problems associated with using an actual dredge and
the limited amount of time available for sampling under baseflow conditions.
Instead, samples at Site 1 were collected upstream and downstream of
the dredge while the dredge was in operation. Five macroinvertebrate
and periphyton samples were collected at each transect, except the 0
m, 5 m, and 10 m transects. Sampling the 0 m, 5 m, and 10 m transects
individually was not practical due to the narrow width of the dredge
piles; collection of five samples across their limited width was not
possible. Therefore, ten macroinvertebrate and periphyton samples were
collected from the 0-10 m area to document conditions immediately below
the dredge. At Site 2a, sampling was limited to recent dredge piles
located 25, 35, and 70 m below the moored dredge, and a reference transect
located 250 m upstream of the dredge. Although the dredge was not in
operation during sampling at Site 2a, it had been in operation during
the preceding week. Finally, the samples from the reference area at
Site 2a were used with similarly collected samples from the mouth of
the NF to compare conditions in the two forks of the Fortymile River.
In 1998, five macroinvertebrate and periphyton samples
were taken from the reference, mined, 20 m, and 40 m locations at Site
1 to determine the extent of recovery after one year. No mining occurred
at Site 1 during the 1998 study period. At Site 2a, samples were taken
from the reference, 35 m, and 70 m transects. At Site 2b, slightly downstream
of Site 2a, samples were taken from three locations that had been dredged
along the inside of a meander bend. Ten samples were taken from an "Upper"
location that had been dredged in late September 1997. Five samples
were taken from two dredged areas slightly downstream of the upper location
that had been dredged within the preceding week. We sampled a single
dredge site on the NF that had been dredged with a 10 inch dredge within
the previous 10 days of our sampling. Samples were taken at locations
that had been dredged, no attempt was made to document the downstream
extent of mining disturbance at this site because of inconsistent (patchy)
dredge operations by the Site 3 dredge operators. Ten samples were taken
from a location not affected by mining in the NF, as well as from each
of three transects within the mined area. In addition to the dredged
locations within the Fortymile basin, ten samples were taken from unmined
locations in both the SF and NF near their junction with the mainstem
(Sites 4 and 6). A second NP location was sampled on request by the
US Geological Survey after an upwelling of groundwater containing arsenic
and other heavy metals was located on the North Fork and is described
in detail below. Ten samples were taken from this location and were
compared to samples taken from upstream of the upwelling.
Field and Laboratory Methods - The methods used throughout
this study are standard and widely accepted techniques in stream ecology.
Published reference sources provide detailed instructions regarding
these methods (Hauer and Lamberti 1996, APHA 1995, Cuffney et al. 1993,
Porter et al. 1993, Platts et al. 1983). These references often provide
multiple methods for sampling a given variable. We selected the techniques
that were most applicable to our work on the Fortymile; specific details
and modifications used on the Fortymile are described below.
Turbidity, the inverse of water clarity, and specific
conductance, a measure of the amount of total dissolved mineral salts
in the water, were measured on location with portable meters (Hach model
2100P and Orion model 135, respectively) immediately after collection
of the water samples. The meters were calibrated on a regular basis,
as indicated in the manufacturer's instructions. Water samples for alkalinity
and hardness were stored in insulated containers after collection to
minimize chemical and biological activity in the water. For analysis,
the samples were sent to the Stream Ecology Center, Idaho State University.
The alkalinity and hardness of each sample was determined in the laboratory
using standard titration methods (APHA 1995).
Samples for total filterable solids were filtered
on location within 3 hours of collection. The filters containing the
samples were stored in insulated containers to minimize bacterial degradation
of filtered organics. Upon completion of the field sampling, the samples
were sent for analysis to the Stream Ecology Center, Idaho State University.
These samples were analyzed by determining the amount of mass lost on
combustion at 550°C for 3 hours. The amount of mass lost on combustion
is equivalent to the organic mass of the sample and is referred to as
ash-free dry mass (AFDM). Standard procedures were used to determine
the AFDM of the samples (APHA 1995). Total settleable solids were measured
on-site immediately after sample collector using Imhoff cones; settleable
solids were measured only while the dredge was in operation.
Water samples from the Fortymile River were collected
for determination of heavy metal concentrations using the "clean hands/dirty
hands" procedure as prescribed by the US Environmental Protection Agency.
All materials (sample containers, filters, coolers, etc.) and protocols
used in the collection of heavy metal samples were provided by US EPA.
Samples were sent for analysis to the US EPA laboratory in Manchester,
WA. In 1998, macroinvertebrates were collected to examine the potential
of these organisms to concentrate heavy metals within their tissues.
Macroinvertebrates were collected from four locations: Alder Creek,
Polly Creek, and two locations on the NP Fortymile. Alder and Polly
creeks are tributaries to the mainstem of the Fortymile; Alder served
as the reference site and Polly as a site that has been mined historically
and currently experiences some mining activity. On the NF Fortymile,
the USGS has identified an area of upwelling groundwater that potentially
is a source for dissolved heavy metals in that river. One of the NF
Fortymile sites from which macroinvertebrates were collected was located
above this possible heavy metal source, the other downstream of it.
After collection, the invertebrates were immediately frozen and kept
frozen until analysis. Analysis of the metal concentrations within the
invertebrate tissues was conducted by James Crock at the USGS, Mineral
Resources Program, Denver. To obtain a sufficient mass of tissue for
analysis, all individuals from a site were combined; thus the results
are based on a single measurement per site. The invertebrates were dried,
pulverized, and weighed. The material was then transferred to a Teflon™
vessel and digested in 10 mL of concentrated nitric acid. One mL of
the solution was diluted to 10 mL and analyzed using the USGS standard
ICP-MS method. Mercury was determined using a cold vapor-atomic fluorescence
spectrometry on a separate 1 mL aliquot diluted to 10 mL in sodium dichromate/nitric
acid (James Crock, personal communication).
Description of streambed morphology was accomplished
by developing cross-sectional profiles (see Platts et al. 1983) of the
river at the transects described above (Fig. 2). Distance out from a
fixed location on the bank was measured along a (Kevlar) cable stretched
taut across the river. At numerous points across the width of the river,
the distance from the cable to the water surface and the total water
depth were measured.
All macroinvertebrate sampling was done with a Portable
Invertebrate Box (PIB) sampler that was modified for use in water deeper
than the height of the sampler. The PIB sampler encompassed 0.093 m2
of streambed (the sampler was approximately 30 cm on a side). The sampler
was placed into position on the streambed and held in place by one operator
while the second operator disturbed the substrata enclosed by the sampler
to dislodge the organisms. A removable 250p.m mesh net was attached
to the downstream end of the sampler to collect the dislodged organisms.
Although designed to be used in deep water, the current velocity of
the Fortymile precluded use of the sampler at most deep-water locations,
particularly those in the center of the river. At some deep-water locations,
SCUBA techniques were used to collect the samples; SCUBA was required
for collection of approximately 5% of the samples collected within the
sediment plume. In general, all macroinvertebrate samples were collected
from near-shore habitats, approximately 2-30 meters from the bank. This
is the same distance from the bank in which the dredge was operating.
Following collection, each sample was placed into
a labeled plastic bag (Whirl-pak brand) to which approximately 10-15
ml of concentrated formalin was added to preserve the organisms. In
the laboratory, the contents of each macroinvertebrate sample were spread-out
in a white sorting tray and all organisms removed. The sorting was accomplished
with the aid of a dissecting microscope of 10X magnification. The organisms
were then identified to the lowest feasible taxonomic level, usually
genus, using published taxonomic references, primarily Merritt and Cummins
(1996), Wiggins (1996), and Stewart and Stark (1993). A reference collection
was established and voucher specimens are located in the Stream Ecology
Center, Pocatello at Idaho State University.
Periphyton samples were collected from individual
rocks located just upstream of each macroinvertebrate sample. Processing
was done immediately after collection of the rock and followed the procedures
of Robinson and Minshall (1986). Briefly, the process involved removing
all material within an enclosed area (3.14 cm2) from the rock surface.
The removed material was then suctioned onto a pre-fired, glass microfiber
filter (Whatman GF/F). Filters were frozen with liquid nitrogen in a
modified dewar flask (Taylor-Wharton model 3DS) and sent to the Stream
Ecology Center, Idaho State University for processing. Periphyton samples
were extracted with reagent grade methanol (Holm-Hansen and Riemann
1978) and the 1997 chlorophyll-a content was determined with a spectrophotometer
(Gilford Instruments model 2600). The 1998 chlorophyll-a samples were
analyzed using a fluorometer in order to detect very low concentrations.
Following centrifugation, approximately 3 ml of the sample was removed
and used in the chlorophyll-a determination, the remaining material
was used for measuring the AFDM of the sample as described above under
total filterable solids.
Water Chemistry and Clarity
At Site 1, dredge operation had no discernable effect
on alkalinity, hardness, or specific conductance in the Fortymile (Fig.
3). Alkalinity ranged from <20 to >50 mg CaCO3/L, regardless of whether
or not the dredge was operating. Hardness ranged from approximately
80 to 115 mg CaCO3/L. Both alkalinity and hardness displayed a large
amount of variability in the immediate vicinity of the dredge whether
or not the dredge was operating. Values of alkalinity and hardness measured
at 320 m below the dredge were similar during operation of the dredge
to values measured when the dredge was not in use (Fig. 3). Specific
conductance showed only slight spatial and temporal variation during
our sampling. Values ranged from 131 to 135 µS/cm, with a small decrease
immediately downstream of the dredge, when in operation (Fig. 3). Turbidity
and total filterable solids (TFS) both displayed an increase below the
dredge (Fig. 4). During operation of the dredge, turbidity increased
from values around 1 NTU upstream of the dredge to values of approximately
25 NTU immediately downstream of the dredge. The elevated turbidity
declined rapidly downstream and by 160 m ( 525 ft) turbidity had returned
to values measured upstream of the dredge. No such increase in turbidity
was recorded when the dredge was not in operation. TFS showed a pattern
similar to that of turbidity, increasing from 3 mg AFDM/L upstream of
the dredge to 46 mg AFDM/L immediately downstream of the dredge (Fig.
4). As with turbidity, TFS did not display an increase downstream of
the dredge when the dredge was not operating. Regardless of whether
or not the dredge was operating, a longitudinal increase in TFS was
measured from 80 m to 320 m downstream of the dredge. At 160 m downstream
of the dredge, values of TFS were 28 and 23 mg AFDM/L during operation
and non-operation, respectively. Total settleable solids showed a pattern
very similar to that observed for TFS (Fig. 5).
During operation of the dredge, specific conductance
and turbidity were measured across the width of the Fortymile at 0,
5, 10, 20, and 320 m downstream of the dredge to identify the proportion
of the river width affected by the dredge plume. Specific conductance
was unaffected by the dredge plume which was located along the right
bank, but did decrease near the left bank (Fig. 6). This decrease was
most likely due to groundwater and/or a small tributary that joined
the Fortymile on the left bank just upstream of the study area.
Unlike specific conductance, cross-sectional measurements
of turbidity from within the dredge plume showed a large increase, relative
to areas outside the plume (Fig. 7). However, at 320 m downstream of
the dredge, cross-sectional variation in turbidity was quite low, ranging
from 1.2 to 2.5 NTU. During this sampling, the dredge was operating
in close proximity to the right bank. Under these conditions, the plume
tended to remain near the right bank and did not extend to the center
of the river. In terms of turbidity, approximately 7% of the river width
was affected by the dredge plume for a distance of less than 320 m.
For the unfiltered samples, two metals, copper and
zinc, showed distinct increases downstream of the dredge (Fig. 8). Total
copper increased approximately 5-fold and zinc approximately 9-fold
at the transect immediately downstream of the dredge, relative to the
concentrations measured upstream of the dredge. For both metals, the
concentrations declined to near upstream values by 80 m downstream of
the dredge. The pattern observed for total copper and zinc concentration
is similar to that for turbidity and TFS (see Fig. 4), suggesting that
the metals were in particulate form, or associated with other sediment
particles. The results of sampling for dissolved heavy metals area are
shown in Table 1. Zinc, arsenic, and copper displayed an average value
downstream of the dredge that was greater than the average value measured
upstream of the dredge (note that samples sizes are low, particularly
upstream of the dredge). Copper displayed the greatest change, increasing
by approximately 3-fold downstream of the dredge. Dissolved lead concentrations
did not appear to be affected by operation of the dredge. Values of
dissolved mercury actually were greater upstream of the dredge, suggesting
that any effect of the dredge was likely within the range of natural
variation. (The operator reported observing deposits of liquid mercury
within the sediments he was working.) For both dissolved and total concentrations,
budgetary limitations precluded multiple sampling across either space
or time, thus the results of heavy metal sampling are only indicative
of likely conditions.
Due to the low densities of macroinvertebrates in
the dredge plume (and in the Fortymile in general) and the short exposure
times, no macroinvertebrates were collected for heavy metal tissue analysis
downstream of the suction dredge. However, results from the 1998 analysis
of macroinvertebrate tissues suggest that these organisms are capable
of concentrating heavy metals at least under conditions of chronic exposure.
Although the data are preliminary in nature, several metals showed substantially
greater concentration in the invertebrates from Polly Creek (mined)
than from Alder Creek (reference), including mercury, zinc, molybdenum,
and arsenic (Table 2). Other metals, such as copper and nickel, did
not exhibit substantial differences between the two sites. The upwelling
area identified by the USGS as a potential source of metals in the NF
Fortymile did not appear to be influencing metal concentrations in macroinvertebrates.
For the metals listed above, nickel was the only metal that showed a
substantial increase (Table 2).
Site 1- Cross-sectional profiles were mapped to quantify
the extent of the dredge piles relative to the width of the river. At
Site 1 only the pile created most recently, 0 m downstream of the dredge,
was visible with our profile mapping (Fig. 9). At the transects 5 and
20 m downstream of the dredge the piles were visually obvious due to
the light color of the excavated material compared to undisturbed riverbed.
However, the piles did not appear as distinct "mounds" in the measurements
made at these transects. One year after active dredging occurred, the
distinct mounds seen in Figure 8 at the 0 m transect were no longer
apparent. There was no discernable dredge pile at the 5 and 20 m areas.
Figure 9 is based on detailed mapping along the right bank of the river
and is drawn to scale to represent the conditions within the streambed
relative to the depth of the river in that area. There is a large width:depth
ratio for Site 1 as indicated by Figure 10. Discernable dredging activity
can be seen within the first 5 m from the right bank. The area that
this particular dredge operation affected was about 6% the width of
Site 2a- In August 1997 partial cross-sectional profiles
were measured every 5 meters, beginning slightly downstream of dredging
activity and continuing for 110 meters, to map a series of dredge piles
along the right bank of the South Fork of the Fortymile (Appendix A).
In July 1998 three transects were re-measured to map the change in location
of the dredge piles (Fig. 1). The dredge pile at 30 m shows a shift
towards the center of the stream, though the overall size remained essentially
the same after one year. A profile of the 40 m transect produced similar
results. Remaining partial cross-sectional profiles are presented in
Site 2b- In July 1998 a second site on the South Fork
was included in our sampling to determine if there are spatial differences
in dredging effects on biota. Cross sectional profiles were measured.
Full cross-sectional profiles were completed for the "upper" pile in
1998 which had been dredged in September of 1997 (Fig. 12) and partial
cross-sections were measured for the upper, middle, and lower locations
(Figs. 13 and 14). Easily discernable dredge piles were observed and
measured between 0, 5, and 10 m below a reference transect at the upper
location for Site 2b. Partial cross-sectional profiles also were measured
to determine the longitudinal extent of the upper dredge pile (Fig 13).
According to our measurements, the upper dredge pile tapered off at
about 35 m. Profiles for the middle and lower dredge areas show another
dredge pile beginning between 80 and l00 m. The lower dredge pile begins
at about 130 m and continues slightly past 140 m (Fig 14). The middle
and lower dredge areas were mined about 7 days prior to our sampling
at Site 2b.
Site 3- Cross-sectional profiles also were measured
at Site 3 in the North Fork. Entire width profiles were measured every
20 m along this reach (Fig. 15) and partial profiles were measured at
various distances between each full profile (Fig. 16). Dredging was
active at the 0 m and 10 m locations and between the 40 and 60 m locations.
There is a large width:depth ratio for Site 3. Figure 13 shows the size
of the dredge piles relative to the entire width of the river for Site
3. The full width profile measured for Site 3 shows distinguishable
channel forms where mining activity had occurred within 10 days of our
sampling at 20 m, 60 m, and 80 m though the 80 m location may simply
be due to natural bed forms. The lack of obvious dredge piles at the
0 m and 40 m locations are most likely because the dredge pile began
slightly upstream of these locations. Dredge piles accounted for approximately
15% of the total channel width at Site 3.
The partial profiles show very distinct dredge piles
5 m downstream of mining activity which can be seen nearly 4 m from
the right bank. 10 m downstream another relatively distinguishable streambed
"rise" is discernable between 4 and 6 m from the right bank. There is
no discernable effect on the streambed 15 m downstream of mining activity
according to these profiles.
Periphyton Standing Crop
At Site 1, 1997 periphyton AFDM was greatest at the
transect upstream of the suction dredge, with a mean value of 1.8 mg
AFDM / cm2 (Fig. 17). Periphyton standing crop was reduced by approximately
2-4 fold at the transects downstream of the dredge. The lowest value,
>0.5 mg AFDM / cm2, occurred in the first 10 m immediately below the
dredge. Unlike other variables, periphyton standing crop did not appear
to recover at subsequent transects downstream of the dredge. At the
320 m transect, for example, AFDM was only 50% of the value measured
upstream of the dredge. Chlorophyll-a concentrations are reduced to
unmeasurable values within the areas dredged and 20 m below the operating
dredge. Measured chlorophyll-a concentrations follow the results of
periphyton standing crop biomass downstream of the operating dredge.
After one year, chlorophyll-a concentrations and periphyton standing
crop biomass in the mined area had returned to values near those from
the unmined reference location, indicating that periphyton is unaffected
by dredging the previous year at this location (Fig 18).
Both periphyton standing crop and chlorophyll-a at
Site 2a showed little response to dredging in comparison to the upstream
reference location in 1997. In 1998, mean chlorophyll-a concentrations
were nearly identical at the reference location to those values in 1997;
however, mean chlorophyll-a concentrations were greater at each of the
dredged locations in 1998 than in 1997 (Fig 19). Periphyton standing
crop in 1998 also increased 2-4 fold in the reference and 25 m locations
and increased slightly less in the 70 m and 100 m locations after one
year (Fig 19).
At Site 2b, periphyton standing crop biomass averaged
between 3 and 4 mg/cm2 for all locations regardless of the year in which
they were dredged. However, mean chlorophyll-a was 2.5 times greater
in the "Upper" location, which had been dredged late in the previous
year, than either of the other two nearby locations that had been dredged
in 1998. The Upper location was dredged late in the 1997 mining season
but sampled only during 1998. The greater amount of chlorophyll-a in
the upper location, compared to the other two (1998) dredge piles is
most likely due to the additional time of recovery (Fig. 20).
Comparisons between the NF and SF Fortymile were conducted
to document differences in background conditions and the potential for
recovery of mined areas in two tributaries with different mining pressures
within the same basin. Mean periphyton biomass was three times greater
in the NF site (Site 4) than in the SF site (Site 6) in 1997. Mean chlorophyll-a
concentrations were 4 times greater in the NF than, in the SF for the
same year (Fig 21).
Site 1- The short-term influence of the suction dredge
on macroinvertebrates appeared to be limited to the first 20-40 m downstream
of the dredge. Two locations were examined upstream of the dredge at
Site 1, the first was approximately 80 m upstream and the second approximately
200 m upstream. In terms of water velocity and substrate characteristics,
the -200 m site was considerably more similar to the habitat downstream
of the dredge than was the -80 m site. For this reason, only the -200
m transect was used as the reference for Site 1.
The abundance of macroinvertebrates at Site 1 was
low, relative to large rivers in other parts of North America (e.g.,
Royer and Minshall 1996). A mean of 270 individuals per m2 was collected
at the reference site; approximately 370 individuals per m2 were found
at the site 160 m downstream of the dredge (Fig. 22). Diversity averaged
6-7 taxa per sample at the reference site and ranged from 1 to 7 taxa
per sample at the sites downstream of the dredge. Taxa within the orders
of Ephemeroptera (mayfly), Plecoptera (stonefly), and Trichoptera (caddisfly)
are considered sensitive to habitat degradation and are used commonly
in aquatic bioassessment. The mean number of EPT taxa was 5 per sample
at the reference site and ranged from <1 to 5 per sample at the sites
downstream of the dredge.
The abundance and diversity of macroinvertebrates
at Site 1 was greatly reduced in the first 10 m below the dredge, relative
to the reference site. Immediately below the dredge (0-10 m) macroinvertebrate
abundance was reduced by 97%, number of taxa by 88%, and number of EPT
taxa by 92%, relative to the site 200 m upstream of the dredge. The
abundance and diversity of macroinvertebrates returned to values seen
at the reference site by 80 to 160 m downstream of the dredge.
The relative abundance of all taxa collected from
the Site 1 in 1997 are presented by transect in Table 3. The order Trichoptera
was the most abundant, in terms of richness, with seven genera represented.
Five genera of Ephemeroptera and two genera of Plecoptera were collected.
Two families of Diptera were found, Simuliidae (blackflies) and Chironomidae
(midges). Other groups included: one genus of Coleoptera (beetles),
Acarina (water mites), Collembolla (springtails), Oligochatea (aquatic
earthworms), and Ostracoda. For all transects, 50% or greater of all
taxa were members of the Chironomidae and the Ephemeroptera.
The sampling conducted in 1998 indicated substantial
recovery at Site 1 from the dredging that occurred in 1997, in terms
of macroinvertebrate diversity. Diversity was notably reduced downstream
of the dredge in 1997 (see above) but in 1998 the difference in diversity
among the four transects was minimal (Fig. 23). For example, at the
location 20 m downstream of the dredge macroinvertebrate diversity was
approximately 6 taxa in 1997 but 17 taxa in 1998. A similar increase
in the number of taxa was observed at all Site 1 transects that were
sampled in both 1997 and 1998. Macroinvertebrate density and the number
of EPT taxa also increased after one year (Fig. 24).
Site 2a- Sampling in 1997 revealed patterns at Site
2a similar to those observed at Site 1. Macroinvertebrate density at
the reference transect was approximately 200 individuals per m2 (Fig.
25). At the transect 25 m downstream of the dredge, density decreased
to approximately 20 individuals per m2 and then increased to about 100
individuals per m2 at the transect 70 m downstream of the dredge. The
number of taxa at the reference transects was equal for Site 1 and Site
2a and showed a similar downstream pattern at both sites. The number
of EPT taxa, however, was considerably less at Site 2a in 1997, although
the downstream pattern was the same as that for Site 1. Recovery of
macroinvertebrate diversity at Site 2a was nearly complete one year
after dredging with approximately 20 taxa at each of the transects (Fig.
26). One year after dredging with a 10 inch dredge at Site 2a, macroinvertebrate
density, richness, and number of EPT taxa also had recovered to pre-mining
conditions (Fig. 27).
Site 2b- A second site was established on the South
Fork of the Fortymile River in 1998 to evaluate the effects of dredging
on a nearby site with different water flow and possibly substrate composition.
This site was on the inside bank of a meander bend, about 800 m downstream
of Site 2a. Site 2b was also used to evaluate the effects of dredging
late in the fall on macroinvertebrate composition. In Figures 28 and
29, locations labeled "Upper" represent an area dredged with a 10-inch
dredge in late September 1997. Locations labeled "Middle" and "Lower"
represent adjacent areas mined within a week of our sampling in July
1998. Comparing Site 2a results with the Upper location of Site 2b revealed
that there were in fact differences in macroinvertebrate density between
the Upper site of Site 2b and the reference area of Site 2a. Mean macroinvertebrate
density at the reference location of Site 2a was 26% of the "Upper"
location of Site 2b, 40% of the "Middle" and nearly 30% of the "Lower"
locations (Fig 28A). The number of EPT taxa per sample present in the
Site 2a reference location were 74% that of the "Upper" location of
Site 2b (Fig 29A). Likewise, the number of Diptera present in each sample
from Site 2a were 72% those present at Site 2b (Fig. 29B) Diptera comprised
between 40 and 80% of the macroinvertebrates per sample at all of our
Site 3- We sampled a single dredge site on the North
Fork in which a 10-inch dredge was operated by an experienced miner
and was actively dredged within 10 days prior to our sampling. This
site consisted of three dredged areas, one beginning at the head of
our study reach (T0), the second stretching the length between 10 and
20 m from the T0 location (T10), and the third encompassing the distance
between 40 and 60 m (T40) from the T0 location. The mined areas at 0
m, 10 m, and 40 m were compared to a reference location in an unmined
area of similar substrate type and water velocity. We were not able
to determine the distance downstream affected by dredging because of
inconsistent dredge operations by the North Fork miners which were caused
by relatively high flows over the duration of our sampling. The study
reach chosen here allowed us to determine the short term recovery (>10
days) of these dredged areas in the North Fork. Our results suggest
that all measures except macroinvertebrate density appeared to fully
recover within 10 days since dredging. Macroinvertebrate density at
the reference location averaged about 1600 organisms per m2 while densities
within the mined areas averaged between 1200 and 1400 organisms/m2 (Fig.
30A). Macroinvertebrate taxa ranged from 10 to 12 per sample for all
locations (Fig. 30B). Mean numbers of EPT taxa ranged from 5 to 6 per
sample (Fig. 30C). Diptera, which comprised the majority of the macroinvertebrate
community at all of the sites sampled, ranged from 60 to 80% in the
NF sites (Fig. 30D).
North Fork/South Fork Comparison - Comparisons between
the North Fork and South Fork were made to determine if the South Fork
macroinvertebrate populations were depauperate due to degraded water
quality from increased mining activity on the South Fork itself and
some of its major tributaries. In 1998 we sampled a different reference
location on the South Fork (Site 6, see Fig. 1) that was nearly 500
m upstream of its confluence with the North Fork and compared this data
with those from an unimpacted reference site several kilometers upstream
on the North Fork (Site 5). We also compared this North Fork reference
site to a location downstream of an upwelling of heavy metals noted
by the USGS near the confluence of the North and South Forks (Site 4).
The upwelling of heavy metals between Sites 4 and
5 appears to have little effect on macroinvertebrate populations in
the North Fork. The number of taxa, number of EPT taxa, and overall
relative abundance of Diptera are nearly identical for both Sites 4
and 5. Macroinvertebrate density was nearly 2500/m2 downstream of the
upwelling and nearly 1500/m2 upstream (Fig 31A). The number of taxa
per sample at all locations ranged from 11 to 12 (Fig 31B). The number
of EPT taxa ranged from 5 at the NF and SF reference areas, to 6 at
the NF confluence area (Fig 31C). Diptera comprised 60 to 80 % of the
macroinvertebrates at all locations (Fig 31D).
Although we did not sample the South Fork confluence
site in 1997, there may be some degree of yearly variation in macroinvertebrate
populations in the South Fork as seen from comparison of reference conditions
from Site 2a (see Fig. 26). In the North Fork however, there appears
to be less yearly variation in macroinvertebrate populations in the
years that we sampled. Even though taxa richness was similar at the
NF and 2a sites in both years, the relative dominance of taxa differed
among the sites (Fig. 32). There was a greater difference in the taxa
abundance of some taxa between years at the SF reference location whereas
there is almost no change in the relative dominance of taxa in the NF
site. The difference is seen in the shape of the curves. Table 4 shows
that the Chironomidae (order Diptera) comprised over 75% of all the
macroinvertebrates present in our samples at Site 4 in 1997 and 82%
in 1998. Baetis comprised 0.5% in 1998, and 5.5% in 1997. In the SF
Diptera comprised about 34% of the macroinvertebrates in 1997 and about
35% in 1998. However, Oligochaeta (Annelida) comprised 32% of the macroinvertebrates
in 1998 and only 8% in a 1997. Baetis, a mayfly, comprised 1.3% of the
macroinvertebrates in 1998 and 5% in 1997.
Benthic Organic Matter
Benthic organic matter (BOM) is a primary source of
carbon and energy for organisms that live on and within the substrate
of the river. In general, the amount of BOM found in the Fortymile was
lower than values from many streams in the contiguous United States
(see Minshall et al 1982), but are similar to other studies from the
interior arctic and subarctic Alaska region (for example, see Miller
and Stout 1989).
Site 1- In 1998, mean amounts of BOM within the mined
area were slightly lower than those found at the reference and downstream
(20, 40 m) areas. BOM at the 20 m location is also much more spatially
variable than at the other locations (Fig. 33). This increased patchiness
may be a result of the downstream redistribution of BOM from upstream
BOM concentrations at Site 2a in 1997 were similar
between reference and mined locations, averaging 5 g per m2 at the reference
location and 9 and 11 g per m2 at the 35 m and 70 m locations, respectively
(Fig. 34). Mean amounts of BOM in 1997 at the reference area was 15%
that of 1998. In 1998, mean BOM at Site 2a ranged from an average of
33 g per m2 at the reference area to 25 and 37 g per m2 at the 35 m
and 70 m areas, respectively. BOM at Site 2b ranged from 23 g per m2
at the locations mined in 1998 (Middle and Lower areas) and averaged
53 g per m2 at the location mined in the late fall of 1997 (Upper area).
These values were similar to those from 1997 for Site 2a, indicating
a yearly variation in BOM of between 15 and 30%. BOM from Site 3 averaged
between 6 and 7 g per m2, and showed little difference in average amounts
between locations (Fig. 35). However, the coefficients of variation
in the mined locations showed considerable variability, particularly
at the 35 m location.
Mean amounts of BOM in both the NF and the SF confluence
locations show considerable differences. At the SF confluence site (Site
6), BOM was more spatially variable and averaged more than twice the
amount found at the NF confluence site (Site 4, Fig. 36).
The primary effect of suction dredging on water chemistry
of the Fortymile River, as detected at Site 1, was increased turbidity,
total filterable solids (TFS), and copper and zinc concentrations downstream
of the dredge. Turbidity and TFS were substantially elevated downstream
of the dredge and the plume of sediment-laden water created by the dredge
was visually obvious. But, although the plume was visually dramatic
it was spatially confined to within 160 m (= 525 ft.) of the dredge
and was restricted to the portion of those days that the dredge was
operating. Furthermore, the effect of the plume was limited to approximately
7% of the width of the river. The results from this sampling revealed
a relatively intense, but very localized, decline in water clarity during
the time the dredge was operating. Wanty et al. (1997) reported turbidity
values of 19 NTU 30.5 m (100 ft) downstream of a 10 inch dredge located
below Wilson Creek on the North Fork Fortymile River. Values returned
to near background levels (3.7 NTU) within the next 30.5 m but remained
slightly above background levels (2.2 - 2.3 NTU) as far as 150 m downstream
(furthest sampling transect). Turbidity values downstream of an 8-inch
dredge operating in the same vicinity were lower because less sediment
was being disturbed and the sediments were coarser and hence settled
more rapidly. The 19 NTU at 30.5 m is comparable to the value we found
at 20 m at Site 1.
Wanty et al. (1997) examined dissolved metal concentrations
60.8 m (200 ft) downstream of a 10-inch and an 8-inch dredge and found
no difference between the sides and center of the dredge plume. In our
study, dissolved metals displayed no clear pattern in relation to the
dredge suggesting the increased concentrations of total copper and total
zinc at Site 1 were likely a result of metals associated with the sediments
excavated by the dredge. As the metal-laden sediments were transported
downstream and deposited on the riverbed, total copper and zinc concentrations
declined. By 80 m downstream of the dredge, copper and zinc concentrations
were similar to those measured upstream of the dredge (see Fig. 8).
These results suggest the need for examining heavy metal accumulation
on the riverbed, rather than instantaneous measures of heavy metal concentrations
in the water column. The examination of heavy metal concentrations in
aquatic macroinvertebrates indicated that at some locations, such as
Polly Creek, the chronic effects of mining may be reflected in the physiological
condition of the biota. However, the degree to which metals within the
tissues of the macroinvertebrates may influence life-history or other
biological traits is unknown.
Discussions with local miners indicated that the amount
of material in the plume is, in part, a function of the type of sediment
that is being excavated from the riverbed. Thus, the impact of suction
dredging on water clarity and heavy metal concentrations may be greater
or lesser than that reported here, depending on the type of material
being excavated. In general, the observed decrease in water clarity
was unlikely to have altered ecosystem function in that area of the
Fortymile. However, the increased sediment load and rapid reduction
in light could cause aquatic organisms to drift (Allan 1995:221-237,
Wiley and Kohler 1984), resulting in reduced macroinvertebrate abundance
and/or delayed re-colonization of dredge piles. The effect of suction
dredging on the abundance of drifting macroinvertebrates was not addressed
in the present study, but drifting is likely an important mechanism
in the interaction between macroinvertebrate abundance and suction dredging.
In particular, organisms capable of drifting may be displaced, but not
killed, by the dredging activities. Those organisms that are entrained
by the dredge will not necessarily be killed. For example, Griffith
and Andrews (1981) examined >3,600 organisms and reported less than
1% mortality for macroinvertebrates entrained through a 3-inch suction
The cross-sectional profiles indicate the impact of
the dredge piles relative to the width of the river was small (see Fig.
10). Assuming widths of 2 m for the dredge pile and 80 m for the river,
the dredge pile would represent 2.5% of the river width. Our results
show that in all four of the dredge sites studied, there were substantial
changes to the bed morphology where dredging had occurred, but there
was no discernable change toward the center of the river. There also
did not appear to be any downstream influence on bed morphology by dredged
sediments, indicating that dredging strongly influenced immediately
adjacent substrates but had little effect beyond, either laterally or
downstream of the dredged area. Though no measurements of substrate
composition were made directly in the Fortymile, it seems likely that
suction dredging has little effect on the size and distribution of bed
sediments. Local miners claim that much of the Fortymile River system
has been mined in recent history and though this is an unsubstantiated
claim, it appears reasonable as we observed no striking differences
between sediment compositions within mined areas and those in reference
areas particularly in the amount of deposited fines. We did observe
that at Site 1, downstream gravels were covered with a fine sediment
within the plume caused by the dredge. Given the shallow depth of bedrock
and the intense scouring action by ice-flows and spring runoff, it is
likely that sediments of all sizes may be well mixed and that fine sediments
do not accumulate at the bed surface.
After one year discernable dredge piles remained at
one of the two sites studied in both years, though reduced in size and
in the South Fork site, shifting toward the stream's center. Thomas
(1985) studied suction dredging in a stream in Montana and reported
that spring flows eliminated dredge piles created along the stream margin.
Likewise, Somer and Hassler (1992) examined the effect of suction dredging
in two northern California streams and observed that dredge piles existed
only seasonally and did not persist beyond springtime high-flows. Based
on our observations and results, it appears likely that the dredge piles
at the locations we examined will remain in place no longer than 1 to
3 years. In many cases the stream channel will return to its pre-dredge
condition in a year as a result of river freezing and the succeeding
ice-action and springtime flows that accompany snow-melt in the Fortymile
The abundance and diversity of aquatic macroinvertebrates
at a given site are closely related to the size, stability, and surface
complexity of the substrata at that site (e.g., Minshall 1984, Hart
1978). In addition, the magnitude of impact a particular disturbance
has on a macroinvertebrate community may be mediated by substrate size;
small rocks are more easily tumbled (i.e., disturbed) than are larger
rocks (Gurtz and Wallace 1984). Thus, the effect that suction dredging
has on the macroinvertebrate community of the Fortymile depends on the
characteristics of the substrata being disturbed. The rate at which
dredge piles are re-colonized also will depend on stability of the individual
substratum. A detailed study requiring a longer period of time than
was available would be required to accurately determine the rate at
which macroinvertebrates re-colonized dredged areas. Studies of smaller
scale dredging impacts have shown complete recolonization within 30
days of the cessation of mining activity. Given the northern extent
of the Fortymile region, the harsh climate and short time available
for production and recolonization, the depauperate macroinvertebrate
structure, and the likely low quality and quantity of available food
resources typical of sub-arctic rivers, recolonization would likely
be extended beyond 30 days. It also is possible that the initially low
abundance and diversity of macroinvertebrate taxa in the Fortymile would
cause rapid recolonization due to the low numbers of organisms required
to call an area "substantially recovered". Without detailed recolonization
studies for longer periods of time, it is difficult to "guess" at potential
times of recovery.
As with water clarity, the effect of suction dredging
on macroinvertebrate abundance and diversity at the locations we examined
was confined spatially to a relatively small area downstream of the
dredge. Other researchers also have documented the localized nature
of suction-dredge effects (Somer and Hassler 1992, Harvey 1986, Thomas
1985), although each of these studies was conducted using smaller, recreational
dredges. In the present study, both abundance and diversity were notably
reduced for 10 m downstream of the dredge at Site 1. By 80 m below the
dredge, however, abundance and diversity appeared unaffected by the
dredge plume. Site 2a displayed a similar pattern, although the sampling
was more spatially limited. The short-term, downstream impact of suction
dredging on macroinvertebrates probably was limited to the same area
in which the dredge plume was visible. Therefore, the percent of the
riverbed being affected by the dredge was small: approximately 7% of
the width for <80 m downstream. The cumulative effect of suction dredging
on the biota of the Fortymile cannot yet be assessed fully, but likely
will depend on the number of dredges operating concurrently and the
distance between them, the size of the dredges, the strategy of the
dredge operators, and the extent of re-colonization that occurs on the
excavated dredge piles. Clearly, the effect of suction dredging will
not be the same for all locations in the Fortymile and/or sizes of dredge.
The results from 1998 indicate that substantial recovery
of the macroinvertebrate community occurs within one year after suction
dredging. At both Site 1 and Site 2a, the transects dredged in 1997
showed, in 1998, taxa abundance curves very similar to the reference
transects (see Figures 23 and 26). Although suction dredging is a very
intense, local disturbance to benthic organisms. the biological and
chemical effects of suction dredging do not appear to extend for more
than a year. However, conditions at these two sites after two years
and at sites 2b and 3 after one year could not be determined prior to
the termination of the project.
The comparison of conditions in the North Fork versus
the South Fork suggests that macroinvertebrate density in this river
system may be a function of annual variation in food resources and physical
conditions, especially flow and suspended sediment (likely caused by
additional mining activity in the SF tributaries). Results from 1997
suggested that greater food abundance (e.g., periphyton and BOM) in
the NF corresponded to an approximately 5-fold greater density of macroinvertebrates.
These comparisons were made under the assumption that the reference
location at Site 2a was representative of the South Fork conditions.
However, our 1998 comparison of the North and South Forks, using an
undredged site in the SF nearest to the confluence of the two streams
(Site 6) and that we believe is more representative of conditions in
the tributary, showed no clear difference in biotic conditions between
the two sites. The results suggest that conditions may vary markedly
among locations and years and suggest that in addition to differences
in food resources differences in physical conditions may be important.
We suggest that other mining activities within the basin, primarily
those in the South Fork tributaries may be important causes of decreased
biotic integrity in some years and locations. However, suction dredge
mining clearly reduces macroinvertebrate densities, diversity, BOM,
and periphyton immediately below dredge activity regardless of the background
conditions, though these effects are local and short lived.
Part II - Recreational Dredging
in Resurrection Creek and the Chatanika River
Recreational gold mining is a popular activity throughout
much of Alaska and suction dredging is a common method used in recreational
mining. Recreational dredges are smaller than those examined on the
Fortymile and typically have intake lines of 2-6 inches in diameter.
Despite the relatively small size of the dredges, streams that are popular
with hobbyists may experience a more intensive mining disturbance than
do larger rivers such as the Fortymile because of the concentrated and
repetitive nature of the mining in these areas. Part II of this report
describes the results of our research into the effects of recreational
suction dredging in several Alaskan streams.
This research was conducted on Resurrection Creek
located on the Kenai Peninsula in 1997 and on the Chatanika River, located
along the Steese Highway north of Fairbanks, in 1998. Resurrection Creek
is designated as a recreational mining site by the State of Alaska and
the U.S. Forest Service and is open to recreational dredging from about
May 15 through July 15 of each year. The Chatanika River is not officially
designated for mining, but is a popular recreational site with few accessible
areas that are open to mining during approximately the same time period.
Our sampling on Resurrection was conducted on 22 August
1997; approximately 5 weeks after recreational dredging in the Resurrection
Creek had ended for the year. The general design was similar to that
described above for sampling on the Fortymile. Four locations were sampled:
(1) within the reach of stream that suction dredging is permitted, (2)
approximately 500 m upstream of the dredged area, (3) approximately
35 m downstream of the dredged area, and (4) an area >500 m downstream
of the dredged area. In each of these locations, five macroinvertebrate
samples and three periphyton samples were collected. Water samples were
collected at the location within the dredged area, but as active dredging
was not occurring, these samples are indicative of conditions in the
stream as a whole. All samples were collected, preserved, and processed
as described above for samples from the Fortymile River.
Sampling on the Chatanika River occurred during July
1 and 2, 1998 approximately two weeks prior to the end of the mining
season for that region. Because there was no designated downstream mining
boundary as there had been for Resurrection Creek, a slightly different
sampling regime was used. Samples were taken at approximate distances
downstream of last distinguishable active mining location within the
river. Transects at "Mined", 50, 100, 150, 300 and 500 m were sampled
on two different days. However, an intense rain within the Chatanika
basin on the second day caused the river to rise and alter conditions
from the first day and therefore the samples beyond 100 m were discarded.
Samples from the Mined ( "0 m" transec location were taken from representative
locations within the entire actively mined area. An area upstream of
any active mining was used as our reference location. Substrate measurements
were also made to document any changes in substrate size or sorting
caused by mining. Approximately 25 stones were chosen at random from
near the location of each macroinvertebrate sample. Each stone was measured
to the nearest cm and embeddedness was determined. Embeddedness is the
portion of stone covered by fine sediments and is an indication of the
amount of interstitial filling.
One-way ANOVA was used to test for statistically significant
differences among the four locations in Resurrection Creek. Prior to
analysis, the data were transformed using either natural log (X) or
arcsin (square root (X)) as appropriate (Zar 1984). Pairwise comparisons
were conducted using the Tukey HSD test.
At the time of sampling, total alkalinity, total hardness,
and specific conductance in Resurrection Creek were 29 mg CaCO3/L, 69
mg CaCO3/L, and 110 µS / cm, respectively. Mean benthic organic matter
(BOM) ranged from approximately 15 to 30 g / m2 among the four sampling
locations (Fig. 37), but ANOVA indicated no significant differences
(p=0.252). Mean chlorophyll-a was greatest in the mining area and the
location immediately downstream, but the differences among the means
were not significant (p=0.182) (Fig. 37). Periphyton AFDM showed a pattern
similar to chlorophyll-a, with the greatest mean values in the mined
area, but here too the differences were not significant (p=0.064) (Fig.
37). The reach of Resurrection Creek in which suction dredging occurs
is bordered by a campground and numerous foot trails along the stream.
The riparian canopy along that section of Resurrection Creek appeared
reduced, relative, to areas downstream, by the activities associated
with recreational mining (e.g., stream-side camping). The reduced riparian
shading (= increased solar radiation) may be responsible for the trend
towards greater periphyton AFDM and chlorophyll-a observed in the mined
area and the location immediately downstream. Additionally, these results
suggest that activities other than the actual dredging, such as long-term
camping, firewood collection, trampling of vegetation, etc., also may
have an impact on streams open to recreational suction dredging.
The pattern seen with periphyton was not observed
for macroinvertebrates in Resurrection Creek. Mean density was 3,700
individuals per m2 in the mined area, and ranged from 4,300 to 4,500
individuals per m2 in the other three locations, although the variability
was large and the differences not significant (p=0.581) (Fig. 38). Total
taxa richness from about 17 to 19 among the four locations (p=0.811).
The number of EPT taxa was not significantly different among the sites
(p=0.415), although the mean values increased from 9.5 at the upstream
location to 11 taxa at the most downstream location (Fig. 38).
Results from the Chatanika River showed a trend toward
decreasing macroinvertebrate density as well as less variable distribution
of those macroinvertebrates with distance from active dredging (Fig
39). Average densities decreased from 6000 per m2 at the reference location,
to 2000 per m2 150 m downstream of the mined area. The number of taxa
per sample was more even among locations, ranging from 10 to 13 taxa
per sample. EPT taxa per sample also showed a slight trend toward decreasing
numbers downstream of the mined area, ranging from 6 EPT taxa at 150
m, to 8 EPT taxa at the reference area. Mean amounts of BOM were greater
within the mined area (10 g/m2) than within the reference area (6 g/m2)
or the 50 and 100 m areas (7 g/m2 each) (Fig. 40). Substrate measurements
showed little change among locations, ranging from 11 to 15 cm. Substrate
embeddedness also averaged 15 to 24 percent (Table 5). The mined areas
showed no discernable trends toward any significant change from the
Based on density, taxa richness, and EPT richness,
there was no difference in the macroinvertebrate community between the
mined area and the locations downstream. The relative abundance of Plecoptera
(stoneflies) was significantly greater at the two downstream locations
than in the mined area (p=0.037) (Fig. 32). However, if the observed
reduction was a result of recreational suction mining, downstream recovery
was rapid (i.e., by 35 m).
In general, other studies on the effects of recreational
suction dredging have reported only localized reductions in macroinvertebrate
abundance (Somer and Hassler 1992, Harvey 1986, Thomas 1985). Studies
that examined temporal recovery have found that macroinvertebrates return
to pre-dredging densities within 30-45 days (Harvey 1986, Thomas 1985).
Our sampling in Resurrection Creek occurred approximately 35 days after
suction dredging had ended for the year. Thus, it is not surprising
that the abundance and diversity of macroinvertebrates was not significantly
different between the mining area and the locations downstream. Results
from a concurrent but separate study not funded by the EPA in 1998 also
suggest considerable redistribution of BOM downstream of mining areas
and reduced numbers of macroinvertebrates (both richness and density)
within those mined areas immediately following the end of the mining
season (A.M. Prussian, pers. comm.).
The results presented here on the effects of recreational
suction dredging on macroinvertebrates are derived from a one-time sampling
of only two streams. All of the streams specified in the litigation,
plus an additional 13 streams were examined for compatibility with the
study design. The two sites presented here represent the best examples
of concentrated mining activity we could find and should be considered
"worst-case" scenarios because both streams receive considerable mining
activity and have relatively well-defined downstream boundaries. The
remaining sites suggested in the litigation were either not as intensively
mined or do not contain easily identified mining boundaries. Together
with the results of other studies, we suggest that the impacts by small-scale
dredging activity are primarily contained within mined areas and persist
for about one month after the mining season. However, other studies
suggest a high degree of variability among streams in terms of impact
caused by small-scale dredges (A.M. Prussian, pers. comm.) confounding
our ability to draw broad conclusions for small-scale mining impacts
on stream ecosystems in the State of Alaska. Additional study is needed
to fully quantify the impact of suction dredge mining on the environment
of Alaska before final conclusions are reached regarding the effects
of this activity on Alaskan streams and their associated plant and animal
This project could not have been completed without
the help of numerous individuals. Gretchen Hayslip (EPA), Steve McGroarty
(Alaska DNR), and Phil North (EPA) assisted in the planning and development
of the study. Field sampling was accomplished with the help of Jeff
Davis, Mike Monaghan, Eric Snyder, and Steve Thomas. Larry Taylor, Pat
Scofield, and Scott Reed and Dave Hatch kindly allowed us access to
their mining sites. In the laboratory, Angela Bright, Christine Fischer,
Jamie Larson, Cary Myler, Cecily Nelson, Mark Overfield, Kelly Sant,
Amanda Rugenski, Maria Blackhorse, Mark Sjostrom, Chris Seeley, and
Jacklynn Johnson helped with sample processing and data entry. Taxonomic
identification of the aquatic macroinvertebrates was performed with
the help of Christina Relyea. Jim Crock and Larry Gough of the USGS
facilitated sample processing and heavy metal analysis of the macroinvertebrates.
The EPA laboratory in Manchester, WA conducted heavy metal analysis
of suspended and dissolved samples. Judy Minshall and Marie Davis assisted
with logistics and purchasing of field supplies. Funding for this research
was provided through a contract with the US Environmental Protection
Allan, J.D. 1995. Stream Ecology. Chapman & Hall,
APHA. 1995. Standard methods for the examination of
water and wastewater, American Public Health Association, Washington,
Barbour, M.T., J. Gerritsen, G.E. Griffith, R. Frydenborg,
E. McCarron, J.S. White, and M.L. Bastian. 1996. A framework for biological
criteria for Florida streams using benthic macroinvertebrates. Journal
of the North American Benthological Society 15:185-211.
Cairns, J., Jr. and J.R. Pratt. 1993. A history of
biological monitoring using benthic macroinvertebrates. Pages 10-27
in (D.M. Rosenberg and V.H. Resh, editors) Freshwater Biomonitoring
and Benthic Macroinvertebrates, Chapman & Hall, New York.
Cuffney, T.F., M.E. Gurtz, and M.R. Meador. 1993.
Methods for collecting benthic invertebrate samples as part of the National
Water-Quality Assessment Program, US Geological Survey, Open-File Report
Green, R.H. 1979. Sampling design and statistical
methods for environmental biologists. Wiley, New York.
Griffith, J.S. and D.A. Andrews. 1981. Effects of
a small suction dredge on fishes and aquatic invertebrates in Idaho
streams. North American Journal of Fisheries Management 1:21- 28.
Gurtz, M.E. and J.B. Wallace. 1984. Substrate-mediated
response of stream invertebrates to disturbance. Ecology 65:1556-1569.
Hart, D.D. 1978. Diversity in stream insects: regulation
by rock size and microspatial complexity. Verh. Internat. Verein. Limnol.
Harvey, B.C. 1986. Effects of suction gold dredging
on fish and invertebrates in two California streams. North American
Journal of Fisheries Management 6:401-409.
Hauer, F.R. and G.A. Lamberti. 1996. Methods in stream
ecology. Academic Press, San Diego.
Holm-Hansen, 0. and B. Rieman. 1978. Chlorophyll a
determinations: improvements in methodology. Oikos 30:438-447.
Merritt, R.W. and K.W. Cummins. 1996. An introduction
to the aquatic insects of North America, Kendall/Hunt Publishing, Dubuque,
Miller, M.C., and J.R. Stout. 1989. Variability of
macroinvertebrate community composition in an arctic and subarctic stream.
Minshall, G.W. 1984. Aquatic insect-substratum relationships.
Pages 358-400 in (V.H. Resh and D.M. Rosenberg, eds.) The ecology of
aquatic insects. Praeger, New. York.
Minshall, G.W. 1982. Characterization and dynamics
of benthic organic matter and invertebrate functional feeding group
relationships in the upper Salmon River, Idaho (USA). Int. Revue ges.
Peterson, C.G. 1996. Response of benthic algal communities
to natural physical disturbance. Pages 375-402 in (R.J. Stevenson, M.L.
Bothwell, and R.L. Lowe, editors) Algal Ecology, Academic Press, San
Platts, W.S., W.F. Megahan, and G.W. Minshall. 1983.
Methods for evaluating stream, riparian, and biotic conditions. USDA
Forest Service, General Technical Report INT-138.
Porter, S.D., T.F. Cuffney, M.E. Gurtz, and M.R. Meador.
1993. Methods for collecting algal samples as part of the National Water-Quality
Assessment Program, US Geological Survey, Open-File Report 93-409.
Resh, V.H., A.V. Brown, A.P. Covich, M.E. Gurtz, H.W.
Li, G.W. Minshall, S.R. Reice, A.L. Sheldon, J.B. Wallace, and R.C.
Wissmar. 1988. The role of disturbance in stream ecology. Journal of
the North American Benthological Society 7:433-455.
Robinson, C.T. and G.W. Minshall. 1986. Effects of
disturbance frequency on stream benthic community structure in relation
to canopy cover and season. Journal of the North American Benthological
Royer, T.V. and G.W. Minshall. Development of biomonitoring
protocols for large rivers in Idaho. Report to State of Idaho, Division
of Environmental Quality, Boise, 55 pages.
Smith, E.P., D.R. Orvos, and J. Cairns, Jr. 1993.
Impact assessment using the before-afterontrol-impact (BACI) model:
concerns and comments. Canadian Journal of Fisheries and Aquatic Sciences
Somer, W.L. and T.J. Hassler. 1992. Effects of suction-dredge
gold mining on benthic invertebrates in a northern California stream.
North American Journal of Fisheries Management 12:244-252
Stewart-Oaten, A., W.W. Murdoch, and K.R. Parker.
1986. Environmental impact assessment: "psuedoreplication" in time?
Thomas, V.G. 1985. Experimentally determined impacts
of a small, suction gold dredge on a Montana stream. North American
Journal of Fisheries Management 5:480-488.
Wanty, R.B., B. Wang, and J. Vohden. 1997. Studies
of suction dredge gold-placer mining operations along the Fortymile
River, eastern Alaska. U.S. Geological Survey Fact Sheet FS-154-97.
Wiggins, G.B. 1996. Larvae of the North American caddisfly
genera (Trichoptera),University of Toronto Press, Toronto.
Wiley, M. and S.L. Kohler. 1984. Behavioral adaptations
of aquatic insects. Pages 101-133 in (V.H. Resh and D.M. Rosenberg,
eds.) The ecology of aquatic insects. Praeger, New York.
Zar. J.H. 1984. Biostatistical Analysis, second edition.
Prentice-Hall, Inc., Englewood Cliffs, New Jersey.