Vicki
J. Watson
Assistant Professor Botany/Environmental Studies
University of Montana
Missoula. MT 59812
Abstract--The
Clark Fork River ecosystem is not just that water that maps call the Clark Fork
River. This ribbon of surface water interacts with ground water, with the local
climate, with the landscape through which it passes, and with the tributaries
that feed it. These interactions acting over time deter- mine the river's nature
or overall condition. As users of the river we tend to be most concerned about
certain aspects of the river's condition, including its water quality, the
quantity and pattern of flow, and the nature and abundance of its biota. The
Clark Fork is a complex river system of the Northern Rockies with a history of
abusive uses, a present of multiple stressful uses, and a future that could be
characterized by a number of scenarios. The river appears to be recovering from
some of the past abuses, and suffering various types of impacts from the current
uses. The future condition of this river system is dependent (1) on its own
ability to recover from the past and to assimilate the present, and (2) on the
wisdom and concern of those who shape its future management.
SUMMARY
OF WATER QUALITY ISSUES IN THE CLARK FORK ECOSYSTEM
Let
us concentrate on the current water quality issues in the Clark Fork, with
occasional looks at the past and future when the need arises to understand the
river's present behavior or our future options. Because of the size and
complexity of this system, I will divide it into three rivers for the purpose of
discussion.
The upper river begins as little more than a creek (160 ft3/s mean daily flow)
born of the union of several smaller creeks. It grows to a medium sized river
(3,000 ft3/s), swollen by three major tributaries (Little Blackfoot River, Rock
Creek, and the Blackfoot River) by the time it reaches the Milltown Reservoir
near Missoula. The principal problem of the upper river is the toxicity
associated with heavy metals leaching and eroding from sites contaminated by
historic mining and smelting operations in the headwaters. The present situation
is greatly improved over the past as a result of treatment systems and settling
ponds installed in the upper river over the past 30 years. Prior to these
control efforts, the upper river was a biological desert. With reduction in
metals loads. the dilution provided by tributaries allowed Clark Fork water
quality to improve to the point that organisms fro~ the tributaries could
recolonize the mainstem. Populations of aquatic insects (see Canton and Chadwick
paper) and fish (see Phillips paper) have shown Considerable recovery. Still,
there continue to be frequent exceedances of the water quality criteria for the
protection of aquatic life and occasional catastrophic events (treatment system
failures or floods that wash toxic sediments from the flood plain into the
river).
Two Superfund sites have been designated in the headwaters
and it is hoped that remedial actions at these sites will improve water quality
and the fishery. However, a number of issues raised at this symposium cause us
to look carefully at these efforts. Paradoxically, fish populations in the Clark
Fork actually decrease as one moves farther downstream from the Superfund sites
(see Phillips paper). A number of explanations for this have been offered. Water
chemistry changes in the downstream direction may increase toxicity or
bioavailability of metals (see Babb and Pagenkopf paper). Tailings washed down
the river years ago and deposited in the flood plain may con- tribute
significantly to the river's metal load (see Rice and Ray paper). Eutrophication
of the uppermost reaches of the river with associated increases in fish food
organisms may make it possible to support greater fish populations. None of
these explanations are mutually exclusive. Although contributing to fishery
problems, siltation, dewatering by irrigation, and loss of habitat caused by
channelization during highway and railroad construction do not seem sufficient
to explain the downstream decline in fish populations according to the regional
fishery biologist (D. Workman, personal communication).
The upper river tributaries, so important to the upper river's condition and
ability to recover, also have problems. The Little Blackfoot with its
respectable brown trout population suffers from dewatering and streambed
manipulation by irrigators. A tributary of Flint Creek which enters the Clark
Fork at Drummond has metal contamination from historic mining and continues to
have trout contaminated with mercury to a level of some health concern. Rock
Creek, a blue-ribbon trout stream of national significance, may be threatened by
sediment loading and siltation of its bed from logging operations. The Blackfoot
River with its high water quality and excellent rainbow trout population (1,000
to 1,500/mile), has numerous tributaries that may be stressed by logging, placer
mining, development, and herbicide spraying occurring in this basin. Sediment
and nutrient loadings to the lakes of the Clearwater River, a tributary of the
Blackfoot, are a particular concern. Continued impacts to these tributaries
could reduce their ability to function as nurseries and sources of recolonizers
for the mainstem. The Blackfoot River's ability to interact beneficially with
the mainstem may already be reduced by the small dam near its confluence with
the mainstem and by the Milltown Reservoir where it joins the mainstem (tagged
Blackfoot River fish have not been found in the Clark Fork--D. Workman, personal
communication). Any future reconstruction of these two structures should address
how a new structure might reduce this impact.
The Milltown Reservoir, which divides the upper river from
the middle river, is another focus of concern. Another Superfund site, the
reservoir is characterized by metal-contaminated sediments that washed downriver
and collected here, contaminating the drinking water of Milltown with arsenic.
The reservoir's dam is slated for reconstruction beginning after high flow in
1986. In the past, annual maintenance drawdowns of the reservoir released
quantities of sediments to the middle river occasionally causing fish kills and
severely exceeding water quality criteria to protect aquatic life. However, at
the request of the Department of Fish, Wildlife and Parks, Montana Power
Company, owner of the dam, adopted a new method of lowering the reservoir that
greatly decreased loss of sediments. Subsequent fish population inventories
suggest that this has had a positive effect on the fishery. Although the pro-
posed reconstruction will release heavy metals to the middle river, studies of
recent drawdowns suggest that sediment release can be kept to a level that
should not cause acute toxicity to fish. Engineering studies of the dam indicate
that reconstruction is necessary to prevent dam failure. Since the new structure
is expected to survive only 50 years, reconstruction does not represent a
permanent solution to the problem of the contaminated sediments in the
reservoir. Montana Power has stated that reconstruction will eliminate the need
for annual drawdowns, which, if true, should be beneficial to water quality.
Middle
River
Water from
the Blackfoot, Bitterroot, and St. Regis Rivers swell the Clark Fork's discharge
to nearly 10,000 ft3/s just above its confluence with the Flathead River. Past
water quality concerns in this section focused on organic wastes originating
from the Missoula Sewage Treatment Plant and a large pulp and paper mill.
Historically, loadings of organic matter and nutrients exceeded the river's
assimilative capacity and produced excessive foaming and growths of Sphaerotilus.
The pulp mill's untreated effluent was sufficiently toxic to cause fish
kills. The mill added primary treatment (settling ponds) soon after it opened,
and the addition of secondary treatment at the mill (1974) and at the sewage
plant (1978) greatly reduced most of these problems. Monitoring of aquatic
insect populations in the middle river over the past 30 years by The Institute
of Paper Chemistry (see Rades paper) suggests that the river has recovered from
the earlier severe pollution problems but is growing steadily more eutrophic,
perhaps as a result of continued urban growth in the Missoula Valley. Given
this, the pulp mill owners' request in 1983 for permission to increase its
discharge and to discharge year round is a cause for some concern.
The public's perception of water quality degradation
throughout the river system, the view that the middle river fishery was below
that which would be supported, and the lack of historic data on this river reach
resulted in wide-spread public support for more information on the possible
impacts of the new pulp mill discharge before a permit was granted. Of
particular concern were the possible effects of year-round discharge during
summer low flow conditions, the cumulative effect of the additional nutrient
loading on downstream reservoirs and lakes, synergistic effects with upper river
metal loadings, and the legal precedent set for the Nondegradation Policy. The
Water Quality Bureau responded with a temporary permit, a promise to prepare an
EIS before issuing a longer permit, and a monitoring program designed to fill in
some of the information gaps. Some of the results obtained during the first year
of monitoring included: no violations of the dissolved oxygen standard were
observed in the middle river (except briefly at the bottom of a reservoir
further downriver); some toxic organics were identified in mill effluent (at
concentrations approaching those lethal to aquatic life); during high flow,
copper exceeded water quality criteria for the protection of aquatic life in the
entire middle river, otherwise metals were at acceptable levels (note the
detection level for cadmium was not low enough to determine if it violated
standards); also during high flow, algal nutrients exceeded EPA criteria for
prevention of nuisance algae.
Perhaps
the three issues of greatest concern here are esthetics, nutrients, and the
fishery. Year-round discharge means that the highly colored mill effluent is
entering the river when it is having its greatest recreational use. Since it
takes several miles for the effluent to mix and become unnoticeable, the
esthetic impact may be greater than that associated with seepage alone.
Nutrients are a concern because of their potential contribution to enrichment of
reservoirs and lakes downstream and because the point sources in this stretch of
river (sewage plant and mill) may be among the most controllable sources of
nutrients to the river. Finally, the fishery is 1ess
than that expected for a river with the characteristics of the middle Clark Fork
according to fishery biologists with the Montana Department of Fish Wildlife and
Parks. Several explanations may be offered: the occasional water quality
criteria violations for metals are sufficient to reduce fish populations; water
quality parameters not presently monitored (such as toxic organics) are limiting
the fishery; water quality problems during extreme 1.Ow flow years are limiting
(the year of monitoring was an average flow year); spawning and rearing habitat
is inadequate. This last possibility brings us to a discussion of the
tributaries of the middle river.
Middle
River Tributaries
The
middle river has several tributaries known to have significant impacts as well
as a number that are relatively pristine. Rattlesnake Creek is a high quality
stream flowing from a protected watershed; however, there is a potential for
impact as it flows through Missoula to reach the Clark Fork. Joining the river
just below Missoula, the Bitterroot River and its tributaries receive
significant inputs of sediment and nutrients from logging, agriculture, and
urban development and contribute a substantial load of these pollutants to the
Clark Fork. Probably of greater significance, however, is the dewatering of the
Bitterroot and its tributaries for irrigation, especially during low flow years,
with obvious fishery impacts. At Darby, the Bitterroot supports about 300 to 350
fish/mi; this drops to 200 to 250 in dewatered sections where young age classes
of fish are especially reduced (D. Workman, personal communication).
A series of tributaries between the Bitterroot and the
Flathead Rivers have sufficient flow and appropriate habitat to function as
important nursery areas for the Clark Fork (Six Mile. Nine Mile. Petty. Fish.
Trout and Cedar Creeks, and the St. Regis River). Most of these suffer some
combination of the following impacts: overgrazing, logging, urban development,
dewatering, highway construction. Other small creeks in this reach of the river
pass through culverts that are not passable by fish. Every effort should be made
to mitigate these impacts or to hold the line at the present level of impact. A
better understanding of the changing condition of these tributaries and their
role in the Clark Fork system is essential.
Problems surfacing in the middle Clark Fork basin include
growing problems of contamination on the Missoula Valley aquifer involving
septic systems, leaking gas tanks and pipelines and pesticide spills. The
interaction of this aquifer and the river is poorly understood. and the
aquifer's impact on river water quality needs to be evaluated, especially
downgradient of Missoula. Additional concern has been expressed for the
potential for acid rain impacts on the poorly buffered waters of lakes and
streams in the Bitterroot Mountains.
Lower
River
The
lower river is a very large river (20,000 ft3/s) formed by the union of the
Clark Fork with the Flathead River. The river flows through a series of
reservoirs and into Lake Pend Oreille in Idaho. Many of the water quality
problems of the lower river are a result of the flow regime of the reservoirs
and the lake. Unlike some reservoirs, these fast-flushing reservoirs offer a
poor environment both for river and lake fisheries. Slowing river flow causes
particles to settle, forming the muddy bottom of a lake--poor habitat for river
bottom insects that support a river fishery. Yet flow is fast enough to inhibit
the development of lake biota (algae and small crustaceans) that sup- port a
lake fishery. And both river and lake fisheries suffer from sudden drastic
drawdowns that leave spawning and rearing areas high and dry. Earlier attempts
to establish a river fishery in the largest reservoir, Noxon, had been
unsuccessful. Maintenance of more stable water levels in recent years has
permitted the development of a bass and perch fishery (see Rumsey and Huston
paper); however, this fishery was surely set back by the recent (April 1985)
drastic drawdown ordered by the Bonneville Power Administration for power
generation purposes.
Other concerns associated with the reservoirs include the
settling of metals from the upper river, eutrophic conditions associated with
nutrient loading, and possible synergistic interactions between these. Sediments
in the lower river reservoirs appear to have higher metal concentrations than do
the sediments of tributaries of the Clark Fork (see Johns and Moore paper).
However, concentrations are much lower than those observed at Milltown, probably
as a result of sediment dilution from tributaries and eroded bank material. It
is not clear whether these sediments represent a hazard to ground water (as at
Milltown) or to aquatic life. Limited sampling in the reservoirs by the Water
Quality Bureau in 1984 revealed reduced oxygen conditions and pH levels for a
short time in the deepest part of Noxon Reservoir, but not to the point that
substantial releases of metals would be expected. However, metal concentrations
were slightly higher in the lower river reservoirs during summer when compared
to spring and fall samples (and bottom samples were slightly higher than surface
samples); hence, more monitoring at this time of year is warranted. Unless the
reservoirs become more eutrophic or show lower dissolved oxygen and pH values in
some years, release of metals to the water column from sediments does not appear
to present a problem.
Depending on their flow regimes, reservoirs, like lakes, can
experience eutrophication as a result of excess nutrient loading. Whether water
quality in the Clark Fork reservoirs will degrade as a result of increased
nutrient loading depends on whether the algal populations are limited by
nutrients or by flushing time. In the case of the Thompson Falls and Cabinet
Gorge Reservoirs, flushing time is almost certainly limiting. In Noxon
Reservoir, nutrients may sometimes be limiting during low flow summer
conditions. Water Quality Bureau data from August 1984 did not find severe algal
blooms in the main channel (visibility was about 13 feet). However, most aquatic
systems display considerable variation from year to year, and without data from
sever- al earlier years it is not possible to determine whether the Champion
discharge increased algal populations significantly. However, it is relevant to
ask if current water quality in Noxon Reservoir is acceptable. Several years of
monitoring on the reservoir should make it possible to determine its aver- age
condition. If water quality is unacceptable, or if it appears to be degrading,
or if the nutrients in the reservoir's sediments appear to be increasing, it
would be appropriate to determine what sources of nutrients could be reduced.
Reservoirs can also support
nuisance growths of aquatic macrophytes or pondweeds in shallow areas. Cabinet
Gorge Reservoir has extensive weed beds that have elicited complaints from
boaters. Such beds often receive most of their nutrients from the sediments and
are limited mainly by the area of sediments shallow enough for sufficient light
to reach the bottom to support their growth. It is likely that the construction
of Noxon Reservoir greatly de- creased the rate at which the Cabinet Gorge
Reservoir is filling in, slowing the development of these shallow areas. Noxon
Reservoir has some weed beds in shallow areas in the Finian Flats region and
where tributaries enter and form shallow deposits. The rate at which Noxon fills
in and the amount of nutrients in the sediments filling it will have the
greatest effect on the formation of such beds. Except where they interfere with
boating or swimming, weed beds are usually not a problem; indeed, they are
beneficial to lake fisheries. However, sudden die offs of large areas as a
result of a drawdown can result in oxygen depletion as the plants decompose.
The Flathead River, the largest tributary of the Clark Fork, has a number of
important water quality concerns. The best known is the enrichment of Flathead
Lake, which appears to have reached a threshold with respect to its ability to
assimilate phosphorus. Like many large lakes, Flathead appears to be phosphorus
limited and its phosphorus load has increased significantly from natural
background levels to a point where this once oligotrophic lake has experienced
blue-green algal blooms for the past few summers (J. Stanford, personal
communication). Control of point sources has begun and the significance of
nonpoint sources is being evaluated.
The Flathead River fisheries downstream of Kerr and Hungry Horse Dams suffer
from water level fluctuations and occasional dewatering of the river. The recent
introduction of the opossum shrimp Mysis relicta to the Flathead region
lakes may have serious impacts on lake fisheries, as it appears to compete
with salmon fry for food. Another concern is the planned development of a
surface coalmine in Canada on Cabin Creek, a headwater tributary of the North
Fork of the Flathead. The release of sediment and heavy metals could cause
significant impacts to the biota of this pristine part of the Flathead
ecosystem.
Prospect
Creek (which enters the Clark Fork at the Thompson Falls Reservoir) flows past
the U.S. Antimony mine/mill. While ground water near the tailing ponds exhibits
elevated levels of dissolved sulfates, sodium, and antimony, surface water is
less affected (only the antimony water quality criteria is approached--see
Shapley and Woessner paper). This loading is probably not significant for the
Clark Fork, and, given the intermittent nature of the creek near the ponds,
there is little damage to fish habitat. However, the mill is operating at a very
low level relative to historic operations. In the event that the mill resumed
larger scale operations, or in the event of a spill from the ponds, more
significant impacts could occur.
One of the greatest concerns for the future of this part of
the Clark Fork is the possible impact of the proposed mining in the Cabinet
Mountains Wilderness. Several mining companies (principally ASARCO and U.S.
Borax) have filed claims for as many as nine mines. If the mining is permitted,
the first mine is likely to be ASARCO's proposed silver mine near Rock Creek
which joins the Clark Fork downstream of Noxon Dam. The biota of pristine Rock
Creek should be very susceptible to sediment and metal loading (metals are most
toxic in extremely soft water such as that of Rock Creek).
One of the greatest water quality concerns on the lower river is the long-term
quality of Lake Pend Oreille--a deep oligotrophic lake in Idaho that receives
most of its inflow from the Clark Fork. Lake area residents perceive that the
lake's water quality is degrading, citing as evidence increased growth of
periphyton (attached algae) on boats and increased growth of algae and aquatic
plants in shallow areas. Lake residents feel that nutrient loading from the
Clark Fork is responsible, although such phenomena might be explained by natural
variation or by increases in shoreline loading associated with increased local
development. However, measurements of midlake water clarity over a number of
years show a disturbing downward trend though the data are insufficient to give
a high level of confidence in this trend (Mike Beckwith. personal
communication).
The algal productivity of such a lake is almost certainly
limited by phosphorus loading to the lake. The magnitude and timing of the
response of a lake to a given change in phosphorus loading depends largely on
its flushing time. In the case of Fend Oreille, flushing time is more difficult
to estimate meaningfully than was true of the reservoirs. The lake's overall
flushing time is 3 years. However, the northern part of the lake, which receives
the inflow of the Clark Fork, also contains the lake outlet. While it receives
most of the loading. this northern arm also flushes more quickly than the
southern part of the lake. Thus. the lake's behavior may not be predict- able
simply from models that relate loading to trophic state. Despite this. our first
objective should be to determine the phosphorus load to the lake and to predict
the likely extent and rate of change in water quality from existing models. If
these models suggest that the lake is in no imminent danger, drastic nutrient
control actions could await several years of frequent assessments of water
clarity (and chlorophyll content) over the summer; this information should help
to determine whether a trend in water quality degradation is discernible.
Additionally, loadings should continue to be assessed to determine the response
of this lake to various loading levels. If models suggest that the lake may be
receiving excess loading, control efforts could begin while the studies are
conducted. By the time any substantial reduction in loading is achieved, several
years of monitoring of the lake's conditions under present loading should be
available to compare to its condition after load reduction.
Given experience with other lake improvement programs, it is
likely that near shore water quality problems will respond most to control of
shoreline sources of nutrients while midlake problems will respond to the
largest overall source, which may well be the Clark Fork. It is important to
realize that nutrients may not be the only threat to the Pend Oreille ecosystem.
Development around the lake is very likely affecting wetlands, littoral areas,
incoming streams, and other areas critical to fisheries and bird populations.
Local concern must address these impacts as well as the Clark Fork loadings.
IN
CONCLUSION--NOT JUST AN INDUSTRIAL SEWER
Like
most rivers with high quality tributaries, the Clark Fork is a resilient system
that responded quickly and strongly with improved water quality and fisheries to
the treatment of point sources of metals and organic matter and to stabilization
of water levels in reservoirs. However, the system and its improved state are
fragile because: (1) diffuse sources of metals in the flood plain of the upper
river wash into the river during floods and storms, (2) treatment system
failures allow occasional toxic shocks to the system, (3) occasional extreme
drawdowns in the lower river reservoirs can wipe out fisheries that have been
building over several years. In addition to such catastrophic events, the river
suffers from the creeping degradation brought on by greater and greater impacts
on its tributaries--especially the effects of dewatering and siltation. The
lakes of the Clark Fork system appear to be suffering from enrichment. The
greatest long-term concerns for the system are the long-term control of metal
loading in the upper river, the loss of fish habitat associated with reservoirs
and with tributary impacts, the enrichment of lakes, and the proposed mine
district in the lower river.
Clearly, the Clark Fork River is a complex natural system
with many problematic interactions with the artificial systems of man. However,
the Clark Fork is not "just an industrial sewer" for a number of
reasons. Many of the water quality problems discussed above do not stem from
traditional industrial point sources. Many come from municipal sources or from
land uses such as logging and agriculture or from flow manipulation for power
generation. Reducing industrial effluents and controlling the metal problem in
the headwaters will not solve all the Clark Fork's problems (although these are
important for certain problems in some reaches).
Additionally, unlike an industrial sewer, the Clark Fork is a
living system, with potential for recovery or further degradation. Although it
is a hard-working river, it is still a river of startling beauty with tremendous
opportunities for fishing and floating, scenery and serenity, peace and
excitement. If we write it off as a sewer, or decide that its problems are too
complex, too expensive, too political, we have much to lose. If we allocate
resources to identify the causes of the most serious problems and deal with
those causes, we have much to gain.
ACKNOWLEDGMENTS
I would like to thank the following for supplying information or ideas used in the above paper {all errors are my responsibility): Glenn Phillips, Peter Rice, Gary Ray, Dennis Workman, Phil Smith, Steve Mehring, Ron Marcoux, Larry Weeks, Loren Bahls, Rod Berg, Bill Woessner, Roger Wordsworth, Fred Runkel, JaGk Stanford, Jill Davies, Gary Engman, Gary Eudaily, Greg Munther, Chris Kronberg, Bill Good, Dave Odell, Robert Bukantis, Mike Beckwith, and others.