|
|
|
|
|
|
|
|
|
|
Volcanic Events, pg. 2 Mount St. Helens History, pg. 3-15 Eyewitnesses, pg. 53-67 Absolute Times, pg. 81-82, 86 Activity Sequence, pg. 127-134 Gas Studies, pg. 190-191 |
Chemical
Compositions, pg.
233-250
Ash Clouds, pg. 323-333 Blast Dynamics, pg. 379-400 Rapid Deposition, pg. 466-478 Phreatic Explosions, pg. 509-511 New Lava Dome, pg. 540-544 Ash-Fall Deposits, pg. 568-584 Water Chemistries, pg. 659-664 River Water Quality, pg. 719-731 |
THE 1980 ERUPTIONS OF MOUNT ST. HELENS, WASHINGTON
SOME EFFECTS OF THE MAY 18 ERUPTION OF
MOUNT ST. HELENS ON RIVER-WATER QUALITY
By JOHN M. KLEIN
ABSTRACT
The May 18 eruption of Mount St. Helens had a pronounced effect on the water quality in rivers draining areas affected by the blast, debris avalanche, mudflows, and ash. Comparison of preemption and posteruption data shows marked increases in selected rivers of the levels of sulfate, chloride, turbidity, total nitrogen, total organic nitrogen, and total iron, manganese, and aluminum.
In streams affected solely
by ash, the changes were short lived, and the intensity of change decreased
with distance from Mount St. Helens. In contrast, the effects were more
persistent in streams that drain the areas affected by the blast and mudflows.
INTRODUCTION
After 100 yr of dormancy, Mount St. Helens, in southwestern Washington, erupted on March 27, 1980; and ash and steam were emitted. Eruptions of ash and steam continued during March and April, and continual seismic activity was recorded.
On May 18, at 0832 PDT, Mount St. Helens erupted violently. Following an explosive north-trending lateral blast, ash and steam were shot vertically to more than 15 km. Immediately north of the volcano, a once-productive forest was pyrolyzed and destroyed. At a greater distance from the mountain, trees were uprooted and stripped of bark, branches, needles, and leaves.
Airborne ash was carried eastward and dispersed over eastern Washington and other States farther downwind (fig. 414). Watersheds just east of the volcano received heavy ash fall. However, streams draining to the south tributary to the Columbia River (fig. 415) received only traces of ash.
At the beginning of the eruption, part of the north flank of the volcano collapsed, depositing a massive debris avalanche of rock, ash, pumice, and ice in the upper 25 km of the North Fork Toutle River valley (fig. 415). A mudflow quickly developed in the South Fork Toutle River, and several hours later, a much larger mudflow, originating from the massive debris-avalanche deposit, flowed down the North Fork Toutle River. The North Fork mudflow moved downstream through the lower Toutle and Cowlitz Rivers, and a considerable part of the sediment was conveyed through the Cowlitz into the Columbia River.
The massive alterations caused by the eruption could be expected to
have marked effects on river-water quality. This report describes some
of the effects caused by ash deposition in rivers east of the volcano,
and the effects of massive debris-avalanche and mudflow deposits in rivers
draining to the west and south.
ASH COMPOSITION
Fruchter and others (1980) have reported some general compositional trends of the volcanic ash at various distances from the volcano. Silica content increased with distance, whereas iron, magnesium, calcium, and manganese decreased. Potassium and sodium showed little variability.
Taylor and Lichte (1980) analyzed volcanic-ash samples for major, minor, and trace-element composition. They also observed variations in chemical composition of ash with distance from Mount St. Helens, and related these to changes in the physical characteristics of the ash. In addition, Taylor and Lichte reported water-soluble components after column leaching the ash with known volumes of distilled water. A summary of some of the major, minor, and trace soluble-constituent concentrations of the ash at specified locations is in table 96.
From these analyses alone, one might expect that a major impact on river-water quality would be an increase, from preemption levels, in the concentrations of certain soluble constituents, especially chloride and sulfate. The timing of such compositional changes with respect to the eruption would depend on whether the ash fell on land or water, and the occurrence and intensity of precipitation on ash that was deposited on land. If ash fell on water, the soluble constituents would dissolve rapidly and be observed in analyses of samples collected soon after deposition. Heavy precipitation on land-deposited ash might wash large quantities of soluble constituents into streams, whereas light precipitation would leach those constituents into the soil where they would react with soil materials. In the latter case, changes in stream-water composition might not be measurable. Moreover, changes would occur after sufficient water had leached through the soil and eventually entered the water course.
Calcium and bicarbonate are normally the predominant dissolved constituents
in streams of the Pacific Northwest. Variations in concentrations are small,
occur slowly, and are correlated with stream discharge. A major geologic
event such as this eruption and subsequent ash fall might be expected to
produce noticeable changes in the chemical composition of waters.
DATA COLLECTION
Following the eruption of March 27, a program of water-quality data collection was initiated at selected stream sites surrounding Mount St. Helens (fig. 415). Measurements at these stations supplemented data collected monthly at several USGS National Stream Quality Accounting Network Stations1 and at other sites sampled periodically.
1The NASQAN consists of more than 500 surface-water monitoring stations at which water-quality measurements of the Nation's streams and rivers are made monthly or bimonthly.
Following the May 18 eruption, additional measuring stations (fig. 415) were established. Initially, samples were collected daily at some of the sites that received heavy ash fall or were affected by the debris avalanche or mudflows. After several days, the sampling frequency was decreased to weekly until the end of June. Since then, samples have been collected monthly.
All water samples were collected from river cross sections using depth-integrating
samplers and methods (Guy and Norman, 1970). Samples to be analyzed for
dissolved trace elements and major chemical constituents were filtered
through 0.45-m (micrometer) membrane filters, placed in acid-washed polyethylene
bottles, and acidified with double-distilled, analytical-grade nitric acid
to prevent absorption and chemical precipitation. Unfiltered samples were
analyzed for total trace-element concentrations. Samples for analysis of
nitrogen and phosphorus were chilled at the time of collection and during
shipment to the USGS laboratory in Arvada, Colo. Constituents were analyzed
by methods described by Skougstad and others (1979). Data from only some
of the stations in the overall sampling network are used in this preliminary
report.
RESULTS AND DISCUSSION
The following discussion is supplemented with several figures to illustrate
the basic findings. Variation in chemical composition is shown for several
sites by use of trilinear diagrams (for example, fig. 416). Within each
triangle and the center quadrangle, the percentage of the cations (calcium,
magnesium, sodium, and potassium) and the anions (bicarbonate, carbonate,
chloride, and sulfate) totals 100 thus, changes in the relative amounts
of these constituents are shown concisely. Variation in the levels of sulfate,
chloride, total nitrogen, total organic nitrogen, and turbidity2
are shown in time-series plots for selected sites (for example, fig. 417).
For each parameter, the same scale is used for all sites to facilitate
comparison of the relative magnitudes of change.
EFFECTS OF ASH ON RIVER-WATER QUALITY
As expected from the distribution of ash fall (fig. 414), streams to the south of the volcano, tributary to the Columbia River (fig. 415, stations 4 and 5), were unaffected. A small tributary to the Klickitat River (fig. 415, station 1) was also unaffected.
East of Mount St. Helens, the magnitude of the observed effects of ash
deposition was a function of the pattern of ash depths and the relative
volumes of water draining from heavily and slightly affected areas down
to the sampling sites. In general, the magnitude of the observed effects
ranged from pronounced in streams close to the volcano to slight in streams
in eastern Washington. In the following sections, the effects are described
in streams located progressively eastward from the volcano.
KLICKITAT RIVER
Samples from the Klickitat River near Glenwood (fig. 415, station 2) and near Pitt (fig. 415, station 3) showed that ash deposition had a pronounced effect on water composition. The changes at the Pitt station downstream were not as pronounced as those at the Glenwood station, near the headwaters, owing to dilution from tributary drainages that received very light ash fall. However, the length of the historic record of measurements at the Pitt station makes it the better location for discerning the ash-fall effects.
The shift in chemical composition of the water from May 18 until several days later is illustrated in a trilinear diagram (fig. 416). The sample taken on May 18 showed no effects of ash fall, but the sample collected the following day showed much higher percentages of chloride and sulfate, which suggests the addition of soluble chloride and sulfate compounds from the ash. By May 20, the anion composition had shifted back toward the normal preemption percentages, and subsequent samples confirmed that the composition had returned to the normal range. The cation composition did not vary to any measurable extent, probably because the percentage of soluble cations in the ash (table 96) was not substantially different from that of the native waters.
Changes in the anion concentrations in the Klickitat River at Pitt were extremely short lived, as illustrated by the time-series plots of sulfate and chloride (figs. 417 and 418). Samples collected just after May 18 showed increases in the concentration of these constituents. The increase in sulfate, although significant, did not exceed previous high concentrations. The chloride increase noted in the May 19 sample was notably greater than in preemption and other posteruption samples. By the end of June, the concentrations of both sulfate and chloride had returned to within the preemption ranges of values.
The most dramatic effects of the ash fall on water quality in the Klickitat are illustrated in the time-series plots of total nitrogen, total organic nitrogen, and turbidity (figs. 419 and 420). Samples collected just after May 18 contained levels of these three parameters well above the highest observed during the previous 2 yr. Because the nitrogen analyses were performed on unfiltered samples, the correspondence between the peaks in nitrogen and turbidity suggests that the nitrogen was closely associated with the particulate material. The peaks observed on June 17 are probably related to ash fall from the eruption of June 12.
The high nitrogen concentrations cannot be as easily correlated to the
ash as were the high sulfate and chloride concentrations because the ash
was not analyzed for organic nitrogen. However, the increase in nitrogen
corresponding to the time of ash deposition is not surprising. The major
blast on May 18 denuded the nearby environment, destroying the forest and
other vegetation. Pyrolysis (a major chemical change due to intense heat)
of the forest may have resulted in vaporization of much forest organic
material, which included a large quantity of nitrogen. The vaporized organic
material probably condensed, attached to the ash particles, and was transported
in the ash plume. For a short time, organic nitrogen conveyed by ash into
water would stay attached to the ash particles. After some time, dissolution,
mineralization, and nitrification of this nitrogen would result in increased
amounts of dissolved nitrate. The effects of ash fall on the nitrogen concentrations
and turbidity were short lived. By July, the variations had returned to
within the normal range.
AHTANUM CREEK
In streams farther east of the volcano, such as the North Fork and South Fork Ahtanum Creek (fig. 415, stations 6 and 7), variations in element content and turbidity were similar to those in the Klickitat River. Data from the North Fork Ahtanum Creek site near Tampico (fig. 415, station 6) show the similarity (fig. 421).
Higher percentages of sulfate and chloride were observed in the sample
taken May 18, rather than in the sample taken on May 19, when the change
was seen in the Klickitat River. The time difference is probably the result
of the proximity of Ahtanum Creek to the heaviest ash-fall area (fig. 414),
and a shorter travel time of water from within the watershed to the sampling
site. The degree of change in composition over normal values was also greater,
because the entire drainage received substantial ash.
YAKIMA RIVER
In streams and rivers still farther east of the volcano, the effects of ash fall on surface-water quality were much less apparent than in the Klickitat River and Ahtanum Creek. Besides the Columbia River, the Yakima River is the major river draining central Washington, and much of the land it drains received significant ash fall. Of the two stations on the Yakima having long historical water-quality records (fig. 415, stations 8 and 9), data from station 8 near Union Gap best illustrate the water-quality changes observed following the eruption.
At the Union Gap station, sulfate and chloride concentrations in samples varied sharply over a brief period following the major ash fall of May 18 (figs. 422, 423) and subsequent eruptive events, but the magnitudes of change were small relative to preemption variations. Indeed, it is difficult to decide whether the short-term fluctuations were due to the ash fall, or were merely artifacts of increased sampling frequency. The much lower effect on water quality in the Yakima than in the Klickitat River and Ahtanum Creek probably resulted from dilution of water in tributaries that received much ash by water from tributaries that received little or no ash.
In contrast to the undramatic changes in sulfate and chloride, the levels of total organic nitrogen, total nitrogen, and turbidity increased abruptly shortly after the May 18 ash fall to values well above the highest values observed during the previous year (figs. 424, 425). Also, the increase in total nitrogen amount was due largely to total organic nitrogen, which, in the preceding year's samples, had accounted for only about 50 percent of the total nitrogen concentrations.
As at the stations on the Klickitat River and Ahtanum Creek, the increases
in nitrogen and turbidity were short lived. By July, the variations were
within the range of those observed during the previous year.
SPOKANE RIVER
Still farther east, the effects of ash fall on water quality were much less apparent than those at the Yakima River near Union Gap. In eastern Washington, the Spokane River drains an area of northwestern Idaho and eastern Washington that received substantial ash fall; effects of ash deposition on the river were evaluated using data from one site on the Spokane River near Riverside State Park in Spokane (fig. 415, station 10) and Hangman Creek (fig. 415, station 11).
At both stations, the percentages of sulfate and chloride in samples
collected after ash fall were small and well below the highest values observed
during the previous 2 yr. Here again, an increase in turbidity was accompanied
by increases in total organic nitrogen and total nitrogen, with organic
nitrogen accounting for most of the increase. The increases occurred on
May 27, during and following a major storm.
EFFECTS OF VOLCANIC DEBRIS AND
MUDFLOWS ON RIVER-WATER QUALITY
The Toutle River headwater drainages were strongly altered by debris-avalanche deposits and pyroclastic-flow and mudflow deposits from the May 18 eruption. The downstream reaches of the Toutle River and the lower Cowlitz River below its confluence with the Toutle were devastated by massive mudflows and flooding. Water samples collected regularly since the eruption from the Toutle and Cowlitz Rivers show that these events have had a pronounced and persistent effect on the surface-water quality.
A water sample collected from the Toutle River near Castle Rock (fig. 415, station 12) on March 29, before the major eruption, contained calcium and bicarbonate as the predominant dissolved chemical constituents (fig. 426). The sample collected on May 20 showed higher percentages of sulfate and chloride. Samples collected at this station through July showed continuing high sulfate and chloride concentrations, indicating that these constituents continued to leach from the ash, debris, and mudflow deposits. The mudflows on May 18 prohibited collection of satisfactory water-quality samples. On May 19, the specific conductance in the Toutle River near Castle Rock was 1,200 umhos (micromhos), compared with normal values of about 90 umhos. On May 20, when the first samples could be collected, the specific conductance was 560 umhos. These values indicate dissolved-solids concentrations of about 6 to 12 times above normal for the Toutle River.
Following the major eruption, sampling frequency was increased in the Cowlitz River at Castle Rock (fig. 415, station 13) to determine the effects of inflow from the Toutle River drainage. Starting on May 30, samples were also routinely collected at Kelso (fig. 415, station 14) to monitor changes in quality between the two sites and to document the quality of water flowing into the Columbia River.
The compositional variations in the Cowlitz River at Castle Rock were similar to those observed in the Toutle River; however, the overall effect was reduced because of dilution by flow from the upper Cowlitz. The variability of the composition of water at the Castle Rock station was and is being determined by the character and magnitude of discharge from the devastated Toutle River system, by upstream dredging activities in the Cowlitz and Toutle, and by dilution effects of discharge from the upper Cowlitz.
In the Cowlitz River at Kelso (fig. 415, station 14), high concentrations of sulfate and chloride were measured in late May and early June, although maximum values for both parameters were below those in samples collected monthly during the previous 2 yr. However, the mudflow deposits had a dramatic effect on the levels of total organic nitrogen, total nitrogen, and turbidity (figs. 427, 428). Samples collected in late May and early June at Kelso showed turbidity values well above the previous maximum value observed at this station. Again, turbidity increases were accompanied by dramatic increases in the total nitrogen concentration, with most of the increases due to total organic nitrogen.
Figure 429 shows the major ion composition of water samples collected before and after the eruption at miscellaneous stations close to the volcano. In the South Fork Toutle River (fig. 429, station 15), a major increase occurred in chloride concentration, the increase in sulfate was negligible, and only minor cation changes were observed. Similar changes were observed in Pine Creek at its confluence with the Lewis River (fig. 429, station 16). Samples collected before the major eruption indicated that the water was characterized by sodium and bicarbonate, but posteruption changes resulting from mudflows in Pine Creek showed a higher percentage of chloride.
The Kalama River (fig. 429, station 17) was relatively unaffected by the May 18 eruption, but received ash deposits from the May 25 event. Only minor increases in sulfate and chloride were observed.
Posteruption water-chemistry data from Coldwater Creek (fig. 429, station
18) and the Toutle River near Kid Valley (fig. 429, station 19) showed
nearly equal percentages of the major dissolved-chemical constituents.
These stations had greater percentages of sulfate and chloride than all
other sites, indicating a greater chemical effect of the eruption on streams
whose headwaters are within the area of the blast and debris-avalanche
deposits.
OCCURRENCE OF TRACE METALS
Analysis of the ash (table 96) showed the presence of iron, manganese, and several other metals which, in the investigated streams and rivers, normally occur in very low concentrations. To determine the effects of ash fall on concentrations of these trace metals, both filtered and unfiltered samples were analyzed to measure the dissolved and total metal concentrations. The difference between the two values represents the concentration in the suspended sediments.
Concentrations of suspended sediments were high in most of the samples, and most of the trace metals determined were associated with these sediments. The dissolved-metal concentrations (table 97) were generally much lower than the suspended-sediment metal concentrations, and were only slightly higher than preemption levels in those streams affected solely by ash depostion. In the Toutle and Cowlitz Rivers, posteruption concentrations of dissolved iron, manganese, and aluminum increased and persisted, owing to continued leaching from the debris-avalanche and mudflow deposits.
Total concentrations of iron, manganese, and aluminum increased greatly over preemption levels (table 97). Total concentrations of these metals at the Toutle and Cowlitz stations increased 100 to 500 times over preemption values, whereas the concentrations increased only about 5 to 50 times at the ash-affected stations on the Klickitat River and the North Fork Ahtanum Creek. There was also a marked difference between the mudflow- and ash-affected stations in the duration of total metal-concentration increases. At the Klickitat and North Fork Ahtanum stations, the increases in total metal concentrations were short lived. In contrast, increases at the Toutle and Cowlitz River stations persisted at least through June, owing to the continuing influx of suspended sediment.
Concentrations of other trace metals of interest--arsenic, cadmium,
chromium, and cobalt--were not noticeably higher than levels before the
ash fall in either the filtered or unfiltered samples.
SUMMARY
The May 18 eruption of Mount St. Helens in southwestern Washington had a pronounced effect on river-water quality. The effects were greatest in the Toutle River system, which was devastated by the blast, a massive debris-avalanche deposit, and mudflows. The Cowlitz River, into which the Toutle drains, was also affected severely. Relatively small but still marked changes in water quality were observed in drainage basins affected by ash east of the volcano.
The major changes in water quality associated with the eruptive events included increases in the levels of sulfate, chloride, total nitrogen, total organic nitrogen, turbidity, iron, manganese, and aluminum. In the ash-affected basins east of Mount St. Helens, the observed water-quality changes were short lived and, in general, decreased in magnitude with distance from the volcano. Changes in water quality in the Toutle and Cowlitz Rivers persisted longer, owing to continuing influx and leaching of sediment in the devastated headwater areas.
Total concentrations of iron, manganese, and aluminum increased greatly over preemption levels to as much as 100 to 500 times in the Toutle and Cowlitz Rivers, compared to 5 to 50 times at the reported sites east of the volcano. At all sampled sites, the dissolved metal concentrations were generally much lower than the suspended metal concentrations.
It is especially interesting that high concentrations of total nitrogen
and total organic nitrogen were observed in posteruption water samples.
In all samples, the high nitrogen concentrations were associated with high
levels of turbidity, and the greater proportion of the nitrogen was organic.
The high concentrations of organic nitrogen probably resulted from pyrolysis
of forest material within the blast zone. The vaporized organic products
may have immediately become attached to volcanic-ash particles and then
were transported to the point of deposition.
START & IMAGES |
|