This section summarizes the sediment-transport data collected at gaging stations near Mount St. Helens that have been made available for general use. The range of sediment concentration is examined, from low-flow conditions to lahars. Basic data records for gaging stations are categorized by drainage basin and described. Summaries of changes in sediment discharge and concentration are presented. Long-term trends of sediment concentration in six streams are compared by dominant sediment source. Changes in sediment size in different streams are illustrated with size-distribution data from suspended-sediment, bed-material, and bedload samples.
Range of Sediment Concentration
Sediment samples were collected at miscellaneous sites in the study area during the 2 months immediately preceding the May 18, 1980, eruption (data in Dinehart and others, 1981). Sediment data were seldom collected at streams in the Toutle River basin before 1980. The Toutle River at Highway 99 had been a water-quality sampling site, although suspended sediment was not an analyzed constituent. Sediment samples were collected on May 18-19, 1980, from the Toutle River lahars. Daily stream- and sediment- discharge measurements continued through the summer of 1980, during which suspended-sediment concentrations ranged from 3,000 to 10,000 mg/L. Sediment concentrations greater than 300,000 mg/L were measured during storm flow on November 21, 1980, in the North Fork Toutle River and on February 19, 1981, in the Toutle River. Sediment concentrations greater than 100,000 mg/L during storm flows were measured for several years at gaging stations.
High sediment concentrations unrelated to storm flow were measured during unique, sediment-laden flows. On August 27, 1980, a breached pond on the debris-avalanche deposit produced sediment concentrations greater than 500,000 mg/L in the North Fork Toutle River. On March 19-20, 1982, sediment concentrations around 1 million mg/L were measured at three gaging stations along the Toutle River course of a lahar generated by a minor eruption onto snowpack. On May 14, 1984, a small lahar flowed from the crater of Mount St. Helens and entered the North Fork Toutle River where sediment concentrations at Kid Valley exceeded 80,000 mg/L. Such flows were infrequent, and most high sediment concentrations in the rivers were the result of storm flows. During the study period, sediment concentrations (excluding lahars) measured at gaging stations ranged over six orders of magnitude (figs. 32, 33, 36, 38, 39, 41, and 43-46).
Basic Data Records at Gaging Stations
Major drainage basins in the Mount St. Helens area were each monitored by one or more gaging stations during the study period. Daily values of sediment discharge are plotted for 10 gaging stations (figs. 32, 33, 36, 38, 39, 41, and 43-46); most figures include a parallel plot of cumulative sediment discharge. In the plots of daily values by water year, the solid vertical lines represent October 1 of the previous calendar year, and the vertical tics are at April 1 of the water year.
On the same figures, concentrations of instantaneous sediment samples are plotted by time above records of daily mean stream discharge. Concentrations of depth-integrated samples (cross-sectional and box samples) are plotted together by time to show the frequency of sampling and the range of sediment concentration. The basic data records used for sediment-discharge computations at various gaging stations are discussed below for each drainage basin.
Depletion and Dormancy of Sediment Sources
Sediment discharge in the affected drainage basins can diminish by either depletion or dormancy of sediment sources. "Depletion" indicates that a sediment source has been removed by transport, and "dormancy" indicates that a source has not been transported, but is not available for transport by the dominant range of stream discharges. Changes in sediment discharge and concentration reflect depletion or dormancy of sediment sources, and the length of the 11-year study period may not be adequate to reveal the distinction.
The dominant sediment sources created by the 1980 eruption were readily erodible and were supplemented by erosion of existing bank deposits. Stream turbidity and sediment concentration were reduced as two sediment sources were rapidly eroded, those being (1) the lahar deposits in river channels that provided material for fluvial transport through mass wasting, and (2) the volcanic ash deposits overlying large areas of the Toutle and the upper Lewis River basins. Depletion of lahar deposits in channels was documented periodically by cross-sectional surveys. The surveys established that the largest changes in sediment storage occurred during storm flows in water year 1981 (Martinson and Meyer, 1987). The annual cycle of streamflow eroded lahar deposits from channels in the Toutle and the upper Lewis River basins.
Only a small proportion of another sediment source, the debris-avalanche deposit of the North Fork Toutle River, had been eroded during the study period. The percentage by volume of various sediment sizes in the debris-avalanche deposit is shown in figure 47 (data from U.S. Army Corps of Engineers, 1981). At least 85 percent of the material is medium gravel or finer, most of which is readily transportable by the North Fork Toutle River. Planners estimated that one-third of the debris-avalanche deposit, or about 1 billion yd3, would be eroded by 2001 (Cowlitz County, 1983). Based on annual sediment discharge from the upper North Fork Toutle River valley and deposition figures for the SRS, the debris-avalanche deposit had less than 10 percent of its volume eroded by 1990 (J.E. Costa, U.S. Geological Survey, written commun., 1992).
Channel widening and armoring reduced the access of moderate river flows to easily erodible sediments. Sediment that is not removed with ease becomes dormant or inactive and is transported only at successive extreme flows during the annual cycle. Therefore, much of the sediment source in the upper Toutle River was not depleted, but was made dormant through evolution of the drainage systems (Meyer and Martinson, 1989). Channel widening and coarsening contributed to the isolation of sediment from access by most streamflows. This coarsening process is usually described as "armoring." The mean diameter of sediment on the streambed surfaces became increasingly coarse, producing a decreased mobility of sediment at lower ranges of flow.
Changes in Sediment Discharge and Concentration
A purpose of this report is to identify any broad, time-related changes in sediment-transport quantities measured at gaging stations near Mount St. Helens. "Change" can have two meanings for sediment transport by streams near Mount St. Helens. On May 18, 1980, the mountainous, bouldery gravel-bed channels of the forested Toutle and upper Lewis River basins were changed suddenly to muddy, braided, sand- and gravel-bed streams flowing through devastated valleys. Concentrations of suspended-sediment samples, collected in March, April, and early May 1980 at the Toutle River at Highway 99 and Pine Creek at mouth, ranged from 4 to 30 mg/L. After the 1980 eruption, concentrations were greater consistently by one to five orders of magnitude. The transformation of streams by the 1980 eruption was dramatic and significant. The change discussed here, though, is the trend of the affected streams to assume pre-eruption conditions during the following 11 years.
Rivers in the Cascades Range are usually clear at low flow and become turbid during storm flow, as the Toutle River did before the 1980 eruption ( fig. 48 ). If water quality in the Toutle and the upper Lewis River streams returns to pre-eruption conditions, streamflows should exhibit this pattern. During the first few years following the eruption, turbid conditions were common in the Toutle River throughout the range of river flow (fig. 48). Even at suspended-sediment concentrations as low as 100 mg/L, the prevalence of fine sediment gave streams an opaque brown or gray color. By 1990, most streams near Mount St. Helens were clear at low flow, but many of the same streams were clear briefly in the summer of 1981. Therefore, more specific criteria than "turbid" and "clear" are needed to define long-term changes in sediment transport. Possible correlations with time are presented here for peak sediment discharge, mean discharge and concentration, and sand concentration sampled at similar discharges.
Changes in sediment transport are examined in several ways. The annual sediment discharges show an obvious decrease during the 11-year period. The changing relation between daily mean discharge and daily mean concentration for gaging stations is illustrated with logarithmic regressions on scatter plots. Then, concentration values (derived from regression equations) are plotted by year for discharge exceedance values of 1 and 50 percent. Finally, sampled concentrations are separated into ranges of stream discharge and are tested for correlation with time.
Comparisons of Similarly Affected Basins
Distinctive long-term trends of sediment concentration can be seen by comparing drainage basins having similar types of volcanic deposits. Six gaging stations, each having about 10 years of sediment data, are grouped by dominant sedimentation effects. Sediment concentrations of cross-section and single- vertical samples are shown again to compare the similar trends.
Sizes of bed material, suspended sediment, and bedload at streams near Mount St. Helens were measured periodically. Trends in particle-size distributions during the study period are discussed in this section. A general coarsening of bed material and bedload can be discerned from the size data, although the suspended-sediment size data did not show a simple relation with time.
