Before reliable values of daily sediment discharge can be computed, continuous water-discharge records are computed from river stage, and samples that describe the changes in sediment concentration are collected. The instrumentation used to accomplish those tasks is described here.
When a datum is assigned to an elevation at a gaging station, the river stage relative to the datum is referred to as "gage height." To record gage height at a typical, stable stream, a stilling well usually is mounted on the river bank. An intake from the main flow is connected to the well, in which a mechanical float traces changes in river stage. Stilling wells were not used at Mount St. Helens, because they easily trap sediment through their intakes. High sediment concentrations would have created constant maintenance problems in stilling wells, so mercury-column stage manometers were installed exclusively in the study area.
Gage height was recorded continuously with chart recorders linked to stage manometers. Digitized gage heights, acquired at 15-minute intervals, were transmitted by satellite telemetry to receiving stations and computer storage on a current basis. Satellite transmittal of gage height was invaluable for anticipating storm flows and for providing redundancy in the collection of stage records in case of recording failures. Equipment malfunctions at the gaging stations also were detectable immediately from the satellite data. The detail provided by pen traces of river stage was used to evaluate the behavior of storm flow and short-term trends in sediment discharge.
Automatic pumping samplers (U.S. PS-69) were installed at all gaging stations at streams near Mount St. Helens where sediment discharge records were needed (fig. 8). A pumped sample contained about one liter of river water pumped from a fixed point above the streambed. Pumping from the stream was started by a timer at regular intervals ranging from daily to hourly. As river water flowed through a 0.6-in.-diameter hose during the 3-minute pumping cycle, a small amount was diverted to the sample container. The sampling frequency was increased during rising stage according to preset thresholds. For example, a 2-ft rise in gage height would trigger a change from a daily sampling frequency to an hourly rate. The time of sample collection was indicated with a printed mark on the chart record of gage height.
Bed-material samples were collected during sediment-discharge measurements whenever feasible. Bed material was sampled in wadable streams with a U.S. BMH-53, or a metal container of similar volume. Deep, swift streams were sampled from bridges or cableways with a U.S. BM-54 bed-material sampler. This cable-suspended sampler rotates a 10.7-in3 bucket into the streambed to a depth of 1.7 in. when tension on the suspension cable is released. This procedure often was ineffective during storm flows because fluid drag on the cable and sampler would resist attempts to release tension. A solenoid-activated mechanism later was developed for the BM-54 to close the bucket directly. Size distributions of bed material are summarized in the section "Changes in Sediment Sizes."
Experimental attempts at bedload sampling with Helley-Smith samplers were made in the first few years after the 1980 eruption. Extreme fluid drag on the samplers caused inadvertent dredging of the streambed during retrieval of the sampler. Tether lines to increase sampler stability and to avoid the dredging were first available in 1984. A bedload-sampling program with suitable equipment was instituted in 1985. After that time, bedload samplers and associated equipment were developed that provided reasonable transport estimates of coarse bedload (Childers, 1992). Size distributions of bedload samples are summarized in the section "Changes in Sediment Sizes."