Note the guide bracket for the sediment sampler and the walkway which is raised to the winter position. April 9, 1952, water discharge about 450 cfs. The tallest tree is adjacent to the left edge of the section. July 14, 1953, water discharge about 350 cfs. The water-stage recorder for the gaging station is on the right bank behind the bridge. The ditch at the left connects with a pond and contributes no inflow. April 9, 1952, water discharge about 450 cfs. sampler. (See pl. 3B.) A portable bracket, which can be hooked over the measuring sill, guides the sampler so that, when the sampler is moved vertically along the bracket, the nozzle of the sampler traverses the entire depth of flow from the water surface down to the top of the measuring sill. As long as the measuring sill is free of sand deposits, all the sediment passing over the sill can be sampled regardless of the vertical distribution of sediment. Thus, concentrations determined from depth-integrated samples represent the total sediment concentration, which is used in computing the total sediment discharge. Two different types of samples were collected at section D. Daily samples consisted of depth-integrated samples collected at stations 10, 30, 50, and 70; cross-section samples consisted of depth-integrated samples collected at 15 verticals or more, spaced either at equal increments of width or according to water discharge. The daily samples were collected once or twice a day by a local observer, and daily samples and cross-section samples were collected periodically by engineers. Concentrations from the daily samples were used to define a continuous concentration curve from which a time-weighted mean concentration for each day was determined. However, because the time-weighted mean concentrations were sometimes unrepresentative of the cross section as a whole, they were corrected by using a determined relation between the cross-section samples and the daily samples. Forty-four ratios of the concentrations from daily samples to the concentrations from cross-section samples during part of the 1950 water year and all the 1951 and 1952 water years averaged 1.03 and ranged from 0.72 to 1.61. The relation between the concentrations of cross-section samples and daily samples fluctuates. The relation does not seem to be affected by any of the common flow parameters; but it is, of course, affected by variations in the lateral distribution of sediment in the cross section, which is in turn affected by dune movement into the flume. An analysis of data collected on June 27, 1951, (Benedict, Albertson, and Matejka, 1955) indicates the extent of the variations of daily sample concentrations with time. Concentrations from depth-integrated sediment samples collected at 5-minute intervals over a 4-hour period at the 4 daily sampling stations indicate that the concentration of any sample at any time might vary as much as ±83.5 percent from the true mean for the vertical. Also, the concentration determined from 1 bottle at each of the 4 stations might vary from the average concentration as much as ±25 percent, and a concentration determined from 2 bottles at each of the 4 stations might vary from the average concentration as much as 19 percent. Two bottles at each of the four verticals are normally collected to obtain daily sample concentrations. Records of sediment loads (total load) passing section D are published in a continuing series of water-supply papers entitled, Quality of Surface Waters of the United States. The published daily loads are computed from daily mean water discharges and from daily mean concentrations that are corrected to represent cross-section concentrations. Monthly and annual sediment loads passing section D during the period March 22, 1950, through September 30, 1952, are given in table 6. During most of the year, monthly loads are fairly constant, although they are usually highest in late winter and early spring. Data from measurements of water and suspended-sediment discharges indicate that Qs, the total sediment discharge, varies about MEASURED SUSPENDED-SEDIMENT DISCHARGE, IN TONS PER DAY FIGURE 8.-Sediment-transport curve for section D at the turbulence flume. 10,000 as the cube of Qw, the water discharge, (Qts=C1Qw3); however, the relationship is not well defined. (See fig. 8.) Size analyses of depth-integrated samples represent the size distribution of the total load of the stream and show that about 90 percent of the sediment is coarser than 0.062 millimeter and that about 35 percent of the total is in the 0.125- to 0.25-millimeter size range. Figure 5 shows the size distribution of the material passing the flume on 4 different days of medium water discharge. The percentage of suspended sediment in individual size ranges is fairly constant except possibly at relatively high and low concentrations (fig. 9). Generally at section D, the depth decreases and the velocity of flow increases from the left to the right bank. However, the measured suspended-sediment concentration follows a random pattern; the concentrations have a tendency to be highest in the center bay of the |