WSP00684 - Laserfiche WebLink

Home Browse Search

collected only those par1icles having diameters <64 mm. We analyzed particle-size distributions using standard techniques (Kellerhals and Bray, 1971; Folk, 1974). Samples were dried, then a representative split was sieved using standard, brass sieves at I ljl intervals from 4 to -5ljl (0.064 to 32 mm). Par1icles retained on each sieve were weighed and the percent of the subsample in each ljl class was determined. Particle-size distribution of debris flows Par1icle-size distributions were determined by reconstructing the percentage of par1icles in each ljl class on the basis of sample weight or by occurrence in point counts. If particle diameters were measured in the field, the par1icle-size distribution determined using sieve analysis was adjusted for these par1icles after the par1icle weight was calculated. If point counts were made on the surface of the deposit from which the sample was collected, the two types of data were combined. Although point counts are made using surface exposure and dry-sieve analyses are based on weight percent of a sample, the order of magnitude of the resulting percentages is similar (Kellerhals and Bray, 1971). We assumed that point counts accurately measure par1icle diameters in excess of 64 mm; therefore, the distribution of par1icles >64 mm was determined using point counts, whereas the distribution of par1icles <64 mm was determined by combining point count and dry-sieve data. The percentage of par1icles <64 mm determined by point count was adjusted by the par1icle-size distribution of the collected sample. We determined par1icle-size distribution for 41 fresh, unaltered debris-flow deposits left by debris flows that occurred between 1965 and 1999 (fig. 14). The deposits are very poorly sorted. Pebbles are the most abundant par1icles at 41 percent by weight (table 16). Boulder content is highly variable, but typically accounts for about 14 percent of debris-flow deposits. On average, about 22 percent of all particles are smaller than gravel, and par1icles finer than sand account for only 4 percent of the distribution. The average sand content of debris flows is about 18.2 percent with a range of 2.4 to 47 percent. We found no significant statistical relation between sand content and other factors that might contribute to the high variability, such as drainage area, watershed lithology, or the volume of the debris flow. The strongest correlations obtained were between sand-and-finer par1icles and debris- flow volume (R2 = 0.20) and tributary drainage area (R2 = 0.20). For the sand fraction, alone the highest R2 value with any variable was 0.04. Table 16. Percentage of boulderl, cobblel, pebblea, land, and IlIt+clay In debrll flow In Grand Canyon. Size Cle.. Number of samples Boulders (>256 mm) Cobbles (64,256 mm) Pebbles (2-64 mm) Sand (0.063-2 mm) Silt + Clay (< 0.0063 mm) Deb,lo-flow depoatte (welght% ~ 1 SD) 41 13.9"'18.7 24.4'" 19.3 40.6", 20.6 18.2", 11.3 3.7", 3.1 Bulk density of debris flows The bulk density of debris flow deposits is required to convert deposit volume, as calculated using equation (10), to deposit mass. We calculate bulk density from par1icle-size information, by assuming that the volume occupied by par1icles larger than 2 mm has a density equivalent to rock, about 2.65 Mg/m3. We further assume that the aggregate debris finer than gravel has a density of 1.5 Mglm3, a density typical of a non-compacted sandy soil. The bulk density of a debris-flow deposit can therefore be estimated from: y = 2.65' r(W", '" < -I) + 1.50' I(W",'" > -I), (II) where y = the density of debris-flow deposits and W" = a weight percent fraction for a par1icle-size range. Using the values presented in table 16, we calculated an average value of y = 2.4 Mg/m3, which corresponds to a solids concentration of 85 percent by volume. This estimate is high compared to measured values for debris-flow deposits in other regions (Pierson, 1980; Major, and Voight, 1980; Gallino and Pierson, 1985; Iverson, 1997), which DEBRIS-FLOW SEDIMENT YIELD 31