Excerpts from Dresser, A. T., C. Cook, and M. Smith. 2001. Long Term Trend Monitoring Program for the South Fork Trinity River watershed. U.S.D.A. Forest Service, Six Rivers National Forest. Eureka, CA. 89 pp.
"Pebble counts are the cheapest, easiest way to monitor gross changes in size distribution of streambed sediment. The protocol for pebble counts is well established and has been used extensively since Wolman introduced the technique in 1954. Substrate monitoring is required for the TMDL process, and pebble counts are the best way to meet that goal.
Generally, pebble count data are displayed as percent finer than a given size. Particle sizes are assigned a percentage based on their rank. For example, D35 is the size in millimeters at which 35 percent of the sampled particles are smaller. Changes in the size of a given percentile can show if the streambed is getting coarser or finer over time. It also can be used to monitor the amount of sediment finer than a given size.
Pebble counts are inexpensive but they cannot completely characterize the size
distribution of streambed sediment. The principal shortcoming of pebble counts
is that they tend to under-represent the finer particle sizes. This is particularly
true for sizes less than about 5 mm, which is in the range of fine sediment
than can affect fish survival. Secondly, pebble counts are not performed within
spawning redds because the size distribution of the streambed is changed by
the spawning fish. Therefore, sampling outside of redds may not reflect the
actual condition of spawning gravels in the study reaches. Furthermore, size
distribution of the streambed has not been well correlated with fish abundance
or survival. Finally, pebble counts are somewhat imprecise and the data resolution
may be insufficient to document subtle changes.
Field Methods
Pebble counts were performed underneath the cross section tape. Distance across the active channel (usually between pin #2 and pin #3) was divided by 100 to determine the sampling interval. In a few cases, the sampling interval was deemed too small (i.e., it was likely the same particle could be sampled twice). In these cases the active channel width was divided by 50 to determine the sampling interval and a second transect located a few feet downstream. Pebbles were selected randomly along the tape transect. The intermediate diameter of each pebble was recorded in millimeters.
Pebble Count Interpretation
The repeated sampling of streambed gravels should illustrate trends in sediment size and whether the channel is impacted by fine sediments. If the South Fork Trinity is recovering from sedimentation associated with the 1964 flood, we would expect an increase in the size of the D50, and a decrease in the proportion of sediment finer than 2 mm.
Changes in the pebble count were highly variable with no consistent trend.
About half the cross sections got coarser and half got finer. The most consistent
trend was the reduction in the portion of bed material finer than 2 mm. Reduction
in the proportion of fines was observed on 15 of 18 cross sections (83%)."