Sediment characteristics of northwestern Wisconsin’s Nemadji River, 1973–2016

In 2015–16, a comparison study of stream sediment collection techniques was done for a U.S. Geological Survey streamgage on the Nemadji River near South Superior, Wisconsin (U.S. Geological Survey station number 04024430) to provide an adjustment factor for comparing suspended-sediment rating curves for two historical periods 1973–86 and 2006–16. During 1973–1986, the U.S. Geological Survey used the equal-width-increment technique to collect suspended-sediment concentration data (EWI SSC). The Wisconsin Department of Natural Resources and Minnesota Pollution Control Agency collected grab samples for total suspended solids (grab TSS) concentration starting in 2006 and continuing beyond 2016. In addition to the comparison study of suspended-sediment concentrations, bedload and bed material samples were collected in 2015–16, and the modified Einstein procedure was run to further characterize total sediment loads. The 2015–16 study indicated that the EWI SSC and grab TSS concentrations were different, but not as much as expected, especially on the high end where grab TSS concentrations were sometimes higher than EWI SSC concentrations, possibly due to a combination of a high percentage of fines in suspension and higher concentrations in the center of the channel than the margins. The 2015–16 measured bedload made up a small percentage of total sediment load, and bedload and streambed particle sizes are 90 to 100 percent sand sized or smaller. The relative proportion of measured bedload to total load decreased with increased streamflow, and for streamflows greater than 1,800 cubic feet per second, the suspended load made up 98 percent of the total load. Calculated 2015–16 instantaneous total sediment loads from the modified Einstein procedure were up to 70 percent of the measured loads for flows less than 1,000 cubic feet per second and near or more than 100 percent for flows greater than 1,000 cubic feet per second. The sediment rating curve developed for the 2006–16 adjusted grab TSS data had a similar slope but a lower intercept than its 1973–86 EWI SSC counterpart, indicating that for a given streamflow, suspended-sediment concentrations were lower for 2006–16 compared to 1973–86. The negative offset equates to estimates of annual suspended-sediment loads in 2006–16 being on average 87 percent of the 1973–86 loads. Over the period 2009–16, annual suspended-sediment loads ranged from a low of about 21,000 tons per year in 2015 to a high of 167,000 tons per year in 2012 with a mean of 85,000 tons per year. However, reductions in suspended-sediment concentrations are likely obscured by large loads during years with flooding.