Effect of Historical Logging on Geomorphology, Hydrology, and Water Quality in the Little Fork River Watershed

The Little Fork River begins in lowlands near Lake Vermilion in northeast Minnesota. It drains a watershed of 1,843 square miles, flowing 160 miles before it enters the Rainy River on the U.S. – Canadian border. Land cover in the watershed is predominantly forest and wetland where the river flows through lacustrine and reworked till of the former glacial Lake Agassiz basin. The watershed is sparsely populated and remote where principal industries include forest products harvesting and tourism. Extensive stands of pine and pulp-wood were historically logged from the 1890's to 1937. Currently the Little Fork River is impaired for turbidity in the lower (most downstream) reach from the town of Littlefork to the Rainy River. Because several waterbodies in Minnesota have been listed as impaired without sufficient monitoring and assessment, this study was designed to further document the turbidity and sediment concentrations in the mainstem of the River from the Rainy River confluence to the headwaters; and to provide information needed to determine the scope and likely sources of the turbidity impairment for the upcoming Total Maximum Daily Load study. Our study included water quality sampling at seven representative reaches, a detailed analysis of the long term climatology and US Geological Survey streamflow datasets, an analysis of watershed landcover, and initial stream geomorphology surveys. Our results indicate that the turbidity impairment extends from the Rainy River confluence upstream for at least 142 river-miles. The primary cause of the excessive turbidity is suspended sediment likely resulting from erosion of the mainstem river banks- which, in turn, is due to increased streamflows. A detailed analysis of the US Geological Survey streamflow gage dataset indicates that 1.5 year 'bankfull' flows (i.e. the streamflow that defines the size and shape of the channel) have significantly changed over the 80 year period of record. Using a weight of evidence approach, we hypothesize that these changes in bankfull flow are due to the impacts of historical logging and the influence of local geology and vegetation. Trends in bankfull discharge in the Little Fork River are independent of the effects of annual precipitation; we infer the changes to be driven by watershed factors (i.e. land cover). Normalized bankfull flows in the adjacent and similarly sized Big Fork River watershed are significantly lower (p<0.001) than those on the Little Fork River. We hypothesize that the factors explaining this include: a greater percentage of peatlands in the Big Fork watershed (that were never historically logged) which dampen flood flows and reduce flashiness; a greater number of headwater lakes in the Big Fork providing more watershed storage; earlier and comparatively less historical logging in the Big Fork; and a greater proportion of pasture / open land in the Little Fork watershed. A regional curve analysis indicated that the Little Fork River and its tributaries have approximately twice the water yield when compared to other area streams. The Little Fork River is likely still recovering from the hydrological and geomorphic impacts associated with historical logging- by the years of increased runoff (water yield) following initial harvest, as well as the geomorphic impact of using the river to transport logs. The former is supported by numerous visual observations, an aerial over-flight, and initial geomorphology surveys; however significant study remains to substantiate this hypothesis. Several proposed next steps for the upcoming TMDL and long-term Little Fork / Big Fork Rivers paired watershed study are identified.