Abstract

In response to concerns about water-quality impairments that may affect habitat degradation in Agassiz National Wildlife Refuge in northwest Minnesota, the U.S. Geological Survey, in cooperation with the U.S. Fish and Wildlife Service collected streamflow data, discrete nutrient and suspended- sediment samples, and continuous water-quality data from 2008 to 2010. Constituent loads were estimated for nutrients and suspended sediment using sample data and streamflow data. In addition, a potential water-quality and streamflow monitoring program design was developed for Agassiz National Wildlife Refuge. Results from this study can be used by resource managers to address identified impairments and protect wildlife habitat and public water supply, and may contribute toward developing more effective water-management plans for Agassiz National Wildlife Refuge.

Streamflow was measured by the U.S. Geological Survey at four inflow and two outflow sites located on rivers and drainage ditches in and near Agassiz National Wildlife Refuge during open-water (no ice cover) periods in 2008, 2009, and 2010. Discrete samples were collected and analyzed for nutrients (total ammonia plus organic nitrogen, dissolved nitrate plus nitrite, dissolved ammonia, total nitrogen, dissolved orthophosphorus, and total phosphorus) and suspended-sediment concentration. Continuous water-quality measurements were collected for water temperature, specific conductance, dissolved-oxygen concentration, pH, and turbidity.

In 2006, the Thief River from Thief Lake to Agassiz Pool was listed as impaired for high ammonia concentrations. Results from this study indicate that concentrations at all sites did not exceed the 0.04-mg/L water-quality standard for un-ionized ammonia. Compared with the four inflow sites, the two outflow sites generally had significantly greater dissolved ammonia concentrations, significantly smaller nitrate plus nitrite concentrations, and no major differences in total ammonia plus organic nitrogen and total nitrogen. Small differences in suspended-sediment concentration were observed among inflow sites, but outflow sites had significantly greater suspended-sediment concentrations than inflow sites. At the primary outflow site, during the scheduled drawdown of Agassiz Pool from October 2009 into 2010, suspended-sediment concentrations were high compared to concentrations prior to the scheduled drawdown of Agassiz Pool. Overall, orthophosphorus and total phosphorus concentrations were significantly greater at inflow site A1 (located on Branch 1 of Ditch 11) than any other site. In 2010, although this site accounted for only 3 percent of the total streamflow from inflow sites, this same site accounted for 31, 27, and 13 percent of the inflow load for nitrate plus nitrite, orthophosphorus, and total phosphorus, respectively.

Among the sites, for most constituents, annual (open-water period) nutrient and sediment loads generally were greatest at the site with greatest volume of streamflow (the primary outflow site) and greatest in the year with the greatest amount of streamflow (2010). Large loads at the primary outflow site in 2010, particularly for sediment, likely resulted from the combination of greater flows in 2010 and scheduled drawdown of Agassiz Pool. Of the three inflow sites to Agassiz Pool, two of the sites accounted for at least 97 percent of the total annual sediment load from 2008 to 2010. Although loads were greater in 2010, in many cases the annual flow-weighted concentrations for nutrients and suspended-sediment were greatest in 2009, which may have been related to differences in the streamflow patterns between 2009 and 2010. For most sites and constituents, mean monthly nutrient and sediment loads were greatest in April, May, June, September, and October, which corresponded with months of greater streamflow volume. For the primary outflow site, the greatest sediment load occurred in October, which is likely related to high concentrations of suspended sediment at the start of scheduled drawdown of Agassiz Pool in October 2009 and large streamflow volume in October of 2010. For sites located downstream from Thief Lake and Agassiz Pool, the seasonal pattern of most mean monthly nutrient loads and mean monthly flow-weighted nutrient concentrations were affected by releases from these water bodies and the vegetative growing season. For inflow sites not located directly downstream from impoundments, much less variability in the flow-weighted concentrations of nitrate plus nitrite and orthophosphorus was observed. Continuous water-quality monitor data from 2010 indicated instances when water-quality standards for dissolved oxygen, pH, and turbidity were not met. For all sites, spikes in turbidity occurred related to rainfall, with as little as 2 percent of the values exceeding the 25 nephelometric turbidity units water-quality standard and at most 38 percent of the values exceeding the standard.

A recent (2011) radioisotope study indicates that Agassiz Pool has been experiencing a net gain of sediment (more inflow load than outflow load) in the last 68 years, but during the 3-year period of this study (2008 to 2010), a net loss of sediment from Agassiz Pool occurred. A net loss from 2008 to 2010 was likely related to a combination of several atypical water-management activities that occurred at the two outflow sites including: the first year of operation of the water control structure at the smaller outflow site in 2008; construction downstream from the primary outflow site in 2008 and 2009; and scheduled drawdown of Agassiz Pool in fall 2009 through 2010, which occurs only once every 10 years.

A future water-quality monitoring program for Agassiz National Wildlife Refuge could include data collection at 2 indicator sites (one inflow and one outflow site) with a total of 7 discrete samples and 7 streamflow measurements consisting of the following: 5 samples, along with a streamflow measurement, collected during the same week each month in April, May, June, July, and October combined with 2 supplementary samples and streamflow measurements during periods of storm runoff. In addition to the discrete samples, continuous water-quality monitors could be deployed at each site. Future water-quality monitoring in Agassiz National Wildlife Refuge would provide information that can be used to assess the changes in water quality with time, changes in management conditions, effects of upstream mitigation practices (for example, buffer strips, side-channel inlets) within the Thief River watershed, as well as other variables.

First posted July 27, 2012