Data were collected at 11 water-quality sampling sites in the upper Red River of the North (Red River) Basin from May 1997 through September 1999 to describe the water-quality characteristics of the upper Red River and to estimate constituent loads and flow-weighted average concentrations for major tributaries of the Red River upstream from the bridge crossing the Red River at Perley, Minn. Samples collected from the sites were analyzed for 5-day biochemical oxygen demand, bacteria, dissolved solids, nutrients, and suspended sediment.

Concentration data indicated the median concentrations for most constituents and sampling sites during the study period were less than existing North Dakota and Minnesota standards or guidelines. However, more than 25 percent of the samples for the Red River at Perley, Minn., site had fecal coliform concentrations that were greater than 200 colonies per 100 milliliters, indicating an abundance of pathogens in the upper Red River Basin. Although total nitrite plus nitrate concentrations generally increased in a downstream direction, the median concentrations for all sites were less than the North Dakota suggested guideline of 1.0 milligram per liter. Total and dissolved phosphorus concentrations also generally increased in a downstream direction, but, for those constituents, the median concentrations for most sampling sites exceeded the North Dakota suggested guideline of 0.1 milligram per liter.

For dissolved solids, nutrients, and suspended sediments, a relation between constituent concentration and streamflow was determined using the data collected during the study period. The relation was determined by a multiple regression model in which concentration was the dependent variable and streamflow was the primary explanatory variable. The regression model was used to compute unbiased estimates of annual loads for each constituent and for each of eight primary water-quality sampling sites and to compute the degree of uncertainty associated with each estimated annual load. The estimated annual loads for the eight primary sites then were used to estimate annual loads for five intervening reaches in the study area.  Results were used as a screening tool to identify which subbasins contributed a disproportionate amount of pollutants to the Red River. To compare the relative water quality of the different subbasins, an estimated flow-weighted average (FWA) concentration was computed from the estimated average annual load and the average annual streamflow for each subbasin.

The 5-day biochemical oxygen demands in the upper Red River Basin were fairly small, and medians ranged from 1 to 3 milligrams per liter. The largest estimated FWA concentration for dissolved solids (about 630 milligrams per liter) was for the Bois de Sioux River near Doran, Minn., site. The Otter Tail River above Breckenridge, Minn., site had the smallest estimated FWA concentration (about 240 milligrams per liter). The estimated FWA concentrations for dissolved solids for the main-stem sites ranged from about 300 to 500 milligrams per liter and generally increased in a downstream direction.

The estimated FWA concentrations for total nitrite plus nitrate for the main-stem sites increased from about 0.2 milligram per liter for the Red River below Wahpeton, N. Dak., site to about 0.9 milligram per liter for the Red River at Perley, Minn., site. Much of the increase probably resulted from flows from the tributary sites and intervening reaches, excluding the Otter Tail River above Breckenridge, Minn., site. However, uncertainty in the estimated concentrations prevented any reliable conclusions regarding which sites or reaches contributed most to the increase.

The estimated FWA concentrations for total ammonia for the main-stem sites increased from about 0.05 milligram per liter for the Red River above Fargo, N. Dak., site to about 0.15 milligram per liter for the Red River near Harwood, N. Dak., site. The increase resulted from a decrease in flows in the Red River above Fargo, N. Dak., to the Red River near Harwood, N. Dak., intervening reach and the large load for that reach.

The estimated FWA concentrations for total organic nitrogen for the main-stem sites were relatively constant and ranged from about 0.5 to 0.7 milligram per liter. The relatively constant concentrations were in sharp contrast to the total nitrite plus nitrate concentrations, which increased about fivefold between the Red River below Wahpeton, N. Dak., site and the Red River at Perley, Minn., site.

The Red River near Harwood, N. Dak., to the Red River at Perley, Minn., intervening reach had the largest estimated FWA concentration for total nitrogen (about 2.9 milligrams per liter), but the estimate was highly uncertain. The Otter Tail River above Breckenridge, Minn., site had the smallest concentration (about 0.6 milligram per liter). The estimated FWA concentrations for total nitrogen for the main-stem sites increased from about 0.9 milligram per liter for the Red River at Hickson, N. Dak., site to about 1.6 milligrams per liter for the Red River at Perley, Minn., site.

The Sheyenne River at Harwood, N. Dak., site had the largest estimated FWA concentration for total phosphorus (about 0.5 milligram per liter). The Otter Tail River above Breckenridge, Minn., site had the smallest concentration (about 0.1 milligram per liter). The estimated FWA concentrations for total phosphorus for the main-stem sites increased from about 0.15 milligram per liter for the Red River below Wahpeton, N. Dak., site to about 0.35 milligram per liter for the Red River at Perley, Minn., site.

The estimated FWA concentrations for suspended sediment for the main-stem sites increased from about 50 milligrams per liter for the Red River below Wahpeton, N. Dak., site to about 300 milligrams per liter for the Red River at Perley, Minn., site. Much of the increase occurred as a result of the large yield of suspended sediment from the Red River below Wahpeton, N. Dak., to the Red River at Hickson, N. Dak., intervening reach.