Northern Prairie Wildlife Research Center
In the summer of 1912, R.T. Young, a zoologist employed by the North Dakota Biological Station, traveled by boat from Fargo to Pembina and collected fishes and other biota at 29 sites on the Red River and in the mouths of tributary streams in North Dakota and Minnesota (Brannon 1912). Due to many snags and rubbish in the river, most sampling was conducted with a 100 ft gill net and small 15 ft seine. Sampling success was poor with the seine, and only six species were reported, with the maximum number of fish taken in the gill net on any one night as 8 or 9. Locals blamed the construction of the St. Andrews Locks at Winnipeg for the scarcity of fish. Young emphasized the "shocking disregard of all sanitary decency by some of the inhabitants of the riverbanks," who evidently dumped dead cattle into the streams and left them there to rot. Detailed habitat information was not provided for sites sampled by Young (Brannon 1912).
A cooperative biological survey by the University of North Dakota Biological Station and the UMMZ began in 1922 under the direction of M.C. Thompson. Fishes were collected in Devils Lake and from lakes in the Turtle Mountains and from the Red and James Rivers, with T.L. Hankinson in charge of fieldwork (Hankinson 1928). In June 1926, several collections were made by C.L. Hubbs and L.P. Schultz during an expedition they made to the Pacific Coast (Hankinson 1928, UMMZ 1994). A summary of these collections along with several made by R.T. Young (who visited about 30 localities in North Dakota) in 1921 was provided by Hankinson (1928). Overall, 33 fish species were collected by Young, Hubbs, and Schultz, and/or Hankinson from 11 sites on the Red, Pembina, Park, Maple, and Sheyenne Rivers (UMMZ 1994). At the time, the Red River was described as "sluggish and turbid" with a "margin of soft clay and firm bottom out from shore." The Sheyenne River was found to have excellent fish habitat and was described as "a beautifully winding river...with a succession of pools and riffles" (Hankinson 1928).
At the request of the Minnesota Department of Conservation, T.A. Olson, biologist, Minnesota Department of Health, surveyed fishes at 18 sites on the Buffalo, Wild Rice, Sandhill, Red Lake, Clearwater, and Snake Rivers (Olson 1932). In all, 33 species were collected by Olson (1932). This survey was the first ever completed on the streams of the eastern Red River basin. Most mentions of habitat at sites visited by Olson (1932) regarded stream width, depth, turbidity, or flow. The Marsh, Middle, and Tamarac Rivers were not sampled because they were dry and/or completely grown over with vegetation.
A single site in the Red River basin was sampled in 1949 by R.C. Gibbs and D. Allen during a survey of fishes of South Dakota (Bailey and Allum 1962). Eleven species were collected from Lake Traverse, at the headwaters of the Bois de Sioux River, in Roberts County, South Dakota.
Before the closing of the Baldhill Dam on the Sheyenne River north of Valley City in 1950, a study was conducted by H.W. Wilson, UMN, under the guidance of S. Eddy (Wilson 1950). The short (30 mi) reach of the river which became Lake Ashtabula was intensively sampled for fishes at 15 sites, while the remainder of the upper Sheyenne River northward and eastward to Coal Mine Lake was sampled at six sites. Wilson (1950) reported the presence of 21 species in the upper Sheyenne River, five which had not previously been found.
The last survey of fishes during the early period of sampling in the Red River basin was completed by personnel of the BMNH, UMN in 1955. The primary investigator was R. Tasker, but he was assisted by T. Collins, professor emeritus of biology, Moorhead State University, who at the time was a student at the UMN. During the investigation, 39 species were collected at 52 sites on the Mustinka, Otter Tail, Pelican, Buffalo, Wild Rice, Sandhill, Red Lake, Clearwater, and Two Rivers and their tributaries (BMNH 1994). Recorded were several species never before taken in the Red River basin. This study of stream fishes was the most comprehensive research ever completed on tributaries to the Red River in Minnesota. Notes from this survey can be found in Eddy et al. (1972).
Several streams were surveyed during the 1960s by members of the Biology Department, University of North Dakota (UND). During the spring and fall of 1962, the fishes of the Forest River were sampled at nine sites and related to the Pleistocene geology of the region (Feldman 1963). Of the 23 species reported, six species preferred waters of the glacial Lake Agassiz beach ridge area, where sandy bottoms and clear water rapids were common, and nine species preferred the muddy bottom conditions found in the Lake Agassiz plain. In 1964, 11 tributaries to the Red River in North Dakota were sampled for fishes by Copes (1965). During the investigation, 60 species were reported from 90 sampling stations. A paper regarding only the Sheyenne River was published from this work by Tubb et al. (1965), and the entire study was summarized by Copes and Tubb (1966).
In 1967, a stream survey was published by the Minnesota Department of Conservation, now known as the Department of Natural Resources (MDNR). The entire Buffalo River watershed, lakes, water quality, wildlife, and fishes were described by Reedstrom (1967). A total of 28 fish species were found in the lakes and streams of the watershed. A similar study was conducted on the Pelican River in 1968-1969, where 42 fish species were collected (Reedstrom and Carlson 1969).
Before the proposed Garrison Diversion project in North Dakota, several streams throughout the state were sampled to assess the impacts of a connection between the Missouri River drainage in the west and the Red River drainage in the east, as interbasin transfer of biota would then be a possibility. Another UND student, G.W. Russel, surveyed the fishes of the James, Wild Rice, Sheyenne, and Souris Rivers and the Devils Lake basin. Thirty-five species were collected from the Sheyenne River and 12 species from the Wild Rice River during the summer of 1974 (Russel 1975). Detailed distribution maps for several species collected in these streams during this survey can be found in Owen et al. (1981). These results, as well as data from several other streams surveyed in North Dakota in 1974 by personnel of the North Dakota Game and Fish Department (NDGF), were provided by Duerre (1975).
A major survey of several tributaries to the Red River in Minnesota was conducted by personnel of the BMNH primarily from 1974-1976, although a few streams were sampled during the 1960s (BMNH 1994). A total of 241 sites were visited on the Rabbit, Mustinka, Pelican, Otter Tail, Buffalo, Wild Rice, Sandhill, Clearwater, Red Lake, Middle, Snake, Tamarac, Two, and Roseau Rivers or their tributaries. The collections included 70 different fish species, several that had not previously been reported from the basin.
In the late 1970s, several stream surveys were conducted by the Minnesota Department of Natural Resources, Ecological Services Section (MDNR ECO). A stream "reconnaissance" of the Roseau River was conducted from 1975-1977; and the physical characteristics, water quality, benthos, vegetation, wildlife, and fisheries were recorded (Enblom 1982). A total of 22 fish species were collected during the survey; and information was provided regarding species diversity, population size, and the relationships of populations to streamflow. Naplin et al. (1977) sampled fishes from 13 stations on the Wild Rice River near the proposed Twin Valley reservoir in 1976. Thirty-one fish species were reported along with detailed information regarding habitat, population size, and length-weight relationships for game-fish species. A stream survey of the Red Lake River was conducted from 1976-1977, and eight study "sectors" were electrofished, providing information on 38 fish species (Renard et al. 1983). The distribution of fishes was primarily influenced by stream gradient and substrate and, to a lesser extent, water clarity and depth. The report also provides data regarding water quality and wildlife characteristics of the drainage.
Collections of fishes were made in the Sheyenne River downstream from the Baldhill Dam by Peterka (1978). Several sites occurred in spring fed streams in the Sheyenne Delta which provided refuges for several species requiring clean-water conditions. A total of 53 species were reported from collections made in 1964-1977, with six of them being known introductions.
In 1985, the North Dakota Natural Heritage Inventory and the NDGF sampled fishes from 15 sites in the Pembina River watershed (Kreil and Ryckman 1987). A total of 18 fish species were collected for inventory only. No habitat information was provided for the sites. During 1987, the Elm, Rush, Maple, Sheyenne, and Wild Rice Rivers in North Dakota were sampled for fishes by NDGF personnel (Duerre 1988). Species richness of these tributaries ranged from a low of 10 for the Rush River to a high of 43 for the Sheyenne River. The high richness of the Sheyenne River was thought to be due to the environmental quality of several spring creeks and their role as refugia during catastrophic events. Peterka (1992) conducted a survey of fishes in six streams which are tributary to the Red River in North Dakota. Species richness ranged from a low of 13 for the Pembina River downstream from Walhalla to a high of 18 for the Park River. Field notes provided for each site included stream width, depth, flow, clarity, substrate, and vegetation.
A survey of fishes was conducted in the Tamarac National Wildlife Refuge, which includes portions of the Buffalo and Otter Tail Rivers, by Schmidt (1993). Thirty-two species were reported from sampling at 29 stations in the summer of 1992 (Schmidt 1993). No habitat information was provided in this report.
The most recent investigations of stream fishes in the Red River basin have occurred during the summers of 1993 and 1994 as a part of two major studies. Several sites throughout the basin have been sampled for fishes using electrofishing gear by the MDNR, Minnesota Pollution Control Agency (MPCA), North Dakota Department of Health (NDDH), U.S. Environmental Protection Agency (EPA), and the U.S. Geological Survey (USGS). These studies are a part of the USGS National Water Quality Assessment program (Stoner et al. 1993) and the development of an index of biotic integrity for fishes in the basin (Goldstein et al. 1994). A total of 67 species were reported from 113 sites during a two-year sampling period. Additional sampling by these agencies and final reports of these studies are expected soon. In a related study, Feigum (1995) determined the biological integrity of disturbed and undisturbed sites for fishes in the Buffalo River.