USGS - science for a changing world

Northern Prairie Wildlife Research Center

  Home About NPWRC Our Science Staff Employment Contacts Common Questions About the Site

Effects of a Legal Drain Clean-Out on Wetlands and
Waterbirds: A Recent Case History

Results and Discussion

Water Regimes at BBM and Kraft Slough

Prior to Drain Clean-Out.  In the 1960-94 period, precipitation was highest in 1962, 1969, and 1975 (NOAA 1892-1994), which paralleled periods of maximum water depths in the sloughs (Figure 2). The lowest water levels came after several years of below-average precipitation in 1974 and 1984 (Figure 2, Figure 3a). Over the 20th century, the driest years in Sargent County were in 1936, 1952, and 1976 (NOAA 1892-1994). Information is lacking for 1936, but aerial photographs in 1937 showed BBMK dry (U.S. Department of Agriculture photos). The area of wetland vegetation in BBM in 1937 was approximately the same as in 1984 (U.S. Department of Agriculture photos). In 1952, aerial photographs showed BBM holding water (U.S. Department of Agriculture unpubl. photos), and in 1976 BBM remained filled despite severe drought because of high precipitation and storage during 1975 (Figure 2, G. Krapu, unpubl. obs.). During the period from 1960-94, droughts occurred principally during the early to mid-1970s and early and late 1980s (Figure 2). The sustained capacity of BBM to retain water in years of below average precipitation continued through 1984, the year the clean-out began (Figure 2). No records were found of drawdowns in Bruns Slough from 1940 to 1984, one record (1959) exists for Big Slough (H. F. Duebbert, unpubl. notes), two records (1947, 1959) for Meszaros Slough (USA v. SCWRB, Civil No. A3-88-125, D.N.D. Transcript of Proceedings of Nov. 9, 1994, p.163-168, testimony of J. Hayen; H. F. Duebbert, unpubl. notes), and two records (1959, 1977) for Kraft Slough (H. F. Duebbert, unpubl. notes; U.S. Water and Power Resource Service 1980). Information on water depths in Kraft Slough and BBM, where available for comparable periods, suggest that maximum water depth in Kraft Slough usually was lower than in Bruns Slough prior to the clean-out. In October 1974, Kraft Slough had 31 cm of water (Krapu and Duebbert 1974), whereas Bruns Slough held about 46 cm (G. Krapu, unpubl. notes). In 1979, water depth was 1.0 m in the center of Kraft Slough on 23 August, 1.5 m in Bruns Slough on 24 July, and "chin deep", presumably about 1.5 m, in Big Slough on 27 June 1979 (U.S. Bureau of Reclamation unpubl. data).

Figure 2
Figure 2.  Average annual precipitation (in cm with calibration on left axis) based on measurements taken at gaging stations located at Forman, Oakes, Verona, and Lisbon, North Dakota from 1960-94 is compared to water depth (cm) in Bruns Slough (calibration on right axis). The long-term average of 50 cm of precipitation per year is indicated by background darker shading. Water-depth measurements for Bruns Slough (diamonds) were taken during summer and fall. An arrow identifies 1984 when the Drain No. 11 clean-out began.

Post Drain Clean-Out.  Although 1986 was the wettest year since 1975 (Figure 2) and despite 22.4 cm of precipitation falling during April (NOAA 1892-1994), only mudflats and puddles remained in Bruns Slough in April and June. Similar conditions were reported by other biologists (S. Zschomler, U.S. Fish and Wildlife Service, unpubl. memo and photos). Mudflat conditions persisted throughout the summer of 1986 despite continuing high rainfall, and Big Slough was dry in September. In 1987, Bruns (Figure 2) and Meszaros sloughs were dry from the start of spring, and Big Slough was nearly drained. BBM remained without water throughout the fall. In April 1988, BBM were dry, and by late spring, a major part of Bruns Slough had been tilled and planted to sunflowers and corn. In early April 1989, major parts of these wetlands were inundated temporarily by runoff from snowmelt to a maximum depth of about 60 cm. Waters began receding rapidly in mid-April, and by mid-June, BBM were largely drained. A rain in early July added a few centimeters of water in Bruns and Big sloughs, but by August, the sloughs were dry. In 1990, BBM were dry in spring and Bruns and Meszaros sloughs remained without water throughout the rest of the year (Figure 2). In Big Slough, about 15 cm of water accumulated in the center in mid-July from backup behind a culvert on Drain No. 11 that had been filled with sandbags by an unknown party. By fall, Big Slough also was dry. In contrast, Kraft Slough held water continuously throughout the 1986-90 period and had depths of 45 and 40 cm on 10 July and 1 November 1990. The inability of BBM in 1986 to store the high runoff as in previous years left the wetlands without any carry-over in spring 1987 and, with below average precipitation from 1987-90 (Figure 2), the basins remained dry during most of that period.

Runoff from above-average precipitation in 1993 (Figure 2) filled wetlands across southeastern North Dakota. High snowfall during the 1993-94 winter, followed by another wet summer in 1994 (NOAA 1892-1994), created the wettest conditions since 1975. On 28 October 1994, maximum water depth in Kraft Slough was 122 cm and 38 and 46 cm in Bruns and Big sloughs. Surface water in Bruns and Big sloughs was from water backing up behind a beaver dam that bad been built across Drain No. 11 downstream from Big Slough (D. M. Frick, pers. comm.). The failure of water levels in BBM to rebound with the return of a major wet cycle (Figure 2) confirmed that the cleaned drain operates efficiently to remove water from BBM except to the extent that major obstructions block flows.

An estimated 32,760 ha of basin wetland habitat existed in Sargent County in May 1979, of which 4,263, 12,943, 10,758, and 4,796 ha were in temporary, seasonal, semipermanent, and lake basins, respectively. The clean-out of Drain No. 11 changed the water regime of BBM from permanently-flooded to temporarily-flooded causing changes in functions and values of 1762 ha of wetland habitat. When impacts to 133 smaller basins that are bisected by or are contiguous to the drain are included, the Drain No. 11 clean-out severely impacted about 2093 ha (U.S. Fish and Wildlife Service 1987) or about 6% of basin wetland area and about 37% of lake area in the county.

Changes in Vegetative Cover to Open Area

Emergent cover to open water ratios on BBM during summer 1979 were 5.6:1, 2.2:1, and 3.9 to 1, respectively (U.S. Power and Water Resource Service 1980). In 1990, there were no openings in Bruns Slough, and the ratios of emergent cover to open water had increased to 1259.8:1 and 39.2:1 in Big and Meszaros sloughs. In comparison, ratios remained constant at 4.3:1 from 1979 to 1990 in Kraft Slough. The loss of open area at BBM resulted after frequent drawdowns during 1986-90 stimulated the germination and growth of dense stands of emergent aquatic plants (Figure 3b). As a result, 96% of the open area of BBM reverted to dense stands of emergent perennials by 1990, while remaining unchanged at Kraft Slough (Table 1). An open water to emergent vegetation ratio of about 1:1 is considered optimal for attracting high densities of waterfowl and other waterbirds (Weller and Spatcher 1965, Kaminiski and Prince 1984).

Bruns Slough before clean-out (1974)   Bruns Slough after clean-out (1990)
Figure 3.  (Left) Aerial view of Bruns Slough before the clean-out of Drain No. 11 in June 1974 and (Right) after the clean-out in December 1990. The 1974 photo shows a mosaic of water and emergent cover in 1974, reflecting productive habitat conditions for waterbirds despite effects of drought brought on by several preceding years of below-average precipitation (see Figure 2). Bruns Slough was dry and most was under cultivation in December 1990.

Table 1.  Percent change in open area of Bruns, Big, Meszaros, and Kraft sloughs in Sargent County, North Dakota from measurements taken based on conditions existing in July 1979 and 1990 before and after the clean-out of Drain No. 11.
Site Wetland Basin Area
Open Area (ha) Change (%)
1979b 1990c
Bruns Slough 332.7 50.1 0.0 -100.0
Big Slough 629.9 198.6 0.5 -99.0
Meszaros Slough 466.8 94.4 11.9 -87.0
Kraft Slough 339.5 78.1 78.3 +0.3
a Basin area in July 1979 (U.S. Water and Power Resources Service 1980).
b Standing water in openings.
c Moist soils in openings.

Waterbird Use of BBM and Kraft Slough

Prior to Drain Clean-Out.  BBM were key diving duck and colonial-nesting waterbird breeding areas prior to 1984. The complex of BBM together with Kraft Slough probably supported the highest concentration of breeding colonial-nesting waterbirds and diving ducks in southeastern North Dakota. Differences in waterbird species diversity among BBMK were small before the clean-out, with the most species sighted at Kraft Slough; the number of species with eggs or dependent young was highest at Big Slough (Table 2). Among diving ducks, redheads were most abundant, followed by ruddy ducks (Oxyura jamaicensis Gmelin), and canvasback (Aythya valisineria Wilson). The number of canvasbacks and ruddy ducks were highest in Bruns and Meszaros sloughs, respectively, whereas redheads were most abundant in Big Slough (Table 2). The most common dabbling ducks in BBMK were blue-winged teal (Anas discors Linnaeus), northern shoveler (Anas clypeata Linnaeus), and mallard (Table 2); 13 active mallard nests were encountered at Big Slough during 1979 waterbird censuses. Giant Canada geese (Branta canadensis maxima Delacour) with broods were seen on Bruns Slough during the survey and had nested on all four wetlands in past years.

Table 2.  Characteristics of waterbird use of Bruns, Big, Meszaros, and Kraft sloughs in Sargent County, North Dakota based on surveys conducted during 27 May-10 June 1979.
Site No. Species of
Mean Density Rank
of Waterbirds
No. Species With
No. Diving Ducks No. Dabbling Ducks
Bruns Slough 21 2.32b 9 6 16 36 4 4 3
Big Slough 21 2.70 12 51 26 c 15 9 34
Meszaros Slough 22 2.68 9 21 39 8 33 4 12
Kraft Slough 23 2.30 10 37 21 4 45 65 13
a RED = redhead, RUD = ruddy duck, CAN = canvasback, BWT = blue-winged teal, SHO = northern shoveler, and MAL = mallard.
b Pooled SE = 0.187.
c No canvasbacks were sighted during the survey of Big Slough but five nests were located on the wetland during spring and early summer.

Coots and eared grebes (Podiceps nigricollis Brehm) were the most common breeding waterbirds in 1979. Coots were most abundant on Bruns Slough (n = 140); 358 immatures were counted on 7 August 1979, indicating high recruitment. A colony of eared grebes on Big Slough had 445 nests; 150 and 104 eared grebe nests were counted in colonies on Kraft and Meszaros sloughs. Colonies of black-crowned night herons (Nycticorax nycticorax Linnaeus) with 100 (estimated) and 15 nests were observed on Kraft and Big sloughs, respectively; 14 black tern (Chlidonias niger Linnaeus) nests also were observed on Big Slough. Franklin's gulls (Larus pipixcan Wagler) were not present on the 6 June count at Kraft Slough, but by late June, an estimated 10,000 pairs were nesting on the east, north, and northwest sides of the central opening (U.S. Water and Power Resource Service 1980). Yellow-headed blackbirds (Xanthocephalus xanthocephalus Bonaparte) and long-billed marsh wrens (Cistothorus palustris Wilson) were abundant on BBMK. Overall, no differences existed in mean ranks of waterbird density among the four sloughs (F = 1.41, df = 3, 116; P = 0.2433) (Table 2). Similarity among BBM and Kraft Slough in species diversity and waterbird densities reflects that BBM provided high quality breeding habitat for waterfowl and other waterbirds prior to the clean-out.

BBM were major staging areas for lesser snow geese (Chen caerulescens Linnaeus) and Canada geese during spring and fall prior to the drain clean-out. Use tended to be highest during years when most other wetlands in southeastern North Dakota were dry. In the fall of 1961, for example, a year of severe drought over much of North Dakota, 50,000 to 60,000 lesser snow geese staged in Big Slough (U.S. Fish and Wildlife Service 1987). In spring 1974 (Figure 2), approximately 100,000 geese staged in Bruns and Big sloughs (G. Krapu, unpubl. obs). In November of 1981, another dry year, 15,000 and 7,000 geese staged on Bruns and Big sloughs (U.S. Fish and Wildlife Service 1987). Bruns and Big sloughs also were used as staging habitat by tundra swans (Cygnus columbianus Ord), probably because large sago pondweed beds provided attractive foraging sites (Earnst 1994). Bald eagles (Haliaeetus leucocephalus Linnaeus) often accompanied the large flocks of staging waterfowl during spring and fall before the clean-out. The last observation of bald eagles was on Big Slough on 10 November 1984.

Waterfowl hunting traditionally has been a major source of recreation during fall in Sargent County because the county's wetlands are located within a major waterfowl flight corridor of the Central Flyway (Bellrose 1980). At Big Slough during fall in the 1970s, up to 50 hunters would hunt from a railroad embankment on days when conditions were favorable for waterfowl shooting (H. Bellin, pers. comm.). At Meszaros Slough, two camps with 4 to 5 hunters each often were present. Numerous other hunters stayed at motels in nearby towns or resided within driving distance of BBM. The presence of state-owned game management areas at Big and Meszaros sloughs provided access to the general public.

Post Drain Clean-Out.  Waterfowl and other waterbirds avoided BBM for most of 1986-90 because of lack of water, resulting in a near total loss of waterbird production. In 1989, runoff was sufficient in April to cause ducks and other waterbirds to settle in BBM, but the water drained out by June before most nests had hatched and before any young could fledge. Loss of suitable habitat for waterbird breeding in BBM, combined with limited comparable habitat in the surrounding area, has sharply reduced production of diving ducks and several other species of waterbirds in western Sargent County. Kraft Slough from 1986 to 1990 held adequate water to support breeding populations of most species of ducks and most other waterbirds that were present at BBMK in 1979.

BBM lacked water and waterbirds during 4 of 5 early spring and 5 of 5 fall migration periods from 1986 to 1990. At Kraft Slough on 10 July 1990 at the peak of the drought and with BBM dry, 459 waterbirds were counted, including 312 waterfowl (J. Austin and M. Schwartz, unpubl. data). Also, several thousand geese were seen on Kraft Slough during each visit made annually in April 1987-90 (G. Krapu, unpubl. notes and photos). On 10 November 1990, Kraft Slough supported about 5000 lesser snow geese (K. Nelson, unpubl. notes and photos).

Fifty-four percent of the 114 North Dakota-shot of ducks banded in Sargent County were taken by hunters in Sargent County. With the drainage of three of the four principal waterfowl staging areas in western Sargent County as a result of the clean out of Drain No. 11, duck and goose hunting opportunities in Sargent County have diminished substantially during drought periods when water is limited primarily to lakes.

Potential Effects of Habitat Loss on Disease-Related Mortality

The loss of BBM as staging habitat, combined with drying of most other ponds due to drought, may have contributed to a dieoff of geese at Kraft Slough in November 1990. A total of 487 lesser snow geese died from a flock of about 5,000 birds (K. Nelson, Biological Technician, Tewaukon National Wildlife Refuge [NWR], unpubl. rept.), and autopsies revealed the cause of death was necrotic enteritis (J. C. Franson, National Wildlife Health Science Center [NWHSC], Madison, Wisconsin, unpubl. rept.). Necrotic enteritis is associated with proliferation of the bacteria, Clostridium perfringens (Veillon and Zuber) and the build-up of its toxins in the intestine. Damage to the intestinal lining or manipulation of the diet are considered to be predisposing factors in this disease (Ficken 1991). Geese seem to be most susceptible when staging on alkaline lakes, particularly during periods when limited fresh water is available (R. M. Windingstad, NWHSC, pers. commun.). Snow goose mortality from necrotic enteritis in North and South Dakota previously has been reported in dry years at sites when waters contained high concentrations (µg/l) of sodium (1,575-3,900), magnesium (2,030-2,200), sulfate (10,525-16,300), and TDS (17,250-24,400) (R. Windingstad, unpubl. data); concentrations of these ions in the waters of Kraft Slough under drought conditions in November 1990 were 3,090, 1,580, 10,990, and 18,200 µg/l, respectively (K. Koppinger, North Dakota State Department of Health and Consolidated Laboratories, Bismarck, ND, unpubl. data). With the drainage of Bruns and Big sloughs, geese staging in western Sargent County during Fall 1990 had limited access to water with lower salt concentrations than existed at Kraft Slough. Lake Taayer, a shallow saline wetland near Kraft Slough (Figure 1) was dry in 1990. Change in diet may also have contributed to the dieoff from necrotic enteritis by disrupting the intestinal microflora, allowing C. perfringens to proliferate (Wobeser and Rainnie 1987). All outbreaks of this disease reported to date have occurred during the fall (NWHSC, unpubl. data), which corresponds to when geese experience major changes in diets following migration from the arctic.

Grain residues remaining from the previous harvest are plentiful on cropland in the midcontinent region in early spring. These residues support massive flocking of migrant geese in areas where wetland habitat suitable for roosting is scarce due to wetland drainage, drought, or a combination of both (Krapu et al. 1995). In early spring 1991 with limited pond habitat available in eastern North Dakota (U.S. Fish and Wildlife Service 1995) including Sargent County (F. Giese, unpubl. memo), lesser snow geese congregated at high densities on remaining water areas. On 26 March 1991, an estimated 700,000 geese were roosting on Lake Tewaukon in Sargent County (F. Giese, Manager, Tewaukon NWR, pers. comm.), reflecting the magnitude of flocking that is possible under contemporary conditions.

Waterfowl may be more vulnerable to disease when crowded into limited habitat under harsher than normal environmental conditions (Siegal 1980, Wobeser 1981). In the Central Flyway, high losses to avian cholera (Pasteurella multocida Lehmann and Neumann) epizootics in the Rainwater Basin Area of southcentral Nebraska (an estimated 200,000 waterfowl have died since 1975 [J. Kauffeld, unpubl. data]) have occurred primarily when roosting habitat has been most restricted due to drought and wetland drainage. In the U.S. part of the prairie pothole region (PPR), an average of about 15 mortality events from disease are reported annually (1985-95). Botulism type C is the most common disease identified, but avian cholera epizootics have been documented in 4 of the past 5 years (NWHSC, unpubl. data). Of nine reported cases of avian cholera since 1989, most have been within a radius of 150 km of BBM, including an outbreak on Tewaukon NWR in Sargent County (NWHSC, unpubl. data), when about 500,000 lesser snow geese were present on 1 April 1991 (F. Giese, pers. comm.). A total of 79 dead waterfowl (including 76 lesser snow geese) were picked up on Lake Tewaukon on 2 April 1991 (F. Giese, unpubl. memo), of which 3 were examined and found to have died from avian cholera (J. Franson and L. Glaser, NWHSC, unpubl. rept.). This outbreak posed a risk to approximately one quarter of the midcontinent lesser snow geese population (based on the 1990-91 winter goose survey count of 2.1 million lesser snow geese [U.S. Fish and Wildlife Service and Canadian Wildlife Service 1994]). Additional losses occurred at Kraft Slough, where 8 dead lesser snow geese were seen on 2 April and 16 on 5 April; about 15,000 lesser snow geese were present on Kraft Slough on 4 April (F. Giese, unpubl. memo).

Because Sargent County wetlands are located on a major migration corridor of the Central Flyway, ducks and geese that stage there come from a large geographic area. As a result, high losses of waterfowl in Sargent County or elsewhere in the PPR from epizootics occurring under crowded conditions could reduce hunting opportunities and other values provided by midcontinent waterfowl populations over a wide area. Ducks banded in Sargent County have been reported from 34 states, 7 provinces of Canada, and 13 other countries (Figure 4). The highest proportion of band recoveries (16%) were from North Dakota. Aside from North Dakota, Minnesota, the Mississippi Delta area of eastern Arkansas, and the Louisiana Gulf Coast were the principal recovery locations (Figure 4). Mallards (n = 315) were recovered principally from North Dakota (18%), Arkansas (14%), and Minnesota (11 %) and northern pintails (n = 69) were taken most often in Louisiana (28%), North Dakota (19%), and Texas (12%). However, some pintails (7%) were recovered as far west as the Central Valley of California (Figure 4). Blue-winged teal (n = 169) were most often recovered from Louisiana (24%) and dispersed the farthest from Sargent County of all waterfowl; banded individuals were recovered in Canada (9), Mexico (9), Venezuela (9), Columbia (6), Cuba (5), Dominican Republic (5), El Salvador (2), and one each from Bahama Islands, Belize, French Guiana, Jamaica, Lesser Antilles, Panama, and Nicaragua (Figure 4).

Figure 4

Figure 4.  Distribution of band recoveries from ducks banded in Sargent County, North Dakota, 1951-89.

Of 316 recoveries of lesser snow geese banded in Sargent County, most occurred in eastern Texas (98), North Dakota (60), South Dakota (41), Missouri (21), Louisiana (19), Iowa (19), Minnesota (12), and Ontario (11). Most small Canada geese (B.c. parvipes Cassin) and B.c. hutchinsii Richardson) (n = 387), were recovered from North Dakota (116), Texas (87), Oklahoma (41), South Dakota (40), Manitoba (27), Montana (20), and Nebraska (10). The distribution of small Canada geese extends from the arctic to Mexico primarily in a corridor through the central plains, whereas lesser snow geese had a more easterly migration route.

Previous Section -- Methods
Return to Contents
Next Section -- Conclusions

Accessibility FOIA Privacy Policies and Notices

Take Pride in America logo logo U.S. Department of the Interior | U.S. Geological Survey
Page Contact Information: Webmaster
Page Last Modified: Saturday, 02-Feb-2013 06:56:36 EST
Menlo Park, CA [caww54]