Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

High Duck Nesting Success in a Predator-Reduced Environment Author(s): Harold F. Duebbert and John T. Lokemoen Source: The Journal of Wildlife Management, Vol. 44, No. 2 (Apr., 1980), pp. 428-437 Published by: Allen Press Stable URL: http://www.jstor.org/stable/3807974 Accessed: 04/12/2008 12:17 Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showpublisher?publishercode=acg. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that promotes the discovery and use of these resources. For more information about JSTOR, please contact support@jstor.org. Allen Press is collaborating with JSTOR to digitize, preserve and extend access to The Journal of Wildlife Management. http://www.jstor.org

HIGH DUCK NESTING SUCCESS IN A PREDATOR-REDUCED ENVIRONMENT HAROLD F. DUEBBERT, Northern Prairie Wildlife Research Center, U.S. Fish and Wildlife Service, Jamestown, ND 58401 JOHN T. LOKEMOEN, Northern Prairie Wildlife Research Center, U.S. Fish and Wildlife Service, Jamestown, ND 58401 Abstract: Duck nesting and production were studied during 1969-74 on a 51-ha field of undisturbed grass-legume cover and a surrounding 8.13-km2 area in north-central South Dakota. The principal mammalian predators of ducks were reduced within a 259-km2 zone from May 1969 through August 1971. Dabbling duck nest densities, hatching success, and breeding populations attained high levels. Seven duck species produced 1,062 nests on the 51-ha field during 6 years; 864 (81%) hatched, 146 (14%) were destroyed, and 52 (5%) had other fates. During 1970-72, when predator reduction was most effective, the hatching success for 756 nests was 94%. The number of mallard (Anas platyrhynchos) nests increased from 37 (0.7/ha) in 1969 to 181 (3.5/ha) in 1972. Mallard pairs increased from 2.8/km2 to 16.8/km2 on the 8.13-km2 area during the same period. A minimum of 7,250 ducklings hatched on the 51-ha field during the 6 years, including 2,342 ducklings in 1972. Exceptionally high duck nesting densities and hatching rates occurred when predators were controlled. J. WILDL. MANAGE. 44(2):428-437 Studies in the prairie pothole region of North America have shown that predation is often a primary cause of reduced nesting success of dabbling ducks (Keith 1961, Smith 1971, Stoudt 1971, Higgins 1977). Conversely, other studies have shown that duck nesting success or production increased when predation was temporarily reduced (Kalmbach 1939, Ellig 1955, Balser et al. 1968, Lynch 1972, Schranck 1972, Duebbert and Kantrud 1974). Waterfowl biologists, managers, and administrators have often speculated about the effects of organized predator control on duck production. During 1969-74 we studied a population of breeding dabbling ducks that included mallards, gadwalls (Anas strepera), pintails (A. acuta), green-winged teal (A. crecca), blue-winged teal (A. discors), northern shovelers (A. clypeata), and American wigeon (A. americana) on an 8.13-km2 study area near Hosmer, South Dakota, here termed the Hosmer Study Area (HSA). Primary emphasis was placed on nesting studies in a 51-ha field of undisturbed grass-legume cover. The objective of this study was to investigate the nesting ecology and population re- sponses of dabbling ducks in a glaciated prairie wetland region where the principal predators were rigidly controlled. We express our appreciation for assistance during the study to our supervisor, H. W. Miller; to F. Ulmer for allowing access to the study area; to P. Dosch who conducted the predator control for the South Dakota Department of Game, Fish, and Parks; to K. F. Higgins, L. M. Kirsch, and A. T. Klett and several other biologists for assistance with field work; to D. A. Davenport and D. H. Johnson for help with data analysis; and to G. A. Swanson and L. M. Kirsch, who reviewed our manuscript and gave editorial assistance. STUDY AREA The nesting study was conducted on a 51-ha field (here termed nesting field, NF) of tall, dense cover composed primarily of intermediate wheatgrass and alfalfa. This cover was established on a cropland field in 1965 under the U.S. Department of Agriculture's Cropland Adjustment Program, and remained undisturbed for the duration of our study. The study area was within a 259-km2 (16 x 16-km) area (here termed predator 428

DUCK NESTING IN A PREDATOR-REDUCED AREA Duebbert and Lokemoen 429 Fig. 1. The 51-ha nesting field and the 8.13-km2 Hosmer Study Area, 16 May 1972. control area, PCA), where red foxes (Vulpes vulpes), striped skunks (Mephitis mephitis), raccoons (Procyon lotor), and badgers (Taxidea taxus) were intensively controlled from May 1969 to August 1971. These mammals were the major predators of duck nests, hens, and ducklings in the area (Duebbert and Lokemoen 1976). Data on dabbling duck breeding populations and broods, wetland conditions, and land use were recorded on the 8.13-km2 circular HSA, with a radius of 1.6 km, centered on the NF (Fig. 1). The HSA contained a complex of 74 wetland basins, of which 14 were Class I (ephemeral), 21 were Class II (temporary), and 39 were Class III (seasonal) according to Stewart and Kantrud (1971). Wetlands ranged in size from 0.04 ha to 4.05 ha (c = 0.92 ha). Total area of the wetland basins was 67.9 ha, of which 0.9 ha was Class I, 6.5 ha were Class II, and 60.5 ha were Class III. Water conditions during mid-may and mid-july varied widely among years (Table 1). Wetland habitats were excellent for duck production in 1969 and 1970, good in 1971, excellent in 1972, and poor in 1973 and 1974. Agricultural drainage had not affected the area, and original wetland complexes remained intact. Aquatic vegetation in the wetlands was mostly undisturbed, grazed, or hayed. Only a few ephemeral or temporary wetlands were tilled. Giant burreed (Sparganium eurycarpum), slough sedge (Carex atherodes), and marsh smartweed (Polygonum coccineum) dominated the vegetation of undisturbed seasonal wetlands. Invertebrate populations were not quantified, but our observations indicated abundant snails (Lymnaeidae), midges (Chironomidae), and other food organisms used by breeding hens and ducklings associated with seasonal wetlands (Swanson et al. 1974). Land use on the HSA was typical of the diversified grain-livestock farms in the region. Percentages of the total area in various land uses were: grazed mixedgrass prairie, 28; cropland, 24; domestic

430 DUCK NESTING IN A PREDATOR-REDUCED AREA Duebbert and Lokemoen Table 1. Number and area of wetlands on the 8.13-km2 Hosmer Study Area in May and July 1969-74.a Mid-May Mid-July Year N Area (ha) N Area (ha) 1969 40 53.4 66 60.9 1970 36 42.3 15 15.4 1971 19 19.5 15 12.2 1972 53 55.6 16 18.2 1973 11 7.0 3 0.2 1974 16 3.7 5 0.4 a Based on 74 wetland basins with a total area of 67.9 ha. pasture, 9; wetlands, 8; domestic hay, 8; townsite, 7; undisturbed grass-legume cover (NF), 6; and miscellaneous, 9. The land use pattern and farming practices were consistent throughout the study. METHODS Intensive predator control was conducted continuously from May 1969 through August 1971 on the PCA. The HSA was situated within the PCA, which was part of a study of relationships between predation and ring-necked pheasants (Phasianus colchicus) conducted by the South Dakota Department of Game, Fish, and Parks (Trautman et al. 1974). Predator control methods included strychnine-treated baits, trapping, and shooting. Ducks breeding on our study area were in an essentially predator-free environment during 4 nesting seasons, as indicated by few observations of predators, tracks, scats, active dens, or roadkilled animals. There were few avian predators on the area; and small mammals, other than those controlled, were of minor importance as nest or duck predators. Nests were found by flushing hens with a 53-m cable-chain device towed between 2 vehicles (Higgins et al. 1969). Two complete searches were made of the NF each year at about 30-day intervals. Dates ranged from 25 May to 3 June for the 1st search and 25 June to 8 July for the 2nd. Eggs were candled to determine their stage of development (Weller 1956). The duck population was counted in mid-may and in early June according to techniques outlined by Hammond (1969). Wetland and land use conditions were recorded during the mid-may and mid- July surveys. Nest densities were totaled from all nests of each species that were found. We calculated hatching rates for each species by the conventional method of dividing the number of hatched nests by the sum of hatched plus destroyed nests. Nests that were deserted or destroyed because of our disturbance, or whose markers were lost, were omitted in calculations of nest success. Mayfield (1961, 1975) pointed out that conventional methods of analyzing nesting data tend to overestimate success and underestimate density because all nests were not found. Recently, Miller and Johnson (1978) discussed problems inherent in interpreting the results of nesting studies, and recommended use of the Mayfield method. In addition to the conventional method, our data were analyzed by the Mayfield method with a computer program developed by D. A. Davenport at the Northern Prairie Wild- life Research Center to determine the probability of a nest hatching. Tests were performed to determine the spatial relationships of mallard nests that were simultaneously active about 1 June each year. The principal method was the nearest-neighbor test, which compared the average distance between nearest nests to the value expected if nests were located randomly (Clark and Evans 1954). Nest searches were timed so that each hen would be flushed only once or twice, and nest fates were not checked until af-

DUCK NESTING IN A PREDATOR-REDUCED AREA Duebbert and Lokermoen 431 Table 2. Numbers and percentages of nests hatching for duck nests on 51-ha field on the Hosmer Study Area, 1969-74.a 1969 1970 1971 1972 1973 1974 Total/Average Species N % N % N % N % N % N % N % Mallard 37 79 93 99 79 95 181 90 96 56 13 69 499 84 Gadwall 47 71 50 90 60 100 82 90 24 59 23 86 286 86 Pintail 14 43 25 91 14 100 23 91 11 89 2 100 89 84 Green-winged teal 3 100 3 100 2 100 2 100 2 100 0 12 100 Blue-winged teal 17 59 61 98 14 91 19 84 3 33 0 114 87 Northern shoveler 7 71 18 100 5 100 10 90 3 67 0 43 90 American wigeon 1 0 4 75 6 83 5 100 2 50 1 0 19 78 Total/average 126 69 254 96 180 96 322 90 141 59 39 78 1,062 85 Nests/100 ha 247 498 353 631 276 76 347 a Percentages of hatching success may be based on slightly fewer nests because some were lost, deserted, or destroyed by searchers. ter calculated hatch dates. Fates of nests were determined by methods outlined by Rearden (1951) and Einarsen (1956) and our own extensive field experience. Ducks were observed from a blind located in the NF during 2 hours after sunrise on 4 and 14 May 1971 to obtain information on pursuit flights, aerial displays, and other behavior. Vegetation in the NF was analyzed by a series of 1-m2 quadrats and cover board readings along line transects as described by Duebbert and Lokemoen (1976). RESULTS AND DISCUSSION Hatching Success and Other Nesting Factors High success of dabbling duck nests occurred on the Hosmer Study Area (HSA) during 4 nesting seasons in the near-absence of mammalian predation (Table 2). Eggs hatched in 758 of 842 nests (90%) during 1969-72 when predator control was in effect. Predator reduction was most nearly complete during the 1970 and 1971 nesting seasons, and in those years, 395 of 411 nests (96%) hatched successfully. In 1969, a few predators were present because control had just been initiated. In 1972, preda- tors began to reinvade the area, but populations remained low from residual effects of control activities during the previous 3 years. Comparative data on duck nesting success in similar habitats not subject to predator control are available from a concurrent study in 9 fields of undisturbed grass-legume cover with a total area of 271 ha outside of the predator control area (PCA) (Duebbert and Lokemoen 1976). The vegetation on these fields was much like that on the HSA, and the nesting plots were surrounded by similar land-use patterns and wetland complexes. Fields were situated an average of 20 km from the HSA and from 3 to 10 km from the boundary of the PCA. In 1971, 1972, and 1973, eggs in 320 of 570 nests hatched, for an average success rate of 56% in the 9 fields without predator control (Duebbert and Lokemoen 1976). During 1971-72, nesting success on the HSA averaged 93%; on the 9 fields outside of the PCA, it averaged 63%. In 1971, average duck nesting success was 51% in a variety of habitats on agricultural land outside the PCA about 10 km from the nesting field (NF) (Duebbert and Kantrud 1974). Nest densities were low on the agricultural lands, averaging

432 DUCK NESTING IN A PREDATOR-REDUCED AREA Duebbert and Lokemoen Table 3. Hatching success (%) of nests as calculated by conventional (C) and Mayfield (M) methods, 1969-72. Blue-winged Year Mallard Gadwall teal 1969 C 79 71 59 M 46 53 35 1970 C 99 90 98 M 96 81 100 1971 C 95 100 91 M 84 100 76 1972 C 90 90 84 M 78 74 77 14/km2. Higgins (1977) reported hatching success of 25% on untilled habitats and 17% on tilled habitats for duck nests in intensively farmed areas of North Dakota. Nest destruction rates were relatively low during most years of our study. Thus, there was less difference between nesting success as calculated by the conventional or the Mayfield method on the HSA, than in areas with higher nest losses (Table 3). For example, in 1971, mallard nesting success for 79 nests was 95% by the conventional method and 84% by the Mayfield method. In 1971, gadwall nesting success for 60 nests was 100% by both methods. In 1969, when nest destruction rates were higher, there was a greater difference between nest successes determined by the 2 methods. Abandonment rates were low: 21 of 1,062 nests (2%) were abandoned during all years, indicating that nest-site fidelity was not adversely affected by the high nest density. Nearly all abandonments were a result of our interference when nests contained 1-5 eggs. Another 31 nests had other fates: 6 had complete clutches of infertile eggs, 4 were destroyed by search vehicles, and 21 had unknown fates because markers were lost. Clutch sizes were normal for the species of ducks that nested on the HSA, indicating that egg laying was not influenced by high-density nesting, concentration of pairs, or long-distance flights. Average clutch size in 1971 for mallards was 8.1 in the NF and 8.2 in fields of similar cover outside of the PCA (Duebbert and Lokemoen 1976), and gadwall mean clutch size was 10.7 on both areas. In 1972, clutch sizes for mallards and gadwalls were slightly lower in the NF than in the other fields (8.1 vs. 8.6 and 10.1 vs. 10.8, respectively). Mallard clutch size during the study averaged 8.4 (338 nests), which is the same average reported for long-term studies in Saskatchewan by Stoudt (1971). Hatchability rates of eggs were high in successful nests on the HSA. During all years, 7,250 of 7,594 eggs (95%) hatched in successful nests. The 7,250 figure represents the minimum number of ducklings hatched on the field. Only 5 of 1,062 nests contained an abnormally large number of eggs or eggs of different species, and that indicated a low incidence of intraspecific or interspecific egg laying. For mallard hens in the predator-reduced environment on the HSA, the nesting chronology was relatively early. In 1970 and 1971, 89% and 84% of the mallard nests were started before 1 June. In 1972, when predator populations remained low, 97% of the mallard nests were started before i June and 42% before 1 May. About 1/2 of the young mallards would be fledged by 1 August under this early nesting schedule. Nest Densities and Spacing Dabbling ducks normally nest solitarily, but aggregations of some species may be formed where the nesting environment is safe (Lack 1968:118). The nesting field on the HSA contained unusually

DUCK NESTING IN A PREDATOR-REDUCED AREA Duebbert and Lokernoen 433 Table 4. Indicated dabbling duck pairs on the 8.13-km2 Hosmer Study Area, 1969-74. 1969 1970 1971 1972 1973 1974 Total Species N % N % N % N % N % N % N % Mallard 23 12 90 23 59 30 137 27 67 32 18 50 394 26 Gadwall 39 20 61 16 28 14 54 11 48 23 6 17 236 15 Pintail 35 18 80 21 39 20 107 21 54 25 7 19 322 21 Green-winged teal 6 3 6 2 3 2 8 2 2 1 0 0 25 2 Blue-winged teal 63 33 109 28 40 20 130 26 13 6 4 11 359 23 Northern shoveler 25 13 41 11 24 12 60 12 24 11 1 3 175 11 American wigeon 1 tra 3 tr 4 2 6 1 4 2 0 0 18 1 Total 192 99 390 101 197 100 502 100 212 100 36 100 1,529 99 Pairs/km2 24 48 24 62 26 4 31 a Less than 1%. dense aggregations of nests during 1969-72, when the ducks were in a predatorreduced environment (Table 2). Nests of all species of resident dabbling ducks were found, but mallard nests made up 47% (499 nests) and gadwalls 27% (286 nests) of the total of 1,062 nests. In 1968, the year before intensive predator reduction began, 61 nests (120/100 ha) were found on the NF (Duebbert 1969). In 1969, 126 nests (247/100 ha) were found, and the density increased to 322 nests (631/100 ha) in 1972. The number of mallard nests increased from 37 in 1969 to 181 in 1972. Gadwall nests increased from 47 to 82 during the same period. Comparison between density of mallard nests on the NF and pairs counted in the surrounding HSA indicated that nests were aggregated in the NF. During all years, the number of mallard nests on the 51-ha NF closely approximated the pair population in the 8.13-km2 HSA. Mallard pairs were observed flying to the NF from wetlands up to 3.2 km away. When hens were flushed from nests, they did not always go to the nearest wetland, but often flew 1.6 km or more. Mallard nests were spaced in various patterns throughout the NF in different years. Results of the nearest-neighbor tests showed the following nest distri- bution patterns for the respective years: 1969, random; 1970, aggregated (P = 0.14); 1971, random; 1972, aggregated (P = 0.013); 1973, aggregated (P < 0.001); 1974, spaced (P < 0.01). On the basis of 6-year averages, there was a general tendency toward higher nest densities (2.29/ha) on the soil zones with higher capability ratings (according to the U.S. Soil Conservation Service classification) and lower (0.65/ha) on those with lower capability ratings. Soils with higher capability ratings supported taller, more dense cover, and slopes ranged 0-6%. Soils with lower capability ratings supported shorter, less dense cover, and slopes ranged 6-12%. This indicated a tendency for nesting hens to select nest sites in zones of taller, more dense cover within the NF. The highest known density of active mallard nests in the NF at 1 time was 154 (3.0/ha) on 24 May 1972. Then, the average distance between nests was 27 m, and 119 (78%) of the nests were less than 30 m apart. We found 3 mallard nests that were closer to other nests than 6 m; the intervals were 2.1 m, 3.0 m, and 4.6 m. Newton and Campbell (1975) found that 2 m was about the closest nest spacing mallard hens would tolerate without strife. We concluded there was unoccu-

434 DUCK NESTING IN A PREDATOR-REDUCED AREA Duebbert and Lokernoen Table 5. Densities of dabbling duck pairs in relation to surface water in mid-may on the 8.13-km2 Hosmer Study Area, 1969-74. Pairs per ha of surface water Species 1969 1970 1971 1972 1973 1974 Mallard 0.43 2.12 3.03 2.46 9.57 4.86 Gadwall 0.73 1.44 1.44 0.97 6.86 1.62 Pintail 0.66 1.89 2.00 1.92 7.71 1.89 Green-winged teal 0.11 0.14 0.15 0.14 0.29 Blue-winged teal 1.18 2.58 2.05 2.34 1.86 1.08 Northern shoveler 0.47 0.97 1.23 1.08 3.43 0.27 American wigeon 0.02 0.07 0.21 0.11 0.57 Total 3.60 9.22 10.10 9.03 30.29 9.73 pied nesting cover for mallards on the NF even with the high densities we recorded. Breeding Populations Breeding populations of dabbling ducks, especially mallards, increased to high densities on the HSA (Table 4). In 1969, populations of all dabblers were similar to those generally found in the northern Coteau du Missouri in South Dakota (Brewster et al. 1976, Duebbert and Lokemoen 1976). Blue-winged teal were the most numerous, followed by gadwalls, pintails, mallards, and northern shovelers. Between May 1969 and May 1970, we recorded a 103% increase in dabbling duck pairs and a 291% increase in mallard pairs, but a 10% decrease in HSA wetlands with water. Mallards changed from 12% to 23% of the breeding population and from 5th to 2nd in abundance. During the same period, in eastern South Dakota, Pospahala et al. (1974:67), reported a 32% decrease in wetlands with water, a 5% increase in total ducks, and an 8% increase in breeding mallards. Mallards became the most abundant breeding species in 1971, and maintained that position until the end of the study. Gadwalls and other dabblers also increased, but less markedly than mallards. Numbers of ponds in May 1971 decreased 12% in eastern South Dakota (Pospahala et al. 1974:67) and 48% on the HSA (Table 1). In 1972, the dabbling duck breeding population increased to 502 pairs, or 62 pairs/km2, which was the highest density recorded on our area. Mallard pairs increased to 137, for a density of 16.8 pairs/ km2 (Table 4). In 1972, use of available water by breeding pairs of all species averaged 9.0/ha; mallard pairs averaged 2.5/ ha (Table 5). Wetlands with water and total breeding pairs also increased on nearby study areas (Duebbert and Lokemoen 1976) and in eastern South Dakota (Pospahala et al. 1974). The large increases in breeding populations, especially mallards, occurred only on the HSA. Some wetlands on the HSA contained extremely high densities of breeding ducks. For example, on 17 May 1972 we counted 40 indicated pairs of ducks on a 1.4-ha seasonal wetland. Other high populations on seasonal wetlands were 47 pairs on 3.2 ha, 45 pairs on 2.4 ha, and 65 pairs on 4.0 ha. In 1973 there was a large reduction in water area (Table 1), and effects of predation increased after control ceased in August 1971. The area of surface water decreased 88%, and only 11 of 76 wet-

DUCK NESTING IN A PREDATOR-REDUCED AREA Duebbert and Lokermoen 435 lands contained water in mid-may. Numbers of breeding ducks declined, but crowded on the remaining water in dense aggregations. In 1973, total duck pairs averaged 30.3/ha of water and mallard pairs averaged 9.6/ha of water (Table 5). Water conditions deteriorated further in 1974, and in mid-may there was only 3.7 ha of water. The highest mallard pair density on the HSA was 16.8/km2 in 1972. The highest recorded mallard pair density we found in our literature search was 36.2/km2 at Lousana, Alberta, in 1958 (Smith 1971). The pair density of all dabbling duck species was 84.5/km2 at Lousana in 1958, compared with the 1972 density of 61.8/ km2 on the HSA. On the basis of density per area of water, the 1973 HSA mallard pair population of 9.6/ha and dabbler pair population of 30.3/ha were higher than any we found in the literature. When Smith (1971) recorded the high populations in 1958, the mallard pair density per area of water was 5.4/ha, and the dabbler pair density was 12.8/ha. During the May 1971 observations, we obtained information on behavioral interactions of ducks that were nesting in high densities. Of 63 mallard hens that arrived at the field as members of a pair or a group, 46 (73%) were not accompanied or chased by male mallards. The entire sequence of 5 pursuit flights was observed, and hens eventually landed in the cover, suggesting that behavioral interactions did not prevent those hens from nesting in the field. Of 53 gadwall hens that arrived at the field as members of a pair or group, 38 (72%) were not accompanied or chased by male gadwalls. The great amount of duck activity that accompanied the high concentration of nests is emphasized by the fact that a dabbling duck pair, lone drake, or group flight arrived at the field on the average of 1 each 0.7 min- ute. A mallard pair, lone drake, or group flight was seen each 1.4 minutes. Mallards and gadwalls in the breeding population apparently adapted in 2 ways to the high nesting densities: sizes of defended areas were reduced so that more pairs were accommodated on the available wetlands near the NF, and some birds used wetlands located more than 1.6 km from the NF. Many birds flew to the NF from several wetlands about 2.4 km away. On the HSA, pairs were spaced throughout the available wetland habitat, but at higher than normal densities. Hens selected nest sites in the most attractive and secure nesting cover, and that contributed to aggregations of nests. After hatching of nests, the hens and broods dispersed into the complex of wetlands within 1.6-3.2 km of the nesting field. Nesting Cover Vegetation in the NF was composed of about 90% intermediate wheatgrass, 8% alfalfa, and 2% other plants. Loosely arranged litter composed of residual vegetation from previous growing seasons covered 100% of the ground and averaged 13 cm (range 6-52) in depth. The number of dead plant stems averaged 104/m2 (range 12-254), with an average height of 80 cm (range 50-133). The average height of dead leaves on plants that remained standing from the previous year was 60 cm (range 31-79), and that of green leaves 30 cm (range 28-37). Mallard nests were in tall cover. Of 499 nests, 490 (98%) were in cover that exceeded 30 cm in height. Four hundred four (81%) were in 30- to 60-cm cover, 86 (17%) were in cover taller than 60 cm, and 9 (2%) were in 15- to 30-cm cover. No mallard nests were in cover shorter than 15 cm, but such cover was available in the field. Gadwall nests were in tall cover simi-

436 DUCK NESTING IN A PREDATOR-REDUCED AREA. Duebbert and Lokemnoen lar to that used by mallards. Of 286 nests, 147 (51%) were in 30- to 60-cm cover, 135 (47%) were in cover taller than 60 cm, 4 (1%) were in 15- to 30-cm cover, and none was in cover shorter than 15 cm. MANAGEMENT IMPLICATIONS For many years, managers of waterfowl production areas have sought management practices to increase duck production on small units of habitat. In the future, maintenance of desirable duck populations may require greater emphasis on intensive management of lands dedicated to waterfowl production. As Bellrose and Low (1978:64) stated: "If waterfowl are to be maintained at anywhere near existing levels, it is absolutely imperative to increase the waterfowl yield per unit of habitat." Cooch (1969:6) suggested additional study of the degree to which breeding ducks can be crowded on available habitat. Our study showed that a dabbling duck nest density of 631/100 ha (1972) and hatching success of 96% (1971) occurred on a 51-ha field of undisturbed grass-legume cover. The major environmental factors that contributed to the high nest density and hatching success included a complex of high quality natural wetlands; a field of undisturbed tall, dense vegetation for nesting; and rigid control of predators. Habitats available for ducks in the production of 7,250 hatched ducklings in 6 years included 51 ha of nesting cover and 74 wetlands totaling 68 ha. High nest densities and hatching success of dabbling ducks on islands are well documented (Hammond and Mann 1956, Duebbert 1966, Vermeer 1968). This study indicated that by the judicious use of habitat management and predator management, it is possible to create highly productive islands of habitat for duck nesting on the mainland. We document- ed 2 facts relevant to the management of habitats to increase duck production above that which occurs normally. First, mammalian predator populations were effectively reduced and maintained at low levels on a 259-km2 area including the HSA by 1 man for 3 years. However, regulations are now in effect that restrict the use of strychnine on federal lands, and strychnine was the most effective control method. Second, dabbling ducks, especially mallards, responded to the reduced predation with increased nest densities, hatching success, and breeding populations. The manipulation of factors that influence the inherent rate of increase of game populations is a basic concept of wildlife management. During the past 20 years, there has been a steady decline in the quantity and quality of waterfowl production habitat on private lands in the prairie pothole region of the United States. Therefore, we suggest that maintenance of high-quality nesting cover, protection and management of wetlands, and predator management should be em- phasized on areas intensively managed for waterfowl production. In our opinion, predator management should not be limited to direct control of animals, but could include the use of mechanical and natural barriers to protect nesting habitats and other management techniques to reduce predation. LITERATURE CITED BALSER, D. S., H. H. DILL, AND H. K. NELSON. 1968. Effect of predator reduction on waterfowl nesting success. J. Wildl. Manage. 32:669-682. BELLROSE, F. C., AND J. B. Low. 1978. Advances in waterfowl management research. Wildl. Soc. Bull. 6:63-72. BREWSTER, W. G., J. M. GATES, AND L. D. FLAKE. 1976. Breeding waterfowl populations and their distribution in South Dakota. J. Wildl. Manage. 40:50-59. CLARK, P. J., AND F. C. EVANS. 1954. Distance to

DUCK NESTING IN A PREDATOR-REDUCED AREA. Duebbert and Lokemoen 437 nearest neighbor as a measure of spatial relationships in populations. Ecology 35:445-453. COocH, F. G. 1969. Waterfowl production habitat requirements. Pages 5-10 in Saskatoon wetlands seminar. Can. Wildl. Serv. Rep. Ser. 6. Ottawa. DUEBBERT, H. F. 1966. Island nesting of the gadwall in North Dakota. Wilson Bull. 78:12-25..1969. High nest density and hatching success of ducks on South Dakota CAP land. Trans. North Am. Wildl. Nat. Resour. Conf. 34:218-229., AND H. A. KANTRUD. 1974. Upland duck nesting related to land use and predator reduction. J. Wildl. Manage. 38:257-265.,AND J. T. LOKEMOEN. 1976. Duck nesting in fields of undisturbed grass-legume cover. J. Wildl. Manage. 40:39-49. EINARSEN, A. S. 1956. Determination of some predator species by field signs. Oreg. State Monogr. Stud. Zool. 10. 34pp. ELLIG, L. J. 1955. Waterfowl relationships to Greenfields Lake, Teton County, Montana. Mont. Fish Game Comm Bull. 1. 35pp. HAMMOND, M. C. 1969. Notes on conducting waterfowl breeding population surveys in the north central states. Pages 238-254 in Saskatoon wetlands seminar. Can. Wildl. Serv. Rep. Ser. 6. Ottawa.,AND G. E. MANN. 1956. Waterfowl nesting islands. J. Wildl. Manage. 20:345-352. HIGGINS, K. F. 1977. Duck nesting in intensively farmed areas of North Dakota. J. Wildl. Manage. 41:232-242.,L. M. KIRSCH, AND I. J. BALL, JR. 1969. A cable-chain device for locating duck nests. J. Wildl. Manage. 33:1009-1011. KALMBACH, E. R. 1939. Nesting success: its significance in waterfowl production. Trans. North Am. Wildl. Conf. 4:591-604. KEITH, L. B. 1961. A study of waterfowl ecology on small impoundments in southeastern Alberta. Wildl. Monogr. 6. 88pp. LACK, D. 1968. Ecological adaptations for breeding in birds. Methuen and Co., Ltd., London. 409pp. LYNCH, G. M. 1972. Effect of strychnine control on nest predators of dabbling ducks. J. Wildl. Manage. 36:436-440. MAYFIELD, H. 1961. Nesting success calculated from exposure. Wilson Bull. 73:255-261..1975. Suggestions for calculating nest success. Wilson Bull. 87:456-466. MILLER, H. W., AND D. H. JOHNSON. 1978. Interpreting the results of nesting studies. J. Wildl. Manage. 42:471-476. NEWTON, I., AND C. R. G. CAMPBELL. 1975. Breeding of ducks at Loch Leven, Kinross. Wildfowl 26:83-102. POSPAHALA, R. S., D. R. ANDERSON, AND C. J. HEN- NY. 1974. Population ecology of the mallard: II. Breeding habitat conditions, size of the breeding populations, and production indices. U.S. Fish Wildl. Serv. Resour. Publ. 115. 73pp. REARDEN, J. D. 1951. Identification of waterfowl nest predators. J. Wildl. Manage. 15:386-395. SCHRANCK, B. W. 1972. Waterfowl nest cover and some predation relationships. J. Wildl. Manage. 36:182-186. SMITH, A. G. 1971. Ecological factors affecting waterfowl production in the Alberta parklands. U.S. Fish Wildl. Serv. Resour. Publ. 98. 49pp. STEWART, R. E., AND H. A. KANTRUD. 1971. Classification of natural ponds and lakes in the glaciated prairie region. U.S. Fish Wildl. Serv. Resour. Publ. 92. 57pp. STOUDT, J. H. 1971. Ecological factors affecting waterfowl production in the Saskatchewan parklands. U.S. Fish Wildl. Serv. Resour. Publ. 99. 58pp. SWANSON, G. A., M. I. MEYER, AND J. R. SERIE. 1974. Feeding ecology of breeding bluewinged teals. J. Wildl. Manage. 38:396-407. TRAUTMAN, C. G., L. F. FREDRICKSON, AND A. V. CARTER. 1974. Relationship of red foxes and other predators to populations of ring-necked pheasants and other prey, South Dakota. Trans. North Am. Wildl. Nat. Resour. Conf. 39:241-255. VERMEER, K. 1968. Ecological aspects of ducks nesting in high densities among larids. Wilson Bull. 80:78-83. WELLER, M. W. 1956. A simple field candler for waterfowl eggs. J. Wildl. Manage. 20:111-113. Received 2 December 1977. Accepted 11 January 1979.