Raymond J. Greenwood

Predator selection of prairie landscape features and its relation to duck nest success

Mammalian predation is a major cause of mortality for breeding waterfowl in the U.S. Northern Great Plains, and yet we know little about the selection of prairie habitats by predators or how this influences nest success in grassland nesting cover. We selected 2 41.4-km 2 study areas in both 1996 and 1997 in North Dakota, USA, with contrasting compositions of perennial grassland. A study area contained either 15-20% perennial grassland (Low Grassland Composition LGC]) or 45-55% perennial grassland (High Grassland Composition [HGC]). We used radiotelemetry to investigate the selection of 9 landscape cover types by red fox (Vulpes vulpes) and striped skunk (Mephitis mephitis), while simultaneously recording duck nest success within planted cover. The cover types included the edge and core areas of planted cover, wetland edges within planted cover or surrounded by cropland, pastureland, hayland, cropland, roads, and miscellaneous cover types. Striped skunks selected wetland edges surrounded by agriculture over all other cover types in LGC landscapes (P-values for all pairwise comparisons were <0.05). Striped skunks also selected wetland edges surrounded by agriculture over all other cover types in HGC landscapes (P < 0.05), except for wetland edges within planted cover (P = 0.12). Red foxes selected the edge and core areas of planted cover, as well as wetland edges within planted cover in LGC landscapes (i.e., they were attracted to the more isolated patches of planted cover). However, in HGC landscapes, red foxes did not select interior areas of planted cover (i.e., core areas of planted cover and wetland edges in planted cover) as frequently as edges of planted cover (P< 0.05). Red foxes selected core areas of planted cover more frequently in LGC than in HGC landscapes (P< 0.05) and selected pastureland more frequently in HGC than in LGC landscapes (P< 0.05). Furthermore, red foxes selected the isolated patches of planted cover more than pastureland in LGC landscapes (P< 0.05). Duck nest success was greater in HGC landscapes than in LGC landscapes for planted-cover core (P < 0.0001), planted-cover edge (P< 0.001) and planted cover-wetland edge (P< 0.001). Both the increased amount of planted-cover core area and the increased pastureland selection in HGC landscapes may have diluted predator foraging efficiency in the interior areas of planted cover and contributed to higher nest success in HGC landscapes. Our observations of predator cover-type selection not only support the restoration and management of large blocks of grassland but also indicate the influence of alternative cover types for mitigating nest predation in the Prairie Pothole Region.

Effects of harness transmitters on behavior and reproduction of wild mallards

Radio telemetry has been an important research tool in waterfowl studies for >20 years, yet little effort has been made to evaluate potential effects of transmitters on the birds that carry them. As part of a 4-year mallard (Arras platyrhynchos) study in the prairie pothole region of North Dakota and Minnesota, we compared radio-marked and unmarked female mallards in terms of percent time observed feeding, resting, and preening; nest initiation date; and clutch size and egg volume. Radio-marked females carried a 23-g back-mounted transmitter attached with a 2-loop harness (Dwyer 1972). On average, radio-marked females tended to feed less, rest and preen more, initiate nests later, and lay smaller clutches and eggs than unmarked females

Relationships of habitat patch size to predator community and survival of duck nests

We studied duck nest success and predator community composition in relation to size of discrete patches of nesting cover in the Prairie Pothole Region (PPR) of the United States in 1993-95 We focused on nests in uplands that were seeded to perennial grasses and forbs and enrolled in the Conservation Reserve Program (CRP) in Minnesota, North Dakota, and South Dakota. We estimated daily survival rates (DSRs) of upland duck nests and indices of activity for red foxes (Vulpes vulpes), coyotes (Canis latrans), American badgers (Taxidea taxus), striped skunks (Mephitis mephitis) and Franklins ground squirrels (Spermophilus franklinii) and related these variables to habitat patch size. The effect of patch size (small vs. large) on estimated annual mean DSR was dependent on date of nest initiation (early vs. late) and year. Examination of within-year comparisons for early and late nests suggested that DSR was generally greater in larger habitat patches Activity indices for the 5 mammalian nest predators were influenced differently by year, location, and patch size. Activity indices of the red fox were greatest in small patches. Coyote indices were the most inconsistent, demonstrating a year × location × patch size interaction. Activity indices of the striped skunk and American badger varied only among years. Franklins ground squirrel indices were affected by study area location, with higher indices in the southeast than the northwest. Red fox activity was weakly correlated with that of the striped skunk and coyote. Although a positive relationship between habitat patch size and nest success probably exists, we believe the experiment to fully test this hypothesis will continue to be elusive.

Analysis of Predator Movement in Prairie Landscapes with Contrasting Grassland Composition

Abstract Mammalian predation influences waterfowl breeding success in the U.S. northern Great Plains, yet little is known about the influence of the landscape on the ability of predators to find waterfowl nests. We used radiotelemetry to record nightly movements of red foxes (Vulpes vulpes) and striped skunks (Mephitis mephitis) in two 41.4-km2 study areas in North Dakota. Study areas contained either 15–20% grassland (low grassland composition) or 45–55% grassland (high grassland composition). Grasslands included planted cover, pastureland, and hayland. We predicted that the type and composition of cover types in the landscape would influence both predator movement across the landscape (as measured by the fractal dimension and displacement ratio) as well as localized movement (as measured by the rate of movement and turning angle between locations) within patches of different cover types. Red fox movements were straighter (lower fractal dimensions and higher displacements) across landscapes with a low grassland composition, indicating directed movement between the more isolated patches of planted cover. Striped skunk movements did not differ between landscape types, illustrating their movement along wetland edges, which had similar compositions in both landscape types. The high variability in turning angles by red fox in planted cover and pastureland in both landscape types is consistent with restricted-area foraging. The high rate of movement by red foxes in planted cover and by striped skunks in wetland edges suggests that spatial memory may influence movement patterns. Understanding the behavior of predators in fragmented prairie landscape is essential for managing breeding habitat for grassland birds and for predicting the spatial and temporal dynamics of predators and their prey.

LANDSCAPE COMPOSITION, PATCH SIZE, AND DISTANCE TO EDGES: INTERACTIONS AFFECTING DUCK REPRODUCTIVE SUCCESS

Prairies and other North American grasslands, although highly fragmented, provide breeding habitat for a diverse array of species, including species of tremendous economic and ecological importance. Conservation and management of these species re- quires some understanding of how reproductive success is affected by edge effects, patch size, and characteristics of the landscape. We examined how differences in the percentage of grassland in the landscape influenced the relationships between the success of nests of upland-nesting ducks and (1) field size and (2) distance to nearest field and wetland edges. We collected data on study areas composed of 15-20% grassland and areas composed of 45-55% grassland in central North Dakota, USA during the 1996 and 1997 nesting seasons. Daily survival rates (DSRs) of duck nests were greater in study areas with 45-55% grassland than with 15-20% grassland. Within study areas, we detected a curvilinear relationship between DSR and field size: DSRs were highest in small and large fields and lowest in moderately sized fields. In study areas with 15-20% grassland, there was no relationship between probability of hatching and distance to nearest field edge, whereas in study areas with 45-55% grassland, there was a positive relationship between these two variables. Results of this study support the conclusion that both landscape composition and config- uration affect reproductive success of ground-nesting birds. We are prompted to question conservation strategies that favor clustering moderately sized patches of nesting habitat within agricultural landscapes because our results show that such patches would have low nest success, most likely caused by predation. Understanding the pattern of nest success, and the predator-prey mechanisms that produce the pattern, will enable design of patch configurations that are most conducive to meeting conservation goals.

Evaluation of Mark-Recapture for Estimating Striped Skunk Abundance

The mark-recapture method for estimating striped skunk (Mephitis mephitis) abundance was evaluated by systematically livetrapping a radio-equipped population on a 31.4-km2 study area in North Dakota during late April of 1977 and 1978. The study population was 10 females and 13 males in 1977 and 20 females and 8 males in 1978. Skunks were almost exclusively nocturnal. Males traveled greater nightly distances than females (3.3 vs. 2.6 km, P < 0.05) and had larger home ranges (308 vs. 242 ha) although not significantly so. Increased windchill reduced night-time activity. The population was demographically but not geographically closed. Frequency of capture was positively correlated with time skunks spent on the study area. Little variation in capture probabilities was found among trap-nights. Skunks exhibited neither trap-proneness nor shyness. Capture rates in 1977 were higher for males than for females; the reverse occurred in 1978. Variation in individual capture rates was indicated among males in 1977 and among females in 1978. Ten estimators produced generally similar results, but all underestimated true population size. Underestimation was a function of the number of untrapped skunks, primarily those that spent limited time on the study area. The jackknife method produced the best estimates of skunk abundance. J. WILDL. MANAGE. 49(2):332-340 The estimation of animal abundance is often difficult. Mark-recapture is used commonly to estimate the abundance of secretive mammals, including striped skunks (Verts 1967, Bailey 1971, Lynch 1972). Although there are many estimators available (Cormack 1968, 1979; Otis et al. 1978; Pollock 1981; Seber 1982), there are few data on their application to populations of known size. This paper assesses the applicability of several estimators with data for populations of striped skunks of known size in 2 years and discusses aspects of skunk biology that influence capture probability. Although captures in this s udy are fewer than desired for most markrecapture evaluations, the approach represents a practical application of the mark-recapture method. W. B. Bicknell and J. L. Piehl provided field assistance. Radiotelemetry equipment was obtained from the James Ford Bell Mus., Cedar Creek Natl. Hist. Area, Univ. of Minnesota, St. Paul. D. R. Anderson and G. C. White provided the CAPTURE computer program. Appreciation is extended to D. W. Sparling, who verified This content downloaded from 157.55.39.201 on Sat, 17 Dec 2016 05:24:49 UTC All use subject to http://about.jstor.org/terms J. Wildl. Manage. 49(2):1985 STRIPED SKUNK MARK-RECAPTURE * Greenwood et al. 333 some of the calculations and reviewed the manuscript, and to D. R. Anderson, K. P. Burnham, R. M. Cormack, L. L. Eberhardt, D. L. Otis, K. H. Pollock, and G. C. White, who provided constructive comments on the manu-

Effects of radio transmitters on nesting captive mallards

Radio packages may subtly affect bird behavior and condition, and thus could bias results from studies using this technique. To asses effects on reproduction of mallards (Anas platyrhynchos), we tested 3 types of back-mounted radio packages on captive females. Eight paired females were randomly assigned to each of 4 treatments: 4-g transmitter attached with sutures and glue, 10-g or 18-g transmitter attached with a harness, and no transmitter (control). All mallards were fed ad libitum. No differences were detected among treatments in number of clutches, clutch size, nesting interval, egg mass, or body mass; powers (range = 0.15-0,48) of tests were low. Feather wear and skin irritation around radio packages were minimal

Age-related reproduction in striped skunks (Mephitis mephitis) in the upper Midwest

Reproductive data from the upper Midwest are meager for the striped skunk ( Mephitis mephitis ), a common North American carnivore. We provide data on some age-related reproductive attributes of 178 female striped skunks collected at 19 sites in eastcentral North Dakota and westcentral Minnesota in 1979–1981 and 1987–1991. Seventy-four percent of the females were 1 year old; 95% were pregnant or parous when collected. Thirteen of 873 (1.5%) embryos in 123 pregnant females were being resorbed. The overall mean (±1 SE ) litter size estimated from live embryos was 7.2 ± 0.4. Means of litter-size estimates were similar for females ≥1 year old, but annual estimates of litter size differed among years for all females combined. For females from the interval 1979–1981 and 1990, the mean implantation date based on embryo size was 4 March (±1.6 days). Potential litters were composed of a mean of 55 ± 3% females. Estimates of litter size based on counts of corpora lutea averaged 0.9 young per female less than estimates for the same females based on counts of live embryos, indicating that some skunks may have produced polyovular follicles or identical twins.

Alula Removal Technique for Identifying Wings of Released Waterfowl

A technique for marking waterfowl ducklings less than 1 day old was developed to permit subsequent identification in the hand. Marking was accomplished by removing the alulae from wings of mallard (Anas platyrhynchos) ducklings with a small scissors or clipper. No differences in growth rate, behavior, or flight capability were noted between clipped and nonclipped birds. Thirty-five hun- dred ducklings were clipped when less than 1 day old, and 1,880 of these birds were subsequently examined at 41/2 weeks of age. Significant (P <0.01) differences in successful removal of right and left alulae indicate that right-handed persons should exercise more care when clipping left wings than when clipping right wings. Alula removal appears to be effective for marking the wings of waterfowl. The Duck Wing Survey of the U. S. Bu- reau of Sport Fisheries and Wildlife is con- ducted annually throughout the nation to gather data on the species, age, and sex of ducks harvested by licensed waterfowl hunters ( Carney and Geis 1960 ) . Duck wings received from cooperating hunters are collected and identified at regional cc . 1 ,,