David R. Luukkonen

GEOGRAPHICAL GENETICS: CONCEPTUAL FOUNDATIONS AND EMPIRICAL APPLICATIONS OF SPATIAL GENETIC DATA IN WILDLIFE MANAGEMENT

Abstract Molecular-genetic technology and statistical methods based on principles of population genetics provide valuable information to wildlife managers. Genetic data analyzed in a hierarchical, spatial context among individuals and among populations at micro- and macro-geographic scales has been widely used to provide information on the degree of population structure and to estimate rates of dispersal. Our goals were to (1) provide an overview of spatial statistics commonly used in empirical population genetics, and (2) introduce analytical designs that can be employed to extend hypothesis-testing capabilities by incorporating space-time interactions and by using information on habitat quality, distribution, and degree of connectivity. We show that genetics data can be used to quantify the degree of habitat permeability to dispersal and to qualify the negative consequences of habitat loss. We highlight empirical examples that use information on spatial genetic structure in areas of harvest derivation for admixed migratory species, wildlife disease, and habitat equivalency analysis.

Movements and Survival of Molt Migrant Canada Geese From Southern Michigan

Abstract We studied movements and survival of 250 female giant Canada geese (Branta canadensis maxima) marked during incubation with either satellite-monitored platform transmitting terminals or very high frequency radiotransmitters at 27 capture areas in southern Michigan, USA, in 2000–2003. We destroyed nests of 168 radiomarked females by removing eggs after day 14 of incubation, and we left nests of 82 incubating hens undisturbed after capture and marking. Of females whose nests we experimentally destroyed, 80% subsequently migrated from breeding areas to molt remiges in Canada. Among 82 nests left undisturbed, 37 failed due to natural causes and 51% of those females departed. Migration incidence of birds that nested in urban parks was low (23%) compared with migration incidence of birds that nested in other classes of land use (87%). Departure of females from their breeding areas began during the second and third weeks of May, and most females departed during the last week of May and first week of June. Based on apparent molting locations of 227 marked geese, birds either made long-distance migratory movements >900 km, between latitudes 51° and 64° N, or they remained on breeding areas. Molting locations for 132 migratory geese indicated 4 primary destinations in Canada: Western Ungava Peninsula and offshore islands, Cape Henrietta Maria, Northeast James Bay and offshore islands, and Belcher Islands, Hudson Bay, Canada. Following molt of remiges, Canada geese began to return to their former nesting areas from 20 August through 3 September, with 37% arriving on or before 15 September and 75% arriving on or before 1 October. Migration routes of geese returning to spring breeding areas were relatively indirect compared with direct routes taken to molting sites. Although overall survival from May through November was 0.81 (95% CI: 0.74–0.88), survival of migratory geese marked on breeding sites where birds could be hunted was low (0.60; 95% CI: 0.42–0.75) compared with high survival of birds that remained resident where hunting was restricted (0.93; 95% CI: 0.84–0.97). Nest destruction can induce molt migration, increase hunting mortality of geese returning from molting areas, and reduce human–goose conflicts, but managers also should consider potential impacts of increasing numbers of molt migrants on populations of subarctic nesting Canada geese.

Genetic methods for determining racial composition of Canada goose harvests

We used molecular genetic markers and established statistical methods to estimate proportional contributions of subspecies and breeding populations to admixed wintering and migratory Canada goose (Branta canadensis) harvests. We compared harvest estimates across spatially and temporally explicit sampling intervals. We characterized 997 individuals from breeding populations in Canada representing interior Canada geese (B. c. interior; n = 4) and in Michigan representing giant Canada geese (B. c. maxima; n = 5) for 5 microsatellite loci. We determined that microsatellite loci coupled with maximum-likelihood methods provided accurate and precise proportional contribution estimates of samples from each subspecies and population. We first conducted simulation analyses and derived harvest estimates for unknown individuals representing a range of plausible harvest mixture scenarios using blind tests. Based on harvested individuals collected over a 4-year period (1993-1996), we found that the racial composition of Canada goose harvests varied significantly among years and across early, regular, and late seasons within a year. Harvest composition varied spatially between management areas in different regions and between managed and private lands in close (<40 km) geographic proximity. Higher proportions of resident giant Canada geese were harvested during early hunting seasons and on private lands relative to migratory interior Canada geese. Harvest estimates suggest that individuals from different subspecies and populations are differentially abundant or susceptible to harvest at different times of the fall season, during different years, and populations across different geographic locations. Given that baseline genetics data are available for subspecies of management interest, genetic methods can provide harvest composition estimates at many spatial and temporal scales, including enumeration of statistical confidence.

Evaluation of pheasant crowing rates as a population index

Assumptions of many call-count indexes remain untested. One assumption of the pheasant crowing count index is that crowing rates are similar among populations. We measured crowing rates of ring-necked (Phasianus colchicus torquatus), Sichuan (P. c. strauchi), and hybrid rooster pheasants from 1 hour before until 1 hour after sunrise, 1992-93. We tested hypotheses that crowing rates were unrelated to subspecies, age, year, and weather variables. A model predicting crowing rates included subspecies and time from sunrise as significant effects (R 2 = 0.41). This model described a parabola where crowing rates reach maximum at 9 minutes after sunrise, regardless of subspecies. Temperature, wind velocity, and cloud cover were not significant in explaining crowing rate variation. Recommendations for improving the crowing count survey include: conducting surveys from 30 minutes before to 30 minutes after sunrise, increasing the listening period length from 2 to 4 minutes for ring-necked pheasant populations, and adjusting for subspecific differences in crowing rates.

Survival and Recovery Rates of American Woodcock Banded in Michigan

American woodcock (Scolopax minor) population indices have declined since U.S. Fish and Wildlife Service (USFWS) monitoring began in 1968. Management to stop and/or reverse this population trend has been hampered by the lack of recent information on woodcock population parameters. Without recent information on survival rate trends, managers have had to assume that the recent declines in recruitment indices are the only parameter driving woodcock declines. Using program MARK, we estimated annual survival and recovery rates of adult and juvenile American woodcock, and estimated summer survival of local (young incapable of sustained flight) woodcock banded in Michigan between 1978 and 1998. We constructed a set of candidate models from a global model with age (local, juvenile, adult) and time (year)-dependent survival and recovery rates to no age or time-dependent survival and recovery rates. Five models were supported by the data, with all models suggesting that survival rates differed among age classes, and 4 models had survival rates that were constant over time. The fifth model suggested that juvenile and adult survival rates were linear on a logit scale over time. Survival rates averaged over likelihood-weighted model results were 0.8784 ± 0.1048 (SE) for locals, 0.2646 ± 0.0423 (SE) for juveniles, and 0.4898 ± 0.0329 (SE) for adults. Weighted average recovery rates were 0.0326 ± 0.0053 (SE) for juveniles and 0.0313 ± 0.0047 (SE) for adults. Estimated differences between our survival estimates and those from prior years were small, and our confidence around those differences was variable and uncertain. Juvenile survival rates were low.

Survival and Reproduction of Female Sichuan, Ring-Necked, and F 1 Hybrid Pheasants

Sichuan pheasants (Phasianus colchicus strauchi) were introduced in Michigan because it was believed they would exhibit higher survival rates than existing ring-necked pheasants (P. c. torquatus). We evaluated survival and reproduction of female Sichuan pheasants, ring-necked, and F 1 hybrid pheasants (P. c. strauchi × P. C. torquatus) hatched and reared in captivity and released in southern Michigan. Spring-to-fall survival probabilities of hybrids (0.275) were greater (P = 0.048) than those of ring-necked (0.116) and Sichuan females (0.106). First nests of ring-necks had more eggs than first nests of both Sichuans and hybrids. Daily nest survival rates did not differ among pheasant types. Greater survival by hybrid females resulted in seasonal production of 2.45 chicks/female, which was greater than the 1.37 chicks/female for ring-necks and 0.83 chicks/ female for Sichuans (P = 0.007). While releases of pen-reared pheasants to establish or increase wild pheasant populations are often ineffective, use of pheasants with potential for diverse gene pools may increase chances of success.

Genetic Analysis of Standardized Collections of Cackling and Canada Goose Harvests

Abstract Many states have established special harvest seasons and hunting zones to target overabundant populations of temperate-nesting Canada geese (Branta canadensis) and protect less abundant northern-nesting populations of Canada geese and cackling geese (B. hutchinsii). To meet management needs for spatially and temporally explicit harvest estimates, we utilized established methods of genetic stock identification and standardized harvest sample collections to estimate proportional contributions of cackling (B. h. hutchinsii) and Canada geese (B. c. maxima and B. c. interior) to 5 consecutive annual harvests (1998–2002) in Michigan, USA. Harvest samples (n = 2,272) were analyzed for several seasons at statewide, regional, and local spatial scales. We expanded upon previous studies that used genetic methods to monitor cackling goose and Canada goose harvests by analyzing harvests within regional and local goose management areas. Likelihood ratio tests were also employed to compare harvest composition among spatial and temporal sampling groups. Tests revealed that proportions of giant and interior Canada geese within local harvests varied significantly during the fall in different hunt zones of Michigan and during different time periods. Adaptive management of temperate-nesting and northern-nesting geese would benefit from accurate estimates of harvest composition, as provided by genetic-based methods. Quantification of changes in harvest composition as a function of variation in season opening date and duration, and bag limits can provide valuable insight into goose migratory behavior and population dynamics. Harvest composition estimates may be used to predict impacts of management prescriptions on mortality rates of specific breeding populations.

Influence of Weather, Wetland Availability, and Mallard Abundance on Productivity of Great Lakes Mallards (Anas platyrhynchos)

Waterfowl rely on breeding habitat availability for successful reproduction. Breeding habitat availability likely changes throughout the season and among years as weather patterns change and thus productivity rates are likely susceptible to these changes. We used data from 1961 to 2011 to investigate effects of weather, breeding habitat availability and abundance of breeding mallards (Anas platyrhynchos) on productivity rates of mallards breeding in the Great Lake states (Michigan, Minnesota, and Wisconsin; hereafter GLS). We hypothesized that productivity rates would increase with wetter and warmer conditions however, extreme temperatures may have a negative impact and that high breeding density may negatively impact productivity rates. Specifically, we looked at the effects of average June and July temperature and precipitation, the Palmer Hydrological Drought Index (hereafter PHDI), and wetland counts to model productivity rates across the three states for the time series. We used a reduced time series model set to evaluate the impacts of wetland counts on productivity. We found that in general, wetter conditions, as indexed by high positive PHDI values and relationships with pond abundance, positively affected productivity. We believe that breeding habitat availability is likely a reasonable predictor of mallard productivity rates in the GLS.