USGS Science⌕ Search

Geology topics

A.F. O'Connell

Publications and source records attributed to A.F. O'Connell.

At least 19 recordsLinked to original sources
USGS2011

Variation in wind and piscivorous predator fields affecting the survival of Atlantic salmon, Salmo salar, in the Gulf of Maine

Observations relevant to the North American stock complex of Atlantic salmon, Salmo salar L., suggest that marine mortality is influenced by variation in predation pressure affecting post‐smolts during the first months at sea. This hypothesis was tested for Gulf of Maine (GOM) stocks by examining wind pseudostress and the distribution of piscivorous predator fields potentially affecting post‐smolts. Marine survival has declined over recent decades with a change in the direction of spring winds, which is likely extending the migration of post‐smolts by favouring routes using the western GOM. In addition to changes in spring wind patterns, higher spring sea surface temperatures have been associated with shifting distributions of a range of fish species. The abundance of several pelagic piscivores, which based on their feeding habits may predate on salmon post‐smolts, has increased in the areas that serve as migration corridors for post‐smolts. In particular, populations of silver hake, Merluccius bilinearis (Mitchell), red hake, Urophycis chuss (Walbaum), and spiny dogfish, Squalus acanthias L., increased in size in the portion of the GOM used by post‐smolts. Climate variation and shifting predator distributions in the GOM are consistent with the predator hypothesis of recruitment control suggested for the stock complex.

Gulf Of Maine↗
USGS2007

Deer exposed to exceptionally high concentrations of lead near the Continental Mine in Idaho, USA

Habitat surrounding the inactive Continental Mine in northern Idaho, USA, supports bear ( Ursus arctos, Ursus americanus ), moose ( Alces alces ), elk ( Cervus elaphus ), woodland caribou ( Rangifer tarandus caribou ), and abundant mule ( Odocoileus hemionus ) and white-tailed deer ( Odocoileus virginianus ). Tailings on the mining site were capped and remediated in 2003 to reduce environmental exposure of surrounding soil and sediments of Blue Joe Creek, downslope of the mine. Before capping, the mean Pb concentration in deer pellets collected on-site was 920 mg/kg of Pb (dry wt). This exposure, if chronic, would be comparable to an exposure that could be lethal to cattle or horses. Surprisingly, the mean pellet Pb concentration of 950 mg/kg in 2004 was as high as it was before remediation, and it was related to a high rate of soil ingestion. Mean soil content of the pellets collected from the capped site in 2004 was 22% dry weight, estimated from the acid-insoluble ash, a marker of soil ingestion. Clumps of sand and bits of rock were observed inside some of the pellets, and Pb concentrations in the pellets were correlated (p < 0.05) with soil content. Although terrestrial risk assessments generally estimate exposure from diets and from incidentally ingested soil, the deer at this site were directly ingesting contaminated soil or mining waste. The mean Pb concentration of this ingested soil was estimated as 6,700 mg/kg and the maximum as 25,000 mg/kg, well above the Pb concentrations measured in the remediated cap. The deer seemed to be ingesting soil or mining waste from one or more small but highly contaminated sources located beyond the remediated cap.

Idaho↗
USGS2007

Design and implementation of estimation-based monitoring programs for flora and fauna: A case study on the Cherokee National Forest

Science-based monitoring of biological resources is important for a greater understanding of ecological systems and for assessment of the target population using theoretic-based management approaches. When selecting variables to monitor, managers first need to carefully consider their objectives, the geographic and temporal scale at which they will operate, and the effort needed to implement the program. Generally, monitoring can be divided into two categories: index and inferential. Although index monitoring is usually easier to implement, analysis of index data requires strong assumptions about consistency in detection rates over time and space, and parameters are often biasednot accounting for detectability and spatial variation. In most cases, individuals are not always available for detection during sampling periods, and the entire area of interest cannot be sampled. Conversely, inferential monitoring is more rigorous because it is based on nearly unbiased estimators of spatial distribution. Thus, we recommend that detectability and spatial variation be considered for all monitoring programs that intend to make inferences about the target population or the area of interest. Application of these techniques is especially important for the monitoring of Threatened and Endangered (T&E) species because it is critical to determine if population size is increasing or decreasing with some level of certainty. Use of estimation-based methods and probability sampling will reduce many of the biases inherently associated with index data and provide meaningful information with respect to changes that occur in target populations. We incorporated inferential monitoring into protocols for T&E species spanning a wide range of taxa on the Cherokee National Forest in the Southern Appalachian Mountains. We review the various approaches employed for different taxa and discuss design issues, sampling strategies, data analysis, and the details of estimating detectability using site occupancy. These techniques provide a science-based approach for monitoring and can be of value to all resource managers responsible for management of T&E species.

Proceedings of the Annual Conference of the Southe↗
USGS2004

Contribution of natural history collection data to biodiversity assessment in national parks

There has been mounting interest in the use of museum and herbaria collections to assess biodiversity; information is often difficult to locate and access, however, and few recommendations are available for effectively using natural history collections. As part of an effort to inventory vertebrates and vascular plants in U.S. national parks, we searched manually and by computer for specimens originating within or adjacent to 14 parks throughout the northeastern United States. We compared the number of specimens located to collection size to determine whether there was any effect on detection rate of specimens. We evaluated the importance of park characteristics (e.g., age since establishment, size, theme [natural vs. cultural]) for influencing the number of specimens found in a collection. We located >31,000 specimens and compiled associated records (hereafter referred to as specimens) from 78 collections; >9000 specimens were park-significant, originating either within park boundaries or in the local township where the park was located. We found >2000 specimens by means of manual searches, which cost $0.001–0.15 per specimen searched and $0.81–151.95 per specimen found. Collection effort appeared relatively uniform between 1890 and 1980, with low periods corresponding to significant sociopolitical events. Detection rates for specimens were inversely related to collection size. Although specimens were most often located in collections within the region of interest, specimens can be found anywhere, particularly in large collections international in scope, suggesting that global searches will be necessary to evaluate historical biodiversity. Park characteristics indicated that more collecting effort occurred within or adjacent to larger parks established for natural resources than in smaller historical sites. Because many institutions have not yet established electronic databases for collections, manual searches can be useful for retrieving specimens. Our results show that thorough, systematic searching of natural history collections for park-significant specimens can provide a historical perspective on biodiversity for park managers.

Conservation Biology↗
USGS2002

Evaluation of genetic variability in a small, insular population of spruce grouse

Using microsatellite markers we determined genetic variability for two populations of spruce grouse in eastern North America, one on a coastal Maine island where breeding habitat is limited and highly fragmented, the other in central New Brunswick (NB), where suitable breeding habitat is generally contiguous across the region. We examined six markers for both populations and all were polymorphic. Although the number of alleles per locus and the proportion of unique alleles were lower in the island population, and probably a result of small sample.size, heterozygosity and a breeding coefficient (Fis) indicated slightly more variability in the island population. Deviation from Hardy-Weinberg equilibrium also was more evident in loci for the mainland population. Several traits previously documented in the island population: relatively long natal dispersal distances, reproductive success, territoriality, adult survival, and longevity support the maintenance of hetrerzygosity, at least in the short-term. Sample collection from two small (500 ha), separate areas in NB, and the predicted importance of immigration density to supplement this population demonstrate the need for behavioral and ecological information when interpreting genetic variation. We discuss the relevance of these issues with respect to genetic variability and viability.

Book chapter↗
USGS2000

[Book review] Ecological Scale - Theory and Applications by D.L. Peterson and V. Thomas Parker, editors

Covering the complexity of the scale topic, this volume represents an important compilation of information on a topic that is often misunderstood, and one for which little attention is paid (although, thankfully, this seems to be on the decline). Although technical, this book provides full exposure to the scale issue in ecology and is an important reference for researchers and resource managers who are working to understand and preserve ecological function in parks. The authors and editors have combined to provide a needed examination of a very important topic. In summary, why purchase all those books on scale when just one will do? The mountain of information alone stuffed into this one book should prompt all parks to get a copy and have it on hand as a quick and ready reference.

Park Science↗
USGS1998

Mortality and survival of white-tailed deer Odocoileus virginianus fawns on a north Atlantic coastal island

Mortality and survival of white-tailed deer Odocoileus virginianus fawns (N = 29) were studied from birth to one year of age during 1991–95 on Mount Desert Island (MDI), Maine, where deer hunting is prohibited, coyotes Canis latrans have become recently established, and protected U.S. National Park lands are interspersed with private property. The rate of predator-caused mortality was 0.52, with coyote predation (N = 8) accounting for at least 47% of mortalities from all causes (N = 17). Mortality rate from drowning was 0.24 (N = 3), and from vehicles 0.14 (N = 3). For fawns radio-collared as neonates, 10 of 14 mortalities occurred during the first two months of life. Annual rate of fawn survival was 0.26. Survival rate from six months to one year was 0.65 and four mortalities (two predation, two drowning) were observed during this interval. A subgroup of fawns (N = 11) captured near a residential area and along the periphery of a coyote territory had a higher rate of survival to one year of age (S = 0.67) than did fawns from all other areas (N = 18, S = 0.00). Recruitment to one year of age was lower than that observed in other deer populations in the northeastern United States. Low recruitment associated with coyote predation and mortality sources associated with humans appears to be limiting white-tailed deer populations in some segments of this insular landscape.

Maine↗
USGS1996

Patch occupancy and dispersal of spruce grouse on the edge of its range in Maine

We surveyed 18 habitat patches (black spruce (Picea marinana) - tamarack (Larix larcina) wetlands) for spruce grouse (Dendragapus canadensis canadensis) on Mount Desert Island, Maine, during April-May in 1992 and 1993 to determine patch occupancy relative to patch area. We also equipped nine juvenile grouse with radio transmitters to determine movement and habitat use outside of patches during autumn dispersal. The 2 large patches (77 and 269 ha), 5 of 6 medium-sized (11-26 ha) patches, and 1 of 10 small (4-8 ha) patches were occupied. Spruce grouse occupied smaller habitat patches than previously reported, and occupied patches were closer (P < 0.05) to the nearest occupied patch (x = 1.2 km) than were unoccupied patches (x = 2.5 km). Eight of nine juvenile grouse left their natal habitat patch during autumn dispersal, and net dispersal distance (x = 2.3 km) was greater than that reported for grouse in areas with more contiguous habitat. Dispersing juveniles used all major forest types and 33 % of relocations were in deciduous forest. Thus, deciduous forest was not an absolute dispersal barrier.

Maine↗
USGS1996

Productivity of the spruce grouse in fragmented habitat at the edge of its range

We measured productivity of the Spruce Grouse (Dendragapusc anadensicsa nadensis) in patchy black spruce (Picea mariana) habitat along the southeastern limit of its range in mid-coastal Maine. We captured grouse and attached necklace-mounted radio transmitters to hens prior to nesting. Of 19 females monitored, only 26% raised chicks to the late brood-rearing period. Predation was high on hens (37%) and five were killed before hatching eggs. Six (55%) entire broods were lost and only 30% of chicks survived to late summer. Production ( No. of chicks/female), an index of productivity, was < 1 and lower in Maine and Minnesota study areas in black spruce than areas dominated by jack pine (Pinus banksiana) or a mixture of jack pine and spruce with dense undergrowth. Where Spruce Grouse breed in patchy black spruce communities, immigration from neighboring populations or inter-patch movement by local individuals may be required to maintain viable populations.

Journal of Field Ornithology↗