SEARCH · Search USGS

Results for “SELECT Journal”

Search indexed USGS publications on groundwater, aquifers, geologic maps, mineral resources and earthquakes. Explore source records by subject and place.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 919 records · Page 51Linked to original sources

USGS2019

Multi‐scale habitat selection of elk in response to beetle‐killed forest

Forests of the Rocky Mountains (USA and Canada) have experienced a large‐scale bark‐beetle ( Dendoctronus ponderosae ) epidemic that has led to widespread mortality of pine trees, followed by structural and compositional changes to the forest. The millions of dead trees resulting from this event likely have an effect on ecosystem processes, but currently those effects are mostly unclear. Changes to forest canopy and understory structure are likely to affect behavior and space use of large ungulates as forage abundance, thermal cover, and locomotive costs are presumably altered. We developed and tested hypotheses for how resource selection in summer of female elk ( Cervus canadensis ; n = 47; 2012–2016), the primary large mammal in our study area, was influenced by changes in canopy, understory vegetation, and downed logs in bark‐beetle affected forest in south‐central Wyoming, USA. We employed global positioning system (GPS) technology, an imagery‐derived land cover classification specifically depicting beetle‐affected forest, and on‐the‐ground forest measurements to develop resource and step selection functions. At the study area scale, elk avoided beetle‐killed forest during nearly all parts of the day and selected for intact conifer forest during the day. At the micro‐habitat scale, as canopy cover decreased in beetle‐killed areas, there was a concomitant increase in grass biomass and downed logs. Nevertheless, while in the forest, elk did not alter resource selection relative to changes in understory vegetation or downed logs. The boost in forage abundance within beetle‐killed areas was insufficient to overcome the presumed increases in energy expenditures associated with locomotion and thermoregulation in beetle‐killed forests. The bark‐beetle epidemic has altered how elk use the landscape and has resulted in a potential loss of forest habitat that elk use during the day. Our results indicate that habitat treatments (i.e., fire or harvest) that remove standing dead trees and downed logs in beetle‐killed forest would further facilitate understory growth while reducing the locomotion costs associated with foraging in beetle‐killed forest.

Wyoming, Colorado↗

USGS2011

Challenges of using electrical resistivity method to locate karst conduits-A field case in the Inner Bluegrass Region, Kentucky

Conduits serve as major pathways for groundwater flow in karst aquifers. Locating them from the surface, however, is one of the most challenging tasks in karst research. Geophysical methods are often deployed to help locate voids by mapping variations of physical properties of the subsurface. Conduits can cause significant contrasts of some physical properties that can be detected; other subsurface features such as water-bearing fractures often yield similar contrasts, which are difficult to distinguish from the effects of the conduits. This study used electrical resistivity method to search for an unmapped karst conduit that recharges Royal Spring in the Inner Bluegrass karst region, Kentucky, USA. Three types of resistivity techniques (surface 2D survey, quasi-3D survey, and time-lapse survey) were used to map and characterize resistivity anomalies. Some of the major anomalies were selected as drilling targets to verify the existence of the conduits. Drilling near an anomaly identified by an electrical resistivity profile resulted in successful penetration of a major water-filled conduit. The drilling results also suggest that, in this study area, low resistivity anomalies in general are associated with water-bearing features. However, differences in the anomaly signals between the water-filled conduit and other water-bearing features such as water-filled fracture zones were undistinguishable. The electrical resistivity method is useful in conduit detection by providing potential drilling targets. Knowledge of geology and hydrogeology about the site and professional judgment also played important roles in locating the major conduit. ?? 2011 Elsevier B.V.

Journal of Applied Geophysics↗

USGS1982

Age-class determination of canvasbacks

A technique was developed to distinguish yearling from adult canvasbacks ( Aythya valisineria ) in the field during spring on the basis of white flecking on the distal ends of selected wing feathers. Covert feathers from adults had well-defined vermiculation patterns whereas feathers from yearlings lacked such markings. These age-related characters were confirmed by paired comparisons of feathers from the same captive birds in consecutive years and by discriminant analysis of feathers using densitometric measurements from known-age wild birds. Reflective densitometric measurements of greater secondary coverts for females and males were significantly different between 1- and 2-year-old and between 2- and 3-year-old canvasbacks. Greater secondary coverts were the best feathers for recognizing age-classes of males and females. Densitometric values indicate low variability among different observers and within samples.

Journal of Wildlife Management↗

USGS2011

Resource selection by black-footed ferrets in South Dakota and Montana

The black-footed ferret (Mustela nigripes), once extinct in the wild, remains one of the most critically endangered mammals in North America despite 18 years of reintroduction attempts. Because black-footed ferrets are specialized predators of prairie dogs (Cynomys sp.), a better understanding of how black-footed ferrets select resources might provide insight into how best to identify and manage reintroduction sites. We monitored ferret resource selection at two reintroduction sites with different densities of prairie dog populations-one that contained a high density of prairie dogs (Conata Basin, South Dakota) and one that was lower (UL Bend, Montana). We evaluated support for hypotheses about ferret resource selection as related to the distribution of active burrows used by black-tailed prairie dogs (Cynomys ludovicianus), interactions between ferrets, and habitat edge effects. We found support for all three factors within both populations; however, they affected ferret resource selection differently at each site. Ferrets at Conata Basin tended to select areas with high prairie dog burrow density, closer to the colony edge, and that overlapped other ferret ranges. In contrast, ferrets at UL Bend tended not to select areas of high active prairie dog burrow density, avoided areas close to edge habitat, and females avoided areas occupied by other ferrets. The differences observed between the two sites might be best explained by prairie dog densities, which were higher at Conata Basin (119.3 active burrows per ha) than at UL Bend (44.4 active burrows per ha). Given the positive growth of ferret populations at Conata Basin, management that increases the density of prairie dogs might enhance ferret success within natural areas. To achieve long-term recovery of ferrets in the wild, conservationists should increasingly work across and outside natural area boundaries to increase prairie dog populations.

Natural Areas Journal↗

USGS1992

Evaluation of the effects of candidate molluscicides on two nontarget bivalves

A variety of molluscicides have been proposed for use in control of zebra mussels (Dreissena polymorpha), but their effect on nontarget aquatic organisms has not been evaluated. Standard methods were adapted for assessing the toxicity of candidate molluscicides to two nontarget bivalves. Fingernail clams, Musculium transversum, and the fawnfoot mussel, Truncilla donaciformis, were selected to represent the two families of native bivalves. Test organisms were collected from pools 6 to 9 of the Upper Mississippi River near La Crosse, WI. Static acute toxicity tests were conducted for 48 hours followed by a 96-hour monitoring period in untreated water to more fully assess survival and mortality. Toxicity data were analyzed by probit analysis to give LC sub(50) values and 95% confidence limits. The same chemicals as those tested at Ohio State University were evaluated against zebra mussels. Results from these studies and those conducted at Ohio State University will be used to evaluate the effectiveness of chemicals in zebra mussel control and their potential hazard to nontarget organisms.

Journal of Shellfish Research↗

USGS2001

Ground-nesting waterbirds and mammalian carnivores in the Virginia barrier island region: Running out of options

We examined changing patterns of distribution of two large mammalian predators, the raccoon (Procyon lotor) and red fox (Vulpes vulpes), and beach-nesting terns and Black Skimmers (Rynchops niger) along ca. 80 km of the Virginia barrier island landscape between the periods 1975-1977 and 1998. Based on evidence from trapping, scent stations, den observations and sightings of the two predators, there has been a marked increase in their island ranges. In 1975-77, only 6 of the 11 surveyed barrier islands definitely harbored at least one of the two mammals, but by 1998, 11 of 14 islands showed evidence of one or both during the spring and summer. Concurrently, annual beach-nesting bird surveys have been conducted since the mid 1970s during June. From 1977 to 1998, the number of colonies of terns [Common (Sterna hirundo), Gull-billed (S. nilotica), Least (S. antillarum), Royal (S. maxima), and Sandwich (S. sandvicensis)] and Black Skimmers declined from 23 colonies on 11 barrier islands to 13 colonies on 10 islands. In addition, the populations decreased dramatically for all species except the marginal Sandwich Tern and Least Tern. This pattern suggests that mammalian predation may be a major factor in colony site selection or success, although we have no data on success at most locations. The only consistently large colony over the years has been the Royal Tern colony on Fisherman Island, one of the few with no resident large mammals. Because these declining waterbirds appear to be running out of options for safe colony sites in coastal Virginia, we discuss the prospects of conducting limited predator removals on certain islands. In addition, considerations of strict management and enforcement of protection at critical manmade colony sites that now attract large numbers of certain species, are timely. Lastly, where dredged material disposal projects are planned, providing nesting sites for these colonial species and roosting sites for migrant birds may be appropriate.

Journal of Coastal Research↗

USGS2016

Validation of a side-scan sonar method for quantifying walleye spawning habitat availability in the littoral zone of northern Wisconsin Lakes

Previous research has generally ignored the potential effects of spawning habitat availability and quality on recruitment of Walleye Sander vitreus , largely because information on spawning habitat is lacking for many lakes. Furthermore, traditional transect-based methods used to describe habitat are time and labor intensive. Our objectives were to determine if side-scan sonar could be used to accurately classify Walleye spawning habitat in the nearshore littoral zone and provide lakewide estimates of spawning habitat availability similar to estimates obtained from a transect–quadrat-based method. Based on assessments completed on 16 northern Wisconsin lakes, interpretation of side-scan sonar images resulted in correct identification of substrate size-class for 93% (177 of 191) of selected locations and all incorrect classifications were within ± 1 class of the correct substrate size-class. Gravel, cobble, and rubble substrates were incorrectly identified from side-scan images in only two instances (1% misclassification), suggesting that side-scan sonar can be used to accurately identify preferred Walleye spawning substrates. Additionally, we detected no significant differences in estimates of lakewide littoral zone substrate compositions estimated using side-scan sonar and a traditional transect–quadrat-based method. Our results indicate that side-scan sonar offers a practical, accurate, and efficient technique for assessing substrate composition and quantifying potential Walleye spawning habitat in the nearshore littoral zone of north temperate lakes.

Wisconsin↗

USGS2019

Inversion of airborne EM data with an explicit choice of prior model

Inversion of airborne electromagnetic (AEM) data is an under-determined inverse problem, in that infinitely many resistivity models exist that will be able to explain the observed data, within measurement errors. Therefore, additional information or constraints must be taken into account to solve the inverse problem. In deterministic approaches, the goal is to locate one optimal model that can be obtained by using some form of smoothness constraints implied through a number of regularization choices. This model, however, will not necessarily represent realistic geological features. Probabilistic methods offer an alternative in which the solution is not one model, but a collection of models, whose variability represents the uncertainty. The probabilistic approach can also rely on implicit model assumptions, representing prior information (a type of regularization information) that may or may not be consistent with the actual available information. Here, we present an approach for AEM inversion in which the prior model is explicitly chosen by a user, preferably selected based on actual prior information available and then integrated with AEM data using a general Monte Carlo based sampling approach. This approach leads to a new workflow to AEM inversion in which geological prior information is independently and explicitly chosen before inversion is carried out. The main benefit of this approach is that each model obtained will, by construction, be consistent with prior (geological) information as well as geophysical data. Through examples based on synthetic and real AEM data, we will demonstrate the methodology, not least that the choice of prior information cannot be avoided: Either it is done explicitly, or it will be chosen implicitly by the choice of method used to invert the AEM data.

Geophysical Journal International↗

USGS2014

Evidence against a Pleistocene desert refugium in the Lower Colorado River Basin

Aim The absence of Sonoran Desert plants in late Pleistocene-aged packrat middens has led to speculation that they survived glacial episodes either in refugia as intact associations (Clementsian community concept) or in dry microsites within chaparral or woodland according to individualistic species responses (Gleasonian community concept). To test these hypotheses, we developed a midden record from one likely refugium in north-eastern Baja California, Mexico. We also measured stomatal guard cell size in fossil leaves to further evaluate site-level individualistic responses of Larrea tridentata (creosote bush) ploidy races to climatic changes, including monsoonal history, over the late Quaternary. Location Sierra Juárez, Lower Colorado River Basin, north-eastern Baja California, Mexico. Methods Packrat ( Neotoma ) middens were collected from <300 m elevation on the eastern piedmont of the Sierra Juárez. Plant macrofossils and pollen were analysed from 50 dated middens, including determination of Larrea tridentata ploidy races. Results Pleistocene middens dating back to >55,000 cal. yr BP contained a mix of extralocal species characteristic of chaparral and pinyon–juniper–oak woodland, along with some modern desert elements. Many other desert taxa were absent during the Pleistocene, although most had arrived by the beginning of the Holocene 11,700 years ago. Main conclusions The assemblage of chaparral, woodland and select desert elements refutes the hypothesis that the Lower Colorado River Basin served as a late Pleistocene refugium for Sonoran Desert flora. The rapid arrival of most missing desert species by the early Holocene suggests they did not have far to migrate. They probably survived the last glacial period as smaller, disparate populations in dry microsites within chaparral and pinyon–juniper–oak woodlands. Diploid and tetraploid races of Larrea tridentata were present during the Pleistocene, but hexaploids did not appear until the mid-Holocene. This demonstrates that individualistic responses to climate involved genetic variants, in this case cytotypes, and not just species.

Baja California↗

USGS2004

Use and interpretation of logistic regression in habitat-selection studies

Logistic regression is an important tool for wildlife habitat-selection studies, but the method frequently has been misapplied due to an inadequate understanding of the logistic model, its interpretation, and the influence of sampling design. To promote better use of this method, we review its application and interpretation under 3 sampling designs: random, case-control, and use-availability. Logistic regression is appropriate for habitat use-nonuse studies employing random sampling and can be used to directly model the conditional probability of use in such cases. Logistic regression also is appropriate for studies employing case-control sampling designs, but careful attention is required to interpret results correctly. Unless bias can be estimated or probability of use is small for all habitats, results of case-control studies should be interpreted as odds ratios, rather than probability of use or relative probability of use. When data are gathered under a use-availability design, logistic regression can be used to estimate approximate odds ratios if probability of use is small, at least on average. More generally, however, logistic regression is inappropriate for modeling habitat selection in use-availability studies. In particular, using logistic regression to fit the exponential model of Manly et al. (2002:100) does not guarantee maximum-likelihood estimates, valid probabilities, or valid likelihoods. We show that the resource selection function (RSF) commonly used for the exponential model is proportional to a logistic discriminant function. Thus, it may be used to rank habitats with respect to probability of use and to identify important habitat characteristics or their surrogates, but it is not guaranteed to be proportional to probability of use. Other problems associated with the exponential model also are discussed. We describe an alternative model based on Lancaster and Imbens (1996) that offers a method for estimating conditional probability of use in use-availability studies. Although promising, this model fails to converge to a unique solution in some important situations. Further work is needed to obtain a robust method that is broadly applicable to use-availability studies.

Journal of Wildlife Management↗

USGS2010

Repeated use of an abandoned vehicle by nesting Turkey vultures ( Cathartes aura )

Turkey Vultures ( Cathartes aura ) lay their eggs on an existing substrate in the dark recesses of a variety of natural sites ( Kirk and Mossman 1998 ). Although an important requirement of Turkey Vulture nest-site selection is isolation from human disturbances ( Kirk and Mossman 1998 ), their nests have been reported in abandoned buildings since at least the early 1800s ( Nuttall 1832 ). Depopulation of rural areas in North America in recent decades has resulted in many abandoned buildings within the Turkey Vulture's breeding range ( Peck 2003 ). Increased use of abandoned buildings by nesting Turkey Vultures has been implicated in the species' recent northward range expansion ( Peck 2003 , Nelson et al. 2005 , Houston et al. 2007 ). Although abandoned or inoperative vehicles also are widespread in rural areas, we found no published literature documenting Turkey Vultures' use of these potential nest sites. Herein, we summarize the first documented incidence of a Turkey Vulture nesting in an abandoned vehicle.

South Dakota↗

USGS1998

Nesting habitat selection by sage grouse in south-central Washington

To characterize western sage grouse ( Centrocercus urophasianus phaios Bonaparte) nesting habitat in sagebrush-steppe habitat in Washington, we initiated a study on the Yakima Training Center to determine nesting habitat characteristics and whether these characteristics differed between successful and depredated nests. Most nests (71%) were in big sagebrush ( Artemesia tridentata Nutt.)/bunchgrass communities. Nest habitat was characterized by greater shrub cover, shrub height, vertical cover height, residual cover, and litter than at random locations. Successful 1-m 2 nest sites within big sagebrush/bunchgrass in 1992 had less shrub cover (51%) and shrub height (64 cm) than depredated nest sites (70% and 90 cm, respectively). Successful 77-m 2 nest areas in big sagebrush/bunchgrass in 1993 had more tall grass ( ≥ 18 cm) than depredated nest areas. Management that protects the big sagebrush/bunchgrass community is essential for maintaining nesting habitat for sage grouse.

Journal of Range Management↗

USGS2012

American hydrogeology at the millennium: An annotated chronology of 100 most influential papers

Hydrogeology developed as scientists undertook activities to describe how a groundwater system functions to explain why it is that way, in order to solve practical problems of water supply. This paper demonstrates the evolutionary nature and growth of hydrogeology in the United States on the basis of a selection of one hundred papers that had a significant impact on subsequent activities. We have identified three revolutionary concepts that resulted directly from this evolutionary understanding and have selected papers that demonstrate important consequences. These three concepts are 1) that the mathematical expression for heat flow can be paraphrased for groundwater and used in transient flow conditions to determine aquifer characteristics; 2) that the distribution of fluid potential can be formulated in mathematical equations suitable for solution by various analytical techniques; and 3) that chemical thermodynamics can be applied to hydrogeologic systems in order to understand the processes controlling the chemical character of groundwater. One purpose of this paper is to encourage scientists to gain an additional dimension of satisfaction from their work by being aware of the contributions of those who went before them and to see how their own work fits into the current understanding of hydrogeology.

Hydrogeology Journal↗

USGS2007

Nutrient concentrations and yields in undeveloped stream basins of the United States

Data from 85 sites across the United States were used to estimate concentrations and yields of selected nutrients in streams draining relatively undeveloped basins. Flow-weighted concentrations during 1990-1995 were generally low with median basin concentrations of 0.020, 0.087, 0.26, 0.010, and 0.022 milligrams per liter (mg/L) for ammonia as N, nitrate as N, total nitrogen, orthophosphate as P, and total phosphorus, respectively. The flow-weighted concentration of nitrate exceeded 0.6 mg/L in only three basins. Total nitrogen exceeded 1 mg/L in only four basins, and total phosphorus exceeded 0.1 mg/L in only four basins. The median annual basin yield of ammonia as N, nitrate as N, total nitrogen, orthophosphate as P, and total phosphorus was 8.1, 26, 86, 2.8, and 8.5 kilograms per square kilometer, respectively. Concentrations and yields of nitrate tended to be highest in northeastern and mid-Atlantic coastal states and correlated well with areas of high atmospheric nitrogen deposition. Concentrations and yields of total nitrogen were highest in the southeastern part of the nation and in parts of the upper Midwest. In the northeast, nitrate was generally the predominant form of nitrogen, and in the southeast and parts of the upper Midwest, organic nitrogen was the dominant form. Concentrations of total phosphorus were generally highest in the Rocky Mountain and Central Plain states.Data from 85 sites across the United States were used to estimate concentrations and yields of selected nutrients in streams draining relatively undeveloped basins. Flow-weighted concentrations during 1990-1995 were generally low with median basin concentrations of 0.020, 0.087, 0.26, 0.010, and 0.022 milligrams per liter (mg/L) for ammonia as N, nitrate as N, total nitrogen, orthophosphate as P, and total phosphorus, respectively. The flow-weighted concentration of nitrate exceeded 0.6 mg/L in only three basins, Total nitrogen exceeded 1 mg/L in only four basins, and total phosphorus exceeded 0.1 mg/L in only four basins. The median annual basin yield of ammonia as N, nitrate as N, total nitrogen, orthophosphate as P, and total phosphorus was 8.1, 26, 86, 2.8, and 8.5 kilograms per square kilometer, respectively. Concentrations and yields of nitrate tended to be highest in northeastern and mid-Atlantic coastal states and correlated well with areas of high atmospheric nitrogen deposition. Concentrations and yields of total nitrogen were highest in the southeastern part of the nation and in parts of the upper Midwest. In the northeast, nitrate was generally the predominant form of nitrogen, and in the southeast and parts of the upper Midwest, organic nitrogen was the dominant form. Concentrations of total phosphorus were generally highest in the Rocky Mountain and Central Plain states.Data collected across the US from 85 streams draining relatively undeveloped basins were used to identify broad regional and national patterns in nutrient concentrations and yields. The basins of interest were selected from three USGS programs: the Hydrologic Benchmark Network, the National Water-Quality Assessment, and the Research Program. Water samples from most basins were collected on a weekly to bimonthly schedule. While the flow-weighted concentrations of nutrients varied, concentrations were low in most basins. Median flow-weighted concentrations of ammonia, nitrate, total nitrogen, orthophosphate, and total phosphorus were 0.020, 0.087, 0.26, 0.010, and 0.022 mg/l, respectively. Nitrate concentrations tended to be highest in the northeastern US, while southeastern and north-central basins had some of the highest NH3 concentrations. Flow-weighted concentrations of total P were generally highest in the Rocky Mountain and Central Plain states and in parts of the southeast.

Journal of the American Water Resources Associatio↗

USGS1980

Opaque minerals as aids in distinguishing between source and sorting effects on beach sand mineralogy in southwestern Oregon

Both source area and wave sorting affect the heavy-mineral distributions of beach sands. Although source and sorting effects are difficult to distinguish, they can be separated on four Oregon beaches by studying the magnetic fraction of the sand. Prominent headlands bound the four beaches, which are located on the southwestern Oregon coast. On three beaches the percentage of magnetite in the sand from the upper swash zone consistently increases toward the north end of each beach, apparently owing to selective sorting during littoral transport. However, the percentages of chromium and titanium in the magnetite are generally independent of the relative location of the samples along the beach and therefore independent of sorting effects. Each beach appears to be characterized by a fairly distinct range of Ti/Cr in the magnetic fraction, and the range differs from beach to beach. The magnetite percentage and the Ti/Cr range suggest that sorting and source effects can be distinguished on beaches rich in "black" sands, and that sand transport around headlands in this area is not extensive enough to mask source differences. The results of this study suggest that sorting and source effects can be distinguished wherever there are chemical or mineralogical differences in a restricted density fraction.

Journal of Sedimentary Petrology↗

USGS1993

Plasmid-mediated romet resistance of Edwardsiella ictaluri

Romet®, a potentiated sulfa drug composed of five parts sulfadimethoxine and one part ormetoprim, is used to treat channel catfish Ictalurus punctatus infected with Edwardsiella ictaluri , the causal agent of enteric septicemia of catfish (ESC). Recently, several Romet-resistant isolates of E. ictaluri were isolated from channel catfish that had died from ESC in Virginia and Mississippi. This antimicrobial resistance was determined to be plasmid-encoded, as shown by agarose gel electrophoresis of plasmid DNA, and by the fact that the plasmid and associated antimicrobial resistance could be transferred to a recipient by single-step conjugation. Size of the R plasmid was approximately 55 kilobase pairs. Plasmids also conferred resistance to tetracycline, oxytetracycline, streptomycin, trimethoprim, and SXT (another potentiated sulfonamide, composed of trimethoprim and sulfamethoxazole). Each Romet-resistant E. ictaluri isolate was mated with a plasmid recipient, the Romet-sensitive Escherichia coli isolate 1932. Transconjugates were selected by plating on Mueller-Hinton agar that contained antimicrobials, and the mean transfer frequency was 1.483 × 10 −3 transconjugates per donor cell in the mating mixture. Each E. coli transconjugate was then mated with 20 Romet-sensitive isolates of E. ictaluri to assess the propensity of these E. ictaluri isolates to accept the R plasmid. The average transfer frequency when E. ictaluri isolates of channel catfish origin (N = 14) were used as recipients was 1.197 × 10 −2 transconjugates per donor cell in the mating mixture; mean transfer frequency when E. ictaluri isolates not originating from channel catfish were used was 1.019 × 10 -3 transconjugates per donor cell, which was significantly less (P = 0.0002).

Journal of Aquatic Animal Health↗

USGS2010

Landscape correlates along mourning dove call-count routes in Mississippi

Mourning dove (Zenaida macroura) call-count surveys in Mississippi, USA, suggest declining populations. We used available mourning dove call-count data to evaluate long-term mourning dove habitat relationships. Dove routes were located in the Mississippi Alluvial Valley, Deep Loess Province, Mid Coastal Plain, and Hilly Coastal Plain physiographic provinces of Mississippi. We also included routes in the Blackbelt Prairie region of Mississippi and Alabama, USA. We characterized landscape structure and composition within 1.64-km buffers around 10 selected mourning dove call-count routes during 3 time periods. Habitat classes included agriculture, forest, urban, regeneration stands, wetland, and woodlot. We used Akaike's Information Criterion to select the best candidate model. We selected a model containing percent agriculture and edge density that contained approximately 40% of the total variability in the data set. Percent agriculture was positively correlated with relative dove abundance. Interestingly, we found a negative relationship between edge density and dove abundance. Researchers should conduct future research on dove nesting patterns in Mississippi and threshold levels of edge necessary to maximize dove density. During the last 20 years, Mississippi lost more than 800,000 ha of cropland while forest cover represented largely by pine (Pinus taeda) plantations increased by more than 364,000 ha. Our results suggest observed localized declines in mourning dove abundance in Mississippi may be related to the documented conversion of agricultural lands to pine plantations.

Journal of Wildlife Management↗

USGS2009

On selecting a prior for the precision parameter of Dirichlet process mixture models

In hierarchical mixture models the Dirichlet process is used to specify latent patterns of heterogeneity, particularly when the distribution of latent parameters is thought to be clustered (multimodal). The parameters of a Dirichlet process include a precision parameter ?? and a base probability measure G 0 . In problems where ?? is unknown and must be estimated, inferences about the level of clustering can be sensitive to the choice of prior assumed for ??. In this paper an approach is developed for computing a prior for the precision parameter ?? that can be used in the presence or absence of prior information about the level of clustering. This approach is illustrated in an analysis of counts of stream fishes. The results of this fully Bayesian analysis are compared with an empirical Bayes analysis of the same data and with a Bayesian analysis based on an alternative commonly used prior.

Journal of Statistical Planning and Inference↗