Statistics as viewed by biologists

A survey of colleagues and recent periodicals was used to characterize how biologists use statistics, with an emphasis on applications in ecology and evolutionary biology. Biologists use statistics to describe and uncover patterns in data and to ensure that they are obtaining reliable knowledge. Universally, sampling and study design problems plague biological research programs, and this continues to be one of the greatest needs for statistical training. Biologists traditionally have used a Neyman-Pearson approach to hypothesis testing, combined with an overbearing focus on the Fisherian notion of statistical significance and P-values. Multivariate statistics became popular 20 years ago but were used less following the publication of a few critical reviews that questioned the validity of statistical inferences. Advances in nonlinear dynamics that grew from computer technology have stimulated new methods for time-series analysis capable of capturing underlying structures in temporal data. Similarly, computer technology supporting geographical information systems (GIS) has led to widespread application and development of spatial statistics. Recently, Bayesian statistics have found support in a few applications, but it is my impression that most biologists are either suspicious of Bayesian methods, do not understand them, or have a difficult time implementing them. Computer-intensivesimulation methods are commonly used for inference in volving complex models. Biologists are recognizing that statistics have applications beyond hypothesis testing, and we are seeing increased use of likelihood approaches, such as the Akaike information criteria, for model selection. Repeatedly, we find that technological advances in computing power have revolutionized how biologists use statistics.     

Effects of conventional and grass-feeding systems on the nutrient composition of beef

The objectives of this study were to determine the nutrient composition of grass-fed beef in the United States for inclusion in the USDA National Nutrient Database for Standard Reference, and to compare the fatty acid composition of grass-fed and conventionally fed (control) beef. Ground beef (GB) and strip steaks (SS) were collected on 3 separate occasions from 15 grass-fed beef producers that represented 13 different states, whereas control beef samples were collected from 3 regions (Ohio, South Dakota, and Texas) of the United States on 3 separate occasions. Concentrations of minerals, choline, vitamin B(12), and thiamine were determined for grass-fed beef samples. Grass-fed GB samples had less Mg, P, and K (P < 0.05), and more Na, Zn, and vitamin B(12) (P < 0.05) than SS samples. Fat color, marbling, and pH were assessed for grass-fed and control SS. Subjective evaluation of the SS indicated that grass-fed beef had fat that was more yellow in color than control beef. Percentages of total fat, total cholesterol, and fatty acids along with trans fatty acids and CLA were determined for grass-fed and control SS and GB. Grass-fed SS had less total fat than control SS (P = 0.001), but both grass-fed and control SS were considered lean, because their total fat content was 4.3% or less. For both GB and SS, grass-fed beef had significantly less (P = 0.001 and P = 0.023, respectively) content of MUFA and a greater content of SFA, n-3 fatty acids, CLA, and trans-vaccenic acid than did the control samples. Concentrations of PUFA, trans fatty acids, n-6 fatty acids, and cholesterol did not differ between grass-fed and control ground beef. Trans-vaccenic acid (trans-11 18:1) made up the greatest concentration of the total trans fats in grass-fed beef, whereas CLA accounted for approximately 15% of the total trans fats. Although the fatty acid composition of grass-fed and conventionally fed beef was different, conclusions on the possible effects of these differences on human health cannot be made without further investigation.     

Winter water relations of New England conifers and factors influencing their upper elevational limits. II. Modeling

We used the water relations model, WINWAT, to model winter water relations of three conifer species-eastern hemlock (Tsuga canadensis (L.) Carr.), eastern white pine (Pinus strobus L.) and red pine (P. resinosa Ait.)-growing at their upper elevational limits on Mt. Ascutney, Vermont, USA, in the winters of 1997 and 1999. Modeled relative water contents remained above 60% in the two youngest foliar age classes of all three species during both winters, indicating that desiccation stress in winter is not responsible for setting the upper elevational limits of these species at this site under present climatic conditions. WINWAT indicated that winter water relations of these low-elevational species were sensitive to low relative humidity, which increased transpiration rates, and low temperatures, which inhibited recharge, but are much less sensitive to summer climate than in the case of subalpine conifers in Colorado. Our results indicate that summer and winter temperatures and relative humidities (or precipitation/potential evapotranspiration ratios) should be incorporated into climate change models designed to simulate future tree distributions.     

A habitat-based framework for grizzly bear conservation in Alberta

Grizzly bear (Ursus arctos L.) populations in Alberta are threatened by habitat loss and high rates of human-caused mortality. Spatial depictions of fitness would greatly improve management and conservation action. We are currently challenged, however, in our ability to parameterize demographic rates necessary for describing fitness, especially across gradients of human disturbance and for land cover types. Alternative approaches are therefore needed. We describe here a method of estimating relative habitat states and conditions as surrogates of fitness using models of occupancy and mortality risk. By combining occurrence and risk models into a two-dimensional habitat framework, we identified indices of attractive sinks and safe harbour habitats, as well as five habitat states: non-critical habitats, secondary habitats (low-quality and secure), primary habitats (high-quality and secure), secondary sinks (low-quality, but high risk), and primary sinks (high-quality and high risk). Primary sink or high attractive sink situations were evident in the foothills where bears were using forest edges associated with forestry and oil and gas activities on Crown lands, while primary habitats or safe harbour sites were most common to protected alpine/sub-alpine sites. We suggest that habitat states and indices be used for setting baseline conditions for management and comparison of habitat conditions over time and identification of grizzly bear conservation reserves. A no net loss policy of critical habitats could be used to maintain existing habitat conditions for landscapes threatened by human development. Under such a policy, conversions of primary habitat would require restoration of equivalent amounts of primary sinks through decommissioning of roads.     

Subcutaneous filariasis (Pelecitus sp.) in a yellow-collared macaw (Ara auricollis)

Clinical manifestations of subcutaneous filariasis in a yellow-collared macaw (Ara auricollis) included lameness induced by subcutaneous nodule formation, which was attributed to the presence of the filarial parasite Pelecitus sp. Following anthelmintic and corticosteroid therapeutic failure, the problem was resolved by surgically removing the adult worms.     

Teratocarcinoma of the ovary in a mare

A 5-year-old Appaloosa mare had a history of mild intermittent abdominal discomfort and clinical signs that were suggestive of intestinal obstruction. Palpation per rectum revealed a large mass attached to the left uterine horn, with smaller masses extending dorsally and cranially and causing constriction of the rectum. At necropsy, numerous multilobulated masses were observed in the abdominal cavity and several nodules were seen in the lungs. The left ovary was large and contained both cartilage and hairlike material. Sections from all masses had similar histologic features and confirmed the tentative diagnosis of teratocarcinoma.     

Grizzly bears and forestry: II: Grizzly bear habitat selection and conflicts with road placement

Timber harvesting and oil and gas extraction create ecological change beyond just the footprint of the resource extraction. These activities also create a permanent network of roads that can have lasting effects on forest ecology. Grizzly bears (Ursus arctos) suffer higher mortality when in close proximity to roads, yet bears in the foothills of west-central Alberta, continue to use these high-risk areas. We examined the hypothesis that bears were not necessarily attracted to logging roads but that these roads were placed in habitats that bears preferred. Using a resource selection function, we examined patterns in road placement and identified areas that had characteristics similar to roaded habitats (henceforth road-like habitats). We then quantified grizzly bear response to roads and road-like habitats. Of the 30 bears sampled, we found that 17 selected areas closer to roads than random and 11 selected road-like habitats. Road-like habitats were selected by subadults and some adult females but were not a good predictor of adult male habitat selection. Seasonally, grizzly bears selected habitats closer to roads in spring and early summer but selected road-like habitats more in late summer and fall, indicating that bears may be attracted to disturbed habitats in spring and undisturbed habitats in fall. We conclude that roaded habitats were selected by some grizzly bears but road placement in low-elevation valleys alone does not account for the strong selection preference. Although road placement likely plays a role, other factors such as clearcut association with roads may compound the attractiveness of roaded habitats to grizzly bears. The tradeoff between mortality risk and high-quality forage near roads must be addressed to prevent these areas from acting as attractive sinks.