Trespassing on the tracks: a review of railway pedestrian safety research

INTRODUCTION: Train-pedestrian collisions have been shown to be the leading cause of fatality in train-related accidents worldwide, yet there is remarkably little research in this area. METHOD: In this paper, the major types of railway transportation accident research are briefly highlighted to indicate the general context of research concerning train-pedestrian collisions, which are then reviewed. Themes emerging from the diverse research are identified, the various strategies that have been proposed for prevention of railway pedestrian accidents are discussed, and the empirical evidence for their efficacy examined in the light of the much more extensive literature on road pedestrian accidents. Finally, it is proposed that application of current theory in behavioral and cognitive psychology may usefully inform future research in transportation safety.     

Habitat Improvements and Fish Community Response Associated with an Agricultural Two‐Stage Ditch in Mower County,Minnesota

Water quality and stream habitat in agricultural watersheds are under greater scrutiny as hydrologic pathways are altered to increase crop production. Ditches have been traditionally constructed to remove water from agricultural lands. Little attention has been placed on alternative ditch designs that are more stable and provide greater habitat diversity for wildlife and aquatic species. In 2009, 1.89 km of a conventional drainage ditch in Mower County, Minnesota, was converted to a two‐stage ditch (TSD) with small, adjacent floodplains to mimic a natural system. Cross section surveys, conducted pre‐ and post‐construction, generally indicate a stable channel with minor adjustments over time. Vegetation surveys showed differences in species composition and biomass between the slopes and the benches, with changes ongoing. Longitudinal surveys demonstrated a 12‐fold increase in depth variability. Fish habitat quality improved with well‐sorted gravel riffles and deeper pool habitat. The biological response to improved habitat quality was investigated using a Fish Index of Biological Integrity (FIBI). Our results show higher FIBI scores post‐construction with scores more similar to natural streams. In summary, the TSD demonstrated improvements in riparian and instream habitat quality and fish communities, which showed greater fish species richness, higher percentages of gravel spawning fish, and better FIBI scores. This type of management tool could benefit ditches in other regions where gradient and geology allow.     

Can dispersal mode predict corridor effects on plant parasites?

Habitat corridors, a common management strategy for increasing connectivity in fragmented landscapes, have experimentally validated positive influences on species movement and diversity. However, long-standing concerns that corridors could negatively impact native species by spreading antagonists, such as disease, remain largely untested. Using a large-scale, replicated experiment, we evaluated whether corridors increase the incidence of plant parasites. We found that corridor impacts varied with parasite dispersal mode. Connectivity provided by corridors increased incidence of biotically dispersed parasites (galls on Solidago odora) but not of abiotically dispersed parasites (foliar fungi on S. odora and three Lespedeza spp.). Both biotically and abiotically dispersed parasites responded to edge effects, but the direction of responses varied across species. Although our results require additional tests for generality to other species and landscapes, they suggest that, when establishing conservation corridors, managers should focus on mitigating two potential negative effects: the indirect effects of narrow corridors in creating edges and direct effects of corridors in enhancing connectivity of biotically dispersed parasites.     

Edge effects, not connectivity, determine the incidence and development of a foliar fungal plant disease

Using a model plant-pathogen system in a large-scale habitat corridor experiment, we found that corridors do not facilitate the movement of wind-dispersed plant pathogens, that connectivity of patches does not enhance levels of foliar fungal plant disease, and that edge effects are the key drivers of plant disease dynamics. Increased spread of infectious disease is often cited as a potential negative effect of habitat corridors used in conservation, but the impacts of corridors on pathogen movement have never been tested empirically. Using sweet corn (Zea mays) and southern corn leaf blight (Cochliobolus heterostrophus) as a model plant-pathogen system, we tested the impacts of connectivity and habitat fragmentation on pathogen movement and disease development at the Savannah River Site, South Carolina, USA. Over time, less edgy patches had higher proportions of diseased plants, and distance of host plants to habitat edges was the greatest determinant of disease development. Variation in average daytime temperatures provided a possible mechanism for these disease patterns. Our results show that worries over the potentially harmful effects of conservation corridors on disease dynamics are misplaced, and that, in a conservation context, many diseases can be better managed by mitigating edge effects.     

Diversity, Seasonality, and Context of Mammalian Roadkills in the Southern Great Plains

Thousands of mammals are killed annually from vehicle collisions, making the issue an important one for conservation biologists and environmental managers. We recorded all readily identifiable kills on or immediately adjacent to roads in the southern Great Plains from March 2004–March 2007. We also recorded distance traveled, whether a road was paved or divided, the number of lanes, and prevailing habitat. Surveys were opportunistic and were conducted by car during conditions of good visibility. Over our 239 surveys and >16,500 km traveled, we recorded 1412 roadkills from 18 different mammal species (size ranged from Sciurus squirrels to the white-tailed deer, Odocolieus virginianus). The overall kill rate was 8.50 / 100 km. Four species were prone to collisions: the Virginia opossum (Didelphis virginiana), nine-banded armadillo (Dasypus novemcinctus), striped skunk (Mephitis mephitis), and northern raccoon (Procyon lotor). Together they accounted for approximately 85% (1198) of all roadkills. Mortality rate differed significantly between 2- and 4-lane roads (8.39 versus 7.79 / 100 km). Kill rates were significantly higher on paved versus unpaved roads (8.60 versus 3.65 / 100 km), but did not depend on whether a road was divided. Roadkills were higher in spring than in fall (1.5×), winter (1.4×), or summer (1.3×). The spring peak (in kills / 100 km) was driven chiefly by the armadillo (2.76 in spring/summer versus 0.73 in autumn/winter) and opossum (2.65 versus 1.47). By contrast, seasonality was dampened by a late winter/early spring peak in skunk mortalities, for which 41% occurred in the 6-week period of mid-February through March. The raccoon did not exhibit a strong seasonal pattern. Our data are consistent with dispersal patterns of these species. Our results underscore the high rate of highway mortality in the southern plains, as well as differences in seasonality and road type that contribute to mortality. Conservation and management efforts should focus on creating underpasses or using other means to reduce roadkill rates.     

Acute toxic hazard evaluations of glyphosate herbicide on terrestrial vertebrates of the oregon coast range

GOAL, SCOPE AND BACKGROUND: The degree to which dose responses of model organisms (lab rodents) can adequately predict dose responses of free-ranging wild mammals or amphibians is unknown, and the relative sensitivity of such species to body loading of a toxicant such as glyphosate is seldom reported. For relative effects of dosage, we compare sensitivity of nine wild vertebrate species to effects of high doses of glyphosate in Swiss-Webster laboratory mice both by gavage and by intraperitoneal injection. We also evaluate sublethal effects of herbicide exposure on behavior and reproductive success of one mammal and one amphibian species. METHODS: Comparisons of acute toxicity of glyphosate were made with intraperitoneal dosings of technical glyphosate isopropylamine salt to nine species of terrestrial vertebrates (five amphibians, four mammals) and compared with responses in Swiss-Webster laboratory mice. Animals collected from sites that had no recent herbicide application were allowed 7-14 days to equilibrate in captivity before treatment. RESULTS: Median lethal dose ranged from 800 to 1,340 mg kg(-1) in mammals, and 1,170 to >2,000 mg kg(-1) in amphibians, with Oregon vole being the most sensitive. White lab mice were in the middle of the mammalian range. Tailed frog, at >2,000 mg kg(-1) was the least sensitive. Calibration of IP sensitivity to oral administration by gavage indicated that roughly four times as much glyphosate must be administered to obtain a comparable estimate of lethality. Administration by gavage in highly concentrated solutions tended to cause physical injury, hence may prove less useful as a relative indicator of toxicity. When sublethal dosages were given to roughskin newts or chipmunks, mobility and use of cover appeared largely unaffected. DISCUSSION: Direct toxic effects of spraying glyphosate under normal forest management seem unlikely for the nine species examined. Nor could we detect significant indirect effects of exposure on behavior and use of cover features in two species. There may be effects on other aspects of the field biology of these animals, such as reproductive rates, which we did not investigate. Recent field data indicate that changes in habitat quality following herbicide application can result in high reproductive activity in species associated with the grasses and forbs that proliferate following field applications. CONCLUSIONS: When compared to field data on body burdens of wild mammals exposed after aerial application of glyphosate at maximum rates in forests, there seems to be a large margin of safety between dosages encountered and those causing either death or limitation of movement, foraging or shelter. RECOMMENDATIONS AND PERSPECTIVES: Margins of safety for small mammals and amphibians appear to be large under any probable exposure scenarios, however our results indicate high variability in responses among species. Uncertainty introduced into field studies from unknown sources of mortality (e.g, likely predation) must be considered when interpreting our results.     

Potential effects of forest policies on terrestrial biodiversity in a multi-ownership province.

We used spatial simulation models to evaluate how current and two alternative policies might affect potential biodiversity over 100 years in the Coast Ranges Physiographic Province of Oregon. This 2.3-million-ha province is characterized by a diversity of public and private forest owners, and a wide range of forest policy and management objectives. We evaluated habitat availability for seven focal species representing different life histories. We also examined how policies affected old-growth stand structure, age distributions relative to the historical range of variability, and landscape patterns of forest types. Under the current policy scenario, the area of habitat for old-growth forest structure and associated species increased over time, the habitat for some early-successional associates remained stable, and the area of hardwood vegetation and diverse early-successional stages declined. The province is projected to move toward but not reach the historical range of variation of forest age classes that may have occurred under the wildfire regimes of the pre-Euroamerican settlement period. Ownership explained much of the pattern of biodiversity in the province, and under the current policy scenario, its effect increased over time as the landscape diverged into highly contrasting forest structures and ages. Patch type diversity declined slightly overall but declined strongly within ownerships. Most of the modeled change in biodiversity over time resulted from policies on public forest lands that were intended to increase the area of late-successional forests and species. One of the alternative policies, increased retention of wildlife trees on private lands, reduced the contrast between ownerships and increased habitat availability over time for both early- and late-successional species. Analysis of another alternative, stopping thinning of plantations on federal lands, indicated that current thinning regimes improve habitat for the Olive-sided Flycatcher, but the no-thinning alternative had no effect on the habitat scores for the late-successional species in the 100-year simulation. A comparison of indicators of biological diversity suggests that using focal species and forest structural measures can provide complementary information on biodiversity. The multi-ownership perspective provided a more complete synthesis of province-wide biodiversity patterns than assessments based on single ownerships.     

Cumulative ecological and socioeconomic effects of forest policies in coastal Oregon.

Forest biodiversity policies in multi-ownership landscapes are typically developed in an uncoordinated fashion with little consideration of their interactions or possible unintended cumulative effects. We conducted an assessment of some of the ecological and socioeconomic effects of recently enacted forest management policies in the 2.3-million-ha Coast Range Physiographic Province of Oregon. This mountainous area of conifer and hardwood forests includes a mosaic of landowners with a wide range of goals, from wilderness protection to high-yield timber production. We projected forest changes over 100 years in response to logging and development using models that integrate land use change and forest stand and landscape processes. We then assessed responses to those management activities using GIS models of stand structure and composition, landscape structure, habitat models for focal terrestrial and aquatic species, timber production, employment, and willingness to pay for biodiversity protection. Many of the potential outcomes of recently enacted policies are consistent with intended goals. For example, we project the area of structurally diverse older conifer forest and habitat for late successional wildlife species to strongly increase. 'Other outcomes might not be consistent with current policies: for example, hardwoods and vegetation diversity strongly decline within and across owners. Some elements of biodiversity, including streams with high potential habitat for coho salmon (Oncorhynchus kisutch) and sites of potential oak woodland, occur predominately outside federal lands and thus were not affected by the strongest biodiversity policies. Except for federal lands, biodiversity policies were not generally characterized in sufficient detail to provide clear benchmarks against which to measure the progress or success. We conclude that land management institutions and policies are not well configured to deal effectively with ecological issues that span broad spatial and temporal scales and that alternative policies could be constructed that more effectively provide for a mix of forest values from this region.     

Transfer of natural and anthropogenic radionuclides to ants, bryophytes and lichen in a semi-natural ecosystem

Few data are available to quantify the transfer of both natural and anthropogenic radionuclides to detritivorous invertebrates to facilitate estimation of the internal dose to such biota in models used to assess radiation exposure. To enhance the available data, activity concentrations of ¹³⁷Cs, ⁴⁰K, ⁹⁰Sr, ^239?+?240Pu, ²⁴¹Am, ²³⁵U and ²³⁸U were measured in ants (Formicidae) and corresponding undisturbed soil collected from the Zlatibor mountain in Serbia and ant/soil concentration ratios (CR) calculated. The ²⁴¹Am concentration ratios for ants were fourfold higher than those calculated for ants in a previous study whereas they are similar to the more numerous data previously reported for a range of detritivorous invertebrates in other studies. CR values for ¹³⁷Cs in ants were similar to the few other reported values and slightly lower than those for a range of detritivorous invertebrates. Those for ^239?+?240Pu were slightly higher than those for ants in two other studies but they were close to upper limit of a range of data reported for detritivorous invertebrates. All the CR values will be included in a future revision of the ERICA Tool database and will particularly improve the information available for uranium.