An ability to predict species' sensitivities to habitat loss and fragmentation has important conservation implications, and numerous hypotheses have been proposed to explain interspecific differences observed in human-dominated landscapes. We used occupancy data collected on

We studied the effects of agriculturally induced fragmentation of forests and wetlands on amphibian assemblages and their distribution in a landscape of the midwestern United States. Potential breeding pools and upland areas in 30 forest patches of various sizes and degrees of isolation were intensively sampled for amphibians during April through August 1996 and March through August 1997 in Indiana. Species presence was documented using pitfall traps, anuran vocalization surveys, and cover-board sampling for adults and minnow traps and dip nets for larvae. Amphibian, anuran, and salamander assemblages were nonrandomly distributed across the landscape. American toads (Bufo americanus) and gray treefrogs (Hyla versicolor) were ubiquitous, whereas the distributions of several other species were ordered in a predictable manner. Logistic regression was used to develop predictive models of probabilities of occurrence for species in response to forest and wetland patch and landscape variables. Occurrence of redback salamanders (Plethodon cinereus) was positively associated with the area of a forest patch. Occurrence of ranid frogs was positively associated with proximity of wetlands for three of four species, and occurrences of smallmouth salamanders (Ambystoma texanum), spring peepers (Pseudacris crucifer), and western chorus frogs (Pseudacris triseriata) were related to the degree of wetland permanency. Multiple linear regression revealed that species richness was greatest for wetlands with intermediate degrees of permanency. The observed nonrandom distribution exhibited by several amphibians suggests that they respond to landscape-level attributes. Moreover, species differed substantially in the nature of their responses to fragmentation, consistent with differences in their life history and ecology.

The recovery of threatened and endangered species is complicated by the number, severity, and tractability of the threats facing each species. We investigated the nature and the treatment of threats in recovery plans for 181 threatened and endangered species. We examined the types of threats facing species, as well as the degree to which threats were understood and addressed. We found that >85% of all species faced at least four out of nine distinct types of threats. The most common threats were those related to resource use, exotic species, construction, and the alteration of habitat dynamics. Recovery plans lacked basic information about the magnitude, timing, frequency, or severity of 39% of all threats facing the 181 species. Likewise, 37% of all threats were not directly addressed with recovery tasks. Threats from pollution were more poorly understood than other threats, and threats from exotics were better addressed than other types of threats. Finally, we found that threats that were better understood were assigned recovery tasks more often than threats that were more poorly understood. Thus our results suggest that a lack of basic understanding of the nature of the threats facing threatened and endangered species may, in part, be undermining our recovery efforts.

No abstract

Landscape-scale alterations that accompany urbanization may negatively affect the population structure of wildlife species such as freshwater turtles. Changes to nesting sites and higher mortality rates due to vehicular collisions and increased predator populations may particularly affect immature turtles and mature female turtles. We hypothesized that the proportions of adult female and immature turtles in a population will negatively correlate with landscape urbanization. As a collaborative effort of the Ecological Research as Education Network (EREN), we sampled freshwater turtle populations in 11 states across the central and eastern United States. Contrary to expectations, we found a significant positive relationship between proportions of mature female painted turtles (Chrysemys picta) and urbanization. We did not detect a relationship between urbanization and proportions of immature turtles. Urbanization may alter the thermal environment of nesting sites such that more females are produced as urbanization increases. Our approach of creating a collaborative network of scientists and students at undergraduate institutions proved valuable in terms of testing our hypothesis over a large spatial scale while also allowing students to gain hands-on experience in conservation science.