Clearcuts, Salamanders, and Field Studies

Abstract

Environmental impact assessment is an extremely difficult area in which to conduct field research studies. There are many design problems due to difficulties in finding controls and in randomization and replication of study plots. Often, the need for results in a short time and with little expenditure poses practical difficulties. Currently, examining effects of clearcutting and other forest management techniques on herpetofauna is a very active field. The studies in question deal specifically with Plethodontid salamanders of the Southern Appalachians. We note that there are (at least) two fundamentally different approaches considered in the literature for the estimation of salamander population recovery following clearcutting. One approach, used by Ash (1988, 1997), is a direct longitudinal approach. Matched control and treated (clearcut) plots at different locations are followed over time-ideally from before clearcutting until many years later-to monitor the recovery of the salamander populations. This is an example of the well-known beforeafter-control-impact (BACI) design (e.g., Green 1979; Eberhardt & Thomas 1991). Assuming that the control and treated plots (in absence of clearcutting) would have had the same time trends since clearcutting, recovery can be measured directly, provided that the different sites are measured over a long enough time period. At intermediate time points, extrapolation of trends forward in time is necessary and requires the additional assumption that past recovery trends will hold in the future. Ash (1997) assumes that the recovery trends can be extrapolated linearly. Another approach, used by Petranka and coworkers (Petranka et al. 1993, 1994), is a time-specific approach. Sites of different ages since clearcutting are all sampled at one specific point in time to monitor the current status of salamander populations. Recovery of clearcut