Evidence of large-scale source-sink dynamics and long-distance dispersal among Wood Thrush populations

Source-sink dynamics are commonly thought to occur among Wood Thrush (Hylocichla mustelina) and other songbird populations, allowing for the persistence of populations with negative growth rates ("sinks") through immigration from populations with positive growth rates ("sources"). Knowledge of source-sink dynamics is important for management and conservation because the removal of source habitat should result in the extinction of dependent sinks. However, since research has focused on identifying individual sources/sink populations, not source-sink pairs, we cannot predict these effects or the scale over which they occur. We posit that, when dispersal occurs from a source to a sink year after year, there will be a one-year time-lagged correlation in abundance between the two populations. This should occur for populations separated by distances over which juveniles disperse. Using the North American Breeding Bird Survey data, we tested for such time-lagged correlations between paired Wood Thrush populations from 10 to 200 km apart. Populations were linked with a one-year time lag over distances from 60 to 80 km, indicating that dispersal and source-sink dynamics may occur over these long distances. There was also a declining trend in forest cover from sources to sinks. Conservation and management strategies should therefore be designed at large scales, with consideration for source-sink dynamics and forest cover.     

Selecting indicator species to monitor ecological integrity: a review

We review critical issues that must be considered when selectingindicator species for a monitoring program that aims to maintainor restore ecological integrity. First, we examine the pros andcons of different management approaches on which a conservationprogram can be based and conclude that ecosystem management ismost appropriate. We then identify potential indicators ofecological integrity at various levels of the ecosystem, with aparticular emphasis on the species level. We conclude that,although the use of indicator species remains contentious, it canbe useful if (1) many species representing various taxa and lifehistories are included in the monitoring program, (2) theirselection is primarily based on a sound quantitative databasefrom the focal region, and (3) caution is applied wheninterpreting their population trends to distinguish actualsignals from variations that may be unrelated to thedeterioration of ecological integrity. Finally, we present anddiscuss different methods that have been used to select indicatorspecies.