Approaches to investigate effects of forest management on birds in eastern deciduous forests: How reliable is our knowledge?

We reveiw some key features of scientific inquiry and experimental design and apply them to studies of the effects of forest management on songbirds. We use examples from contemporary studies in eastern deciduous forests. Scientific methods, observational versus experimental studies, replication and randomization, choice of factors and models, and response variables are important elements in designing research approaches that address effects of forest management. There are significant gaps in our knowledge on the effects of forest management on birds. Many studies have addressed effects of management on species abundance, but we can make only limited inferences from most of these. The design of studies is complicated because of the range of forest management practices, variation in bird species responses, differences among forest types, and the effects of confounding factors such as landscape effects. Few studies have addressed effects of forest management on the reproductive success of forest songbirds. We believe the reliability of our knowledge in this area will be improved most quickly if we use current knowledge to generate hypotheses, use a mix of well-designed observational and manipulative experiments to test them, and more frequently measure reproductive success in addition to bird abundance.