Bird Migration: Methodologies and Major Research Trajectories (1945-1995)

Abstract

As a new graduate student at Louisiana State University in the early 1960s I was greatly impressed by a paper published by Niko Tinbergen (1963). In that paper he pointed out that the first phase in the study of animal behavior is that of observation and description. Once this phase is completed, if it is ever really completed, the scientist can then proceed to ask questions that Tinbergen grouped into four categories: immediate causation, development, evolutionary history, and function. These categories have proven very useful to me in "classifying" the types of bird migration studies that have been conducted through time, and I use them to organize my review of important research tra-jectories in the study of bird migration over the past fifty years. Tinbergen (1963) cautioned that observation and description should be the foundation for the formulation of hypotheses, but this has not always been the case. Unfortunately many young investigators today have the impression that descriptive studies are a waste of effort and money. They have been told that hypothesis testing is where the action is, and if you are not testing hypotheses, your interests and efforts are surely misguided or at the very least outmoded. Nothing could be farther from the truth, and Lorenz (1973) said it best when he referred to the trend as the "fashionable fallacy of dispensing with description." There can be no debate about the importance of testing hypotheses, but this phase of scientific endeavor should be predi-cated on strong underpinnings of observation, description , and discovery. Hypotheses should not be based on unfounded assumptions. The historical perspective that follows first reviews the important role of methodology in the "observation and description" phase of migration as studied in the field and in the laboratory. It then briefly addresses what I consider the major trends in migration research relative to Tinbergen' s four classes of questions in the last half century. Because of space limitations, the citations that are included are representative and are not intended to be exhaustive. numbered all other types, undoubtedly because of the ease of conducting such studies. Most of these studies have used straightforward methods to gather data on migratory birds: 1) observations of visible migration in progress, 2) observations of migrant birds at stopover areas, 3) the capture and banding of migrants, and 4) the gathering of dead and injured migrant birds after collision with man-made structures. The vast majority of the studies of the visible migration of birds have used field observations with binoculars and telescopes to monitor the low-altitude passage of birds at a particular location (Thomson 1953). Several long-term studies have been conducted at observatories in Europe (Edelstam 1972, Ulfstrand et al. 1974), however similar work has not been undertaken to any great extent on the western side of the Atlantic. The only studies in North America that resemble those of visible migration in Europe are at sites where migrating raptors pass in large numbers (Heintzelman 1975). The usefulness of the data gathered during studies of the visible migration of birds cannot be disputed, but one must be cautious when interpreting these data because of possible hidden biases and inherent limitations of the methods. For example, birds may be passing over the same observation point several times, and an unknown amount of migration might be "in-visible" to the observer because the birds are flying too high or too far away to be seen with binoculars (Al-erstam 1978, Kerlinger 1989). Studies of visible migration have contributed considerable data on the seasonal and daily temporal patterns of migration for many different species, and it has been argued that these data may be important in monitoring population sizes of the various species (but see Svensson 1978). Many of the studies of visible migration have spanned decades and have generated sizable data sets. Because of the difficulty of publishing compilations of essentially "raw data," I suspect that many data sets are archived in bird observatory libraries or associated institutes throughout the world, and the data are used mostly in local and regional publications on bird migration. The observations and censusing of migrant birds at stopover areas has also been an important part of migration studies in the field. Systematic observations of the behavior of birds at migration stopover areas have produced important findings on feeding behavior during migration (Ferns 1975), territoriality while enroute (Rappole and Warner 1976), and bird-habitat associations during migration (Pamell 1969, Hutto 1985a, Moore et al. 1990) to mention a few. Additional data