The oldest, simplest, and most frequently used method of studying migration is by direct observation. Size, color, song, and flight of different species all aid the amateur as well as the professional in determining when birds are migrating. Studies begun by Wells W. Cooke and his collaborators (Cooke 1888-1915) and continued by his successors in the U.S. Bureau of Biological Survey (later U.S. Fish and Wildlife Service) were of particular importance in the earlier years of these investigations in North America. Some of the largest and most interesting routes and patterns were sorted out by tediously compiling and comparing literally thousands of oberservations on whether a species was or was not seen in a given locality at a particular time of the year. More recently, "The Changing Seasons" reports by many amateur bird observers in Audubon Field Notes (now American Birds) have been a most important source of information on direct observation of migration. In the agregate, direct observation has contributed much to our knowledge of migration, but, as will be pointed out in other sections, until a few years ago, observers were not aware of some of the biases in this technique.
The "moon watch" is a modification of the direct observation method. It has long been known that many species of birds migrate at night. Until recently, it was not apparent just how important nocturnal migration really is. Significant information has been derived from watching, through telescopes, the passage of migrating birds across the face of a full moon. Since the actual percent of the sky observed by looking through a telescope at the moon is extremely small (approximately one-hundred thousandth of the observable sky), the volume of birds recorded is small. On a night of heavy migration, about 30 birds per hour can be seen. The fact that any birds are observed at all is testimony to the tremendous numbers passing overhead. Large-scale, cooperative moon-watching studies have been organized and interpreted by George H. Lowery, Jr. (1951; Lowery and Newman 1966).
Another specialized direct observation approach which has yielded
important information on the spatial and altitudinal distribution of night
migrating birds has been the use of small aircraft equipped with auxiliary
landing lights (Bellrose 1971). Major disadvantages of night observation
are that species cannot be identified and that birds continue to migrate
without a full moon. However, these techniques do give information on the
nocturnal migration movements that could not be obtained by other methods.
An adjunct to the previously described nocturnal observation methods, which has potential for species identification, is the use of a parabolic reflector with attached microphone to amplify call (chip) notes (Ball 1952; Graber and Cochrane 1959). This device, when equipped with a tape recorder, can record night migrants up to 11,000 feet on nights with or without a full moon. A primary disadvantage is that one cannot tell the direction a bird is traveling and there is considerable difficulty in identifying the chip notes made by night migrants. In addition, the bird may not call when it is directly over the reflector and consequently it would not be recorded. These calls are quite different from the notes we hear given by familiar birds during the daytime while they are scolding an intruder or advertising their territory.
Reference material consisting of preserved bird skins with data on time and place of collection exist in many natural history museums. The essential ingredient in studying migration by this method is to have an adequate series of specimens taken during the breeding season so differences in appearance between geographically separated breeding populations of the same species can be determined. Such properly identified breeding specimens may be used for comparison with individuals collected during migration to associate them with their breeding areas (Aldrich 1952; Aldrich, Duvall, and Gels 1958). This supplies a convenient way of recognizing and referring to individuals representative of known populations wherever they may be encountered.
If birds can be captured, marked, and released unharmed, a great deal of information can be learned about their movements. Many different marking methods have been developed to identify particular individuals when they are observed or recaptured at a later date. A few of the general methods are summarized in this section.
The records of banded birds are also yielding other pertinent information relative to their migrations such as arrival and departure dates, the length of time different birds pause on their migratory journeys to feed and rest, the relation between weather conditions and starting times for migration, the rates of travel for individual birds, the degree of regularity with which individual birds return to the summer or winter quarters used in former years, and many other details. Many banding stations are operated systematically throughout the year and supply much information concerning the movements of migratory birds that heretofore could only be surmised. The most informative banding studies are those where particular populations of birds are purposely banded to produce certain types of information when they are recovered. Examples of such planned banding are the extensive marking of specific populations of ducks and geese on their breeding grounds by the U.S. Fish and Wildlife Service and the Canadian Wildlife Service, as well as in "Operation Recovery," the cooperative program of banding small landbirds along the Atlantic Coast (Baird et al. 1958). When these banded birds are recovered, information concerning movements of specific populations or the vulnerability to hunting is gained. Colored leg bands, neck collars, or streamers can be used to identify populations or specific individuals, and birds marked with easily observed tags can be studied without having to kill or recapture individuals, thus making it a particularly useful technique.
We have learned about the migratory habits of some species through banding, but the method does have shortcomings. If one wishes to study the migration of a particular species through banding, the band must be encountered again at some later date. If the species is hunted, such as ducks or geese, the number of returns per 100 birds banded is considerably greater than if one must rely on a bird being retrapped, found dead, etc. For example, in mallards banded throughout North America the average number of bands returned the first year is about 12 percent. In most species that are not hunted, less than 1 percent of the bands are ever seen again.
In 1935, Lincoln commented that, with enough banding, some of the winter ranges and migration routes of more poorly understood species would become better known. A case in point is the chimney swift, a common bird in the eastern United States. This is a nonhunted species that winters in South America. Over 500,000 chimney swifts have been banded, but only 21 have been recovered outside the United States (13 from Peru, 1 from Haiti, and the rest from Mexico). The conclusion is simply this: Whereas banding is very useful for securing certain information, the volume of birds that need to be banded to obtain a meaningful number of recoveries for determining migratory pathways or unknown breeding or wintering areas may be prohibitive. One problem in interpretation of all banding results is the fact that recoveries often reflect the distribution of people rather than migration pathways of the birds.
Other methods used to mark individuals in migration studies include
clipping the tip end off a feather (not a major flight feather) with a
fingernail clipper or touching the feather with colored paint or dye. This
marking technique is obviously good for only as long as the bird retains
the feather (usually less than one year), but allows the investigator to
recognize whether the bird has been handled previously or not.
One of the most promising methods of tracking the movements of individual birds in migration has been developed in recent years. It is called radio tracking, or telemetry, and consists of attaching to a migrating bird a small radio transmitter that gives off periodic signals or "beeps". With a radio receiving set mounted on a truck or airplane, it is possible to follow these radio signals and trace the progress of the migrating bird. One of the most dramatic examples of this technique was reported by Graber in 1965. He captured a grey-cheeked thrush on the University of Illinois campus and attached a 2.5-gram transmitter to it (a penny weighs 3 grams). The bird was followed successfully for over 8 hours on a course straight across Chicago and up Lake Michigan on a continuous flight of nearly 400 miles at an average speed of 50 mph (there was a 27 mph tail wind aiding the bird). It is interesting to note that while the little thrush flew up the middle of Lake Michigan, the pursuing aircraft skirted the edge of the lake and terminated tracking at the northern end after running low on fuel while the bird flew on. The limitations of radio telemetry, of course, are the size of the transmitter that can be placed on birds without interfering with flight and the ability of the receiving vehicle to keep close enough to the flying bird to detect the signals. Despite this difficulty there has been considerable development in the technology, and encouraging results to date give promise for the future, particularly when receivers can be mounted on orbiting satellites (Graber 1965; Bray and Corner 1972; Southern 1965).
One of the developments of our modern age of electronics has been the discovery that migrating birds show up on radar screens used in monitoring aircraft. At first. the screen images caused by flying birds were a mystery to radar operators, and they designated the dots "angels." Later when their nature was understood, students of bird migration seized on the unique opportunity to obtain information on movements of birds over extensive areas (Sutter 1957; Drury 1960; Lacke 1963a, b; Bellrose 1967; Graber 1968; and Gauthreaux 1972a, b).
Three types of radar have been used for studing birds: 1) general surveillance radar, similar to ones located at airports, that scans a large area and indicates the general time and direction of broad movements of birds; 2) a tracking radar that records the path of an airplane (or bird) across the sky by "locking on" to a designated "target" and continuously following only that object; and 3) a Doppler radar similar to those operated by law enforcement agencies for measuring the speed of a passing automobile. The latter radar set is useful in determining the speed of flying birds.
The use of radar in migration studies has been invaluable in determining direction of mass movement, dates and times of departure, height of travel, and general volume, especially at night. One interesting fact to come out of current radar work is the discovery of relatively large movements of warblers and other land birds migrating over the seas rather than along the coastlines and in directions observers were completely unaware of a few years ago.