You are not currently logged in.
Access JSTOR through your library or other institution:
If You Use a Screen ReaderThis content is available through Read Online (Free) program, which relies on page scans. Since scans are not currently available to screen readers, please contact JSTOR User Support for access. We'll provide a PDF copy for your screen reader.
Estimating Breeding Season Bird Densities from Transect Counts
John T. Emlen
Vol. 94, No. 3 (Jul., 1977), pp. 455-468
Published by: American Ornithological Society
Stable URL: http://www.jstor.org/stable/4085213
Page Count: 14
Since scans are not currently available to screen readers, please contact JSTOR User Support for access. We'll provide a PDF copy for your screen reader.
Preview not available
In this paper I propose that for each species in an area the number of birds detected along a transect trail can be translated into an estimate of absolute density (birds per unit of area) by counting all the detectable (cue-producing) birds in a trailside strip narrow enough to permit detection of all cues produced (the specific census strip), and adjusting this count for the undetectable (silent) birds in the strip by applying a locally and concurrently derived index of the frequency of cue production for the species. Cue frequency is apparently impossible to measure in nonbreeding birds, but if all cues (sightings, calls, etc.) are used, most of the birds in the strip presumably will be detected when the observer advances slowly enough to allow each bird a good chance to make its presence known. In the nonbreeding season the strip width is set for each species at the distance from the trail at which total cue detection starts to decline. In the breeding season cue frequency may be determined for song cues, and strip widths for each species set at a convenient distance within the relatively great distance at which song detection starts to decline. Values obtained by these transect procedures reflect the density of each species at the time the traverse is run; a series of traverses may be made to provide mean values for selected periods through the season. The method is similar to that described in an earlier paper. New or modified procedures are described for recording detections, establishing specific strip widths, bypassing the calculation of "coefficients of detectability," estimating distances in the field, determining an optimum rate of progress, and measuring the frequency of singing in a representative sample of the population. Problems of converting adjusted transect counts of singing males to total population density and of applying a combination of all-cue and song-cue procedures to mixed populations of breeding and nonbreeding species are discussed. A comparison of transect and plot map census methods is presented. In the transect method density estimates are of birds present at a particular time rather than of birds wholly or partially resident at one time or another during a season. The sampling quadrats of transect censuses are elongate rectangles spanning extensive tracts of habitat rather than truncated blocks of representative habitat. The transect method is applicable at any season while the plot map method can be used only during the breeding season when birds are singing on territories. Problems of reliability in the plot map method stemming from individual movements during a survey period and from questions on how to interpret clusters of observation points on territory maps and how to evaluate boundary line territories are replaced in the transect method by problems of accuracy in assuming complete coverage in the all-cue operations and in assigning birds as inside or outside the lateral boundary lines of the census strips. More area can be covered per unit of time in the transect than in the mapping method.
The Auk © 1977 American Ornithologists' Union