Sampling bird communities in bottomland hardwood forests of the Mississippi Alluvial Valley: Number of points visited versus number of visits to a point

Abstract

Within each of 4 forest stands on Delta Experimental Forest (DEF), 25 points were visited 5 to 7 times from 8 May to 21 May 1991, and 6 times from 30 May to 12 June 1992. During each visit to a point, all birds detected, visuallyor aurally, at any distance were recorded during a 4-minute interval. Using these data, our objectives were to recommend the number of point counts and the number of visits to a point which provide the greatest efficiency for estimating the cumulative number of species in bottomland hardwood forest stands within the Mississippi Alluvial Valley, and to ascertain if increasing the number of visits to points is equivalent to adding more points. Because the total number of species detected in DEF were different between years, 39 species in 1991 and 55 species in 1992, we considered each year independently. Within each stand, we obtained bootstrap estimates of the mean cumulative number of species obtained from all possible combinations of six points and six visits (i.e., 36 means/stand). These bootstrap estimates were subjected to ANOVA; we modelled cumulative number of species as a function of the number of points visited, the number of visits to each point, and their interaction. As part of the same ANOVA we made an a priori, simultaneous comparison of the 15 possible reciprocal treatments (i.e., 1 point-2 visits vs. 2 points-1 visit, etc.). Results of analyses for each year were similar. Although no interaction was detected between the number of points and the number of visits, when reciprocals were compared, more points visited yielded significantly greater cumulative number of species than more visits to each point. Significant differences were detected among both the number of points visited and among the number of visits to a point. Scheffe's test of differences among means indicated that the cumulative number of species increased significantly with each added point, through five points, but six points did not differ from five points in 1991. Similarly, the cumulative number of species increased significantlywith each revisit, up to four visits, but four visits did not differ significantly from five visits. Starting with one point, which yielded about 33 percent of the total species pool when averaged among one through six points, each subsequent point resulted in an increase of about 9 percent, 5 percent, 3 percent, and 3 percent, respectively. Each sequential increase in the number of visits, however, only resulted in increases of 7 percent, 4 percent, 2 percent, and 2 percent of the total species pool.