19 



Discussion 



Photo-Iden tification Catalog 



The ability to identify individuals over time using natural markings has 

 proved to be a valuable and benign research tool and a standard in population 

 studies of marine mammals. Maintaining a photographic database of individual 

 dolphins enables researchers to monitor not only population parameters but habitat 

 use, social association and distribution patterns. 



The high proportion of marked dolphins and the high frequency of 

 resightings underscores the importance of including only excellent quality images of 

 distinctively marked individuals in the photo-ID catalog. This minimizes 

 subjectivity in the matching process and reduces the chance of making incorrect 

 identifications or missing them altogether. 



The development and use of our photo-identification catalog has been tested 

 in three study areas, including Charlotte Harbor, and has proven effective in each 

 case. However, as the catalogs grow and we expand into different study areas, we 

 recognize the utility of developing computer-assisted matching and archiving 

 abilities. 



Abundance Estimates and Trends 



Comparison of the point abundance estimates from Methods 2, 3, and 4 

 indicates reasonable consistency across methods, and an indication of change from 

 the first three years to the last two years of the study (Figure 7). In all cases the lower 

 95% CLs were greater than or equal to the minimum count provided by the catalog- 

 size method. Thus, if we consider the most extreme 95% CL values to be the limits 

 to our estimates, the number of dolphins using the Charlotte Harbor study area 

 during the surveys was between 198 and 369 during 1990 - 1992, and between 315 and 

 463 during 1993 - 1994. 



We attempted to identify the reasons for the apparent increase in abundance 

 of dolphins in Charlotte Harbor during the later years of the survey. Contra- 

 indicative results for Methods 2 and 3 in 1990 confound evaluation of the 

 significance of differences between 1990 and later years (Figure 8). An apparent 

 increase from 1992 to 1993 and 1994 was also evident, but field effort limitations 

 brought about by Hurricane Andrew complicate interpretation of this year's 

 estimate. Consistent patterns were obtained for both Methods 2 and 3 for 

 comparisons between 1991, and 1993 and 1994, however. Based on Method 2, the 

 abundance estimate from 1991 increased 31% and 61% in 1993 and 1994, respectively. 

 For Method 3, the comparable increases were 40% and 45%. For perspective, this 

 increase, within the summer season across years, is much smaller than the summer 

 to winter increases of 176% and 223% reported by Thompson (1981) and Scott et al. 

 (1989) for Charlotte Harbor and Pine Island Sound. 



