The cost comparison shown in Table 22 is based upon using trained profes- 

 sionals. It would reduce expenses by depending upon qualified volunteer 

 workers, an approach the British have employed in "Operation Seafarer" (Cramp 

 et al. 1974). However, this approach requires extensive coordination and 

 standardization of technique to insure that, over large geographic areas, all 

 volunteers can achieve 100% coverage. 



Based on experience in 1976 and after consultation with others (P. A. and 



F. G. Buckley and others), it was decided that helicopters would be used for 

 the coastal study area in 1977. 



Aerial-Ground Comparisons 



Because population estimates in 1977 were made from helicopters, "ground- 

 truthing" was performed at selected colonies to establish the numerical rela- 

 tionship between the number of adult birds and the number of active nests 

 found in the colony. Kadlec and Drury (1968b) suggested that, for herring 

 gull colonies, proper interpretation requires photographing several (5-10) 

 islands with known numbers of nests to establish the gull/nest ratio (correc- 

 tion factor) for that particular census. Aerial photography was used with 

 limited success in both 1976 and 1977 using small, medium, and large format 

 cameras (see Aerial Photography). Instead of deriving bird/nest ratios from 

 photos, I compared numbers visually estimated to nest numbers in colonies (or 

 marked sample plots in colonies) and used the slope of the linear regression 

 equation as the correction factor. In using this measure, one must be aware 

 that the variability observed is additive, partly due to errors in estimation 

 and partly due to actual variation in bird attendance at the colony. 



Because of the logistics involved, sufficient samples of colonies of all 

 species could not be ground counted. Instead, comparisons were made for the 

 most common species and those most difficult to census were treated. For some 

 species, adult-to-nest ratios were derived from other sources (discussed 

 below). 



The aerial censuses were flown using small, 3-person Bell or Enstram 

 helicopters, except in New York where a 5-person Bell jet "Ranger" was used. 

 Censuses in each State required from 1-4 days during the period 31 Hay to 10 

 June. During the flight, two observers recorded species composition and 

 numbers of adults present. After the flights, ground verification and census- 

 ing were performed in selected colonies. The ground (nest) counts were usually 

 made within 10 days after the aerial census. In most cases, one-search nest 

 counts were made. In Massachusetts, a two-search, "mark-recapture" method was 

 used. 



The results of the aerial estimate-nest count regressions are shown for 

 several species in Table 23. The aerial estimates used were usually composite 

 "best estimates" if there was more than one observer. Not only does the bird/ 

 nest ratio (b value in Table) vary "among species" as expected, but also 

 "within species" for different censuses. Black-crowned night herons are 

 grossly underestimated in aerial counts. Four nests were found for every 

 adult flushed from the colony by helicopter (b = 0.24). Considering the very 

 dense colony shrubbery obscuring the observer's view and the heron's propen- 

 sity to remain on nests during disturbances (Portnoy pers. comm. , P. A. and F. 



G. Buckley pers. comm.), this result is not surprising. 



42 



