LEATHERWOOD AERIAL SURVEY (IF DOLPHINS AND MANATEES 



herd density are mutually independent, the data 

 by day were examined for correlation. Using Ken- 

 dall's rank correlation coefficient (Conover 1971) 

 at a = 0.05, mean herd size and mean herd density 

 were demonstrated to be uncorrelated within the 

 area surveyed. 



The dolphin densities per square kilometer were 

 then multiplied by the area surveyed and a factor 

 of 5 (since the survey covered 20' V of the total area) 

 and the 95'^i confidence limits calculated for the 

 estimate. The figures support an estimate of 

 438±127 dolphins for the Indian and Banana Riv- 

 ers during the time of the survey. 



As an alternate method for estimating dolphin 

 densities, I took the average density over repli- 

 cates from column 3, Table 2. This procedure re- 

 sults in a density estimate of 0.40 dolphin km- 

 (1.36 dolphins/n.mi.^), a value very close to the 

 estimate obtained using the method described 

 above (0.41 dolphin km-, 1.41 dolphins/n.mi.^), 

 but having a variance twice as large (0.1837 vs. 

 0.0941. Because of the higher variance, it can be 

 argued that the first method used, because it takes 

 into account both average herd size and average 

 herd density, is preferable in this case. 



The numbers of dolphins entering or leaving the 

 river at Sebastian (4 groups totaling 15 animals) 

 and Fort Pierce Inlets (none sighted) were negligi- 

 ble and were judged as insignificant to the total 

 population size. Two of those groups were entering 

 the river against an outgoing tide, one moved from 

 the river into the inlet on an ebbing tide, then 

 turned around and reentered the river, and one 

 was milling within the inlet (Table 3). 



The surprisingly low estimate does, of course, 

 raise an important question. Is the population of 

 bottlenosed dolphins in the river complex always 

 this small (and only appears larger because of 

 periodic concentrations of animals in limited 

 areas) or is it augmented seasonally by influxes of 

 animals from other areas migrating into,the rivers 

 in response to the movement of fishes? 



Caldwell ( 1955) and others have suggested lim- 

 ited home ranges for bottlenosed dolphins. Wells 

 et al.," Irvine et al.,'- and Shane and Schmidley 



"Wells. R. S.. A, B. Irvine, and M. D. Scott 1977, Home 

 range characteristics and group composition of the Atlantic 

 bottlenosed dolphin Tur>ii(>ps tnirtcatus on the west coast of 

 Flonda. In Proceedings lAbstr.l of the Second Conference on 

 the Biology of Marine Mammals, San Diego, Calif., 12-15 Dec. 

 1977, p. l.i 



'■^Irvine, A. B., M D, Scott, and R. S Wells, 1977, Move- 

 ments and activities of Atlantic bottlenosed dolphins. In Pro- 

 ceedings lAbstr.l of the Second Conference on the Biology of 

 Marine Mammals, San Diego. Calif, 12-15 Dec. 1977. p. 16. 



(see footnote 10) have all clearly demonstiated 

 limited home ranges for portions of the popula- 

 tions in their study areas; Wells et al. (see footnote 

 1 1 ) have shown differences in size and locations of 

 home ranges based on age and sex classes, and all 

 these authors have reported some movements ol' 

 animals into and out of their study areas. 



Caldwell and Caldwell (1972) summarize the 

 views of the fishermen from eastern Florida that 

 there are "river" and "ocean" T. Iriniciiliis popula- 

 tions. Caldwell et al. (1975) presented evidence 

 from the distribution of cases of "Lobos" disease 

 (lobomycosis) in bottlenosed dolphins that indi- 

 cate greatest susceptibility to the disease in 

 riverine-estuarine stocks and suggest isolation of 

 river from ocean stocks. 



Shane''' reported that the offshore population of 

 bottlenosed dolphins off Texas rarely interacted 

 with the bay population but that the winter popu- 

 lation in the Port Aransas area was at least twice 

 as large as that in summer, because the bay popu- 

 lation was augmented by "large numbers" of dol- 

 phins entering that area for the winter either from 

 the adjacent gulf or from adjacent bay systems. 

 Whether or not a similar influx occurs in the In- 

 dian River is unclear. Additional surveys during 

 the peak seasons of the most important midwinter 

 fisheries (king and Spanish mackerel, bluefish, 

 spots, and pompano) might provide answers. 



In considering the questions of the dolphins' 

 population size and alleged damage to nets, it 

 should be remembered that bottlenosed dolphins, 

 at least in some areas, are not uniformly distrib- 

 uted but tend to concentrate in areas of high fish 

 productivity (Leatherwood and Platter see foot- 

 note 9) which are often areas of highest human use 

 (Leatherwood 1975). Irvine et al. (see footnote 12), 

 for example, reported that short-term movements 

 of bottlenosed dolphins near Tampa Bay appear to 

 correlate with movements of mullet. Frequent 

 joint use of resources by dolphins and humans 

 make the dolphins highly visible and could result 

 in inflated estimates of their numbers. 



Even if not augmented seasonally by immigra- 

 tion from other areas, the relatively small dolphin 

 population in Indian and Banana Rivers could be 

 responsible for net damage of the types reported by 

 Cato and Prochaska ( 1976). Feeding by dolphins 

 near seine and gill net fisheries is well known 



'^Shane. S, H. 1977, Population biology of Twr-'iio/w /ra(i- 

 latus in Texas. In Proceedings (Ab.str.l of the Second Confer- 

 ence on the Biologvof Marine Mammals. San Diego. Calif. 12-15 

 Dec. 1977. p. .57. ' 



57 



