NOTE Wiley et al.: Stranding and mortality of Megaptera novaeangliae 



197 



the length of coastline along the state. This is referred 

 to as the stranding incidence ratio (SIR). Length of 

 coastline was calculated from Ringold and Clark ( 1980). 



*ANJ 



MD 



Chesapeake '. 

 Bay 



VA 



, Ca P e . oa 

 «»\Hatteras * 



NC 



Figure 1 



Locations of humpback whale, Megaptera nov- 

 aeangliae, strandings from New Jersey to south- 

 ern Florida, 1985 through 1992. 



Reproductive class was inferred from body-length 

 data. Animals of less than 8 m in length were con- 

 sidered to be dependent, nursing calves (Nishiwaki, 

 1959; Rice, 1963). We considered newly independent 

 calves to be animals greater than 8.0 m but less than 

 9.9 m (calculated from Katona et al., 1983 2 ). Males 

 between 9.9 m and 11.6 m and females between 9.9 

 m and 12.0 m were considered sexually immature 

 but not newly independent. Animals greater than 

 11.6 m (males) and 12.0 m (females) were considered 

 sexually mature (Nishiwaki, 1959; Rice, 1963). 



The Mann-Whitney [/-test (Sokal and Rohlf, 1981) 

 was used to test for differences between the number 

 of strandings that occurred in the period 1985—88 

 versus 1989-92. Time periods were chosen to coin- 

 cide with reported changes in observations of live 

 animals in the same region (Swingle et al., 1993). 

 The hypothesis that strandings occurred randomly 

 throughout the study area was tested by chi-square 

 analysis in a 2x2 contingency table (Sokal and Rohlf, 

 1981). The hypothesis that stranding events were not 

 influenced by season was tested by chi-square analy- 

 sis. Seasons were winter (January-March), spring 

 (April-June), summer (July-September) and fall 

 (October-December). Seasonal groupings were con- 

 structed so that the winter season would approxi- 

 mately coincide with the period of peak humpback 

 occupancy of the breeding grounds, as reported by 

 Mattila and Clapham (1989). The hypothesis that 

 stranding occurrence was not influenced by sex was 

 tested by chi-square analysis in a 2x2 contingency table. 



Factors relating to mortality were taken from the 

 written reports of on-site stranding response person- 

 nel from the Northeast and Southeast Regional 

 Stranding Networks or, when not available, from the 

 synthesis of such reports contained in the MMER 

 The experience of stranding network response per- 

 sonnel is variable, and factors contributing to death 

 or interpretation of bodily injury can be subject to 

 debate. If on-site investigators recorded references 

 to rope marks, propeller marks, broken bones, large 

 gashes, etc., or directly suggested ship strike or en- 

 tanglement as a potential cause of death, we attrib- 

 uted the death to possible anthropogenic causes. All 

 mortality not suggesting anthropogenic trauma were 

 grouped into a "natural" mortality category. This in- 

 cluded animals that were euthanized but showed no 

 other indications of human interaction. If a necropsy 

 was conducted and no mention was made of body 

 trauma, we assumed natural mortality. Carcasses 

 that were reported to be in advanced stages of de- 

 composition were eliminated from consideration. 



2 Calculated as length at birth, 4.5 m; growth rate, 45 cm per 

 month; 12 month growth period = 9.9 m. 



