after tagging and showed no swelling whatsoever despite some attachments bending 60 

 degrees. "Wart" (PTT #839, NEA #1140) was seen 59 days after tagging (16 days after 

 tag loss). There were no signs of swelling or infection and only a single 1 cm diameter 

 white scar where the tag had been. 



Loss of tags may be ascribed (at least in part) to: 1) tissue rejection of a foreign 

 object, 2) high levels of vigorous breeding activity and, 3) females with active calves. We 

 also saw evidence of bottom contact when whales surfaced with mud on their heads. 



Transmitter Performance 



In 1989 and 1990 there were no PTT electronic failures, and frequency stability 

 (necessary for accurate locations) was very good. While the transmitters performed 

 exceedingly well in general (see Table 2), transmitted data were subject to error. Errors 

 can be induced by interference during transmission from the PTT to the satellite or from 

 the satellite to an earth station receiver, and can distort individual bits or whole sections 

 of the message. The resulting erroneous message may or may not have reasonable 

 values and must, therefore, be confirmed by independent means. Summary data for each 

 summary period were confirmed when we had multiple transmissions. Alternative 

 strategies were used to confirm discrete information. 



1989 



In 1989 a CRC code was a part of each message and was used to determine 

 errors in the discrete portion of the message (Appendix A). The CRC code was 

 calculated by the microprocessor in the PTT and sent as a part of each transmission so 

 errors could be detected easily. Figure 10 illustrates the number of messages received 

 during each 4-hour summary period from PTT #843. This was the only 1989 PTT to 

 achieve any long-term track. The transmission scheme (synchronization of transmissions 

 with satellite passes) worked well for south-north passes, and resulted in a high number 

 of messages between 1600 - 2400 GMT. However, a failure of the software to 

 coordinate with north-south orbits resulted in very few data points between 0400 and 

 1200 GMT. A total of 304 messages were received for PTT #843, and of these 33 

 (11%) contained errors detected by the CRC code. 



1990 



Of nine whales tagged in 1990, two did not provide any useful data, while seven 

 provided from 22 - 665 messages each for a total of 1,466 messages (Table 3). Individual 

 PTTs operated from 3 - 43 days for a total of 160 "whale-tracking days." This body of 

 information represents the largest data set available on dive durations, movements and 

 long-term monitoring of any species of cetacean. In 1990, the Telonics ST-6 software did 

 not include error detection codes. Instead, we checked to determine whether discrete 

 dive durations were feasible and also used range limiters (see Appendix B). If any of the 

 discrete or summary information was found to be in error, the entire line of discrete 



24 



