TRACKING HERRING SCHOOLS WITH A HIGH-RESOLUTION SONAR - VARIATIONS INHORIZONTAL AREA AND RELATIVE ECHO INTENSITY

Fourteen herring schools off northern Norway were tracked for about 1 hour each by the 95 kHz Simrad SA950 sonar onboard RN ''G.O. Sars'' in May 1994. The sonar was connected to a work station that contained so ftware for reading the echo telegrams of the sonar, printing of an ech ogram, automatic detection and measurement of schools, and logging of the sonar data. The horizontal area and relative echo intensity of the schools were recorded ping by ping as well as swimming depth and dist ance and bearing to the vessel. The position, speed and heading of the vessel were also recorded. Inter-and intra-school events as interpret ed from the sonar display were recorded in a separate protocol during the school tracking. The recorded horizontal area and relative echo in tensity of the schools varied considerably. Linear models with school area or relative echo intensity as dependent variables, and with range , tilt, speed and swimming angle relative to the sonar beam as continu ous effects, did not explain more than 15% and 30% of the observed var iations for most schools, respectively. There was a negative correlati on between relative echo intensity and range for all schools. Inter-an d intra-school events occurred at average rates of about 14 minutes, a nd inter-school events such as split and joint influenced school size. The sound absorption and the degree to which the sonar beam insonifie s the schools in the vertical plane are proposed as the major sources of variation for recorded horizontal area and relative echo intensity of the schools.