A decline in the abundance of 

 harbor porpoise, Phocoena 

 phocoena, in nearshore waters 

 off California, 1986-93 



Karin A. Forney 



Southwest Fisheries Science Center 

 National Marine Fisheries Service, NOAA 

 RO. Box 271, La Jolla. California 92038 



Harbor porpoise, Phocoena pho- 

 coena, have been caught incidentally 

 in set gill nets off the central Cali- 

 fornia coast since at least 1958 

 (Norris and Prescott, 1961). The 

 annual mortality of harbor porpoise 

 caught in gill nets in this region 

 peaked in the mid-1980's and then 

 gradually declined (Barlow and 

 Forney, 1994) as fishing effort de- 

 creased following the implementa- 

 tion of restrictions and area clo- 

 sures in order to protect marine 

 mammals, sea birds, and sport fish- 

 eries (Barlow et al., 1994). In 1986, 

 a series of aerial line-transect sur- 

 veys was initiated jointly by the 

 California Department of Fish and 

 Game and the National Marine 

 Fisheries Service to monitor trends 

 in abundance of the central Cali- 

 fornia harbor porpoise population. 

 Harbor porpoise in this region are 

 managed separately from animals 

 found off northern California and 

 Oregon, because movement of ani- 

 mals along the U.S. West Coast 

 appears limited, and fishery-in- 

 duced mortality is restricted to cen- 

 tral California (Barlow and Hanan, 

 in press). An analysis of covariance 

 model applied to the first five an- 

 nual surveys (1986-90) failed to 

 detect a significant trend in abun- 

 dance (Forney et al., 1991); how- 

 ever, simulations revealed that sta- 

 tistical power to detect trends, given 

 the level of variability observed in the 



741 



time series, was low with only five 

 survey years. A minimum often sur- 

 vey years was estimated as necessary 

 to provide sufficient power. 



Additional surveys utilizing the 

 same methodology were conducted 

 in 1991 and 1993, completing an 

 eight-year time series. In updating 

 the analysis of trends in central 

 California harbor porpoise abun- 

 dance for the period 1986-93, 1 an- 

 ticipated either that 1) no signifi- 

 cant trend would be identified be- 

 cause of low power, or 2) an increase 

 in abundance might be detected 

 because the population was ex- 

 pected to be recovering after the 

 reduction in fishery-induced mor- 

 tality. However, a declining trend 

 in central California harbor por- 

 poise abundance was identified for 

 the period 1986-93. Because of the 

 surprising nature of this result and 

 because of the management impli- 

 cations, this report presents the 

 updated 1986-93 analysis and in- 

 cludes additional data from the 

 1989-1993 aerial surveys in north- 

 ern California for the first time. 



Methods 



A complete description of both field 

 and analytical methodology can be 

 found in Forney et al. (1991), and 

 only a brief summary of the meth- 

 ods used is provided here. 



Field methods 



Aerial line-transect surveys were 

 conducted from late summer to 

 early fall (15 August through 15 

 November) of the years 1986-91, 

 and 1993. In each survey year, a set 

 of 26 transects between Point Con- 

 ception and the Russian River (Fig. 

 1A) was replicated as often as 

 weather permitted (generally 4-8 

 times) to monitor the central Cali- 

 fornia harbor porpoise population. 

 Beginning in 1989, a set of 17 ad- 

 ditional transects between the Rus- 

 sian River and the California-Or- 

 egon border (Fig. IB) was surveyed 

 1-3 times per field season to moni- 

 tor the northern California popu- 

 lation. The transects followed a zig- 

 zag pattern designed to survey sys- 

 tematically between the coast and 

 the 92-m (50-fathom) isobath, 

 which is the depth range in which 

 the majority of harbor porpoise are 

 expected to be found in this region 

 (Barlow, 1988). The only deviation 

 from this design occurred outside 

 San Francisco Bay, where the 92- 

 m (50-fathom) contour is located too 

 far offshore for safe operation of the 

 survey aircraft; in this region, the 

 transect lines extended only to the 

 55-m (30-fathom) contour. Total 

 transect length was 916 km, and 

 under good weather conditions all 

 transects could be surveyed in two 

 days. The survey platform was a 

 high-wing, twin-engine Partenavia 

 P-68 aircraft outfitted with two 

 bubble windows for lateral viewing 

 and with a belly port for downward 

 viewing. The survey team consisted 

 of three observers (situated left, 

 right, and belly) and one data re- 

 corder. Line-transect methods were 

 followed with sighting distances 

 calculated from the angle of decli- 

 nation to the sighting (obtained 

 with a hand-held clinometer) and 

 from the aircraft's altitude. Surveys 

 were flown at about 167-185 km/hr 

 (90-100 knots) airspeed and 213 m 



Manuscript accepted 8 June 1995. 

 Fishery Bulletin: 741-748 (1995). 



