272 
Fishery Bulletin 96(2), I 998 
methods (Lennert et al., 1994; Perkins et al. 5 6 7 8 ; 
Julian 6, 7> 8 ). Separate estimates of entanglement are 
provided for species that had individuals released 
alive. This paper documents incidental marine mam- 
mal, turtle, and seabird kill estimates in these two 
California gillnet fisheries, based on data from the 
NMFS observer program and CDFG effort estimates 
for the period July 1990 through December 1994, and 
documents the process and methods leading to these 
estimates. 
Methods 
Data collection 
National Marine Fisheries Service observer data, 
daily logbooks of commercial gillnet fishermen, and 
receipts of landed fish sales were used in marine 
mammal mortality estimation. NMFS observer data 
were collected by trained technicians aboard com- 
mercial gillnet fishing boats that had a Marine Mam- 
mal Protection Act (MMPA) Exemption Permit and 
that met minimum U.S. Coast Guard safety stan- 
dards. There were two general observation catego- 
ries: observation of randomly selected trips and ob- 
servation of approximately every fifth vessel trip. In 
the setnet fishery, systematically selected trips were 
further divided. Notification prior to the setting of 
nets resulted in a preset, systematic observation 
rather than a postset, systematic observation where 
notification to the vessel was given after the nets were 
set (see “Discussion” section). NMFS observers re- 
corded data on location, date, marine mammal en- 
tanglements, including location of mammals in the 
5 Perkins, P., J. Barlow, and M. Beeson. 1992. Pinniped and 
cetacean mortality in California gillnet fisheries: 1991. Inter- 
national Whaling Commission Scientific Committee working 
paper SC/44/SM14. [Available from Southwest Fisheries Sci- 
ence Center, National Marine Fisheries Service, PO. Box 271, 
La Jolla, CA 92038.] 
6 Julian, F. 1993. Pinniped and cetacean mortality in Califor- 
nia gillnet fisheries: preliminary estimates for 1992, rev. 2/ 
94. International Whaling Commission Scientific Committee 
working paper SC/45/022. [Available from Southwest Fisher- 
ies Science Center, National Marine Fisheries Service, P.O. Box 
271, La Jolla, CA 92038.] 
7 Julian, F. 1994. Pinniped and cetacean mortality in Califor- 
nia gillnet fisheries: preliminary estimates for 1993. Interna- 
tional Whaling Commission Scientific Committee working pa- 
per SC/46/011. [Available from Southwest Fisheries Science 
Center, National Marine Fisheries Service, P.O. Box 271, La 
Jolla, CA 92038.] 
8 Julian, F. 1995. Cetacean and pinniped mortality in Califor- 
nia gillnet fisheries: preliminary estimates for 1994. Interna- 
tional Whaling Commission Scientific Committee working pa- 
per SC/47/05. [Available from Southwest Fisheries Science 
Center, National Marine Fisheries Service, P.O. Box 271, La 
Jolla, CA 92038.] 
net (by thirds of the net — vertically and horizontally), 
gear, bycatch, and target species catch for each net 
pull observed during a trip (Lennert et al., 1994). 
Observers recorded twelve net-related parameters 
for drift and set nets. They were net type (set, drift, 
float, or trammel), net material (monofilament, multi- 
filament, or a combination), net strength (pounds test 
or twine size depending on strength code), strength 
code, net length (fathoms), net depth (number of 
meshes), stretched mesh size (inches), extender 
length (feet, float, and drift nets only), hanging line 
material (synthetic or natural fiber), percent slack 
in net, number of meshes hanging (between knots to 
the cork line), and hanging length (distance between 
knots on the cork line in inches). Not infrequently, a 
drift or set net will consist of panels of varying char- 
acteristics. In this case, observers would record char- 
acteristics on up to 5 different panels. Net charac- 
teristics for both fisheries are summarized in Table 1. 
Although the variability in these characteristics con- 
tributes somewhat to the variability in mortality es- 
timates, the significant factors for mortality estima- 
tion are the amount of effort and the general loca- 
tion of the effort. For some species, e.g. pinnipeds, 
quarter of the year is also significant (Perkins et al. 5 ). 
Collected data were entered into a database file, 
checked for accuracy, and tabulated for mortality 
estimation (Tables 2-5). After an initial six month 
period, this observation method continued unchanged 
in both fisheries. Realized observation rates varied 
between 4.4% and 17.9% yearly, but observation rates 
were more variable if stratified by area and quarter. 
Observation in the driftnet fishery continued through 
December 1995, whereas observation in the setnet 
fishery terminated by July 1994 because of a signifi- 
cant decrease in fishing effort in that fishery (due to 
regulations that restricted areas open to gillnet fish- 
ing). Observer data were complemented by informa- 
tion from vessel logbooks and landing receipts (i.e. 
receipts from sales of landed fish). 
Vessel logbooks were submitted monthly to CDFG 
and constituted the major source of information for 
estimation of total effort. Data for each logbook en- 
try included date, vessel and permit identification, 
area fished by CDFG block number (Lennert et al. 
1994), gear, number of sets made, and number and 
species of fish caught (CDFG blocks are typically 10' 
square; larger blocks are defined for areas further 
from shore). Logbook information was entered into a 
database by technicians and checked for accuracy by 
biologists. Fish species targeted for catch by the fish- 
ermen were determined and assigned to each data 
entry by CDFG personnel according to fish caught, 
gear used, and other pertinent factors. Purchases of 
landed fish by commercial fish buyers were recorded 
