146 
Fishery Bulletin 120(2) 
1000 2000 km 
140°W 120°W 
Figure 5 
Coefficient of variability (CV) estimated by using a jackknife procedure and the generalized 
additive model with 6 knots for the thin-plate regression spline, for each 1°-by-1° block over 
the study area in the eastern North Pacific Ocean where opah (Lampris spp.) were caught 
with deep-set pelagic longlines during 2009, 2010, 2017, and 2018. Darker areas of the map 
show higher levels of variability in the data. The CVs were calculated as (standard deviation/ 
mean)100. 
Table 2 
Number of sets and catch per unit of effort (CPUE), mea- 
sured as number of individual opah per 1000 hooks, for 
the deep-set pelagic longline fishery operating in areas 
in the eastern North Pacific Ocean dominated by bigeye 
Pacific opah (Lampris megalopsis), smalleye Pacific opah 
(L. incognitus), or mixed species. Values are given by the 
number of knots (k) for the isotropic thin-plate regres- 
sion spline in each generalized additive model: 4 (low), 
6 (medium), and 8 (high). Some cells are blank because 
values are included only when data for more than 20 sets 
were available. 
Bigeye Smalleye Mixed species 
No. of No. of No. of 
k sets CPUE sets CPUE sets CPUE 
Sampling period: 1996-2013 
4 236,152 0.43 
6 233,340 0.42 
8 227,167 0.40 
Sampling period: 2014-2018 
4 80,382 0.34 364 
6 79,114 0.32 616 
8 77,311 0.31 584 
of the main Hawaiian Islands and that the range of small- 
eye Pacific opah does not extend across the Pacific Ocean. 
Although pelagic longline fisheries cannot operate within 
200 nautical miles of the U.S. West Coast, data are available 
for this region from other fisheries and research cruises. 
Catch composition in the California Current from the state 
of Washington to Mexico is nearly 100% smalleye Pacific 
opah (Hyde et al., 2014). It should be noted that most opah 
sampled were relatively large (>80 cm fork length) and pre- 
sumed to be adults, and even less information is available on 
the geographic range of larvae and juveniles. Additional 
species-specific information across size classes is needed to 
determine the full distribution of each species. 
The movement patterns of highly migratory species and 
the dynamic nature of their habitat make it difficult, if 
not impossible, to establish static boundaries of species 
ranges (Maxwell et al., 2015; Dunn et al., 2016). Although 
this study classified fixed areas for each species within the 
range sampled, the distribution of species can shift over 
time with variations in abundance, natural climate vari- 
ability on interannual (e.g., El] Nifio—Southern Oscillation) 
and decadal (e.g., Pacific Decadal Oscillation) timescales, 
as well as with longer-term climate change. 
A number of factors indicate the potential for shifts in 
the distribution of opah. First, the distributions of bigeye 
tuna and swordfish, which have habitats and diets that 
overlap with those of opah, have shifted with climate 
