Rodgveller et al: Evidence of hook competition in longline surveys 
371 
- 70 -* 
1978 1982 1986 1990 1994 1998 2002 
Figure 4 
Percent deviation from the average catch rate of the time 
series for grenadier (giant grenadier [ Albatrossia pectoralis]), 
rockfish (shortraker [Sebastes borealis ] and rougheye [Sebastes 
aleutianus] rockfish), and sablefish (Anoplopoma fimbria) 
caught during the National Oceanic and Atmospheric Admin- 
istration Alaska Fisheries Science Center annual sablefish 
longline surveys from 1979 to 2003. (A) includes only the 
grenadier preferred depths and (B) includes only the preferred 
rockfish depths. Data from the East Yakutat and Southeast 
management area are presented as an example. The lines 
that intersect the y-axis at 0 indicate the average catch rate; 
anything below this line is lower than average and anything 
above is higher than average. 
fish and grenadier are found in the same habitats 
consistently. Additionally, the correlation between 
catch rates of these species in the trawl gear was 
positive, also indicating they both use the same 
habitats. Because catch rates from multiple fixed 
stations, where the same habitats are sampled 
each year in several geographic areas for 25 years, 
consistently show negative correlations, it is un- 
likely that the negative relationship is due to dif- 
fering habitat preferences. For example, the time 
series of catch rates at all stations in the East 
Yakutat and Southeast area shows that grenadier 
and rockfish densities are above average when 
sablefish densities are below average and vice 
versa (Fig. 4). This result cannot be explained by 
differing habitat preferences because the same 
habitats were sampled each year. 
Sablefish and rockfish catch rate correlations 
in the trawl gear showed that there may be three 
distinct habitat types at rockfish preferred depths, 
and that negative correlations on the longline sur- 
veys may be partially influenced by habitat pref- 
erences of sablefish and rockfish at these depths. 
Hov/ever, although there appeared to be some dif- 
ferent habitat preferences, relatively large num- 
bers of both species were caught in most hauls. 
Also, nonparametric correlations of trawl catch 
rates were not significantly different from zero, 
indicating that any differences in habitat prefer- 
ences were not great enough to cause negative cor- 
relations for the longline gear. Additionally, just 
as with sablefish and grenadier catch rate trends 
for longline gear, the rockfish catch rates were 
above average in years when sablefish were be- 
low average, and vice versa, even when the same 
habitats were sampled each year. 
It is also unlikely that sablefish are directly 
competing with grenadier and rockfish, caus- 
ing negative correlations. If this were true there 
would be negative correlations for the trawl gear 
as well as the longline gear. Also, if sablefish 
were pushing grenadier or rockfish out of their 
preferred depth range, grenadier and rockfish 
catch rates would increase at other depths when sable- 
fish catch rates increased; however, this has not been 
observed. Or, if their population numbers were actu- 
ally being depressed because of competition, it would 
likely take longer time periods for adult populations 
of long-lived fish species like grenadier and rockfish 
to rebound. 
Although selective differences between longline and 
trawl gear for species and size have been reported for 
the Atlantic Ocean (e.g., Hovgard and Riget, 1992; Huse 
et ah, 2000), both gear types caught adult cohorts in 
these studies. Although gear selectivity of longline and 
trawl gear in Alaska has not been compared in field 
studies, all four of the species analyzed in this article 
were caught in longline and trawl gear in significant 
numbers. Very few small fish are caught, indicating 
that both gears catch adult fish (Fig. 6). Moreover, this 
finding indicates that the same cohorts are being se- 
lected by both gear types. Therefore, correlation coef- 
ficients between these species in these gear types can 
be compared fairly. 
Some environmental and fish specific variables, such 
as feeding history (e.g., Lpkkeborg et al., 1995; Stoner 
and Strum, 2004) may affect the strength of the effect 
of competition for hooks; however, we could not address 
these variables. It is possible, however, that at warmer 
water temperatures fish would have greater metabolic 
demands and increased hunger, causing more intense 
competition if prey availability was similar in each 
scenario (i.e., at colder and warmer temperatures). It 
is likely that the ability of fish to locate bait and their 
hunger affects catch rates. To assess the effects of these 
variables on competition, more laboratory studies of fish 
behavior would be needed. 
