132 
Fishery Bulletin 11 6(2) 
Table 3 
Information on Dungeness crab (Cancer magister) that were tagged and released between 
October 2012 and April 2014 and differences between crab that were and were not recap¬ 
tured and their tags returned (all release events combined) for the commercial ocean fish¬ 
ery off Oregon and recreational fisheries in Yaquina Bay, Oregon, by boat and shoreside. 
Information includes the number of days at large (time between release and recapture 
of crab for which tags were returned and between release and end of the study for crab 
for which tags were not returned), carapace width, reflex impairment score (score), water 
depth at the location the tagged crab was released, and the number of days from the 
opening of the fishery in which the tagged crab was released (commercial fishery only). 
Tag returned Tag not returned 
Mean 
Range 
Mean 
Range 
Commercial 
Days at large 
107 
2-468 
499 
209-590 
Carapace width (mm) 
155 
138-171 
154 
52-193 
Score 
0.14 
0-2 
0.25 
0-6 
Depth at release (m) 
49 
5-150 
51 
5-150 
Days from opening of fishery 
66 
0-198 
83 
0-198 
Recreational—boat 
Days at large 
78 
0-449 
448 
136-674 
Carapace width (mm) 
137 
104-183 
127 
82-167 
Score 
0.15 
0-3 
0.12 
0-6 
Depth at release (m) 
14 
4-28 
14 
4-28 
Recreational—shoreside 
Days at large 
79 
0-163 
268 
146-672 
Carapace width (mm) 
124 
106-154 
119 
86-159 
Score 
0.03 
0-1 
0.16 
0-3 
Logistic regression modeling in R software was used 
to determine whether there is a relationship between 
drop height and the probability that a crab died within 
5 d of holding, and whether any other variables influ¬ 
enced the likelihood of survival, including 1) the No¬ 
vember 2013 vs. April 2014 experiment (the difference 
between the 2 being whether or not the crab was kept 
in water before being dropped), 2) the side on which 
the crab landed (dorsal, ventral, or side), 3) carapace 
width (in millimeters; continuous), 4) sex, 5) shell 
hardness (soft or hard), and 6) whether or not the cara¬ 
pace cracked as a result of the drop. Model coefficients 
were estimated by using maximum likelihood (Ramsey 
and Schafer, 2002) based on the fate (mortality or sur¬ 
vival) of individual crab that were held after the drop 
experiments. Akaike information criteria were used to 
determine the most parsimonious model for the data. 
Results 
Tag-return study 
A total of 4093 live crab intended for discard were 
tagged and released, and 430 tags were returned by 
15 August 2014 (11%; Table 2). Tags were returned by 
207 different fishermen, ranging from 1 to 60 tags/fish¬ 
erman (average: 2 tags/fisherman; mode: 1 tag/fisher¬ 
man). When the date of recapture was known, 3 tagged 
crab were recaptured the same day on which they were 
released, 53 within the first week at large (12% of the 
returns), 142 within the first 30 d (33%), 295 within 
the first 100 d (69%), and 415 within a year (97%). 
The longest time between release and return was 468 
d (Table 3). On average, returned tags were at large 
for 98 d. With respect to movement, the Euclidean dis¬ 
tance from release to recapture locations ranged from 
0 to 150 km (Fig. 2), and there was no relationship 
between days-at-large and total distance. Some of the 
farthest distances (>100 km) occurred within a week at 
large, and some of the shortest distances (<1 km) were 
detected after a year. The majority (73%) of recaptured 
crab, however, were caught less than 10 km from the 
release location (59% for the commercial fishery; 98% 
and 100% for the recreational by boat and shoreside 
fisheries, respectively). For the recreational fisher¬ 
ies, 43% (by boat) and 58% (shoreside) of recaptures 
were within 1 km. For the commercial fishery, 93% of 
crab were recaptured within 50 km, indicating that 
the probability of recapture was likely not reduced by 
animals moving out of the area. There was a pattern 
that recaptures were closer to shore at the start of the 
