Yochum et al. A comparison of methods for evaluating mortality of discarded Cancer magister 
127 
mortality predictor (RAMP) approach has been devel¬ 
oped (Davis and Ottmar, 2006). This method relates 
impairment in reflex actions to the probability of mor¬ 
tality. This relationship is established by first select¬ 
ing reflexes that give a consistent, instantaneous, and 
involuntary response to a stimulus. After enduring the 
set of stressors associated with fishing and discarding 
(either directly during fishing operations or through 
simulation in a laboratory), the animals can be evalu¬ 
ated by determining whether each of these reflexes is 
present. To relate the levels of reflex impairment (i.e., 
the total number of missing reflexes) to the probability 
of mortality, survival must be determined by using one 
of the methods described previously (captive holding, 
tagging, etc.). A “RAMP relationship” is subsequently 
created by determining the number of animals, for 
each level of impairment, that die out of those moni¬ 
tored for mortality. Once established, this relationship 
can be applied to reflex-impairment data collected over 
the spatial and temporal extent of a fishery, making 
the mortality rates more representative of the fishery 
at large. 
The efficacy of a RAMP relationship is linked with 
the reliability of the predicted delayed mortality rates. 
Therefore, it is essential to consider method-specific 
biases and limitations. For example, it is important to 
evaluate the contribution to mortality from tagging for 
identification (Tegelberg and Magoon, 1971; Wassen- 
berg and Hill, 1993) or telemetry studies, or the contri¬ 
bution that is due to the effect of captivity in research 
that determines mortality through holding animals 
(Yochum et al., 2015; Yochum et al., 2017). In addition, 
methods that do not allow long-term monitoring may 
underestimate mortality rates caused by chronic, en¬ 
during impairment (Wassenberg and Hill, 1993; Berg- 
mann and Moore, 2001). 
Mortality rates estimated through captive holding 
are biased by the unnatural environment or holding 
conditions (or both) (Yochum et al., 2015; Yochum et 
al., 2017). Mortality rates could be overestimated be¬ 
cause of agonistic interactions or predation among cap¬ 
tive animals, suboptimal temperature or water quality, 
density of animals in holding enclosures, or failure to 
meet other biological or environmental requirements 
of the captive animal (Simonson and Hochberg, 1986; 
Kondzela and Shirley, 1993; Wassenberg and Hill, 
1993; Spanoghe and Bourne, 1997; Portz et al., 2006; 
Weltersbach and Strehlow, 2013). Alternatively, mor¬ 
tality rates could be underestimated because mortal¬ 
ity resulting from an animal’s inability to obtain food 
or avoid predation is not incorporated in such stud¬ 
ies (Durkin et al., 1984; Uhlmann et al., 2009; Benoit 
et al., 2010; Urban, 2015). Similarly, the effect of dis¬ 
placement from suitable habitat and the impact from 
the return to water after capture and handling onboard 
are not often considered. 
Despite limitations of captive holding, RAMP rela¬ 
tionships have commonly been created by using on¬ 
board holding tanks or laboratory-based holding fa¬ 
cilities (Davis, 2007; Stoner et al., 2008; Humborstad 
et al., 2009; Barkley and Cadrin, 2012; Stoner, 2012; 
Hammond et al., 2013; Rose et al., 2013; Depestele et 
al. 2 ; McArley and Herbert, 2014; Humborstad et al., 
2016). Preference for this approach is largely due to 
advantages over alternative methods, which include 
providing scientists with control and allowing them to 
differentiate causes of mortality, observing degradation 
in health and changes in behavior, and knowing the 
time of death (Davis and Ryer 3 ). Because short-term 
laboratory holding is frequently used to estimate de¬ 
layed discard mortality rates, we conducted a field vali¬ 
dation study to assess the limitations of this approach. 
This was done by comparing results from 2 RAMP 
studies that evaluated delayed mortality, one through 
tag-returns and one by using laboratory holding. The 
former study and comparison are described here, and 
the latter was reported by Yochum et al. (2017). Tag¬ 
ging was selected for the comparison because it allows 
an evaluation of long-term mortality rates and allows 
animals to experience more natural conditions after 
release. 
The Oregon (U.S.A.) commercial and recreational 
Dungeness crab (Cancer magister) fisheries were se¬ 
lected for this study because of their high level of dis¬ 
card, and because Oregon fishermen have experience 
with tagging studies for these crab, which yielded high 
tag-return rates (Jow, 1965; Snow and Wagner 4 ; Dem- 
ory 5 ; Hildenbrand et al. 6 ). Additional factors that make 
Dungeness crab a good candidate for comparing dis¬ 
card mortality in situ and in the laboratory include ev¬ 
idence that they 1) are agonistic and cannibalistic (Ja¬ 
coby, 1983; Fernandez, 1999; Barber and Cobb, 2007), 
2) are often preyed upon by seabirds and California sea 
lions (Zalophus californianus) upon their return to the 
water, and 3) like many crustaceans, can be difficult 
to maintain in captivity owing to stress, disease, and 
sensitivity to temperature and water quality (Burton, 
2001; Barrento et al., 2008). 
In Oregon, only male Dungeness crab at or above 159 
mm (6.25 in) carapace width (measuring the straight 
line distance across the carapace, shell edge to shell 
2 Depestele, J., E. Buyvoets, P. Calebout, M. Desender, J 
Goossens, E. Lagast, D. Vuylsteke, and C. Vanden Ber- 
ghe. 2014. Calibration tests for identifying reflex action 
mortality predictor reflexes for sole (Solea solea ) and plaice 
(Pleuronectes platessa ): preliminary results. ILVO-com- 
mun. Rep. 158, 30 p. [Available from website.] 
3 Davis, M. W., and C. H. Ryer. 2003. Understanding fish 
bycatch discard and escapee mortality. AFSC Q. Rep. 2003, 
Jan-Mar, 9 p. Alaska Fisheries Science Center, Seattle, WA. 
[Available from website.] 
4 Snow, C. D., and E. J. Wagner. 1965. Tagging of Dunge¬ 
ness crabs with spaghetti and dart tags. Fish Comm. Or¬ 
egon, Res. Briefs 11:5—13. 
5 Demory, D. 1971. Crab movement off Port Orford, Ore¬ 
gon. Shellfish Invest. Inf. Rep. 70-7. Res. Div., Fish Comm. 
Oregon, Salem, OR. 
6 Hildenbrand, K., A. Gladics, and B. Eder. 2011. Crab tag¬ 
ging study: adult male Dungeness crab (Metacarcinus ma¬ 
gister) movements near Reedsport, Oregon from a fisheries 
collaborative mark-recapture study, 21 p. Oregon Wave En¬ 
ergy Trust, Portland, OR. 
