Elements of vitality testing and delayed mortality in fisheries

Conceptual diagram outlining elements for vitality testing and delayed mortality in fisheries. Fish are captured and environment sampled. Fish become stressed which is measured as impairment from control health by observing reflex actions and injury types. Stressed fish are held for captive observation to determine delayed mortality. Bias and error can be introduced by initial impressions of vitality before testing reflex actions and injury, by differing observer scoring opinions, and by holding conditions that are stressful for the fish. 

Scoring vitality impairment is most difficult when observer decision is used. Training observers is a key part of RAMP development. Reflex actions (RA) are clearly present in control animals, and observers do not need decisions to score present. As impairment increases, scoring RA requires increasing observer decisions about whether sampled RA are present. The decision can be based on how control RA appear to trained observers. Each observer will have different opinions that can be influenced by their initial impressions of the animal and of the stressor treatments the animal has been exposed to.

Initially after stress induction, RA impairment increases and mirrors stress levels, while mortality is not evident. When animals reach a critical impairment level, replicates begin to show mortality, which increases rapidly over small changes in RA score. At highest levels of impairment decisions are less frequent as the animal ceases general movement and responsiveness.

Belly up: Righting reflex action time to recovery correlated with delayed mortality?

Upside down fish in market tank (Hong Kong)

RAMP incorporates presence/absence of several reflex actions and injuries to measure vitality impairment and potential delayed mortality. A simpler method may be possible by measuring time for recovery of orientation when fish are placed upside down in water. This method can be tested.

Place a fish upside down in water and observe the time until the fish returns to normal orientation. This duration is a measure of vitality impairment. Longer recovery times indicate greater vitality impairment and data can be included in statistical models for relationships among fishery stressors, injury, righting time, and delayed mortality. We can test the relationship between righting impairment and delayed mortality. 

Righting reflex action is a central behavior that is the nexus of neural, muscle, and organ actions and is intimately linked with loss of physiological regulation associated with stressor exposure.  Olfactory impairment is another example of a central function that is correlated with delayed mortality (in humans, Pinto 2014).

Body orientation is a sensitive measure of fish consciousness. Presence or absence of righting can be included in the RAMP score. Loss and recovery of orientation is a well known symptom for induction of and recovery from fish anesthesia and is used as an indicator of morbidity and delayed mortality in stress experiments (Davis and Ottmar 2006, Szekeres et al. 2014, Raby et al. 2015).  

Measuring replicate animals for the time to righting recovery and delayed mortality after a stressor experiment can test the correlation between righting impairment and delayed mortality. If the correlation between righting and delayed mortality is valid and strong, then we have a rapid method for predicting discard mortality on board fishing vessels without need for holding or tagging fish to confirm their survival. Research groups on fishing vessels can observe fish during catching, landing, sorting, and discarding under differing stressors; seasons, water temperatures, tow durations, catch quantities, species mixes, and sorting times.

Survival of schooling small pelagic fish discarded from purse seine fisheries

Greenback horse mackerel, Trachurus declivis 

Purse seine fishing

Vitality impairment and RAMP can be used to determine survival of discarded schooling small pelagic fish in purse seine fisheries.  When catch is too large, fish are “slipped” for the net and discarded. These discarded fish are usually exposed to some level of hypoxic conditions associated with crowding in the purse seine. Elevated temperature in surface water may be a stressor. Skin abrasion and scale loss can occur in the net. Many small pelagic species (mackerel, sardine, anchovy, smelt, herring) caught in purse seines are obligate or facultative schoolers that reflexively form groups mediated by the optomotor response. Vitality impairment can be tested for individual fish or groups. See Davis and Ottmar, 2006 for testing groups of free-swimming fish. Schooling fish seek the company of species mates, so testing groups of schooling fish is probably the most informative method. How is this testing done and linked with delayed mortality in captive observation tanks?    

RAMP links vitality impairment scores with delayed mortality scores. The RAMP estimate for delayed mortality is only as good as the mortality estimates from captive observation or tagging experiments. How many replicates are needed in captive observation experiments? The purse seine schooling species need to be held in groups. A replicate group size of ten fish is good for schooling. These fish must be held in good water quality and circulation, in a circular tank size that allows schooling. For initial RAMP formulation, you will need 10 replicate groups of ten fish each. Observations of reflex action impairment and delayed mortality should be made over a range of stressor intensities that result in delayed mortality of 0 to 100%. Then replicate vitality impairment scores are linked with replicate delayed mortality scores to form the RAMP which can be validated with further experiments and replication.

Suuronen, 2005  Stressors in capture and escape of fisheries.

Fish can be sampled from any point on the fishing process, depending on the stressors of interest. Reflex action testing can be made on a group of fish held in a circular observation tank big enough for schooling (See Davis and Ottmar, 2006).  Possible reflex actions for testing include: orientation; schooling; rheotaxis; startle response to sound or light; swimming to bottom of tank. Injuries can also be noted; abrasion, scale loss. After testing the replicate group can then be placed in a holding tank and monitored for delayed mortality through five to ten days. 

Herring lose schooling, orientation, and tail beat frequency increases as the purse seine is drawn smaller (Morgan, 2014). Fatigue and hypoxia are possible stressors in purse seines (Tenningen 2014).

For discard species caught in purse seines that are not schooling fish, or are larger schooling fish, individual fish can be tested for vitality impairment.  Reflex actions tested can include: body flex, orientation, eye roll, operculum or mouth clamp, tail grab, righting, startle.  These fish can be tagged for identification and held together for five to ten days in tanks to determine delayed mortality.

Flexibility of using RAMP to determine bycatch mortality rates for Tanner crab caught in Alaska bottom trawls

Tanner crab (Chionoecetes bairdi), AFSC

Yochum et al. 2015 evaluated the flexibility of RAMP methodology by creating a RAMP for Tanner crab (Chionoecetes bairdi) discarded from the groundfish bottom trawl fishery in the Gulf of Alaska and comparing it to a previously established RAMP for unobserved Tanner crab bycatch (encountered gear and remained on the seafloor) from the bottom trawl fishery in the Bering Sea. The authors found that: “The two RAMPs and the overall mortality rates calculated using these predictors were comparable. However, we detected significant differences between RAMPs. While probabilities of mortality were similar between the two studies for crab with all or no reflexes missing, discarded crab with intermediate reflex impairment had lower mortality probabilities than those from the unobserved-bycatch study. Our results indicate that a RAMP may produce more accurate mortality estimates when applied to animals experiencing similar stressors as those evaluated to create the RAMP, through similar methodology.”

Conditions for holding crab after exposure to stressors can control delayed mortality and should be considered in experiments. “There were differential mortality rates by holding type. Higher mortality rates occurred in the on-board tanks (where the crab were held for the first few days) and in the laboratory tank. Moreover, Score-zero crab died in the holding tanks, but not in the at-sea cages. These results indicate that holding tanks contribute additional stressors, either due to transport, additional handling, or stress from being held in an unnatural setting or at temperatures greater than what was experienced in their natural environment. 

Our holding duration of two weeks was sufficient to determine mortality for all Scores. Given that it can take longer for Score-zero animals to die than those with higher Scores, our holding period allowed us to sufficiently capture Score-zero mortalities. However, the death of a Score-zero crab at day 12 may indicate that holding for more than a week confuses mortality attributed to fishing stressors with that from captivity.”

Evaluation of RAMP flexibility was made by comparing results from different studies. “To evaluate the divergence between the RAMPs we analyzed the differences between the studies. The primary difference was in experimental methods, namely the treatment of the crab before assessment. Crab from the Discard-mortality study were exposed to air for 90min on average (range from 9 to 230min) without any “recovery” in water. In contrast, crab from the Unobserved-mortality study had only brief air exposure and were held in water while awaiting assessment (generally less than 15 min), which may have allowed some recovery. 

These differences in air exposure and recovery in water probably affected the relationship between observed reflex impairments and delayed mortality and hence accounted for the discrepancy between RAMPs. Prolonged air exposure and experiencing cold temperatures was linked with increased delayed and instant mortality, number of autonomies for crab, as well as reduced vigor, juvenile growth, and feeding rates (Carls and O’Clair, 1995; Giomi et al., 2008; Grant, 2003; Stoner, 2009; Warrenchuk and Shirley, 2002). Stoner (2009) found that reflex impairment score and exposure to freezing temperatures were nearly linearly related for Tanner crab. Moreover, he found that the different RAMP reflexes had variable sensitivity to freezing temperatures, namely that the chela closure reflex was the most sensitive reflex, and mouth closure was least. Similarly, Van Tamelen (2005) found that the legs and eyes of snow crab cooled faster than the body, perhaps making them more susceptible to cold air exposure. We hypothesize that the prolonged air exposure for the Discard-mortality study likely impaired the crabs’ reflexes and resulted in higher Scores.”

Recommendations made by the authors. “Results from this study indicate that bias can be introduced in mortality rate estimates when using a RAMP created for one study to estimate mortality rates for a different study where the experimental methods differ, especially with respect to air exposure and recovery in water before assessment. However, when RAMP is used only to approximate mortality rates or to make comparisons between gear types or uses, a previously established RAMP could be used with caution, especially if animals with intermediate Scores are not predominant. For more accurate bycatch mortality rate estimates, our results indicate the importance of using a RAMP that was created by assessing animals that experienced similar stressors to those which the RAMP will be applied. Namely, the procedure for assessing the animals should be similar. We feel that the amount of time the animal spends out of water before assessment be standardized within a time range, along with whether or not the animal is allowed to recover in water before assessment, unless these variables are the treatments being studied.” 

“Our results indicate that consistency in methodology and relevance with respect to mimicking actual fishing stresses for the RAMP approach increases the flexibility of RAMP. It is therefore important, when creating a RAMP, to create repeatable methods that are well documented when publishing. RAMP reflexes should be assessed in a specified order to prevent bias from reflexes that are physiologically linked. If there is a reflex that influences the determination of other reflexes it should be assessed last or not at all. Reflexes that are difficult to determine presence or absence should not be used, and it should be clear in the methods what constitutes an “absent” reflex and how immediate mortalities are treated (are they given a Score or classified separately?). In addition, when a RAMP is being created, data should be recorded on all possible stressors, including injury, and evaluated for their contribution to mortality. Moreover, effort should be made (within the logistical constraints of field and laboratory research) to minimize additional stressors that are unrelated to the fishing stressors of interest.”

Making and Using RAMP in Fisheries

A video is available that explains making and using RAMP in fisheries.

Why is vitality impairment related to mortality?

By definition, healthy animals have full vitality. Vitality becomes impaired as animals become stressed by capture and handling. Severe vitality impairment can result from the effects of physical injury or other stressors, e.g., fatigue, temperature, light, sea state, and air exposure. Maladaptive stress responses or critical injury associated with severe vitality impairment can result in immediate and delayed mortality.

Why score reflex actions and injury?

Reflex actions are fixed behavior patterns that are directly related to vitality impairment, without control by volitional behavior factors, e.g., motivation, hunger, fear, shelter seeking, migration, and reproduction. Reflex actions reflect the state of neural, muscle, and organ functions.

Injuries are directly related to vitality impairment because they can control neural, muscle, and organ functions.

Scoring vitality impairment in general

Any type of reflex action or injury that is related to vitality can be summed to score vitality impairment. The important point is that a sum of presence/ absence scores for vitality characteristics produces an index of vitality impairment. This vitality index can then be used as a measure of variability for sublethal stressor effects in fisheries, as well as a validated indicator and predictor of mortality and survival.

Steps for making and using RAMP in fisheries.

Methods for estimating discard survival in fisheries: an integrated approach

Discarding Pacific halibut, FAO

ICES has published a report on methods for estimating discard survival in fisheries. The report details the results of the February, 2014 ICES WKMEDS workshop on discard survival.

“This report will:

-  describe the concepts behind assessing discard survival (Sections 2 and 3);

-  describe three different approaches for estimating survival (vitality assessment, captive observation and tagging) (Sections 4, 5 and 6); and 

-  provide guidance on the selection of the most appropriate approaches and experimental designs, as well as how to integrate and utilize information from them, with respect to specific discard survival objectives (Sections 3, 7, 8 and 9). 

Later versions of this report will cover in more detail: 

-  techniques for assessing survival using tagging and biotelemetry; and 

-  the most appropriate methods for analyzing and reporting survival data. 

It is assumed that the user of these guidance notes has sufficient scientific training, or at least access to suitable scientific support, to be able to conduct the techniques described in these notes in an appropriately systematic and disciplined manner. However, these guidance notes are intended also to be informative for other stakeholders associated with fishing (primarily fishers and managers) who wish to support and understand discard survival estimates.”

The ICES WKMEDS report is a summary of an integrated approach for estimating discard survival. The approach uses various combinations of vitality assessment, captive observation, and tagging to achieve realistic estimates for discard survival in fisheries. The combinations of methods are determined by scientists, stakeholders, and managers using evaluation and prioritization:

“the choice of which species in which fisheries to study depends upon several criteria: existing survival information, the biological traits of the species, its population status, magnitude of discarding, fishery characteristics, environmental characteristics, socio-economic value of the fishery, available resources, and management policy. The process of prioritizing is unlikely to be simple and may involve a number of iterations, where results of preliminary studies inform the final choice.”

The ICES WKMEDS report represents a new approach for estimating discard survival. Sources of information about objectives, priorities, resource implications, and time frames are included in a decision matrix. Managers can use the matrix to make informed choices about discarding in key fisheries and management units and what methods can be used for further study of discard survival. Initial calibration of vitality assessment using delayed mortality observations of discards creates validated indicators for survival. Then use of validated vitality assessment indicators such as RAMP (Reflex Action Mortality Predictors) can provide rapid real-time assessment of potential discard mortality on-board fishing vessels.

ICES. 2014. Report of the Workshop on Methods for Estimating Discard Survival (WKMEDS), 17–21 February 2014, ICES HQ, Copenhagen, Denmark. ICES CM 2014/ACOM:51. 114 pp.

The importance of vitality in fishing experiments

Key fishing stressor factors, Davis, 2002

Knowledge of key factors controlling fisheries is necessary for sustainable management of fishery stocks. Scientific hypothesis testing in the form of fishing experiments is a necessary component of fisheries knowledge development and validation. Fishing experiments are performed in the field by simulating actual fishing conditions, by actual fishing, and during survey cruises. Fishing experiments can be used to identify key stressor factors that control and contribute to the survival and mortality of captured, discarded, or escaped animals as well as identifying the key factors controlling fishing gear capture efficiency and selectivity.

Trawl captured animals, Robert A. Pawlowski, NOAA Corps

While field fishing experiments represent realistic conditions, they are a matrix of confounded factors which cannot be easily separated into mechanistic hypothesis tests and explanations of factor importance. Effects of factors are often synergistic and prior animal stressor history can alter relative effects of subsequent exposure to factors, e.g., depth changes, injury, elevated temperature, air exposure, and size and species differences.

Flow chart of experimental fishing stressor factors, Davis and Olla 2001

Simulated fishing experiments with factors in controlled laboratory conditions is one way to test hypotheses about mechanistic effects of individual factors and their interactions. However these laboratory experiments are generally viewed as not realistic to field conditions and they are used to identify factors that may be important in the field. Furthermore, modern requirements of animal care laws and committees restrict the use of laboratory fishing experiments by not allowing human application of experimental stressor factors on animals and the use of mortality outcomes. These same laws and committees do not have jurisdiction over field fishing experiments. 

Laboratory trawl tow tank, NOAA RACE

Given that factors are confounded in field fishing experiments, how can we test for effects of factors in the traditional mechanistic hypothesis test? We can test for changes in animal vitality. Since vitality has been shown to be correlated with survival and mortality, it is a useful indicator of animal outcomes before and after exposure to experimental stressor factors. For example, we generally do not know the exposure of animals to stressors prior to experimental manipulation of factors. Not knowing the complete stressor profile is not an obstacle since the animal knows the complete stressor profile and presents vitality levels that have integrated the effects of that profile. Then we can expose animals to additional stressor factors and measure further changes in vitality from their initial levels. 

Important to shift mechanistic thinking from needing to know the effects of individual factors to knowing the effects of fishing variability. Manipulations of time in air and elevated temperature represent differences in fisher sorting and handling behavior on deck and are appropriate for defining the effects of fishing variability. Effects of variation in tow time and catch quantity can be manipulated and are included in the mix of animals landed. The questions of associations among individual fishing stressor factors is left for another day and are more of interest to mechanistic scientists than to managers and fishers. Fishing variability will give a picture of the fishery and its potential effects on animal vitality. By measuring animal vitality, which integrates the effects of stressor factors, you have measured a key master variable that indicates the important effects of fishing.

Vitality is the key variable that can be used to indicate and predict delayed survival and mortality outcomes for discards and escapees from fishing. The relationships between vitality and survival and mortality are defined by captive observation or tagging and biotelemetry experiments. During exposure of animals it is important to insure that all stressor types normally in the fishery in question are present for the population of tested animals (e.g., temperature, air exposure, fatigue, injury) and that a full range (0-100%) of vitality impairment and mortality are observed. Then relationships can be calculated for each species of interest that do not extrapolate beyond available data ranges and that apply to the fishery of interest. These relationships can then be used to predict survival and mortality for animals under any condition of interest in the fishery without the need for further captive observations or tagging.

Consider how scientific peer-reviewers may see this shift from mechanistic thinking and develop thoughts that elaborate the importance of vitality from the animal’s point of view. Some resistance is expected from mechanists who believe that they can attribute cause and effect to individual factors. There is always a matrix of interactions, even under the most restrictive and controlled experimental conditions. There are always interactions and synergisms to account for. As a result, there are associations among factors, rather than cause and effect. In other words, there are causes and conditions associated with effects. 

From the fishing experiment perspective, we set up fishing conditions that are real or that simulate fishing and then measure animal vitality, which is an integrated measure of the effects of interacting factors. It is useful to identify the important stressors by experimentally changing them in fishing experiments; changes in time in air, trawl time, trap retrieval time, depth, season, temperature, catch amount, and injuries. Always remember that there is a hidden context of conditions, i.e., the animals are prestressed by other factors not being controlled. But this hidden context can be accounted for by observing and comparing vitality impairment among animals observed in all treatments, including simply captured animals without additional stressor exposures (using positive controls). This experimental approach is useful both for fishers who wish to modify fishing gear and practices, as well as managers who wish to observe animal vitality and correlate that with mortality and survival.

Survival of the fittest

Bob van Marlen, IMARES

Work is in progress at ICES to define methods for estimating discard survival in reference to the new European Commission Common Fisheries Policy. The ICES newsletter states, “Under the recently reformed European Commission Common Fisheries Policy (CFP), the practice of discarding fish will be phased out, replaced instead with landing obligations. Under the landing obligation, all catches of regulated species must be landed and counted against quotas unless it has been scientifically proven that the species can survive the discarding process. Species that display a high level of discard survivability will be awarded an exemption, meaning that fishers can return these fish to the sea. Unregulated and protected species will continue to be released.”

The ICES WKMEDS focuses on developing guidelines and identifying best practice for undertaking experiments to investigate the survival of organisms discarded from the catches of commercial fisheries. 

Importantly ICES states, “Techniques for estimating survival under review include captive observation, vitality assessment, and tagging and biotelemetry, each with its own advantages and disadvantages. By using a combination of techniques, as WKMEDS suggests, clear synergies can be achieved and challenges overcome.”

"It's an exciting time", state the workshop Chairs Michael Breen and Tom Catchpole, "this group will be central to an international community that are working together to address the important issue of discard survival."

The WKMEDS group is producing a synthesis of previous discard survival and mortality research with the goal to develop an integrated approach for estimating discard survival. The integrated approach is designed to guide scientists and managers in their evaluation of potential candidate fisheries and species for discarding exceptions to the landing obligation of the CFP. The integrated approach incorporates information about the role of fishing conditions as stressors, species sensitivity to stressors, vitality impairment of captured species, estimated discard survival rates, and cost-benefit analysis of methods for survival assessment in fisheries as well as impacts of discarding on fisheries stocks.

On the importance of considering sublethal stress and injury effects in discard and escapee survival and fitness

Shark release, NOAA

Sublethal stress and injury are important factors to consider in discard and escapee survival and fitness (Wilson et al. 2014). Fishing gears are often designed to enhance escape of animals that might otherwise become bycatch. Captured animals are discarded from fisheries for economic, regulatory, conservation ethic, and political reasons. Observation of animal impairment (Davis 2010) and immediate mortality for discards on board fishing vessels is straightforward. For animals that are alive when discarded or that escape from fishing gear, the effects of stress and injury can alter behavior, growth, and reproduction, or result in delayed mortality. Delayed mortality of escapees (Suuronen 2005) and discards (Revill 2012) has been documented for commercially important species. However, few studies have measured potential fitness effects of sublethal stress and injury on surviving discards or escapees.

Fig. 1. Conceptual diagram outlining the immediate and long-term effects of escape or release from commercial fishing gear and how it relates to each level of biological organization. Question marks (?) denote areas for which no primary literature exists, and present future avenues of research (Wilson et al. 2014).

Sublethal effects of capture, escaping, and discarding can occur at individual, community, and population levels of organization. For individuals, immediate sublethal effects are physiological responses, injury, and reflex impairment. Delayed sublethal effects are behavioral impairment, altered energy allocation, wound healing, immune function and disease, reproductive success, and offspring quality and performance. Few studies have been conducted for responses at community and population levels, and clearly these are important to consider.

Wilson et al. 2014 summarize:

“The obvious gap that emerges is the lack of research linking at-release measurements with latent sublethal fitness outcomes such as foraging, energetics, growth, reproduction and offspring quality. The dearth of knowledge in this area is likely based on two realities: (1) a justifiable focus on simply quantifying and reducing bycatch mortality, and (2) the difficulty of long-term monitoring of fitness outcomes in wild animals. Of the reviewed studies, several indicated that physiological disturbance, injury or behavioural impairments may have had long-term implications for growth and reproductive fitness. Further study of sublethal effects could clarify previously unaccounted-for population level consequences of fisheries and better conservation practices to mitigate the impacts of fisheries.”

Also of importance to discard survival and fitness is consideration of predation that can occur after escape or discarding of captured animals (Raby et al. in press). Controlling factors for predator-induced mortality include fishery type, stress and injury, barotrauma, predator behavior and abundance, fish size, and temperature. Summary of the Raby et al. review suggests research directions:
  “The important first step is for fishers, managers and researchers to identify systems where predation is likely to be a substantial contributor to unobserved fishing mortality. Most study of capture-and-release mortality involves quantifying the effects of factors such as temperature, capture depth or fight time. Predator type and abundance could be considered new ‘phantom’ factors that are dynamic and would be challenging to incorporate into research. A conservative approach would be to assume a constant level of predation threat for a given fishery and focus on examining the capacity of released fish to evade predators and the accompanying rates of predator-induced mortality. PRP is a unique contributor to mortality because it is probably most often characterized by a short period (minutes or hours) of risk, which could simply be overcome by using pre-release techniques that reduce the impairment of fish being released (Farrell et al. 2001; Broadhurst et al. 2009).”

Future studies that can be used to assess the presence of delayed sublethal fitness effects in fisheries escapees and discards include allostasis, biotelemetry, reproductive success of individuals, measurement of genetic material contributions to next generations, and tank or net pen holding studies to determine behavior, growth rates, and reproduction. In all these types of studies, ongoing collection of fisheries observer data on reflex impairment and injury using a vitality scoring system (RAMP) would be needed to link vitality scores of at-release discards or escapees back to fitness outcomes for individuals.

Philosophy of using RAMP to measure vitality, survival, and mortality of animals

Blue sharks and food, WASC

Mortality can occur over varying time frames after an animal is exposed to potentially lethal stressors. The problem of mortality prediction is made more difficult by animal mobility, as animals can become hidden from observation, especially over longer time frames. Then indicator measures must be used to predict cryptic delayed mortality. What is an effective indicator for predicting mortality? Do we observe the animal immediately after stress induction and before leaving our presence, or do we observe the conditions in which the animal was stressed? 

A common approach to predicting delayed animal mortality is to observe the conditions in which stress is induced and use this information as an indicator for mortality.  Animals are experimentally exposed to important stressors and their combinations in a matrix of interactions. Then animals are sampled for mortality after holding them captive for short periods or tagging, releasing, and recapturing or using biotelemetry over longer time frames. Mortality, and its inverse, survival are then modeled from sampled combinations of risk factors. Since there are relatively unlimited sets of risk factors and their interactions, indicator models for mortality based on stressors often will not give realistic estimates or not include important conditions for stress induction.

Alternatively, animal impairment can be observed as an indicator for delayed mortality after exposure to risk factors. Reflex impairment occurs immediately in an animal when it’s neural, muscular, or organ systems are stressed. Summing presence or absence of several reflex actions calculates an index called RAMP (reflex action mortality predictor) which is a direct measure of reflex impairment and vitality. Correlation of RAMP with immediate and delayed mortality make it an indicator for mortality and survival. With RAMP, the approach of predicting mortality is based on direct observations of animal vitality. The animal continually integrates all the effects of experienced risk factors as reflex impairment and communicates it’s health state, vitality, and fitness through the language of RAMP. Other types of animal impairment that have been tested as potential indicators for mortality include physiological variables (cortisol, glucose, lactate, and electrolytes) and injury.  However these measures are not consistently correlated with delayed mortality.

In an effort to ameliorate mortality risk factors, a hybrid approach can be used for predicting cryptic delayed mortality that conserves and integrates information. Instead of asking the question “Does the animal die?” we can ask “When, where, and under what conditions does the animal die?” Animals are observed in experimentally controlled conditions of mortality risk (Davis 2002, Suuronen 2005). Then initial stressor conditions are sampled, as well as time courses for animal impairment and delayed mortality. Relationships between stressor factors, animal impairment, and delayed mortality can be identified and modeled. The resulting knowledge base can be used to test hypotheses about importance of mortality risk factors and efficacy of predicting cryptic delayed mortality using animal impairment as an indicator. Previous research has shown that reflex impairment measured as RAMP is a powerful predictor for cryptic delayed mortality (Davis 2010). After validation, RAMP can be used to test the effects of experimental or natural changes in mortality risk factors such as design of fishing gears, aquaculture rearing conditions, aquarium trade, pollution exposure, climate change, and other potentially risky situations.

Trawl bycatch reduction device, FRDC

The problem of using indicators to predict cryptic delayed mortality is simplified by shifting from modeling mortality in potentially unlimited sets of risk factors to direct, real time measurement of animal impairment and prediction of delayed mortality. This shift in focus to reflex impairment allows for real time testing of animal fitness in systems of interest and is a cheaper, more efficient use of limited research resources than using risk factor indicators for mortality prediction.

High discard survival merits exemptions to European Union ban on fishery discards

Atlantic cod, NOAA

The introduction of the obligation to land all catches (eliminate discards) in the recent reform of the Common Fisheries Policy (CFP) represents a fundamental shift in the management approach to European Union fisheries from regulation of landings to regulation of catch. Research has shown that not all discards die. In some cases, the proportion of discarded fish that survive can be substantial, depending on the species, fishery and other technical, biological and environmental factors. If these surviving animals are discarded instead of landed, they can contribute to future stock recruitment.

Article 15 paragraph 4(b) of the CFP regulation allows for the possibility of exemptions from the landing obligation for species for which "scientific evidence demonstrates high survival rates". Taking the first element of this "scientific evidence"- it is important that managers have guidance on protocols and methodologies that should be followed in order to ensure the results of such experiments are scientifically robust. Presently there are no such internationally agreed guidelines. EWG 13-16 has provided guidance on best practice to undertake survival studies. In this regard EWG 13-16 has identified three methodologies for conducting survival experiments i.e. captive observation experiments, vitality/reflex assessments, and tagging/biotelemetry experiments.

Captive observation experiments involve holding animals that have been captured after exposure to fishery stressors. Holding can be in tanks or net pens while short-term survival is observed.  Holding periods typically range from 3-21 days until mortality associated with experimental fishery stressors has abated. Discard survival rates in specific fisheries conditions are then modeled using data from holding experiments. Davis (2002) reviewed an array of potential explanatory variables for discard survival, which can be classified into three broad categories: biological (e.g. species, size, age, physical condition, occurrence of injuries), environmental (e.g. changes in: temperature, depth, light conditions) and operational (e.g. fishing method, catch size & composition, handling practices on deck, time exposed to air). The complexity and interactions of explanatory variables for discard survival could present a problem to fisheries managers because instead of simply asking “Can we discard this species?” it may be necessary to ask “when, where, and under what conditions can we discard this species?” A potentially unlimited variety of fishery condition combinations would need to be modeled for determining discard survival.

Effects of fish size, fishing gear type, and temperature on sablefish mortality, NOAA

Tagging/biotelemetry experiments are similar to captive observation experiments in that animals are captured after exposure to fishery stressors. Then animals are tagged, released, and monitored for survival either by recapture or by biotelemetry.  Survival observations can be made over periods of weeks, months, and years. In addition to the complications of fishery stressor variable interactions, these experiments have the additional complications of including sources of mortality associated with predation and food and habitat availability that are independent of the effects of the initial capture stressors.    

Vitality/reflex assessments are real time in situ determinations of animal vitality and health.  The animal integrates the effects of fishery stressors and lives or dies according to it’s level of vitality impairment. Davis (2010) has shown that calculation of an index for reflex impairment (RAMP, reflex action mortality predictor) based on summing observations of several reflex actions is a robust, quantitative measure of animal vitality that can include the effects of fishery stressors on animal survival. When animals are exposed to fishery stressors and captured, as described above for captive observation experiments and tagging/biotelemetry experiments, they exhibit various degrees of stress and impairment of vitality which can be associated with mortality and survival.  Correlation of RAMP scores with mortality or survival levels observed in captive observation experiments or tagging/biotelemetry experiments makes RAMP a proxy for mortality or survival (Raby et al. 2012). Further RAMP validation can be made by testing with additional holding or tagging experiments in fisheries of interest.

RAMP curves for Atlantic cod, Humborstad et al. 2009

Once validated, RAMP assessments could be used to identify species in a fishery that may have the potential to survive discarding, and that merit an exemption to the Landings Obligation. Where a large majority of individuals of a particular species demonstrated consistently high RAMP scores, and there were very few examples of immediate mortality, this would indicate that species may warrant further investigation to demonstrate its potential for short & long term survival, post-discarding. Using this approach, a large number of species could be assessed (quickly & inexpensively), over a wide range of conditions and for a variety of boats (& discarding practices) throughout the fishery. 

At the same time, continued development of innovative fishing gears and fisher avoidance of high bycatch areas and times can help reduce capture of unwanted species. RAMP can be used to evaluate the survival of animals that are impacted by fishing gears and escape before landing on fishing vessels. The banning of discarding will make the evaluation of mortality rates for animals escaping from fishing gears especially important. "Out of sight and out of mind" will not be a viable strategy with regards to evaluating fishing mortality for gears engineered to enhance escape of bycatch species.

Survival of bycatch species escaping purse seines

Many species of schooling fish are caught using purse seines.

Purse seine

Fish in schools caught by purse seines are generally in aggregations of one species.  However there can also be fish that are too small, too many, or other species such as turtles, porpoise, and tuna that are not designated for capture in the fishery. For management purposes, these non-target animals are considered bycatch and are discarded, either from the fishing gear or from the catching vessel.  

Porpoise and tuna bycatch in purse seine

What are the mortality rates for bycatch species that are discarded or escape from purse seines? Experiments have been designed to answer this question for mackerel that are "slipped" from a purse seine because too many are caught in a net set.

Huse and Vold 2010

Mortality rates of mackerel were related to a stress index of crowding density and time.

Huse and Vold 2010  Stress indices (fish density (kg m⁻¹) times crowding time) from Lockwood et al. (1983) (diamonds) and from our own experiments (triangles). The exponential line is fitted to the data from Lockwood et al.  

Reflex impairment and RAMP can be used in field survival and mortality experiments. Observations of reflex impairment for fishes, turtles, and porpoise can be made in holding nets and related to vitality and mortality using RAMP calculations. Note that the mortality curve for mackerel is similar to RAMP curves for other species.  

Use of RAMP in field experiments and fishing operations can result in large amounts of real time, high quality data on discarded or escaped bycatch vitality and mortality.  This data is key for the effective management of fisheries stocks and conservation of ocean ecosystems.

Survival of bycatch species escaping trawls

New fishing trawl gears are designed to facilitate the escape of animal species that would otherwise be landed as bycatch and discarded. Knowledge of animal behavior such as startle, rheotaxis, phototaxis, avoidance, and sheltering is used to facilitate escape of potential bycatch species from trawls.

He, editor

Survival of animals that escape from trawls is an important aspect of conservation. Any mortality of escapees must be accounted for as bycatch mortality in fisheries management.

Rahikainen et al. 2004

Measurement of survival and mortality rates for escapees is a difficult research problem. One approach is to design trawls and nets that can sample escapees for later observation of behavior, recovery from capture, or mortality.

 Suuronen 2005

Reflex actions and volitional behavior can be observed in cages after animals have escaped from trawls and through their recovery period. RAMP can validated during these experiments and used to predict mortality rates for animals observed to escape from fishing gears.

Project Survival

Banning fishery discards and using RAMP

European Union fishery ministers have agreed to phase out the practice of discarding unwanted or regulated animals (bycatch) from landed catches.  The practice of discarding bycatch can be tremendously wasteful of fishery resources including fish, elasmobranchs, invertebrates, birds, amphibians, reptiles, and mammals.

Discarding, Richard

Banning discarding from fisheries requires total retention of animals caught, which must be landed and processed.  As many of these discarded species are of low economic value, efforts are made to design fishing gears that avoid catching bycatch species in the first place.

Suuronen 2005

A key assumption in the ethical design of fishing gears that do not catch bycatch and discarded species is that animals survive gear encounters. Escaping animals must have significant survival rates after gear encounters if they are to continue contributing to recruitment and ecosystem function. If animals escape from fishing gears and do not survive, they are the same problem as discards in fisheries, except that they are hidden.

Suuronen 2005

Measurement of mortality rates for discards and for animals that escape from fishing gears is vital to the management of fisheries, as they represent a significant form of fishing mortality. Discard and escapee mortality rates have been difficult to measure and new, effective methods are needed.

Viability estimates for Pacific halibut bycatch, based on vitality codes (1-4) for injury and activity have been incorporated into fisheries management for several years.  Recent research results by Benoît et al. 2012 on discard mortality have suggested methods based on fishery-scale sampling with semi-quantitative vitality codes (excellent-1, good-2, poor-3, and moribund-4) and conditional reasoning.

Benoît et al. 2012. Post-capture survival probability over time (h) for five southern Gulf of St. Lawrence marine fish taxa (panels), as a function of their pre-holding vitality class score (colours). The shaded areas represent the 95% confidence band for the Kaplan–Meier empirical survival curve for each vitality class, plotted up to the time at which the last observation was made for a given taxon and vitality level. The lines represent the fits of the selected model for each species and vitality class. For cod and plaice, the fits for models M3 and M4 are presented using solid lines and dashed lines respectively (note that these lines largely overlap). The location of the circles along the line and the size of the circles indicate respectively the times at which observations were censored and the proportion of censored observations for the taxon and vitality level at that time.

Reflex impairment measured by RAMP is a quantitative measure of vitality that gives increased resolution and accuracy to the determination of health and survival of discards and animals encountering and escaping fishing gears. Future research on this subject can benefit from the incorporation of fishery-scale sampling of RAMP for discards and for animals escaping from fishing gears.

The banning of discarding will make the evaluation of mortality rates for animals escaping from fishing gears especially important. "Out of sight and out of mind" will not be a viable strategy with regards to evaluating fishing mortality for gears engineered to enhance escape of bycatch species.