Physiological and Behavioral Effects of Angling Stress on Kelp Bass, an Important Game Fish in Southern California

Abstract

Game fish populations in North America face increasing pressure from recreational anglers, yet sublethal effects from mandatory catch-and-release regulations remain unknown for many targeted species. In southern California, Kelp Bass Paralabrax clathratus populations have significantly declined in recent decades, prompting changes in management and increasing release rates. To assess acute effects of current fishing regulations, we collaborated with recreational anglers to evaluate short-term, sublethal impacts of capture stress on Kelp Bass physiology and behavior. To evaluate the timeline and magnitude of physiological stress, blood samples were collected at time points (10–120 min) after angling and confinement and compared to control fish sampled in less than 3 min. Postrelease recovery was determined by comparing biomarker levels between control fish and recaptured fish that were rapidly sampled after a time at liberty ranging from 3 h to 186 d. Biomarker levels in fish recaptured multiple times were compared to previous samples to evaluate repetitive angling effects on physiological responses. Circulating cortisol, glucose, and lactate were elevated and steadily increased in the hour after capture, although angling duration and handling duration were not correlated with biomarker response. Fish size significantly affected fish stress, with larger fish experiencing less stress and rapid recovery within 24 h. Behavioral control fish, which ingested acoustic accelerometers hidden inside bait, exhibited strong diel activity that was reduced for 30 h in angled fish. Tracked individuals exhibited high individual variation in rate of movement and area use, with no noticeable postrelease impacts. Although this study found Kelp Bass to be resilient to angling stresses, evaluation of long-term effects from catch and release is warranted. This fishery may also benefit from slot limit regulations, improved angler engagement programs, and education on best practices that minimize fish stress during catch and release.

The impacts of recreational fishing activities on globally declining fish populations are now broadly acknowledged, as is the need for effective management strategies to restore and ensure healthy fish stocks for the future. Fisheries managers and scientists increasingly recognize the widespread effects of rapidly growing recreational fisheries on game fish populations as well as the new—and complex—challenges this has generated for fisheries management. Much like the impacts of commercial fishing, recreational activity has often resulted in fishing-induced trophic changes, biomass reductions, and truncations in age and size at maturity for highly sought-after species (Coleman et al. 2004; Cooke and Cowx 2006; Lewin et al. 2006, 2019; Arlinghaus et al. 2007; Hamilton et al. 2007; Radford et al. 2018; Huddart 2019). Most fishing regulations are implemented to minimize fish mortality while maintaining the important social and economic value of the fishery (Arlinghaus et al. 2007; Hughes 2015; Lyach and Čech 2018). Various harvest control strategies (e.g., size and bag limits) have been instrumental in sustainably managing many fish stocks, but they often result in more fish being caught and released, with the assumption that released individuals survive and reproduce (Cooke and Schramm 2007; Hansen et al. 2015; Sass and Shaw 2020). Roughly 60% of the global recreational catch is estimated to be caught and released annually (Cooke and Cowx 2004), with release rates varying by species, region, and angler (Policansky 2002; Stensland et al. 2013; Stensland and Aas 2014; Gupta et al. 2015; Danylchuk et al. 2018; Grilli et al. 2020). Voluntary release is increasing among some anglers, offering proof that changing human behavior through regulation, education, or citizen science can be a powerful tool for promoting fish release, recovery, and, ultimately, population growth.

Similar to many game fish species, rapid growth of the recreational fishery for Kelp Bass Paralabrax clathratus in southern California has outpaced science-based management, resulting in regulations that have increased release rates for angled fish (Radomski et al. 2001; Cooke and Schramm 2007; Camp et al. 2015; Adams 2017). Since the early 19th century, the Kelp Bass has been a highly sought-after nearshore game fish, with overfishing concerns causing the closure of commercial fisheries for Paralabrax spp. (including Barred Sand Bass P. nebulifer and Spotted Sand Bass P. maculatofasciatus) in 1953 (Dotson and Charter 2003). However, rapid growth of the recreational industry soon yielded annual catch rates consistently exceeding historical commercial landings by the end of the century (CDFW 2013, 2019a). In recent decades, Paralabrax spp. have shown clear evidence of overexploitation, including severe declines in abundance, truncations in size and age structures, and the disappearance of annual spawning aggregations (Erisman et al. 2011; Jarvis et al. 2014; Bellquist et al. 2017; Won 2018). New size and bag limits established in 2013 have stabilized population declines, yet stock recovery may be inhibited by the unknown consequences of increasing mandatory release rates (Bellquist et al. 2017; CDFW 2019a). Although the spatial ecology (Lowe et al. 2003), life history (Love and Ebeling 1986; Love et al. 1996; Erisman and Allen 2005; Johnson 2005), stock dynamics (Steele et al. 2002; Erisman et al. 2011; Jarvis et al. 2014), and socioeconomic value (Siddall 2013; Bellquist et al. 2017) of this fishery have been investigated, there is little knowledge of the physiological and behavioral responses of Kelp Bass to catch-and-release practices. Understanding the impacts of angling on individuals and populations would help fisheries managers to implement effective management strategies and support future sustainable fisheries harvest.

There has been an increasing emphasis among fisheries scientists on evaluating the sublethal effects of catch and release on fish physiology and behavior as well as on identifying angling and handling techniques (i.e., “best practices”) that reduce animal stress and improve survival and recovery after release (Cooke and Suski 2005; Delle Palme et al. 2016; Brownscombe et al. 2017a, 2017b; Sass and Shaw 2020; Watson et al. 2020). Fish have evolved complex physiological responses to maintain homeostasis when disrupted by natural stressors, and these responses are primarily mediated by the hypothalamus–pituitary–interrenal (HPI) axis, which releases the stress hormone cortisol into circulation at elevated levels (Schreck and Tort 2016; Sadoul and Geffroy 2019). However, during extreme stress these mechanisms can be maladaptive and can affect metabolism, hydromineral balance, respiration, immune function, and other cellular processes, thus generating negative impacts on animal growth, fitness, behavior, and survival (Barton and Iwama 1991; Wendelaar Bonga 1997; Sadoul and Vijayan 2016; Schreck and Tort 2016; Yada and Tort 2016; Mateus et al. 2017; Valenzuela et al. 2018; Jerez-Cepa et al. 2019; Sadoul and Geffroy 2019). Sublethal effects can be minimized when anglers follow best practice guidelines that reduce fish stress during catch and release; therefore, angler education can be an effective strategy for fishery sustainability that complements more traditional harvest control regulations (Cooke and Schramm 2007; Brownscombe et al. 2017a, 2017b; Holder et al. 2020).

Technological advances in telemetry and biologging techniques have made it increasingly feasible to monitor wild fish and evaluate behavioral responses to angling that are sensitive indicators of physiological stress (Beitinger 1990; Archard et al. 2012; Brownscombe et al. 2017a, 2017b; Whitney et al. 2018; Sharma 2019). Postrelease behaviors observed in wild-caught fish have included delayed feeding (Meka and Margraf 2007; Sneddon et al. 2016) and migration (Lennox et al. 2016a, 2016b; Bass et al. 2018), inhibited reproduction (Schreck et al. 2001; Schreck 2010; Papatheodoulou et al. 2022), altered habitat use (Hoolihan et al. 2011; Brownscombe et al. 2014; Raby et al. 2018), and reduced swimming performance and/or predator avoidance (Brownscombe et al. 2013; Struthers et al. 2018; McLean et al. 2020). Addressing physiological and behavioral metrics together can provide valuable insight into the relationship between biochemical and organismal stress responses as well as the potential impacts on population-level processes (Fefferman and Romero 2013; Cooke et al. 2016; Louison et al. 2017; Balasch and Tort 2019). Biologging and telemetry technologies have provided researchers with the tools necessary to monitor the fine-scale activity of wild animals in their natural environments and to avoid the influences of chronic stress from confinement impacting normal behavior (Lowe and Kelley 2004; Donaldson et al. 2008, 2013; Cooke et al. 2012, 2016; Brooks et al. 2019; McGarigal et al. 2020).

The goal of this study was to evaluate the short-term, sublethal effects of recreational catch-and-release angling on Kelp Bass. By collaborating with anglers to catch fish, we established physiological and behavioral baselines, responses, and recovery from capture stresses using fine-scale acoustic telemetry tracking and stress-related blood biomarker concentrations. We hypothesized that angling and handling of this species result in significant endocrine disturbance (i.e., elevation in circulating cortisol, glucose, and lactate) and that the magnitude of this response is proportional to the duration for which the fish are fought on the line and exposed to air. We also hypothesized that physiological stress would lead to behavioral responses, including reduced area use, movement, and postrelease feeding activity. However, given the hardy nature of Kelp Bass, we expected to find low mortality, rapid recovery, and a general resilience to catch-and-release fishing practices.

METHODS

This study was conducted in Big Fisherman's Cove (BFC), a relatively protected area (~0.13 km²) within the no-take Catalina Island Marine Life Reserve (CIMLR) on Santa Catalina Island in southern California (33°26′N, 118°29′W; Figure 1). Previous study by Lowe et al. (2003) found that Kelp Bass from the CIMLR rarely left the boundaries of the reserve, which suggests that fish in this study were not stressed from angling prior to or between sampling events. Habitat in BFC encompasses a range of depths (0–40 m), seafloor substrata, vertical relief, and artificial structures (e.g., pier, floating dock, submerged mooring blocks; Figure 1). Fishing effort was consistent throughout the study (May 2015–December 2017) and was concentrated around the floating dock by 10 volunteer anglers, which ensured that angler behavior (i.e., how fish were fought and handled) was representative of the larger recreational fishing community. However, to minimize fish mortality, anglers used circle hooks and single barbless J-hooks with live and artificial baits (Meka 2004; Cooke and Suski 2005; Serafy et al. 2012; Jha et al. 2020).

FIGURE 1. Satellite imagery of Big Fisherman's Cove in the Catalina Island Marine Life Reserve (CIMLR), with inset map showing the location of the CIMLR on Santa Catalina Island off the coast of southern California. Yellow points represent Vemco radio-linked acoustic positioning array (VRAP) buoy locations and the control base station in the University of Southern California's Wrigley Marine Research Laboratory building. The blue line represents the approximate edges of the VRAP detection area, and the white line designates the area where position error is less than 1 m.

Physiological stress and recovery

Kelp Bass were landed rapidly using standard hook-and-line practices. To evaluate baseline levels of cortisol, glucose, and lactate, blood samples (0.2–0.8 mL) were collected in less than 3 min (i.e., rapid sampling) from the time of hooking (Barton 2002; Schreck and Tort 2016; Lawrence et al. 2018; McGarigal et al. 2020). To evaluate the magnitude and rate of the physiological response, angled fish were randomly assigned to a “stress treatment” group and subjected to a confinement period of 10, 15, 20, 30, 60, 90, or 120 min in an onboard holding tank (50 L) before blood sampling. The size of the holding tank was determined based on California Department of Fish and Wildlife (CDFW)-recommended live-well guidelines for tournament anglers and the estimated maximum size of fish that we expected to catch (Gilliland and Schramm 2009). Although confinement can be an additional stress on fish, it was necessary to observe the temporal aspects of physiological response (Lowe and Kelley 2004; McGarigal et al. 2020). Because the confinement period was standardized within each stress treatment group, we could evaluate how capture conditions influenced fish responses to identify best practices.

Blood was collected via caudal vein puncture using heparinized tuberculin syringes. Whole-blood glucose and lactate were measured in the field using portable meters (OneTouch Ultra Blood Glucose Monitoring System, LifeScan; Lactate Plus, Nova Biomedical) that were calibrated with appropriate standards as per manufacturers' instructions (Leclercq et al. 2014). Prior to use in the field, meters were validated for fish plasma using standard calorimetric assay methods (lactate calorimetric assay, Sigma Aldrich; glucose colorimetric assay, Cayman Chemical). After testing, blood was centrifuged in the field (3 min at 10,000 rpm) and plasma was separated and frozen in liquid nitrogen to maintain protein integrity during transfer to long-term storage in a − 80°C freezer. Cortisol was measured using thawed samples and commercially available enzyme-linked immunosorbent assay kits (Cayman Chemical), which have been previously validated for use on fish plasma (Sink et al. 2008; Mills et al. 2010; Metcalfe et al. 2018; Carbajal et al. 2019).

Detailed notes on the capture conditions for individual fish were recorded to identify the influence of angling and handling conditions on fish stress. Fight duration and air exposure have been correlated with physiological stress in a variety of teleosts (Cooke et al. 2002; Meka and McCormick 2005; Lennox et al. 2017; McGarigal et al. 2020); therefore, we recorded fight time and handling time for every individual. These two metrics were summed (i.e., landing time) to evaluate the cumulative effects of fighting and air exposure. Differences in handling and air exposure during the blood sampling process were evaluated by recording sample time (i.e., from the time of hooking to the time at which blood was drawn). Physical injuries (i.e., wound location, status [new or healed], and severity), visible parasites, and traumas experienced during handling (e.g., unintentionally dropped on deck) were recorded to evaluate the impacts of these additional stressors on fish biochemistry. All fish were weighed (g) and measured for SL (cm) to evaluate potential size-dependent responses. To visually identify recaptured individuals, all fish were externally tagged through the dorsal musculature with a uniquely numbered Floy tag (Floy Tag and Manufacturing, Seattle, Washington) prior to release. Because Paralabrax spp. do not exhibit sexual dimorphism outside of the breeding season (Erisman and Allen 2005), we were unable to determine the sex of angled individuals.

By recapturing previously angled fish, we could evaluate their physiological recovery at different postrelease time points in their natural environment. Recaptured individuals were randomly assigned to either a rapid blood sampling (<3 min from hooking) treatment or a confinement period (i.e., 10, 15, 20, or 30 min) before sampling. Rapid sampling provided a snapshot of the fish's physiological state before stress from the second angling event could initiate a biochemical response. Sampling of recaptured Kelp Bass after a confinement period allowed us to evaluate the effects of repetitive angling events on the rate and magnitude of the stress response. Individuals that were recaptured multiple times were subject to alternating rapid sampling and confinement sampling for consecutive recaptures, which allowed us to evaluate both recovery from a previous angling event and the effects of repetitive angling on biochemistry for the same individual.

Physiology data analysis

Because differences in measured cortisol, glucose, and lactate levels were not normally distributed among treatment groups, biomarker elevation was analyzed using Kruskal–Wallis and Dunn's pairwise comparison tests. The importance of capture conditions on biomarker levels was assessed using general linear models (GLMs), and model fit was compared using Akaike's information criterion. Variation in fish stress due to size was addressed using the best-fit GLM to predict cortisol elevation for a continuous range of fish sizes (15–51 cm) in each confined treatment group. Postrelease recovery was evaluated by using paired t-tests to compare the biomarker concentrations from fish that were caught for the first time and rapidly sampled to biomarker levels in fish that were recaptured and rapidly sampled. For fish that were recaptured multiple times, the effects of repetitive angling on the cortisol response were evaluated by using repeated-measures ANOVA to compare cortisol levels after confinement between initial and recapture angling events.

Behavioral stress and recovery

To evaluate normal, “unstressed” fish behavior (control), acoustic accelerometers (V9A-2H, Vemco Ltd.; 42 × 9 mm, 69 kHz, 151 dB, ~75-d battery life) were hidden inside squid bait so that fish freely ingested the transmitter package, thus eliminating stresses associated with capture and handling (Lowe and Kelley 2004; McGarigal et al. 2020). Larger individuals (visually estimated by divers as exceeding 40 cm SL) were targeted for feeding because they were more aggressive when feeding and were likely to retain the transmitter longer than smaller individuals before regurgitating. Other studies have successfully used this method and found that transmitter retention varies by species and individual. In most studies, transmitter retention is days to weeks, but Engås et al. (1998) reported tracking an Atlantic Cod Gadus morhua for 68 d. Behavioral responses to capture stress were determined by angling similar-sized fish (>40 cm SL), lightly anesthetizing the animals in tricaine methanesulfonate (MS-222; Finquel, 0.15 g/L) for 1 min, and placing the transmitter—along with several small pieces of squid—into the stomach through a lubricated, small-diameter tube gently inserted down the esophagus. Fish were then weighed, externally tagged, and revived (~10 min or until fully responsive to chasing) in saltwater tanks before being released. Because none of the transmitters was immediately regurgitated and because Kelp Bass often consume relatively large prey, we are confident that the procedure did not cause excessive stress or esophageal damage to any of our subjects (Bridger and Booth 2003).

Radio-acoustic monitoring system

Fine-scale movement and activity for tagged Kelp Bass were quantified using a Vemco radio-linked acoustic positioning array (VRAP). The VRAP buoys were deployed in a triangular configuration to provide the most efficient coverage of BFC while remaining line of sight to an onshore base station located in the University of Southern California (USC) Wrigley Marine Research Laboratory (Figure 1). Range testing confirmed that most areas in the BFC were within VRAP detection limits (Figure 1), and position accuracy was generally less than 1 m within the array (O'Dor et al. 1998, 2001; Jadot et al. 2006). Each buoy was composed of an omnidirectional hydrophone and Vemco VR20 acoustic receiver for detecting ultrasonic transmitter signals, as well as a VHF radio modem and antennae for two-way communications with the base station (O'Dor et al. 1998, 2001; Jadot et al. 2006). Fish positions were rendered in real time using a trilateration algorithm. Using this system, multiple tagged fish could be tracked simultaneously without signal collision or interference.

Behavioral data analysis

Acceleration, rate of movement (ROM), and area use size were considered relevant metrics for evaluating behavioral responses to capture stresses. Kelp Bass were often found along the perimeter of BFC, so data were filtered to remove detections with position error greater than 5 m to prevent location estimates from occurring over land. Acceleration measures along three axes (10 Hz) were used to calculate a root mean square acceleration value (m/s²), which provided a single index of whole-body movement as proxy for activity. Rate of movement (m/s) was calculated using the straight-line distance and time between consecutive detections, which was filtered to remove detections with time lags greater than 10 min. Core range and home range (m²) were defined as the areas that encompassed 50% and 90% of an individual's detections, respectively. These areas were calculated using Brownian bridge utilization distributions to accommodate the variability in time lag between detections. Mean acceleration, ROM, and area use were calculated hourly during a track to identify changes in behavior and recovery after release; to normalize variation in detection frequency among individuals, a bootstrapping method was applied using a random subset (n = 30) of data from each hour bin. The rate of recovery was evaluated by comparing behavior in the initial 3 h postrelease to behavior 24 and 48 h later using repeated-measures ANOVA. The effects of treatment (control versus angled), time of day (hours), sea surface temperature (mean hourly values [°C] from three HOBO pendant loggers deployed in BFC), and hour postrelease on fish behavior were evaluated as fixed effects in a linear mixed-effects model and were compared using Akaike's information criterion. Individual fish identity was included as a random factor because Kelp Bass exhibited high intraspecific variation in behavior. Behavior was evaluated for a 36-h postrelease period using the selected linear mixed-effects model and a sliding 3-h window as a smoothing parameter to reduce signal noise from individual and temporal variation.

RESULTS Physiological Stress and Recovery

Between May 2015 and December 2017, a total of 392 Kelp Bass were caught and sampled in BFC, with a mean SL ± SE of 33.3 ± 0.5 cm (range = 19–50 cm), mean fight time of 26.3 ± 2.0 s (range = 3–650 s), and mean handling time of 47.0 ± 3.1 s (range = 5–87 s). Fight time was positively correlated with fish length (P < 0.001), but handling time was not.

Sample time for control fish averaged 2.28 ± 0.06 min (n = 58) so that measured cortisol (median = 1.00 ng/mL, interquartile range [IQR] = 0.22–4.39 ng/mL), glucose (median = 1.20 mmol/L, IQR = 1.10–1.52 mmol/L), and lactate (median = 0.80 mmol/L, IQR = 0.70–1.0 mmol/L) levels reflected the baseline physiological state (Figure 2). Narrow IQRs for all three biomarkers reflect that there was very little variance in biomarker concentrations for the control group. Cortisol was significantly elevated above baseline in all angled and confined (i.e., stressed) groups (P < 0.001; Figure 2A), with hormone levels increasing steadily up to 60 min in confinement and peaking at concentrations that were 96% above baseline levels (median = 61.13 ng/mL, IQR = 43.45–130.15 ng/mL; P < 0.01) before slightly declining in fish that were confined for 90 min (median = 53.23 ng/mL, IQR = 42.70–69.79 ng/mL; P = 0.35) and 120 min (median = 54.31 ng/mL, IQR = 36.71–60.37 ng/mL; P = 0.44). Similarly, cortisol variance increased across treatment groups as confinement duration increased from 10 to 60 min (median values skewed toward the lower limits of IQRs; Figure 2A), followed by decreasing variance (narrower IQRs) for treatment groups that were confined for more than 60 min. Similar to cortisol, glucose levels were significantly elevated after 15 min in confinement (3.06 ± 0.22 mmol/L, n = 24; P < 0.01; Figure 2B) and increased with confinement duration, peaking after 60 min at levels that were 83.5% above baseline (8.12 ± 0.56 mmol/L, n = 26) before declining in the 90- and 120-min treatment groups. Lactate was also significantly elevated above baseline levels in all confinement groups (Figure 2C), but unlike cortisol and glucose, lactate levels increased steadily with confinement duration and the highest concentrations were recorded in fish that were confined for 120 min (median = 11.9 mmol/L, IQR = 7.955–13.32 mmol/L, n = 16; P = 0.001). Like cortisol, variance in glucose and lactate levels generally increased across treatment groups as confinement duration increased (Figure 2B, C).

FIGURE 2. Box plot of (A) plasma cortisol, (B) whole-blood glucose, and (C) whole-blood lactate concentrations for Kelp Bass that were angled and held in confinement for 10–120 min before blood sampling. Control samples were collected in less than 3 min from the time of hooking. Letters (a–e) designate significantly different treatment groups as determined by Kruskal–Wallis and Dunn's pairwise comparison tests (α = 0.05). Horizontal bars represent mean values (sample sizes are shown in parentheses within panel A; n = 9–58).

Landing time and fish size were both significant variables explaining the biomarker variance observed within treatment groups. Within each treatment, glucose and lactate levels increased as landing time (fight time plus handling time) increased (Figure 3B, C), and GLM output confirmed that sample time (landing time plus holding time) was a significant factor in the glucose and lactate response (P < 0.001; Table 1). The GLM output also determined that sample time significantly affected cortisol levels (P < 0.01; Table 1); however, cortisol decreased within each treatment group as landing time increased (Figure 3A), unlike glucose and lactate. Fish size (cm SL) was also a significant factor in the cortisol response (P = 0.001; Table 1), with an inverse relationship in which larger fish exhibited lower hormone concentrations (Table 1; Figure 4A). The GLM-predicted cortisol concentrations for fish ranging from 15 to 51 cm SL indicated that fish below the legal size limit (<35.5 cm TL) may experience cortisol responses that are an order of magnitude higher than levels predicted for legal-sized (>35.5-cm TL) individuals (Figure 5). The predicted peak cortisol response (indicated by the dark-red points in Figure 5) were consistently low (<80 ng/mL) for legal-sized fish but increased exponentially as size declined. We did not run similar predictive analyses for glucose or lactate because fish size was not found to be a significant factor for either biomarker, which may be due to highly variable responses in 30–40-cm fish (Figure 4B, C).

FIGURE 3. Cumulative effect of angling time and handling time (i.e., time to land) on Kelp Bass plasma cortisol and whole-blood glucose and lactate concentrations. Biomarker levels were measured for fish that were angled and held in confinement for 10–120 min before blood sampling.

TABLE 1 General linear model output of cortisol, glucose, and lactate response variables for Kelp Bass after angling, handling, and confinement. Predictive parameters considered in model development were fight time (s), handling time (s), time to land (s), sample time (min), and fish length (cm SL). Significant capture parameters for each response variable are shown.

Variable	Parameter	Coefficient	SE	Deviance explained (%)	t	P	Error family (link)
Cortisol				18.8			Gamma (inverse)
	Sample time	−0.008	0.0001		−7.686	<0.001
	Fish length	0.001	0.0003		3.209	0.001
Glucose				53.5			Gamma (inverse)
	Sample time	−0.005	0.0004		−13.62	<0.001
Lactate				42.3			Gamma (inverse)
	Sample time	−0.005	0.0006		−9.317	<0.001

FIGURE 4. Effect of Kelp Bass size (cm SL) on plasma cortisol and whole-blood glucose and lactate concentrations. Biomarker levels were measured for fish in each treatment group that were angled and held in confinement for 10–120 min before blood sampling. Fish size was a significant explanatory variable for the cortisol response but was not significant for glucose or lactate.

FIGURE 5. Predicted cortisol concentrations (ng/mL) for various sizes (cm SL) of Kelp Bass after angling and confinement in a live well for 10–60 min. The dashed line represents the current minimum size limit for Kelp Bass (35.5 cm [14 in]).

Regardless of the time at liberty (3 h to 186 d), all recaptured and rapidly sampled Kelp Bass (n = 54) exhibited biomarker levels that were comparable to baseline concentrations (P > 0.05; Figure 6). Fish that were recaptured multiple times exhibited baseline levels when rapidly sampled (n = 8, P > 0.05), whereas they demonstrated significant elevations when angled and confined for 10–30 min (P < 0.05; Figure 7).

FIGURE 6. (A) Plasma cortisol, (B) whole-blood glucose, and (C) whole-blood lactate levels in recaptured Kelp Bass after an initial angling and handling event. Time represents hours and days at liberty since initial catch and release. Control fish (“C” on the x-axis) were sampled in less than 3 min from the time of hooking. Error bars represent fish initially caught and held in a live well for 10, 15, and 20 min before blood collection. Asterisks indicate a significant difference from the control group as determined by Kruskal–Wallis and Dunn's pairwise comparison tests (P [less than] 0.001). Control and stress treatment values are presented as mean ± SE (n = 30–46).

FIGURE 7. Plasma cortisol levels for eight Kelp Bass that were recaptured multiple times after varying days at liberty (1–46 d). Dark bars indicate blood collected in less than 3 min from the time of hooking; light bars indicate blood collected after 10–30 min of confinement in a live well. Control samples (“C” on the x-axis) were from fish initially caught and sampled in less than 3 min from the time of hooking, and stressed samples (“S” on the x-axis) were from fish initially caught and sampled after 20 min of confinement in a live well. Asterisks indicate a significant difference as determined by repeated-measures ANOVA (P [less than] 0.001).

Behavioral Stress and Recovery

Between June 2016 and November 2017, the movement and activity of control fish (n = 9) and angled fish (n = 8) were tracked in BFC (range = 1–8 d). Control fish showed strong diel periodicity in behavior, including frequent bouts of rapid acceleration (i.e., activity, m/s²) at night, particularly between 2300 and 0100 hours (Figure 8A). Individual variation was also greater at night compared to daytime, when control fish were less active. In contrast, control fish exhibited a greater ROM (m/s) and greater area use (core and home range, m²) during the day than at night (Figure 8B, C). Individual variation was also higher during the day, but both behavior metrics generally increased from morning to evening. Compared to control fish, angled fish exhibited reduced activity levels during the initial 30 h postrelease, including 50% fewer bouts of high acceleration per hour (Figure 8D). Angled fish activity steadily increased between 30 and 36 h postrelease, surpassing control fish activity levels (Figure 8D). Similar to control fish, angled fish also exhibited high individual and temporal variation in postrelease ROM and area use, with no clear behavioral responses to catch-and-release fishing (P > 0.05; Figure 8E, F).

FIGURE 8. Effects of angling and handling on Kelp Bass diel activity (m/s2), rate of movement (ROM; m/s), and area use size (m2). The left column represents unstressed control fish (A) acceleration (i.e., activity), (B) ROM, and (C) area use size averaged over each hour (mean ± SE) for a 24-h period. In panel C, the core range size (red line) and home range size (black line) were determined by Brownian bridge 50% and 90% utilization estimates, respectively. Gray shaded areas represent nighttime hours; unshaded areas represent daytime hours. The right column represents the effect of angling (black line and points) relative to control behavior (gray horizontal line) on (D) acceleration, (E) ROM, and (F) area use size, with effect size calculated using a sliding 3-h window smoothing parameter for the initial 36 h postrelease. Gray polygons represent 95% confidence intervals.

DISCUSSION

Significant elevations in circulating cortisol, glucose, and lactate concentrations indicate that Kelp Bass experience physiological stress during catch-and-release fishing. Control fish (i.e., those rapidly sampled in <3 min) exhibited biomarker levels that were similar to or less than the baseline levels reported for other teleosts (Figure 2), which confirms rapid sampling as a viable method for evaluating the nonstressed physiological condition in wild-caught individuals (Barton 2002; Gerber et al. 2017; Lawrence et al. 2018; Pringle et al. 2020). Kelp Bass also exhibited a gradual increase in cortisol levels over time, peaking after 1 h (Figure 2A), which reflects the slow, negative-feedback response of the HPI axis to stress stimuli that results in elevated cortisol levels circulating through the bloodstream (Donaldson 1981; Wendelaar Bonga 1997; Kalamarz-Kubiak 2018; Kiilerich et al. 2018). Delayed elevation in glucose levels after angling is likely due to the mediating effects of elevated corticosteroid concentrations, like cortisol, on gluconeogenesis during stress (Figure 2B; Wendelaar Bonga 1997; Begg and Pankhurst 2004; Leclercq et al. 2014; Lawrence et al. 2018; McGarigal et al. 2020; Pringle et al. 2020). Unaffected by the HPI axis, lactate concentrations in Kelp Bass were elevated within 10 min of angling and continued to increase steadily throughout the 120-min observation period (Figure 2C); such gradually increasing lactate levels in the bloodstream were likely due to the continuous release (i.e., “washout”) of lactate from the white muscle tissue, which was built up during exertion (i.e., angling; Boutilier et al. 1993; Weber et al. 2016; Martins et al. 2018; Pringle et al. 2020). High lactate levels have been linked to greater postcapture mortality rates in some species of fish and sharks (Hoffmayer and Parsons 2001; Davis 2002; Skomal 2007; Mohan et al. 2020), but this seems unlikely to affect Kelp Bass, which have exhibited very high postrelease survival in previous studies (Lowe et al. 2003). Overall, the physiological response timeline is similar between Kelp Bass and other teleosts; for example, gradual elevation in cortisol and lactate and 10–20-min delays in glucose response have been reported for Brown Trout Salmo trutta (Pickering et al. 1982), Rainbow Trout Oncorhynchus mykiss (Meka and McCormick 2005), Bronze Bream Pachymetopon grande (Pringle et al. 2020), and Bluegill Lepomis macrochirus (Louison et al. 2017) after exposure to acute stressors.

Interestingly, fight time and handling duration may not have as strong an influence on physiological stress in angled Kelp Bass as has been reported for other game fish species. Fight time, air exposure (i.e., handling time), and their summed duration (i.e., landing time) were not significantly correlated with any of the measured biomarkers (Table 1). Although the relationship was not significant, glucose and lactate both exhibited general increases when landing time increased (Figure 3), whereas cortisol declined. Some angling studies have found that fight duration and air exposure were directly correlated with the degree of physiological stress experienced by fish during catch and release (e.g., Gingerich et al. 2007; Brownscombe et al. 2015), while other studies have reported that no strong relationship was evident (e.g., Brownscombe et al. 2014). The lack of significant relationships between capture conditions and Kelp Bass stress in this study may be due to the general brevity of the stressors and the minimal variation in conditions among individuals (Lamansky and Meyer 2016; Lennox et al. 2016a, 2016b). This may also explain why the magnitude of the physiological response in Kelp Bass was less than that reported for species in other angling studies where fish were subjected to more intense and/or prolonged angling and handling stressors. In one study (Lowe and Kelley 2004; McGarigal et al. 2020), California Sheephead Semicossyphus pulcher were fought on the line for up to 20 min, compared to the 26.3-s average fight time for Kelp Bass in this study, so it is not surprising that the peak cortisol response for California Sheephead was three times that of Kelp Bass. Likewise, bonefish Albula spp. in a study by Brownscombe et al. (2015) experienced prolonged air exposure (2 min), whereas Kelp Bass were handled for an average of 47 s. Variability in the literature regarding the magnitude of stress responses in fish may also be due to interspecific differences in life history characteristics (Barton 2002; Fanouraki et al. 2011; Bordin and Freire 2021) or the environmental conditions during capture (e.g., water temperature; Cooke and Suski 2005; Gingerich et al. 2007; Brownscombe et al. 2015). Water temperature has sometimes been identified as a contributing factor to fish stress and postrelease mortality (Cooke and Suski 2005), but temperature did not significantly influence fish stress in the present study, either because Kelp Bass are resilient to thermal changes or because extreme temperature conditions were never experienced. Nevertheless, careful consideration of the methods and capture conditions is important when comparing physiological stress responses between studies and/or species.

Sensitivity to capture stress in Kelp Bass appears to be partially size dependent, which has important implications for size-based fishing regulations. Fish size may play a role in the glucose and lactate responses of angled Kelp Bass, but high variance among similar-sized fish (within a treatment group) suggests that physiological mechanisms may be responding to cumulative differences in capture conditions, prior stressors experienced, and intrinsic difference between the sexes. Circulating glucose levels may also depend on the timing and/or quantity of food consumed prior to capture, which would influence the energy stores available for glucose production and the rate of glucose depletion during confinement (Bever et al. 1977; Black and Love 1986; Jiang and Zhang 2003). Lactate exhibited a stronger—but still not significant—correlation with fish size (Figure 4), possibly because smaller fish often experienced relatively shorter fight times (i.e., less exertion). Our holding tanks exceeded the recommended live-well size of 3.78 L (1 gal) per 0.453 kg (1 lb) of fish (Gilliland and Schramm 2009), but smaller individuals were better able to maintain moderate activity levels in the confinement tanks than larger individuals (personal observation), which would aid the rate of lactate washout from muscle tissue and clearance from circulation (Choi et al. 1994; Milligan et al. 2000; Currey et al. 2013; Weber et al. 2016; Zhang et al. 2018). Fish size was clearly a significant factor in the cortisol response (Table 1; Figure 4), with larger fish exhibiting relatively lower cortisol levels within each treatment group. Predictive modeling results indicated that Kelp Bass below the current legal size limit (35.5 cm TL) are likely to experience cortisol elevations up to an order of magnitude higher than legal-sized individuals (Figure 5). Similar size dependence has been reported in the aquaculture literature, where the sensitivity of the HPI axis to acute stress diminishes as fish grow and mature (Barton et al. 1985; Schreck et al. 1997; Schreck and Tort 2016). Based on age and growth curves (Love et al. 1996), size at maturity (22–27 cm TL; Love et al. 1996), and assumptions of annual reproduction, the current minimum size limit is intended to provide Kelp Bass with two to five spawning seasons before they enter the fishery (CDFW 2013, 2019a). However, because large females are highly valuable for recruitment (Beldade et al. 2012; Gwinn et al. 2015) and smaller fish appear to be more stressed by angling, slot limits may be a viable alternative to minimum size limits, although this needs further investigation.

Kelp Bass of all sizes showed a resiliency to capture stress and a capacity for rapid physiological recovery even after repetitive angling events. All Kelp Bass that were recaptured and rapidly sampled had cortisol, glucose, and lactate levels consistent with baseline concentrations, and several of those fish were recaptured within 24 h, indicating that recovery may occur within hours of release (Figure 6). Kelp Bass also appeared to be resilient to repetitive angling events because all individuals that were recaptured multiple times exhibited both physiological recovery (i.e., baseline biomarker levels) and an ability to mount normal endocrine responses (i.e., biomarker elevation) after angling and confinement (Figure 7). Several studies have reported similar rapid recovery rates in captive or cultured fish (Pickering and Pottinger 1989; Pankhurst and Dedualj 1994; Arkert et al. 2020); however, game fish often experience chronic stress during confinement and only recover to their baseline physiological state when returned to their natural environment (Lowe and Wells 1996; Galima 2004; McGarigal et al. 2020). Repetitive acute stressors in some species, like Olive Flounder Paralichthys olivaceus (Lim and Hur 2018), can result in symptoms similar to chronic stress, including consistently elevated cortisol and an inhibited ability to respond to additional stressors (Schreck 2000; Pankhurst 2011; Schreck and Tort 2016; Samaras et al. 2021). Time at liberty between recaptures in this study varied from hours to months, so it is plausible that repetitive angling over a shorter time frame would have greater, and possibly long-term, impacts on Kelp Bass physiology (Mugnier et al. 1998; Schreck and Tort 2016). However, because our focus was on short-term individual angling effects, we did not evaluate other tertiary end points (e.g., reproduction, growth, and disease resistance) and we can only speculate about long-term and population-level impacts. We also suggest that future research should evaluate the impacts of cumulative anthropogenic (e.g., angling) and environmental (e.g., climate change) stressors on Kelp Bass physiology. Understanding the short- and long-term impacts of these sources of stress on fish physiology will help to inform regulatory decisions and promote the sustainability of this valuable fishery.

The most significant behavioral response exhibited by Kelp Bass to angling stress was a short-term decline in activity, although the fish resumed normal activity levels within the 36-h tracking period. Unstressed (i.e., control) Kelp Bass demonstrated strong diel behavior, with higher activity levels at night compared to the daytime (Figure 8A; Hobson 1979; Sabino and Zuanon 1998; Young and Winn 2003). Adult Kelp Bass are ambush predators and perform rapid burst swimming when chasing prey; therefore, acceleration can be a useful proxy for foraging-associated activity in wild fish (Sabino and Zuanon 1998; Skinner et al. 2009; Horie et al. 2017; our in situ observation). Angled Kelp Bass exhibited reduced activity levels for approximately 30 h postrelease, which was followed by rapid recovery and elevated activity (above baseline) for the last 6 h of the tracking period (Figure 8D). These results suggests that foraging may be inhibited for angled fish after release, although the underlying cause of this change in behavior may be biochemical, physical, or both. Intragastric insertion of the transmitters could block or reduce gut capacity, thereby hindering foraging behavior (Bridger and Booth 2003), but this would have affected both angled and control fish equally. Few studies have properly controlled for the effects of anesthesia on fish behavior (Ross and Ross 2009; Topic Popovic et al. 2012), so it is possible that the treatment of angled fish with MS-222 may have contributed to the reduced activity observed after release. However, we consider this an unlikely explanation because Kelp Bass were lightly sedated and appeared very active during the revival period in holding tanks prior to release. Additionally, delayed (or reduced) feeding periods after acute stress have been reported for several species to which anesthesia was not administered, including Rainbow and Brown trouts (Pickering et al. 1982; Pankhurst et al. 2008a, 2008b), Coho Salmon O. kisutch (Wedemeyer 1976), and European Bass Dicentrarchus labrax (Leal et al. 2011). An alternative explanation is that elevated corticosteroid circulation during stress via activation of the HPI axis acts to suppress ghrelin (an appetite-stimulating peptide) and increase corticotropin-releasing hormone concentrations (an appetite inhibitor), which would both result in reduced food intake and delayed feeding behavior (Unniappan and Peter 2005; Bernier 2006; Volkoff and Peter 2006; Pankhurst et al. 2008a, 2008b; Volkoff 2016; Conde-Sieira et al. 2018). Jaw damage can also occur during angling and handling when anglers use a lip-grip device to hold fish horizontally out of water, which stresses jaw ligaments and can damage tendons and muscles, impairing a fish's ability to grip and manipulate prey (Meka 2004; Meka and Margraf 2007; Danylchuk et al. 2008; Lennox et al. 2016a, 2016b; Skaggs et al. 2017; Thompson et al. 2018). These injuries can be prevented by using a rubber dip net to handle fish in water, which has the added benefits of reducing air exposure and avoiding abrasions that remove the protective mucus layer (Suski et al. 2007; Danylchuk et al. 2008; Colotelo and Cooke 2011; Lizée et al. 2018; Foster et al. 2020). After the period of reduced food intake, elevated foraging activity exhibited by Kelp Bass may be an effort to replace energy stores depleted while fasting (Lim and Hur 2018; Hvas et al. 2022). A better understanding of these bioenergetic consequences from the angling of Kelp Bass is a subject that warrants further study and would require longer monitoring periods that evaluate the duration of elevated foraging and the potential negative impacts on growth and fitness of repetitive angling and frequent fasting.

Kelp Bass showed high individual variation in ROM and area use, with no significant postrelease impacts, which supports our conclusion that this species is quite resilient to catch-and-release practices. Control fish at night moved shorter distances and stayed primarily within the core area of their home range compared to daytime, when fish moved farther, covered a larger area, and exhibited greater individual variation (Figure 8B, C). This diel behavior is consistent with typical ambush predator behavior (Hobson 1979; Love and Ebeling 1986; Sabino and Zuanon 1998; Young and Winn 2003; Abecasis et al. 2013; Özgül et al. 2019) and with Kelp Bass behavior observed previously by Lowe et al. (2003). Because we were unable to identify the sex of tracked individuals, natural differences between males and females may account for some of the individual variability that was observed. Sexual differences may be especially pronounced during the main spawning season (June–August), when males may be more aggressive and active and may patrol larger areas, as has been reported for other group spawners like Atlantic Cod (Dean et al. 2014). Reduced movement and impaired mobility after release have been well documented in other teleosts, such as the Common Carp Cyprinus carpio (Rapp et al. 2012, 2014), bonefish Albula spp. (Brownscombe et al. 2013, 2014), White Sturgeon Acipenser transmontanus (McLean et al. 2020), and black bass Micropterus spp. (, Cooke et al. 2002, 2021), but such effects were not apparent in this study. Likewise, home-ranging species often keep to their smaller core range when stressed, which minimizes movement and energetic demands during recovery, but angled Kelp Bass showed no observable reduction in home range area (Kawabata et al. 2008; Le Pichon et al. 2015; McGarigal et al. 2020). It is possible that subtle behavioral impacts went undetected in this study due to the high behavioral variability; however, the lack of any significant changes in postrelease movement or area use suggests that these impacts are likely negligible.

Despite the evidence indicating that Kelp Bass are resilient to catch-and-release practices, this fishery would still benefit from conservation of critical habitat, a shift in management toward slot limits, and more widespread angler engagement and education on best practices. While current regulations have prevented further population declines (Bellquist et al. 2017), fisheries managers have not accounted for the potential sublethal impacts of increasing catch-and-release rates on individual physiology and behavior or the size-dependent responses of Kelp Bass to angling stress. We suggest that implementing slot limits (e.g., 26–36 cm TL [10–14 in TL]) would promote stock sustainability by helping to maintain natural age- and size-structured populations (Gwinn et al. 2015; Bellquist et al. 2017; Ahrens et al. 2020; Kasper et al. 2020), and these limits have been proven as a successful management strategy for White Sturgeon (CDFW 2019b) and Northern Pike Esox lucius (Pierce 2010). However, setting the appropriate limits for Kelp Bass requires further consideration, as does the potential socioeconomic impact of these regulations on the fishing community—particularly on anglers who consider “trophy fishing” (i.e., catching very large individuals) to be their primary motivation (Holland and Ditton 1992; Arlinghaus 2006; Beardmore et al. 2011). The success of new fishing regulations can also be aided by changing the perspective of recreational fishing communities from “harvest” to “sustainability,” which has numerous challenges but can be a valuable alternative to mandatory regulation (see Cooke et al. 2013; Delle Palme et al. 2016; Danylchuk et al. 2018). One way to accomplish this is through outreach programs that engage anglers in fisheries science and education on best practices (Cooke and Suski 2005; Holder et al. 2020; Scyphers et al. 2021). In this study, participation by recreational anglers ensured that capture methods were consistent with the wider fishery and anglers felt greater stewardship for the resource and support for regulatory changes after helping to conduct the science themselves. While differences in angler behavior had no noticeable impact on Kelp Bass stress during catch and release, we still recommend following general best practice guidelines, such as minimizing fight times, supporting fish weight when handling, limiting air exposure, avoiding warm water temperatures, and using circle hooks or barbless J-hooks (Cooke and Suski 2005; Brownscombe et al. 2017a, 2017b). The overall resiliency of the Kelp Bass fishery in southern California would also be supported by conserving and/or establishing critical habitat areas (i.e., high vertical relief and rocky boulders with lots of refuge cavities), which aid in postrelease recovery and provide both shelter and foraging habitat (Love et al. 1996; Lowe et al. 2003; Ginther and Steele 2018; McGarigal 2018; Miller et al. 2018). As fish populations at the local, regional, and global scales face escalating anthropogenic and environmental stressors, it will be increasingly important to rely on science-based management decisions to ensure the long-term sustainability of game fish populations and their valuable recreational fisheries.

ACKNOWLEDGMENTS

We thank our collaborators at CDFW for their funding and support in conducting this research in the CIMLR. We also thank the USC Sea Grant Program and the USC Dornsife Wrigley Graduate Fellowship program for supporting the graduate student involved in this project, C. R. McGarigal, as a Sea Grant Trainee and Fellow. Assistance and collaboration in the immunoassay and metabolite measures by Jesus Reyes (Pacific Coast Environmental Conservancy) are gratefully acknowledged. We thank the many recreational anglers and angling clubs who assisted the project with funding and fishing efforts. This project would not have been possible without the support from the staff at the USC Wrigley Marine Institute on Catalina Island, where the field work was conducted.There is no conflict of interest declared in this article.