Skip to main content
NCBI home page
Conservation Physiology logoLink to Conservation Physiology
. 2024 Oct 28;12(1):coae074. doi: 10.1093/conphys/coae074

Repeatability of swimming activity of the Patagonian grouper Acanthistius patachonicus based on accelerometry

Leonardo A Venerus 1,#,✉, Paolo Domenici 2,3,4,#, Stefano Marras 5,6, Lucas E Beltramino 7, Javier E Ciancio 8
Editor: Rodgers Essie
PMCID: PMC11519044  PMID: 39474043

Abstract

The study of repeatability in behaviour and activity level can be used to evaluate inter-individual differences, which are fundamental to assess the resilience of populations to environmental variation. Previous work on repeatability in wild fish populations has largely been based on acoustic telemetry or mark-and-recapture and has revealed repeatable activity patterns over relatively long periods in a number of species. Although accelerometry is a promising tool for investigating the swimming activity of fish in the wild, little is known about the repeatability of accelerometry-based traits in wild fish. Here, we used external accelerometers to investigate the swimming activity of the Patagonian grouper Acanthistius patachonicus, a rocky-reef fish with high site fidelity, which ensures a high recapture rate of accelerometer tags. Accelerometry was used to investigate the short-term repeatability of a number of activity traits, including swimming, hovering, daily median tailbeat frequency, percentage of high tailbeat frequency and total number of tailbeats at different times of the year. We found that all of the variables are repeatable over the daily scale and four out of five variables are repeatable over weekly periods. Overall, our work suggests that these traits are individual-specific for the short time period investigated. In addition, the percentage of time spent in swimming and hovering was greater in the warm season compared to the cold season, suggesting higher activity levels related to higher temperatures. These results suggest that activity traits related to swimming are repeatable and likely related to the physiological state of each individual. Finally, our work shows that accelerometry can be considered a valuable tool to explore inter-individual differences with potential applications for assessing the resilience of wild populations.

Keywords: Behaviour, free-living marine fishes, Northern Patagonia, rocky-reef fishes, tailbeats, tri-axial accelerometers

Introduction

Swimming activity can vary both amongst and within species (Kolok et al., 1998; Claireaux et al., 2007; Marras et al., 2010). Certain pelagic species such as tuna and clupeids are known to be very active, whilst benthic species tend to be more sedentary (Videler, 1993; Webb, 1994). However, even within a given species, it is now known that activity level can vary amongst individuals and that this is correlated with physiological and behavioural characteristics (Killen et al., 2013). The tendency of an individual to express a certain behaviour or activity level can be investigated by assessing its trait repeatability (Killen et al., 2016). The assessment of repeatability is a fundamental issue in ecology and evolution, because it allows to quantify the extent to which behavioural or physiological traits can be shaped by selection (Hitchcock et al., 2015). For this reason, trait repeatability has been suggested to provide the upper limit for trait heritability (Dochtermann et al., 2015). Within the context of conservation, the level of within-individual variability in behaviour and physiology is related to the vulnerability of individuals that must cope with anthropogenic environmental change (e.g. pollution, climate warming, habitat loss, etc.). Understanding the role of within-individual variation on population responses to the environment is an open area for research (Killen et al., 2016).

In the past couple of decades, there has been a growing interest in the repeatability of traits in fishes, with much work being based on laboratory observations of metabolism, locomotor activity and behavioural traits such as boldness and aggression (Nespolo and Franco, 2007; Williams, 2008; Bell et al., 2009). Repeatability has been defined as the proportion of within-population variation, as opposed to within-individual variation (Biro and Stamps, 2015). A highly repeatable trait therefore is a trait that is consistently repeated in each individual of the population. Although most evidence of trait repeatability in fishes is based on laboratory studies (Nelson and Claireaux, 2005; Oufiero and Garland, 2009; Marras et al., 2010, 2011; Killen et al., 2016) or on studies of individuals who lived at sea between measurements (e.g. Handelsman et al., 2010; Vandamm et al., 2012), there is a growing interest in quantifying repeatability in the field (e.g. Claireaux et al., 2007; Forsythe et al., 2012; Taylor and Cooke, 2014; Harrison et al., 2015; Killen et al., 2016; Alós et al., 2017; Villegas-Ríos et al., 2017; Griffin et al., 2023).

Fieldwork poses a number of challenges that include partitioning the repeatability resulting from environmental effects from the one intrinsic to individual animals (Killen et al., 2016), as well as the technical difficulties of measuring behaviour and activity in the wild. In particular, a number of studies have investigated the repeatability of activity levels and spatial patterns in fishes in natural settings, mainly based on acoustic telemetry. For example, Hanson et al. (2010) found that the daily distance travelled and mean daily swimming speed in largemouth bass Micropterus salmoides were repeatable on a daily basis, but not amongst seasons or across multiple winters. Working on a different species (Salvelinus confluentus), Taylor and Cooke (2014) found that the inter-individual movements were consistent across two seasons. More recently, Alós et al. (2017) found that activity levels and circadian rhythmicity (i.e. awaking time rest onset and rest duration) was repeatable for up to 14 days in the free-ranging pearly razor fish, Xyrithchys novacula. Harrison et al. (2015) found repeatable, cross-contextually consistent, personality-dependent home range, movement, dispersal from release and site fidelity (i.e. caused by individual differences in spatial use behaviours) over 2 years in Lota lota. Furthermore, Villegas-Ríos et al. (2017) found that diel vertical migration, activity, home range and dispersal of a marine species (Atlantic cod Gadus morhua) showed moderate to high individual repeatability in all behavioural traits.

However, common proxies for activity based on acoustic telemetry data proved to be poor descriptors of the activity levels (Pereñíguez et al., 2022), at least for certain species that spend a large proportion of their time inside crevices and move without significant displacement, like the Patagonian grouper A. patachonicus (Ciancio et al., 2016). The use of data loggers, including high-frequency tri-axial accelerometers, has provided a promising tool to monitor activity and behaviour of animals when they cannot be visually tracked (e.g. Murchie et al., 2011; Kawabata et al., 2014; Vázquez Diosdado et al., 2015; Beltramino et al., 2019; DeSantis et al., 2020). This method also has the advantage that accelerometers can record large amounts of detailed information as they can register at frequencies of up to hundreds of Hertz (Hz). Additionally, accelerometers can be used as a proxy to monitor energy expenditure by relating animal activity with energy consumption (Wilson et al., 2006, 2020).

Although fish are particularly suited for behavioural studies employing tethered accelerometers, one of the mayor challenges is represented by the recovery of the external accelerometer. For this reason, most of the studies have been performed in mesocosms (Brownscombe et al., 2014) or aquaria (Wright et al., 2014), where the recovery of the loggers is highly reliable. In recent years, there has been a growing interest in the use of external accelerometers in the field. This is mainly due to the use of novel technologies and fishing techniques that increase the chances of recovering the tags and the collected data. Some examples include the development of tags that acoustically transmit averages or ranges of acceleration values (Murchie et al., 2011; Carbonara et al., 2021; Pereñíguez et al., 2022; Lennox et al., 2023), pop-off systems coupled with a VHF transmitter (Gleiss et al., 2019) and also more specialized expertise of researchers to recapture the fish from the field and recover the tags (Gleiss et al., 2013; Ciancio et al., 2016; Beltramino et al., 2019). This has led to an increasing number of studies using accelerometers to assess a diverse array of questions, such as the description of the post-release effects of fishing (Whitney et al., 2016), the activity landscape for some species (Papastamatiou et al., 2018), the swimming kinematics of large pelagic predators (Gleiss et al., 2019) and the foraging strategies in billfish (Marras et al., 2015). Despite the high-definition information they provide, accelerometers have been used infrequently to study repeatability of traits. Some examples include their use in free-living fishes (Rudd et al., 2021), terrestrial mammals (Keegan et al., 2011; Gottlieb et al., 2013; MacKay et al., 2014; Stiegler et al., 2022) and birds (Grémillet et al., 2018; Rotics et al., 2021).

The Patagonian grouper A. patachonicus (Jenyns, 1840) is a sedentary, long-lived species that inhabits temperate rocky reefs in the southwest Atlantic Ocean (Irigoyen et al., 2013). These fish are mostly observed in an upright body position outside reef crevices, up to a few metres above the seabed at near neutral buoyancy, or resting inside crevices at different roll angles (Ciancio et al., 2016; Beltramino et al., 2019). Previous studies showed that the activity budget for this species is modulated by water temperature and suggested that other environmental forcing variables (e.g. tide and ambient light levels) might have some effects at the individual level (Beltramino et al., 2019). For example, some fish displayed a diel activity rhythm that closely followed ambient light levels, whilst the behaviour of other individuals was not so strongly affected by light (Beltramino et al., 2019). Furthermore, this species has a high site fidelity, particularly during the warmer months when the Patagonian grouper is more abundant in shallow rocky reefs, and therefore has a high chance of recapture (Irigoyen, 2010). For these reasons, the Patagonian grouper was chosen as an appropriate model to study repeatability of activity traits in the field using accelerometer tags.

Here, we used external accelerometers to detect different swimming patterns (including hovering, swimming and other metrics related to tailbeat frequency) of the Patagonian grouper in the field in the short term (i.e. day-to-day and 1 week) within two periods (hereafter ‘seasons’ for simplicity): cold (July–October) and warm (December–May). We hypothesize that (1) swimming patterns are repeatable within the short term and (2) activity is higher during the warm season. This will provide insights on the repeatability of activity traits in marine fishes, as well as an assessment of the feasibility of accelerometry to explore inter-individual variability in activity in wild populations.

Materials and Methods

Animals and experimental protocol

A total of 26 adult individuals of the Patagonian grouper, mean total length (TL) ± SD = 30.7 ± 3.6 cm, were captured between 2017 and 2019, on two reefs, at depths between 6.5 and 9.2 m, in the Golfo Nuevo, Northern Patagonia, Argentina (~43°S 65°W) (Supplementary Material S1). Fish were captured by scuba divers inside reef crevices by pole-hooking (Irigoyen and Venerus, 2008) or with a fish-landing net. After capture, fish were slowly brought to the surface in a dip net, and equipped with an accelerometer X16-mini (Gulf Coast Data Concept Company), with dimensions of 50 × 21 × 7 mm and mass of 17 g in air (Fig. 1). Accelerometers were encapsulated in heat-shrink tubes with enough air to achieve slight negative buoyancy of the whole package, and fixed to the right side of the body, at ~0.4 length, which is known to be near the centre of mass for teleosts (Webb, 1978), by using plastic straps or surgical thread. In both cases, we used surgical needles to pass the strap or the thread under the fish skin. The plastic straps were closed and the surgical thread was firmly knotted, in both cases by pressing against the heat-shrink tubes. The attachment procedure was preceded by spreading a solution of lidocaine 10% (Fullcaina, LAFEDAR) in the tagging area. In addition, the length of each fish was recorded and a spaghetti tag (Floy FD-68B) was attached at the base of the dorsal fin for individual identification. After a minimum of 8 days of free-living period, fish were recaptured using the same fishing technique and the accelerometers were removed from the fish. Fish were then gently released back into their crevices.

Figure 1.

Figure 1

Open in a new tab

Patagonian grouper A. patachonicus tagged with an external accelerometer in the Golfo Nuevo, Argentina. Photograph by Lic. Agustín Biasotti (UNPSJB).

Mean water temperature in the area reaches a maximum in March and a minimum in September (Dellatorre et al., 2012). We defined two sampling seasons based on mean water temperature: warm (ca.16–18°C) and cold (ca. 8–10°C). Tagging, undertaken in December, and from February to May, was considered representative of the warm season, whereas fish tagged from July to October were considered representative of the cold season.

Ethical declaration

In Argentina there is not a legal instrument in place to ensure animal care during experimentation. However, the procedure for tagging and recapturing groupers was approved by the Institutional Animal Care and Use Committee (IACUC) from the CCT CONICET-CENPAT, Puerto Madryn, Chubut, Argentina.

Data analysis

We used five dependent variables to explore repeatability in swimming activity of the Patagonian grouper; two of them (‘Swimming’ and ‘Hovering’), estimated by a classification algorithm, represent the percentage of time invested in swimming and hovering, respectively. The other three variables (‘Median tailbeat frequency’, ‘Percentage of high tailbeat frequency’ and ‘Total number of estimated tailbeats’) were based on proxies related to tailbeat frequency. The variables used are the following:

  • 1) Swimming, defined as the percentage of time that fish spent swimming during a given day. We considered swimming as all the data points classified as such by a machine learning algorithm (see below). This can be considered as a proxy for when the fish was actively swimming at any speed, using the tail or the pectoral fins.

  • 2) Hovering, defined as the percentage of time fish spent holding its position in the water column, during a given day. This was estimated using the data points classified as such by a classification algorithm (see below).

  • 3) Median tailbeat frequency (MTBF), which was calculated as the median for each day, using a class width of 0.5 Hz, between 0.5 and 12 Hz (zero tailbeats were excluded from the calculation of the median). This lower limit is in line with previous work on fish of similar size, which yielded low speed ~0.6 length/s (L/s) or less, at ~1.2 Hz (Herskin and Steffensen, 1998; Steinhausen et al., 2005; Ohlberger et al., 2007). This is a proxy of how fast fish were swimming whilst active.

  • 4) The percentage of high tailbeat frequency (i.e. >4 Hz) (HTB%) out of the total tailbeats per day. HTB is a proxy for swimming behaviour higher than sustainable. The threshold was based on previous work, which shows that maximum sustained speed for fish of similar size, is ~2–3 L/s (Videler, 1993). Using a stride length of 0.75 length (L), which is common in fish (Videler, 1993), this swimming speed corresponds to a tailbeat frequency ~2.6–4 Hz. Therefore, using a frequency of 4 Hz as a threshold should yield all events in which fish swam at a speed above their maximum sustained. This is an indication of how often fish are swimming in a burst-like mode whilst they were active.

  • 5) The total number of estimated tailbeats (TTB) per day and individual, using 0.5 Hz as the minimum value as for MTBF. This is an indication of the activity level implying BCF swimming (Body Caudal Fin swimming sensuWebb, 1984) with tailbeat frequency (TBF) ≥0.5 Hz. This variable was log-transformed to deal with heteroscedasticity in the data (see Results).

Acceleration data were extracted from the accelerometers by using the procedure described in Beltramino et al. (2019). Although acceleration data were recorded for up to 14 days in a few individuals, a period of 8 days was dictated by the need of maximizing the number of fish and repeated days in the sample. Data from the first recording day (Day 1) were not included in the analysis to avoid potential manipulation effects and because they did not include a full day of recording. Similarly, the last day (Day 8) was not included because of the lack of a full day of recording in most cases. Hence, the longest period available for testing repeatability was from Day 2 to Day 7.

The details of the classification procedure are given in Supplementary Material S2. A training dataset, compiled from a total of seven fish (wild and captive) tagged with accelerometers and video recorded whilst performing different swimming activities, was used to classify the acceleration signals of the free-living fish into four categories selected for this species (details in Beltramino et al., 2019). Amongst those categories, only ‘Swimming’ and ‘Hovering’, implying active movement rather than posture, were used in the present study to test repeatability in swimming activity in the Patagonian grouper. The daily percentage of time spent in each of these two activities was estimated by using the K-nearest neighbour classification algorithm (KNN, Bidder et al., 2014). The other two categories, not used in this study, were ‘Static on bottom’, which implies that fish were resting on the bottom and movement was restricted, and ‘Lying on its side’, which includes fish resting at extreme roll angles (Ciancio et al., 2016; Beltramino et al., 2019).

To estimate a proxy for TBF, we split the ‘Sway’ raw acceleration (using data at 25 Hz for this analysis) into its static and dynamic components (Shepard et al., 2008). Then, we used the dfreq function (2-s window with 50% overlap, and a 10% threshold) of the package ‘seewave’ (Sueur et al., 2008) of the R software (R Core Team, 2020) to identify tailbeat cycles in the sway dynamic acceleration. The estimated TBF was validated in a swim tunnel, in which we swam tagged fish at controlled speeds (regression between the means of video-counted TBF and accelerometer-derived TBF: R2 = 0.76, P < 0.001, n = 19; unpublished data).

To perform all the analyses described, we coded ad hoc routines in R (R Core Team, 2020). The KNN classification requires the ‘class’ package (Venables and Ripley, 2002). To estimate the variances of roll and sway, we used the packages ‘xts’ (Ryan and Ulrich, 2023), ‘TTR’ (Ulrich, 2021) and ‘caTools’ (Tuszynski, 2021).

Analyses of repeatability

The repeatability in the five activity variables ‘Swimming’, ‘Hovering’, ‘MTBF’, ‘HTB%’ and ‘TTB’ was tested using two different approaches. The first approach is based on the Spearman correlation coefficient (Marras et al., 2010). This coefficient was used to test the repeatability of the daily value for each variable versus the same value for the successive day, e.g. Day 2 versus Day 3, Day 3 versus Day 4 and so on (Hanson et al., 2010). We also focused on the repeatability on a weekly basis by comparing Day 2 versus Day 7, i.e. the longest time interval available. All Spearman coefficients were estimated in i) individuals tagged in the cold and in the warm seasons separately and ii) all individuals pooled together without considering the factor ‘Season’ (see below). P-values were adjusted for multiple comparisons amongst successive days for each variable and season (warm, cold and both) by using the Holm sequential correction (Holm, 1979) with the p.adjust function of the R software (R Core Team, 2020).

The second approach used to evaluate repeatability was based on the Intra-class Correlation Coefficient (ICC), which ranges from 0 to 1: it is 0 when all individuals have the same mean, and it is 1 when all individuals have different means but all measurements on the same individual are identical (i.e. perfect repeatability) (Becker, 1984; Lessells and Boag, 1987). ICC values were estimated from a series of repeated-measures models, fitted for each of the four activity variables under study. We used linear mixed-effects models because they present some advantages over other techniques (e.g. analysis of variance (ANOVA)); mainly, they do not hinge on either sphericity assumptions or complete data (Howell, 2009). In our study, e.g. the dfreq function failed to identify tailbeat cycles during Day 7 for one of the tagged fish during the cold season; hence one observation was missing for the variables ‘MTBF’, ‘HTB%’ and ‘TTB’ and an ANOVA approach was not feasible.

We fitted a series of models for each activity variable by including ‘Season’ with levels cold and warm, and ‘Day’, with levels Day 2,…Day 7, as fixed explanatory variables, TL as a covariate and ‘Fish identity’ (i.e. subject) as the random variable. Four models were fitted for each activity variable; all of them included the explanatory variables ‘Season’ and ‘Day’. In addition, two models included the covariate TL (with and without the interaction term ‘Season × Day’), and the other two did not include TL (with and without the interaction term ‘Season × Day’). Models were fitted by Maximum Likelihood (ML) (Zuur et al., 2009) and were compared by using the Akaike Information Criterion, AIC (Akaike, 1973). For each activity variable, the model bearing the minimum AIC value was considered the best model. Those models differing by more than 2 in the AIC value with respect to that from the best model were discarded. The chosen models for each activity variable were refitted by restricted maximum likelihood (REML) before estimating ICC (Zuur et al., 2009). A Gaussian error distribution with identity link was used for all models. The total number of tailbeats (‘TTB’) were log-transformed because data were clearly heteroscedastic. Diagnostic plots for the fitted models were obtained with the simulateResiduals function of the ‘DHARMa’ package (Hartig, 2022) of the R software (R Core Team, 2020).

We estimated the adjusted ICC (Inline graphic), which represents repeatability that controls for confounding effects (Nakagawa and Schielzeth, 2010; Nakagawa et al., 2017; Stoffel et al., 2017), and is calculated by the expression:

graphic file with name DmEquation1.gif

where: Inline graphic is the variance of the random effect, which represents the between-subject variation, and Inline graphic, the residual variance, which represents the within-subject variation. In addition, confidence intervals for repeatability were estimated by means of a parametric bootstrap (Stoffel et al., 2017). We interpreted the Inline graphic values following Liljequist et al. (2019), who suggested making qualitative interpretations of repeatability based on the categories: ‘poor’ (ICC < 0.5), ‘moderate’ (0.5–0.75), ‘good’ (0.75–0.9) and ‘excellent’ (ICC > 0.9).

We used the R packages ‘lme4’ (Bates et al., 2015) and ‘rptR’ (Stoffel et al., 2017) to fit linear mixed-effects models, and to estimate Inline graphic values and their confidence intervals, respectively. Significance for statistical tests was set at P < 0.05 level.

Results

All five dependent variables were significantly repeatable when Spearman correlations were analysed on a day-successive day basis for the pooled seasons, and in most cases when they were evaluated for each season separately, particularly for the warm season (Fig. 2 and Table 1). During that season, all correlations on a day-successive basis were significant for ‘Hovering’, ‘MTBF’ and ‘TTB’, and most were significant for ‘HTB%’ (four out of five). However, only two out of five correlations were significant for ‘Swimming’. During the cold season, all correlations on a day-successive basis were also significant for ‘Hovering’ and ‘TTB’, and most were significant for ‘Swimming’ (four out of five) and ‘MTBF’ (three out of five). The only exception to this general pattern was ‘HTB%’, which showed non-significant repeatability for most day-to-day pairings during the cold season (four out of five). Overall, this means that within the very short term (1 day), the swimming activity of fish was highly consistent. The Spearman correlations estimated on the longest time interval (Day 2 vs Day 7, i.e. weekly basis), were significant for four out of five variables: ‘Hovering’, ‘MTBF’, ‘HTB%’ and for ‘Swimming’, but in the latter case, only when both seasons were pooled together, probably due to the higher number of cases (Fig. 3 and Table 2).

Figure 2.

Figure 2

Open in a new tab

Relationship in daily percentage of time spent (a) swimming and (b) hovering, and in daily (c) median number of tailbeats (‘MTBF’), (d) percentage of high tailbeat frequency (‘HTB%’) and (e) logarithmized total number of tailbeats (‘TTB’), between two consecutive days, for the Patagonian grouper, discriminated by season. Open red circles: hot season; open blue circles: cold season. A bisector (dotted) line is included as reference in each plot.

Table 1.

Results of the Spearman correlations between successive days for five dependent variables used to evaluate repeatability, based on acceleration data collected from 26 free-living Patagonian groupers tagged in the Golfo Nuevo, Península Valdés. The asterisks show significant correlations (P < 0.05)

Dependent variable Warm season Cold season Both seasons pooled
rho (n = 13) P rho (n = 13) P rho (n = 26) P
Swimming
Day 2 vs Day 3 0.500 0.123 0.346 0.123 0.698 <0.001*
Day 3 vs Day 4 0.401 0.162 0.835 0.001* 0.832 <0.001*
Day 4 vs Day 5 0.692 0.022* 0.681 0.011* 0.765 <0.001*
Day 5 vs Day 6 0.412 0.162 0.703 0.011* 0.626 <0.001*
Day 6 vs Day 7 0.654 0.031* 0.775 0.004* 0.720 <0.001*
Hovering
Day 2 vs Day 3 0.901 <0.001* 0.714 0.009* 0.791 <0.001*
Day 3 vs Day 4 0.923 <0.001* 0.775 0.004* 0.898 <0.001*
Day 4 vs Day 5 0.841 <0.001* 0.808 0.002* 0.856 <0.001*
Day 5 vs Day 6 0.549 0.026* 0.687 0.010* 0.628 <0.001*
Day 6 vs Day 7 0.901 <0.001* 0.599 0.015* 0.795 <0.001*
MTBF
Day 2 vs Day 3 0.660 0.035* 0.400 0.175 0.569 0.001*
Day 3 vs Day 4 0.656 0.035* 0.729 0.011* 0.675 <0.001*
Day 4 vs Day 5 0.598 0.046* 0.628 0.043* 0.620 0.001*
Day 5 vs Day 6 0.514 0.046* 0.619 0.043* 0.680 <0.001*
Day 6 vs Day 7 0.561 0.046* 0.381 a) 0.175 0.610 b) 0.001*
HTB%
Day 2 vs Day 3 0.659 0.021* 0.445 0.255 0.597 0.003*
Day 3 vs Day 4 0.451 0.061 0.643 0.044* 0.572 0.003*
Day 4 vs Day 5 0.583 0.036* 0.443 0.255 0.533 0.003*
Day 5 vs Day 6 0.697 0.016* 0.182 0.276 0.578 0.003*
Day 6 vs Day 7 0.726 0.012* 0.448 a) 0.255 0.685 b) <0.001*
TTB
Day 2 vs Day 3 0.648 0.017* 0.602 0.044* 0.650 <0.001*
Day 3 vs Day 4 0.747 0.007* 0.791 0.003* 0.818 <0.001*
Day 4 vs Day 5 0.927 <0.001* 0.553 0.044* 0.806 <0.001*
Day 5 vs Day 6 0.680 0.016* 0.774 0.004* 0.805 <0.001*
Day 6 vs Day 7 0.643 0.017* 0.622 a) 0.044* 0.802 b) <0.001*
Open in a new tab

a) n = 12

b) n = 25

Figure 3.

Figure 3

Open in a new tab

Relationship in daily percentage of time spent (a) swimming and (b) hovering, and in daily (c) median number of tailbeats (‘MTBF’), (d) percentage of high tailbeat frequency (‘HTB%’) and (e) logarithmized total number of tailbeats (‘TTB’), between the second and seventh day post-tagging, for the Patagonian grouper, discriminated by season. Open red circles: hot season; open blue circles: cold season. A bisector (dotted) line is included as reference in each plot.

Table 2.

Results of the Spearman correlations (between Days 2 and 7 post-tagging) for five dependent variables used to evaluate repeatability, based on acceleration data collected from 26 free-living Patagonian groupers tagged in the Golfo Nuevo, Península Valdés. The asterisks show significant correlations (P < 0.05)

Dependent variable Warm season Cold season Both seasons pooled
rho (n = 13) P rho (n = 13) P rho (n = 26) P
Swimming −0.220 0.235 0.467 0.055 0.335 0.047*
Hovering 0.533 0.032* 0.632 0.012* 0.707 <0.001*
MTBF 0.502 0.040* 0.803 <0.001* 0.667 <0.001*
HTB% 0.527 0.032* 0.762 a) 0.002* 0.659 b) <0.001*
TTB 0.368 0.120 0.352 a) 0.120 0.326 b) 0.056
Open in a new tab

a) n = 12

b) n = 25

The linear mixed-effects models for the variables ‘Swimming’, ‘Hovering’, ‘MTBF’, ‘HBT%’ and ‘TTB’ met normality and homoscedasticity assumptions (results not showed). In all cases, the models that incorporated TL as a covariate showed ΔAIC values between 1.7 and 1.9, with respect to the same models without TL, so that for parsimony, we finally chose the models that did not include the TL covariate. In the selected models for ‘Swimming’, ‘Hovering’, ‘MTB’ and ‘HTB%’, the interaction term ‘Season × Day’ did not reduce the residual deviance significantly so it was not retained. However, in the model for ‘TTB’, the interaction reduced the residual deviance significantly (ΔAIC = 6.17) and hence it was included, implying that the effect of the season varied through the days since tagging (Fig. 3). The percentage of time spent swimming or hovering was higher during the warm months: they increased 83.7% (P < 0.001) and 59.4% (P < 0.05), respectively, with respect to the cold period (Fig. 4). Conversely, the HTB% (>4 Hz) was 22.4% lower during the warm season (P < 0.05). There were no differences in the median number of tailbeats between seasons (P = 0.07); neither did we detect any effects of the day on any of the four variables: ‘Swimming’, ‘Hovering’, ‘MTBF’ or ‘HTB%’. The analysis of repeatability based on Inline graphic showed that it was poor for ‘Swimming’ and ‘HTB%’ and moderate for the remaining three activity variables studied: ‘Hovering’, ‘MTBF’ and ‘TTB’ (Table 3).

Figure 4.

Figure 4

Open in a new tab

Data distribution for the five dependent variables studied: daily percentage of time spent (a) swimming and (b) hovering, and daily (c) median number of tailbeats (‘MTBF’), (d) percentage of high tailbeat frequency (‘HTB%’) and (e) logarithmized total numbers of tailbeats (‘TTB’), for the Patagonian grouper, discriminated by day and season.

Table 3.

Adjusted ICCs (Inline graphic estimates) and their corresponding 95% confidence intervals (95% CI) for five dependent variables studied, based on acceleration data from 26 free-living Patagonian groupers tagged in the Golfo Nuevo, Peninsula Valdés

Dependent variable Inline graphic estimate 95% CI
Swimming 0.403 [0.214–0.580]
Hovering 0.657 [0.473–0.787]
MTBF 0.547 [0.345–0.697]
HTB% 0.448 [0.261–0.640]
TTB 0.573 [0.388–0.727]
Open in a new tab

Discussion

Our results showed that most activity variables considered are repeatable in the short term (up to 1 week). We did not find large differences in the day-to-day repeatability between seasons. For each season, most variables showed significance in all, or in all but one or two, of the day-to-day pairings. The only exceptions to this general pattern were ‘HTB%’ and ‘Swimming’, for which most day-to-day pairings were weakly correlated during the cold and the warm seasons, respectively. When both seasons were pooled together, day-to-day pairings were significantly correlated for all the variables, possibly due to a greater statistical power resulting from the larger amount of data. The possibility that trait repeatability is maintained across different temperature regimes (i.e. cold vs warm seasons), beyond an increase in the average activity level during the warm season (particularly evidenced by the variables ‘Swimming’ and ‘Hovering’), is consistent with previous results by Claireaux et al. (2007), who found that sprint performance was repeatable after acclimatizing European seabass (Dicentrarchus labrax) at different temperatures (12 and 22°C), and with those by Hanson et al. (2010), who reported that both daily distance travelled and mean daily swimming speed in the largemouth bass (M. salmoides) were repeatable during most seasons. However, repeatability based on Spearman correlations on a week interval was higher in the cold season for four of the variables, ‘TTB’ being the only exception to this pattern, due to increasing values with days post-tagging. It is therefore possible that a general lower level of activity during the cold season diminishes the variability in activity, which may cause an increase in repeatability. Amongst the measured variables, ‘Hovering’, ‘MTBF’ and ‘TTB’ were the ones with the highest repeatability based on the ICC analysis. The repeatability of TTB suggests that the activity level (Body Caudal Fin locomotion, BCF sensuWebb, 1984) is individual-specific.

Whilst swimming in this species is mainly supported by the tail, hovering may also contribute importantly to oxygen consumption, as it was shown for a labriform fish (Tudorache et al., 2009), and therefore to the overall energy budget. Individuals with a high hovering score may therefore have a higher daily energetic demand. The Patagonian grouper is a sedentary species that spends large periods in crevices to avoid predators, and it is also considered that this behaviour may reduce the costs of buoyancy control and posture maintenance (Ciancio et al., 2016). Individuals that spend longer periods outside their crevices, hovering in the water column, may be more vulnerable to predation or fishing, which would lead to a selective removal from the reef (Lewin et al., 2006). Therefore, the assessment of hovering repeatability may be valuable from a conservation viewpoint, as it would allow to predict which individuals are more likely to be caught repeatedly with potential effects on their overall survivorship. The selective removal of individuals with higher hovering scores in reefs subject to high fishing effort might modify the average energetic demand of fish at the local (i.e. reef) scale.

The high repeatability of ‘MTBF’ implies that individuals that swim faster than others did so day after day. Interestingly, ‘Swimming’ (which is a proxy for general activity per day) was less repeatable across the week period and it yielded the poorest correlation for both seasons pooled together, as well as the lowest ICCadj amongst the variables tested, followed closely by ‘HTB%’ (both having an ICCadj value slightly <0.5). This suggests that ‘MTBF’ is decoupled from the percentage of time spent in swimming. The HTB% > 4 Hz can be interpreted as the use of swimming speeds higher than sustainable, such as those commonly used in predator–prey interactions (Domenici and Blake, 1997). The repeatability of ‘MTBF’ and specifically that of ‘HTB%’ (though significant only for seasons pooled) suggests that certain individuals consistently perform more predatory attack or avoidance manoeuvres, as it is known that, e.g. high likelihood to respond to threat may be a personality trait (Jones and Godin, 2010). The Argentine sea bass tends to feed mainly on benthic prey that are not highly mobile such as polychaetes and crabs (Goldstein and Cousseau, 1987; Galván et al., 2009), although they also opportunistically feed on pelagic fish such as anchovy (Engraulis anchoita). Therefore, their high-speed swimming activity recorded could be related to anti-predator behaviour but also, at least occasionally, to feeding strikes aimed at catching manoeuvrable fish in the water column.

The high repeatability found in the very short term (day-to-day) is in line with previous work on a freshwater species (Largemouth bass M. salmoides) based on activity level measured through whole-lake acoustic telemetry array with sub-metre accuracy (Hanson et al., 2010). These authors found that the daily distance travelled and mean daily swimming speed was repeatable day-to-day within each season. A recent work also found repeatability in short (daily) and medium (monthly) time periods of locomotion kinematics and swimming performance variables for a Gymnotiform fish (Oufiero et al., 2021). The repeatability of a similar trait (median swimming speed) has also been proved during the early stages of a clonal freshwater fish, Poecilia formosa, in periods up to 10 weeks (Laskowski et al., 2022).

Accelerometers have proved to be useful tools to describe and quantify different aspects of animal behavioural ecology, with studies focused on the occurrence and intensity of behaviour, movement characteristics, body posture and energy expenditure (e.g. Ciancio et al., 2016; Wilson et al., 2020; Garde et al., 2022; Lennox et al., 2023). A few studies also focused on circadian rhythms (Alós et al., 2017) and foraging patterns (Sala et al., 2015), or evaluated how the behaviour of aquatic animals changes over time in response to changing environments (Stieglitz and Dujon, 2017). Admittedly, the primary current limitation of the present work is the restricted time frame, which is limited to a week interval. Previous work suggests that experiments performed at short time intervals tend to show higher repeatability than behaviours recorded at longer intervals (Bell et al., 2009; Hanson et al., 2010). Overall, telemetry studies on fishes strongly suggest the presence of repeatable activity patterns over relatively long periods in a number of species, although context-specific performances may change with differing environmental conditions across seasons and years (Hanson et al., 2010). Therefore, it would be interesting to test if the activity traits that can be measured using accelerometry and are related to specific swimming patterns (e.g. hovering, swimming), show some degree of long-term (e.g. across seasons and years) repeatability. Although the need for long-term investigations based on accelerometry is challenging because of some technical issues such as the memory available and the size of the tag, new accelerometry tags currently under development will allow longer recording intervals in the near future (e.g. Lennox et al., 2023).

Compared to previous work based on other methodologies (e.g. ultrasonic telemetry, mark-recapture, observation, video recording, hormone titration, etc.), which have focused on studying the repeatability of activity levels, spatial pattern, boldness and stress levels (e.g. Cook et al., 2011; Harrison et al., 2015; White et al., 2015; Villegas-Ríos et al., 2017), the use of accelerometry can widen the spectrum of activities that can be monitored. Here, for example, we found that ‘Hovering’, an activity that cannot be detected via telemetry because it does not imply displacement, is repeatable within the short term (week). Similarly, accelerometry allows different swimming activities with relatively high resolution, such as low- versus high-speed swimming activity, to be detected, thus revealing how often fish are likely to be engaged in escape responses, mating or predator–prey-type interactions. Further work could investigate the possibility of partitioning burst-swimming activity into feeding versus anti-predator behaviour, as in laboratory studies (Broell et al., 2013). On the other hand, certain fish traits such as home range and depth use are not detected by accelerometers, though they are detected by other methods such as acoustic telemetry. Therefore, the use of different techniques combined (e.g. Kawabata et al., 2014; Meese and Lowe, 2020) is a promising avenue for monitoring the behaviour, the spatial patterns and the activity levels of fish, thereby providing a useful tool for assessing the inter-individual variability of fish populations. Maintaining inter-individual variation in movement patterns is important for a population’s ability to respond to future environmental change (Killen et al., 2016). Hence, further research using accelerometry for measuring both the repeatability of behavioural traits and the effect of season and temperature on activity levels would be fundamental for assessing the resilience of fish populations to environmental disturbances, which can be considered a major advancement in the field of conservation physiology.

Supplementary Material

Web_Material_coae074
web_material_coae074.zip (96.7KB, zip)

Acknowledgements

We thank A. Biasotti, S. Fernández, A. Lhose, M. Matos, N. Ortíz, M. Piunca, D. Pollicelli, M. Pollicelli and F. Quiroga for their help during fieldwork. Tech. M. Moris helped us with the aquarium work. The Área Náutica de la Universidad de la Patagonia San Juan Bosco provided logistic support.

Contributor Information

Leonardo A Venerus, Centro para el Estudio de Sistemas Marinos, Consejo Nacional de Investigaciones Científicas y Técnicas (CESIMAR, CCT CONICET-CENPAT), Blvd. Brown 2915, U9120ACD Puerto Madryn, Chubut, Argentina.

Paolo Domenici, Consiglio Nazionale delle Ricerche, Istituto per lo Studio Degli Impatti Antropici e Sostenibilità in Ambiente Marino (CNR-IAS), Località Sa Mardini, 09070 Torregrande, Oristano, Italy; Consiglio Nazionale delle Ricerche, Istituto di Biofisica (CNR-IBF), Area di Ricerca San Cataldo, Via G. Moruzzi 1, 56124 Pisa, Italy; National Biodiversity Future Center (NBFC), Piazza Marina 61, 90133 Palermo, Italy.

Stefano Marras, Consiglio Nazionale delle Ricerche, Istituto per lo Studio Degli Impatti Antropici e Sostenibilità in Ambiente Marino (CNR-IAS), Località Sa Mardini, 09070 Torregrande, Oristano, Italy; National Biodiversity Future Center (NBFC), Piazza Marina 61, 90133 Palermo, Italy.

Lucas E Beltramino, Universidad Nacional de la Patagonia San Juan Bosco (UNPSJB), Blvd. Brown 3051, U9120ACE Puerto Madryn, Chubut, Argentina.

Javier E Ciancio, Centro para el Estudio de Sistemas Marinos, Consejo Nacional de Investigaciones Científicas y Técnicas (CESIMAR, CCT CONICET-CENPAT), Blvd. Brown 2915, U9120ACD Puerto Madryn, Chubut, Argentina.

Author contributions

L.A.V., P.D., S.M., L.E.B. and J.E.C. developed the research questions and experimental design. L.A.V., L.E.B. and J.E.C. conducted fieldwork. L.A.V., P.D., S.M., L.E.B. and J.E.C. developed the statistical analysis. L.A.V. and P.D. led the writing of the manuscript. S.M., L.E.B. and J.E.C. provided revisions.

Conflicts of interest

The authors have no conflicts of interest to declare.

Funding

This work was supported by PADI Foundation [grant number 21720 to J.E.C.]; the Agencia Nacional de Promoción Científica y Tecnológica [grant number PICT 2015–3340 to J.E.C.]; the Youth Activity Fund (The Explorers Club) [grant number NA to L.E.B.]; the Secretaría de Ciencia, Tecnología e Innovación Productiva del Chubut [grant number NA to L.E.B.] and Fitfish Cost Action [grant number FA 1304 to P.D.].

Data availability

The data underlying this article will be shared on reasonable request to the corresponding author.

Supplementary material

Supplementary material is available at Conservation Physiology online.

References

  1. Akaike H (1973) Information theory and an extension of the maximum likelihood principle. In Petrov BN, Caski F, eds, Proceedings of the Second International Symposium on Information Theory. Akademiai Kiado, Budapest, pp. 267–281. [Google Scholar]
  2. Alós J, Martorell-Barceló M, Campos-Canela A (2017) Repeatibility of circadian behavioural variation revealed in free-ranging marine fish. R Soc Open Sci 4: 60791. 10.1098/rsos.160791. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bates D, Maechler M, Bolker B, Walker S (2015) Fitting linear mixed-effects models using lme4. J Stat Softw 67: 1–48. 10.18637/jss.v067.i01. [DOI] [Google Scholar]
  4. Becker WA (1984) Manual of Quantitative Genetics, Ed 4th. Academic Enterprises, Washington, USA, p. 190. [Google Scholar]
  5. Bell AM, Hankison SJ, Laskowski KL (2009) The repeatability of behaviour: a meta-analysis. Anim Behav 77: 771–783. 10.1016/j.anbehav.2008.12.022. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Beltramino LE, Venerus LA, Trobbiani GA, Wilson RP, Ciancio JE (2019) Activity budgets for the sedentary Argentine sea bass Acanthistius patachonicus inferred from accelerometer data loggers. Austral Ecol 44: 397–408. 10.1111/aec.12696. [DOI] [Google Scholar]
  7. Bidder OR, Campbell HA, Gómez-Laich A, Urgé P, Walker J, Cai Y, Gao L, Quintana F, Wilson RP (2014) Love thy neighbour: automatic animal behavioural classification of acceleration data using the K nearest neighbour algorithm. PloS One 9: e88609. 10.1371/journal.pone.0088609. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Biro PA, Stamps JA (2015) Using repeatability to study physiological and behavioural traits: ignore time-related change at your peril. Anim Behav 105: 223–230. 10.1016/j.anbehav.2015.04.008. [DOI] [Google Scholar]
  9. Broell F, Noda T, Wright S, Domenici P, Steffensen JF, Auclair JP, Taggart CT (2013) Accelerometer tags: detecting and identifying activities in fish and the effect of sampling frequency. J Exp Biol 216: 1255–1264. 10.1242/jeb.077396. [DOI] [PubMed] [Google Scholar]
  10. Brownscombe JW, Gutowsky LFG, Danylchuk AJ, Cooke SJ (2014) Foraging behaviour and activity of a marine benthivorous fish estimated using tri-axial accelerometer biologgers. Mar Ecol Prog Ser 505: 241–251. 10.3354/meps10786. [DOI] [Google Scholar]
  11. Carbonara P, Alfonso S, Dioguardi M, Zupa W, Vazzana M, Dara M, Spedicato MT, Lembo G, Cammarata M (2021) Calibrating accelerometer data, as a promising tool for health and welfare monitoring in aquaculture: case study in European sea bass (Dicentrarchus labrax) in conventional or organic aquaculture. Aquac Rep 21: 100817. 10.1016/j.aqrep.2021.100817. [DOI] [Google Scholar]
  12. Ciancio JA, Venerus LA, Trobbiani GA, Gleiss A, Wright S, Beltramino LE, Norman B, Holton M, Wilson RP (2016) Extreme roll angles in Argentine sea bass: could refuge ease posture and buoyancy control of marine coastal fishes? Mar Biol 163: 90. 10.1007/s00227-016-2869-z. [DOI] [Google Scholar]
  13. Claireaux G, Handelsman C, Standen E, Nelson JA (2007) Thermal and temporal stability of swimming performance in the European sea bass. Physiol Biochem Zool 80: 186–196. 10.1086/511143. [DOI] [PubMed] [Google Scholar]
  14. Cook KV, O'Connor CM, Gilmour KM, Cooke SJ (2011) The glucocorticoid stress response is repeatable between years in a wild teleost fish. J Comp Physiol A 197: 1189–1196. 10.1007/s00359-011-0680-3. [DOI] [PubMed] [Google Scholar]
  15. Dellatorre FG, Pisoni JP, Barón PJ, Rivas AL (2012) Tide and wind forced nearshore dynamics in Nuevo Gulf (Northern Patagonia, Argentina): potential implications for cross-shore transport. J Mar Syst 96: 82–89. 10.1016/j.jmarsys.2012.02.004. [DOI] [Google Scholar]
  16. DeSantis DL, Mata-Silva V, Johnson JD, Wagler AE (2020) Integrative framework for long-term activity monitoring of small and secretive animals: validation with a cryptic pitviper. Front Ecol Evol 8: 169. 10.3389/fevo.2020.00169. [DOI] [Google Scholar]
  17. Dochtermann NA, Schwab T, Sih A (2015) The contribution of additive genetic variation to personality variation: heritability of personality. Proc R Soc B Biol Sci 282: 20142201. 10.1098/rspb.2014.2201. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Domenici P, Blake R (1997) The kinematics and performance of fish fast-start swimming. J Exp Biol 200: 1165–1178. 10.1242/jeb.200.8.1165. [DOI] [PubMed] [Google Scholar]
  19. Forsythe PS, Crossman JA, Bello NM, Baker EA, Scribner KT (2012) Individual-based analyses reveal high repeatability in timing and location of reproduction in lake sturgeon (Acipense fulvescens). Can J Fish Aquat Sci 69: 60–72. 10.1139/f2011-132. [DOI] [Google Scholar]
  20. Galván DE, Botto F, Parma AM, Bandieri L, Mohamed N, Iribarne OO (2009) Food partitioning and spatial subsidy in shelter-limited fishes inhabiting patchy reefs of Patagonia. J Fish Biol 75: 2585–2605. 10.1111/j.1095-8649.2009.02453.x. [DOI] [PubMed] [Google Scholar]
  21. Garde B, Wilson RP, Fell A, Cole N, Tatayah V, Holton MD, Rose KAR, Metcalfe RS, Robotka H, Wikelski Met al. (2022) Ecological inference using data from accelerometer needs careful protocols. Methods Ecol Evol 13: 813–825. 10.1111/2041-210X.13804. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Gleiss AC, Schallert RJ, Dale JJ, Wilson SG, Block BA (2019) Direct measurement of swimming and diving kinematics of giant Atlantic bluefin tuna (Thunnus thynnus). R Soc Open Sci 6: 190203. 10.1098/rsos.190203. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Gleiss AC, Wright S, Liebsch N, Wilson RP, Norman B (2013) Contrasting diel patterns in vertical movement and locomotor activity of whale sharks at Ningaloo Reef. Mar Biol 160: 2981–2992. 10.1007/s00227-013-2288-3. [DOI] [Google Scholar]
  24. Goldstein HE, Cousseau MB (1987) Estudios sobre el regimen alimentario del mero (Acanthistius brasilianus) y su relación con las características morfométricas del sistema digestivo (Pisces, fam. Serranidae). Rev Invest Desarro Pesq 7: 85–103. [Google Scholar]
  25. Gottlieb DH, Capitanio JP, McCowan B (2013) Risk factors for stereotypic behaviour and self-biting in rhesus macaques (Macaca mulatta): animal’s history, current environment, and personality. Am J Primatol 75: 995–1008. 10.1002/ajp.22161. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Grémillet D, Lescroël A, Ballard G, Dugger KM, Massaro M, Porzig EL, Ainley DG (2018) Energetic fitness: field metabolic rates assessed via 3D accelerometry complement conventional fitness metrics. Funct Ecol 32: 1203–1213. 10.1111/1365-2435.13074. [DOI] [Google Scholar]
  27. Griffin LP, Brownscombe JW, Adams AJ, Wilson JK, Casselberry GA, Holder PE, Filous A, Lowerre-Barbieri SK, Cooke SJ, Danylchuk AJ (2023) Individual variation and repeatability of Atlantic tarpon Megalops atlanticus migrations in the southern US: implications for conservation and management. Mar Biol 170: 168. 10.1007/s00227-023-04311-3. [DOI] [Google Scholar]
  28. Handelsman CA, Claireaux G, Nelson JA (2010) Swimming ability and ecological performance of cultured and wild European sea bass in coastal tidal ponds. Physiol Biochem Zool 83: 435–445. 10.1086/651099. [DOI] [PubMed] [Google Scholar]
  29. Hanson KC, Hasler CT, Donaldson MR, Cooke SJ (2010) Stability of swimming performance and activity hierarchies among wild largemouth bass at multiple temporal scales: evidence for context-dependent shuffling between seasons. Can J Zool 88: 324–333. 10.1139/Z10-006. [DOI] [Google Scholar]
  30. Harrison PM, Gutowsky LFG, Martins EG, Patterson DA, Cooke SJ, Power M (2015) Personality-dependent spatial ecology occurs independently from dispersal in wild burbot (Lota lota). Behav Ecol 26: 483–492. 10.1093/beheco/aru216. [DOI] [Google Scholar]
  31. Hartig F (2022) DHARMa: Residual Diagnostics for Hierarchical (Multi-level/mixed) Regression Models. https://cran.r-project.org/web/packages/DHARMa/vignettes/DHARMa.html (date last accessed 7 September 2024).
  32. Herskin J, Steffensen JF (1998) Energy savings in sea bass swimming in a school: measurements of tail beat frequency and oxygen consumption at different swimming speeds. J Fish Biol 53: 366–376. 10.1111/j.1095-8649.1998.tb00986.x. [DOI] [Google Scholar]
  33. Hitchcock AC, Chen T, Connolly E, Darakananda K, Jeong J, Quist A, Robbins A, Ellerby DJ (2015) Trade-offs between performance and variability in the escape responses of bluegill sunfish (Lepomis macrochirus). Biol Open 4: 743–751. 10.1242/bio.201511577. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Holm S (1979) A simple sequentially rejective multiple test procedure. Scand J Stats 6: 65–70. [Google Scholar]
  35. Howell DC (2009) Repeated-measures designs. In Statistical Methods for Psychology, Ed7th. Wadsworth CENGAGE Learning, Belmont, USA, pp. 461–514 [Google Scholar]
  36. Irigoyen AJ (2010) Efecto del Alga Invasora Undaria pinnatifida Sobre la Comunidad de Peces de Arrecife en los Golfos Norpatagónicos Doctoral Thesis,. Universidad Nacional del Comahue, Río Negro, Argentina, 157 p [Google Scholar]
  37. Irigoyen AJ, Galván DE, Venerus LA, Parma AM (2013) Variability in abundance of temperate reef fishes estimated by visual census. PloS One 8: e61072. 10.1371/journal.pone.0061072. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Irigoyen AJ, Venerus LA (2008) The 'pole-hooking' method: a novel and echonomical technique for in situ tagging small to medium-sized fishes. Fish Res 91: 349–353. 10.1016/j.fishres.2007.12.015. [DOI] [Google Scholar]
  39. Jones KA, Godin J-GG (2010) Are fast explorers slow reactors? Linking personality type and anti-predator behaviour. Proc Royal Soc B 277: 625–632. 10.1098/rspb.2009.1607. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Kawabata Y, Noda T, Nakashima Y, Nanami A, Sato T, Takebe T, Mitamura H, Arai N, Yamaguchi T, Soyano K (2014) Use of a gyroscope/accelerometer data logger to identify alternative feeding behaviours in fish. J Exp Biol 217: 3204–3208. 10.1242/jeb.108001. [DOI] [PubMed] [Google Scholar]
  41. Keegan GG, Kramer J, Yonezawa Y, Maki H, Pai F, Dent EV, Kellerman TE, Wilson DA, Reed SK (2011) Assessment of repeatability of a wireless, inertial sensor-based lameness evaluation system for horses. Am J Vet Res 72: 1156–1163. 10.2460/ajvr.72.9.1156. [DOI] [PubMed] [Google Scholar]
  42. Killen SS, Adrianssesns B, Marras S, Claireaux G, Cooke SJ (2016) Context dependency of trait repeatability and its relevance for management and conservation of fish populations. Conserv Physiol 4: cow007. 10.1093/conphys/cow007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Killen SS, Marras S, Metcalfe NB, McKenzie DJ, Domenici P (2013) Environmental stressors alter relationships between physiology and behaviour. Trends Ecol Evol 11: 651–658. 10.1016/j.tree.2013.05.005. [DOI] [PubMed] [Google Scholar]
  44. Kolok AS, Plaisance EPP, Abdelghani A (1998) Individual variation in the swimming performance of fishes: an overlooked source of variation in toxicity studies. Environ Toxicol Chem 17: 282–285. 10.1002/etc.5620170221. [DOI] [Google Scholar]
  45. Laskowski KL, Bierbach D, Jolles JW, Doran C, Wolf M (2022) The emergence and development of behavioral individuality in clonal fish. Nat Commun 13: 6419. 10.1038/s41467-022-34113-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Lennox RJ, Eldøy SH, Dahlmo LS, Matley JK, Vollset KW (2023) Acoustic accelerometer transmitters and their growing relevance to aquatic sciences. Mov Ecol 11: 45. 10.1186/s40462-023-00403-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Lessells CM, Boag PT (1987) Unrepeatable repeatabilities: a common mistake. Auk 104: 116–121. 10.2307/4087240. [DOI] [Google Scholar]
  48. Lewin W-C, Arlinghaus R, Mehner T (2006) Documented and potential biological impacts of recreational fishing: insights for management and conservation. Rev Fish Sci 14: 305–367. 10.1080/10641260600886455. [DOI] [Google Scholar]
  49. Liljequist D, Elfving B, Roaldsen KS (2019) Intraclass correlation– a discussion and demonstration of basic features. PloS One 14: e0219854. 10.1371/journal.pone.0219854. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. MacKay JRD, Haskell MJ, Deag JM, Reenen K (2014) Fear responses to novelty in testing environments are related to day-to-day activity in the home environment in dairy cattle. Appl Anim Behav Sci 152: 7–16. 10.1016/j.applanim.2013.12.008. [DOI] [Google Scholar]
  51. Marras S, Claireaux G, McKenzie DJ, Nelson JA (2010) Individual variation and repeatability in aerobic and anaerobic swimming performance of European sea bass, Dicentrarchus labrax. J Exp Biol 213: 26–32. 10.1242/jeb.032136. [DOI] [PubMed] [Google Scholar]
  52. Marras S, Killen SS, Claireaux G, Domenici P, McKenzie DJ (2011) Behavioural and kinematic components of the fast-start escape response in fish: individual variation and temporal repeatability. J Exp Biol 214: 3102–3110. 10.1242/jeb.056648. [DOI] [PubMed] [Google Scholar]
  53. Marras S, Noda T, Steffensen JF, Svendsen MBS, Krause J, Wilson ADM, Kurvers RHJM, Herbert-Read J, Boswell KM, Domenici P (2015) Not so fast: swimming behavior of sailfish during predator–prey interactions using high-speed video and accelerometry. Integr Comp Biol 55: 719–727. 10.1093/icb/icv017. [DOI] [PubMed] [Google Scholar]
  54. Meese EN, Lowe CG (2020) Active acoustic telemetry tracking and tri-axial accelerometers reveal fine-scale movement strategies of a non-obligate ram ventilator. Mov Ecol 8: 8–17. 10.1186/s40462-020-0191-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Murchie KJ, Cooke SJ, Danylchuk AJ, Suski CD (2011) Estimates of field activity and metabolic rates of bonefish (Albula vulpes) in coastal marine habitats using acoustic tri-axial accelerometer transmitters and intermittent-flow respirometry. J Exp Mar Biol Ecol 396: 147–155. 10.1016/j.jembe.2010.10.019. [DOI] [Google Scholar]
  56. Nakagawa S, Johnson PCD, Schielzeth H (2017) The coefficient of determination R2 and intra-class correlation coefficient from generalized linear mixed-effects models revisited and expanded. J R Soc Interface 14: 20170213. 10.1098/rsif.2017.0213. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Nakagawa S, Schielzeth H (2010) Repeatability for Gaussian and non-Gaussian data: a practical guide for biologists. Biol Rev 85: 935–956. 10.1111/j.1469-185X.2010.00141.x. [DOI] [PubMed] [Google Scholar]
  58. Nelson JA, Claireaux G (2005) Sprint swimming performance of juvenile European sea bass. Trans Am Fish Soc 134: 1274–1284. 10.1577/T04-087.1. [DOI] [Google Scholar]
  59. Nespolo RF, Franco M (2007) Whole-animal metabolic rate is a repeatable trait: a meta-analysis. J Exp Biol 210: 2000–2005. 10.1242/jeb.02780. [DOI] [PubMed] [Google Scholar]
  60. Ohlberger J, Staaks G, Holker F (2007) Estimating the active metabolic rate (AMR) in fish based on tail beat frequency (TBF) and body mass. J Exp Zool Part A Ecol Genet Physiol 307: 296–300. 10.1002/jez.384. [DOI] [PubMed] [Google Scholar]
  61. Oufiero C, Garland T (2009) Repeatability and correlation of swimming performances and size over varying time-scales in the guppy (Poecilia reticulata). Funct Ecol 23: 969–978. 10.1111/j.1365-2435.2009.01571.x. [DOI] [Google Scholar]
  62. Oufiero C, Kraskura K, Bennington R, Nelson J (2021) Individual repeatability of locomotor kinematics and swimming performance in a gymnotiform swimmer. Physiol Biochem Zool 94: 22–34. 10.1086/712058. [DOI] [PubMed] [Google Scholar]
  63. Papastamatiou YP, Watanabe YY, Demŝar U, Leos-Barajas V, Bradley D, Langrock R, Weng K, Lowe CG, Friedlander AM, Caselle JE (2018) Activity seascapes highlight central place foraging strategies in marine predators that never stop swimming. Mov Ecol 6: 9. 10.1186/s40462-018-0127-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Pereñíguez JM, Venerus LA, Gutiérrez-Cánovas C, Abecasis D, Ciancio JE, Jiménez-Montalbán P, García-Charton JA (2022) Acoustic telemetry and accelerometers: a field comparison of different proxies for activity in the marine environment. ICES J Mar Sci 79: 2600–2613. 10.1093/icesjms/fsac190. [DOI] [Google Scholar]
  65. R Core Team (2020) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria, https://www.R-project.org/ (date last accessed 7 September 2023). [Google Scholar]
  66. Rotics S, Turjeman S, Kaatz M, Zurell D, Wikelski M, Sapir N, Fiedler W, Eggers U, Resheff YS, Jeltsch Fet al. (2021) Early-life behaviour predicts first-year survival in a long-distance avian migrant. Proc R Soc B 288: 20202670. 10.1098/rspb.2020.2670. [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. Rudd JL, Exeter OM, Hall J, Hall G, Henderson SM, Kerry C, Witt MJ, Hawkes LA (2021) High resolution biologging of breaching by the world's second largest shark species. Sci Rep 11: 5236. 10.1038/s41598-021-84670-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Ryan JA, Ulrich JM (2023) xts: eXtensible Time Series. R package version 0.13.0. https://CRAN.R-project.org/package=xts (date last accessed 7 September 2023).
  69. Sala JE, Quintana F, Wilson R (2015) Foraging effort in Magellanic penguins: balancing the energy books for survival. Mar Biol 162: 501–514. 10.1007/s00227-014-2581-9. [DOI] [Google Scholar]
  70. Shepard ELC, Wilson RP, Halsey LG, Quintana F, Laich AG, Gleiss AC, Liebsch N, Myers AE, Norman B (2008) Derivation of body motion via appropriate smoothing of acceleration data. Aquat Biol 4: 235–241. 10.3354/ab00104. [DOI] [Google Scholar]
  71. Steinhausen MF, Steffensen JF, Andersen NG (2005) Tail beat frequency as a predictor of swimming speed and oxygen consumption of saithe (Pollachius virens) and whiting (Merlangius merlangus) during forced swimming. Mar Biol 148: 197–204. 10.1007/s00227-005-0055-9. [DOI] [Google Scholar]
  72. Stiegler J, Lins A, Dammhahn M, Kramer-Schadt S, Ortmann S, Blaum N (2022) Personality drives activity and space use in a mammalian herbivore. Mov Ecol 10: 33. 10.1186/s40462-022-00333-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  73. Stieglitz TC, Dujon AM (2017) A groundwater-fed coastal inlet as habitat for the Caribbean queen conch Lobatus gigas—an acoustic telemetry and space use analysis. Mar Ecol Prog Ser 571: 139–152. 10.3354/meps12123. [DOI] [Google Scholar]
  74. Stoffel MA, Nakagawa S, Schielzeth H (2017) rptR: repeatability estimation and variance decomposition by generalized linear mixed-effects models. Methods Ecol Evol 8: 1639–1644. 10.1111/2041-210X.12797. [DOI] [Google Scholar]
  75. Sueur J, Aubin T, Simonis C (2008) Seewave: a free modular tool for sound analysis and synthesis. Bioacoustics 18: 213–226. 10.1080/09524622.2008.9753600. [DOI] [Google Scholar]
  76. Taylor MK, Cooke SJ (2014) Repeatibility of movement behaviour in a wild salmonid revealed by telemetry. J Fish Biol 84: 1240–1246. 10.1111/jfb.12334. [DOI] [PubMed] [Google Scholar]
  77. Tudorache C, Jordan AD, Svendsen JC, Domenici P, DeBoeck G, Steffensen JF (2009) Pectoral fin beat frequency predicts oxygen consumption during spontaneous activity in a labriform swimming fish (Embiotoca lateralis). Environ Biol Fishes 84: 121–127. 10.1007/s10641-008-9395-x. [DOI] [Google Scholar]
  78. Tuszynski J (2021) caTools: Tools: Moving Window Statistics, GIF, Base64, ROC AUC, etc. R package version 1.18.2. https://CRAN.R-project.org/package=caTools (date last accessed 7 September 2023).
  79. Ulrich J (2021) TTR: Technical Trading Rules. R package version 0.24.3. https://CRAN.R-project.org/package=TTR (date last accessed 7 September 2023).
  80. Vandamm JP, Marras S, Claireaux G, Handelsman CA, Nelson JA (2012) Acceleration performance of individual European sea bass, Dicentrarchus labrax measured with a sprint performance chamber: comparison with high-speed cinematography and correlates with ecological performance. Physiol Biochem Zool 85: 704–717. 10.1086/666463. [DOI] [PubMed] [Google Scholar]
  81. Vázquez Diosdado JA, Barker ZE, Hodges HR, Amory JR, Croft DP, Bell NJ, Codling EA (2015) Classification of behaviour in housed dairy cows using an accelerometer-based activity monitoring system. Anim Biotelemetry 3: 15. 10.1186/s40317-015-0045-8. [DOI] [Google Scholar]
  82. Venables WN, Ripley BD (2002) Modern Applied Statistics with S, Ed 4th. Springer, New York, USA, p. 497. [Google Scholar]
  83. Videler JJ (1993) Fish Swimming. Chapman & Hall, London, UK, p. 260 [Google Scholar]
  84. Villegas-Ríos D, Réale D, Freitas C, Moland E, Olsen EM (2017) Individual level consistency and correlations of fish spatial behaviour assessed from aquatic animal telemetry. Anim Behav 124: 83–94. 10.1016/j.anbehav.2016.12.002. [DOI] [Google Scholar]
  85. Webb PW (1978) Fast-start performance and body form in seven species of teleost fish. J Exp Biol 74: 211–226. [Google Scholar]
  86. Webb PW (1984) Body form, locomotion and foraging in aquatic vertebrates. Am Zool 24: 107–120. [Google Scholar]
  87. Webb PW (1994) Exercise performance of fish. Adv Vet Sci Comp Med 38B: 1–49. [PubMed] [Google Scholar]
  88. White JR, Meekan MG, McCormick MI (2015) Individual consistency in the behaviors of newly-settled reef fish. PeerJ 3: e961. 10.7717/peerj.961. [DOI] [PMC free article] [PubMed] [Google Scholar]
  89. Whitney NM, White CF, Gleiss AC, Schwieterman GD, Anderson P, Hueter RE, Skomal GB (2016) A novel method for determining post-release mortality, behavior, and recovery period using acceleration data loggers. Fish Res 183: 210–221. 10.1016/j.fishres.2016.06.003. [DOI] [Google Scholar]
  90. Williams TD (2008) Individual variation in endocrine systems: moving beyond the ‘tyranny of the golden mean’. Philos Trans R Soc Lond B Biol Sci 363: 1687–1698. 10.1098/rstb.2007.0003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  91. Wilson RP, Börger L, Holton MD, Scantlebury DM, Gómez-Laich A, Quintana F, Rosell F, Graf PM, Williams H, Gunner Ret al. (2020) Estimates for energy expenditure in free-living animals using acceleration proxies; a reappraisal. J Anim Ecol 89: 161–172. 10.1111/1365-2656.13040. [DOI] [PMC free article] [PubMed] [Google Scholar]
  92. Wilson RP, White CR, Quintana F, Halsey LG, Liebsch N, Martin GR, Butler PJ (2006) Moving towards acceleration for estimates of activity-specific metabolic rate in free-living animals: the case of the cormorant. J Anim Ecol 75: 1081–1090. 10.1111/j.1365-2656.2006.01127.x. [DOI] [PubMed] [Google Scholar]
  93. Wright S, Metcalfe JD, Hetherington S, Wilson R (2014) Estimating activity-specific energy expenditure in a teleost fish, using accelerometer loggers. Mar Ecol Prog Ser 496: 19–32. 10.3354/meps10528. [DOI] [Google Scholar]
  94. Zuur AF, Ieno EN, Walker NJ, Saveliev AA, Smith GM (2009) Mixed Effects Models and Extensions in Ecology with R. Springer, New York, USA, p. 574. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Web_Material_coae074
web_material_coae074.zip (96.7KB, zip)

Data Availability Statement

The data underlying this article will be shared on reasonable request to the corresponding author.

Articles from Conservation Physiology are provided here courtesy of Oxford University Press

RESOURCES