Search

Introduction: Diadromous fish populations have strongly declined over decades, and many species are protected through national and international regulations. They account for less than 1% of fish biodiversity worldwide, but they are among the most perceptible linkages between freshwater and marine ecosystems. During their migration back and forth, diadromous fish species are subjected to many anthropogenic threats, among which river damming can severely limit access to vital freshwater habitats and jeopardize population sustainability. Here, we developed a method based on a double-observer modeling approach for estimating the abundance of diadromous fish during their migration in rivers. Methods: The method relies on two independent and synchronous records of fish counts that were analyzed jointly thanks to a hierarchical Bayesian model to estimate detection efficiencies and daily fish passage. We used simulated data to test model robustness and identify conditions under which the developed approach can be used. The approach was then applied to empirical data to estimate the annual silver eel run in the Touques River, France. Results: The analysis of simulated datasets and the study case gives evidence that the model can provide robust,accurate, and precise estimates of detection probabilities and total fish abundance in a set of conditions dependent on the information provided in the data (annual distribution of fish passage, annual number of observation, pairing period, etc.). Discussion: Then, the method can be applied to various species and counting systems, including nomad acoustic camera devices. We discuss its relevance for programs on river continuity restoration, notably to quantify population restoration associated with dam removals.

La pêche à l’électricité est une méthode classiquement utilisée pour les inventaires de biodiversité des poissons et crustacés dans le cadre de la surveillance réglementaire de l’état écologique des cours d’eau. Elle pose un certain nombre de difficultés en territoires ultramarins tropicaux, de par certaines spécificités physiques (conductivité de l’eau basse à très basse) et écologiques (communautés mixtes de poissons et de crustacés, biodiversité forte à très forte) qui les caractérisent. Issues du partenariat OFB-INRAE, les recherches menées en collaboration avec l’entreprise HYDRECO-Guyane ont permis de définir des conditions spécifiques d’application de la pêche à l’électricité pour les petits cours d’eau (< 6 mètres de large) tropicaux. Grâce à des essais au laboratoire et in situ, en Guyane, aux Antilles et à La Réunion, nous avons identifié les réglages électriques les plus efficaces pour la capture, l’estimation de la richesse spécifique et de l’abondance des espèces et le respect de la santé animale. Ces recherches se placent dans l’objectif plus large d’améliorer la qualité des données bancarisées pour le suivi environnemental des masses d’eau. Elles donnent lieu à des recommandations pour la maitrise d’œuvre et la maitrise d’ouvrage de la pêche à l’électricité en petits cours d’eau tropicaux.

La pêche à l’électricité est une méthode classiquement utilisée pour les inventaires de biodiversité des poissons et crustacés dans le cadre de la surveillance réglementaire de l’état écologique des cours d’eau. Elle pose un certain nombre de difficultés en territoires ultramarins tropicaux, de par certaines spécificités physiques (conductivité de l’eau basse à très basse) et écologiques (communautés mixtes de poissons et de crustacés, biodiversité forte à très forte) qui les caractérisent. Issues du partenariat OFB-INRAE, les recherches menées en collaboration avec l’entreprise HYDRECO-Guyane ont permis de définir des conditions spécifiques d’application de la pêche à l’électricité pour les petits cours d’eau (< 6 mètres de large) tropicaux. Grâce à des essais au laboratoire et in situ, en Guyane, aux Antilles et à La Réunion, nous avons identifié les réglages électriques les plus efficaces pour la capture, l’estimation de la richesse spécifique et de l’abondance des espèces et le respect de la santé animale. Ces recherches se placent dans l’objectif plus large d’améliorer la qualité des données bancarisées pour le suivi environnemental des masses d’eau. Elles donnent lieu à des recommandations pour la maitrise d’œuvre et la maitrise d’ouvrage de la pêche à l’électricité en petits cours d’eau tropicaux.

Acoustic cameras, or imaging sonars, are high-potential devices for many applications in aquatic ecology, notably for fisheries management and population monitoring. However, how to extract such data into high-value information without a time-consuming entire data set reading by an operator is still a challenge. Moreover, the analysis of acoustic imaging, due to its low signal-to-noise ratio, is a perfect training ground for experimenting with new approaches, especially concerning deep learning techniques. We present hereby a novel approach that takes advantage of both convolutional neural network (CNN) and classical computer vision (CV) techniques, able to detect fish passages in acoustic video streams. The pipeline pre-treats the acoustic images to localise the signals of interest and to improve the detection performances. The YOLOv3-based model was trained with fish data from multiple species recorded by the two most frequently used models of acoustic cameras, the DIDSON and ARIS, including species of high ecological interest, as Atlantic salmon or European eels. The pre-treatment of images greatly improves the model performance, increasing its F1-score from 0.52 to 0.69. The model we developed provides satisfying results detecting almost 80% of fish passages and minimising the false-positive rate. On a validation data set, 40 h of videos and around 1 800 fish passages, the efficiency increases with the fish sizes, notably reaching a recall higher than 95% for Atlantic salmon. Conversely, the model appears much less efficient for eel detections on ARIS videos than on DIDSON data (31% recall vs 75%).

Estimation of abundance with wide spatiotemporal coverage is essential to the assessment and management of wild populations. But, in many cases, data available to estimate abundance time series have diverse forms, variable quality over space and time and they stem from multiple data collection procedures. We developed a hierarchical Bayesian modelling (HBM) approach that take full advantage of the diverse assemblage of data at hand to estimate homogeneous time series of abundances irrespective of the data collection procedure. We apply our approach to the estimation of adult abundances of 18 Atlantic salmon (Salmo salar) populations of Brittany (France) from 1987 to 2017 using catch statistics, environmental covariates, and fishing effort. Additional data of total or partial abundance collected in four closely monitored populations are also integrated into the analysis. The HBM framework allows the transfer of information from the closely monitored populations to the others. Our results reveal no clear trend in the abundance of adult returns in Brittany over the period studied. lations d'animaux sauvages. Dans de nombreux cas, toutefois, les donnees disponibles pour l'estimation de series chronologidifferentes procedures. Nous avons elabore une approche de modelisation bayesienne hierarchique (MBH) pour tirer pleinement parti de l'assemblage varie de donnees disponibles pour estimer des series chronologiques homogenes d'abondance, quelle que soit la procedure de collecte de donnees. Nous appliquons l'approche a l'estimation de l'abondance d'adultes dans 18 populations de saumon atlantique (Salmo salar) de Bretagne (France) de 1987 a 2017, en utilisant des statistiques sur les prises, des covariables faisant l'objet d'une surveillance etroite sont aussi integrees a l'analyse. Le cadre de MBH permet le transfert d'information l'abondance des adultes de retour en Bretagne au cours de la periode etudiee. [Traduit par la Redaction], L’estimation de l’abondance à vaste couverture spatiotemporelle est nécessaire pour l’évaluation et la gestion de populations d’animaux sauvages. Dans de nombreux cas, toutefois, les données disponibles pour l’estimation de séries chronologiques d’abondance prennent différentes formes, sont de qualité variable dans le temps et l’espace et ont été obtenues par différentes procédures. Nous avons élaboré une approche de modélisation bayésienne hiérarchique (MBH) pour tirer pleinement parti de l’assemblage varié de données disponibles pour estimer des séries chronologiques homogènes d’abondance, quelle que soit la procédure de collecte de données. Nous appliquons l’approche à l’estimation de l’abondance d’adultes dans 18 populations de saumon atlantique (Salmo salar) de Bretagne (France) de 1987 à 2017, en utilisant des statistiques sur les prises, des covariables environnementales et l’effort de pêche. D’autres données sur l’abondance totale ou partielle obtenues pour quatre populations faisant l’objet d’une surveillance étroite sont aussi intégrées à l’analyse. Le cadre de MBH permet le transfert d’information sur des populations étroitement surveillées aux autres populations. Nos résultats ne font ressortir aucune tendance nette de l’abondance des adultes de retour en Bretagne au cours de la période étudiée. [Traduit par la Rédaction]

Estimation of abundance with wide spatiotemporal coverage is essential to the assessment and management of wild populations. But, in many cases, data available to estimate abundance time series have diverse forms, variable quality over space and time and they stem from multiple data collection procedures. We developed a hierarchical Bayesian modelling (HBM) approach that take full advantage of the diverse assemblage of data at hand to estimate homogeneous time series of abundances irrespective of the data collection procedure. We apply our approach to the estimation of adult abundances of 18 Atlantic salmon (Salmo salar) populations of Brittany (France) from 1987 to 2017 using catch statistics, environmental covariates, and fishing effort. Additional data of total or partial abundance collected in four closely monitored populations are also integrated into the analysis. The HBM framework allows the transfer of information from the closely monitored populations to the others. Our results reveal no clear trend in the abundance of adult returns in Brittany over the period studied. lations d'animaux sauvages. Dans de nombreux cas, toutefois, les donnees disponibles pour l'estimation de series chronologidifferentes procedures. Nous avons elabore une approche de modelisation bayesienne hierarchique (MBH) pour tirer pleinement parti de l'assemblage varie de donnees disponibles pour estimer des series chronologiques homogenes d'abondance, quelle que soit la procedure de collecte de donnees. Nous appliquons l'approche a l'estimation de l'abondance d'adultes dans 18 populations de saumon atlantique (Salmo salar) de Bretagne (France) de 1987 a 2017, en utilisant des statistiques sur les prises, des covariables faisant l'objet d'une surveillance etroite sont aussi integrees a l'analyse. Le cadre de MBH permet le transfert d'information l'abondance des adultes de retour en Bretagne au cours de la periode etudiee. [Traduit par la Redaction], L’estimation de l’abondance à vaste couverture spatiotemporelle est nécessaire pour l’évaluation et la gestion de populations d’animaux sauvages. Dans de nombreux cas, toutefois, les données disponibles pour l’estimation de séries chronologiques d’abondance prennent différentes formes, sont de qualité variable dans le temps et l’espace et ont été obtenues par différentes procédures. Nous avons élaboré une approche de modélisation bayésienne hiérarchique (MBH) pour tirer pleinement parti de l’assemblage varié de données disponibles pour estimer des séries chronologiques homogènes d’abondance, quelle que soit la procédure de collecte de données. Nous appliquons l’approche à l’estimation de l’abondance d’adultes dans 18 populations de saumon atlantique (Salmo salar) de Bretagne (France) de 1987 à 2017, en utilisant des statistiques sur les prises, des covariables environnementales et l’effort de pêche. D’autres données sur l’abondance totale ou partielle obtenues pour quatre populations faisant l’objet d’une surveillance étroite sont aussi intégrées à l’analyse. Le cadre de MBH permet le transfert d’information sur des populations étroitement surveillées aux autres populations. Nos résultats ne font ressortir aucune tendance nette de l’abondance des adultes de retour en Bretagne au cours de la période étudiée. [Traduit par la Rédaction]

Underwater acoustic cameras are high potential devices for many applications in ecology, notably for fisheries management and monitoring. However how to extract such data into high value information without a time-consuming entire dataset reading by an operator is still a challenge. Moreover the analysis of acoustic imaging, due to its low signal-to-noise ratio, is a perfect training ground for experimenting with new approaches, especially concerning Deep Learning techniques. We present hereby a novel approach that takes advantage of both CNN (Convolutional Neural Network) and classical CV (Computer Vision) techniques, able to detect a generic class ''fish'' in acoustic video streams. The pipeline pre-treats the acoustic images to extract 2 features, in order to localise the signals and improve the detection performances. To ensure the performances from an ecological point of view, we propose also a two-step validation, one to validate the results of the trainings and one to test the method on a real-world scenario. The YOLOv3-based model was trained with data of fish from multiple species recorded by the two common acoustic cameras, DIDSON and ARIS, including species of high ecological interest, as Atlantic salmon or European eels. The model we developed provides satisfying results detecting almost 80% of fish and minimizing the false positive rate, however the model is much less efficient for eel detections on ARIS videos. The first CNN pipeline for fish monitoring exploiting video data from two models of acoustic cameras satisfies most of the required features. Many challenges are still present, such as the automation of fish species identification through a multiclass model. 1 However the results point a new solution for dealing with complex data, such as sonar data, which can also be reapplied in other cases where the signal-to-noise ratio is a challenge.

Quantifying connectivity within fish metapopulations is an important component in understanding population dynamics and providing an evidence base for assessment and management. We investigate meta-population connectivity of the common sole (Solea solea L.) in the Eastern English Channel (EEC). The EEC common sole stock is currently assessed as a single and spatially homogeneous population but connectivity induced through adult movements within this stock and with nearby stocks remains unknown. To fill this knowledge gap, we developed a state-space mark-recovery model, designed to estimate adult connectivity, using mark-recapture data from multiple release experiments from 1970 to 2018 across the EEC and adjacent management areas. The model estimates seasonal fish movements between five pre-defined areas, Western English Channel, Eastern English Channel (split into three discrete sub-areas) and North Sea. Over 32000 fish were tagged, 4036 of which were recovered via fisheries. Our results suggest minimal large-scale adult movements between these areas: movements among spatial units within the EEC were very low with even lower levels of immigration from areas adjoining the EEC. Our results support the hypothesis of segregated populations within the EEC. The importance of accommodating population substructure in the fisheries management is considered.

Quantifying connectivity within fish metapopulations is an important component in understanding population dynamics and providing an evidence base for assessment and management. We investigate meta-population connectivity of the common sole (Solea solea L.) in the Eastern English Channel (EEC). The EEC common sole stock is currently assessed as a single and spatially homogeneous population but connectivity induced through adult movements within this stock and with nearby stocks remains unknown. To fill this knowledge gap, we developed a state-space mark-recovery model, designed to estimate adult connectivity, using mark-recapture data from multiple release experiments from 1970 to 2018 across the EEC and adjacent management areas. The model estimates seasonal fish movements between five pre-defined areas, Western English Channel, Eastern English Channel (split into three discrete sub-areas) and North Sea. Over 32000 fish were tagged, 4036 of which were recovered via fisheries. Our results suggest minimal large-scale adult movements between these areas: movements among spatial units within the EEC were very low with even lower levels of immigration from areas adjoining the EEC. Our results support the hypothesis of segregated populations within the EEC. The importance of accommodating population substructure in the fisheries management is considered.

Quantifying connectivity within fish metapopulations is an important component in understanding population dynamics and providing an evidence base for assessment and management. We investigate meta-population connectivity of the common sole (Solea solea L.) in the Eastern English Channel (EEC). The EEC common sole stock is currently assessed as a single and spatially homogeneous population but connectivity induced through adult movements within this stock and with nearby stocks remains unknown. To fill this knowledge gap, we developed a state-space mark-recovery model, designed to estimate adult connectivity, using mark-recapture data from multiple release experiments from 1970 to 2018 across the EEC and adjacent management areas. The model estimates seasonal fish movements between five pre-defined areas, Western English Channel, Eastern English Channel (split into three discrete sub-areas) and North Sea. Over 32000 fish were tagged, 4036 of which were recovered via fisheries. Our results suggest minimal large-scale adult movements between these areas: movements among spatial units within the EEC were very low with even lower levels of immigration from areas adjoining the EEC. Our results support the hypothesis of segregated populations within the EEC. The importance of accommodating population substructure in the fisheries management is considered.

A major challenge in understanding the response of populations to climate change is to separate the effects of local drivers acting independently on specific populations, from the effects of global drivers that impact multiple populations simultaneously and thereby synchronize their dynamics. We investigated the environmental drivers and the demographic mechanisms of the widespread decline in marine survival rates of Atlantic salmon (Salmo salar) over the last four decades. We developed a hierarchical Bayesian life cycle model to quantify the spatial synchrony in the marine survival of 13 large groups of populations (called stock units, SU) from two continental stock-groupings (CSG) in North America (NA) and Southern Europe (SE) over the period 1971-2014. We found strong coherence in the temporal variation in post-smolt marine survival among the 13 SU of NA and SE. A common North Atlantic trend explains 37% of the temporal variability of the survivals for the 13 SU and declines by a factor 1.8 over the 1971-2014 time series. Synchrony in survival trends is stronger between SU within each CSG. The common trends at the scale of NA and SE capture 60% and 42% of the total variance of temporal variations, respectively. Temporal variations of the post-smolt survival are best explained by the temporal variations of sea surface temperature (SST, negative correlation) and net primary production indices (PP, positive correlation) encountered by salmon in common domains during their marine migration. Specifically, in the Labrador Sea/Grand Banks for NA populations 26% and 24% of variance is captured by SST and PP, respectively and in the Norwegian Sea for SE populations 21% and 12% of variance is captured by SST and PP, respectively. The findings support the hypothesis of a response of salmon populations to large climate induced changes in the North Atlantic simultaneously impacting populations from distant continental habitats.

Tropical cyclones are renowned for their destructive nature and are an important feature of marine and coastal tropical ecosystems. Over the last 40 years, their intensity, frequency and tracks have changed, partly in response to ocean warming, and future predictions indicate that these trends are likely to continue with potential consequences for human populations and coastal ecosystems. However, our understanding of how tropical cyclones currently affect marine biodiversity, and pelagic species in particular, is limited. For seabirds, the impacts of cyclones are known to be detrimental at breeding colonies, but impacts on the annual survival of pelagic adults and juveniles remain largely unexplored and no study has simultaneously explored the direct impacts of cyclones on different life-history stages across the annual life cycle. We used a 20-year data set on tropical cyclones in the Indian Ocean, tracking data from 122 Round Island petrels and long-term capture–mark–recapture data to explore the impacts of tropical cyclones on the survival of adult and juvenile (first year) petrels during both the breeding and migration periods. The tracking data showed that juvenile and adult Round Island petrels utilize the three cyclone regions of the Indian Ocean and were potentially exposed to cyclones for a substantial part of their annual cycle. However, only juvenile petrel survival was affected by cyclone activity; negatively by a strong cyclone in the vicinity of the breeding colony and positively by increasing cyclone activity in the Northern Indian Ocean where they spend the majority of their first year at sea. These contrasting effects raise the intriguing prospect that the projected changes in cyclones under current climate change scenarios may have positive as well as the more commonly perceived negative impacts on marine biodiversity.

Tropical cyclones are renowned for their destructive nature and are an important feature of marine and coastal tropical ecosystems. Over the last 40 years, their intensity, frequency and tracks have changed, partly in response to ocean warming, and future predictions indicate that these trends are likely to continue with potential consequences for human populations and coastal ecosystems. However, our understanding of how tropical cyclones currently affect marine biodiversity, and pelagic species in particular, is limited. For seabirds, the impacts of cyclones are known to be detrimental at breeding colonies, but impacts on the annual survival of pelagic adults and juveniles remain largely unexplored and no study has simultaneously explored the direct impacts of cyclones on different life-history stages across the annual life cycle. We used a 20-year data set on tropical cyclones in the Indian Ocean, tracking data from 122 Round Island petrels and long-term capture–mark–recapture data to explore the impacts of tropical cyclones on the survival of adult and juvenile (first year) petrels during both the breeding and migration periods. The tracking data showed that juvenile and adult Round Island petrels utilize the three cyclone regions of the Indian Ocean and were potentially exposed to cyclones for a substantial part of their annual cycle. However, only juvenile petrel survival was affected by cyclone activity; negatively by a strong cyclone in the vicinity of the breeding colony and positively by increasing cyclone activity in the Northern Indian Ocean where they spend the majority of their first year at sea. These contrasting effects raise the intriguing prospect that the projected changes in cyclones under current climate change scenarios may have positive as well as the more commonly perceived negative impacts on marine biodiversity.

Author Posting. © Inter-Research, 2012. This article is posted here by permission of Inter-Research for personal use, not for redistribution. The definitive version was published in Marine Ecology Progreee Series 454 (2012): 285-307, doi:10.3354/meps09616., Over the last century, major climate changes and intense human exploitation of natural living resources have occurred in the Southern Ocean, potentially affecting its ecosystems up to top marine predators. Fisheries may also directly affect seabirds through bycatch and additional food resources provided by discards. The past 20 yr of research has seen an increasing number of studies investigating the effects of climate change and fisheries activities on Southern Ocean seabirds. Here, we review these studies in order to identify patterns in changes in distribution, phenology, demography and population dynamics in response to changes in climate and fisheries bycatch. Shifts in distribution and breeding phenology were documented in parallel to increases in sea-surface temperatures and changes in sea-ice cover. Above all warm sea-surface temperatures negatively affected demographic parameters, although exceptions were found. Relationships suggest non-linear effects of sea-ice cover on demographic parameters and population dynamics, with optimum sea-ice cover conditions appearing to be the rule. Fishing efforts were mainly negatively related to survival rates, and only for a few species positively related to breeding success. A handful of studies found that chronic mortality of immature birds due to fisheries negatively affected populations. Climate factors and fisheries bycatch may simultaneously affect demographic parameters in a complex way, which can be integrated in population models to project population trajectories under future climate or fisheries scenarios. Needed are studies that integrate other environmental factors, trophic levels, foraging behaviour, climate−fisheries interactions, and the mechanisms underlying phenotypic plasticity, such as some pioneering studies conducted elsewhere., We thank the French Polar Institute (IPEV), the Terres Australes et Antarctiques Françaises, the Zone Atelier de Recherches sur l’Environnement Antarctique et Subantarctique (Centre National de la Recherche Scientifique—Institut Ecologie et Environnement), and the program Behavioural and Demographic Responses of Indian Ocean Marine Top Predators to Global Changes (REMIGE) funded by ANR Biodiversity for their ongoing support.

Author Posting. © Inter-Research, 2012. This article is posted here by permission of Inter-Research for personal use, not for redistribution. The definitive version was published in Marine Ecology Progreee Series 454 (2012): 285-307, doi:10.3354/meps09616., Over the last century, major climate changes and intense human exploitation of natural living resources have occurred in the Southern Ocean, potentially affecting its ecosystems up to top marine predators. Fisheries may also directly affect seabirds through bycatch and additional food resources provided by discards. The past 20 yr of research has seen an increasing number of studies investigating the effects of climate change and fisheries activities on Southern Ocean seabirds. Here, we review these studies in order to identify patterns in changes in distribution, phenology, demography and population dynamics in response to changes in climate and fisheries bycatch. Shifts in distribution and breeding phenology were documented in parallel to increases in sea-surface temperatures and changes in sea-ice cover. Above all warm sea-surface temperatures negatively affected demographic parameters, although exceptions were found. Relationships suggest non-linear effects of sea-ice cover on demographic parameters and population dynamics, with optimum sea-ice cover conditions appearing to be the rule. Fishing efforts were mainly negatively related to survival rates, and only for a few species positively related to breeding success. A handful of studies found that chronic mortality of immature birds due to fisheries negatively affected populations. Climate factors and fisheries bycatch may simultaneously affect demographic parameters in a complex way, which can be integrated in population models to project population trajectories under future climate or fisheries scenarios. Needed are studies that integrate other environmental factors, trophic levels, foraging behaviour, climate−fisheries interactions, and the mechanisms underlying phenotypic plasticity, such as some pioneering studies conducted elsewhere., We thank the French Polar Institute (IPEV), the Terres Australes et Antarctiques Françaises, the Zone Atelier de Recherches sur l’Environnement Antarctique et Subantarctique (Centre National de la Recherche Scientifique—Institut Ecologie et Environnement), and the program Behavioural and Demographic Responses of Indian Ocean Marine Top Predators to Global Changes (REMIGE) funded by ANR Biodiversity for their ongoing support.

A major challenge in ecology is to understand the impact of increased environmental variability on populations and ecosystems. To maximize their fitness in a variable environment, life history theory states that individuals should favor a bet-hedging strategy, involving a reduction of annual breeding performance and an increase in adult survival so that reproduction can be attempted over more years. As a result, evolution toward longer life span is expected to reduce the deleterious effects of extra variability on population growth, and consequently on the trait contributing the most to it (e. g., adult survival in long-lived species). To investigate this, we compared the life histories of two Black-browed Albatross (Thalassarche melanophrys) populations breeding at South Georgia (Atlantic Ocean) and Kerguelen (Indian Ocean), the former in an environment nearly three times more variable climatically (e. g., in sea surface temperature) than the latter. As predicted, individuals from South Georgia (in the more variable environment) showed significantly higher annual adult survival (0.959, SE = 0.003) but lower annual reproductive success (0.285 chick per pair, SE = 0.039) than birds from Kerguelen (survival = 0.925, SE = 0.004; breeding success = 0.694, SE = 0.027). In both populations, climatic conditions affected the breeding success and the survival of inexperienced breeders, whereas the survival of experienced breeders was unaffected. The strength of the climatic impact on survival of inexperienced breeders was very similar between the two populations, but the effect on breeding success was positively related to environmental variability. These results provide rare and compelling evidence to support bet-hedging underlying changes in life history traits as an adaptive response to environmental variability.

A major challenge in ecology is to understand the impact of increased environmental variability on populations and ecosystems. To maximize their fitness in a variable environment, life history theory states that individuals should favor a bet-hedging strategy, involving a reduction of annual breeding performance and an increase in adult survival so that reproduction can be attempted over more years. As a result, evolution toward longer life span is expected to reduce the deleterious effects of extra variability on population growth, and consequently on the trait contributing the most to it (e. g., adult survival in long-lived species). To investigate this, we compared the life histories of two Black-browed Albatross (Thalassarche melanophrys) populations breeding at South Georgia (Atlantic Ocean) and Kerguelen (Indian Ocean), the former in an environment nearly three times more variable climatically (e. g., in sea surface temperature) than the latter. As predicted, individuals from South Georgia (in the more variable environment) showed significantly higher annual adult survival (0.959, SE = 0.003) but lower annual reproductive success (0.285 chick per pair, SE = 0.039) than birds from Kerguelen (survival = 0.925, SE = 0.004; breeding success = 0.694, SE = 0.027). In both populations, climatic conditions affected the breeding success and the survival of inexperienced breeders, whereas the survival of experienced breeders was unaffected. The strength of the climatic impact on survival of inexperienced breeders was very similar between the two populations, but the effect on breeding success was positively related to environmental variability. These results provide rare and compelling evidence to support bet-hedging underlying changes in life history traits as an adaptive response to environmental variability.