The Mediterranean Sea, despite covering less than 1% of the global ocean surface, hosts a disproportionately high level of marine biodiversity, including a genetically distinct and endangered subpopulation of fin whales (Balaenoptera physalus). As the only regularly occurring mysticete in the basin, the Mediterranean fin whale is increasingly exposed to rapid oceanographic changes driven by climate change, which may alter prey availability, disrupt migratory patterns, and modify the distribution of suitable habitat. Understanding how environmental change influences fin whale presence is therefore critical for developing effective conservation strategies. This study investigates the relationship between fin whale occurrence and key environmental variables, and models future distribution suitability under projected climate scenarios. Presence data from 2006 to 2022 were compiled from ACCOBAMS and OBIS databases, while pseudo-absences were generated using occurrences of striped dolphins (Stenella coeruleoalba) recorded in locations where fin whales were not detected. Environmental predictors were sourced from the Copernicus Climate Data Store and included sea surface temperature (SST), phosphate concentration, phytoplankton biomass, and zooplankton biomass, variables known to influence euphausiid availability, the primary prey of Mediterranean fin whales. Two species distribution modelling approaches were applied: Generalized Additive Models (GAMs) and Random Forest (RF). Model performance was assessed through cross-validation and comparison of predictive accuracy. The Random Forest model outperformed GAMs, capturing non-linear responses and complex interactions among variables; SST emerged as the most influential predictor. Phosphate, phytoplankton, and zooplankton contributed to model performance but exhibited weaker or more variable relationships with fin whale presence. Future projections were developed using two Representative Concentration Pathways (RCPs), RCP 4.5 and RCP 8.5, to estimate habitat changes for mid- century (2050) and late-century (2099). Under the RCP 4.5 scenario, the predicted probability of occurrence remains generally higher than under RCP 8.5, highlighting a progressive deterioration of suitable habitat under more severe warming conditions. Overall, this study highlights the vulnerability of Mediterranean fin whales to climate-driven oceanographic change and underscores the urgency of integrating climate projections into conservation planning. Identifying future habitat refugia will be essential for mitigating risks to this already fragile and isolated population.
The Mediterranean Sea, despite covering less than 1% of the global ocean surface, hosts a disproportionately high level of marine biodiversity, including a genetically distinct and endangered subpopulation of fin whales (Balaenoptera physalus). As the only regularly occurring mysticete in the basin, the Mediterranean fin whale is increasingly exposed to rapid oceanographic changes driven by climate change, which may alter prey availability, disrupt migratory patterns, and modify the distribution of suitable habitat. Understanding how environmental change influences fin whale presence is therefore critical for developing effective conservation strategies. This study investigates the relationship between fin whale occurrence and key environmental variables, and models future distribution suitability under projected climate scenarios. Presence data from 2006 to 2022 were compiled from ACCOBAMS and OBIS databases, while pseudo-absences were generated using occurrences of striped dolphins (Stenella coeruleoalba) recorded in locations where fin whales were not detected. Environmental predictors were sourced from the Copernicus Climate Data Store and included sea surface temperature (SST), phosphate concentration, phytoplankton biomass, and zooplankton biomass, variables known to influence euphausiid availability, the primary prey of Mediterranean fin whales. Two species distribution modelling approaches were applied: Generalized Additive Models (GAMs) and Random Forest (RF). Model performance was assessed through cross-validation and comparison of predictive accuracy. The Random Forest model outperformed GAMs, capturing non-linear responses and complex interactions among variables; SST emerged as the most influential predictor. Phosphate, phytoplankton, and zooplankton contributed to model performance but exhibited weaker or more variable relationships with fin whale presence. Future projections were developed using two Representative Concentration Pathways (RCPs), RCP 4.5 and RCP 8.5, to estimate habitat changes for mid- century (2050) and late-century (2099). Under the RCP 4.5 scenario, the predicted probability of occurrence remains generally higher than under RCP 8.5, highlighting a progressive deterioration of suitable habitat under more severe warming conditions. Overall, this study highlights the vulnerability of Mediterranean fin whales to climate-driven oceanographic change and underscores the urgency of integrating climate projections into conservation planning. Identifying future habitat refugia will be essential for mitigating risks to this already fragile and isolated population.
Modelling the impact of Climate Change on the habitat distribution of the fin whale in the Mediterranean Sea.
GAVIN, MARIA GIOVANNA
2024/2025
Abstract
The Mediterranean Sea, despite covering less than 1% of the global ocean surface, hosts a disproportionately high level of marine biodiversity, including a genetically distinct and endangered subpopulation of fin whales (Balaenoptera physalus). As the only regularly occurring mysticete in the basin, the Mediterranean fin whale is increasingly exposed to rapid oceanographic changes driven by climate change, which may alter prey availability, disrupt migratory patterns, and modify the distribution of suitable habitat. Understanding how environmental change influences fin whale presence is therefore critical for developing effective conservation strategies. This study investigates the relationship between fin whale occurrence and key environmental variables, and models future distribution suitability under projected climate scenarios. Presence data from 2006 to 2022 were compiled from ACCOBAMS and OBIS databases, while pseudo-absences were generated using occurrences of striped dolphins (Stenella coeruleoalba) recorded in locations where fin whales were not detected. Environmental predictors were sourced from the Copernicus Climate Data Store and included sea surface temperature (SST), phosphate concentration, phytoplankton biomass, and zooplankton biomass, variables known to influence euphausiid availability, the primary prey of Mediterranean fin whales. Two species distribution modelling approaches were applied: Generalized Additive Models (GAMs) and Random Forest (RF). Model performance was assessed through cross-validation and comparison of predictive accuracy. The Random Forest model outperformed GAMs, capturing non-linear responses and complex interactions among variables; SST emerged as the most influential predictor. Phosphate, phytoplankton, and zooplankton contributed to model performance but exhibited weaker or more variable relationships with fin whale presence. Future projections were developed using two Representative Concentration Pathways (RCPs), RCP 4.5 and RCP 8.5, to estimate habitat changes for mid- century (2050) and late-century (2099). Under the RCP 4.5 scenario, the predicted probability of occurrence remains generally higher than under RCP 8.5, highlighting a progressive deterioration of suitable habitat under more severe warming conditions. Overall, this study highlights the vulnerability of Mediterranean fin whales to climate-driven oceanographic change and underscores the urgency of integrating climate projections into conservation planning. Identifying future habitat refugia will be essential for mitigating risks to this already fragile and isolated population.| File | Dimensione | Formato | |
|---|---|---|---|
|
Gavin_Maria_Giovanna.pdf
accesso aperto
Dimensione
4.83 MB
Formato
Adobe PDF
|
4.83 MB | Adobe PDF | Visualizza/Apri |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/101702