Ecological dependencies make remote reef fish communities most vulnerable to coral loss

Ecosystems face both local hazards, such as over-exploitation, and global hazards, such as climate change. Since the impact of local hazards attenuates with distance from humans, local extinction risk should decrease with remoteness, making faraway areas safe havens for biodiversity. However, isolation and reduced anthropogenic disturbance may increase ecological specialization in remote communities, and hence their vulnerability to secondary effects of diversity loss propagating through networks of interacting species. We show this to be true for reef fish communities across the globe. An increase in fish-coral dependency with the distance of coral reefs from human settlements, paired with the far-reaching impacts of global hazards, increases the risk of fish species loss, counteracting the benefits of remoteness. Hotspots of fish risk from fish-coral dependency are distinct from those caused by direct human impacts, increasing the number of risk hotspots by ~30% globally. These findings might apply to other ecosystems on Earth and depict a world where no place, no matter how remote, is safe for biodiversity, calling for a reconsideration of global conservation priorities.


nature research | reporting summary
April 2020 Field-specific reporting Please select the one below that is the best fit for your research. If you are not sure, read the appropriate sections before making your selection.

Life sciences
Behavioural & social sciences Ecological, evolutionary & environmental sciences For a reference copy of the document with all sections, see nature.com/documents/nr-reporting-summary-flat.pdf Ecological, evolutionary & environmental sciences study design All studies must disclose on these points even when the disclosure is negative.

Study description
Research sample Sampling strategy

Data collection
Timing and spatial scale Data exclusions 5. the friction surface map needed to compute accessibility: https://malariaatlas.org/research-project/accessibility-to-cities; 6. human settlement data: http://data.europa.eu/89h/jrc-ghsl-ghs_smod_pop_globe_r2016a; 7. bleaching alert data: https://coralreefwatch. We investigated the relationships between remoteness and different sources of (local extinction) risk for fish communities. We combined data on reef fish and coral distribution with data on global impacts on oceans. We built networks of potential interactions between fish and corals and between fish and fish (trophic interactions). We used those networks to quantify fish-coral dependency at the global scale. We then combined all the data into a risk assessment framework. We found that while the risks stemming from global hazards from global change and local anthropogenic hazards decrease with remoteness, the risk stemming from ecological dependencies (i.e. risk of cascading effects through interaction networks) increases. This put remote fish communities at risk, obliterating the potential beneficial effect of isolation.
12. An alternative dataset of reef fish distribution (GASPAR) to validate our fish range maps (https://rs.figshare.com/collections/ Supplementary_material_from_Coral_reef_fishes_reveal_strong_divergence_in_the_prevalence_of_traits_along_with_the_global_d iversity_gradient_/5647995/) We did not predetermine sampling size, and used all available data.
Data were downloaded from online sources (using custom Python scripts when possible) by Giovanni Strona using a personal computer and stored both in a physical hard disk and in a remote server owned by the University of Helsinki.
Data were collected in January 2020. Data cover the global scale.
From the full list of occurrence records for fish, we excluded all the species for which we could not retrieve ecological information (which resulted in a final list of 5862 fish species). As for the occurrence data, we used a geo-statistical procedure (fully described in methods) to exclude outliers.