Island biogeography of marine organisms

For this research we have explored through technical diving a chain of seamounts and oceanic islands in the South Atlantic. Some of the reefs we found were 900 km from the coast and at depths of up to 85 meters. We sampled endemic reef fishes during the dives, and by studying their evolutionary history we developed a new conceptual model to explain the biogeography and evolution of marine organisms in isolated islands.

Go to the profile of Hudson Tercio Pinheiro
Sep 08, 2017
2
1
Upvote 2 Comment

Our paper in Nature can be found here: http://go.nature.com/2jxVLir

Studying the biodiversity of seamounts of the Vitória-Trindade Chain was an old dream. Most of us had the opportunity to visit Trindade Island before, which is situated at the end of the chain, but the seamounts were sampled for the first time. Our results from Trindade disclosed unexpectedly high biodiversity, and a high endemism rate. The high biodiversity was always thought to be facilitated by the presence of the seamount chain, which fishes could use as stepping-stones to reach the islands. However, high connectivity would translate into a low endemism rate, and this paradox of seamounts at the same time helping connectivity but also promoting endemics, was our main motivation. 

                                                                              Trindade Island, off the coast of Brazil.

Reaching the seamounts and diving on those deep and isolated environments was not an easy task. We were all experienced divers from the Universidade Federal do Espírito Santo (Brazil), however, we had to re-learn how to dive from scratch, engaging in several technical diving courses to learn how to perform mixed-gas and decompression dives. We also spent months modifying a large fishing boat for research and diving operations. The boat conditions were not that great, but the fishing crew was fantastic. They used to fish along the Vitória-Trindade Chain for decades, knowing the region very well. Most of the reefs we visited were known fishing spots, so the local fishermen also deserve credit for our discoveries.

Boat and equipment used during our expedition to the Vitória-Trindade Chain

All the seamounts of the Vitória-Trindade Chain were previously described as having their tops as flat platforms, around 60-70 meters depth. But when we did our first dive we discovered complex reef environments, rising from 70 m all way to 30-20 m depth. Almost every fish we observed on the seamounts were new records, including some endemics that were previously only found around Trindade Island and Martin Vaz Archipelago, at the far east of the Chain. 

Reef environments on the top of the Vitória Seamount, Vitória-Trindade Chain, Brazil.

We had the opportunity to collect most of the endemic fishes from the seamounts and islands, and bring them to the lab at the California Academy of Sciences (USA), where we extracted their DNA and sequenced them. With their DNA sequence we were able to compare their genes with the genes of their sister species, from the Brazilian continental coast. The comparison gave us the information about how different the lineages are, and using the mutation rates for those genes we were able to estimate the approximate age of the species from the Vitória-Trindade Chain.


To our surprise, most of the endemic species had a very recent origin, colonizing the seamounts and islands and diverging from the continental populations less than 2 million years ago. This geological period is known as the Pleistocene, when the Earth experienced several climatic variation events that changed the sea levels. During the Ice Ages for example, the sea level was much lower than now. The last Ice Age, which ended around 10,000 years ago, presented one of the lowest sea levels, about 120 meters below the present level. The low sea level exposes at least nine seamounts, increasing the number of islands from two to 11. Back to the fishes, when we looked at the ecological traits of the endemics we discovered that the younger lineages were composed of small species with very limited dispersal abilities. Most of these species were dependent on the shallow waters presented by the exposed seamounts to reach the islands at the end of the chain during sea-level lows, and had the seamount populations extirpated during the sea level highs. So isolation, driven by sea level change, was important for the origin of the endemic species today only found around the islands.

Brazil blenny Hypleurochilus brasil, one of the young endemic species with limited dispersal capability.

The population genetics of these young endemics showed divergent evolutionary histories too. For instance, the neon goby is widespread along the entire chain, probably colonized the seamounts first and the islands later, and had its populations expanding since their initial colonization events. The Brazil blenny likely colonized the islands after a mainland population expansion. Interestingly, we found a specimen at Davis Seamount, which because of its morphology, we thought would cluster with the island population. Differently from what we were expecting, this individual is closely related to the continental coast population, indicating a secondary colonization event. It seems that the high morphological similarity between seamount and island populations is either a signal of convergent evolution (similar characteristics that evolve independently) or hybridization.

The few older endemic species are considered relict species, or paleoendemics, because they do not have sister species in the Brazilian Province. Their seamount and island populations survived the harsh environmental conditions of the Ice Ages, that wiped out their continental sister lineages. These species have ecological traits such large size, long pelagic larval duration, and schooling and/or nocturnal activity, what makes them good dispersers and colonizers. 

Trindade wrasse Halichoeres rubrovirens, one of the paleoendemic species found in the Vitória-Trindade Chain.

The ecological traits and the evolutionary history of the endemics matched the premises of the theory of island biogeography, where species with good dispersal abilities colonize islands first, with a high immigration rate. Weak dispersers arrive later, associated with a higher extinction rate – as they do not maintain a connection with the mainland, they either adapt or go extinct. Differently from the terrestrial realm, marine organisms are well known by their high dispersal capabilities. They have a pelagic larval phase that drift with oceanic currents for many days or weeks. Thus, marine organisms are able to quickly occupy most of island niches and usually not leaving anything open to be filled by adaptations. The few terrestrial organisms that make to the islands, on the other hand, find many open niches, commonly diversifying into many species, as it happened, for example, to the Darwin finches in Galapagos – a process known as adaptive radiation.     

We then conceptualized the idea to a general dynamic theory of island biogeography for marine organisms, which relates changes on immigration, extinction and speciation rates to the geological history of islands and seamounts. After drawing and sketching models on paper towels at the Long Marine Lab (University of California at Santa Cruz, USA), we ended up developing two main hypotheses. The first, following prior ideas, has immigration slowly decreasing with the geological history of the island, with extinction increasing. In a negative relationship to immigration, speciation follows the extinction rate, mainly for two reasons: the extinction of weak disperses that cannot maintain connectivity to mainland is well known, and the extinction of less adapted species would open niches and reduce competitive interactions to nascent species. The second hypothesis considers immigration as an ecological process, happening quicker for marine organisms, and thus closely related to the species richness of the islands. Following this idea, speciation and extinction of weak dispersers are constantly high during island formation and erosion, maintaining a high turnover of weak dispersers. Speciation and richness would drop with island erosion and seamount subsidence.

Vitória-Trindade Chain, in the central coast of Brazil, Southwestern Atlantic.

The Vitória-Trindade Chain is a perfect scenario to test biogeographic and evolutionary theory. Composed by a line of seamounts and islands perpendicular to the mainland, it is isolated from other islands and archipelagos. In addition to the scientific potential, the Vitória-Trindade Chain also provides other ecosystems services such as fisheries. Many Brazilian fishers use the Chain as fishing ground. Nevertheless, despite its uniqueness, environmental and socio-economic importance, part of the Vitória-Trindade Chain is being targeted for mining activities: a company has extracted carbonate from shallow coralline reefs to produce fertilizers. This activity has destroyed part of the seamount ecosystems situated outside the Brazilian Exclusive Economic Zone, and has the potential to disrupt ecological and evolutionary processes, affecting the sustainability of fishing activities as well. Scientists, environmentalists and Brazilian citizens are trying to show to the Brazilian government the importance of the region and its potential for sustainable development. Proposals of a Marine Biosphere Reserve and Marine Protected Areas are being suggested and we hope they will be materialized soon.

Go to the profile of Hudson Tercio Pinheiro

Hudson Tercio Pinheiro

Researcher, California Academy of Sciences

1 Comments

Go to the profile of Sergio R Floeter
Sergio R Floeter 3 months ago

Very nice behind the paper! Thanks for sharing it. Keep the good work!