Marine biodiversity at the end of the world: Cape Horn and Diego Ramírez islands

Abstract

The vast and complex coast of the Magellan Region of extreme southern Chile possesses a diversity of habitats including fjords, deep channels, and extensive kelp forests, with a unique mix of temperate and sub-Antarctic species. The Cape Horn and Diego Ramírez archipelagos are the most southerly locations in the Americas, with the southernmost kelp forests, and some of the least explored places on earth. The giant kelp Macrocystis pyrifera plays a key role in structuring the ecological communities of the entire region, with the large brown seaweed Lessonia spp. forming dense understories. Kelp densities were highest around Cape Horn, followed by Diego Ramírez, and lowest within the fjord region of Francisco Coloane Marine Park (mean canopy densities of 2.51 kg m-2, 2.29 kg m-2, and 2.14 kg m-2, respectively). There were clear differences in marine communities among these sub-regions, with the lowest diversity in the fjords. We observed 18 species of nearshore fishes, with average species richness nearly 50% higher at Diego Ramírez compared with Cape Horn and Francisco Coloane. The number of individual fishes was nearly 10 times higher at Diego Ramírez and 4 times higher at Cape Horn compared with the fjords. Dropcam surveys of mesophotic depths (53–105 m) identified 30 taxa from 25 families, 15 classes, and 7 phyla. While much of these deeper habitats consisted of soft sediment and cobble, in rocky habitats, echinoderms, mollusks, bryozoans, and sponges were common. The southern hagfish (Myxine australis) was the most frequently encountered of the deep-sea fishes (50% of deployments), and while the Fueguian sprat (Sprattus fuegensis) was the most abundant fish species, its distribution was patchy. The Cape Horn and Diego Ramírez archipelagos represent some of the last intact sub-Antarctic ecosystems remaining and a recently declared large protected area will help ensure the health of this unique region.

Kelp forests at the end of the earth: 45 years later

“…Our data show similarly low densities of L. albus between the two time periods and recent surveys in the Cape Horn Archipelago recorded even lower densities of this species ( � X = 0.6 ± 1.3 individuals . m -2 ) [22], which is consistent with Dayton's prediction [18].…”

“…Kelp canopy fraction has been shown to be linearly correlated to canopy biomass density at sites in California [31,32] and this canopy pixel fraction to biomass relationship has been applied to M. pyrifera sites at several locations across the Americas (Baja California, SE Alaska, Cape Horn [33,34]). The vast majority of available Landsat imagery in the region was acquired starting in 1998, likely due to data storage limitations from this remote area prior to that date.…”

“…A recent global analysis of changes in kelp abundances over the past half-century identified declines in 38% of ecoregions for which there are data, with increases in 27% of ecoregions, and no detectable change in 35% of ecoregions [22]. These spatially variable trajectories reflect regional differences in the drivers of changes in kelp abundance and highlight the dynamic nature of kelp populations.…”

“…Unlike other regions of the world, the M. pyrifera beds in southern Chile do not appear to be controlled by sea urchins; however, sea urchins and the gastropod Tegula atra have been shown to influence kelp recruitment dynamics in lower latitudes within the region [19][20][21]. In addition to M. pyrifera, understories of other kelp taxa (e.g., Lessonia) are also noted to be important to the ecology and ecosystem function of southern Chile [3,18,22].…”

A High-Resolution Global Map of Giant Kelp (Macrocystis pyrifera) Forests and Intertidal Green Algae (Ulvophyceae) with Sentinel-2 Imagery

“…The above-mentioned methods have been applied at local to regional scales, e.g., in the Santa Barbara Channel (Southern California) or the Francisco Coloane Marine Park, Cape Horn archipelago and Diego Ramirez islands in southern Chile [29]. These studies were not scaled up to the global distribution due to the high operative costs of human hours required for pre-processing and classifying images, as well as providing pure endmember data, especially along extensive coastlines (but see [30]).…”