The Abrolhos reefs (16º40ʹ,19º 40ʹ S – 39º10ʹ, 37º20ʹ W) are distributed in two arcs – the Coastal and Outer arcs – lying almost parallel to the mainland shore. The inner shelf, also called the Coastal arc, is located between 10 and 20 km offshore and is formed by shallow bank reefs and isolated coral pinnacles of varying dimensions (Moura et al., 2013). The mid shelf, or Outer arc, is ∼70 km offshore and formed by isolated giant coral pinnacles (popularly called “chapeirões”) and the rocky reefs of the Abrolhos Archipelago. The tops of these giant pinnacles are deeper (>10 m) than the bank reefs of the Coastal arc (Dutra et al., 2006). The term “chapeirões” was introduced by geologist Charles Hartt in 1870, which alludes to wide-brimmed hats, hats being translated to “chapéus”, in Portuguese (“chapeirões” is the augmentative) (Hartt, 1870).

The historical chart of the Abrolhos Bank was developed by the born-Spanish cartographer and astronomer Admiral Ernest Amédée Barthélemy Mouchez during expeditions along the Brazilian coast, in 1861 (Fig. S1). Admiral Ernest carried out numerous expeditions to South America between 1856 and 1861. His surveys included the Paraguay river, from Paraná to Asunción, and numerous regions on the coast of Brazil, from Bahia to Rio de Janeiro (12º58ʹ16ʺ S, 38°30ʹ39ʺ W – 22°54ʹ10ʺ S, 43°12ʹ 27ʺ W) and even reaching the austral Lobos Island (29°20ʹ48.0ʺ S, 49°42ʹ17.4ʺ W) and Tramandaí (29º59ʹ 05ʺ S, 50º08ʹ 01ʺ W).

We obtained this chart digitalized and in high resolution from the digital collection of the Brazilian National Library (Cart525992 (bn.br)) in Rio de Janeiro (Fig. S1). The chart contains notes on the benthic habitat of the region, including sand bottoms, coral, shells, rocks, gravel, mud, clay, and seagrass beds. In addition, this nautical chart contains information such as specific reference points, hazards, and water depths that contextualize data extracted from the charts. The nautical chart of Mouchez (1861) was georeferenced using fixed reference points on the coast such as estuaries and rivers through Quantum Gis software (QGis; version 3.16.2).

To estimate spatial and temporal changes in the extent of coral reef habitats in the Abrolhos Bank, we compared historical reef areas with areas from three modern reef layers : (1) the Millennium Coral Reef Mapping Project (UNEP-WCMC, WorldFish Centre, WRI, TNC, 2021); (2) shapefiles of coral reefs of the Abrolhos Bank from Moura et al. (2013); and (3) shapefiles of emerged and submerged reefs from Magris et al. (2017). These three layers were derived from high-resolution satellite images, multispectral Landsat 7 images, resolution 30 meters (UNEP-WCMC, WorldFish Centre, WRI, TNC, 2021), Landsat TM (spatial resolution 30 meters) combined with side-scan sonar records (Moura et al., 2013), and Landsat 5 ETM (resolution 30 m) supplemented by multispectral IKONOS (high spatial resolution 4 m) (Magris et al., 2017).

These modern layers were obtained in shapefile format online (https://data.unep-wcmc.org/datasets/1) (UNEP-WCMC, WorldFish Centre, WRI, TNC, 2021) and provided by the authors of papers through supplemental material (Moura et al., 2013; Magris et al., 2017). The combination of modern layers allowed us to detect the occurrence of emerged and submerged corals as well as smaller reefs (i.e., isolated submerged mushroom reefs/coral pinnacles up to 50 meters in diameter). In addition, we used these three different modern sources of information to minimize potential differences in reef area estimates, making the analysis and results more robust and conservative. Alignment of the historical and present nautical chart (Moura et al., 2013; Magris et al., 2017, see below) was performed using the Quantum Gis software (QGis; version 3.16.2).

To estimate the potential changes in reef extent (area) through time, we analyzed the following reefs of the Coastal arc: Guaratibas, Timbebas, Recifes do canal (which encompasses the Areia, Cabocla, Aranguera, and Lixa reefs), Sebastião Gomes, Coroa Vermelha, and Viçosa (Table S1). These areas were chosen because they are relatively shallow and thus maps (both historical and recent) tend to be more accurate relative to deeper reefs of the Outer arc.

First, we calculated past reef areas by drawing polygons on historical coral reefs indicated in the nautical chart (Fig. 1A, B). The limits of coral reef areas were based on map legends, in which the author described reef contours as follows “The dotted lines indicate the limits in which banks and the chapeirões [mushroom-shaped growth] are enclosed” (in French: Les lignes ponctuées indiquent les limites dans lesquelles sont reenfermés les bancs et les chaperons). We therefore considered these dotted lines as the limits to past coral reef areas. In addition, we created polygons on emerged coral reefs drawn on the nautical chart. Then, these polygons were used as the basis for creating 220 m² grids both in the historical corals (Fig. 1B) and in the modern layers (Fig. 1C, D). Differences between historical and modern reef area were calculated through the number of grids covering the past and modern reef areas.

Fig. 1.

Method used to estimate temporal changes in reef area for the Abrolhos Bank. (a) Reefs represented in the historical nautical chart; (b) Grids (220 m²) created from the polygons limiting historical reef areas; (c) Modern reef layer of the Millennium Coral Reef Mapping Project (superimposed colors); and (d) Grids (220 m²) created from the polygons of the modern layer of the Millennium Coral Reef Mapping Project. Viçosa reefs (orange), Coroa Vermelha reefs (light blue) and Sebastião Gomes reefs (pink).

(0.82MB).

Percent loss (PL) in reef areas were calculated according to the following formula:

To increase understanding of historical changes and the potential drivers of change in Abrolhos reefs, we searched archival and published historical literature including naturalist reports, documents, and books, primarily from the 19th century. Most of the naturalists’ reports and books were written in English or translated to Portuguese. The compilation of historical information was made in museums, libraries, and research institutions of two cities—Rio de Janeiro (Rio Janeiro state, southeast Brazil) and Recife (Pernambuco state, northeast Brazil) —, considering the number of historical collections they host (Table S2). We also searched 14 libraries and archives in digital collections (see Table S2 for a complete list of collections). For online searches, we used the following keywords in English and Portuguese: ‘Brazilian reefs’, ‘Abrolhos reefs’, ‘scientific expedition’, ‘coral reefs’, and ‘Southern Bahia’, ‘recifes brasileiros’, ‘recifes de Abrolhos’, ‘expedição científica’, ‘recifes de corais’ e ‘sul da Bahia’. We used information from these sources to build a timeline of observations, and to assess the timing of early anthropogenic drivers of change to nearshore coral reef habitat.

Overall, the assessed reef sites in the Abrolhos Bank lost 28 ± 13% (mean ± standard deviation) of their area (km²). Reef sites with the greatest declines in area were close to the coast (Fig. 2), such as Guaratibas (49% loss, 6.65 km offshore) and Recifes do Canal (28%, 13.9 km), followed by Timbebas (22%, 15.1 km), Viçosa (20%, 11 km) and Coroa Vermelha (19%, 10.5 km) (Fig. 2). Differences in reef percent loss between sites were not significant (Kruskal-Wallis’s test, P-value = 0.406). In addition, there was no significant correlation between percent loss and distance from the coast (Spearman’s test, P-value = 0.95).

Fig. 2.

The loss of coral area in the Abrolhos Bank. In (a), pie charts represent the percent loss (orange) in coral area across different reef locations of the coastal arc. (b) Percent loss by coral reef area. Orange bars represent the mean loss and error bars represent the standard deviation. Coral reefs: Guaratibas (GUA); Coroa Vermelha (CV); Viçosa (VI); Recifes do Canal (CNR) and Timbebas (TIM).

(0.17MB).

We compiled accounts from navigators and naturalists, which we categorized as observations related to either navigation or ecological attributes. Collectively, these observations show that the Abrolhos Bank has long been a region of interest for travelers and naturalists exploring the Southwestern Atlantic and the Brazilian coast (Fig. S2). In terms of navigation, this region is known for its risks due to the coral formations that caused several shipwrecks. The hazard warnings regarding the Abrolhos reefs were recorded on maps and originated the name of the region, "Abra os olhos" in Portuguese, which can be translated to “Open your eyes”, in English. One of the first shipwrecks recorded in Abrolhos happened to the Priest José de Anchieta, in 1560 (Barros, 2016). Years later, other shipwrecks were recorded, in which the best known were Saint John the Baptist and Prinz Willem, both in 1631, and Guadiana and Rosalinda vessels in 1955 and 1985, respectively (Barros, 2016).

Descriptions of the ecological attributes of Abrolhos reefs occurred more than 200 years ago, when Europeans arrived on expeditions to the coast of Bahia state. European naturalists visited the Abrolhos Bank with an interest in describing its geomorphology, mapping potential navigational hazards as well as documenting the diversity and occurrence of coral reefs in the region (Fig. S2, Table 1). In 1819, the French Admiral Albin Reine Roussin recorded in his work "Pilote du Brésil", instructions for navigation on the Brazilian coast. The work highlights a description of the Abrolhos seabed and the risks to navigation associated to coral habitats (Roussin, 1845). Between 1817 and 1820, the German naturalists Johann Baptist von Spix and Carl Friedrich Philipp von Martius traveled across Brazil to describe the local fauna and flora, including that of the Abrolhos Bank. Some years later, Charles Darwin visited this region and reported in a passage of the book “Geological Observations on the Volcanic Islands” the high abundance of corals in the seabed “the bottom of the sea is entirely coated by irregular masses of coral, which although often of large size, do not reach the surface and form proper reefs” (Darwin, 1851).

In the second half of the 19th century, Canadian-American geologist Charles Frederick Hartt made several notes on Brazilian coral species, with a special passage for corals from southern Bahia. Charles Hartt was a travelling companion of Professor Louis Agassiz during the Thayer Expedition, which produced the first description of coral zonation for the Abrolhos reefs (Hartt, 1870). Between 1874–1879, two other naturalists, James Dana and Richard Rathbun published works on Brazilian coral reefs, focused on Abrolhos. Richard Rathbun described the characteristics and extent of the Brazilian coral reefs. In 1904, the American geologist John Casper Branner published the book “The Stone Reefs of Brazil, their Geological and Geographical Relations”, describing the geology and geography of Brazilian reefs. Along with Hartt and Rathbun, Branner was part of the Imperial Geological Commission, which was created in 1875 and visited the continental reefs in the states of Alagoas, Amazonas, Bahia, Paraná, Pernambuco, Rio Grande do Norte, São Paulo, Santa Catarina, and Sergipe, as well as Fernando de Noronha archipelago. This expedition resulted in the colleciton of approximately 500,000 specimens and important zoological, geological and ethnographic information.

More than a century later, in 1960, another work was fundamental for the understanding of the biological and ecological aspects of Brazilian reefs (Laborel, 1970). French researcher Jacques Laborel wrote his seminal thesis in 1969 entitled “Les peuplements des madréporaires des côtes tropicales du Brésil” (Madreporia populations of the tropical coasts of Brazil). Laborel traveled along the Brazilian coast studying scleractinian corals, which were called “madreporaires”. He visited the reefs between Ceará (3°42ʹ47.5ʺ S, 38°31ʹ39.1ʺ W) and São Paulo states (23°59ʹ24.0ʺ S, 46°20ʹ48.0ʺ W). In the Abrolhos Bank, Laborel described the species composition and abundance (semi-quantitatively) of the chapeirões (Laborel, 1970).

We identified 10 major works that contain information on the potential drivers of changes in reefs from museums, libraries, and digital collections (Table 1). The first evidence on the anthropogenic impacts on these reefs was reported between 1817 and 1820 by the naturalists Johann Baptist von Spix and Carl Friedrich Philipp von Martius, which mentioned coral extraction as an economic activity in the region (Fig. 3, Table 1). This activity, where corals were burned to make lime that was used in construction, was common along the Brazilian coast and in Abrolhos. For example, the authors write: “These banks can be seen in various parts of the lagoon at a depth of six to twelve feet, and are used in the construction of houses, when there is no lime” (Table 1, Spix and Martius, 1828). Spix and Martius further describe coral collection: “On the southernmost coasts, especially those of Pôrto-Seguro and Abrolhos, it seems, the fishermen…often and also capture very large and precious corals” (Spix and Martius, 1828). Abrolhos Bank was known as an important area of fishing and coral removal, and this practice can be tracked to different periods of time over the history of coral reefs in Abrolhos. In 1870, Charles Hartt also described coral removal in Abrolhos and along the Brazilian coast “I have frequently seen [Montastraea cavernosa] there in the heaps of corals brought from the Island of Itaparica to the city for burning into the lime”. Hartt describes coral of the genus Montastraea (in the past named Heliastrea) as commonly burned to make lime, and highlights the lack of limestone in Brazil as the main cause of coral extraction “Limestone is very scarce on the Brazilian coast, and corals are largely used for making lime. They call the coral pedra de cal [“lime rock” in English] in Brazil, for this reason.” In 1879, the American scientist Richard Rathbun published the work entitled “Brazilian corals and coral reefs”, which mentioned coral extraction of Bahia reefs as an activity. “Here and there, the slaves in procuring limestone, have quarried into the low inner part of the reef, and even into the high wall-like portion” (Rathbun, 1879). Years later, Branner (1904) recorded the removal of corals from the Coroa Vermelha reef, demonstrating that the activity continued in the region (Table 1, Fig. 3).

Fig. 3.

Timeline of impacts on reefs of the Abrolhos Bank. Compilation of the main works and researches regarding impacts on Abrolhos reefs from the 19th century to the present.

(0.3MB).

Historical records on coral extraction span over two centuries and point this activity as the main early source of anthropogenic impact on the Abrolhos reefs (Fig. 3). Organized in a timeline of human-derived impacts, overfishing (Previero and Gasalla, 2018), tourism (Leão et al., 1996) increased sedimentation (Dutra et al., 2006; Silva et al., 2013; Sousa et al., 2014), pollution (Coimbra et al., 2020), and climate change (Teixeira et al., 2019; Duarte et al., 2020; Ferreira et al., 2021) may have also contributed to reef loss in the region (Fig. 3). There are written accounts of fisheries in the Abrolhos bank from 1817 (Spix and Martius, 1828; Table 1), more than 200 years ago (Fig. 3). Hartt (1870), in a passage, described grouper fisheries: “The other fishery carried on in the waters of the Abrolhos is that of the garoupa [grouper], an excellent fish exceedingly abundant, and taken with the hook and line”. Apparently, coral extraction for lime was replaced by coral removal for the ornamental trade in the early 90’s (Gasparini et al., 2005). Coral extraction supplied the souvenir trade, where traders and tourists manually pulled out coral heads (Gasparini et al., 2005). This period is also marked by records of the impacts of tourism on local reefs (Leão et al., 1996). Since the 18th century, economic and urban development associated with deforestation have resulted in increased sedimentation of coastal ecosystems (Dutra et al., 2006; Silva et al., 2013; Sousa et al., 2014). Over the last 20 years, research has identified the impacts of climate change to the Abrolhos reef ecosystems (Teixeira et al., 2019; Duarte et al., 2020; Ferreira et al., 2021).

Our study revealed long-term declines in coastal reef areas of the Abrolhos Bank. These findings help understand the impacts imposed to this important coral reef complex over time, and to build a past scenario of reefs in the region. First, our analysis contrasting a 160-year-old nautical chart with modern layers shows a sharp decline in reef extent; we reveal that the coastal reefs of Abrolhos have lost 28% of their areas. Such loss of coastal reefs has been identified in other parts of the world such as the Great Barrier Reef (Ainsworth et al., 2016), Caribbean (Otaño-Cruz et al., 2019), Indonesia (Lazuardi et al., 2021), and Florida Keys (McClenachan et al., 2017). In Brazil, several studies have mentioned coral reef losses along the coast (Silva et al., 2013), but most were focused on the loss of live coral cover and threatened coral fauna (Ferreira et al., 2021). Therefore, this is the first assessment of historical changes in coral reef extent in Brazil and the observed declines are related to the multiple impacts of direct coral extraction for over a century and recent anthropogenic impacts that have increased sedimentation, siltation, and pollution (Dutra et al., 2006; Silva et al., 2013; Previero and Gasalla, 2018).

The decline in reef area leads to the erosion of biodiversity through changes in the richness, abundance, and distribution of reef organisms, such as fish species (Pratchett et al., 2014). Losses in reef area and habitat may have greater effects on populations of Brazilian endemic species, such as the scleractinian corals that occur in the Abrolhos region (Mussismilia leptophylla, Mussismilia braziliensis, Mussismilia harttii, Mussismilia hispida and Siderastrea stellata), many of which are under extinction risk (MMA, 2018). These changes are associated with the loss of key ecological roles (e.g., predation and herbivory), potentially reducing the resilience of the entire ecosystem (Bellwood et al., 2003). In sum, reef degradation can cause the loss of critical ecosystem functions (Pratchett et al., 2014) and services delivered by reefs to human populations, including fisheries, recreation, tourism, and coastal protection (Spalding et al., 2017). Further, habitat loss threatens local biodiversity, including many fish and coral endemics (Duarte et al., 2020; Ferreira et al., 2021).

Our findings reveal an overall loss of at least 28% in area of coastal reefs in the Abrolhos Bank. These estimates of reef loss reflect the long periods of cumulative coastal impacts recorded from the mid-19th century to the present. Human harvesting activities, such as coral extraction for construction, in addition to increased sedimentation, have possibly caused declines both in reef area and in coral abundance in the region. Our results also offer a window into the past of the Abrolhos Bank reefs, revealing that circa 200 years ago there were larger reef areas with greater coral abundances, especially along the coast. Long-term assessments that include historical sources of information can contribute to obtain adequate baselines and improve our understanding of changes in the reef fauna and seascape, with important implications for conservation strategies and restoration goals. This historical perspective provides an important estimate on how much reef habitat was lost in the Abrolhos region and can be used to set tangible goals for reef restoration. Also, by reconstructing the major impacts responsible for reef decline we provide important information to act on the reduction of such pressures and to prevent the misallocation of restoration efforts in areas with low environmental integrity (i.e., coastal reefs with high sedimentation). Only by looking back into a mostly healthier past we can plan and act towards a healthier future for our reefs.