Predicting suitable habitat for the cold-water coral Lophelia pertusa (Scleractinia)

Abstract

Ecological-niche factor analysis (ENFA) was applied to the reef framework-forming cold-water coral Lophelia pertusa. The environmental tolerances of this species were assessed using readily available oceanographic data, including physical, chemical, and biological variables. L. pertusa was found at mean depths of 468 and 480 m on the regional and global scales and occupied a niche that included higher than average current speed and productivity, supporting the theory that their limited food supply is locally enhanced by currents. Most records occurred in areas with a salinity of 35, mean temperatures of 6.2–6.7  °C and dissolved oxygen levels of 6.0–6.2 ml l⁻¹. The majority of records were found in areas that were saturated with aragonite but had low concentration of nutrients (silicate, phosphate, and nitrate). Suitable habitat for L. pertusa was predicted using ENFA on a global and a regional scale that incorporated the north-east Atlantic Ocean. Regional prediction was reliable due to numerous presence points throughout the area, whereas global prediction was less reliable due to the paucity of presence data outside of the north-east Atlantic. However, the species niche was supported at each spatial scale. Predicted maps at the global scale reinforced the general consensus that the North Atlantic Ocean is a key region in the worldwide distribution of L. pertusa. Predictive modelling is an approach that can be applied to cold-water coral species to locate areas of suitable habitat for further study. It may also prove a useful tool to assist spatial planning of offshore marine protected areas. However, issues with eco-geographical datasets, including their coarse resolution and limited geographical coverage, currently restrict the scope of this approach.

Introduction

Several species of cold-water corals are significant autogenic ecological engineers able to form large reef frameworks that provide habitat for many other organisms (Rogers, 1999; Freiwald et al., 2004; Roberts et al., 2006). These reef structures can be over 40 m tall and may extend for several kilometres, yet the environmental factors controlling their occurrence remain poorly understood. Cold-water coral ecosystems are now receiving increased attention, with new reefs being discovered and scientific understanding improving (Roberts et al., 2006). Here, we use predictive habitat mapping to examine the physical environmental factors that determine the distribution of the main reef-forming cold-water coral, Lophelia pertusa (Linnaeus, 1758), and to produce maps of suitable habitat at both regional and global scales.

There are six main reef framework-forming coral species in deep waters (Freiwald et al., 2004). The most widespread is L. pertusa, which forms bush-like colonies that can measure several metres across consisting of many thousands of coral polyps (Freiwald et al., 2004). Over time, continual coral growth can produce large reef structures often dominated by L. pertusa but also containing other secondary framework-constructing Scleractinia of the genera Madrepora, Desmophyllum, Oculina, Goniocorella and Solenosmillia (Rogers, 1999; Tyler and Zibrowius, 1992). The classic definition of a reef is a submerged structure rising from the surrounding seafloor that forms a hazard to shipping (Wood, 1999). However, most records of L. pertusa originate from deeper waters in high latitudes in the north-east Atlantic and have a cosmopolitan distribution throughout the Atlantic Ocean (Zibrowius, 1980). Here, we use the term reef to describe biogenic structures formed by L. pertusa frameworks that alter sediment deposition, provide complex structural habitat, and are subject to the processes of growth and (bio)erosion (Rogers, 1999; Roberts et al., 2006).

There are two main hypotheses for the major environmental drivers of L. pertusa distribution. Most observations of cold-water corals have been in areas with accelerated currents, such as on sloping topography and topographic highs (Dons, 1944; Strømgren, 1971; Genin et al., 1986; Frederiksen et al., 1992). Currents are thought to concentrate food supply to certain areas (“current acceleration hypothesis” Mortensen et al., 2001; Thiem et al., 2006; Kiriakoulakis et al., 2007). On the reef, strong currents may also limit smothering by reducing sediment deposition (White et al., 2005). On some sloping topography, breaking internal waves can resuspend organic matter or mix surface waters, which also results in increased food supply (Frederiksen et al., 1992). A contrasting hypothesis suggests that the occurrence of cold-water coral reefs is directly associated with the seepage of light hydrocarbons through the seafloor (“hydraulic theory” Hovland and Thomsen, 1997; Hovland and Risk, 2003). Yet, even though some reefs have been found close to seeps and pockmarks, most are not (Roberts et al., 2006). Recent drilling of the Challenger Mound showed no evidence of mound genesis or growth being related to seepages (Williams et al., 2006), and isotopic analyses of L. pertusa do not indicate a seep-based food chain (Spiro et al., 2000; Duineveld et al., 2004).

Most cold-water coral reefs are restricted to intermediate depths. L. pertusa reefs have been recorded most frequently between 200 and 400 m (Dons, 1944), at temperatures between 4 and 12 °C (Freiwald et al., 2004; Roberts et al., 2006) with oceanic salinity of around 35 on the practical salinity scale (Freiwald, 2002; Freiwald et al., 2004). Some reefs are associated with iceberg ploughmarks on the continental shelves of the north-east Atlantic where, during the Pleistocene, the passage of icebergs reworked seafloor sediment and deposited glacial dropstones. This glacial activity increased the amount of hard substrata available for colonisation, which is thought to have been an important factor in the development of large L. pertusa reefs on the Norwegian Margin (e.g., Sula Ridge; Freiwald et al., 1999). However, not all occurrences of L. pertusa rely upon large amounts of hard substrata. Settlement may also occur on small cobbles from which corals grow over time (Wilson, 1979) or even on man-made structures such as oil rigs (Gass and Roberts, 2006). At a larger scale, ocean chemistry may control global distribution patterns. The distribution of scleractinian cold-water corals appears to be strongly related to the depth of the aragonite saturation horizon (ASH). Guinotte et al. (2006) found that 95% of modern framework-forming cold-water coral records were in the Atlantic Ocean, which has a much deeper ASH (>2000 m) than the Pacific where the ASH is relatively shallow (50–600 m). Clark et al. (2006) also found aragonite to be a major factor in the distribution of scleractinian corals on seamounts. Currently, little is known about the effect of lowered aragonite concentration on cold-water corals, but experiments on tropical reef-builders show reduced calcification rates at low carbonate ion availability (Langdon et al., 2000, Langdon et al., 2003; Marubini et al., 2003).

The geographical distribution of cold-water corals will be affected by a combination of different physical, chemical, and biological factors. Predictive modelling approaches can be used to generate habitat suitability maps, producing representations of geographical areas that match species’ requirements based on multiple environmental parameters (Guisan and Zimmermann, 2000; Hirzel et al., 2002). Cold-water corals are ideally suited for predictive modelling approaches because of their sessile nature and longevity. However, the majority of predictive mapping approaches (e.g., General Linear Modelling and Principal Component Analysis) require information on both the presence and absence of species (Guisan and Zimmermann, 2000). In the deep sea, absence data is often unavailable or unreliable. The majority of research expeditions are targeted towards areas with known coral occurrences, and sampling methodologies often vary between expeditions. The patchiness of deep-sea habitats also limits confidence for assessing the absence of cold-water corals, which could be easily missed because sampling methods have a limited spatial coverage.

Ecological-niche factor analysis was developed to address both the paucity and unreliability associated with absence data in predictive modelling (ENFA; Hirzel et al., 2002). This technique assumes that a given species has a non-random distribution within an eco-geographical variable and that the majority of individuals would occupy the optimal range. By using only presence data, ENFA calculates habitat suitability based on the niche width of a species in relation to eco-geographical variables, thereby producing useful statistics related to the species niche and predicted habitat suitability over large spatial areas (Hirzel et al., 2002; Hirzel and Arlettaz, 2003). This approach has been used in several terrestrial studies (Zaniewski et al., 2002; Hirzel and Arlettaz, 2003; Reutter et al., 2003; Hirzel et al., 2004; Santos et al., 2006) but to date, only in four deep-sea studies: two of those studies investigated the distribution of gorgonian corals (Leverette and Metaxas, 2005; Bryan and Metaxas, 2007), the third study evaluated the distribution of cold-water corals on seamounts (Clark et al., 2006), and the fourth derived eco-geographical data from multibeam bathymetry that used ENFA to determine habitat suitability for squat lobsters (Wilson et al., 2007).

Here we applied ENFA to the most cosmopolitan framework-forming coral species L. pertusa. Using the latest species presence data and a suite of eco-geographical variables, we determined key habitat and niche widths and produced habitat suitability maps for two spatial scales. We then discuss those results in light of the major environmental determinants that define the distribution of L. pertusa and describe the limitations of applying ENFA to deep-sea species.