Marine environments are highly heterogeneous, varying across scales of a few meters to entire ocean basins. Understanding the relationship between environmental variability and species distribution is essential for area-based management and conservation. However, this requires a precise alignment of seabed mapping, environmental and biological sampling, which is often difficult to achieve in the deep sea. There is thus an urgent need to tackle this challenge to effectively manage high-diversity habitats such as deep-sea coral and sponge (DSCS) aggregations. Relying on multiple subsea platforms, seafloor mapping and imaging techniques, we mapped the distribution of megafaunal communities at Sur Ridge (800-1250 m depth; off Central California) across multiple spatial scales. First, remotely operated vehicle (ROV) video transects were conducted to characterize community distribution along the ridge, in relation to substratum type, environmental conditions and 1-m scale bathymetry. Five distinct communities, located in specific areas of the ridge were identified. These communities were primarily structured by depth, availability of hard substratum and terrain complexity (slope and rugosity). Indicator taxa were identified for each community and their distribution was characterized at the cm scale from co-registered 5-mm resolution photomosaic and 5-cm lateral resolution bathymetry produced during low altitude ROV surveys. High-resolution mapping allowed the identification of associations between DSCS and other benthic taxa and showed that, even at these small scales, different taxa associate with distinct microhabitats. These results highlight the importance of accounting for habitat heterogeneity, and its role in supporting biodiversity, when designing management and conservation strategies.

All data were collected at Sur Ridge, a high-relief topographic feature ~35 km west of Point Sur, California.

Data on megafaunal densities and seafloor substratum type were collected from remotely operated vehicle (ROV) video transects conducted throughout Sur Ridge between July 2014 and December 2021 (ROV Doc Ricketts deployed from the R/V Western Flyer). The ROV was equipped with two parallel laser beams (29 cm apart) to estimate the width of the field of view during transects. A total of 58 video transects (each 50 m long) were annotated using MBARI’s video annotation and reference system (VARS; Schlining and Stout 2006). All species were identified to the lowest taxonomic level possible based on morphology and sorted into morphotypes. Megafaunal densities were then estimated based on transect area. Substratum types were estimated from 40 frame grabs per video transect, which were captured at regular intervals using GOM player (version 2.3.80.5345). Frame grabs were analyzed with PAPARA(ZZ)I v. 2.8 software application (Marcon and Purser, 2017): One point was randomly generated on each image and classified in one of 11 substratum categories based on Wentworth (1922) and MBARI’s Deep-Sea Guide (Jacobsen Stout et al. 2016): bedrock, rocky outcrop, boulder, eroded surface, cobble, pebble, gravel, biogenic, coarse sand, sand and mud. The proportion composition of each substratum type was then calculated for all transects and included in a substratum composition matrix.

Different physical and environmental variables were considered to understand the distribution of megafaunal communities on Sur Ridge:

-          Pressure (depth), salinity (conductivity), temperature and dissolved oxygen were recorded by sensors mounted on the ROV while transecting. These environmental variables were averaged for each transect.

-          The first and second Principal Component scores obtained from the principal component analysis performed on the Hellinger-transformed substratum composition matrix were used to evaluate the effect of substratum type on megafauna diversity patterns.

-          Terrain variables, including rugosity, slope, curvature, eastness, northness, broad- and fine-scale standardized benthic position indexes were extracted from the 20-m resolution bathymetry of Sur Ridge using the BTM ArcGIS toolbox (Walbridge et al. 2018). Note: While Sur Ridge was mapped comprehensively at a 1-m resolution using MBARI Dorado class Autonomous Underwater Vehicles, the bathymetry was subsequently downscaled to 20 m to account for ROV navigation uncertainties associated with the use of USBL positioning during transects.

High-resolution (cm resolution) data on megafauna distribution in relation to terrain were collected using MBARI’s Low Altitude Survey System (LASS). An area of ~ 100 x 100 m on the northern slope of Sur Ridge was mapped on October 13, 2020 using the LASS operated from the ROV Doc Ricketts. The LASS toolsled combined color stereo still cameras illuminated by strobe lights, and multibeam sonar for producing co-registered 5-mm resolution photomosaic and 5-cm resolution bathymetry, respectively. Megafauna visible on the georeferenced photomosaic produced with the LASS were annotated as point shapefiles in ArcMap V10.8.2. Spatial coordinates of all observations were recorded. Terrain parameters including slope, curvature, rugosity, northness, eastness, fine- and broad-scale standardized benthic position indexes were extracted from the 5-cm resolution bathymetry using the BTM ArcGIS toolbox. Terrain parameter values associated with each megafauna annotation were extracted and used in the analyses.

Nearest-neighbor analysis was performed in ArcMap to measure the distance between megafauna (other than deep-sea corals and sponges) and the closest coral and sponge.

References:

Jacobsen Stout, N., L. Kuhnz, L. Lundsten, B. Schlining, K. Schlining, and S. von Thun. 2016. The Deep-Sea Guide (DSG). Monterey Bay Aquarium Research Institute (MBARI), Moss Landing, California, USA.

Marcon, Y., and A. Purser. 2017. PAPARA(ZZ)I: An open-source software interface for annotating photographs of the deep-sea. SoftwareX 6: 69–80. doi:10.1016/j.softx.2017.02.002

Schlining, B. M., and N. J. Stout. 2006. MBARI’s Video Annotation and Reference System. OCEANS 2006. 1–5.

Walbridge, S., N. Slocum, M. Pobuda, and D. J. Wright. 2018. Unified Geomorphological Analysis Workflows with Benthic Terrain Modeler. Geosciences (Basel) 8:94. doi:10.3390/geosciences8030094

Wentworth, C. K. 1922. A Scale of Grade and Class Terms for Clastic Sediments. J Geol 30: 377–392. doi:10.1086/622910