Terrain variables used for ensemble distribution modelling of vulnerable marine ecosystems indicator taxa on data-limited seamounts of Cabo Verde (NW Africa)

Aim: Seamounts are conspicuous geological features with an important ecological role and can be considered Vulnerable Marine Ecosystems (VMEs). Since many deep-sea regions remain largely unexplored, investigating the occurrence of VME taxa on seamounts is challenging. Our study aimed to predict the distribution of four cold-water coral (CWC) taxa, indicators for VMEs, in a region where occurrence data is scarce.

Location: Seamounts around the Cabo Verde Archipelago (NW Africa).

Methods: We used species presence-absence data obtained from Remotely Operated Vehicle (ROV) footage collected during two research expeditions. Terrain variables calculated using a multiscale approach from a 100 m resolution bathymetry grid, as well as physical oceanographical data from the VIKING20X model, at a native resolution of 1/20°, were used as environmental predictors. Two modelling techniques (Generalized Additive Model (GAM) and Random Forest (RF)) were employed and single-model predictions were combined into a final weighted-average ensemble model. Model performance was validated using different metrics through cross-validation.

Results: Terrain orientation, at broad-scale, presented one of the highest relative variable contributions to the distribution models of all CWC taxa, suggesting that hydrodynamic-topographic interactions on the seamounts could benefit CWCs by maximizing food supply. However, changes at finer scales in terrain morphology and bottom salinity were important for driving differences in the distribution of specific CWCs. The ensemble model predicted the presence of VME taxa on all seamounts and consistently achieved the highest performance metrics, outperforming individual models. Nonetheless, model extrapolation and uncertainty, measured as the coefficient of variation, were high, particularly, in least surveyed areas across seamounts, highlighting the need to collect more data in future surveys.

Main conclusions: Our study shows how data-poor areas may be assessed for the likelihood of VMEs and provides important information to guide future research in Cabo Verde, which is fundamental to advise ongoing conservation planning.

Methods
Terrain variables were derived from a 100 m resolution bathymetry grid, created from a compilation of all available bathymetry data collected by multibeam echosounder (MBES) in the Cabo Verde region. We used an analytical multiscale approach to calculate terrain variables by considering, when possible, different neighbourhood sizes (i.e., number of grid-cells (n)) for calculations. In this study, slope, aspect (converted to eastness and northness), and three types of terrain curvature (plan, profile and mean) were calculated following a Fibonacci sequence of four increasing n values (n = 3, 9, 17, 33) (Dolan et al., 2008). For this, the functions ‘SlpAsp’ and ‘Qfit’ of the “Multiscale DTM” library (Ilich et al., 2023) were used in R Studio. Topographic Position Index (TPI) and Vector Ruggedness Measure (VRM) were calculated at two scales, both fine- and broad-scales (n = 3, 33), using the ‘tpi’ and ‘vrm’ functions, respectively, of the “spatialEco” R Package (Evans and Ram, 2021). Roughness and Terrain Ruggedness Index (TRI) were calculated using the ‘terrain’ function from the “raster” R package (Hijmans et al., 2015), using the default n = 3. Final terrain variables and scales considered in the models were chosen after investigating collinearity between variables (see next section on initial variable selection).

The monthly averages of bottom temperature, bottom salinity and bottom zonal (U) and meridional (V) velocity components for the period of 2009 to 2019 were obtained from a hindcast simulation in the high-resolution VIKING20X ocean general circulation model (VIKING20X-JRA-OMIP described in Biastoch et al., 2021), with a native horizontal resolution of 1/20° (~ 5.3 km). Bottom U and V were converted into mean bottom current speed.

References: 

Biastoch, A., Schwarzkopf, F. U., Getzlaff, K., Rühs, S., Martin, T., Scheinert, M., Schulzki, T., Handmann, P., Hummels, R., & Böning, C. W. (2021). Regional imprints of changes in the Atlantic Meridional Overturning Circulation in the eddy-rich ocean model VIKING20X. Ocean Science, 17(5), 1177–1211. https://doi.org/10.5194/os-17-1177-2021

Dolan, M. F. J., Grehan, A. J., Guinan, J. C., & Brown, C. (2008). Modelling the local distribution of cold-water corals in relation to bathymetric variables: Adding spatial context to deep-sea video data. Deep Sea Research Part I: Oceanographic Research Papers, 55, 1564–1579. https://doi.org/10.1016/j.dsr.2008.06.010

Evans, J. S., & Ram, K. (2021). Package ‘spatialEco.’ R CRAN Project.

Hansteen, T. H., Klügel, A., Kwasnitschka, T. (2014). Cape Verde Seamounts – Cruise No. M80/3 – December 4178  29, 2009 – February 1, 2010—Dakar (Senegal)—Las Palmas de Gran Canaria (Spain). METEOR-Berichte, 4179  M80/3, 42 pp., DFG-Senatskommission für Ozeanographie. https://doi.org/10.2312/cr_m80_3

Hijmans, R. J., Van Etten, J., Cheng, J., Mattiuzzi, M., Sumner, M., Greenberg, J. A., Lamigueiro, O. P., Bevan, A., Racine, E. B., & Shortridge, A. (2015). Package ‘raster.’ R Package, 734, 473.

Ilich, A. R., Misiuk, B., Lecours, V., & Murawski, S. A. (2023). MultiscaleDTM: An open-source R package for multiscale geomorphometric analysis. Transactions in GIS, 27(4), 1164–1204. https://doi.org/10.1111/tgis.13067

Langenkämper, D., Zurowietz, M., Schoening, T., & Nattkemper, T. W. (2017). Biigle 2.0-browsing and annotating large marine image collections. Frontiers in Marine Science, 4, 83. https://doi.org/10.3389/fmars.2017.00083

Orejas, C., Huvenne, V., Sweetman, A. K., Vinha, B., Abella, J. C., Andrade, P., Afonso, A., Antelo, J., Austin-Berry, R., Baltasar, L., Barbosa, N., Barnhill, K. A., Barreiro, A., Bettencourt, R., Blanco, S., Buigues, A., Calado, A., Casal, I., Torre, J. de la, ... Vélez-Belchí, P. (2022). Expedition report iMirabilis2 survey. https://doi.org/10.5281/ZENODO.6352141