The relationship between 3D terrain complexity and fine-scale localization and distribution of species is poorly understood. Here we present a very fine-scale 3D reconstruction model of three zones of circalittoral rocky shelf in the Bay of Biscay. Detailed terrain variables are extracted from 3D models using a structure-from-motion (SfM) approach applied to ROTV images. Significant terrain variables that explain species location were selected using general additive models (GAMs) and micro-distribution of the species were predicted. Two models combining BPI, curvature and rugosity can explain 55% and 77% of the Ophiuroidea and Crinoidea distribution, respectively. The third model contributes to explaining the terrain variables that induce the localization of Dendrophyllia cornigera. GAM univariate models detect the terrain variables for each structural species in this third zone (Artemisina transiens, D. cornigera and Phakellia ventilabrum). To avoid the time-consuming task of manual annotation of presence, a deep-learning algorithm (YOLO v4) is proposed. This approach achieves very high reliability and low uncertainty in automatic object detection, identification and location. These new advances applied to underwater imagery (SfM and deep-learning) can resolve the very-high resolution information needed for predictive microhabitat modeling in a very complex zone.

In the present work we focus on the distribution of two species of sponges. One of these is Asconema setubalense, a sponge found in rocky substrate that was sampled with a photogrammetric vehicle through georeferenced images. The other is Pheronema carpenteri, which inhabits soft bottoms and was sampled by beam trawl. For the spatial distribution modeling of both sponges, the geomorphological variables of depth, slope, broad and fine scale bathymetric position index (BPI), aspect, and types of bottoms were used, all with a resolution of 32 m. Additionally, layers of silicates and currents near the bottom were extracted from Copernicus Marine Environment Monitoring Service (CMEMS), with a resolution of ∼4 and ∼9 km, respectively. Due to the low resolution of the layers, it was considered necessary to validate their use by model comparison, where those that included these variables turned out to be more explanatory than the others. The models were developed in a complex continental break of the Central Cantabrian Sea, which comprises several submarine canyons and a seamount (Le Danois Bank). On the one hand, a very high resolution (32 m) spatial distribution model based on A. setubalense presence was developed using the MaxEnt maximum entropy model. On the other, depending on the availability of density data, generalized additive models (GAMs) were developed for P. carpenteri distribution, although in this case the sampler only allowed a maximum resolution of almost 1 Km. For the A. setubalense, the variables that best explained their distribution were ground types and depth, and for P. carpenteri, silicates, slope, northness, and eastward seawater velocity. The final model scores obtained were an AUC of 0.98 for the MaxEnt model, and an R squared of 0.87 for the GAM model.

Gorgonians play a fundamental role in the deep sea (below 200 m depth), composing three-dimensional habitats that are characterized by a high associated biodiversity and playing an important part in biogeochemical cycles. Here we describe the use of a benthic lander to monitoring polyps activity, used as a proxy of gorgonian feeding activity of three colonies of Placogorgia sp. Images cover a period of 22 days with a temporal resolution of 30 min. In addition, this seafloor observatory is instrumented with oceanographic sensors that allows continuous monitoring of the hydrographic conditions in the site. Deep-learning is used for automatic detection of the state of the polyps registered in the images. More than 1000 images of 3 large specimens of gorgonians are analyzed, annotating polyps as extended or retracted, using the semantic segmentation algorithm ConvNeXt. The segmentation results are used to describe the feeding patterns of this species. Placogorgia sp. shows a daily pattern of feeding conduct, depending on the hours of day and night. Using a Singular Spectrum Analysis approach, feeding activity is related to currents dynamics and Acoustic Doppler Current Profile (ADCP) return signal intensity, as proxy of suspended matter, achieving a linear correlation of 0.35 and 0.11 respectively. This is the first time that the behavior of the Placogorgia polyps, directly related to their feeding process, is described.