In 2021, Kraken Robotics (formerly PanGeo Subsea) was contracted to conduct a Sub-Bottom Imager™ (SBI) survey to support an archaeological campaign on the East coast of the United Kingdom. The SBI was mounted to a Leopard remotely operated vehicle (ROV) to carryout the survey. To develop a windfarm there are a several factors that need to be considered and the presence of historically significant wrecks and objects is important for the purpose of safety and preservation. The SBI data was acquired to support the bathymetry data surrounding an existing windfarm where wrecks and other items were previously positioned. Due to the high currents and mobile sediment in the area many of the previously observed features become buried or partially buried underneath the seafloor, rendering bathymetry and side scan sonar datasets ineffective at accurately measuring the imaged wrecks and objects. The SBI data interpretations were performed using EIVA Navimodel and were completed, on a preliminary basis offshore and then reviewed and reported onshore after the survey completion. The SBI survey resulted in the identification of several archaeological finds including a Roman anchor which is believed to be 1600-2000 years old (Cowie, 2022).

Understanding sedimentation processes in shallow waters is crucial for ensuring safety of navigation, optimizing port operations and effective management of dredging activities. Sedimentation refers to the deposition of sediment particles carried by water currents, which can lead to the gradual shallowing of water bodies and hinder the movement of vessels. This paper explains these processes using a real-world scenario at the Port of Tuxpan, Mexico, an inland port with complex sedimentation. Using nearly 20 years of bathymetry collected with multibeam, singlebeam and densitometer devices, and applying environmental explanatory variables to understand the natural sedimentation process that will allow the best possible forecasting, highlighting its implications for dredging operations, navigation safety, and port operations. Sedimentation processes cause the formation of shoals, sandbars, or siltation, altering the natural navigational channels. By comprehending the sediment dynamics, including sediment transport, erosion, and deposition patterns, authorities can implement appropriate navigation aids, or dredging initiatives, to ensure safe passage for vessels. By understanding the sedimentation processes, port authorities and dredging operators can accurately assess the sediment deposition rates, identify high-risk areas prone to sedimentation, and develop effective dredging strategies to remove accumulated sediments. Ports rely on maintaining adequate depths for berthing large vessels, facilitating efficient operations. Understanding sedimentation processes assists port operators in predicting and managing sediment-related challenges, allowing port authorities to schedule dredging maintenance activities, optimize dredging cycles, and allocate resources effectively, minimizing disruptions to port operations. It’s important to expedite this information availability to stakeholders, such as Port Pilots, Port Operators and Mariners. This paper will present three methods for forecasting sedimentation processes, applying GIS tools and providing guidance to determine when, where and how much to dredge, making this process more efficient, highlighting how crucial this is for ensuring the efficient and sustainable functioning of ports, maintaining navigation safety, and preserving marine ecosystems.

In the rapidly evolving hydrospatial sector of the blue economy, projected by Canada’s Ocean Supercluster to nearly double Canada’s Gross Domestic Product (GDP) by 2030 and reach approximately 3 trillion dollars annually, the efficient maintenance of port infrastructures has emerged as a primary concern. Conventional inspection methods, whether diver-conducted or sensor-based, are time-consuming, costly, and often targeted and limited. To address this challenge, a multi-institutional collaboration led by the Interdisciplinary Center for the Development of Ocean Mapping (CIDCO), encompassing academic and industrial partners, has been established. This initiative aims to enhance efficient port infrastructure maintenance and has led to the development of the Hydrospatial Dual-sight Inspector (HDI), an innovative, semi-supervised, autonomous marine vehicle. The HDI employs a Seafloor ASV EchoBoat, integrating advanced sonar and Light Detection and Ranging (LiDAR) systems. This integration enhances vulnerability and safety risk assessment in port structures, significantly reducing operational costs and time. The operation requires simultaneous digitization of emerged and submerged surfaces with high-density LiDAR Velodyne and R2 Sonic 2020 Multibeam Echosounder, enabling a more comprehensive assessment of surface degradation and verticality issues. Field tests conducted at various Canadian ports, including Rimouski, Québec, and Montréal, have successfully demonstrated the integration and functionality of the HDI’s sensor systems. Collaborative work with Laval University focuses on refining calibration and quality control algorithms for the LiDAR and echosounder data, showing promising results. The project has also initiated the development of CIDCO-specific open-source software to enhance independence, transparency, and security within the hydrospatial domain. Future work will focus on the smart navigation of the HDI, implementing computer-assisted navigational strategies to optimize survey efficiency and ensure collision avoidance. Integration into the open-source Robot Operating System (ROS) platform is planned, expected to significantly contribute to the marine robotics community and strengthen port infrastructure inspection methodologies in the dynamic blue economy landscape.

We introduce a pioneering methodology that bridges the gap between theory of complex systems dynamics, data-driven detection and understanding of coherent ocean variability over a range of spatiotemporal scales, and next generation technologies for nonparameteric predictions in a changing world. Leveraging operator-theoretic techniques originally designed for non-autonomous systems, our approach enables the identification of trends and persistent cycles in climate data using a single trajectory of the system. Drawing inspiration from the remarkable properties of eigenfunctions of Koopman and transfer operators, we showcase their prowess in capturing nonlinear trends and coherent modes of internal variability. Our real-world applications to climate data reveal the potential of this methodology to unveil the hidden patterns within climate phenomena. In particular, we will explore how operator-theoretic techniques serve as a powerful lens for dissecting complex climate dynamic. We will explain how the techniques initially designed for autonomous systems have been adapted by us to analyze the non-autonomous nature of climate behavior. In the end, we will provide insights into our discoveries regarding sea surface temperature (SST) variability over the industrial era and impact of trend on internal ocean variability.

The identification and characterization of underwater dunes from modelled bathymetric surfaces represents an important effort to assess the mutual relationship between sediment dynamics and dune morphology. Different authors have proposed automatic approaches to identify these sedimentary structures (Debese et al., 2016; Di Stefano and Mayer, 2018). Cassol et al. (2021, 2022) have proposed an automatic approach for segmentation and characterization of dunes in the estuarine context of the Saint-Lawrence River and modelled by a regular gridded DBM. This approach considers the identification of the morphological units of the dunes (i.e. crest line, lee and stoss side) using the Geomorphon (Jasiewicz and Stepinski, 2012) algorithm, mathematical morphology and image processing operations. The bathymetric surface is then segmented into dune objects, matching each crest line with its respective troughs. Dune morphological descriptors can be estimated from these identified objects. Based on the BRESS method by Masetti et al. (2018), this presentation presents the results of a multisource approach to automatically characterize dunes considering the DBM and backscatter strength mosaic, both derived from measurements made with multibeam echosounder. The approach is based on the method developed by Cassol et al. (2022) allowing the automatic extraction of dunes and its characteristics from a DBM and an exploratory work of backscatter strength characterization of dune fields in the Belgium part of the North Sea (Roche et al., 2023).