As the number of Earth observation satellites has increased in recent years, satellite-derived bathymetry (SDB) methods have evolved significantly, each with its own disadvantages and advantages. Traditional methods analyze the water-leaving spectral radiance or exploit photogrammetry. An alternative method analyzes the propagation of the waves towards the coastline. As it is not necessary to observe the seabed to provide a water depth estimation, this method can be applied in deeper areas or in areas of higher turbidity than the traditional methods. The aim of this study is to evaluate the benefits of this method compared to a more traditional one. This study addresses this issue by evaluating the uncertainty, quantified by the 95% confidence interval, of the estimation of water depth estimation by wave observation. Water depth is inferred from the wave dispersion relation. We assume that the conditions for applying this relationship are met (negligible current, wave deformation linked only to water height, sufficient wave period) and that the waves are heading towards the coast. The proposed methodology is based on the framework of the estimation theory, and therefore on the likelihood function. The core information is the wave power spectral density and its evolution from the open sea to the coast. This evolution being depth-dependent, we manage to define the likelihood function which allows confidence interval assessment. This formalism has been applied to real data, and makes it possible to analyze different sources of noise that affect the estimation of water depth. Two sources of noise were studied: the noise of the image, and the variability of waves in the open sea. The likelihood framework appears to be a powerful tool for quantifying the impact of sensor quality as well as environmental conditions on the accuracy of water depth estimation.

This technical presentation intends to demonstrate and discuss the use of SAR Remote Sensing for identification and extraction of shorelines, intertidal extents/zones in arctic areas like Canada and Greenland. Climate change has been a major concern globally in all aspects of the environmental realms, especially the loss/changes in Arctic sea ice concentrations, thereby impacting coastal erosion rates and a point of concern for coastal communities. On the other hand , the diminishing sea ice is opening new opportunities for shipping, oil and gas exploration, tourism, and other economic activities. Thus, mapping updated shorelines and intertidal zone extents on nautical charts, especially in the remote arctic areas can be very beneficial for the hydrographic community. Along with existing in-situ and ex-situ methods, SAR or Radar remote sensing can be very useful for coastal area feature mapping, (especially in Arctic/sub-arctic regions that have extreme weather conditions, less/varying daylight time, etc.) due to its all weather capability, day and night operation (independent of solar illumination), as well as open availability of SAR datasets through various space programs, and commercial sensors like Capella SAR that have a sub-daily revisit time and very high spatial resolution. It can be an added asset for continuous surveillance, monitoring, and decision making in coastal areas .Moreover,sensitivity of SAR to changes in dielectric properties of features and surface roughness , allow a clear/effective demarcation between land and water boundaries, identification of bright objects in ocean waters, and changes occurring during HT and LT respectively. TCarta’s current methodology makes use of the above SAR properties to allow “tidally synced” identification and extraction of shorelines (High Tide Line/Low Tide Line), intertidal zone and other coastal features like shoals and rocks (observed during varying tidal ranges). While this workflow can be adapted for any SAR sensor, the current work utilizes Capella SAR and Sentinel-1 SAR imagery. The process broadly involves a) selection of SAR images in closest sync to a specific tidal range information, for a chosen AOI; b) SAR image preprocessing; c) Image thresholding based on backscatter or intensity values ; d) Segmentation & Edge extraction processes for final coastal feature extractions (shoreline, intertidal zones, rock features, etc). Results, advantages, limitations, and way forward of the above method of SAR based intertidal zone/extent extraction will be shared and discussed in this presentation.

NASA’s ICESat-2 satellite has been orbiting the earth since 2018. Originally deployed for monitoring glaciers and sea ice, it is found that the green laser mounted on the satellite has the ability to penetrate the water column, allowing the extraction of shallow water bathymetry soundings from the ATL03 global geolocated photon data product. With an accuracy of about ~50 cm and an orbit cycle of 91 days, the ICESat-2 bathymetry soundings can be used to derive frequent and broader information on coastal bathymetry at a global scale. These ICESat-2 soundings can then be used to train/validate Satellite Derived Bathymetry (SDB) in order to obtain a contiguous map of the coastal regions. However, geographical differences and variable water constituents create challenges for automated bathymetric extraction methods, especially in near-arctic regions such as in Newfoundland. ATL03 bathymetric photon returns in the arctic regions are often sparser than their tropical counterparts, making it challenging for data extraction. Furthermore, as the region is mostly covered with ice, there is only a small temporal window to consider for SDB image acquisition, limiting the collection to summer months when there is less ice blocking seafloor visibility. With the growth of the marine industry, there is a need for easy-to-access higher resolution bathymetric data for coastal navigation. In this study, we detail the challenges posed in the arctic environment for SDB and methods to overcome these difficulties. We detail the extraction of ICESat-2 bathymetry soundings, differences with other regions in the tropics, and the SDB methods using the soundings to produce a map of the shallow water coastal area in near-arctic regions. The project is a collaboration between TCarta and Marine Institute to educate students in marine remote sensing applications and data processing. The result of this collaborative effort is a suite of methods and a higher-resolution coastal map of the whole of Newfoundland, with a total area of 2733.64 km2. With this data, we hope to provide better understanding in coastal navigation of arctic waters.

Background Satellite derived bathymetry (SDB) provides a more comprehensive coverage especially in hard to reach areas where acoustic surveying methods may be difficult to complete. SDB production can be used as a cost effective solution for rapid and frequent data production. Multibeam Echosounder (MBES) bathymetry provides highly accurate, high resolution bathymetric data. However, multibeam data collection is costly, time consuming, and restricted to areas which are accessible to survey vessels. Since MBES bathymetry is known for being highly accurate, SDB values will be compared to multibeam bathymetry in the area of Holyrood Bay, Newfoundland, Canada. Objectives The primary objective of this project was to assess the accuracy of SDB compared to MBES bathymetry in depths less than twenty meters in Holyrood Bay, Newfoundland, Canada. Methods Raw data required to produce SDB including satellite imagery and in-situ data were obtained from TCarta Marine. TCartas’ Trident Toolbox was used within ArcGIS Pro to create a ten meter raster bathymetric surface. MBES data was collected with the aid of the Marine Institute in Holyrood Bay, Newfoundland, Canada. CARIS HIPS and SIPS were used to process the raw MBES data and create a ten meter bathymetric surface. Root mean square error (RMSE) statistics were derived for several depth intervals to allow for an understanding of which depths SDB is able to estimate accurately. MBES bathymetry was used as the true value of depths within the area and SDB data was statistically compared to the multibeam bathymetric values. The Canadian Hydrographic Service (CHS) standards were used to assess the accuracy. Results The anticipated results of this project include the difference between each survey depth and SDB value, as well as the RMSE value for several depth intervals (0-1m, 1-5m, etc). Ideally, each RMSE value will be below 1.5m. Discussion This project was limited by the opportunity to collect MBES data and satellite imagery on the same date. Ideally, satellite imagery and multibeam data would be collected on the same day, however, due to scheduling constraints this was not possible. The closer each RMSE value is to zero will represent higher quality data, however, anything below 1.5m will be acceptable for this study. Conclusion This project takes a comprehensive approach to determining the accuracy of SDB in water depths less than twenty meters. SDB works well along coastal areas which are difficult to travel by vessel. So, it is expected that SDB performs well in those areas.

The Government of Canada's Oceans Protection Plan launched a five-year initiative called Community Hydrography (2022-2027). This program empowers coastal communities to gather and utilize bathymetric data for various local purposes, including marine safety enhancement, community planning, undersea hazard identification, sensitive marine environment detection, and fishing and harvesting activities. In its inaugural year, the initiative has already involved 12 communities collecting bathymetric data in their respective marine environments. Most of these communities operate in coastal waters where bathymetric data is either lacking or sometimes nonexistent. Over the past two years, remarkable technological advancements have automated Satellite-Derived Bathymetry (SDB) methods and workflows, primarily used for mapping shallow waters. This has resulted in the latest web application, SDB-Online, which facilitates not only mapping but also the monitoring of coastal shallow water zones. This approach, hosted on an AWS cloud environment, is linked directly to the European Copernicus satellites' data archives and is applicable to more than 80% of Canada's coastal waters, providing bathymetric data from the shoreline down to water depths of 10-12m. To meet community needs, both community-led bathymetric surveys and SDB generated via SDB-Online were combined in the Anguniaqvia Niqiqyuam Marine Protected Area (NWT). This previously unsurveyed site now boasts coverage of 180 km² (80km2 of SDB and 100km2 of MBES) of new bathymetric data. This presentation will demonstrate, through live records, how SDB data have been created, integrated with community-led bathymetric surveys, and converted into "Community Maps". The scalable technology applied in these areas can be utilized in other Canadian waters to enhance safety and support local communities highly dependent on coastal ecosystems.