Towards automated validation of charted soundings: Existing tests and limitations

Abstract

The nautical chart is one of the fundamental tools in navigation used by mariners to plan and safely execute voyages. Its compilation follows strict cartographic constraints with the most prominent being that of the safety. Thereby, the cartographer is called to make the selection of the bathymetric information for portrayal on charts in a way that, at any location, the expected water depth is not deeper than the source information. To validate the shoal-biased pattern of selection two standard tests are used, i.e. the triangle and edge tests. To date, some efforts have been made towards the automation of the triangle test, but the edge test has been largely ignored. In the context of research on a fully automated solution for the compilation of charts at different scales from the source information, this paper presents an algorithmic implementation of the two tests for the validation of selected soundings. Through a case study with real-world data, it presents the improved performance of the implementation near and within depth curves and coastlines and points out the importance of the edge test in the validation process. It also presents the, by definition, intrinsic limitation of the two tests as part of a fully automated solution and discusses the need for a new test that will complement or supersede the existing ones.

Generalization of Soundings across Scales: From DTM to Harbour and Approach Nautical Charts

“…Nautical chart reliability is of paramount importance for safe navigation, and generalization is a critical stage in the chart compilation process. Automation of such a complicated process is still a significant research issue [3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18]. Many of these processes were based on source data collected using single-beam echosounders or legacy fair sheets.…”

“…Soundings generalization procedures conducted in previous studies typically classify the soundings into groups based on their function [5][6][7] and support the cartographer to select the appropriate soundings. In the cartographic literature, there are a number of research proposals for soundings generalization [5,[8][9][10][11][12][13][14][15][16][17][18]. Additionally, constraints for portrayal of soundings on nautical charts appear in standards published by the IHO [19] and in best practices adopted by HOs [7].…”

Validation of the shoal-biased pattern of bathymetric information on nautical charts

“…This includes an algorithmic implementation of the triangle test that eliminates the false positives of previous implementation efforts near and within linear features (see Figure 1), as well as the first automated implementation of the edge test in the literature. For the above implementations we incorporate the available bathymetric information on the chart, e.g., coastlines (natural and man-made), depth curves, soundings, obstructions, and wrecks (see Kastrisios and Calder 2018;Kastrisios et al 2019).The presented work also illustrates the importance of the edge test in the validation of the charted bathymetric information as it can identify shoals that the triangle test may not identify (see Figure 2), thereby proving that the edge test must not be disregarded by cartographers in the validation process.However, the two tests share the intrinsic limitation of missing discrepancies between the charted and source bathymetric information, even if they are significant (see Figure 3). Thus, a fully automated solution based on a verbatim interpretation of the two tests as written in S-4 does not seem feasible.…”

“…This includes an algorithmic implementation of the triangle test that eliminates the false positives of previous implementation efforts near and within linear features (see Figure 1), as well as the first automated implementation of the edge test in the literature. For the above implementations we incorporate the available bathymetric information on the chart, e.g., coastlines (natural and man-made), depth curves, soundings, obstructions, and wrecks (see Kastrisios and Calder 2018;Kastrisios et al 2019).…”

“…Therefore, we propose a surface-based validation test that will account -at the appropriate charting resolution -for the configuration of the seabed that would be mentally reconstructed by the mariner from the charted bathymetric information. This is expected to lead to a more realistic representation of the submarine relief and its navigationally significant features on charts from the available source information (see Kastrisios et al 2019).…”