Estimation of Location Names

Estimation of location names from geographic coordinates

Overview

A common challenge in oceanographic datasets is missing or inconsistent location metadata. While sediment core data usually come with coordinates (latitude and longitude), fields such as the Nearest Country and the Nearest Municipality are often left blank or recorded inconsistently.

To ensure consistency across the dataset, we algorithmically estimated these fields based on the core’s location. This page outlines the data sources and geospatial methods used to populate the country, municipality, and sea_name columns in the Core table.

Note on Accuracy

These values are algorithmic estimates based on geometric proximity. They may differ from historical logbooks or specific political definitions of maritime borders.

Nearest Country

Since sediment cores are taken underwater, they do not technically belong “inside” a country’s land border. To identify the associated nation, we used a “Nearest Neighbor” analysis.

Concept: For every point in the ocean, we calculated the straight-line distance to the coastlines of all surrounding nations. The country with the shortest distance was assigned to the core.
Data Source: We used the Natural Earth dataset (via the rnaturalearth R package). This provides a standardized, moderate-resolution map of global country boundaries suitable for identifying the nearest sovereign nation.

Nearest Municipality

For higher-resolution analysis, identifying the country is not enough. We also estimated the Nearest Municipality (Local Administrative Unit) to provide granular context.

Concept: Similar to the country estimation, this uses a “Nearest Neighbor” approach but requires much more detailed maps that include islands and complex coastal administrative borders.
Data Source: We utilized the Eurostat GISCO database (via the giscoR R package). This provides official administrative boundaries for European countries down to the municipal level (LAU - Local Administrative Units).

Sea & Ocean Names

To determine which body of water a core was taken from (e.g., “Barents Sea” vs. “Norwegian Sea”), we used the “Point-in-Polygon” method.

Concept: Imagine drawing official boundaries on a map. We simply check which boundary “box” each sediment core falls inside.
Data Source: We utilized the International Hydrographic Organization (IHO) Sea Areas v3. This is the globally accepted standard for defining the limits of oceans and seas.
Method: The specific boundaries were downloaded directly from the Marine Regions estimation service. If a core was located slightly “inland” due to map resolution issues (common in narrow fjords), it was assigned to the nearest adjacent sea body.

Summary of Data Sources

Field	Estimation Method	Data Source	R Package / Tool
Nearest Country	Nearest Neighbor	Natural Earth	`rnaturalearth`
Municipality	Nearest Neighbor	Eurostat GISCO	`giscoR`
Sea Name	Spatial Join	IHO Sea Areas v3	`sf` (Direct WFS)