Global datasets describing the coastline and the coastal zone.
ecu)The world's ocean coastline divided into 4,005,842 segments of 1 km or shorter (median about 480 m). Each segment is described by ten variables covering the adjacent water, the adjacent land and the shoreline itself. Produced by the U.S. Geological Survey with Esri and the Marine Biodiversity Observation Network.
CSU_ID, 80,997 distinct), following the Coastal and Marine Ecological Classification Standard (CMECS). Units are grouped into 16 global coastline groups (Cluster), which the source numbers but does not name.REGIONAL_SINUOSITY = 0 (591,757 segments) means no value was computed; exclude it before averaging.The H3 hex table has one row per (segment, cell) pair, so deduplicate on _cng_fid before summing LENGTH_GEO.
The PMTiles source-layer is ecu.
Sayre, R., 2023, Global Ecological Classification of Coastal Segment Units, U.S. Geological Survey data release, https://doi.org/10.5066/P9HWHSPU. Method: Sayre et al. 2021, Oceanography 34(2), https://doi.org/10.5670/oceanog.2021.219.
Public domain (U.S. Government work). Retrieved from ScienceBase on 2026-09-28; the original package is kept at s3://public-coastal/raw/USGSEsriGlobalCoastalSegmentsv1.mpk (2,075,928,715 bytes, MD5 ad4f54e071c343ccbd574f4cff0adf36).
meow-coastal-band)A coastal strip labelled with Marine Ecoregions of the World (MEOW) ecoregion, province and realm: everything within 50 km of the coastline on the sea side and 20 km on the land side. MEOW ecoregion polygons reach far inland and about 200 nautical miles out to sea, so this band is the coastal part of each ecoregion. It comes in two forms:
dist_km), sea or land (side), depth or elevation. Use it for joins with other H3 layers and for choosing any strip width.ECO_CODE, ECOREGION, PROVINCE, REALM, side, dist_min_km, dist_max_km and area_km2. Each class includes its upper edge, so keeping features with dist_max_km <= 10 gives exactly the cells with dist_km <= 10.The PMTiles source-layer is meow-coastal-band.
Derived from MEOW (Spalding et al. 2007; The Nature Conservancy / WWF, CC-BY-NC-3.0), so it carries MEOW's NonCommercial terms. The coastline is the USGS Ecological Coastal Units shoreline (above); land and sea come from the Copernicus GLO-90 Water Body Mask, depth from GEBCO 2025 and elevation from Copernicus DEM GLO-90 (produced using Copernicus WorldDEM-90 © DLR e.V. 2010-2014 and © Airbus Defence and Space GmbH 2014-2018 provided under COPERNICUS by the European Union and ESA).
tidal-marsh-2020)Tidal marshes, the vegetated coastal wetlands flooded by the tides, mapped for 2020 at 10 m between 60 degrees north and 60 degrees south (Worthington et al. 2024, version 2.6).
tidal_marsh_km2 and tidal_marsh_fraction; a missing cell means zero.The rasters render through the NRP titiler service; the colormap draws value 1 and leaves 0 and 255 transparent.
Worthington, T. A., Spalding, M., Landis, E., Maxwell, T. L., Navarro, A., Smart, L. S., & Murray, N. J. (2024). The distribution of global tidal marshes from Earth observation data. Global Ecology and Biogeography, 33(8), e13852, https://doi.org/10.1111/geb.13852. Data: Zenodo record 8420753, version 2.6, downloaded 2026-10-25; the original archive is kept at s3://public-coastal/raw/tidal-marsh-2020/tidal_marsh_v2_6.zip.
Creative Commons Attribution 4.0.
tidal-flats-v1-2, tidal-flats-v1-1)Tidal flats, the mud, sand and rock flats exposed at low tide, mapped at 30 m for three-year periods between 60 degrees north and 60 degrees south (Murray et al.). Two collections, because the producers' two versions use different mapped areas and should not be joined into one series:
tidal-flats-v1-2: version 1.2, seven periods from 1999-2001 to 2017-2019.tidal-flats-v1-1: version 1.1, the five periods from 1984-1986 to 1996-1998 that version 1.2 does not cover, plus the version 1.1 mapped-area layer (datamask).Raster values: 1 = tidal flat, 0 = not tidal flat or outside the mapped area. The H3 tables have one row per (period, resolution 8 cell) holding any tidal flat, with epoch, tidal_flat_km2 and tidal_flat_fraction. Totals differ noticeably between periods (the two latest version 1.2 periods are about 30% above 1999-2013); these differences are in the producers' maps, which they caution are not a consistent time series everywhere.
The rasters render through the NRP titiler service; the colormap draws value 1 and leaves 0 and 255 transparent.
Murray, N. J. et al. (2019). The global distribution and trajectory of tidal flats. Nature, 565, 222-225, https://doi.org/10.1038/s41586-018-0805-8 (version 1.1). Murray, N. J. et al. (2022). High-resolution global maps of tidal flat ecosystems from 1984 to 2019. Scientific Data, 9, 542, https://doi.org/10.1038/s41597-022-01635-5 (version 1.2). Data from the producers' archive gs://uq-intertidal, downloaded 2026-10-25; the original zips are kept under s3://public-coastal/raw/tidal-flats/.
Creative Commons Attribution 4.0.
coastal-typology-durr-2011)A global classification of the coast into geomorphic types that describe how a stretch of coast filters what rivers carry to the sea (Dürr et al. 2011). Two layers share the class column FIN_TYP:
coastline: 9,604 line segments drawn along the edges of a 0.5-degree grid. Segments can lie up to about 25 km from the true shoreline.catchments: 7,057 polygons of the land draining to each stretch of coast, built from the STN-30 version 6 river basins. A basin can span several polygons, so total BASINAREA over distinct BASINID.FIN_TYP values: 0 endorheic or glaciated, 1 small deltas, 2 tidal systems, 3 lagoons, 4 fjords and fjärds, 5 large rivers (Va), 51 large rivers with tidal deltas (Vb, catchments only), 6 karst, 7 arheic, -9999 no data (catchments only). On the coastline layer, code 5 covers both large-river subtypes.
The H3 hex tables have one row per (feature, cell) pair, so deduplicate on _cng_fid before counting features.
The PMTiles source-layer is catchments (or coastline for the line layer).
Dürr, H. H., Laruelle, G. G., van Kempen, C. M., Slomp, C. P., Meybeck, M., and Middelkoop, H. (2011). Worldwide Typology of Nearshore Coastal Systems: Defining the Estuarine Filter of River Inputs to the Oceans. Estuaries and Coasts 34(3), 441-458. https://doi.org/10.1007/s12237-011-9381-y. Data downloaded from http://geotypes.net on 2026-11-03. The authors allow use under conditions similar to CC BY (http://geotypes.net/disclaimer.html): cite the article, state any modifications, and do not imply their endorsement.
mangrove-typology-v3)A global classification of the world's mangroves by the coastal setting they grow in (Worthington et al.). Mangrove patches are grouped into 3,983 geomorphic units, each classed as Delta, Estuary, Lagoon or OpenCoast in the Class column. The typology is matched to the Global Mangrove Watch version 3.12 extent.
There are twelve layers with the same units and classes: composite (every year's extent combined) and one layer per mapped year (1996, 2007, 2008, 2009, 2010, 2015, 2016, 2017, 2018, 2019, 2020). A unit keeps its Type (for example Delta_30028) and ID in every layer, so use either to match units across years. OGC_FID differs between layers. In the 1996 layer two units, Lagoon_8338 and Lagoon_8440, are each split across two rows, so count units over distinct Type.
The H3 hex tables have one row per (unit, cell) pair. A cell belongs to a unit when its centre lies inside the unit's mangroves, or, for a patch too small to hold a cell centre, when it holds a point inside the patch. Use the hexagons to locate units and the polygons for mangrove area.
The PMTiles source-layer is the layer name, for example 2020 or composite.
Worthington, T. A., et al. (2023). A global biophysical typology of mangroves version 3. Zenodo, https://doi.org/10.5281/zenodo.8340259, licensed CC-BY-4.0. Method: Worthington, T. A., et al. (2020). A global biophysical typology of mangroves and its relevance for ecosystem structure and deforestation. Scientific Reports 10, 14652, https://doi.org/10.1038/s41598-020-71194-5. Data downloaded from Zenodo on 2026-11-03.
ar6-sea-level-projections)Projections of relative sea-level change from the IPCC Sixth Assessment Report (Working Group I, chapter 9), version 20210809, at 1,030 tide gauges and 65,160 nodes of a 1-degree global grid. Change is in millimetres since the 1995-2014 baseline and rates in millimetres per year, from 2020 to 2150 under five SSP emissions scenarios and five warming levels, and to 2300 for the low-confidence projections. Each projection carries 107 quantiles (p0 to p100; p0_1 is the 0.1st percentile, p99_5 the 99.5th) for the total and for each contributing process.
regional-confidence: relative sea level including vertical land motion, as a coastal location experiences it.regional-confidence-novlm: the same without vertical land motion and other background processes.coastal-band/hex: every H3 resolution-8 cell of the MEOW coastal band (meow-coastal-band), with the with-VLM total change of the nearest grid node that has a projection in every year of each scenario (p5, p17, p50, p83, p95). This is a nearest-node assignment, not an interpolation; location_id names the node. The node is usually within 40 km of the cell (99% within about 70 km). In the 5 °C warming level (tlim5.0win0.25), upstream has projections at only 2 of the 83 Black Sea grid nodes, so most of the Black Sea coast takes values from nodes outside the sea, up to 982 km away.The tables have one row per location, scenario, confidence level, component, measure and year, so filter on all of them to get one projection. A missing projection is null in every quantile column.
There is no PMTiles layer: the tables are point projections with many rows per location. For a map, draw the hex for one scenario, confidence level and year.
There are no vector tiles for this dataset. To map it, query the hex for one scenario, confidence level and year and draw the cells, for example with deck.gl's H3HexagonLayer:
Garner, G. G., et al. (2021). IPCC AR6 Sea Level Projections. Version 20210809. Dataset accessed 2026-11-03 at https://doi.org/10.5281/zenodo.5914709, licensed CC-BY-4.0. Upstream requires users to cite Fox-Kemper, B., et al. (2021), Ocean, Cryosphere and Sea Level Change, in Climate Change 2021: The Physical Science Basis, pp. 1211-1362, https://doi.org/10.1017/9781009157896.011; Kopp, R. E., et al. (2023), The Framework for Assessing Changes To Sea-Level (FACTS) v1.0, Geoscientific Model Development 16, 7461-7489, https://doi.org/10.5194/gmd-16-7461-2023; and the data set. Works citing the projections should include this acknowledgement: "We thank the projection authors for developing and making the sea-level rise projections available, multiple funding agencies for supporting the development of the projections, and the NASA Sea Level Change Team for developing and hosting the IPCC AR6 Sea Level Projection Tool."