The breathing ocean — copy

Methodology

What the map shows

Colour represents a spring tidal-range proxy, in metres, derived from the HAMTIDE11a harmonic tide model:

Here A_{\mathrm{M}_2} and A_{\mathrm{S}_2} are the amplitudes, in metres, of the principal lunar and solar semidiurnal tidal constituents. The proxy estimates the semidiurnal high-to-low range when these two constituents reinforce. It is not the total high–low tidal range, a dated forecast, observed station data, or a new fluid simulation. Other constituents, nodal corrections, weather and storm surge are omitted.

Data and processing

• Model and access: HAMTIDE11a, distributed by ICDC / CEN / University of Hamburg, retrieved through public OPeNDAP on 20 October 2026.
• Source subset: 15°W–25°E, 48–65°N, at the native 0.125° (7.5 arc-minute) spacing: 137 latitude rows × 321 longitude columns. The source methodology reports 23,181 valid wet nodes across this full subset—not just the visible map.
• Units and masking: source AMPL fields are in centimetres. Negative fill values (-999) are masked before conversion to metres and calculation of the proxy. The published matrix is rounded to integer millimetres; the renderer divides by 1,000. Rounding contributes at most 0.5 mm error.
• CSV structure: the first column is latitude; subsequent columns run from lon_-15 to lon_25 in 0.125° increments. Blank cells indicate missing values, not zero.

Cartography and interpretation

The visible footprint covers approximately 12°W–20°E, 50–63.5°N, using a regional Lambert conformal conic projection with standard parallels at 50° and 62°N and a central longitude of 4°E.

Display interpolation is bilinear only where all four surrounding model values are valid; otherwise the renderer uses nearest-neighbour sampling. The resulting projected field is drawn as SVG isobands. Interpolation smooths the display but does not increase the model’s physical resolution.

• The legend spans 0–8 m on a linear value axis. Values above 8 m use the upper colour, labelled 8+.
• Fine lines are equal-range contours, not currents or tidal propagation paths.
• Hatching means no model data, not zero tidal range. Differences between the model wet mask and the coastline can leave hatched near-shore patches.
• City dots are geographic reference points—not tide gauges or reported local tidal ranges.
• Coastline and international land boundaries are from Natural Earth 10m. Here “10m” means a 1:10-million cartographic scale, not 10-metre spatial resolution. The coastline is regionally clipped, simplified at approximately 0.006° and rounded to 0.001°. Country borders are separately clipped and simplified at approximately 0.008°, then drawn as subtle reference lines. The tidal colour scale is identical in light and dark mode.

Limitations

This is a static, model-derived field, not an automatically refreshing forecast. It does not resolve every estuary, fjord, island or local coastal extreme. Not for navigation, engineering or flood-risk decisions.

Sources and reproducibility

HAMTIDE documentation · THREDDS catalogue

OPeNDAP base: https://icdc.cen.uni-hamburg.de/thredds/dodsC/ftpthredds/hamtide/

For each of m2.hamtide11a.nc and s2.hamtide11a.nc, request these ASCII subsets and concatenate the western block before the eastern block along longitude:

.ascii?AMPL[1104:1:1240][2760:1:2879]
.ascii?AMPL[1104:1:1240][0:1:200]

Both queries cover 48–65°N. The first spans −15 to −0.125° longitude (345–359.875° in the source coordinates); the second spans 0–25°. Retain model attribution and check the provider’s reuse terms: public access alone does not establish unrestricted redistribution rights. Natural Earth basemap data are public domain.

Citation

Taguchi, E., Stammer, D., and Zahel, W. (2014). Inferring deep ocean tidal energy dissipation from the global high-resolution data-assimilative HAMTIDE model. Journal of Geophysical Research: Oceans, 119, 4573–4592. doi:10.1002/2013JC009766. Distributed in NetCDF by ICDC, CEN, University of Hamburg.