Wave-energy resource climatology from satellite altimetry
Radar altimeters on Sentinel-6, Jason-3, and SARAL/AltiKa have measured significant wave height globally since the early 1990s. Combined with ERA5 reanalysis, these records yield the exceedance statistics and wave-power climatologies that wave-energy converter projects need before a single buoy is deployed.
Sensors
- Sentinel-6 Michael Freilich: Ku- and C-band dual-frequency radar altimeter (POSEIDON-4) providing significant wave height (Hs) along nadir ground tracks with ~7 km along-track posting, 10-day exact-repeat cycle, global coverage. Successor to the TOPEX/Jason series, maintaining continuity of the sea-surface record from 1992.
- Jason-3: Ku/C-band altimeter operated by NOAA/EUMETSAT/CNES/NASA. Interleaved 10-day repeat with Sentinel-6 to give a combined ~5-day effective revisit at mid-latitudes. Hs uncertainty of approximately ±0.25 m or 10% of Hs, whichever is larger, under typical sea states.
- SARAL/AltiKa: Ka-band altimeter (35.75 GHz) on a 35-day repeat orbit reaching 81.5° latitude. Ka-band's narrower pulse footprint (~8 km) reduces sea-surface bias in low-to-moderate sea states, making it particularly useful for nearshore and high-latitude wave climatology where Jason-class sensors can overestimate Hs.
- ERA5 wave reanalysis (ECMWF): Spectral wave model output at approximately 31 km spatial resolution, hourly, from 1940 to near-real-time. Not a satellite sensor, but the essential spatial interpolant: altimeter ground tracks are sparse, and ERA5 is calibrated against the altimeter record to fill the gaps. Provides peak period (Tp) and directional spreading that altimeters alone cannot deliver.
What a nadir altimeter actually measures, and what it cannot
A radar altimeter illuminates a disc of ocean directly beneath the satellite and measures the two-way travel time of the return pulse. The spread of that return in time encodes the root-mean-square wave height of the sea surface, from which significant wave height (Hs, defined as the mean of the highest third of waves) is derived. Wind speed is retrieved simultaneously from the backscatter intensity. The along-track spatial resolution is typically 5 to 10 km after averaging, and the measurement is precise: Sentinel-6 Hs uncertainty is published at roughly ±0.25 m under validation conditions.
The hard constraint is geometry. An altimeter sees only a single ground track, roughly 1 to 6 km wide, on each pass. The cross-track gap between adjacent passes at the equator is around 315 km for the 10-day Jason/Sentinel-6 orbit. A wave-energy developer asking 'what is the wave climate at this specific point?' will find that the nearest altimeter pass may be 50 to 150 km away, and repeat observations at that point arrive only every 10 to 35 days depending on the mission. No single altimeter mission provides the spatial density needed for site-specific climatology without assistance.
Thirty years of ground tracks, stitched into a resource map
The value of altimetry for wave-energy resource assessment lies in its archive, not its instantaneous coverage. The TOPEX/Poseidon mission began in 1992, and the Jason series has maintained an unbroken Hs record ever since. SARAL/AltiKa adds a complementary orbit. Across three decades, the accumulated ground tracks build a statistical sample dense enough to compute Hs exceedance curves, seasonal cycles, and interannual variability at any point within a few hundred kilometres of a track.
The standard workflow combines all available altimeter missions into a multi-mission along-track dataset, then uses ERA5 as a spatial and temporal interpolant. ERA5's wave model is itself constrained by assimilated altimeter data, so the two sources are not independent, but the blending is well-characterised. The result is a gridded climatology of mean wave power density (proportional to Hs² × Tp × ρg/16, where ρ is seawater density and g is gravitational acceleration), expressed as kilowatts per metre of wave crest width. Published global assessments place mean offshore wave power density in high-resource regions such as the north-east Atlantic and Southern Ocean at 40 to 80 kW/m, with the north Atlantic west of Ireland frequently cited above 50 kW/m in the published literature.
From Hs to what a converter actually sees
Wave-energy converters (WECs) are tuned to a target period range. A device optimised for 8 to 12 second swells performs poorly in a 5-second wind-sea. Altimeters measure Hs and wind speed directly; they do not measure wave period or directional spectrum. Peak period must be inferred, either from empirical Hs-to-Tp relationships derived from buoy co-locations, or from ERA5's spectral wave output.
The directional component matters for device orientation and array layout. Altimetry cannot resolve direction at all. ERA5 provides mean wave direction and directional spreading, but at 31 km resolution these are smoothed representations, particularly in areas of complex bathymetry or swell shadowing from island chains. For nearshore sites within a few kilometres of a headland or reef, neither altimetry nor ERA5 is a substitute for a local wave propagation model forced by offshore boundary conditions derived from the satellite record. This is an honest limit of the method, not a deficiency of the satellite data alone.
Exceedance statistics and the siting decision
The primary output for a WEC developer is not a single mean value but a joint probability distribution of Hs and Tp, often displayed as a scatter diagram, alongside an Hs exceedance curve showing the fraction of time the sea state exceeds a given threshold. Structural survivability requires knowing the 50-year return Hs; energy yield requires knowing the fraction of time the device operates within its rated sea state.
From a 30-year multi-mission altimeter record blended with ERA5, it is possible to compute Hs exceedance at the 90th, 95th, and 99th percentiles with reasonable confidence at offshore locations. Interannual variability driven by the North Atlantic Oscillation or ENSO introduces uncertainty of 10 to 20% in annual mean wave power at many sites. A responsible resource assessment quantifies this variability explicitly rather than reporting only the long-term mean.
Seasonal asymmetry is also critical. North Atlantic sites may see mean winter wave power three to five times higher than summer values. A WEC array sized for winter survivability will be oversized for summer generation. Satellite climatology resolves this seasonal structure clearly because the archive spans many complete annual cycles.
Where the satellite record is thinner, and what to do about it
High-latitude sites above 66° are poorly sampled by the Jason/Sentinel-6 orbit, which reaches only 66.15° inclination. SARAL/AltiKa at 81.5° fills much of this gap for polar and sub-polar regions. Coastal waters shallower than roughly 10 to 20 km from shore introduce land contamination into the altimeter footprint, degrading Hs retrievals. Dedicated coastal retracking algorithms (such as ALES or X-TRACK) partially recover nearshore data, but the last 5 to 10 km before the coast remains unreliable.
In these marginal zones, ERA5 at 31 km resolution is also degraded by the same coastal geometry. The practical recommendation is to use the satellite-derived offshore climatology as the boundary condition for a higher-resolution spectral wave model (SWAN or WAVEWATCH III are standard choices) that propagates the resource into the nearshore. Satellize's analytics workflow follows this approach, delivering offshore climatology grids for client-specified regions as GIS-ready inputs to wave propagation modelling. The Tonga crop-estimation programme is a different domain, but the underlying principle of using open satellite archives to characterise environmental conditions for a sovereign client is the same.
Instrument drift and inter-mission calibration are real concerns over a 30-year record. The Jason/Sentinel-6 series has been carefully cross-calibrated through tandem flight phases, but residual biases of a few centimetres in Hs exist between missions. Published multi-mission datasets from CMEMS and AVISO account for these biases, and any resource assessment should document which calibration baseline was used.
What the numbers can and cannot promise a developer
Satellite altimetry provides the only globally consistent, decades-long observational record of ocean wave climate. No buoy network comes close to that spatial and temporal breadth. For pre-feasibility screening across a wide area of ocean, it is the right tool.
It is not a substitute for in-situ measurement at the final selected site. A 12-month directional wave buoy deployment, informed by the satellite climatology, remains standard practice before financial close on a WEC project. The satellite record tells you where to put the buoy, and gives you the long-term context to interpret what the buoy measures. That combination, satellite climatology plus targeted in-situ validation, is more defensible to a project financier than either source alone.
Typical figures
| Along-track Hs posting interval | 1 Hz products: ~7 km (Sentinel-6, Jason-3); 20 Hz high-rate: ~350 m along-track, though noise increases |
| Cross-track gap at equator | ~315 km for 10-day Jason/Sentinel-6 orbit; ~800 km for SARAL 35-day orbit |
| Hs measurement uncertainty | ±0.25 m or ±10% of Hs (whichever is larger) for Jason/Sentinel-6 under validated conditions |
| Wind speed uncertainty (altimeter) | ±1.5 m/s for 10 m neutral wind speed (published Sentinel-6 performance) |
| ERA5 wave reanalysis resolution | ~31 km spatial, 1-hour temporal, global |
| Archive depth (multi-mission) | 1992 to present (TOPEX/Poseidon through Sentinel-6); SARAL from 2013 |
| Latency (near-real-time products) | ~3 hours for NRT Sentinel-6 OGDR; ~1 day for IGDR; ~60 days for final GDR |
| Coastal validity limit | Standard retracking: unreliable within ~10-20 km of coast; coastal retrackers (ALES) recover to ~5 km |
| Orbital inclination / latitude cover | Jason/Sentinel-6: ±66.15°; SARAL/AltiKa: ±81.5° |
| Wave power density derivable parameters | Hs (direct); Tp (ERA5 or empirical); direction (ERA5 only); power density in kW/m |
Analytics Satellize can run
| Offshore wave-power density climatology grid | Multi-mission altimeter Hs blended with ERA5 Tp and direction; wave power formula P = ρg²Hs²Tp/64π | GeoTIFF or NetCDF grid of mean, seasonal, and percentile wave power density for a client-specified ocean region |
| Hs exceedance probability curves | Empirical CDF from 30-year multi-mission along-track record, with ERA5 gap-filling; GEV fitting for return-period extremes | CSV and chart report: P50/P90/P99 Hs and estimated 50-year return Hs at candidate site coordinates |
| Joint Hs-Tp scatter diagram | Co-location of altimeter Hs with ERA5 Tp at coincident grid points; binned joint probability table | Scatter diagram PDF and underlying data table, formatted for WEC performance matrix convolution |
| Seasonal and interannual variability index | Monthly mean wave power time series from blended record; correlation with NAO/ENSO indices from published climate datasets | Time-series chart and statistical summary report quantifying year-on-year variability at candidate sites |
| Regional site-ranking map | Gridded mean annual wave power density ranked by threshold criteria (e.g. mean Hs > 1.5 m, Tp 6-14 s, water depth from GEBCO bathymetry) | GIS polygon layer of ranked candidate zones with tabulated resource statistics per zone |
| Offshore boundary condition dataset for wave propagation modelling | Extracted ERA5 spectral wave parameters (Hs, Tp, direction, spreading) at client-specified offshore boundary points | Time-series CSV or NetCDF formatted as SWAN/WAVEWATCH III boundary input, with documentation |
Who does the work
We can get this done for you. Satellize runs its own analyst desk and a strong science team. You do not buy a data feed and work out what it means; our people source the imagery, run the analysis described on this page, and hand you the answer with its confidence limits stated. Discuss this requirement.