Data sources
Questions
Q: How often do you update your database with model data and satellite data ? (A01062102)
Q: What data sources are used for the various tables and plots ? (A01062106)
Q: Does waveclimate.com cover seas and oceans in any part of the world ? (A01062505)
Q: How do I get quantitative information on tropical storms ? (A01062506)
Q: How do you check the correctness of your data sources and processing ? (A01062603)
Q: What is the maximum windspeed and waveheight stored in your satellite database ? (A01062701)
Q: What about accuracy of windspeed and waveheight stored in your database ? (A01062801)
Q: What period is covered by the data in your database ? (A01062802)
Answers
We update the database once a year. We will bring out a new version of waveclimate.com each spring. Such a new version will be based on data covering the years up to and including the year before. This means that in spring 2011, waveclimate.com is based on data up to and including December 2010. Data up to and including December 2011 will become available in the 2012 spring version.
Satellite wind- and wave data comes from altimeter, with the following exceptions:
- Histograms of wind-sea or swell wave parameters are based on SAR data
- Scatter tables/histograms of wind-sea or swell wave parameters are based on SAR data
- Scatter tables of windspeed versus wind direction come from scatterometer.
- Scatter tables (nearshore) of wave height versus windspeed use SAR wave height and scatterometer windspeed (collocated with SAR)
- Monthly distributions (nearshore) of wave height use SAR wave height.
- Histograms (nearshore) of wave height use SAR wave height.
Alternatively, plots and tables can be based on wavemodel output from a specified model gridpoint (except for vector plots of wind fields). Here is the complete tabular overview of data sources used for the various tables and plots.
Plot or table | Parameter(s) | Part of spectrum | Offshore | Nearshore |
---|---|---|---|---|
Histogram |
Windspeed | Not applicable | Altimeter or wavemodel | wavemodel |
Wave height | Total | Altimeter or wavemodel | wavemodel | |
Wave height | Wind sea/swell | SAR or wavemodel | wavemodel | |
Wave period | Total/wind sea/swell | SAR or wavemodel | wavemodel | |
Monthly distribution |
Windspeed | Not applicable | Altimeter or wavemodel | wavemodel |
Wave height | Total | Altimeter or wavemodel | wavemodel | |
Scatter table 2D |
Windspeed vs. wind direction | Not applicable | Scatterometer | wavemodel |
Wave height vs. wind speed | Total | Altimeter or wavemodel | wavemodel | |
Wave height vs. wave period | Total/wind sea/swell | SAR or wavemodel | wavemodel | |
Wave height vs. wave direction | Total/wind sea/swell | SAR or wavemodel | wavemodel | |
Scatter table 3D |
Wave height, wave period and wave direction |
Total/wind sea/swell | wavemodel | wavemodel |
Windspeed, wave height and wind direction |
Total | wavemodel | wavemodel | |
Data of past events |
Windspeed | Not applicable |
Scatterometer and Altimeter and wavemodel |
Not available |
Wave height | Total | Altimeter and wavemodel | Not available | |
Vector map | Windspeed and -direction | Not applicable | Scatterometer | Not available |
Persistency | Wave height, wave period, windspeed | Total/wind sea/swell | wavemodel | wavemodel |
On open ocean, windspeed from scatterometer is more accurate than altimeter
windspeed. In coastal zones however, altimeter windspeed is more reliable
than scatterometer windspeed: within 50 km of the coastline we prefer
altimeter. Altimeter waveheight is better than SAR waveheight
because more samples are available and because its more
accurate.
Comparison of satellite and model results is discussed
in a separate question (Q03071501).
Yes, the waveclimate.com database covers all seas and oceans. The density of samples may vary, however. You can check the availability of satellite data from the different missions by using one of the data density charts as background for the area selection screen while navigating over the globe. There is a chart ("data density scatterometer") indicating the density of wind data, a chart ("data density altimeter") giving the density of wave data and a chart ("data density SAR") showing the availability of spectral wave data. Finally, the chart "gridpoints wavemodel" shows the location of the wave model output points. Answer A01062106 deals with the question which data sources are being used for the various tables and plots generated by waveclimate.com.
Waveclimate.com warns you if tropical storms are known to occur in your area of interest, because you cannot get reliable information on the tropical storms from waveclimate.com.
The reason is that waveclimate.com derives the wind and wave climate from a database of wind and wave observations at "random" times and positions, not specifically focused on tropical cyclones. Therefore, a tropical cyclone can easily be "missed". The same holds for practically all wave and wind observations. In fact, data from numerical weather and wave forecasting models are also not suitable because tropical storms are not properly resolved in these models. The only data that we find really useful for tropical storm climate assessment are observations of track and intensity (e.g. central pressure or max. windspeed), or hindcast data from dedicated models. These are not included in waveclimate.com, because they cannot be mixed with the other data.
At ARGOSS, we hold worldwide dataset of tropical storm data based on the NOAA GTECCA dataset, which we have screened and validated at a number of locations. Based on these data, we can provide you with crude but fairly reliable estimates of the tropical storm climate for any location. Please contact the ARGOSS helpdesk if you are interested in obtaining such estimates.
The data in the global waveclimate.com database have been validated against buoys. This validation will be repeated once a year when we add new data to the database. Satellites are calibrated with buoys, leading to corrections per mission. Calibration factors for wave model data are derived per model point from the comparison with satellites. Satellite data are further improved before being used for the calibration of the wave model. The online waveclimate.com service applies these corrections to the raw data from the database. Methods and calibration results can be found in a separate document. Section 1.2 of this document provides an executive summary. Proof is given that, with reference to buoys, calibration with these 'best' satellite observations does indeed improve the quality of our wave model. The systematic error in model wave height is reduced by 50% or more. The final bias in model wave height relative to buoys is at most 5cm while wind speed is off by less than 20cm/s.
The waveclimate.com satellite database stores raw and corrected wind speed (both scatterometer and altimeter) in one byte (0-255) in terms of 0.2 m/s. Therefore, the maximum windspeed that could possibly be stored would be 51 m/s. Before being added to the database, scatterometer winds are maximized at 45 m/s inside the latitude range 50S-50N and at 35 m/s outside this range (this range has to do with the presence of ice).
Altimeter raw and corrected waveheights are stored in one byte in terms of 0.1m. This means that the database is capable of storing waveheights upto 25.5 m.
The listed maxima indicate that the saturation value of altimeter measurements of windspeed are less than those of the ERS scatterometers. So scatterometers are the better source of data for high windspeeds. As for significant wave height, Geosat/GFO altimeter wave height data appear to saturate at lower values than ERS, Topex/Poseidon and Jason-1 data.
You cannot assume that the sensors will produce reliable data up to the listed maxima under all conditions because the instruments are calibrated to optimize overall performance, not performance during extreme conditions. Moreover, very few satellite-buoy collocations with high waves and windspeeds are available for calibration. Finally, in tropical cyclones, the highest winds occur in a small region so the spatial resolution of a scatterometer will limit its ability to measure these high winds.
Maximum windspeed found in the waveclimate.com database:
Satellite | Sensor | Date | Position |
Max. raw/corrected Wind speed (m/s) |
Ers-1 | Altimeter | Dec 24, 1991 | 78.42S,167.84W | 20.2 / 21.1 |
Ers-2 | Altimeter | Dec 26, 2001 | 37.26S,25.11E | 32.8 / 31.0 |
Topex/Pos | Altimeter | Jul 21, 2001 | 66.06S,116.84E | 32.8 / 31.3 |
Jason-1 | Altimeter | Feb 11, 2002 | 26.26S,86.76E | 30.2 / 29.8 |
GFO | Altimeter | Feb 7, 2003 | 16.64S,164.36W | 32.8 / 30.2 |
Ers-1 | Scatterometer | Jun 5, 1992 | 45.05S,18.02W | 42.8 / 45.4 |
Ers-2 | Scatterometer | Oct 4, 1999 | 49.91S,7.24E | 37.8 / 40.2 |
Maximum waveheight found in the waveclimate.com database:
Satellite | Sensor | Date | Position |
Max. raw/corrected Waveheight (m) |
Geosat | Altimeter | May 30, 1989 | 66.20S, 124.42W | 12.0 / 11.9 |
Ers-1 | Altimeter | Dec 17, 1991 | 30.29N, 175.15E | 17.0 / 19.6 |
Ers-2 | Altimeter | Dec 27, 2001 | 37.26S, 25.11E | 20.1 / 21.3 |
Topex/Pos | Altimeter | Mar 18, 1999 | 45.94N,176.90W | 20.6 / 21.7 |
Jason-1 | Altimeter | Oct 10, 2003 | 65.41S, 121.56W | 21.6 / 22.9 |
GFO | Altimeter | Jul 31, 2002 | 57.12S, 67.86E | 15.6 / 16.9 |
The accuracy of windspeed and waveheight of our in-house wave model database used for consultancy has been assessed by validating the data against buoys. Results are presented in a separate validation document. In comparison to consultancy practice, waveclimate.com uses some simplified methods and compresses model data. Differences between the online service and consultancy are explained and illustrated here. The quality of satellite data is discussed in this document. Altimeter accuracies are given in section 3.4 table 13-14. Scatterometer results are in section 3.4 table 15. SAR data quality can be found in section 4.1, table 16-19.
The database contains wave model hindcast data for the years 1992-2022 and satellite data for the years 1985-2022.
Wave model data are globally available on the oceans (gridsize 1/2x1/2 degree) and on two finer regional grids. The regional Mediterranean covers the following closed bassins (gridsize 1/4x1/4 degree):
- Mediterranean
- Red Sea
- Black Sea
- Caspian Sea
- Arabian Gulf
The regional EU-shelf model covers the following areas (gridsize 1/6x1/6 degree):
- NE Atlantic
- North Sea
- Baltic Sea
The global database contains satellite observations covering
the years 1985-2022. Windspeed, significant wave height and spectral
wave parameters are available for any region in the world. The
data comes from different satellites, i.e. Geosat, GFO (Geosat
Follow On), Topex/Poseidon, Jason-1/2/3, Ers-1/2, Envisat, Cryosat, Saral, Sentinel3a/b, Quikscat and MetOp-A.
The table below shows the period covered by our records for each satellite-sensor
combination. Geosat, Ers-1, Topex/Poseidon, GFO, Envisat, Jason-1 and Quikscat are no longer operational;
Jason-2, Cryosat, Saral, Sentinel3a/b, Jason-3 and MetOp-A are. Worldwide Ers-2 altimeter is
present until June 2003. As of January 2004, Ers2-altimeter data is provided
for part of the northern hemisphere only. Reliable Ers-2 scatterometer
data stopped coming in January 2001 due to gyroscope problems.
Quikscat scatterometer stopped in November 2009 because of antenna problems.
Envisat stopped transmitting in April 2012. Jason-1 stopped in June 2013 and Jason-2 in October 2019.
Unfortunately, we no longer have access to Sentinel-3b data from 2020 or later.
Currently ongoing altimeter missions are Cryosat, Saral, Sentinel-3a and Jason-3.
MetOp-A provides scatterometer data upto present day.
Satellite mission | Sensor | Begin | End | Mission completed |
Geosat | Altimeter | 03-1985 | 12-1989 | Yes |
GFO | Altimeter | 01-2002 | 09-2008 | Yes |
Ers-1 | Altimeter | 08-1991 | 05-1996 | Yes |
Ers-2 | Altimeter | 04-1995 | 06-2011 | Yes |
Topex/Pos | Altimeter | 09-1992 | 10-2005 | Yes |
Jason-1 | Altimeter | 01-2002 | 06-2013 | Yes |
Jason-2 | Altimeter | 07-2009 | 09-2019 | Yes |
Envisat | Altimeter | 01-2003 | 04-2012 | Yes |
Cryosat | Altimeter | 07-2010 | 12-2022 | No |
Saral | Altimeter | 03-2013 | 12-2022 | No |
Sentinel-3a | Altimeter | 06-2016 | 12-2022 | No |
Sentinel-3b | Altimeter | 06-2018 | 12-2019 | No |
Jason-3 | Altimeter | 02-2016 | 12-2022 | No |
Ers-1 | Scatterometer | 03-1992 | 05-1996 | Yes |
Ers-2 | Scatterometer | 04-1996 | 12-2000 | Yes |
Quikscat | Scatterometer | 10-2000 | 11-2009 | Yes |
MetOp-A | Scatterometer | 01-2010 | 12-2022 | No |
Ers-1 | SAR | 04-1993 | 04-1996 | Yes |
Ers-2 | SAR | 05-1996 | 06-2003 | Yes |