Waveclimate.com

New Features

New features of waveclimate.com "(Apr 28, 2023)" are:

Added satellite data for the year 2022.
The altimeter database holding wave height and wind speed now covers a period of 38 years (1985-2022). Scatterometer wind speed and wind direction are available for 31 years (1992-2022). For details on missions see FAQ on model and satellite data in the database.
Completed altimeter data for the Caspian Sea (data from 2002-2008 were missing).

New features of waveclimate.com "(Mar 16, 2023)" were:

Added model data for the year 2022.
The hindcast database now holds 31 years of data (1992-2022)

New features of waveclimate.com "(Mar 25, 2022)" were:

Added satellite data for the year 2021.
The altimeter database holding wave height and wind speed now covers a period of 37 years (1985-2021). Scatterometer wind speed and wind direction are available for 30 years (1992-2021). For details on missions see FAQ on model and satellite data in the database.

New features of waveclimate.com "(Feb 4, 2022)" were:

Added model data for the year 2021.
The hindcast database now holds 30 years of data (1992-2021)

New features of waveclimate.com "(Apr 10, 2021)" were:

Extended the header of the text file holding time series from the wave ray model.
In addition to nearshore depth, the header now shows water level in the area and bed level at the nearshore site. Please recall that local depth is found as water level (a constant for the whole area) minus local bed level.
Some minor bug fixes concerning the wave ray model
-Winds used for local wave growth were occasionally copied from the nearest boundary point rather than from the grid point nearest to the nearshore site.
-Option to switch off both wave growth from the seaward boundary and wave growth from the coast no longer blocked.

New features of waveclimate.com "(Feb 5, 2021)" were:

Added nearshore depth to the title of the small scale bathymetry plot.
This is done to make the exact value of the nearshore depth available (via the related link appearing in the SWRT progress page) as soon has it has been computed by the wave ray model SWRT.
Added model data and satellite data for the year 2020.
The hindcast database now holds 29 years of data (1992-2020) The altimeter database now covers a period of 36 years (1985-2020). Scatterometer data are available for 29 years (1992-2020). For details on missions see FAQ on model and satellite data in the database.

New features of waveclimate.com "(May 27, 2020)" were:

Added nearshore depth to the header of the time series text file made for downloads.
Adapted depth plots to show total water depth (positive number) rather than seabed level (negative number).
The header of the downloadable time series text file based on a nearshore translation now holds the water depth (height of the water column) at the nearshore location. By default the wave ray model estimates depth from nautical charts. Depth plots now show the total water depth (water level minus seabed level) relative to lowest astronomical tide (LAT). So far the seabed level was plotted which does not include water level. By default water level in the area is set to zero but can be raised to simulate e.g. mean sea level (MSL) conditions.
Added satellite altimeter data for the year 2019.
The altimeter database now covers 35 years (1985-2019). For details on missions see FAQ on model and satellite data in the database.

New features of waveclimate.com "(Mar 1, 2020)" were:

Added wave model data for the year 2019.
All grids now provide 28 years of 3-hourly model data (1992-2019).
Added satellite scatterometer wind data for the year 2019. Altimeter data for 2019 still has to be added.
Scatterometer MetOp-A data from 2019 have been added to the satellite database. The scatterometer now covers 28 years (1992-2019). The altimeter database covers 34 years as before (1985-2018).

New features of waveclimate.com "(Jun 20, 2019)" were:

Added satellite altimeter data for the year 2018.
The altimeter database now covers 34 years (1985-2018). Added data provided by new satellite mission Sentinel-3B launched in 2018.
Also see FAQ on model and satellite data in the database.

New features of waveclimate.com "(Apr 11, 2019)" were:

Added wave model data for the year 2018.
All grids now provide 27 years of 3-hourly model data (1992-2018).
Added satellite scatterometer wind data for the year 2018. Altimeter data for 2018 still has to be added.
Scatterometer MetOp-A data from 2018 have been added to the satellite database. The scatterometer now covers 27 years (1992-2018). The altimeter database covers 33 years as before (1985-2017).
Limited the number of boundary points for a nearshore translation to avoid a silent stop due to lack of memory.
The wave ray model translates wave spectra provided at the offshore boundary to the nearshore location. Over time, the length of the offshore wind and wave time series increase. As a result, a nearshore translation job may come to a halt due to memory overflow (without further notice to the end user). To cicumvent this unwanted behavior, the set of offshore points is thinned if the number of points exceeds a given threshold (set to 15).

New features of waveclimate.com "(December 01, 2018)" were:

Re-processed EU-shelf wave data.
Waveclimate.com provides wind and wave data from three WaveWatchIII grids. Apart from the global grid (30x30 minutes) the data come from two regional grids. The Mediterranean grid (15x15 minutes) covers the Mediterranean, Red Sea, Black Sea, Caspian Sea and the Persian Gulf. The EU-shelf grid (10x10 minutes) covers the Batic, the North Sea and part of the northern Atlantic.

We re-processed the EU-shelf data after detection of a small but unjust shift of spectral wave energy towards the lower spectral frequencies.

We are aware of the fact that, due to data compression and fully automated processing, waveclimate.com wave data are less accurate than project based wave data produced by hand by our consultants. Mainly due to loss of some energy in the high end of the wave spectrum, waveclimate.com may underestimate the average height of relatively low and short wind waves by up to about 5% and overestimate the corresponding zero-crossing wave period by up to say 10%.

As a result of the shift mentioned, introduced with wave model v361 in February 2017, average significant wave height in the EU-shelf was underestimated by up to 10% and (zero-crossing) wave period overestimated by up to 20%.

Now, after re-processing, the accuracy of EU-shelf wave parameters is back to normal again as documented here.

New features of waveclimate.com "(July 04, 2018)" were:

Added option to quickly get monthly statistics for multiple offshore model points, for example for way points along a route.
You can now generate a quick summary of the wind and wave climate for up to 10 offshore locations in one go. The following 2D scatter tables will be written to one result page for each selected month and/or for all-year: This new functionality has been made available for annual (periodical) subscribers.

New features of waveclimate.com "(May 17, 2018)" were:

Added satellite altimeter wind and wave data for the year 2017.
Altimeter Cryosat, Sentinel-3a, Saral, Jason-2 and Jason-3 data from 2017 have been added to the satellite database. The satellite database now covers 33 years (1985-2017) and the model database covers 26 years (1992-2017).

New features of waveclimate.com "(Apr 25, 2018)" were:

Added wave model data for the year 2017.
All grids now provide 26 years of 3-hourly model data (1992-2017).
Added satellite scatterometer wind data for the year 2017. Altimeter data for 2017 still has to be added.
Scatterometer MetOp-A data from 2017 have been added to the satellite database. The scatterometer now covers 26 years (1992-2017). The altimeter database covers 32 years as before (1985-2016).

New features of waveclimate.com "(Dec 05, 2017)" were:

What is the impact of calibrating model waves with altimeter measurements?
To answer the above frequenctly asked question, we added charts illustrating the impact of calibration to the FAQ on the help pages. Automated calibration of waves as done by waveclimate.com not only removes the systematic model-satellite error but also significantly reduces the model-buoy error. The impact of calibrating model waves with altimeter is illustrated by means of charts of the ratio of raw and calibrated mean significant wave height.

Improved calibration of wave periods.
Waveclimate.com corrects model wave periods with the square root of the ratio of calibrated and raw wave height (to keep wave steepness constant). So far, only zero-crossing and mean wave period, found by integration of the (relevant part of the) wave spectrum, were calibrated. We now decided to also calibrate the peak wave period in this way, despite the fact that this measure is more 'jumpy' and becomes ill-defined for double-peaked wave spectra. Also see FAQ on the definition of wave periods.

We added the voucher price to the order vouchers screen.

New features of waveclimate.com "(May 29, 2017)" were:

Added satellite data for the year 2016
Altimeter data and scatterometer data from 2016 have been added to the satellite database. The satellite database now covers 32 years (1985-2016).

New features of waveclimate.com "(Feb 23, 2017)" were:

Upgraded and extended the wave model database (1992-2016)
The WaveWatchIII wave model database has been upgraded from version v352 to version v361. The hindcast database now covers 25 years 1992-2016. Spatial resolution of the global grid is 0.5x0.5 degrees in coastal zones and 1x1 degree on open ocean. CFSR winds are used to drive the wave model version v361.

The previous hindcast v352 database covered 23 years 1992-2014. The global v352 grid was coarser, i.e. 1x1.25 degree in costal zones and 2x2.5 degree on open sea. NCEP winds were used to drive wave model version v352.

In short: new hindcast v361 is up-to-date, has higher resolution and validation shows that winds are better and waves are comparable to previous v352.
Here's the full document on the validation of new hindcast v361 (PDF).

Added satellite data for the year 2015
Altimeter data and scatterometer data from 2015 have been added to the satellite database. The satellite database now covers 31 years (1985-2015).

New features of waveclimate.com "(May 30, 2016)" were:

All scatter tables can now be downloaded in a comma-separated text file.
Scatter tables are now written to readible text files that you can download through the links above the tables on the result pages. The text files contain comma-separated-values and can be readily imported into Excel or any other application for further processing. The new download format has been made available for all types of scatter tables, i.e. for regular 2D tables as well as for 2D tables per sector and month (also referred to as 3D tables).

So far you could only download tables of wind speed versus direction (in WASP format) and 3D tables per sector (one line per table entry). These options and formats are also still available.

New features of version "(July 09, 2015)" were:

Accelerated the nearshore navigation for the wave ray model.
So far the server launched MATLAB to compute the offshore boundary after each zoom or pan action at the nearshore navigation screen. This is relatively slow: each zoom action takes up to five seconds. Nearshore navigation is now done by the server. As a result each zoom action returns almost instantly.

New features of version "(May 01, 2015)" were:

Added hindcast data 2014. All hindcasts now cover a period of 23 years: 1992-2014.
Added satellite data for the year 2014
Altimeter data and scatterometer data from 2014 have been added to the satellite database. The satellite database now covers 30 years (1985-2014).
Updated nearshore wave ray model SWRT
Improved local wave growth. The wave ray model computes fetch lengths for local wave growth, both from land and 'along wave rays' tracing back to the offshore boundary. The computation of fetches per nearshore direction is now based on the distance between the nearshore location and the end of the wave ray associated with the highest frequency (HF). This HF-ray is practically a straight line as refraction of shortest waves is negligible. As a result, the computation of fetch from land is now more robust and less 'jumpy' (the land-water mask used previously occasionally lead to 'jumpy' behaviour). The computation of fetch between the nearshore location and the offshore boundary has been corrected, i.e. the selection of the offshore point providing initial wind-sea conditions. This correction may lead to differences of 10-20% in wind-seas from unsheltered directions.

New features of version "(Jun 02, 2014)" were:

All hindcast databases now cover a period of 22 years: 1992-2013.
Now all hindcast databases cover a period of 22 years (1992-2013), i.e. For the global model and for the Mediterranean model data were added for 2013. For the high-resolution EU-shelf grid, data from the years 2011, 2012 and 2013 have been added to the database. In addition, model data were added for the last two months of 2010. Previously, the EU-shelf model covered the period Jan 1992 up to and including Oct 2010.
Added satellite data for the year 2013
Altimeter data and scatterometer data from 2013 have been added to the satellite database. The satellite database now covers 29 years (1985-2013).
Next generation altimeter mission SARAL-ALTIKA.
As of March 2013, data from altimeter mission SARAL have become available. This altimeter operates in high-frequency Ka-band, leading to better performance (no correction for athmospheric influences needed anymore) and higher accuracy. Disadvantage is the increased sensisitivity for rain and drizzle.
Re-processed and extended data from altimeter mission Cryosat
Altimeter data from Cryosat have been re-processed and extended with data from 2010. Measurements from this mission now start in July 2010 instead of January 2011. Re-processing was carried out to make the Cryosat data consistent over the years 2010-2013.

New features of version "(Mar 4, 2014)" were:

Added high resolution hindcast wave model for the EU-shelf 1992-2010
The high resolution EU-shelf hindcast wave model (1/6x1/6 degree) has been added to the global wave model (1992-2012; 5/4x1 degree) and the local model for the Mediterranean, Red Sea, Persian Gulf, Caspian Sea and Black Sea (1992-2012; 1/4x1/4 degree). The EU-shelf model covers part of the NE Atlantic, the North Sea and the Baltic Sea. The EU-shelf model runs from 15W to 30E and from 40N to 66N. Global grid points in this area have been de-activated.
Added wind-sea and swell wave parameters to time series
Time series text files with 3-hourly records of wind and wave parameters can be downloaded for any model grid point. So far, time series contained wind speed, wind direction, significant wave height, mean wave period and mean wave direction based on the total wave spectrum. The time series records have now been extended with wave height, wave period and wave direction for the wind-sea part and for the swell part of the wave spectrum.
Extended the list of global background charts with hindcast based charts
The user can select a chart to be used as background during offshore navigation. By default, the background shows the presence of altimeter data. Most charts show average wind and wave parameters based on satellite observations. So far, the only hindcast related chart involved the wave model grid points. We now added charts of average wind speed (and direction), average significant wave height (and direction) and average mean wave period based on the global wave model (1992-2012; resolution 5/4x1 degree).

New features of version "(June 20, 2013)" were:

Add hindcast and satellite data for the year 2012.
Model data, altimeter data and scatterometer data from 2012 have been added to the database. Now the hindcast database covers 21 years (1992-2012) and the satellite database covers 28 years (1985-2012).
Upgraded the shallow water wave ray model (SWRT).
The wave ray model has been upgraded (from version 29aug2011 to version 29mar2013): In general, the upgraded wave ray model will provide somewhat higher nearshore waves if wave growth from open sea is switched on. Thanks to the introduction of the detailed nearshore depth chart, results will be less sensitive to the choice/distance of the offshore boundary . In addition, refinement of nearshore bathymetry is believed to give more realistic wave ray patterns.
Clarified voucher administration.
So far, voucher users were able to generate a result page whilst the number of vouchers avialble was smaller than the number of vouchers required for that result page. As a consequence, the number of vouchers available can become negative. On the next voucher order, the number of vouchers lacking would be subtracted from the number of vouchers ordered and thus be paid for with delay. Users find this confusing. Therefore, the current version of the service will only generate a result page if the number of vouchers still available is equal to or exceeds the number of vouchers needed for that particular page.
Minor improvements.
-In the time series text files, the hour field (0,3,6,...,21) no longer gets truncated.
-Added option to preferences to set the bin size (1s or 2s) for wave period in scatter tables.

New features of version "(Jul 10, 2012)" were:

Add hindcast and satellite data for the year 2011.
Model data, altimeter data and scatterometer data from 2011 have been added to the database. Now the hindcast database covers 20 years (1992-2011) and the satellite database covers 27 years (1985-2011).

New features of version "(Feb 06, 2012)" were:

Download of timeseries text files, both offshore and nearshore.
Comma-separated text files with complete (1992-2010) 3-hourly model timeseries of wind and wave parameters can now be generated, either offshore or nearshore, and downloaded to your own machine for further processing. Each record holds wind speed, wind direction, significant wave height, mean wave direction, zero-crossing wave period, mean wave period and peak wave period. The corresponding help page provides the details.

Time series text files can only be payed for with vouchers.

Changed datum for nearshore water-level and bed-level from MSL to LAT.
Nearshore water-level and bed-level are now taken relative to lowest astronomical tide (LAT) instead of mean sea level (MSL). This influences the way in which one would simulate tidal effects with the wave ray model. For consistency, LAT is now also used as reference level for the simple shallow water model. Nothing changes here as long as one uses the same datum for water-level and bed-level.

New features of version "(Jul 20, 2011)" were:

Added persistency analysis, both offshore and nearshore.
Based on operational limits, persistency analysis determines the extent and the occurrence of sufficiently long windows from model hindcast data for each month of the year, including the variation over the years for a particular month. The user specifies the set of conditions to be applied to selected wind/wave parameters and sets the minimum length of a window during which these conditions are constantly met. The analysis is capable of determining continuous periods of favourable or unfavourable conditions.

The analysis offers the following output options Each of the above output options results in one output page containing one table and one graph. The table lists the selected quantity per month and per year, including the monthly mean taken over all years. In the graph, the mean values and their interannual variation are plotted against the month of the year.

New features of version "(Jun 15, 2011)" were:

Replace wave model data 1992-2009 and add wave model data for 2010
Latest hindcast v3.52 now in database. The upgraded model covers 19 years, i.e. 1992-2010.

Updated and improved model calibration
Calibration of model wave parameters with satellite altimeter data has been further improved. At each model point, regression analysis is now based on co-located data from the years 1992-2009 (this used to be 1992-2004). Also, the processing of the satellite data used for calibration has been refined: Chapter 5 of the validation report shows that, relative to buoy measurements, model data improves as a result of calibration with satellite data.

Add satellite data for the year 2010
Altimeter and scatterometer data from 2010 have been added to the database.

Extended and enhanced output

New features of version "(March 28, 2011)" were:

Add directional roses
Directional wave and wind roses were added as an alternative, graphical representation of 2D scatter tables.

New features of version "(February 3, 2011)" were:

Add a more sophisticated shallow water model
In addition to the existing simple shallow water model (depth contours are taken parallel to the coast), we have now added a more sophisticated way ray model. The new wave ray model translates wave spectra from multiple offshore grid points towards the coast along waves rays derived from the underlying high resolution bathymetry map. As opposed to the simple model, the new model offers the option to switch on growth of nearshore waves due to local wind (blowing from sheltered directions), input of the waterdepth at the nearshore location is no longer required and wave propagation direction is found from the underlying bathymetry (no coast normal direction needed any more).

Remove extreme value analysis
To prevent ill-advised application of the results of extreme value analysis for design purposes, we decided to offer this type of analysis by means of in-house consultancy projects only.

New features of version "(May 14, 2010)" were:

Add global and regional wave model data
Add one year of global (2009) and two years of regional wave model data (2008 and 2009). Both the global and the regional model now cover 18 years, i.e. 1992-2009.

Add satellite data for the year 2009.
Altimeter and scatterometer data from 2009 have been validated against buoys and added to the database. For calibration and validation of the updated satellite data refer to chapter 3 of the validation report.

Allow the user to re-define the offshore grid point(s) at the nearshore location screen.
So far, when configuring the simple shallow water model through the nearshore location screen, the user had to return to the offshore location screen in order to change the offshore grid point. Now, both the simple model and the forthcoming wave ray model let the user change the offshore grid point(s) without leaving the nearshore location screen: by clicking on the map. In previous waveclimate versions, clicking on the map was used to let the user change the nearshore location. The nearshore location can of course still be changed but only by means of the editable text fields: in general one knows the exact co-ordinates of the nearshore location.

Now the nearshore translation is always based on offshore conditions supplied by the wave model.
Removed the option to translate SAR wave spectra to the coast in order to avoid lack of data.

New features of version "(June 24, 2009)" were:

Added the option to generate 13 scatter tables at once.
You can now generate a series a 13 scatter tables at once, i.e. in one output page you will see one 2D scatter table for each month of the year preceeded by a 2D scatter table based on samples from all months.

Improved annotation of nearshore output pages.
Unsheltered directions actually used by the nearshore wave model are now printed below tables and plots.

New features of version "(June 5, 2009)" were:

Add one year of global wave model data (2008)
The global model now covers 17 years, i.e. 1992-2008.
The regional model still covers 16 years, i.e. 1992-2007.

Add satellite data for the year 2008.
Altimeter and scatterometer data from 2008 have been validated against buoys and added to the database. For calibration and validation of the updated satellite data refer to chapter 3 of the validation report.

New features of version "(May 21, 2008)" were:

Add one year of global wave model data (2007) and three years of regional regional wave model data (2005-2007).
Previous versions contained 15 years of global model data (1992-2006) and 13 years of regional model data (1992-2004). Both the global and the regional model now cover 16 years, i.e. 1992-2007.

Add satellite data for the year 2007.
Altimeter and scatterometer data from 2007 have been validated against buoys and added to the database. For calibration and validation of the updated satellite data refer to chapter 3 of the validation report.

New features of version "(January 18, 2008)" were:

Upgraded version of both global (1992-2006) and regional (1992-2004) wave model.
Improved database compression in order to be able to store extremely high waves more accurately (e.g. waves above 20m in Gulf of Mexico caused by Katrina).
Latest hindcast v3.40 now in database, with improved wave period data.

New features of version "(June 8, 2007)" were:

Add two years of global wave model data.
The global wave model now covers 15 years, i.e. 1992-2006. Previous versions contained 13 years of global model data (1992-2004).

Add satellite data for the year 2006.
Altimeter and scatterometer data from 2006 have been validated against buoys and added to the database. For calibration and validation of the updated scatterometer data refer to chapter 3 of the validation report.

Improved scatterometer data 1992-2000.
Historical scatterometer data from Ers1 (1992-1996) and Ers2 (1996-2000) have been reprocessed: the CMOD4 geophysical model has been replaced by the CMOD5 model (developed by ECMWF and KNMI). This has improved the processing of high winds, i.e. high winds were underestimated by CMOD4 . In addition, the standard deviation of the error in scatterometer data (relative to buoy measurements) decreased from 1.31 m/s to 1.22 m/s for Ers1 and from 1.22 m/s to 1.16 m/s for Ers2. The relative error decreased by 1 percent for both satellites. Results of validating the improved scatterometer data against buoys can be found in chapter 3 of the validation report.

New features of version "(November 22, 2006)" of waveclimate.com were:

Adjustable wind speed class in 3D scatter tables.
In 3D scatter tables of wave height versus wave period per wind speed class, the width of the wind speed class is no longer fixed to 2 (m/s or knots): according to the user's preferences the wind speed classwidth is taken to be 0.4, 1 or 2 (m/s or knots).

New features of version "(November 10, 2006)" of waveclimate.com were:

Enhanced scatterometer data 2004-2005.
As of January 2004, we receive and use real-time Quikscat scatterometer data for the waveclimate.com database. We improved the filtering of this data to remove any remaining ambiguity in wind direction. This ambiguity emerged in some wind field plots. The impact of this change on statistics proved to be negligible.

New features of version "(May 19, 2006)" of waveclimate.com were:

Extend the five year period of global and regional wave models data to thirteen years.
The global and regional wave models now both cover the same period of 13 years (1992-2004). Previous versions contained 5 years of global model data (1998-2002) while the wave hindcast data for the Mediterranean, the Black Sea, the Read Sea and the Caspian Sea covered the years 1997-2001.

Wave model input winds are calibrated beforehand.
The wind fields driving the wave model are calibrated by means of altimeter wind speed prior to running the wave model. This correction beforehand removes any systematic errors in wind speed. Up to now, model wind speed was calibrated by waveclimate.com after running the wave model.

Improved calibration of wave model wave height and wave period.
Wave model (and SAR) wave height and wave period are still calibrated by means of altimeter data. Wave periods are now scaled by the square root of the quotient of corrected and uncorrected wave height. In this way, the intercept applied to wave height implicitely affects the wave period. Previously, calibration of wave period depended on the scaling factor for wave height only. In addition, scaling of wave height is sligthly different near the origin to ensure a smooth transition between zero and the intercept. For details on calibration and validation methods and results, please refer to chapters 4 and 5 of the validation report.

Less variability in wind-sea/swell splitter for wave model spectra.
Output wave parameters can be related to the total spectrum, the wind-sea part of the spectrum or the swell part of the spectrum. The swell part of the spectrum is defined as the total spectrum minus the wind-sea spectrum. This wind-sea part of the spectrum is set to a Donelan-Pierson wave spectrum which depends on wind speed and wave age. We have now fixed the wave age of model spectra to 1.1, a value corresponding to nearly fully developed wind-sea spectra. Previously, the wave age varied with a somewhat unstable peak frequency measure.

Add satellite data for the year 2005.
Altimeter and scatterometer data from 2005 have been validated against buoys and added to the database. For calibration and validation of satellite data refer to chapter 3 of the validation report.

Upgrade of the graphical user interface.
The graphical user interface of waveclimate.com has been redesigned in order to
  • Get a more up-to-date look-and-feel
  • Remove the dependence on Java (previously, applets were used for menus)
  • Minimize the number of consecutive screens needed to get results
  • Clearify the structure of the site
  • Get a better help