Surface Radiation Parameters: Joule/m^2 to Watt/m^2

Radiation parameters for surface and single levels, as Surface Solar Radiation, are accumulated parameters. Data is Joule per square metre, accumulated over the respective period.

To get Watts per square metre, the accumulated values need to be divided by the time period in seconds over which the data has been accumulated:

  • For ERA5 daily data (stream=oper), the accumulation period is 1 hour (3600 seconds). Hence to convert to W/m2:

     SSR [W/m2] = SSR [J/m^2] / (3600 seconds)

    The result will be the average flux over the one hour ending at 'Step'.

  • For ERA5 daily ensemble data (stream=enda), the accumulation period is 3 hours. Hence to convert to W/m2:

     SSR [W/m2] = SSR [J/m^2] / (3 * 3600 seconds)

    The result will be the average flux over the three hours ending at 'Step'.

  • For ERA-Interim daily data (as for ECMWF operational data), the accumulation period is specified by the 'Step', in hours from the beginning of the forecast. Hence to convert to W/m2:

     SSR [W/m2] = SSR [J/m^2] / ('Step' * 3600 seconds)

    The result will be the average flux over 'Step' hours, from the beginning of the forecast.

For more information on the dataset see the ERA5 and ERA-Interim documentation.

Hi everyone,

what we do for multiple steps, e.g. Step: 3 to 12 by 3,


Hi Jafar,

As described above, the accumulation period depends on the dataset and you will need to take this into account when using the data. The documentation pages (e.g.

ERA-Interim documentation and ERA5 data documentation ) have detailed explanations of the temporal frequency and accumulations for these datasets.




for ERA5 reanalysis the data store simply uses validity times rather than expressing in terms of steps.  Can you confirm whether the time stamps refer to the beginning or end of the accumulation period?  I.e. if I want to create daily accumulations, do I sum from 00Z to 23Z, or from 01Z to 00Z-next day?



Hi Ruth,

For the hourly ERA5 data, the accumulated variables are  for the end of the accumulation period. Hence the daily accumulation totals will be obtained by adding the values from 01Z to 00Z-next day.

Hope that helps,


Thanks Kevin, that was my guess but I wanted to make sure!



Could someone confirm how to convert monthly radiation data from J/m2 to W/m2? I would have thought from reading above that I would need to divide by the number of seconds in each month, but when I do this the result is an order of magnitude out from observations.

Here is an example of my API request:

c = cdsapi.Client()








        'month':['01','02','03','04','05','06','07','08','09','10','11','12' ],




Many thanks,


The information I needed was right here:

i.e. "The energy (turbulent and radiative) and momentum fluxes should be divided by 86400 seconds (24 hours) to convert to the commonly used units of Wm-2 and Nm-2, respectively."

If I convert J/m2 to W/mthe reanalysis variable surface latent heat flux, will there be any difference with the variable mean surface latent heat flux? Or is it basically the same when you do the conversion?