Have you read the documentation? Downward fluxes are positive, upward ones are negative, and they are accumulated from a starting point explained in the documentation.

Dear Jean, 

Thanks a lot for you explanation.

I read the documentation but I have doubt. I downloaded the data and each day has 4 times (0,6,12 and 18 UTC), my question is how can I "convert" this dataset to a "real value"? Do you understand me? I have 6 hours x 3600 second = 21600 second then I divide my value by 21600? Take a look at the file ssr1.png, they're the by factor x10^7, if I divide by 21600 the result is the file ssr2.png, but the values are very high to Surface net solar radiation variable. 

Guilherme Martins.

ssr is the net shortwave radiation at the surface, not the Rn you quote in your original message (Rn is the sum of the net shortwave radiation (ssr) and of the net longwave radiation).

I assume the values you gave in your original message are for a point somewhere in the purple area, with a ssr of about 98 Wm-2. 

Moreover the 7.1395e-11 is zero, likely to be the result of unpacking the original data.

Hello Guilherme,

As Jean-Jacques mentioned, these fields are accumulated from the start of the forecast (date and time in GRIB) to the validation date (date and time + step in GRIB). Can you check for which steps your fields at 00, 06, 12 and 18 are? If you have some fields for step=12,  you will have to subtract the field at step=6 from the same forecast from these, then divide by 21600.

If you still see odd values, can you tell us more precisely which data you've extracted?

Thanks you.

Dominique Lucas

ECMWF User Support.

Jean and Dominique thanks for help me.

I want to calculate the surface net radiation (Rn) as quote by Jean and the same I want sensible and latent heat flux, but is not clear to me the variables names. What is the name of  net longwave radiation variable? I always read the table 128 for search the variable name.

Thanks,

Guilherme.

ssr+str should give you Rn.

Hi Dominique,

I have this file. Thanks.

Guilherme.

Hi Guilherme,

I had similar issue with Le and SH fluxes, which I got for time 12:00 only with the forecast step =3.

So, what I did: a) applied scale and offset to the data given in the original file, and b) devidev by my forecast step 3*60*60 sec.

Values were physical after that,

Hope that helps,

Kind regards,

      Irina

Thanks Irina. Your informations are very important. 

Guilherme.

Hi Jean and everyone,

Is 'str' you mentioned stands for Surface net thermal radiation in ERA-interim?

Thanks a lot.

Guo

Updated link to the document describing the radiation fields in the IFS model (era-interim, era-5, ...:

https://www.ecmwf.int/sites/default/files/elibrary/2015/18490-radiation-quantities-ecmwf-model-and-mars.pdf

also accessible via the FAQ under:

https://www.ecmwf.int/en/forecasts/documentation-and-support

Dominique

Great post! Did someone calculate the net surface heat flux with ERA5 hourly data?, What variables are needed for the ecuación ?..I will appreciate any help.

*Qnet = Qsw(1-a)+Qlw+Ql+Qs))

*Qnet = short wave(1-albedo)+long-wave radiation+latent heat flux+sensible heat flux

List of variables available..

Clear-sky direct solar radiation at surface

Downward UV radiation at the surface

Forecast logarithm of surface roughness for heat

Instantaneous surface sensible heat flux

Near IR albedo for diffuse radiation

Near IR albedo for direct radiation

Surface latent heat flux

Surface net solar radiation

Surface net solar radiation, clear sky

Surface net thermal radiation

Surface net thermal radiation, clear sky

Surface sensible heat flux

Surface solar radiation downward, clear sky

Surface solar radiation downwards

Surface thermal radiation downward, clear sky

Surface thermal radiation downwardsTOA incident solar radiation

Top net solar radiation

Top net solar radiation, clear sky

Top net thermal radiation

Top net thermal radiation, clear sky

Total sky direct solar radiation at surface

UV visible albedo for diffuse radiation

UV visible albedo for direct radiation

Claudio, you might want to check further but I think Qsw is the surface net shortwave flux. So there would be no need to introduce the surface albedo.

Thanks Jean for your response, I'm using the attached function from the paper: Davis, P. E. D., Jenkins, A., Nicholls, K. W., Brennan, P. V., Povl Abrahamsen, E., Heywood, K. J., et al. (2018). Variability in basal melting beneath Pine Island Ice Shelf on weekly to monthly timescales.Journal of Geophysical Research: Oceans,123. https://doi.org/10.1029/ 2018JC014464 , maybe is not the best way tu calculate the surface net heat flux? Best regards from Chile.