I don't know hydrology but it seems that if you multiply each monthly average by the number of days in the corresponding month, you should have the monthly precipitation, so summing these over the year would give you what you want.

Mahboubeh Molavi-Arbshahi

The ERA5 data type is short. So, at First, using the "scale_factor" and "add_offset" attributes, you need to convert them to float:

short tp(time, latitude, longitude) ;

tp:scale_factor = 7.10748075218306e-07 ;

tp:add_offset = 0.023288371432603 ;

tp:_FillValue = -32767s ;

tp:missing_value = -32767s ;

tp:units = "m" ;

tp:long_name = "Total precipitation" ;

.

.

.

Total_precipitation= scale_factor*tp + add_offset

.

Precipitation units are meters. So, you need to multiply by 1000 to have mm.

Total_precipitation = Total_precipitation* 1000

.

And, at the final step, you must multiply them by the number of days in each month.

For example, in February,

Total_precipitation = Total_precipitation * 28

.

After these steps, add all months to calculate the total precipitation in a year.

Dear Morteza Babaei my file is netcdf to grib file

Also, I did this proposed formula before, because the initial tp is so small near 10^-4 when multiplying in 10^-7 the value be near zero so we have for all grid a fix value because of add_offset, my problem is I want to have total precipitation in whole Sea based on Km^3/year to compare with water flow, so I need a scale to multiply precipitation km to km^3/year

thanks for your help.

Mahboubeh Molavi-Arbshahi

It is impossible to convert m to km^3/year.