tlmquintino
commented
2 years ago The units of ssrd, ssr, fdir, strd, and strr as produced from (for example the ECMWF forecast) are (J m-2) as in this reference parameter database https://apps.ecmwf.int/codes/grib/param-db/?id=176
Then, as documented in that page, to obtain fluxes, you need to divided by the accumulation period expressed in seconds. This means that the thermofeel function calculate_mean_radiant_temperature takes in (W m-2).
So the documentation string is wrong and needs to be corrected.
The example code calc_utci.py and utci_wbgt.py deaccumulate ssrd, ssr, fdir, strd, and strr, which leads me to believe the units for these arguments should be fluxes (W/m^2 or J/sm^2) rather than accumulated radiation as stated in the docstring (J/m^2 or Jm^-2). From what I can tell, the Di Napoli paper says the MRT calculation is done with surface fluxes, not accumulated radiation.
Additionally, the docstring for calculate_cos_solar_zenith_angle_integrated says: returns average of cosine of the solar zenith angle during interval [degrees], but the example code in calc_utci.py and utci_wbgt.py "deaccumulate" the value returned by this function as well, which leads me to believe that calculate_cos_solar_zenith_angle_integrated returns the integral of the cos solar zenith angle over a period over time, rather than the average cza during the interval. I'm not familiar enough with the math used to answer this for myself.
To summarize:
ssrd, ssr, fdir, strd, strrincalculate_mean_radian_temperaturesupposed to be W/m^2, and incorrectly labelled as J/m^2?I am new to this so please correct me if I am wrong. Thank you for your help!!