Using real-world weather - what to do for Horiz Infrared Rad (Comfort Workflow)

ladybug-tools
, ,
1

Hi Guys,

I am trying to run the LB thermal comfort workflow with real-world weather that I am downloading from open-meteo api ( 🏛️ Historical Weather API | Open-Meteo.com ) to run the workflow with real world weather rather than the EPW weather.

For the workflow I need the following inputs from an EPW that I am substituting with the real-data, and I have done this with no issue:

air temp, RH, wind speed, wind direction

My problem is when it comes to radiation, these are the rad variables I can get from openmeteo:

image
image1091×228 10.5 KB

Diffuse and Direct are straightforward, but this does not include Horizontal Infrared Radiation. If you plot the two you can see there is nothing that appears to be this:

image
image1387×521 72.1 KB

From the EPW:

image
image886×747 52.9 KB

My work-around has been to just find a day in the EPW around the same time frame, and approx same weather, and use the radiation from the EPW and then the open-meteo data for the other variables. However I would like to stop doing that if possible.

Does anybody have a solution or a source to get this horiz_infrared data?

thanks!

2

hey, try using DF Horizontal Infrared component that uses Cloud cover and DB and WB temperatures as inputs to calculate the Horizontal Infrared radiation.

3

I feel so dumb for not finding this! thank you Dmitry.

4

In case it’s helpful, a couple of additional thoughts:

  • Open-Meteo’s Historical Weather API currently exposes shortwave, direct and diffuse solar radiation, but not downward longwave radiation in its documented hourly variable list.

  • Another alternative is Oikolab/ERA5’s surface_thermal_radiation, which is the downward atmospheric and cloud longwave radiation reaching a horizontal surface. This corresponds to EPW N12 Horizontal Infrared Radiation Intensity. Don’t use surface_net_thermal_radiation, since that includes the upward ground-emitted component. For hourly Oikolab values in W/m², the numerical value can be used as the EPW hourly Wh/m² value, provided the timestamps and hour-ending convention are aligned. Oikolab can also generate an ERA5 actual-year EPW directly.

    p.s. This was top of mind because I’ve just developed a new N12 morphing procedure in my tool FutureWeather.co (using rlds from the climate model data).