Hi @lionpeloux - how did the Oikolab data compare?
There is a lot of technical details in the data source that you’d want to consider, rather than simplly ‘trusting’ a source.
- Radiation data for Climate One Building, JRC, Shiny, and Oikolab are all mostly based on ERA5 reanalysis dataset (JRC also uses NSRDB and SARAH2). We (Oikolab) provided the raw ERA5 data to Climate One Building but not sure how the data was compiled into EPW file.
- Note the EnergyPlus file for Lyon is mostly from 1980’s and early 1990s. Radiation data is not measured by weather stations, so these radiation values were probably empirically calculated, which leads to a lot of inconsistency. We generally recommend against using them, also because they are really outdated; the dates fall outside of the 30 year NOAA climate normals (1991-2020).
In general, I’d recommend going with a dataset that can provide traceability to the source data and the derivation method so you can verify things yourself.
Hope this helps.