Irradiance values acquired via pvlib do not make sense
I have run my code and I get my graph. However, the values in the graph do not make any sense. Most of the time, the dni values surpass the ghi values, which is not supposed to happen. I also compared the result with the irradiation values acquired in the PVGIS and they are completely different.
Do I need to distinguish between true and apparent solar zenith angles in PVLib?
I wish to estimate DHI and DNI from GHI using pvlib.irradiance.erbs. I first compute solar positions at the location in question using pvlib.Location.get_solarposition. Then I use the zenith angles and weather data GHI with erbs. Afterwards I adjust for solar tracking with pvlib.tracking.singleaxis and finally compute irradiance with pvlib.irradiance.get_total_irradiance.
I notice that pvlib.tracking.singleaxis and pvlib.irradiance.get_total_irradiance take apparent solar angles, pvlib.irradiance.erbs takes true zenith angle, while pvlib.Location.get_solarposition does not specify which kind.
This makes me wonder if, and how, I should correct for apparent vs true solar angle between these functions? And if so, which kind should I use with pvlib.Location.get_solarposition?