Irradiance data validity check too restrictive?

Dear SAM team,


I have encountered some errors relating to invalid irradiance data recently. The specific message I got are of this type:
out of range global irradiance 1513 W/m2 at time [y:2010 m:5 d:26 h:13], set to zero

Due to cloud enhancement, it is actually possible that the available irradiance can exceed the extraterrestrial irradiance for minutes at a time in extreme cases, see related reads below for readers of this thread. In my case I was simulating the system with minute data, containing such extreme values.

I agree that this message should be thrown to point people to possibly invalid data. However, I think it would make sense to not set the irradiance to zero in that case, but to run the simulation with the actual data, maybe limited to cases where the time-resolution is sub-hourly.
What do you think?


Best regards,
Timo


Related reads:
www.sciencedirect.com/science/article/pii/S0038092X16001079
www.sciencedirect.com/science/article/pii/S0960148115302913

Dear Timo,

Thanks for your question! This is a very interesting point, and not one that we had thought to address. We will have to check to see if it breaks any of the subsequent algorithms to have values that exceed the extraterrestrial irradiance, but if not, it seems like it might be reasonable to allow the simulation to complete with these flagged high irradiances, particularly for subhourly simulations, as you say. I will add it to our to-do list to investigate further, and we will keep you posted on the resolution!

Best regards,
Janine

Hi Janine,

Thanks for adding this on your to-do list!
For now I am using a workaround by limiting all irradiance values in the meteo file to the permitted maximum value of 1,500. That value is typically not exceeded all too often, so it should not lead to much of an error in the final result. However, of course a proper fix would be much appreciated!

Best regards,
Timo