Altitude choice for (apparent) solar zenith when relative airmass is specified?

[Mark Campanelli]

Greetings,

  I am looking at pvlib.irradiance.perez, whose docstring states

Perez models determine the diffuse irradiance from the sky (ground

reflected irradiance is not included in this algorithm) on a tilted

surface using the surface tilt angle, surface azimuth angle, diffuse

horizontal irradiance, direct normal irradiance, extraterrestrial

irradiance, sun zenith angle, sun azimuth angle, and relative (not

pressure-corrected) airmass. Optionally a selector may be used to

use any of Perez's model coefficient sets.

  My question is if the sun zenith angle is presumed to also be computed at zero altitude since the air mass is relative (i.e., not pressure corrected away from sea level)?

  Followup question: for my obvious test cases, the difference in zeniths (true or apparent) is very small, but maybe there are less obvious situations where they diverge significantly?

  Depending on the answer here, the most technically correct simulation would typically require computing two solar positions (one at site elevation and one at sea level), which would often have significant computational cost. For example, computations using pvlib.irradiance.erbs and pvlib.irradiance.aoi_projection presumably use the site's actual altitude, not sea level.

Thanks for any insight,

Mark

[Anton Driesse]

Hi Mark,

I do not think there is a connection between the two parameters in the way you suggest.  Both parameters feed into the empirical expressions for F1 and F2 so there is no real requirement for any logical consistency: the most important thing is to use the same as was used when the coefficients for those expressions were determined.  Incidentally this also means using the same air mass model that was used originally (see perez_driesse() helper function _calc_delta() ).

Note that altitude has a strong effect on (absolute) air mass, but only a secondary effect on (apparent) zenith angle because it only affects refraction.

Does that help?

Anton

--
You received this message because you are subscribed to the Google Groups "pvlib-python" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pvlib-python...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/pvlib-python/4d49cc5c-5595-4361-8481-3dfb7303c29dn%40googlegroups.com.

-- 
PV Performance Labs
Emmy-Noether-Str. 2
79110 Freiburg
Germany

+49-761-8973-5603 (Office)
+49-174-532-7677 (Mobile)

www.pvperformancelabs.com

[Mark Campanelli]

Thanks Anton. Helpful indeed. I followed up with https://github.com/pvlib/pvlib-python/issues/2879.

Mark Campanelli

LinkedIn

To view this discussion visit https://groups.google.com/d/msgid/pvlib-python/4a956bcc-3e5c-4e2b-80e6-26942577001c%40pvperformancelabs.com.