Computing ionospheric delay from a IONEX file

[AlekLee]

@luc

is is the case in particular for models that are based on a single layer for electronic content or that already assume everything has been integrated along the path, taking a mapping function into account.

There are models that do not make this assumption: NeQuick ITU and NeQuick-G (the second one is a variation of the first one, specialized for Galileo using a slightly different solar flux input with a quadratic polynomial and a slightly different integration method).

Thanks a lot for pointing this out! Incidentally, I already implemented the NeQuickItu model in my MWE script above last night (I’ll probably update it with a better, more comprehensive version later). For now, I’ve been taking the F10.7 flux from the CSSI weather file like so:

from org.orekit.models.earth.atmosphere.data import CssiSpaceWeatherData
from org.orekit.models.earth.ionosphere.nequick import NeQuickItu
space_weather = CssiSpaceWeatherData('SpaceWeather-All-v1.2.txt')
f107 = space_weather.getDailyFlux(epoch)
iono_model_neq = NeQuickItu(f107, TimeScalesFactory.getUTC())
# ...
delay_ground_neq = iono_model_neq.pathDelay(gps_tx_state, ground_topo_frame,
                                            epoch, frequency, None)

Do you know if using the daily flux is the correct way to go here? On the navipedia page (link), they mention “daily flux”, but it isn’t verry clear if that’s just a coincidence or an actual recommendation. The actual model is available from ITU here: https://www.itu.int/dms_pubrec/itu-r/rec/p/R-REC-P.531-16-202509-I!!ZIP-E.zip. In the README, they say that:

The sub-models contained in NeQuick2 use monthly average values of solar activity in two forms: twelve month running mean sunspot number R12 and 10.7 cm wavelength solar radio flux F107. The latter is considered to be the primary input parameter. The following relation between R12 and F107 is used.

This would point to not using the daily flux, but the mean flux instead: space_weather.getMeanFlux(epoch). In the Galileo model spec (link), they also mention the 12-month average. Other ITU recommendations (link1, link2) don’t even mention “flux” or “F10.7”, so I’m not sure. The documentation in the source (link) is no more verbose than what’s in the documenation. I’d greatly appreciate some advice here, @luc .

I noticed that the NeQuickGalileo ( NeQuickGalileo (OREKIT 13.1.2 API) ) and NeQuickModel ( NeQuickModel (OREKIT 13.1.2 API) ) use the effective ionisation level az, as defined in the Galileo spec (link). Is there a way to retrive this metric in a similar way to how one can get the Klobuchar parameters?

Here’s how the two models compare for my tow example, in case it helps anyone: