automated daytime and night time configuration for Kuopio

[ulysses78]

With the new script from NOA, the selection if to use the daytime or night time configuration is done automatically. However, for Kuopio it does not seem to work, as we have a lot of night time configs during times in July when it is still bright at your side! Did we miss any settings?

[HolgerPollyNet]

any idea @thgeorgiou ?

[thgeorgiou]

Hello, sorry for my late response, I was on summer break!

The program always uses the daytime configuration before 16:00 UTC which I now realize won't work around the globe.

I will implement something smarter to guess sunrise/sunset times depending on the location and let you know.

[thgeorgiou]

Hello again,

We thought about this with @mtsichla, the main issue is to have a way to make the program use night-time configuration when the Raman channels are active. Because the Raman channels get activated at some fixed time-of-day that changes seasonally (at least in our station), that will not always correspond to the sunrise/sunset time.

We plan on making the script check whether a Raman channel has data (ie. is not all zeroes) after the pre-trigger height. If that is true for the entire extent of an SCC file, it should use the nighttime configuration, otherwise it will use the daytime configuration. The pre-trigger height bin and which channel is Raman should be configurable per location/station.

Do you think this would suit all stations? Thanks in advance

[HolgerPollyNet]

Hi again,

the ideas sounds indeed good BUT sometimes the Raman channels are switched on even though they actually record only noise. Thus, for the decision if the day or nighttime configuration should be used, one would need to check the SNR and then decide based on the SNR profile - but in my opinion this should be a task of the SCC itself. So in my opinion, easiest for now is to let the user define which times should be which configuration or use the sun calculations allowing some offset, e.g. 30-60 min past sunset.....

[HolgerPollyNet]

To clarify, this holds for all Polly stations and thus would be important to have in the near future...

[HolgerPollyNet]

any plans, how to continue with this issue? @thgeorgiou @ulysses78 @kavoudou @mtsichla

[ulysses78]

@thgeorgiou ... using sun set calculations for the specific site (including an offset of 60 min past sunset) should be possible, don't you think?

[thgeorgiou]

Hello all,

I pushed a new version (1.14.0) that automatically calculates the correct sunset/sunrise times based on location and also includes new options (sunrise_time and sunset_time) to adjust either to a fixed time (HH:MM), or an offset (e.g., +12 or -14 minutes). Could you check if this works for your usecase?

Cheers!

[ulysses78]

Thank you! I will give it a try in the next days.

[HolgerPollyNet]

looks good from the preliminary assessment