Board index FlightGear Development Weather

Discontinuity in the weather

Everything related to weather simulation, visuals should be discussed in the shader subforum.

NOAA/GRIB data; (was Discontinuity in the weather)

Postby Richard » Fri Dec 07, 2018 1:41 pm

Just been having a look at this NOAA/GRIB data and found http://www.zygrib.org/#section_linux - from a quick look at the sources (GPL) it seems to have the basics for decoding the GRIB format files; although I think it uses an intermediary server to interface with NOAA

From a quick inspection it seems like this data (providing that it stays online) could be quite useful;

I'd like to hear Thorsten's opinion on how useful the NOAA data could be, not just the winds but any of the data (current and forecast). I'm also going on the record to say that it is my opinion that AW is great.

What interests me here is expert and scientific views on what could be possible with this data; rather than opinions about how great this data could be if it worked; i.e. the feasibility not the results.

This is the NOAA page http://www.nco.ncep.noaa.gov/pmb/products/gfs/

This is zygrib running; the data for this area was ~1.5mb; parameters as follows:
* lat (82, -43), lon (-158, 179)
* resolution 2
* interval 24h
* period 1 day
* NOAA-GFS standard (all)
* NOAA-GFS altitude (all).
* FNMOC-WW3 waves (all)

Image
Richard
 
Posts: 680
Joined: Sun Nov 02, 2014 10:17 pm
Version: Git
OS: Win10

Re: Discontinuity in the weather

Postby Richard » Tue Dec 18, 2018 4:52 pm

To test this I'm taking the F-15 from Edinburgh across the globe high and fast (FL615 M2.5)

I've seen a few instantaneous changes in pressure altitude of over 1000feet; and it doesn't appear to be related to close by METAR stations as the first few times the console was telling me that there were no stations within 100nm.

Managed to capture one of these on the properties graph and a map of about where I was at the time.

Image

Image

I appreciate that sometimes when flying through clouds or other weather conditions there is actually a rapid change in pressure - but this looks to me to be a fault in the atmosphere/environment simulation rather than a legitimate interpretation of the weather conditions. I shall investigate further; I suspect the method PT_vs_hpt in atmosphere.cxx

-----------------
[1] 6944.52 [WARN]:environment FGEnvironmentMgr::update: No airport within 100NM range
Richard
 
Posts: 680
Joined: Sun Nov 02, 2014 10:17 pm
Version: Git
OS: Win10

Re: Discontinuity in the weather

Postby Thorsten » Tue Dec 18, 2018 5:24 pm

I've seen a few instantaneous changes in pressure altitude of over 1000feet; and it doesn't appear to be related to close by METAR stations as the first few times the console was telling me that there were no stations within 100nm.


Looking at the graph, I agree - there doesn't seem to be an actual change in pressure (the value returns to what is used to be after a minute), it looks more like a temporary perturbation which is then faded out.

If this is METAR-driven, I'd expect to see that it changes to a different value in the end (and we'd need to have a new METAR report in the first place...).

Weird.
Thorsten
 
Posts: 10511
Joined: Mon Nov 02, 2009 8:33 am

Re: Discontinuity in the weather

Postby Richard » Tue Dec 18, 2018 6:12 pm

Thorsten wrote in Tue Dec 18, 2018 5:24 pm:Looking at the graph, I agree - there doesn't seem to be an actual change in pressure (the value returns to what is used to be after a minute), it looks more like a temporary perturbation which is then faded out.


I think that the pressure is returning because the autopilot took the aircraft back to 61500 feet (indicated); you can see this from the red line which is the actual altitude from /position.

but as you say it is weird whatever is happening.
Richard
 
Posts: 680
Joined: Sun Nov 02, 2014 10:17 pm
Version: Git
OS: Win10

Re: Discontinuity in the weather

Postby legoboyvdlp » Tue Dec 18, 2018 7:59 pm

Just at a quick glance, is it possible that it's going from whatever the pressure was to standard pressure? If you notice seemingly the position/altitude was at 63500 - what seemed to happen was the pressure changes (normalizing to standard atmosphere...?) and the indicated / actual altitude move closer together?
User avatar
legoboyvdlp
 
Posts: 6515
Joined: Sat Jul 26, 2014 1:28 am
Callsign: YV-LEGO
Version: 2018.3.1
OS: Windows 10 HP

Re: Discontinuity in the weather

Postby Richard » Wed Dec 19, 2018 2:38 am

It does appear to be when a new value for environment/pressure-inhg comes in from a METAR.

Image

(the flight tracking starts at the top of the Hudson Bay and definitely the reported altitude does seem to be a bit spikey)
Richard
 
Posts: 680
Joined: Sun Nov 02, 2014 10:17 pm
Version: Git
OS: Win10

Re: Discontinuity in the weather

Postby Thorsten » Wed Dec 19, 2018 7:02 am

It's difficult to see what is happening without a time scale on the x-axis.

Certainly the pressure and wind change potentially /fast/ when a METAR comes in - at Mach 2.5 you have 53 seconds to cross each weather tile, so if we ramp steeply betwene values (say within half a minute) that's completely expected and by design. The question is whether this happens instantaneously, i.e. apply a huge jump really within one frame to the next.
Thorsten
 
Posts: 10511
Joined: Mon Nov 02, 2009 8:33 am

Re: Discontinuity in the weather

Postby Richard » Wed Dec 19, 2018 2:13 pm

Thorsten wrote in Wed Dec 19, 2018 7:02 am:It's difficult to see what is happening without a time scale on the x-axis.

Certainly the pressure and wind change potentially /fast/ when a METAR comes in - at Mach 2.5 you have 53 seconds to cross each weather tile, so if we ramp steeply betwene values (say within half a minute) that's completely expected and by design. The question is whether this happens instantaneously, i.e. apply a huge jump really within one frame to the next.


For all intents and purposes it's instant, i.e. within a few frames and within a second; it's fast enough that I can't determine how long it takes as the property monitor and indicated altitude on the HUD both seem to change pretty

Probably if the pressure change did ramp in over a 30 to 60 second period it would be fine. I don't think the probably is inside AW - I think it's something in the interaction between the Metar, FG Environment and possibly the Nasal that is causing an unintended effect, but I won't know unless I figure it out.
Richard
 
Posts: 680
Joined: Sun Nov 02, 2014 10:17 pm
Version: Git
OS: Win10

Re: Discontinuity in the weather

Postby V12 » Wed Dec 19, 2018 11:44 pm

In my recorded data You can see discontinuity when METAR station has been changed. I did not observed what is happen when METAR station is not changed, but report from that station has been changed.
Fly high, fly fast - fly Concorde !
User avatar
V12
 
Posts: 1227
Joined: Thu Jan 12, 2017 4:27 pm
Location: LZIB
Callsign: BAWV12

Re: Discontinuity in the weather

Postby Thorsten » Thu Dec 20, 2018 7:19 am

I don't think the probably is inside AW


It might still be... Though the issue with no METAR in range triggering spikes would argue against the idea.

It smells like some weird emergent issue where two mutually exclusive assumptions are made by different parts of the code.
Thorsten
 
Posts: 10511
Joined: Mon Nov 02, 2009 8:33 am

Re: Discontinuity in the weather

Postby Parnikkapore » Thu Dec 20, 2018 10:13 am

AW seems to be temporarily applying a placeholder weather every time the METAR changes, which can be felt as a ~5s jolt...
There are free alternatives to every program you encounter. You just have to find them.
Parnikkapore
 
Posts: 807
Joined: Thu Oct 29, 2015 10:16 am
Callsign: HS-FGS
Version: next [PPA]
OS: Mint 18

Re: Discontinuity in the weather

Postby Thorsten » Thu Dec 20, 2018 1:37 pm

Definitely it does no such thing.
Thorsten
 
Posts: 10511
Joined: Mon Nov 02, 2009 8:33 am

Re: Discontinuity in the weather

Postby V12 » Thu Dec 20, 2018 4:49 pm

2018.3.1 AW error :
Image
Fly high, fly fast - fly Concorde !
User avatar
V12
 
Posts: 1227
Joined: Thu Jan 12, 2017 4:27 pm
Location: LZIB
Callsign: BAWV12

Re: Discontinuity in the weather

Postby Thorsten » Thu Dec 20, 2018 5:07 pm

Doesn't help me a thing without a recipe to reproduce it (and might be annoying but harmless under some circumstances). What were you trying to do, what happened in response?
Thorsten
 
Posts: 10511
Joined: Mon Nov 02, 2009 8:33 am

Re: Discontinuity in the weather

Postby V12 » Thu Dec 20, 2018 5:54 pm

It was absolutly standard short supersonic flight from EGLL to LOWI without any experiments - tracklog is here - https://fgtracker.ml/modules/fgtracker/ ... ID=7070462. AW did not stop to generate clouds, I did not observed any strange behaviour.
Fly high, fly fast - fly Concorde !
User avatar
V12
 
Posts: 1227
Joined: Thu Jan 12, 2017 4:27 pm
Location: LZIB
Callsign: BAWV12

PreviousNext

Return to Weather

Who is online

Users browsing this forum: No registered users and 1 guest