Radiosonde Launch Site - Launch Times & Parameters

[darksidelemm]

To help provide better predictions of radiosondes launches for all the launch sites that appear on the tracker map, we need information on:

What times of the day, and what days of the week radiosondes are launched. (e.g 00Z/12Z Monday-Friday)
What the target ascent rate, nominal burst altitude, and landing descent rates are. (e.g. 5m/s ascent, 26km burst, 6m/s descent)
Some of this information we can get from other sources (e.g. the giant database of radiosonde telemetry that is SondeHub), but local knowledge can help too!

If you have any of the above information for a launch site near you, please reply to this issue so it can be incorporated in.

EDIT (Luke Prior): Please use the Google Form your response will be automatically posted here.

https://forms.gle/GserZmFAif5z9QXH9

[darksidelemm]

As an example of what information we need:

• Launch Site: Mount Gambier, South Australia (Refer list here: https://github.com/projecthorus/sondehub-tracker/blob/testing/launchSites.json)
• Launch Times: 00Z, Monday and Wednesday
• Flight Profile: 5m/s ascent rate, 26km burst, 6m/s landing rate

[14ri004]

Hello,

Guipavas site (Brest -France)
0h00z 12h00z
Monday to Sunday
mounted at 5m / s
descent between 5m / s and 12m / s
Max Altitude 34000/36000 m

what do you mean by gust of 26 km ?????

[rs1729]

Here is the condensed ogimet data for 2021/06:
ogimet_2021jun.txt.gz

station: [day._timeZ]rs_type(launch_time) ...

[day._timeZ] is from the date/time-block right after TTAA/TTBB.
rs_type(launch_time) are from the blocks after 31313 at the end of TTAA/TTBB.

EDIT:
a quick json version:
soundings_2016jun.json.gz (date/time error)
(not sorted)

[yaakov-h]

I’m not sure how this would best be represented in the JSON that you linked to, but Sydney Airport (YSSY) launches are 6AM local time, which is UTC +10 or +11 depending on the time of year… just to confuse things 🙃

[darksidelemm]

For Sydney we're using an assumption of 18Z launches, which is close enough for prediction purposes.

[LukePrior]

Here is the condensed ogimet data for 2021/06:
ogimet_2021jun.txt.gz

station: [day._timeZ]rs_type(launch_time) ...

[day._timeZ] is from the date/time-block right after TTAA/TTBB.
rs_type(launch_time) are from the blocks after 31313 at the end of TTAA/TTBB.

Hey thanks for this,

I'm having a look through and trying to figure it all out:

04417: [01._11Z]42308(11:15) [02._11Z]42308(11:15) [03._11Z]42308(11:16) [05._11Z]42308(11:15) [06._11Z]42308(11:16)

So I'm assuming this refers to Geosummit (Greenland)

Then we have 5 launches listed which I'm assuming are all targeting 12Z models given the ~11:15 launch time.

I'm a bit confused with the dates so it would appear they were launched on days 1,2,3,5,6.

What does this mean, I see others have entries up until day 30 so I'm assuming it is data from one month?

So if the data stops at 6 does that mean it is not regular?

[rs1729]

@LukePrior
just made json output, just launch-date/time and radiosonde_type.
line by line.
soundings_2016jun.json.gz (date/time error)
I hope I didn't introduce conversion errors.

04417: [01._11Z]42308(11:15) [02._11Z]42308(11:15) [03._11Z]42308(11:16) [05._11Z]42308(11:15) [06._11Z]42308(11:16)

Station 04417, 01._11Z means day 01 in June at 11Z (that's the first raw block in the raw date, more important is the exact launch time at the end of the raw data for that sounding)
Jun 01 at 11Z, rs_type 23, launch time 11:15Z
Jun 02 at 11Z, rs_type 23, launch time 11:15Z
Jun 03 at 11Z, rs_type 23, launch time 11:16Z
Jun 05 at 11Z, rs_type 23, launch time 11:15Z
Jun 06 at 11Z, rs_type 23, launch time 11:16Z

(The "11Z" in the first block is not so useful, it's not really the synoptic time 12Z. Maybe there would be a problem for the midnight soundings, if 23:15Z -> 00Z the next day, the date would change.)

Now in json:

{ "station": "04417", "datetime": "2021-06-01T11:15Z", "rs_type": "23", "datetime": "2021-06-02T11:15Z", "rs_type": "23", "datetime": "2021-06-03T11:16Z", "rs_type": "23", "datetime": "2021-06-05T11:15Z", "rs_type": "23", "datetime": "2021-06-06T11:16Z", "rs_type": "23" }

Here the ogimet data for 04417 and 2021/06:
https://www.ogimet.com/display_sond.php?lang=en&lugar=04417&tipo=ALL&ord=DIR&nil=SI&fmt=html&ano=2021&mes=06&day=01&hora=00&anof=2021&mesf=06&dayf=30&horaf=23&send=send
The uwyo has less reports, maybe the soundings are not so regular:
http://weather.uwyo.edu/cgi-bin/sounding?region=naconf&TYPE=TEXT%3ARAW&YEAR=2021&MONTH=06&FROM=0100&TO=3023&STNM=04417

EDIT:
Don't know if the stations report the date/time all the same way, here is an example for a midnight sounding, where the launch time is probably 2021-06-09T23:53Z:

42971:  [10._00Z]52108(23:53) [11._00Z]52108(23:56) [11._12Z]52108(11:56) [12._00Z]52108(00:01) [12._12Z]52108(11:39) [13._00Z]52108(00:01) [14._00Z]52108(00:03)

https://www.ogimet.com/display_sond.php?lang=en&lugar=42971&tipo=ALL&ord=DIR&nil=SI&fmt=html&ano=2021&mes=06&day=01&hora=00&anof=2021&mesf=08&dayf=30&horaf=23&send=send
TTAA | 10/06/2021 00:00-> TTAA 60001 42971 [...] 31313 52108 82353=
In my json list it might be one day of:

{ "station": "42971", "datetime": "2021-06-10T23:53Z", "rs_type": "21", "datetime": "2021-06-11T23:56Z", "rs_type": "21", "datetime": "2021-06-11T11:56Z", "rs_type": "21", "datetime": "2021-06-12T00:01Z", "rs_type": "21", "datetime": "2021-06-12T11:39Z", "rs_type": "21", "datetime": "2021-06-13T00:01Z", "rs_type": "21", "datetime": "2021-06-14T00:03Z", "rs_type": "21" }

Perhaps I should compare the 00Z_ to the 23:53 if there is a wrap around...

However, if you just want the typical launch times of a station, this shouldn't matter much.

[LukePrior]

@LukePrior
just made json output, just launch-date/time and radiosonde_type.
line by line.
soundings_2016jun.json.gz
I hope I didn't introduce conversion errors.

04417: [01._11Z]42308(11:15) [02._11Z]42308(11:15) [03._11Z]42308(11:16) [05._11Z]42308(11:15) [06._11Z]42308(11:16)

Station 04417, 01._11Z means day 01 in June at 11Z (that's the first raw block in the raw date, more important is the exact launch time at the end of the raw data for that sounding)
Jun 01 at 11Z, rs_type 23, launch time 11:15Z
Jun 02 at 11Z, rs_type 23, launch time 11:15Z
Jun 03 at 11Z, rs_type 23, launch time 11:16Z
Jun 05 at 11Z, rs_type 23, launch time 11:15Z
Jun 06 at 11Z, rs_type 23, launch time 11:16Z

(The "11Z" in the first block is not so useful, it's not really the synoptic time 12Z. Maybe there would be a problem for the midnight soundings, if 23:15Z -> 00Z the next day, the date would change.)

Now in json:

{ "station": "04417", "datetime": "2021-06-01T11:15Z", "rs_type": "23", "datetime": "2021-06-02T11:15Z", "rs_type": "23", "datetime": "2021-06-03T11:16Z", "rs_type": "23", "datetime": "2021-06-05T11:15Z", "rs_type": "23", "datetime": "2021-06-06T11:16Z", "rs_type": "23" }

Here the ogimet data for 04417 and 2021/06:
https://www.ogimet.com/display_sond.php?lang=en&lugar=04417&tipo=ALL&ord=DIR&nil=SI&fmt=html&ano=2021&mes=06&day=01&hora=00&anof=2021&mesf=06&dayf=30&horaf=23&send=send
The uwyo has less reports, maybe the soundings are not so regular:
http://weather.uwyo.edu/cgi-bin/sounding?region=naconf&TYPE=TEXT%3ARAW&YEAR=2021&MONTH=06&FROM=0100&TO=3023&STNM=04417

I started going through all the data and manually updated the first 50 stations but I don't think I will have the time to do them all. The process might be able to be automated if I round all the times to the nearest 3 hours then check if it occurs every day or only once a week.

[rs1729]

@LukePrior
see my EDIT above, there might be some problems around midnight. Or maybe the hours are just rounded...
I made a new list.
soundings_2016jun.json.gz
Now it's

{ "station": "42971", "datetime": "2021-06-10 00Z (23:53Z)", "rs_type": "21", "datetime": "2021-06-11 00Z (23:56Z)", "rs_type": "21",  ... }

where 2021-06-10 00Z is from the first TTAA raw block, and (23:53Z) is the launch time.
Maybe it is just rounded, 23:29 -> 23Z and 23:31 -> 00Z next day.
De Bilt (06260) e.g.

"datetime": "2021-06-05 00Z (23:31Z)", "rs_type": "23", 
"datetime": "2021-06-05 23Z (23:29Z)", "rs_type": "23", 

{ "station": "06260", "datetime": "2021-06-01 00Z (23:36Z)", "rs_type": "23", "datetime": "2021-06-01 11Z (11:16Z)", "rs_type": "23", "datetime": "2021-06-02 00Z (23:33Z)", "rs_type": "23", "datetime": "2021-06-02 09Z (08:48Z)", "rs_type": "23", "datetime": "2021-06-03 00Z (23:35Z)", "rs_type": "23", "datetime": "2021-06-03 12Z (11:35Z)", "rs_type": "23", "datetime": "2021-06-03 23Z (23:28Z)", "rs_type": "23", "datetime": "2021-06-04 12Z (11:32Z)", "rs_type": "23", "datetime": "2021-06-05 00Z (23:31Z)", "rs_type": "23", "datetime": "2021-06-05 23Z (23:29Z)", "rs_type": "23", "datetime": "2021-06-06 23Z (23:25Z)", "rs_type": "23", "datetime": "2021-06-07 23Z (23:25Z)", "rs_type": "23", "datetime": "2021-06-08 23Z (23:21Z)", "rs_type": "23", "datetime": "2021-06-09 23Z (23:27Z)", "rs_type": "23", "datetime": "2021-06-10 12Z (11:31Z)", "rs_type": "23", "datetime": "2021-06-11 00Z (23:34Z)", "rs_type": "23", "datetime": "2021-06-11 23Z (23:21Z)", "rs_type": "23", "datetime": "2021-06-12 23Z (23:22Z)", "rs_type": "23", "datetime": "2021-06-13 23Z (23:27Z)", "rs_type": "23", "datetime": "2021-06-15 00Z (00:01Z)", "rs_type": "23", "datetime": "2021-06-15 23Z (23:27Z)", "rs_type": "23", "datetime": "2021-06-16 23Z (23:25Z)", "rs_type": "23", "datetime": "2021-06-17 12Z (11:31Z)", "rs_type": "23", "datetime": "2021-06-18 00Z (23:36Z)", "rs_type": "23", "datetime": "2021-06-18 11Z (11:29Z)", "rs_type": "23", "datetime": "2021-06-19 00Z (23:38Z)", "rs_type": "23", "datetime": "2021-06-20 00Z (23:45Z)", "rs_type": "23", "datetime": "2021-06-21 00Z (23:35Z)", "rs_type": "23", "datetime": "2021-06-21 11Z (11:14Z)", "rs_type": "23", "datetime": "2021-06-21 23Z (23:25Z)", "rs_type": "23", "datetime": "2021-06-22 23Z (23:19Z)", "rs_type": "23", "datetime": "2021-06-23 23Z (23:24Z)", "rs_type": "23", "datetime": "2021-06-24 12Z (11:34Z)", "rs_type": "23", "datetime": "2021-06-25 00Z (23:34Z)", "rs_type": "23", "datetime": "2021-06-26 00Z (23:41Z)", "rs_type": "23", "datetime": "2021-06-27 00Z (23:39Z)", "rs_type": "23", "datetime": "2021-06-28 00Z (23:41Z)", "rs_type": "23", "datetime": "2021-06-29 00Z (23:40Z)", "rs_type": "23", "datetime": "2021-06-29 11Z (10:48Z)", "rs_type": "23", "datetime": "2021-06-30 00Z (23:38Z)", "rs_type": "23" }

and 23:30 -> 00Z next day:

{ "station": "98223", "datetime": "2021-06-01 00Z (23:30Z)", ...

Don't know how precise you want the launch time data. For stations the do regular soundings for synoptic times 00Z, 12Z, 06Z, 18Z, it is probably easier to find the typical launch time. But some stations operate only on certain days of the week or so (e.g. Uccle, Hohenpeissenberg).

[LukePrior]

@LukePrior
see my EDIT above, there might be some problems around midnight. Or maybe the hours are just rounded...
I made a new list.
soundings_2016jun.json.gz
Now it's

{ "station": "42971", "datetime": "2021-06-10 00Z (23:53Z)", "rs_type": "21", "datetime": "2021-06-11 00Z (23:56Z)", "rs_type": "21",  ... }

where 2021-06-10 00Z is from the first TTAA raw block, and (23:53Z) is the launch time.
Maybe it is just rounded, 23:29 -> 23Z and 23:31 -> 00Z next day.
De Bilt (06260) e.g.

"datetime": "2021-06-05 00Z (23:31Z)", "rs_type": "23", 
"datetime": "2021-06-05 23Z (23:29Z)", "rs_type": "23", 

{ "station": "06260", "datetime": "2021-06-01 00Z (23:36Z)", "rs_type": "23", "datetime": "2021-06-01 11Z (11:16Z)", "rs_type": "23", "datetime": "2021-06-02 00Z (23:33Z)", "rs_type": "23", "datetime": "2021-06-02 09Z (08:48Z)", "rs_type": "23", "datetime": "2021-06-03 00Z (23:35Z)", "rs_type": "23", "datetime": "2021-06-03 12Z (11:35Z)", "rs_type": "23", "datetime": "2021-06-03 23Z (23:28Z)", "rs_type": "23", "datetime": "2021-06-04 12Z (11:32Z)", "rs_type": "23", "datetime": "2021-06-05 00Z (23:31Z)", "rs_type": "23", "datetime": "2021-06-05 23Z (23:29Z)", "rs_type": "23", "datetime": "2021-06-06 23Z (23:25Z)", "rs_type": "23", "datetime": "2021-06-07 23Z (23:25Z)", "rs_type": "23", "datetime": "2021-06-08 23Z (23:21Z)", "rs_type": "23", "datetime": "2021-06-09 23Z (23:27Z)", "rs_type": "23", "datetime": "2021-06-10 12Z (11:31Z)", "rs_type": "23", "datetime": "2021-06-11 00Z (23:34Z)", "rs_type": "23", "datetime": "2021-06-11 23Z (23:21Z)", "rs_type": "23", "datetime": "2021-06-12 23Z (23:22Z)", "rs_type": "23", "datetime": "2021-06-13 23Z (23:27Z)", "rs_type": "23", "datetime": "2021-06-15 00Z (00:01Z)", "rs_type": "23", "datetime": "2021-06-15 23Z (23:27Z)", "rs_type": "23", "datetime": "2021-06-16 23Z (23:25Z)", "rs_type": "23", "datetime": "2021-06-17 12Z (11:31Z)", "rs_type": "23", "datetime": "2021-06-18 00Z (23:36Z)", "rs_type": "23", "datetime": "2021-06-18 11Z (11:29Z)", "rs_type": "23", "datetime": "2021-06-19 00Z (23:38Z)", "rs_type": "23", "datetime": "2021-06-20 00Z (23:45Z)", "rs_type": "23", "datetime": "2021-06-21 00Z (23:35Z)", "rs_type": "23", "datetime": "2021-06-21 11Z (11:14Z)", "rs_type": "23", "datetime": "2021-06-21 23Z (23:25Z)", "rs_type": "23", "datetime": "2021-06-22 23Z (23:19Z)", "rs_type": "23", "datetime": "2021-06-23 23Z (23:24Z)", "rs_type": "23", "datetime": "2021-06-24 12Z (11:34Z)", "rs_type": "23", "datetime": "2021-06-25 00Z (23:34Z)", "rs_type": "23", "datetime": "2021-06-26 00Z (23:41Z)", "rs_type": "23", "datetime": "2021-06-27 00Z (23:39Z)", "rs_type": "23", "datetime": "2021-06-28 00Z (23:41Z)", "rs_type": "23", "datetime": "2021-06-29 00Z (23:40Z)", "rs_type": "23", "datetime": "2021-06-29 11Z (10:48Z)", "rs_type": "23", "datetime": "2021-06-30 00Z (23:38Z)", "rs_type": "23" }

and 23:30 -> 00Z next day:

{ "station": "98223", "datetime": "2021-06-01 00Z (23:30Z)", ...

Don't know how precise you want the launch time data. For stations the do regular soundings for synoptic times 00Z, 12Z, 06Z, 18Z, it is probably easier to find the typical launch time. But some stations operate only on certain days of the week or so (e.g. Uccle, Hohenpeissenberg).

I have now integrated this data into the file:

https://github.com/projecthorus/sondehub-tracker/blob/testing/launchSites.json

Please let me know if you see anything wrong

[markdconner]

For the US launch sites using LMS-6 sondes, the typical profile is:
Ascent rate: 5 m/s
Burst altitude: 33,500 m
Sea-level descent rate: 2.5 m/s

If you need it, I should be able to locate the lat/lon of all US NWS radiosonde sites. I don't know if I can filter for those using LMS-6 vs others, but might be able to make a stab at it with existing Sondehub data collection.

Launch times are normally 1100 and 2300 UTC. Occasionally (like now with a tropical system threatening the US East Coast), soundings will be added with launch times at 0500 and 1700 UTC.

[rs1729]

What does "times":["0:00:00","0:06:00","0:12:00"] mean?
Does "0:" mean "every day"?
What is "times":["3:00:00","4:00:00","4:12:00","7:00:00","1:00:00","2:00:00","6:00:00"] then?
Is 1=Monday or 1=Sunday or something else? I didn't see a description for this field.

Uccle usually does RS41-SGP+O3 soundings 3 times a week, Monday, Wednesday, Friday at 11:30Z.
There are no reports at ogimet.

  {"station":"06447","rs_types":["23"],"lat":50.7969,"lon":4.3581,"alt":100,"station_name":"Uccle (Belgium)","times":["0:12:00"]},

Not sure if "times":["0:12:00"] is right then?

One remark:
When airline traffic was reduced last year, some stations made additional soundings, e.g. Essen usually launched at 10:45Z and 22:45Z, now also at 04:45Z. Not sure for how long this will continue.
De Bilt used to have RS41-SGP launches every night at 23:30Z, and Thursdays at 11:30Z RS41-SGP+O3. I believe there are more soundings now.

[LukePrior]

Does "0:" mean "every day"?

Yes

Is 1=Monday or 1=Sunday or something else?

1 is Monday through to 7 is Sunday.

Uccle usually does RS41-SGP+O3 soundings 3 times a week, Monday, Wednesday, Friday at 11:30Z.

I will update this, the daily 12Z was from an old data source.

[LukePrior]

For the US launch sites using LMS-6 sondes, the typical profile is:
Ascent rate: 5 m/s
Burst altitude: 33,500 m
Sea-level descent rate: 2.5 m/s

Cheers I will get this updated to be the default for LMS6 sondes. We already have the lat/long for all US stations to 4 decimal places.