Radiosondes - Tracking weather balloons

While we claim some association with radio astronomy, we also harbor secret ambitions/dreams of optical astronomy activities.  That makes us active weather watchers and tracking weather balloons involves radio detection of radiosondes, so there is at least some synergy there.  By tracking weather balloons and being able to read the telemetry that the radiosonde transmits from 26km above South Australia, I know it could be cold outside because it was -62°C up there earlier today.

On any day of the year, the Bureau of Meteorology (BOM) realeases two weather balloons from major weather stations around Australia.  Several times a week, the BOM releases weather balloons from other regional locations around Australia.  These balloon flights are used to profile the atmosphere around the launch area, providing important information used for daily weather predictions, atmospheric research and weather modelling development.  In South Australia, there are weather balloon launches from Adelaide Airport at 8:45am and 8:45pm, Australian Central Standard Time.  There are regular (not daily) launches from Mount Gambier and Ceduna and occasional launches from Woomera and other locations, depending upon activities around that particular location.

This image of the BOM's Automatic Meteorological Balloon Launching System was lifted from a BOM webpage.  It shows the Mt Isa launch system, not the one located at Adelaide Airport.
If you want to see weather balloon tracking, you only need to download this free tracking software SondeHub Tracker to do so.  If you are wanting to actually track weather balloons will need some radio receiving equipment, a Raspberry Pi single board computer with some tracking software installed and (an optional commitment) some spare time to watch occasionally.  The radiosonde_auto_rx  free software is available here  and the setup/configuration instructions for it are available here .  There are a number of people actively involved in tracking weather balloons using the above tools and the same, or similar equipment, many of them appear to be licensed amateur radio enthusiasts.  Some of them are members of the Amateur Radio Experimenters Group  but membership is not required, neither is a radio operator's license required to participate. This activity does not require you to transmit radio signals, just receive them.

The image below is an example of a recent weather balloon flight from Adelaide in progress.

Of particular interest to us is identifying regularly occuring and spurious radio signals which we consider to be radio frequency interference (RFI).  The transmission frequency of the Vaisala RS41-SG radiosondes attached to the BOM's weather balloons around South Australia is 401.5MHz, that may not be the case elsewhere but that is someone else's RFI problem.  Our Phase 1 Spectrometer and the new Phase 4 Spectrometers both scan frequency ranges which include 401.5MHz, so for a couple of daylight hours each day of the year, we may have a strange signal appear in our data.  We know that this signal is likely to appear shortly after 2315UT and disappear approximately 2 ~ 3 hours later.  In order to determine the impact and characteristics of this regularly occurring source of RFI, Peter recently decided that it would be worth the effort to reconfigure the Phase 3 Spectrometers at Sunnydale to scan up to 450MHz, when there was a balloon flight heading out that way.  While 401.5MHz is outside the frequency range of the MWA antenna, the spectrometer should still be able to detect the signal from the radiosonde.  The Phase 1 Spectrometer was not used because it has been temporarily shut down for a while.
In the image below we can clearly see the transmissions from the radiosonde, they can be identified as a line of little dots which span the 15 minute observation, just above the 400MHz marker on the left of the image.  The e-Callisto PI Christian has carried out some additional signal processing to the data and has added this information and detailed images to his RFI catalogue for dissemination to the other stations of the e-Callisto network.  Note that there are some other sources of RFI in this image, at 280MHz we have observed a satellite passing and there is some other activity, probably military satellites, between 410MHz and 420MHz.

The last image shows the final stages of this particular weather balloon's flight path. The balloon has burst and the remaining balloon debris, radar reflector, tether and radiosonde are decending at 8.6 meters per second at an altitude of 7600 meters. The distance from our site at Sunnydale was about 8km at closest approach.  Shortly afterward, the remains of the balloon flight crash landed just across the river at Big Bend, north of the town of Nildottie.  That is another story altogether.