NSSL/SWAMP Radiosonde Balloon Launch

Please be patient... there are 18 images on this page and they will take a few minutes to load...

The National Severe Storms Laboratory conducted a series of rawinsonde balloon launches from several locations in southern Arizona and northwest Mexico in July-August 1999. The purpose was to put a microscope to the summer monsoon, which often brings dangerous flash floods to the southwest desert regions.

A key part of the project is radiosonde packages. About the size of a milk carton and weighing less than a pound, these instruments are carried skyward by a helium balloon. During the ascent, this device transmits a datastream indicating its temperature, pressure, humidity, and location. These are recorded to a height of about 8 miles. Even after researchers stop listening to the data, the radiosonde continues up to 15 miles or higher, then the balloon bursts and a small parachute pops out of the radiosonde, slowing its descent to earth. Usually the devices are lost, but occasionally they are found and sent back to the National Weather Service for re-use.

It's 4:05 pm on a hot July day in Ajo, Arizona, and NSSL research assistant Shannon Key must have the balloon well on its way into the atmosphere by 5 pm. First the van is parked in an area away from buildings and power lines. The back parking lot of the Marine Motel is usually the best choice.

Shannon begins preparing for the balloon flight by starting up the data collection computers. Below her are several cylinders of helium gas.

The radiosonde is opened from its packaging, revealing a wet-cell battery.

This battery must be filled with water to activate it. The water is then discarded after 3 minutes.

When the radiosonde is ready to go, it is placed on a mounting box (the white box to Shannon's left). A 15-foot long wire serves to connect the radiosonde to the balloon, and as a radio antenna to receive triangulation data from LORAN stations in the region. To protect the antenna, it is strung out over the pole that Shannon is holding, and then is attached to the balloon, which will be inflated in the yellow tarp below.

The radio reception equipment is then tuned to receive the radiosonde signal.

Shannon begins inflating the balloon with helium.

When the cylinder PSI reading drops 500 psi, from 1200 to 700 psi, the balloon will have the correct amount of helium. Too much and helium will be wasted with the possibility of the balloon exploding before it reaches 8 miles. Too little and the balloon will take too much time to rise, possibly being carried out of radio range before it reaches 8 miles.

The balloon has been inflating inside the yellow tarp to protect it from rocks and pebbles. Shannon finishes inflating the balloon, sealing the neck with a tie. She then attaches the wire antenna.

Everything checks out, and Shannon lets go of the balloon, along with some ants crawling around on its surface. They'll have quite a ride.

The balloon lumbers skyward at about 1200 feet per minute, usually catching the attention of a few passers-by. It then recedes from sight after a few minutes.

Almost two minutes after launch, the data collection computer, compiling a line of data every 10 seconds, shows the balloon is at a height of 3500 ft MSL. The launch temperature was 99 deg F, but the balloon has now risen into pleasant 84-degree air. Ultimately the balloon (and its insect passengers) will encounter frigid minus-50 temperatures in the stratosphere. The computer also shows us that the balloon hasn't encountered any strong winds yet.

The signal is monitored. If there is radio interference, the radio is re-tuned slightly.

The balloon has reached 550 mb (about 17,000 ft), and the data collection computer shows a plot of temperature and dewpoint as the balloon continues rising. The lines zigzag erratically where reception was weak (due to light winds carrying the balloon right over the antenna). The software will filter this out for the final report. Altogether, it usually takes about 45 minutes from launch time for all the required data to be gathered.

The data, a valuable acquisition in this data-sparse region, is then submitted to the NSSL researchers and to area forecast offices for forecasting use. Shannon then gets a 10 hour break until the next launch!

Here's the rawinsonde report that was collected on this launch. It's coded in the standard WMO PILOT format (TTAA/TTBB/PPBB).

72767 TTAA  21002 72767 99951 36275 10504 00/// ///// /////
92690 32071 10502 85440 24865 30001 70094 09638 29003 50578
05968 20507 40748 18557 23008 30956 33775 21012 25082 43172
20016 20228 545// 21020 88/// 77999=

72767 TTBB  2100/ 72767 00951 36275 11950 36476 22938 33272
33900 29469 44665 06224 55603 01237 66583 00457 77556 01748
88508 04977 99470 09557 11437 13559 22402 18357 33394 19363
44377 21960 55300 33775 66221 49768 77194 559//=

PPBB  21006 72767 90123 09002 09003 02002 90467 02003 35502
28001 9089/ 31502 34505 91234 24004 32502 22003 91567 19011
19006 19003 9189/ 20513 21510 92014 22507 24007 22508 92567
23008 24508 22507 9289/ 23506 20506 93025 20510 20516 20014
938// 19517=

And here's the SKEW-T diagram for this launch. It shows that the billowy cumulus clouds in the photos were over 2 miles above the ground!

Return to Tim's personal page.