First of all, I would recommend a group's Tracking and Recovery coordinator listen to the audio recordings of one of EOSS's flights. I decided to make a showcase of EOSS-74a for audio purposes. If you go to the recap for EOSS-74a you will find audio from the pre-flight net the evening before the flight to the recap net a few days after the flight, with audio of the actual recovery process in between. It is several hours in length total but may offer some unique insight into how we do our operations and generate some ideas on how you might like to run your Tracking and Recovery operations. We have by no means perfected the art of recovery, but we get the job done and you may find listening to this audio useful.
Ok, you've invested hundreds to perhaps thousands of dollars in fancy payload systems. You also may have customer payloads that they are expecting to recover after the flight. How is this accomplished? Below is a beginner's guide to recovery operations. Suggestions to augment this are welcome. Submit to the webmaster.
This scenario depends absolutely on two existing systems.
A third system is HIGHLY desirable, and that is an Automatic Position Reporting System (APRS). This is a transceiver, GPS receiver and a terminal node controller (TNC) combination that transmits the latitude, longitude and altitude of the payload periodically throughout the entire flight.
Recovery is essentially a three step process.
Use Balloon Track for Windows to predict the expected flight track of the balloon. You can also easily write your own prediction program. It isn't rocket science and might be a fun adjunct to your balloon groups activities. However, Balloon Track has been around for a while and is pretty good at the prediction angle.
The long winded explanation of how to generate a prediction is available in the Balloon Track Manual available on the download page for that program. However, here is a brief description of the process.
Start Balloon Track
In the setup area of the program check out each tab and enter in the information that pertains to your flight. The most important items are:
Once again, a detailed explanation of how to properly set up the program for operation is contained within the manual mentioned (and linked) above.
Using the menus in the program visit one of the web pages where you can obtain either predicted winds for your flight or the actual RAOB weather sounding for your area the morning of your flight. Save this data to a text file.
On Balloon Track's main screen click [File/Open] and point to the text file of the wind data you saved. Balloon Track should open the file, determine its source, load the proper parsing algorithm, decode the data, write out a Balloon Track formatted winds aloft file, and finally automatically run the prediction.
Click on the button marked "Synopsis" and you should see the final touchdown point indicated in both bearing and range from the launch site and latitude and longitude.
You can "export" the entire prediction by selecting from the main screen menu [File/Print/Flight Prediction]. If you want a disk file containing the prediction select "to file" on the dialog box that opens. If you want a hard copy, select the printer you wish to print to and click OK.
First, you should have a coordinator who manages the recovery team. This person should place the team in a dispersed pattern around the predicted landing point. EOSS runs an on the air net via amateur radio the night before a flight and a prediction is generated at that time and the coordinator then hands out assignments or at least general areas of assignment to the team members.
Many predictions have been made over the years and generally speaking, they almost all are within 20 miles of the actual touchdown point. The current average of all of the folks who are in competition for the record for closest prediction to actual touchdown is around 14 miles from predicted to actual landing. This covers 71 predictions. And, we're getting a bit better. However, predictions can be very strongly affected by various weather phenomena, most notably the passage of a cold or warm front.
Why tell you all that? Because your Tracking and Recovery coordinator needs to make some decisions regarding the disposition of his troops prior to the lift off of your balloon. If you have lots of folks you can lay out a two tier recovery zone. An outer circle of folks 20 miles from the predicted touchdown and an inner tier around 10 miles out. The only way to really know how to lay out your teams is to become familiar with the expected accuracy of your prediction person and the RDF capabilities of your team members. It is just something that has to grow organically within your own group. I would strongly recommend you attempt to gather a large (8 to 10 members) recovery team for your first few flights and just see how things go. As your group progresses, you will find you need them all, you can do with fewer or perhaps you need even a larger crowd to ensure a recovery.
EOSS likes to operate with a minimum of 4 people who are considered reliable in the RDF department by the coordinator. We also want a minimum of 2 people who can receive APRS and plot that data onto maps. One is all that is required but the second is the backup. The APRS stations can double as RDF trackers too as long as they dedicate themselves to primarily taking bearings via RDF and fall back on APRS as a backup. Each member of this small team must be VERY competent as there isn't much in the way of redundancy. It takes a minimum (obviously) of two stations to triangulate a position. If these 4 teams don't take their RDF role seriously and just buzz about the countryside failing to provide bearings, then the coordinator isn't going to be able to generate a known position for the balloon. EOSS always prefers to run with more folks. The more the better. This is because it allows us to nail down a small number of trackers to give bearings while it frees the others to move into new and hopefully better locations for final bearings.
These teams should be IN POSITION at the time your balloon launches. Why? Well, if during the flight you need to reposition them, that new position will be fairly close (hopefully) to their current position. You can usually get them all moved to new locations with a minimum of confusion. If all the teams start out at the launch site (a bad idea generally) then you will have told them all to go somewhere and then in the middle of the ascent phase of the flight, you will be on the move driving down the road and have to rearrange the entire deployment. This is difficult for the coordinator and, most likely, confusing to the tracking stations who are also on the move.
Why was starting the recovery process with all trackers at the launch location a bad idea? Suppose that the prediction says the balloon will land 50 miles away. Then, further suppose that the winds aloft have changed radically since the prediction and now the balloon is going to land 100 miles away. It could happen. And if it did, not one of the trackers would be in a good position to get a cross bearing as the balloon's payloads descended below the horizon in front of your car. Why? Well all of a sudden it is important for EVERYONE to get to the landing location quickly. They all take the most direct route. So, if you do actually take the time to do an RDF bearing while the balloon is in flight each and every tracker will be giving essentially the same bearing pointing down the same road everyone is following to get to that expected landing site. When you have a bunch of stations giving the same bearings you don't get a reliable triangulation fix on the balloon. It might be 10 miles ahead, or 100. There is no cross bearing to limit the distance!
OK, On to the recovery process ...
Ideally, your balloon should have a packet radio based Automatic Position Reporting System (APRS) that is beaconing out its current latitude, longitude and altitude every minute. If you do then it will be possible to compare the actual current position of the balloon to the position predicted for the balloon at that elapsed time, or perhaps better that predicted altitude.
However, suppose you don't have a position reporting system aboard your balloon. Then the job of determining its position will fall on the use of bearings reported from the Tracking and Recovery team to a coordinator who can collate all that information and generate a triangulation for the current position. A good triangulation will depend on the teams being nicely spread out as noted above. Once again, when the know position is calculated it should be compared with the prediction.
This comparison between expected/predicted behavior and the actual flight are important. Pre-launch it is a wise idea to have the Tracking and Recovery teams arrayed around the expected touchdown location. In this way, if the balloon actually goes where you expect, they will be well positioned to give bearings from a multitude of different vectors thus enhancing the accuracy of the triangulation computation. If the balloon looks like it is going to fly long or short or change its angle/vector of flight, you can move your teams in a coordinated manner to cover the new expected landing location.
The single most important bearing taken will be when loss of signal (LOS) occurs. Each Tracking and Recovery station should be advised when the coordinator believes the balloon is nearing the ground. Trackers should seek out an ideal bearing taking location (high above surrounding terrain, little EMI) and stay there. They should constantly maintain a bearing lock on the descending payload and note the direction from their location to the payload when the signal is lost.
Occasionally one or two stations will still be able to hear the payloads after they have landed. However, this is definitely not always the case.
Once the payloads have parachuted to the ground the radio direction finding capabilities of the Tracking and Recovery team will come strongly to the fore. While it is occasionally possible to follow a position reporting (APRS) payload right down to the ground, this is somewhat unusual. The more common scenario is that that signal is lost as the payloads descend into a landing location that is invisible (RF wise) to the entire tracking team. When that happens, it is up to the recovery coordinator to be armed with a decent idea of just where the payloads were just before they hit the ground. That is where that LOS bearing above come in as such an important step in the recovery process. Knowing the general area where he expects the payloads to have landed, the coordinator dispatches his teams to roads that closely surround the "area of interest". It is ardently hoped that someone will reacquire the signal of the payloads while they are cruising these close in roads. If not, we're in deep trouble.
Once a Tracking and Recovery station has received an RF signal from the payloads on the ground, the Tracking and Recovery coordinator uses this information to help him refine his earlier estimate of the touchdown location. He then attempts to get a station into a good position to provide a good cross bearing that will triangulate nicely with the first station. Once this cross bearing has been taken the touchdown location of the payloads is pretty firmly established and the T&R teams start to move towards this spot. If there are roads that will lead them in closer they take them, if not then they start the process of entering land on foot for a recovery.
Suppose you do have APRS and it is working the entire flight. Should you still be taking all these DF bearings? DEFINITELY. EOSS has had an APRS failure on several of our flights during 2004. These failures occurred at various phases of the flight and were all unexpected. Because the Tracking and Recovery team was already tracking by RDF, the loss of the APRS beacon was practically incidental to the smooth operation and recovery of the flight systems. When the APRS failed, we still knew exactly where the balloon was and were constantly updating our expectations as to where it would land. So, our advice is, run with both systems. Never assume that the APRS systems will survive the entire flight. If they do, then the RDF phase of recovery operations turns out to be a practice exercise for flights where things don't go so well. If there is a failure of APRS, then you team is already operating and no adjustments or harried setup of equipment is needed. They are already functioning just fine.
EOSS firmly believes in NOT TRESPASSING. If there is any indication we should NOT be on the land, a fence, a posted notice, any type of barrier at all, we start to canvas the "neighborhood" looking for the land owner to obtain their permission to enter their property and recover our payloads. EOSS has ALWAYS had positive results with interviews with property owners. They are usually quite fascinated by the entire affair. We have a brochure and a generic business card that is really nothing more than an ad for the EOSS website which we give to the land owners so they can feel we are slightly legitimate. And of course, lots of them visit the website after the flight and peruse our flight recaps to see their pictures. :-)
Edge of Space Sciences has a pretty good recovery record. It currently stands at 81 launched and 81 recovered as of July 28, 2004. However, do NOT be fooled into thinking this is the norm. Check out some of our Flight Recaps and specifically look for Photos of the touchdown sites. You will note that in our primary recovery zone, there is practically nothing to hinder our recoveries. No trees to speak of, not much variable terrain like deep valleys. Hey, EOSS lives in recovery heaven. We have even recovered payloads that inadvertently were separated from the parachute and payload train at burst up there at 100,000 feet and did NOT have any radio DF beacons proclaiming their presence. How did we do that? Well, we knew where the balloon burst and with NOTHING to obscure the ground in the area of search we just flew a plane over the area and the pilot and spotter aboard had no problem finding the missing payloads. Of course we were lucky. There are the occasional bushes about the landscape and the errant payloads might have rolled under them. You get the idea here, EOSS is going to have a pretty excellent recovery because we fly into such barren landing sites. You will have to adjust your expectations to your local geographic situation. You may need to add tools to your recovery toolkit. Maybe, a ladder, a boat, who knows? Only you. So, take this guide as something to keep you close in to the touch down location but for the final recovery, your situation will vary depending on your locale.