DATE: Thursday, November 28, 2013
LAUNCH: 2:31 PM MST, Rainbow Tank, Espee Ranch, Hunting Unit 10, Arizona [ 35.669596, -112.56697 ]
WEIGHT: 7 lb (including chute, lines, and radar reflector)
ALTITUDE: 88,963 feet over Aneth, Utah [ 37.18287, -109.21595 ]
BURST: 9:20 PM MST near Cortez, Colorado [ 37.43870, -108.40226 ]
LANDING: 9:51 PM MST, San Juan National Forest, Colorado [ 37.58493, -108.11053 ]
TEMPS: High: 75.74 F (24.3 C), Low: -50.98 F (-46.1 C)
DURATION: 7 hours, 20 minutes
DISTANCE: 300 miles



The idea behind Launch 4 was to photograph the Grand Canyon from the edge of space, but clouds had made a view mostly impossible. It was still an adventure. The pictures of the Painted Desert, East Rim, and Little Colorado River demonstrated promise. When the balloon burst the mission suffered a parachute failure. The payload box plummeted 100,000 feet, hitting the ground in just eleven minutes. Amazingly, we recovered everything intact. I said to my brother, who had tagged along, this would make a great story in National Geographic. As it turned out he knew somebody who'd worked there.


A few weeks later I was exchanging emails with JJ Kelley, an acquaintance of my brother's friend. Some years before JJ had made a documentary while biking across Alaska and sent it off to National Geographic in Washington DC. They took him on as an intern. He worked there for five years, becoming a producer. Along the way JJ formed a production company with a friend called Dudes on Media. They produced several more adventure films, such as Paddle to Seattle, where he and his friend built their own kayaks and then spent months paddling the Inside Passage. Another recent work, Go Ganges, followed them from the headwaters of the Ganges River in the Himalayas to its mouth at the Bay of Bengal. Lately, he's gone undercover in Africa posing as an ivory trader and filmed artists on the shores of Alaska creating works out of plastic flotsam. JJ is an experienced adventurer and professional filmmaker. He considers himself an honorary Wisconsinite, having gone to high school there, but now lives in Brooklyn, New York. I was from Brooklyn and now lived in Wisconsin.

Intrigued, JJ offered to come along on my next outing. "I'll be a fly on the wall," he wrote. Logistically, it was simpler for me to drive somewhere nearby than fly to an unfamiliar place and organize everything. I had a number of new cameras and wanted to find an interesting subject where recovering the payload wouldn't be an ordeal. The Bad Lands of South Dakota seemed to fit the bill. Thanksgiving would afford several days off work.

Several weeks later an epic storm dumped four feet of snow on western South Dakota. If the weather was like this in October, the idea of launching balloons there in November suddenly didn't seem so smart. But a little research showed the weather in Arizona might be mild enough to go back to the Grand Canyon. Thanksgiving was over a month away. I called around the Phoenix area. No welding gas suppliers had helium to sell, but surprisingly I found a party balloon outfit with a ready supply and for a comparatively low price. Aware that some party stores mix nitrogen into their helium, I asked "Is this pure helium?" They assured me it was. I reserved two 220cf cylinders.

JJ was constantly on the road, one week in Alaska, another in Colorado. A few weeks before the trip I asked if he could come. As it turned out, he said, National Geographic magazine was running a story on the Grand Canyon soon. As a tie in, he pitched them the idea of a short video documentary following a near-space balloon mission over the park. A few days before the trip he got a text message. Story approved.

The Launch

I arrived in Phoenix on the evening of the 26th, a Tuesday. The party balloon shop was closing at noon on Wednesday and wouldn't be open on Thursday, Thanksgiving. In the morning I made my way through traffic over to the north-east side of town in a rented Chevy Suburban. The folks running the long-established business were friendly. I expressed my surprise at the relatively low price and again asked if the content was pure helium. Despite the "balloon grade" label, it was 99%, they assured. "Almost medical grade, just not filtered as much." Because they only stored and sold helium and carbon dioxide, they said, their insurance costs were lower. They could charge less. Also, they received helium directly from a supplier, no middlemen. New sources of helium in Wyoming and Qatar were coming on-line soon. And lately, I was told, the price had come down a bit. For years there's been a global shortage. Paying exorbitant prices was the norm. It seemed a stroke of luck. Leaving with two heavy cylinders in the back of the Suburban, I switched on the radio. It was Frank Sinatra singing Come Fly with Me...

JJ was flying out of New York that evening and arriving in Phoenix via Charlotte at around 12:30am. Accompanying him was his girlfriend Katy, who aside of having an unusual Thanksgiving break, would act as an associate producer. While Katy had visited the Grand Canyon many times, JJ had never seen it in person.

Days before arriving a monster storm had swept through the Southwest and was now on the verge of socking the Northeast. Sitting in the hotel Wednesday night, I watched in suspense as JJ's flight got delayed and then delayed again. If they didn't make the Charlotte connection we'd lose an entire day. I had to return the helium cylinders on Monday. Pulling off two launches in four days was already ambitious. To do it in three might be impossible. Finally, their flight left the ground. They'd have just twenty minutes to make their connection. The plane landed in Charlotte. A short time later I got an email. They were on the tarmac waiting to take off for Phoenix. But JJ didn't know if their baggage had made the transfer. I couldn't sleep. Sometime after midnight, JJ emailed again. They were in Phoenix, along with their bags. Relieved, I went to bed. We were to meet at 8am. I slept fitfully.

It was a five-minute drive to JJ's hotel. He and Katy appeared at the appointed hour, dragging a baggage cart piled high with suitcases, bags, and camera equipment. We'd only corresponded via email. It was odd to finally meet in person.

With everything packed in the Suburban, we headed north out of Phoenix on I-17. The plan was to do the first launch that afternoon. The forecast had been for clear skies, but changed to partly cloudy. The aim was to make flight apogee shortly before sunset. This meant a 2:30pm launch. Unlike Launch 4 back in July, the winds were heading from west to east instead of vice versa.

Time passed quickly. In a couple of hours we were in Flagstaff. JJ had just been there following a runner prepping for the New York City marathon. We headed west on I-40 to Williams, then turned north on 64 to Valle, where we'd strike out west on Wilaha County Road (coincidently the same way my brother and I had travelled to recovery the payload in Launch 4).

There was scarcely a wi-fi signal in Valle. The motel was closed. A weather update was impossible. The skies above had clouded over, but looking south it was all blue. The storm passing through a few days before had left twenty-eight inches of snow on the peaks around Flagstaff, but out here on the range there was none at all.

Headed west on Wilaha, the number of vehicles sharing the unpaved road was remarkable. In July, my brother and I hadn't seen anyone. Near the Tin House ranch compound twenty-six miles out, we turned south on Espee Road. Several miles down was a large cattle tank with high banked dirt walls, a perfect wind shield. We pulled off the road and drove a few hundred feet out to the pond. The sky was clear. From this spot the distance to the Grand Canyon was such that the balloon would float well above restricted airspace when it reached the skies over the park. We began unloading the Suburban. A car and then a truck drove down Espee. They slowed, but didn't stop.

JJ had rented tens of thousands of dollars worth of camera equipment. As I setup, he broke out a video camera, huge lens, and enormous tripod. A white truck went by, stopped, then backed up. The driver stared at us, seeming to consider. He made a turn our way. I walked out and intercepted him. He was a thin, stubbly man with a fifty mile stare. A rifle lay across the passenger seat. I explained what we were doing. "I thought you might be stuck in the mud!" he exclaimed. "Elk season is opening tomorrow." That explained all the traffic, and the gun. He lived near Williams now, but had spent thirty years running the gas station out in Cameron. "Well, it'll be an hour before we launch," I said. Fascinated, he wished he could stay and watch, but had to make camp.

Click to view largerEverything proceeded smoothly. I laid out a tarp on which to place the un-inflated balloon, unloaded a helium cylinder, set out the payload box, line, zip ties, duct tape, clamps, scissors, and a knife. JJ and Katy filmed the process. After an hour the skies had completely cleared. I was confident. The balloon now inflated and tied off, the payload stack attached, I started up four still cameras, two video cameras, a Spot GPS Tracker, and backup smart phone tracker. Heat packs were activated. For the first time I was using a data logger, generously donated by High Altitude Science of Colorado Springs. Their Eagle Flight Computer would track the latitude/longitude, altitude, barometric pressure, heading, and temperature. The Spot GPS only records latitude/longitude every ten minutes, and doesn't work at all above 60,000 feet. The Eagle functions up to 250,000 feet and would record data continuously.

Click to view largerThe payload box sealed, JJ positioned himself with the video camera. Katy shot still photos. I walked into the grass holding the payload box, the balloon tugging the line. I let go. Instantly, I saw something was wrong. Instead of rocketing into the air, the balloon drifted lazily from my hands (later calculation showed it was ascending at only 2 meters per second). I was perplexed. Either the helium cylinder hadn't been full or this "balloon grade" gas wasn't pure! Or was I just imagining things? There was no ground wind to speak of. In previous launches I always experienced a bit of a breeze. Could that account for the slow ascent? After Launch 5 it seemed like I finally knew what I was doing, but now I wasn't so sure.

The Chase

Ten minutes later, the gear packed up and driving east, we could still see the balloon in the sky. I said to JJ, "I'm not sure what's going to happen." To follow the balloon, we would need an internet connection in order to access the Spot GPS website for a status report.

The flight prediction assumed a 4.3 meter per second ascent rate and had the payload parachuting down on the Navajo Nation between Tuba City and Page, about eighty-five miles east of the launch site; a treeless area riddled with back-country roads. It seemed like finding the payload there shouldn't be too hard. If we overnighted in Tuba City and retrieved the payload by mid-morning, there'd be plenty of time to turn around, head back, and do another launch that afternoon.

Click to view largerTo reach Tuba City meant a trip through the Grand Canyon National Park. Since working there fifteen years earlier, I had not been back. It was surreal returning with a filmmaker from National Geographic in tow. Just outside the park in Tusayan, I hopped on the internet using an iPad. Checking the Spot GPS website, it was a great relief to see the balloon on the predicted path. Perhaps the unusual lift-off was due to a lack of wind after all? We continued into the park, where like a shock the first view of the Grand Canyon appeared at Pipe Creek Vista. The balloon was floating somewhere tens of thousands of feet above.

Click to view largerWe continued through the park towards Desert View on the East Rim. Just as the sun began creeping below the horizon, the light its golden best, we set-up at the crowded overlook and filmed a spectacular sunset. The rim of the canyon was covered in snow, the red rock below lit up. All around the sound of Japanese and Indian tourists echoed.

In the dark now, we headed out of the park and on to Cameron, where hopefully we'd pick up a wi-fi signal. A half hour later, under the awning at the gas station there, we tried getting online. No luck. The nearby Cameron Trading Post was closed, the motel shuttered. As I walked around the parking lot an employee happened to drive by. I explained our problem. He gave us the username and password for their network. "It works best in the garden out back."

JJ and I trudged down behind the motel. As I started the iPad, he filmed, hoping for a dramatic reaction shot. Hours had passed. The payload must have landed by now, I thought. The Spot website came up. The latest signal was twenty minutes old, from south-east of Page--red rock country. Generally, when the balloon rises above 60,000 feet the Spot goes silent. A long gap follows as it makes apogee, then three or four signals appear as the payload parachutes down. When it lands, the signals repeat from one place. I was puzzled. The Spot had gone quiet after the balloon crossed the East Rim and started signalling again forty miles to the north-east, as though having reached a pinnacle it was coming down. It followed the usual pattern, the trajectory nearly duplicating the prediction. But why no repeat signals from the landing site? We waited, refreshing the webpage. The Spot was mute.

Discouraged, we drove on to Tuba City. After settling into a motel, we checked the Spot site again. A half hour had passed. Nothing new. I wondered, could the box have hit the ground hard enough to dislodge the Spot from the gimbal keeping it face up and preventing communication? In the future, I thought, switching to radio tracking might be an idea.

Dispirited, we ate our Thanksgiving meal at Denny's, the only restaurant open. An hour later we found the Spot still hadn't updated. I couldn't understand. JJ and Katy went to bed. We had agreed to go to the last known location in the morning. If by chance we found it, great, otherwise we'd drive back west and try another launch. I was not optimistic.

Before turning in for the night I checked the Spot site one last time. An update! Without pause, I dashed out the door and roused JJ. We hurried back to the computer. Incredibly, the balloon was now over Colorado, heading toward the Rockies! But we had no idea if it was still going up, coming down, or worse, had become a "floater". Using the same balloon, a group in California had launched a mission that landed off the coast of North Africa. And while their payload could be measured in grams, the idea didn't seem so far fetched that given too little helium, ours might drift for hours. It was deja vu. Something similar had happened in Launch 1.

Every ten minutes the Spot updated, showing the balloon heading further into Colorado. It was getting late. I said to JJ, we'll have to see what happens in the morning. JJ went to bed. But I couldn't sleep, and watched and waited. Finally, the signals began coming from the same place. After three or four repeats, I was sure. Touchdown.

The Recovery

Click to view largerThe balloon had stayed aloft for seven hours and twenty minutes, covering three hundred miles. It crossed Arizona, Utah, and Colorado, and had landed 9,300 feet up in the foothills of the San Juan range, between Durango and Telluride--amazingly, just 3,000 feet from a road. Switching to a contour map view on Google, I could see a thousand feet of elevation gain stood between the road and the balloon. There'd be no trail. From the satellite image, the woods seemed thick. Snow could be a problem. And when we got to the top, there was no way of knowing if the payload was dangling in a tree. I turned in for the night, but barely slept.

We had agreed to meet at 8am. I knocked on their door at 7:45. Smiling, I said, "Grab your camera." Seated in my room, while JJ filmed, I explained what had happened, going over maps, pointing out flight details on the computer. A jet stream over northern Arizona had helped push the balloon east. We made our plans, packed up, and headed out. It was a four hour drive to Colorado.

The trip across the Navajo Nation on US Route 160 took us through Kayenta, Mexican Water, and Teec Nos Pos. The clouds were low, fog thick, the John Ford landscapes uncharacteristically covered in snow. In scenic spots where elsewhere you might find a million dollar home, there sat traditional hogans or trailers. In perspective, Phoenix seemed like another world, in a different time, a million miles away.

We stopped in Kayenta, a muddy red place, dilapidated pickup trucks, overcast skies. At a hardware store I purchased an axe and hand saw. JJ and Katy bought breakfast at a McDonald's. Near Teec Nos Pos we passed Red Mesa High School, an oddly isolated facility that looked more like a prison camp. A signboard extolled their football team, the Redskins. Passing briefly into New Mexico, we drove by the Four Corners Monument, marking the spot where Utah, Arizona, Colorado, and New Mexico meet. Still inside the Navajo Nation, the political lines didn't seem to carry so much weight. On the Ute Mountain Reservation in Colorado we stopped to gas up.

East of Cortez the road began steadily gaining elevation. We had thirty more miles to go, all on the San Juan Scenic Skyway, easily one of the most impressive drives in the state. When working at the Grand Canyon I'd taken exactly one road trip. It was this route.

The Skyway traced its way along the Dolores River Valley. A few miles past Priest Gulch Campground we turned onto Roaring Forks Road, an unpaved spur. Fortunately, it was sunny with temperatures in the 30s. Dense spruce covered the hillsides, aspen on top. Snow coated the ground, but no more than a foot in most places. A mile down the road we pulled over. The payload was on top of the hill opposite, above Roaring Fork Creek. JJ took one look and said, "It will take two hours to climb that." It was 1pm. Once on top, there wouldn't be much time to look around if we want to make it back to the car before dark. We didn't have the proper equipment to spend the night outside at 9,000 feet, in the winter.

Click to view largerNot wasting time, we set out, JJ lugging his camera, Katy following. But before we could ascend we had to climb down and ford the creek. At the bottom of the valley, across the icy stream, I took a waypoint on a handheld GPS. Then we started climbing. Deadfall littered the hillside. JJ suggested we take the steeper ridge line instead of following a debris choked ravine, which might end with a cliff. After fifty feet, I felt discouraged. Coming from Wisconsin, the altitude left me gasping for air. JJ, however, didn't seem affected at all. Wearing only a thin down jacket, no gloves or hat, he seemed untroubled, floating up and down the hill with his heavy camera, while I panted and leaned on the axe for balance. Deer tracks criss-crossed the snow, suggesting the easiest way forward. When one of us was out of sight we kept in contact by shouting. After an hour we'd made it half way. The sky was still blue, the forest absolutely quiet. Standing there, catching my breath, I felt at one with the scene.

Fueled on bananas, granola bars, and chugging water, we continued up and up. The dot representing the payload on the GPS device always seemed a third of a mile away. After reaching a false summit, the terrain levelled out. Finally, we reached the top, Katy a bit behind. JJ setup his camera one last time. This was the money shot. He didn't want to miss the moment I found the payload, whether it be lying on the ground or hanging from a tree.

Click to view largerHolding the GPS in my hand like a homing device I trudged through the snow, scanning the tree tops for any glimpse of orange, the color of the parachute. Suddenly JJ said, "I think you're going to be happy!" I looked around, didn't see anything, then there it was, sitting upright in the snow a few yards ahead, and completely intact: payload box, radar reflector, parachute, and the entire balloon, which instead of exploding into fragments had simply burst at the seam and come down still attached to the chute. It was a huge relief. While I opened the payload box, and JJ filmed, Katy appeared.

The Aftermath

Click to view largerGetting down the hill was a treacherous ordeal. After two hours we were back at the Suburban, exhausted, our feet frozen. We spent the night in Cortez. The following day saw us on US 160 heading back to the Grand Canyon. Along the way JJ became intent on filming wild horses. After a lunch of Navajo Tacos at the Cameron Trading Post, we returned to the South Rim and checked into Yavapai Lodge. For dinner we ate at the Bright Angel Restaurant, where I had worked years before.

Click to view largerEarly Sunday morning there was a knock on my door. It was JJ. Time to shoot the sunrise! We drove through thick fog to nearby Yavapai Point. Not sure there'd be anything to see; upon arriving we were astonished to discover a temperature inversion. The skies above were bright blue, but the entire canyon was filled rim to rim with rolling clouds. As we watched, mist swept up the canyon walls like a wave and then quickly and silently receeded, uncovering the vast hidden depths. Looking east was like peering over an ocean of cloud, a single butte sticking up like an island, the sun rising above, lending color to the scene. Later we discovered the episode was even rarer than thought. It was a total inversion, covering the entire canyon, according to the Park Service a once in a decade event. It even made the news. And by chance someone from National Geographic was there to film it.

Click to view largerOn the way back to Phoenix that afternoon, we stopped in aesthetic Sedona to film an interview segment. We found a nice backdrop near Red Rock State park. An hour and half later, back in sprawling Phoenix, it felt as though returning to America from another country. I dropped off JJ and Katy and headed to a hotel near the airport. That evening, packing everything up, I found the axe and saw wouldn't fit in my luggage. I left them in the trash bin in the room.

After two attempts I still hadn't quite captured the view of the Grand Canyon I'd been looking for. Every launch has been a learning experience. There are many camera variations to try. Clearly, it will take a third expedition.


2000 gram balloon from High Altitude Science

2000 gram balloon from Kaymont sells Totex balloons from Japan, the most popular brand

50" Parachute from Top Flight Recovery in Spring Green, WI

Davis Emergency Deluxe Radar Reflector

  • Cost: already owned
  • Website: purchased on
  • Note: per FAA FAR 101 regulations. You could easily make your own with some cardboard and tin foil.

Spot GPS Personal Tracker Device

  • Cost: already owned
  • Website:
  • Note: You can do without this if you use a smart phone that has excellent coverage, but you run a risk without a backup device.

Samsung Android Smart Phone from US Cellular

  • Cost: already owned, $70 for 1 month of service
  • Note: Failed as a backup due to a lack of coverage in the landing zone

AccuTracking App for smart phone

Eagle Flight Computer from High Altitude Science

Go Pro Hero 3+ Black camera

  • Cost: $422.39
  • Website:
  • Note: Easily the most commonly used camera for a project like this. Used a SanDisk 64gb SD memory card ($44.35) as well as the additional BacPac battery ($48.70) and GoPro 3.5mm Mic Adapter ($12.90).

Anker Astro Mini 3000mAh External Battery Pack

  • Cost: $19
  • Website:
  • Note: I bought the extra battery pack for more recording time for the Go Pro. The Go Pro will not record while connected to a powersource via USB unless you have the extra BacPac battery for the Go Pro itself and plug the USB cable into that.

Audio-Technica ATR-3350 Lavalier Omnidirectional Condenser Microphone

Pentax K-01 w/28mm SMC manual focus lens, positioned for landscapes, ten second intervals

  • Cost: already owned
  • Purchased off eBay

Canon PowerShot SX260 HS, positioned down

  • Cost: $152.51
  • Purchased off eBay

2 x Canon PowerShot A4000 IS, one positioned for landscapes, one for portraits

  • Cost: $181.97
  • Note: Using CHDK I found the battery will last almost 3 hours with the exposure interval set to 15 seconds and the camera settings tweaked to conserve power.

1 Proheat Reusable Hand Warmer, supersaturated liquid (doesn't require air to work) - used to heat the payload box

  • Cost: already owned

1 HotHands Air Activated Handwarmers

  • Cost: $.98

2 Adhesive backed HotHands Toasti Toes

  • Cost: $1.48

Miscellaneous including: duct tape, plastic zip ties, velcro strips, tarp, Everbilt 5/32 in. x 75 ft. Diamond-Braid Poly Cord (all from Home Depot)

  • Cost: $30

220 cf Helium

  • Cost: $150

GRAND TOTAL: $1,693.55

Airfare, gas, hotel, and food count as extra!
Cost: $2,004.54


CUSF Trajectory Prediction:

Spot GPS Map:

Click here to download the data (Google Earth KML file -- rename to .kml)

Eagle Flight Computer 3D Map:

Click here to download the data (Google Earth KML file -- rename to .kml)

Jet Stream Map (refresh page to see animation):


  1. Never buy helium labeled "Balloon Grade". Try and find a welding shop to get helium from, they could never pass anything off on their normal customers that was less than pure helium and still stay in business.
  2. It may be time to investigate using APRS radio tracking, which would allow realtime observation of the entire flight.
  3. Two of the four still cameras I used were accidently shut off when setting them in the payload box, take extra care to make sure the restraints holding them in cannot push on the power buttons.
  4. Save weight and don't bother using a backup smart phone for tracking. So long as the Spot GPS is fixed in a gimbal facing up, it is so reliable a backup may not be needed.
  5. Some cameras may be especially sensitive to cold, while others not at all. In Launch 5 the Pentax K-01 stopped operating normally. This time it worked throughout because I had a hand warmer set right beneath it and a smaller adhesive backed toe warmer on top.
  6. Too many of the pictures had more sky in the shot than desired. Instead of a straight horizontal aspect, the cameras should have been positioned at a slight down angle.




AT 105,000 FEET





© 2014 John Flaig. All Rights Reserved. Sitemap