By Steve Roti, first published in Paragliding Magazine, January 2001
This situation is generally referred to as "cloud suck". Different pilots have different definitions of what the term means, but for the purpose of this article let's use Dennis Pagen's description from his excellent book Understanding The Sky:
Close to thermal clouds the lift can get suddenly stronger in a phenomenon known appropriately as cloud suck. Many a thermaling pilot has found themselves fighting to stay out of the clouds. Cloud suck seems to occur most commonly in low pressure weather and especially in humid conditions.
Why should you care about cloud suck? Two reasons: legality and safety. The legal issues are fairly straightforward -- when we fly in Class E and G airspace under 10,000' MSL (which is where most of us fly most of the time) Federal Aviation Regulation 103 requires us to stay 500' below clouds. The only exception to this is Class G airspace 1,200' or less above the surface where the requirement is simply to stay clear of clouds. Cloud suck can easily pull a paraglider upwards at 1,000 fpm, so if we are flying 500' below cloudbase when we encounter cloud suck we may have 30 seconds or less to avoid whiting out. The safety issues are more complex. Much depends on where you are relative to the terrain and to other pilots. I'll illustrate some of the safety issues with four anecdotes from flights I had during 2000.
Anecdote 1: Woodrat Mountain, southern Oregon, 5/28/2000
During the Starthistle Meet conditions were looking good for cross-country so I followed three other pilots heading south toward Ashland. Launch is 3,800' MSL and cloudbase was around 7,000' MSL, with 3/8 cumulus cover. After the second valley crossing I arrived at the next hill just at ridge level underneath a medium-sized cloud. It took 10 minutes to find a usable thermal and start climbing toward the cloud, and by that time the other pilots were a mile ahead. The climb was slow until about 700' below cloudbase when it started accelerating rapidly. I glanced up at the cloud and saw that it was bigger than I had estimated from 3,000' below. Perhaps it had grown while I was climbing and watching the other pilots ahead of me. I pointed the glider toward the nearest patch of blue sky (opposite the direction of my intended cross-country route) and stepped on the speed bar. I flew out from under the edge of the cloud with 100-200' of clearance and skirted around the side to get back on course. From the side I could see the vertical development that wasn't obvious from below. The lesson here: keep an eye on the size of the cloud you're climbing toward, and if you can't see vertical development from below you can get a rough idea by watching other clouds in the vicinity.
Anecdote 2: Grand Bornand, French Alps, 8/8/2000
The second day of a week of superb paragliding weather in the Chamonix area. The 6,000' MSL launch was crowded when I arrived in the late morning and there was a cloud forming 1,000' directly overhead. The sky had only 1/8 cumulus cover and the clouds were not showing significant vertical development. Some of the local pilots were launching and thermaling straight up into the cloud. I went up to cloudbase, back down to launch level, and was on my way back up to cloudbase again when I decided that I didn't want to deal with the possibility of getting sucked into the cloud in close proximity to other gliders so I left the thermal and set off toward another peak. This illustrates what I consider one of the riskiest aspects of cloud suck -- if you get sucked in along with other pilots you won't be able to see them and there's a very real risk of a mid-air collision in the cloud. (Not to mention the possibility of IFR powered traffic in cloud, remember they're the ones who have a legal right to be in there.)
Anecdote 3: Plaine Joux, French Alps, 8/12/00
The sixth day of good flying weather, and we're just outside the Chamonix valley launching at 4,600' MSL and planning on flying into the valley to the Grand Montet. Cloudbase is over 11,000' MSL and the sky has 4/8 cumulus cover. Thermal climbs are good, mostly 500-800 fpm, and when I get to 10,000' I start on my way toward Chamonix. Half way there it's necessary to fly over a big ridge that forms the mouth of the valley. The problem is that a large oblong cloud has set up over the ridge and when I attempt to fly underneath it my vario instantly pegs. I turn around and head back a mile or so to watch the cloud and plot my strategy. There's a solid line of blue sky next to the cloud so 20 minutes later I return to the edge of the cloud, climb up to 10,500' and then fly into the blue line and parallel to the cloud until I get to the far end (at least a mile). But the story doesn't end there, because by the time I get to the Grand Montet the sky is 6/8 covered with cumulus and some of the clouds are starting to look tall. After a few minutes enjoying the view at the Grand Montet I head toward the Chamonix LZ, noticing on the way that it's awfully bouyant over the valley. Fearing that overdevelopment is on the way, I pull ears to lose altitude faster as I fly down valley. Sure enough, a little while after I land the cloud near the mouth of the valley, the one that tried to suck me up earlier, ODs and drops rain. Another lesson: a cloud big and strong enough to suck up a paraglider has the potential to overdevelop so keep an eye on it even after you've escaped its clutches.
Anecdote 4: Pine Mountain, central Oregon, 9/3/2000
I wasn't there for this one, but a friend of mine got sucked up into a cloud while flying cross-country during the Pine Mountain fly-in (for details see the incident reports column in the November 2000 issue of Paragliding magazine). Briefly, he got sucked into what he thought was just the corner of a cloud at 9,000' MSL and 15 minutes later found himself at 15,000' despite using big ears and speed bar to try to escape. Eventually he switched to a B-line stall and was able to drop out the bottom of the cloud and land safely. According to the pilot here are the mistakes he made: 1) not dressing warm enough; 2) not having warmer gloves; 3) not respecting clouds that are overdeveloping; 4) should have done a B-line stall much sooner; and 5) should have recognized being cold, hypothermic, and hypoxic and landed sooner.
How to avoid cloud suck
This brings us to the heart of the topic -- how to avoid cloud suck. In general I prefer to prevent white-outs from happening rather than being forced to react with an emergency descent maneuver after entering the white room, so I'll approach it from the perspective of things you can do to stay in the blue.
Prevention #1: Check the forecast and get an idea how likely it might be for cloud suck to occur. If the air mass is unstable or the forecast mentions the possibility of overdevelopment and thunderstorms, be warned.
Prevention #2: Keep an eye on the clouds before you launch and while you're flying. If a cloud looks big enough to have the potential to suck me up, I do my thermal climbs near the edge of the cloud rather than under the center. That way I can fly into blue sky whenever I want.
Prevention #3: If I find myself climbing under the center of a cloud I'll leave the thermal early rather than taking it all the way up to cloudbase. It's hard to know exactly how far I am below base, but if I'm under a small to medium sized cloud I try to leave at least 500' below base and under a bigger cloud I try to leave at least 1000' below base and fly toward the edge of the cloud. I try to do this before the vario pegs rather than after. I also try to fly toward the nearest edge or the largest blue hole, unless the nearest edge is directly upwind and the winds aloft are strong in which case I might choose to fly cross-wind toward one of the sides.
Prevention #4: As long as the lift under the cloud isn't excessively turbulent, I use speed bar to help me get out from under the cloud faster. This is the one time when we can be happy that our gliders don't have a flat polar curve. On most gliders the sink rate increases dramatically toward the upper end of the speed bar range, so in addition to going forward faster you're also going down quite a bit faster.
Prevention #5: If speed bar isn't enough, big ears are my next choice to increase descent rate. I usually disengage the speed bar, pull down the big ears, and then, if necessary, step on the speed bar again. I've never had any problem with frontal tucks while using big ears and speed bar, perhaps because big ears increases the angle of attack and speed bar decreases it effectively cancelling each other out.
Prevention #6: If big ears plus speed bar isn't doing the job, my last resort is a B-line stall. I've only needed to do this once in over 1600 flights. I don't consider a B-line stall to be something that should be done frequently because it does stall the glider and recovery from a stall can be unpredictable. Pilots have been injured and even killed releasing B-line stalls close to the ground when the recovery went bad, so unlike big ears, which can be held all the way to the ground in a pinch, B-line stalls should be released with enough altitude to deal with any problems that may occur (parachutage, asymmetric recovery, spin). (Note: some pilots prefer a spiral dive to a B-line stall, and other pilots have suggested the use of a full stall as a last resort to escape from cloud suck.)
Final thoughts: Prevention works best for me. If I can anticipate cloud suck I prefer to fly out from under the cloud sooner rather than be forced to do maneuvers later to escape. One 360 turn can make the difference between an easy escape and a close call or a white out. Be wary of any sudden increases in climb rate while thermaling under a cumulus cloud, particularly as you get closer to cloudbase, and leave the thermal before the vario pegs rather than after. Keep in mind that pilots who habitually try to get all the way up to cloudbase are more likely to get sucked into a cloud than pilots who follow the FAR 103 limitations and strive to stay 500' below cloudbase, so in this case there's demonstrable safety in staying legal.
This article is drawn from my experience with cloud suck, and I'm sure other experienced pilots will have different ideas and suggestions for how to avoid it. If you'd like to read more about cloud suck, see Dennis Pagen's book Performance Flying. Although the book was originally written for hang glider pilots, most of it applies to paraglider pilots as well.