Back to Jérôme's Paragliding page / Technique.

Thermals, Part 2 of 3 :  Thermals and Clouds.  Part 1, Part 3.

by Will Gadd, 2001/1

This article is part two in a three-part series. Part one covered how thermals form and release from the ground; this article covers the relationship between thermals and clouds. The final article in this series will cover thermal flying techniques.)

            First, this article focuses on dealing with our best visible thermal indicators, clouds. There are dozens of books written on lapse rates, instability and the like, so the ideas presented here are more field rules for flying clouds and other sky-based clues rather than a meteorology text, please forgive the gross simplifications I make.

            The basis for understanding what’s going on in the sky comes from watching it; reading books (or articles like this one!) helps, but you need to have your own on-board sky-interpretation system to fly well. Every good pilot I know has spent literally thousands of hours looking at the sky and trying to figure out what’s going on up there. I have spent many blown-out days lying on my back watching the sky swirl over me, and these days are some of the most valuable time I’ve ever put into flying. Are the clouds being blown to bits? Do they remain relatively constant over set points or form over a set point and then drift off downwind decaying as they move? Do they cycle evenly, starting as thin whispies and then forming into ever-more solid masses before decaying, or do some pop up very quickly and then disperse slowly? Do they have hard, flat bottoms or a rounded, mushy appearance? Each answer to these questions provides a wealth of knowledge about the thermals that are generating these clouds. Clouds are infinitely variable, but I believe they do have patterns that can be learned by watching them.

The big concept here is that clouds cycle based on their attached thermals. As a warm air mass rises it eventually reaches an altitude where its moisture condenses out. This process continues only while the cloud is being fed by a thermal (condensation “pumps” basically act the same as thermals, so I’ll treat them the same here for simplicity). At some point the collector or pool of warm air on the ground is exhausted, but the cloud is still being fed by a “bubble”  rising above the ground. Eventually no more rising air feeds the cloud and it starts to decay; at this point there is no more lift under it. This is why many of the best-looking clouds often provide no lift when you fly under them; while pretty, they are at the end of their useful cycle. As clouds decay they will in fact usually produce sinking air, which is annoying if you’ve flown to one expecting an elevator ride back to base. What’s more useful is to connect with the rising air under clouds that are still forming. So how do you tell ‘em apart?

The simplest cloud game is to try and predict whether a cloud is forming or decaying; before doing this in flight, I like to play the cloud prediction game while mowing the lawn, driving, or looking out the office window. Pick one cloud and make a snap decision: is it forming or decaying? Then carefully track that particular cloud through the rest of its cycle; if you think it’s forming, it will grow in size (either vertically or horizontally or both) while becoming ever-more resistant to light (more suspended water means going from whispies to small “clumps” of moisture to solid white to gray). If it’s decaying then it will become ever lighter and slowly fragment into smaller pieces, How long does this process take? Two minutes? Ten? Twenty? Or does it just continue to develop into a monster cumulus savage-your-gliderus? I can seldom make good predictions based on just one look at a cloud, but after watching it for a couple of minutes I can usually tell which direction it’s heading. I believe that it’s absolutely basic to learn the life cycles of clouds if you want to fly XC; this is the aerial equivalent of knowing how to read.

Michael Champlain, one of the better XC pilots I’ve met, taught me a good trick to help understand what clouds are doing while you’re flying. He recommended taking a series of mental snapshots of the sky as I climbed in a thermal. With every circle I look downwind and take a quick “picture” of what all the clouds in my predicted flight direction look like; a long climb may allow for 30 or more good snapshots, and with minimal practice I have learned to memorize which clouds are forming and which decaying based on these snapshots. Over the course of a few climbs my snapshots also give me good clues on how long the clouds are lasting, information which then tells me which ones may still be forming after I glide to them. If the cloud cycles are lasting 30 minutes then I can glide for 10 or 15 minutes and still arrive at a growing cloud with plenty of time to catch a ride. Generally, the more distance between clouds the longer they will last (a larger volume of air is feeding into a single cloud), and the higher cloud base is. If you go on glide toward a cloud that has been forming for 30 minutes and arrive low, the odds are slim that you will find lift no matter how beautiful the cloud over your head is. Many pilots make the mistake of climbing to base, then looking around and heading for whatever cloud looks “best,” regardless of where it is in its life cycle. If you arrive at a cloud after it’s useful lift cycle then it’s worse than gliding into a pure blue hole as there will be sink under it, plus the ground may be shaded, a double hit to your odds of staying in the air.

But if you near the top of your climb and see whispies start to pop within gliding distance and head out on glide toward them, then the odds are much better that you will find useful lift.

OK, you’re on glide toward a fine forming cloud, but where will you connect with the lift? Again, observing the cloud cycles will tell you. If the wind is stronger aloft than on the ground, the clouds will be forming at their upwind edges and decaying at their downwind edges. This tells you that the thermal will be sloped at some angle from upwind of the cloud to it. If you have a GPS or learn to read your groundspeed even while fairly high, you can figure out how strong the wind gradient is and therefore how much the thermal slopes. As a rule of wing, I visualize thermals in wind gradients of 10 MPH or less as sloping at up to 20 degrees, 20 or less MPH at 30 degrees and so on. Also realize that the gradient will often not be linear; there are many days where you will encounter some sort of strong gradient at a particular altitude; the thermals here will often become disorganized, but if you can fight through this barrier then you may continue on to base. Remember this altitude and anticipate doing battle to get through it instead of getting discouraged and heading off.

Some of the most frustrating XC days come when the winds are slower aloft than they are on the ground; I have found this situation surprisingly often and could never understand  how to find thermals until I realized that the clouds were forming on their “downwind” edges and dissipating on their upwind edges! The more moisture-laden areas of the cloud will be on their downwind edges; in this situation you will actually connect with the thermal downwind of the cloud.

The shape and texture of finished clouds also offer a wealth of information. Clouds taller than they are wide generally mean stronger thermals and may lead to over-development later in the day (don’t get me started on instability…). Puffy, closely-spaced clouds that cycle relatively quickly but never attain flat or “hard” bottoms generally don’t have very good lift under them; however, the light lift will be easy to find, just fly downwind and you’ll probably blunder into something. Because these clouds cycle so rapidly it’s almost impossible to time your arrival under one that’s developing. However, they often form up in general areas, and these areas will offer better chances of staying in the air. On humid days the sky will be absolutely filled with evenly spaced clouds; unfortunately, only a few of these clouds will be active while the vast majority are slowly and irritatingly decaying. On dryer days the few clouds that are in the sky will most likely be active, but make sure to get there while they are still in their active cycle. Finally, flat cloud bases indicate well-formed  thermals feeding continuously. Rounded, puffy bases usually indicate less well-formed feeder thermals and weaker lift.

On days with larger clouds, pay careful attention to what part of the base is highest; the best lift will almost always be feeding to the highest part of the cloud. As you climb to base, keep looking around, you may be able to get higher under a different portion of the cloud than you climbed to it under. This is especially common when flying the border between moist and relatively dry air masses; I have seen clouds that are stepped up to 4,000 feet on the Texas dry line.

In addition to understanding what kind of clouds to fly under, most people want to know what kind of clouds to avoid It’s often difficult to tell what your particular cloud is doing as you climb because the cloud tends to block your side-view of it; however, if you’re taking mental snapshots with each circle then you should have a good idea of what’s going on with the other clouds. It’s possible that you are thermalling up under the one giant cu-nimb in the sky, but it’s rare. If the sky is starting to over-develop all around you then it’s probably time to get out of the air regardless of what’s happening over your head. Even large clouds can cycle regularly; some days with cumulus clouds up to five or ten miles across are fine to fly on, but as soon as the clouds start growing much higher than they are wide I usually find myself either running for a much better portion of the sky or landing. After I land and my glider is secure I like to really watch what happens to the clouds I was worried about; did they cycle harmlessly, or are they continuing to blow up? If they did over-develop, how long did it take from the point I called my flight off to when the first gust front hit the ground? I have occasionally been frustrated that I landed early, but the few times I’ve pushed and stayed in the air too long were truly terrifying. The more I fly, the more conservative I become. If the clouds in the sky start “spiking” radically and look like fists on a day when the forecast is for thunderstorms then land immediately. Observing the sky intensely while flying isn’t just about finding the next good climb; it is the basis for safe flying.

This leads me to the broader part of this article: In general, clouds form in related patterns. These patterns may be due to any combination of literally thousands of factors (again, it’s worth understanding the meteorology, buy the book), but these areas of instability are where you want to be flying to connect with the lift. I’ve blundered off into large blue areas only to hit the dirt enough to believe this. It’s almost always worth flying the clouds around the edge of a blue hole rather than jamming straight through it, no matter how much more direct the blue line looks. Sail plane pilots have the luxury of making huge transitions across sky features up to a hundred miles apart, we generally don’t.

Most pilots dream of getting under cloud streets and flying straight until dark; while this does happen occasionally, I’ve found it more useful to treat streets as linked but individual clouds. If the street is set up with flat, hard bottoms and is maintaining good color (dense but not decaying and not over-developing as you fly along it, then stuff the bar and fly as fast as your understanding of speed to fly theory allows. But keep looking ahead and analyzing what is going on; sooner or later the clouds will end, and you need to be paying attention to what’s happening in front of you as well as to either side. I’ve often found it’s better to treat large gaps in streets as blue holes and jump sideways to another street if the gap in front of you is wider than the lateral jump by a significant margin.

Many “blue days” actually offer some very good sky-based clues. For starters, even if clouds don’t form at the top of thermals, “haze domes” often will. These are areas where the light refracts differently through the air due to more moisture, dust or just a different air mass. I’ve seen haze domes most frequently when flying relatively stable blue days in Mexico and the desert southwest; often the haze domes are marked simply by areas of the sky that are less blue. Haze domes are also often the precursors to proper clouds—in the morning you might just get haze domes at an inversion level, but they still mark lift and often are the first areas to pop through an inversion and become clouds. Blue days will often still form dust devils or swirling thermal cores; if you can see hay, fine dust or other debris in the air then that’s a sign of a thermal core as well.


Flying strategies:

            The classic model of thermal formation suggests one rising cylinder of air feeding one cloud.  In reality, I picture the thermals feeding into clouds as trees, with many small thermal “roots” feeding into larger ones until they reach the trunk and lead to the cloud. The higher you are above the ground, the farther apart the “trunks” are and the closer to the clouds you have to fly to truly intercept a large thermal. Anyone who has flown competitions will have seen gliders climbing relatively close together but in different cores before joining and continuing to base. Gliders that are low can take advantage of the smaller “root thermals,” not just the trunks. If you’re in the “low” zone, meaning below half way below cloud base, then you will most likely find relatively small cores. Sailplanes have a relatively hard time taking advantage of these lower-altitude thermals, but we can core up in very small circles, following the individual roots until they expand and join with other thermals. If you’re below half the distance between the ground and the cloud then you can pretty much forget intercepting a large core that connects to the cloud; however, most clouds are fed by multiple smaller cores that join together, so searching over good collectors and triggers upwind of clouds is a good strategy (remember to know the day’s gradients for which way the thermals will slope—the thermals may be “downwind” of the clouds on days with an inverted gradient).

I usually try to connect the collectors and triggers to the clouds they are feeding; this is also useful for predicting where the cloud is in its life cycle. For example, clouds that form over mountain ranges are generally flushed downwind. Once they are flushed past their thermal sources there may still be lift under the cloud as the thermal “bubble” continues to feed it, but you need to arrive relatively high to climb in this bubble no matter how great the cloud looks.

The higher the cloudbase, the longer your glide to the next climb will be (unless you have the good luck to be flying under a street of some kind). Reichmann predicts that the distance between clouds is approximately two and a half times their distance above the ground. If the base is 5,000 feet above the ground then the distance between thermal “trunks” is likely to be 12,500 feet (the distance between the “roots” will likely be somewhat less).  Even if your glider only goes at 5:1 then you should have a reasonably good chance of intercepting a thermal before intercepting the ground! Theoretically, it’s very rare to glide all the way from base to the ground without hitting lift. In reality,  I have done it often, particularly on blue days, but usually in retrospect I went gliding off into a large blue hole or down a sink street and should have turned 90 degrees after sinking more than half the distance between base and the ground to find lift. In the flats I think lift generally forms in lines and so does sink; even on  blue days, the next logical place to look for a thermal is above a good collector/trigger downwind of your last climb.

In the mountains the thermals  and clouds generally form above ranges which may or may not be oriented with your planned flight or wind direction. If you are crossing anything except very narrow mountain valleys on very high-base days then you need to base your decisions less on what the clouds are doing and more on the ground-based tactics covered in the previous article. If you are crossing small gaps while flying along a range then it’s often reasonable to use the clouds to plan your next climb, especially in the American West where base can exceed our FAA-imposed limit of 18,000 feet regularly. Most of our ranges in North America run roughly north-south, while the wind predominantly blows from west to east. One good trick for crossing the valleys between ranges is to climb to base, then drift over the gap with a cloud. This is slow, but XC flying is often more about staying in the air than speed. I’ve used this trick several times at King Mountain and other sites to beat gliders with far better glide ratios. The cloud will eventually start seriously decaying, so it’s better to leave it before this point or you will have to deal with sinking air.

Don’t get too aggravated if you can’t get to base, I generally only get there on days with well-organized climbs leading into flat-bottomed, dense clouds. On more humid days with poor lapse rates (oops, slipping into tech talk here), there may be plenty of clouds but no way in hell to get them. Do note how high you got in your climb before it disintegrated, and roughly how far below base you were. If your first climb of the day ended at 6,000 feet and base looked to be at about 8,000, then expect that the top of your next few climbs may be at a similar altitude unless the clouds start looking better or moving higher. Cloud base usually moves higher throughout the day, and climbs generally improve until late afternoon. If the clouds go to 10,000 feet and start looking really solid, then you might expect to climb higher and closer to the clouds.

The best way to truly understand the sky is to study it with near religious fervor. Read the books and understand the meteorology of any given day, then correlate what was predicted to what actually happened on your flight. If you can’t get into the air due to earthly responsibilities you can still learn a tremendous amount about flying. This will help you immeasurably when it comes time to make decisions while under your glider. My next article will deal with flying your glider in thermals and putting everything in these last two articles together. Happy flights!


Will Gadd lives for XC flying. He has held many site, state, national and one world distance record during the last eight years.