Sounding, Temp, T-Phi , Skew-T for Thermal Hunters

by Ulf Arndt, March 2003

What ???? is this about ???

Idea is to fly high and far. Agreed? No? Go away.... this is not for you.
Still here? Seems I found a reader who loves to core a thermal and the sound of a screaming vario.
Temp, Skew and T-Phi allow you to figure out at 8.00 in the morning if this will be the day to go and fly.
Or decide to fix the garden and take the kids to the Zoo instead of rushing to take-off and end up in turkey.

Good flying for me means strong thermals taking me all the way to a high cloudbase.

Cloudbase - Dew point and temperature and spread

 Let's look at what determines  a high convective cloudbase  for a good flying day.
Clouds get created by the humidity in the air condensating.
As the thermal, which is  warmed up air, goes up, it expands. Because their is less air pressure higher up. And  more space.
Air that expands cools down. In average air cools down 1 degree per 100 meters as it goes up.
Also called adiabatic lapse rate.
The dew point is the temperature where the vapor inside the air will condensate and will become visible as mist, fog or a cloud.
The dew point gets measured in those little white huts next to a weather station, or in some fancy thermometer called hygrometer.
What is of interest for us that we can get it off the Aerosport weather website under the temperature forecast page.
Let's say the forecast gives us for a certain area a maximum temperature of 30 degrees and a dew point of 10 degrees.
And this  30 degree air decides to release at a trigger point and become a thermal.
As it goes up the air cools down by 1 degree per 100m.  When it has cooled down to 10 degree, the humidity inside of this air start condensating and creating a cloud.  We reach 10 degree , when the air has cooled down from 30 degree by 20 degrees.  Which is 2000m above ground.
Cloudbase will be 2000m above ground. Sounds like a good day to fly.

If you see a forecast that reckons that we got a dew point of 20 degree and a maximum temperature of 25 degree, then we can expect a cloudbase of  around 500 meters. Not so great.

Ok, this was a bit simplified. In reality the air mixes as it goes up with some of the air around it. But for our flying it is good enough.

The more correct way is to take the spread, which is max temperature - dew point, and multiply it by 122 to get the cloudbase in meters. Or use the factor 400 to get it in feet.

  Air Pressure  and Altitude

One of the items measured by the balloon going up is the air pressure.
One can say that air pressure is the same as altitude.
 And Altitude in meters is sometimes also represented  as pressure on weather information.
As one goes up the pressure drops. 10 meters corresponds to 1 millibar in the environment that we fly in.
 Air pressure at sea level is 1013 mb, or 1013 hPa hectoPascals.
 850 mb corresponds to 1600 m sea level, or Highveld ground altitude.
But the pressure does not really drop linear as one goes up.
Every 5500 meters the pressure halves.

Sometimes at takeoff you might notice while waiting that the zeroed Altitude on your Vario changes.
Or the next days the zeroed Altitude on the Vario has changed.
This is due to air pressure changes.
If you Vario ALT2 shows 100m altitude gain over an hour, while being stationary, then watch out for any front arriving with a vengeance.
100m corresponds to a 10 mb pressure drop. Pressure drops are an indication of some bad weather and strong winds on their way.

Thermal Strength - Lapse rate

As you go up the surrounding air temperature changes.
Normally it gets colder. Ask those who flew in short and T-shirt and got 2000 meters above takeoff. all they can tell you about is how cold it was.
How much it gets colder as you go up is called the lapse rate.

Let's say we got some warmed up air on the ground at high noon. Ready to rise and become a thermal.
The air on the ground got warmed up to 30 C, while a few meters higher up the air is 20 C.
The bubble decides to release and rise up, since it is warmer and lighter.
Thumbsuck any air parcel what is rising up is cooling down by 1 degree Celcius per 100 meters.
Now it depends what the surrounding air temperature relative to the air parcel that is going up.
And for this we need something and someone to measure that for us.

 Interpreting T-Phi , Skew-T, Stueve Diagrams

Every day at 0 Zulu and 12 Zulu time ( Greenwich mean time ) the weather services all around the world send up weather balloons to collect data on the atmosphere. What is called a sounding. South Africa is 2 hours ahead of Zulu, GMT. Means those sounding were done at 2 o'clock local time.
Those balloons collect as they go up the air pressure, temperature and dew point. And are monitored by a radar reflector attached to them on the wind direction and speed.
Afterwards this data is made available on the Internet at the University of Wyoming website

And we can download the graphs. The graphs are also called a Temp or a  Balloon Sounding.
And the way the Temp  gets drawn is either as a SkewT or a Stueve Diagram.
T-Phi comes from plotting the Temperature T , and phi, what stands for entropy, or how much energy is in the air.

And here we finally got an example of what a Temp diagram can look like.

Horizontal lines is altitude , same as air pressure. We tend to fly up to around 3000 to 4000 meters.
Means only the part from the bottom 1000 mb to 700 mb pressure is of interest for us.
All the stuff above 700 we can ignore.

The blue numbers at the bottom with the blue lines going up vertical is temperature. ( This is a Stueve diagram)

Green curves are the dry adiabats. A warmed up air mass follows those green curves.
The blue curves are wet adiabats, what happens when you air condensates and makes a thunderstorm.
The purple curves are called the saturation mixing rate.

The left curve is the dew point. The right curve is the temperature of the air.


Another URL where you can get sounding data is from   at the SAWB Aerosport Web Site

For this you have to download the RAOB program to view the graph.

The bottom line shows in those blue numbers the temperature.
With the blue lines running sort of diagonal across. We got a skew Temperature diagram here.
The vertical bar, on the left side, in white grey numbers, shows the airpressure.
Airpressure at sea level is around 1013 mb, and it gets less as we go up.


If you are not interested in figuring out the T-Phi SkewT diagrams, use Soarcast  in combination with the Wyoming data to give you an idea of how good the thermals will be.

So you are one of the lucky ones who lives close to a weather station that sends up a balloon to collect sounding data.
Use the closest weather station sounding to your flying site. For Gauteng this is FAIR.
Download from the web at the  Soarcast program. It will tell you if it will be a good day and give you the expected thermal strength. In knots! Just divide the value by 2.  5kts is 2.5 m/s. Feed Soarcast with the text data from the Wyoming sounding. Use MS Internet Exporer IE5 and save the text data of the sounding as a whatever.txt file. Then use Soarcast to open your saved file.

  Stueve and Skew-T Diagram

 Those T-Phi diagrams show temperature of the surrounding air as one goes up as one curve.
and then they got a 2nd curve, which shows the dew point. Dew point means how much one has to cool that air down to have the water in it condensate.
How the info is represented on paper can differ.
One way is to have Temperature go up vertical , the obvious way of doing it, what is called Stueve diagram.
Another way, is to have the temperature isolines run diagonal, which is called Skew-T.
Germans like the Stueve Diagram while in the English world the Skew-T is used.
 More on the different formats  have a look at
Stueve Diagram Example

Here is an empty Stueve and empty SkewT diagram as it used by the Wyoming sounding website.

The bottom numbers are temperature.
On the vertical is height or pressure , same as altitude .
 Left curve shows the dew point .
 Right curve shows the temperature as the balloon went up.

Temperature is vertical blue  lines.
Dry adiabats are the green curves.

A good XC 100 km day  in the Stuve ( Stueve) diagram format


 Skew-T example
An empty SkewT diagram

Again the dew point is the left curve.
And the right curve is the temperature of the air as we go up..

Temperature figures in blue at the bottom. Slanting up towards the right

Dry Adiabat and Isotherm curves intersect under a 90 degree angle in a Skew T.

Example of a 100 km day from Rustenburg in a Skew-T diagram


SkewT are the preferred way for thermal hunters to see if it will be a good day.
We need a good lapse rate and what it does as we go up. And a Skew T shows variation in lapse rate more obvious.

More info on Skew-T see

Some more Examples of Good Temps

A good day will have a Zig Zag temperature curve.
And dry air higher up. dry air means dew point goes to a very low number.
Zig, means a low inversion, but not very strong.
No inversion on the ground becomes a bubble day. The moment the air warms up a little it shoots up.
We need something to hold down the air on the ground for a little while to build up some real hot big bubble of air.
That big bubble will finally break through the inversion and gives a big nice thermal.
Once through the inversion, we want the surrounding air to cool down a lot, so that our thermal always stays a lot warmer
( in real life that is only 1 to 2 degrees ) so that it will take us up higher and higher.
But once the thermal air cools down to dew point and we start getting the cloud then we need an upper inversion. The Zag part.
If there is no second inversion, and the thermal is now condensating and becoming a cloud and  is still warmer to the surrounding air,  then it will keep on going up.
In fact now we cool down a lot less then before. Thankx to the vapor releasing heat.
2 possibilities.  Either surrounding air is very dry , then there is a chance that the thermal will mix as it goes up and dry itself out and the cloud
will dissipate as it goes up.
Or the air is also humid higher up, then we get an Anvil and a thunderstorm.

If we have a Zag, a strong high inversion, if possible at 4000m , then we have a super day.

Above is an example where a pilot flew 70 km from Rustenburg.
Upper wind is west, what is good for Rustenburg, with the takeoff facing East or West.
One can assume that the upper wind will during the day also determine the lower wind direction.
There is an inversion, which will require about 20 Degree to break through.
Where does he get that 20 from?

   We look at the right curve.
   It start of at a horizontal line 850 mb, or it says there 1542 meters. That is the ground.
  Temperature on the ground is around 13 C , in between the 10 and 20 blue skew line.
  That black curve goes to the right as you go up a few meters. Then turns around and heads to the left.
   That is around 810 mb pressure level, 40 mb up, around 400 meters above ground.
  And the temperature is 18 C.
  We got an inversion around 400 meters above ground.
  Now we invent a green curve that runs through that inversion tip , sort of parallel to those green curves.
   This green dry adiabat will intersect the ground line at around a temperature of 21 C
 As the day progresses and the air gets warmed up thermals will start breaking through the inversion from 21 degree onwards.

In April one can expect around 25 degree max.
Find 25 C on the ground and put a green curve through it.The diagram got one roughly there.
 That green curve a warmed up air parcel will coold down as it goes up.
 Once it reaches the temperature of the surrounding air the thermal is finished.
 Which appens here at the 700 mb pressure level, or 3175 meters ASL.

 The dew point curve, the left one, is going more and more to the left.
  Means the dew point is getting lower as we go up.
Upper air is very dry, no risk of thunderstorms.
 Any condensation will be gobbled up by the dry surrounding upper air

And below is the sounding at 14.00 local time.

The temperature curve is leaning away from the green dry adiabat.
Indicating some very good lift.
Maximum temperature worked out better than expected, was in fact more like 28.

 Good Winter day in the Highveld / Rustenburg
In winter due to a strong inversion one can expect thermals to start late.
As long as the temperature curve slants to the left it is worth going out.

 At this day we had light wind conditions. Took off West in Rustenburg, and the thermals were anything from 1 to 4 m/s max.
And up to 2100m ATO. Working until 16.30.

At 12.00 Zulu, or 14.00 local time, the temp shows a straight dry adiabatic temperature drop as on goes up.
At the ground we got around 13-24 degree, cooling of very quick to 20. Any air bubble that gets triggered and is more than 20 will keep on rising until around 3500m.

Here is another example of a very cold day in the high feld after a cold front passed through.
With -2 degree in the morning  one does not expect a good day.

By 11.00 the cumulus clouds were popping.
Soarcast, with a max temperature of only 13 degree, gave 7 m/s thermals.
Even in winter, on one of the coldest days of the year, one can get good flying if the air is right.

The temperature curve is leaning to the left. You can see the thin curve above it, indicating the temperature of a rising airparcel. The temperature is cooling down faster as a warmed up air parcel would. at the 750 pressure level condensation will happen and clouds will start.

A good day in March , autumn in sunny South  Africa, where a pilot flew 100 km from Dunnottar by winch launch

SHOW and KINX do not indicate any overdevelopment.
LIFT and LIFT with 3-4 indicate good thermal activity.
Maximum Temperature for the day was around 30 to be expected.

A Southerly day at The Dam where pilots flew 170 km, 150 km and 100 km.

  A 100 km and 60 km flight from Dunnottar.

 A 7m/s max up day, cloudstreets , from Dunnottar in winter

Weatherforecast predicted trough conditions, air mass changed from the South in the early morning

Good Winter Day
Middle of winter, previous day was 21 June the shortest day of the year, with 2-3 m/s max.
This is 22 June, Soarcast predicted 5m/s and I had 6m/s up over Dunnottar, fighting to stay below ceiling.


Epic weekend from The Dam,  80km max flight

Very high cloudbase, strong thermals at Dunnottar

Multiple 100km- 155km flights from The Dam, but only launchable from 15.00 onwards. Sorcast gave 7m/s

Pilot report on 05 September 2004

Lousy Temps

Not all days are good. Some examples of days, where you can stay at home.
Weak lift , thunderstorm in Dunottar,

This day produced early thuderstorm in Dunottar.
Weak lift in rustenburg area.
But once the gustfront reached the Dam easy XC towards Garankuwa


Rock'n roll at The Dam
12.00 NE wind, wobbly, punchy, gusty. Had 2 strong solid 3-4 m/s thermals. Cloudbase maybe 1000 m ATO.
12km flight, clouds looked threatening and growing fast, but no overdevelopment.

Very humid air
A bad day has the temperature and dew point curves very close together. Stay at home.
A bad day that shows some lower inversion in summer, very humid lower air, no upper inversion, and the dew point running parallel with the temperature is an early thunderstorm day.


 Poor Winter day

In this case the temperature curve is leaning to the right

 No chance for any warmed up air to get very high

Dry adiabatic versus wet adiabatic
 Air that rises cools down by one degree Celsius per 100m. What is called dry adiabatic lapse rate.
But very humid air which is at dew point, and condenses, cools down only 0.5 degree Celsius  per 100m as it goes up.
Why? Let's start at the beginning on the ground. Your vapor is a water drop in the grass.
To convert the water drop to vapor one needs energy.
Like you have to put your pot with water on a stove and use electricity or gas to heat up the pot to bring it to boil.
A lot of energy is required to give those water molecules enough kinetic energy to break the sticky bond with the other water molecules and rather join the
molecules bouncing around in the air and elbow their way in between them.
Once the water bubbles and the vapor goes up, you can not see it.
But put your hand over the pot, and you get burn injuries by the invisible vapor deciding to condensate on your skin.
It is quite stressy for those water molecules getting bounced around and bumping all the time into Nitrogen or Oxygen  molecules around them.
Or into other water molecules in the air.
Much nicer place to rest and settle down on your skin. But for this they have to crash land on your skin, and slow down , and  impact hard on you.
What gives you a painful feeling and damages your skin molecules.
Back to the vapor in the warm air which is going up. Air expands as it goes up. Because we live on shere and as you go up there is more space.
Less collisions with others, means air cools down. And there is some dust particles in the air.
Now those particles are small and normally the pushing and shoving is too much to give a H2O molecule enough time to aim and settle down on one.
But higher up , with less chaos , one can have a chance to settle down on one of them.
That dust particle gets the speed, and happens sooner or later to bounce into some of the other gas, mostly N2, and give them some extra bounce back.
N2 molcules with some more bounce are warmer.
Resulting in humid condensating air  not cooling down aso fast as it rises compared to dry air.
Which then results in a cloud ( the visible condensed water vapor) overdeveloping.

For good flying, we need a strong inversion at condesation level to stop any overdevelopment.
Or lots of dry air above the codensation level to gobble up the humid air.

Other info on the Diagram
 On the side is the wind speed and direction displayed.
Assuming that the upper wind prevails and pushes down during the day it can give some indication what the wind might do during the day.
Add into it some Coriolis deviation , that the wind will swing more counter clockwise in the Southern Hemisphere.
 That CAPE, Convective Available Potential Energy, value gives some indication of the chance for a thunderstorm. The bigger the value, the more energy the air mass got higher up to create a storm. See

If the KINX  is 30 or more than one can expect a thunderstorm, overdevelopment later in the day.

  See  for LIFT,  THERMAL,  K and SHOWALTER Index

Also a good link is

 which gives us this table

 Lift Factor and K-Index for Soaring

Lift Index < 0 Lift Index > 0
K Large > 10 Wet and Very Unstable
Thunderstorms Possible
Stratified Clouds
K Small > 0 
or < 0
Scattered CU's
Perhaps clear - good lift
Good Soaring
dry - stable air
fair weather
blue - some lift

For Wyoming Skew T diagrams see for some short explainations.

Now, is that info really working?
Lets list some of the good days from above versus bad days
-9999 4-5 very good XC
9 5-6 good XC
9 9-10 good thermal day, no XC
-9999 3-4 very good XC
31 wet, early thunderstorm day
3 12 stable
-9999 4 60km+100 km from Dunnottar
-9999 8 170,150,100km from The Dam
-2.9 254 33 -3.2 early healthy clouds growing fast, and disappearing again, then regrowing, finally thunderstorms late, weak lift
1.85 0 29.9 2.11 early low clouds, looked like they overdevelop but nothing happened on overdevelopment

K Index around 30 gives a thunderstorm in the afternoon.

When I look at SA values and conditions I would say ...

SA K and L factors and good thermal days

Lift Index < 10 Lift Index > 10
K  > 10 Thunderstorms 
K   <10 good XC poor in summer,
good in winter?

For which areas can I use the sounding data in South Africa?

In the Gauteng area use the Irene data, which has got the location identifier FAIR.
The air that got measured at 0 Zulu time what corresponds to  2.00 local time, has probably traveled with the wind by the time we start flying.
Predominant wind direction in Pretoria is around North. And around 10-20 km/h wind speed.
By the time the thermals start working, let's say 12.00 lunch time, that air has traveled for 10 hours and is now 100 km to 200 km South of Pretoria.
To apply the data onto your flying site, one has to decide if the air mass which got measured in Irene is also the same in your area.

A MS Powerpoint presentation on this topic

For more info on flying Paragliders in South Africa visit

Prepared and maintained by Ulf Arndt, Copyright Ulf Arndt 2004