Forecasting Thermals and Reading Soundings

Glider pilots use various types of lift to make progress from one part of the country to another. Thermal lift is generated by the rising parcels of air, created by the sun's heat on the ground. The trick is to know if a particular day is going to produce good thermals. This is my summary of how to predict thermals strength from the available weather forecasts (with strong acknowledgement to Weatherjack and others).

How does a thermal work?

When the sun heats the ground, this will also cause the parcel of air immediately above the ground to be warmed up. Since this parcel of air is now warmer than the surrounding air, it is less dense, and therefore more bouyant than its surroundings. It will therefore rise, and as it rises, the pressure it experiences will be reduce. The rising parcel of air is what the gliding community call a thermal and is what provides the energy to allow gliders circling in the thermal to climb. As the parcel of air rises, the reduction in pressure in the atmosphere with height will cause the parcel to expand and cool. If the parcel continues to find itself surrounded by cooler air than itself, it will continue to rise. If it does not, then it will stop. Thus, for good gliding thermals, we look for

  1. Heating of the ground and/or the layer of air just above it
  2. An air mass in which the temperature of the air with height reduces more faster than the rate at which a rising parcel of air cools.

Suprisingly, the temperature profile with height in particular air mass can be very different according to the past history of that air mass.



The rising pockets of warm air that we hang glider pilots call thermals are known as a parcels of air to meteorologists. These parcels of air being warmer and therefore lighter than the surrounding air rise as long as they are warmer than the air around them. While rising, a thermal expands and cools at about 5.5 *F for every thousand feet that it goes up. The rate of temperature change as we change altitude is called the lapse rate and this  5.5 *F or 3 *C rate of cooling is called the dry adiabatic lapse rate. Again, as long as the parcel of air or thermal is warmer than the air around it it will rise but sooner or later it will cool to the temperature of the surrounding air and will stop rising. If the air above us is not cool enough compared to the surface temperature then there will not be any thermals.

Air contains a  percentage of moisture in it and if a thermal cools to a certain temperature based on that amount of moisture it will condense and form a cumulus cloud. This temperature is called the dew point. We see this happen on and near the surface in the form of dew on the grass and fog. As condensation takes place heat is released and the thermal's upward velocity will increase because  it cools slower at the moist adiabatic lapse rate instead of  5.5 *F per thousand feet.

The Upper Level Sounding
If we are to predict what the day's thermals are to be like then we need to know what the temperature of the air above us is. All over the world balloons are sent up from specific weather stations at the same time to measure atmospheric conditions. This is done at 0 Zulu and 12 Zulu for every station. Radio transmitters aboard the balloons then send information about temperature, humidity, and wind velocity back from various altitudes. Because instruments that measure altitude operate on barometric pressure and the rate of pressure change is not always the same, these altitudes are referred to as pressure levels. Inches of mercury are the units we are used to seeing for barometric pressure but upper level soundings use millibars. The information that is sent back from the balloon is only for certain heights or levels and is determined by two criteria. These are called Mandatory Levels and Significant Levels  Mandatory Levels are those used for every sounding and are listed below with approximate altitudes.
Mandatory Levels
Pressure Level
Approximate Height
sea level 0 ft
1000 mb 300 ft
850 mb 5000 ft
700 mb 10000 ft
500 mb 18000 ft
300 mb 30000 ft
200 mb 40000 ft
100 mb 53000 ft

Significant levels are points where there are significant changes in temperature or relative humidity. If the straight-line segment between two plotted points is more than 1 *C in the troposphere then it is considered a significant level and the data is plotted in addition to the mandatory level data. Taking and plotting upper level soundings has been done for many years but now computers are used to analyze the data to create atmospheric profiles which are the heart of weather forecasting.  A variety of methods can be used for graphing these soundings but the most  common in the United States is the skew-T and that is what we will concentrate on in this article.

The Skew-T

As previously stated the skew-T is the sounding data represented graphically and is shown in the example on the right. The black solid lines running diagonally up from left to right are called isotherms and are the temperature in degrees C. The slightly curved orange lines running from right to left diagonally are dry adiabats that represent the cooling that will take place in a dry parcel of warm air (a thermal) while rising. The numbers on the left side of the graph are the pressure levels in millibars and represent height. The green dashed lines sloping slightly to the right are saturation  lines and as will be seen are used to determine the condensation level or cloud base. The orange dashed lines running nearly vertical to about 600 mb. and then curving to the left are moist adiabats and are used to determine the height of the cloud tops.
The thicker vertical red line plotted on the graph is the actual temperature sounding and the thicker green one is the dew point. It should be seen that the temperature and dew point plots start above the bottom of the graph. This is because the graph starts at sea level and our sounding came from a station  around 1200 feet above sea level. This means that we start at about 970 mb. and that all elevations are above sea level. On the far right of the graph are wind arrows pointing at the true wind direction with barbs showing the wind speed in knots. The illustration below the skew-T to the right shows examples of wind symbols and what they mean. The barbs can be used in combination to signify the wind speed at different levels.

Further Definition
dry adiabatic lapse rate When a parcel of dry air is lifted,  it expands and therefore  cools at approximately 5.5 degrees F (3 degrees C) for every one thousand feet that it ascends. In other words it is said to cool adiabatically. The dry adiabat represents this cooling. 
moist adiabatic lapse rate When a parcel of rising air cools to the same temperature as the dew point, condensation occurs forming a cloud and heat is released giving the thermal new life. The thermal will then cool slower at the moist adiabatic lapse rate and will ascend faster. The amount of water vapor present in the parcel determines the rate of cooling that is usually between 2*F and 5*F per 1000 ft. Slower cooling increases the upward rate of the thermal and is what causes towering cumulus and cloud suck.
standard lapse rate The standard or average lapse rate determined by many years of recording weather is 3.5 *F for every 1000 feet of altitude.
temperature inversion If the temperature increases with altitude we are said to have a temperature inversion.
saturation lines  While the temperature of our parcel cools at 5.5 *F, the dew point decreases at about 1 *F per 1000 ft. This means that the dew point and the temperature approach each other at 4.5 *F. Hence, if we subtract the surface dew point from the surface temperature and divide the resultant by 4.5 the answer times 1000 will be the altitude agl. of cloud base. 

Forecasting Thermals With A Skew-T
To make things easier we will use only a blown up section of the skew-T illustrated above to predict the height the thermals will rise to and the height of cumulus cloud base if any.

Step #1 Using the predicted high temperature (23 *C) for the day, plot a line from the surface parallel with a dry adiabat until it intersects the actual temperature sounding. This is illustrated on the graph below with a blue line. The point where the blue line intersects the sounding (the red line) will be the height that a dry thermal will rise to during the time of peak heating. In this case approximately 725 mb. As will be seen in the next step of our example this altitude will be well above cloud base.  This area from the surface to the maximum height of a dry thermal is called the adiabatic layer.

Step #2 The next step of plotting the saturation line from the surface dew point is a bit trickier because the dew point changes as the air warms up and it is difficult to predict what its value will be at the time of peak heating. An easy way to find the predicted dew point is from forecasts such as the NGM-MOS or MM5 models found on the internet at. Using this value (11 *C) we plot a brown line from the surface, this time parallel to the saturation lines until we intersect the blue line that we plotted from the surface temperature. This (810 mb) then is the predicted cloud base at the best time of the day when peak heating occurs. This can be done for several temperatures-dew point spreads to forecast the height of the thermals and cloud base at different times of the day.

Step #3 We can use the following chart to convert the level in millibars that we got from the skew-T to height in feet. The graph is an approximation but it is accurate enough for our purposes. Using the height of the adiabatic layer that we found in our example we find that 725 mb is equivalent to 9,000 ft. msl. Doing the same thing for the value of 810 mb for cloud base we find that it is roughly 6,000 ft. msl. Subtract the msl altitude of the launch at the site we are going to fly from 6,000 and we have the maximum altitude gain we can expect. Remember though that this is a forecast and just as we complain about the weathermen, we will have our own blown forecasts.

Temperature Conversion
Upper level and aviation temperatures are given in *C while many other forecasts are in *F. This means that we will have to convert back and forth to make use of them. Conversion programs are available on the internet and some weather sites even provide one. The following formulas can be used for conversions.

Celsius to Fahrenheit:  F = (9/5 C)+32
Fahrenheit to Celsius:  C = 5/9 (F-32)

Thermal Index
The thermal index is a value that allows us to compare the temperature of a thermal as it rises with the temperature of the air around it. To find the thermal index for a given altitude we find the temperature on the sounding plot (red line) for a specific altitude and then subtract the temperature on the blue line for the same altitude. The result is the thermal index. For an example if we find the temperature at 900 mb on the sounding (11 *C) and then subtract the temperature where the blue line crosses the 900 mb level (17 *C).    11-17= -6 This tells us that the thermal index for 900 mb or 3200 ft. msl is a negative six. A negative number is favorable while a positive number indicates stability. A few more examples are given in the chart below. Sailplane pilots claim that soaring can be done to an altitude where the thermal index is a -3 *F (1.66 *C).

Height Thermal Index *C Thermal Index 
3200 msl 6 -10.8 
5000 msl 4 -7.2 
6000 msl 2.5 -4.5 
8200 msl 1.66 -3 
  From the chart above we can see that soaring is possible to 8200 ft. but that is 2200 ft above cloud base. On a day like this  climb rates should be reasonable and it should be possible to get to cloud base.

Climb Rates
Estimating climb rates is a bit subjective but a general rule of thumb is that the higher the thermals go, the better the climb rates. In other words the height of the adiabatic layer plays an important part in the strength of the thermals. Another indicator for climb rate is the thermal index. Higher negative numbers indicate better climb rates while lower negative numbers tell us to expect poor climb rates.

What Data To Use
Because of the times that the soundings are taken and the delay before we can acquire them it is not always feasible to wait for the 12 Zulu one.  Zulu time  is 5 hours ahead of Eastern Standard Time or 4 hours ahead of Eastern Daylight Time and it is not unusual to wait for one to two hours after the balloon is sent up to get the data.. Fortunately there are other options available to us. One option is to use the the 0 Zulu sounding from the night before but it is sometimes hard to determine which one to use and how far up wind it should be. The other option is to use the forecasted soundings available at several places on the internet.

Links to upper level forecasts
 Noaa FSL
 Storm Machine Forecast Soundings
 Soundings From Maps






A BASIC INTRODUCTION Basically, a tephigram shows the temperature of a vertical profile of the atmosphere. Because a rising thermal acts as an enclosed parcel of air, it cools at a different temperature to the surrounding air. Depending on how the temperature of the surrounding air falls with height, the thermal will either be able to continue to rise, or not. If it can, at some point it will condense and form cloud, if it continues to rise, a shower and rising further a cumulonimbus. By looking at the tephigram we can determine how likely this is to happen. If the temperature profile falls quickly with height (leans to the left), the chances are that a rising thermal will continue to rise, if the temperature falls less quickly or even rises with height the thermal will not rise as much and cloud may not form.

SOME MORE DETAIL... To interpret a tephigram, first you must understand the motion of the air. Air has various properties, the ones we are interested in are moisture content (measured by the dew point) and temperature. When the ground is heated by solar insolation parcels of air are created, or thermals. As these rise they contain the properties of the air at the surface, i.e. a particular temperature and a quantity of water vapour, relative to the overall volume. As the parcel rises, the air pressure reduces (because there is less atmosphere above) and so the parcel expands. This process uses energy from the air within the parcel, causing the temperature to drop. As the temperature drops the dew point remains constant, because the total volume of water in the parcel is (basically) unchanged. If the parcel continues to rise, sooner or later the temperature drops to the same temperature as the dew point... relative humidity has reached 100% and condensation occurs. This point is termed the Convective Condensation Level (CCL). The ascent of the parcel to this point is approximately at a rate of 9.8ºC per kilometre and is termed the dry adiabatic lapse rate (DALR), because the main temperature change is caused by the expansion of the parcel. Above this point, if the parcel continues to rise, condensation continues to occur. Condensation actually releases energy in the form of heat, thus offsetting the fall in temperature that we had been getting due to expansion. This means that the temperature fall with height in saturated air is lower than in unsaturated air. As the amount of water that the air can hold varies with temperature, at high temperatures when it can hold more water, more condensation can occur, and therefore release more heat above the CCL. This means that at high temperatures, the temperature fall with height in saturated air or the Saturated Adiabatic Lapse Rate (SALR) can be as little as 4ºC per kilometre. This increases to nearer 9ºC per kilometre at -40ºC though, because by then the air cannot physically hold as much water. Now, how this relates to the tephigram, the afformentioned parcel of air will only rise if it is warmer than it's surroundings. We can determine whether or not this will be the case, by releasing a balloon carrying instruments to measure the atmosphere with, a radiosonde. This process is commonly called a sounding and gives us the actual temperature with height or environmental lapse rate (ELR). A parcel of air must have a higher temperature than the corresponding point on the ELR to continue rising in normal convective conditions. If a parcel can continue rising beyond the CCL then we will get cloud... if it can rise 3km above the CCL we are at risk of showers... if it can rise to the tropopause (the top of the lower layer of the atmosphere, marked by a sharp and continuous inversion above about 300mb), we will get cumulonimbus and heavy showers and thunderstorms. The way we tell whether it will get cloudy, or if we will get thunderstorms is by using the tephigram (sounding or skew-T diagram). We know the surface temperature, pressure, dew point and ELR. By using the dew point (or absolute humidity) and following the mixing ratio axis up to the ELR from the surface pressure, we can find the CCL. By then tracing back down the DALR axis, we can see the temperature that must be reached before a parcel will be warm enough to rise to the CCL. This temperature is the potential temperature or trigger temperature which must be reached before any convective cloud will form. By then following the SALR axis (the curved one), you can see how the temperature of the parcel will continue to cool with height. If at any point it crosses the ELR, the parcel's temperature is the same as its surroundings, and therefore stops rising and the cloud stops growing. If the ELR drops very quickly with height and a parcel would remain at a higher temperature than its surroundings as it rises, the air is termed unstable. Conversley if the ELR falls slowly with height, a parcel would generally cool more quickly than its surroundings and hence the air is termed stable. If the ELR rises with height then there is an inversion, which generally acts as a very stable cap to cloud growth. the greater the separation between the ELR and DALR/SALR the more unstable the air is (if the ELR is on the left), or the more stable it is (if the ELR is on the right). The process of cloud formation is also confused by lifting mechanisms other than convection, such as forced lifting along an otherwise stable frontal boundary, or up the side of a hill. As the mechanisms are slightly different, cooling of the air occurs at a different rate, giving a different point for cloud formation, the lifting condensation level (LCL). The LCL and CCL are generally at similar heights. Forced lifting is primarily responsible for the initial formation of stratiform clouds and convection for cumuliform clouds.
AND THE OTHER INDICES... Wind Speed: This is plotted by the pink line and read against the appropriate scale on the x axis. This illustrates the change in wind speed with height, or shear. This can be useful to determine the potential strength of any convection (if other conditions are right). A small amount of wind shear doesn't aid convection, and a large amount actually degrades it by literally ripping a thermal/cloud apart. A medium amount of convection can strengthen a thermal/updraught by effectivley creating suction in the upper reaches of the cloud. This also extending the life of the cloud as the updraught can be maintained, even once it has left its warm source at the surface. Wind Direction: This is plotted to the right of the main graph, as many lines running into a vertical line. The top of this line is North, the angle at which the other lines attach therefore determines the wind direction at that altitude. You'll often notice a change in wind direction with height... this is a normal effect of the drag at the surface altering the wind direction. If this effect is coherent enough it can also stabilise convection. It helps this by aiding separation of the updraughts and downdraughts within a convective cell, preventing them from the usual collision and reduction of convection. CAPE (Convective Available Potential Energy): This is a value in joules per kilogram of air, indicating how much energy is potentially available for release within a thermal between the surface and the 500hpa level. This value isn't always realised for a variety of reason i.e. a capping iversion preventing covection from reaching that height, but it is a useful tool to demonstrate the potential within the atmosphere for convection. Lifted Index (LI): This is a parcels temperature at the 500hpa level, subtracted from the environmental temperature. A negative value would therefore indicate the parcel being warmer than it's surroundings, hence having a tendency to rise. A positive value would indicate a parcels temperature is lower than its environment, hence having a tendency to sink. The more positive or negative the value, the stronger the tendency to move in the appropriate direction i.e. an LI of -8ºC would indicate the potential for a very powerful updraught. As with CAPE, this is purely a value at the 500hpa level... a thermal has to get there first. If there is a significant capping inversion in the way, a significantly negative LI might not be realised. Bulk (Richardson) Shear: This is an index demonstrating wind shear... values below 40m/s suggest that convective cells will be dominated by outflow, and will therefore be relativley short lived. Between 40 and 100m/s however, low level mesocyclogenisis could occur if all other conditions are right, producing longer lived and (much) more severe storms. Bulk Richardson Number (BRN): This is a function of CAPE and wind shear. A BRN of <10 indicates strong vertical wind shear and weak CAPE. This often means that the shear is too strong for the CAPE to overcome and any TS are broken down by the high winds at altitude. Between 10 and 45, CAPE and wind shear are very compatible to one another. The high degree of bouyancy allows strong updraughts, the high wind shear helps to seperate these from the downdraughts and can result in rotation of the cell as a whole... severe TS and supercells may result. Above 45 demonstrates very high CAPE relative to wind shear... in these conditions single or multicellular thunderstorms are likely, but rotation and supercells become less likely.
The tephigram therefore enables us to see whether or not, and what kind of, cloud development is possible. The various indices are only useful when used in conjuction with a tephigram.


These diagrams are theoretical, simplified Skew-T's   

Millibar (hpa) pressure levels are on the left in blue
Surface pressure is normally around 1000 mbs
Each millibar corresponds to about 30 feet
(in lowest part of the atmosphere)
850 mbs is about 5,000 feet an important level 
700 mbs is about 10,000 feet
500 mbs is about 18,000 feet
300 mbs is about 30,000 feet cirrus level
Wind arrows (and numbers) are down the right hand side
short dash is 5 knots-----long dash is 10-----triangle 50 knots
The direction is shown by the arrows
difference between true and magnetic can be ignored
---In this example:
At 500 mbs (about 18,000 feet) triangle and short dash means about 55 knots (numbers give 57) from about 320°
At 30,000 feet (300 mbs) the wind is approx 340°/75
Not all charts use quite the same axes nor the same colour schemes and units  

The red arrows point to numbered red diagonal lines.
These are temperatures.  eg, 0, 10, 20 etc
The red line that wiggles its way up the page is the environmental lapse rate line (ELR) 
ie a plot of temperature with height.
It can be seen that the environmental line meets the bottom horizontal 1000 mbs line roughly midway between the two red diagonal lines marked 10 and 20
This is the surface temperature of around 15°C
(before any heating raises it later)
Now follow the ELR up to the 850 mb level.
Here the temperature is about 3°C

At the 700 mbs level, it also happens to be 3°C in this theoretical example 

At 300 mbs (30,000 feet) the temperature is -38°C


You will recall that the red line is the ELR 

The green line represents the dewpoint. 
Thus at 10,000 feet, the dewpoint is -22°C.
At the baseline, it is 8°C, ie surface dewpoint 8°C

The dewpoint is the temperature at which the air can hold no more water vapour, ie it is saturated.
Surface dewpoint is of major importance

On the real sounding on the next page, the dewpoint line is not of course green, but is a the solid line on the left.

The two arrows point to brownish/red lines .
These indicate the mass of water the air can hold at various temperatures.
For example the arrowed example on the right can hold 12 gms of water/kg of air
In the colder air on left, it is much less at eg 3 gm/kg air

At the surface dewpoint (green line intercepting base), the mass of water the air can hold at that temp is indicated.
As that air rises in a thermal, it expands, cools (3°/1000) until it becomes saturated and cloud forms.
But - and this is important - the total water the air is holding has not changed from that it held at the surface

Look at this new graph below.  It shows yet another type of presentation that is available on the internet (Uni Wyoming)
In fact, this is a Skew-T and is very like the demonstration charts seen earlier but with minor variations.
The sounding is the midnight one on what was arguably "the day of the year 2003" when many big flights were achieved
Points to note: 
Very strong inversion at 850 mbs thus convection limited to about 5,000 feet
---An inversion is when the temperature increases with height - more usually it decreases
At 850 mbs the air is very dry (the left dewpoint line is well separated from the right environmental line)
Winds are light at all levels, but near coasts, sea breezes need to be considered  
Sea Breeze rule of thumb
Deep penetration inland only likely if total depth of convection, including cumulus, is between 3,000 and 10,000 feet
Less or greater depth of convection will probably mean any sea breezes confined to very near the coast

So with light winds and appropriate depth of convection on 15th August, sea breezes were forecast to move well inland


Not shown on this Skew-T, but using information from other sources, cooler air between 850 and 800 mbs was expected to move in from the west during the day raising the depth of convection.
That was taken into account when making the forecast for 15th August

Straight lines (arrowed in black) run diagonally up at 45° from the bottom right to the top left.
These are the Dry Adiabatic Lapse Rate (DALR) lines
Adiabatic means no external heat added nor taken away
from the air mass
A rising parcel of air cools (because it expands) at the DALR until such time as it becomes saturated.
DALR can be taken as being about 3°C per 1,000 feet.
So a thermal which leaves the surface with a temperature of 20°C will have cooled to 14°C by the time it has reached 2,000 feet
When the thermal rises far enough and cools sufficiently for condensation to occur, cloud forms.  Condensation takes place if the air continues to rise, and latent heat is given out by the condensation process.  Thus the temperature in the cloudy thermal falls off rather more slowly than it does in a dry thermal.
At low levels, this can be taken to be roughly 1½°C per 1,000 feet
These Saturated Adiabatic Lapse Rates lines (SALR) are curved and indicated by the green arrows
Air behaves either as being dry or saturated.  It is not a gradual process of change between one state and the other  
Now we should know what all the various lines mean.  So how do we use them?  Let's recap
The red ELR line starts from about the 1000 mb level.
Now using those red temperature lines that go diagonally upwards to the right, it can be seen that the ELR reaches the surface at about 15°C
Just to the right of this point, there is one of the black DALR  lines (they go diagonally up to the left) which intercepts the 1000 mb surface at about 17°C
So if the surface temperature reaches 17°C, then a bubble of air (being warmer than the environment) will rise as a thermal and follow that DALR line (temperature decreasing by 3°C every 1,000 feet as it does so)

Eventually, that thermal reaches the red ELR line at about 850 mbs (5,000 feet) and if it were somehow able to rise above that point, would in fact be cooler than the environment.  This cannot happen, (ignoring orographic effects) so where this particular DALR line from a surface temperature of 17°C reaches the environmental line, the air stops rising, and this marks the top of the thermal.

It can be seen by further study that about 15°C would have been needed before any convection can start (the trigger temperature).

We now consider the absurd case of the surface temperature rising to 30°C  A line drawn (or visualized) from 30°C parallel to a DALR will reach the environment line at about 730 mbs  (9,000 feet) - rather a good day!

In this illustration, formation of cumulus has deliberately been ignored as a simplification. 
That will be considered later

This was a midnight sounding.  It looks dreadful, doesn't it? 
Lets blow up the bottom bit.

With some difficulty, it can be seen that the dry and the dewpoint lines are very close or even coincide near the surface.
The surface dewpoint and dry temperature are both about 14°C.

The dewpoint probably won't change much during the day. 
But the dry temperature will rise to an expected 20°C 
(obtained from various forecast sources)

Now imagine that the bottom of the dry (environmental line) reaches the surface at 20° later in the day


Look again at the top chart on previous page
From the expected surface temperature of 20° follow the dry adiabatic line upward diagonally to the left until it meets the environmental line.  This occurs at about 880 mbs.

However, now we have to consider the moisture content of the air.
From the surface dewpoint of about 14°, follow (or parallel) a purple line upwards to the right.
This will intersect the dry adiabatic from 20° surface at about 920 mbs.  This is the expected cloudbase

A cross check using the 400 ft / degree rule would come up with the same result - ie 6° difference means cloudbase of 2,400 feet.  So we have confirmed expected cloud base.

Remember that the 400 ft / degree rule is only appropriate if cumulus develops and cannot be used to determine thermal depth on blue days

In a later lesson, we will see how it can be determined 
(with, it must be said, a certain degree of uncertainty) whether or not it will be a blue day

The relevant part of the sounding is shown
The maximum temperature is assumed to be 20°C

In the previous lesson, we saw how to work out the cloud base.

But the convection does not stop at cloud base.  It continues at the SALR until reaching the ELR at around 400 mbs (~23,000 feet).
Many glider pilots are reluctant to climb in cloud, but the major significance of the height to which the cloud will grow means that widespread showers can be anticipated later

The later midday sounding shows that we were correct with the earlier interpretations.
(It should be clear that once cumulus has formed, it will rapidly grow to great heights)

You will need to work this out for yourselves.

The experts will note the super-adiabatic very near the surface.


Blue or not blue?  This was a fairly predictable blue day.  Let's see why.
The surface dewpoint is around 3°C
Follow the moisture content (purple line) up to the right
It would not intercept the ELR until about 700 mbs
Unless totally impossibly high surface temperatures of around 42° are reached, a DALR (from a realistic temperature) meets the ELR well to the right of the moisture content line; no cumulus can form
If the dewpoint were a little higher, or the air at 900 to 800 mbs a little cooler, then cumulus would develop.
In forecasting, there has to be inspired (and experienced) interpretation in situations like this
Many days are touch and go whether or not blue.  Forecast soundings can be used to see the likelihood of a blue day, but all these factors make predicting with certainty very difficult in some situation.
Many days are easy.  There will definitely be cumulus or it will definitely be blue.  But so many in Britain are borderline 
The dewpoint has to be assessed accurately. 
It might fall during the day 
(as air from aloft mixes with that at the surface) 
Alternatively, the dewpoint might rise a couple of degrees as overnight fog or deposited dew clears.
The 900 to 800 mb temperature might change.
If an anticyclone is pushing in from the west, then the air aloft might warm and thus what initially might be a day with small cumulus becomes blue later 
(temps at 850 mbs shown on the F214 give a clue)


And now for the real experts, a little trick that enables subtle regional differences to be identified. 
Knowledge of blue or not blue might help with task planning  The Met Office has far more information than we have.
Sophisticated computer models would enable distinction between say, areas where there might be cumulus on an otherwise blue day (an aid to task planning).  We don't have that luxury and have to work with what is available

In a nutshell, we want to determine if the moisture content line drawn from the surface dewpoint intercepts the ELR or not?  (Cumulus if it does, blue if it doesn't)

Look at a soundings graph again to refresh some ideas.
Follow a moisture content line.  It effectively decreases by 0.5° each 1,000 ft 
[this is very loose jargon and the term "delta" or D is coined for calculation purposes]
An example should make things a little clearer 
From the F214 at 5,000 feet (or at 850 mbs - the same) the temperature = say 4° in the west, 7° in the east
The expected surface dewpoint Tdew = 8° in all areas
Tdew reduced by 5 x 0.5° = 2.5 at 5,000 feet  So "delta" at 5,000 = 8 - 2.5 = 5.5
F214 gives at 5,000 air temp of 4° (see above)
delta is more than T5000 so cumulus expected
F214 gives at 5,000 air temp of 7° (see above)
delta is less than T5000 so it is likely to be blue
So in the west, the moisture content line intercepts the ELR.  In the east, it does not intercept.
All other things being equal, there is a better chance of cumulus in the west.  The east is likely to be blue
In practice, simply subtract 3 from surface dewpoint and call this D
If D > T5000 then cumulus.  Use Bradbury for cloudbase If D < T5000 then blue.  Bradbury DOES NOT apply