Ice warning - in VMC, below freezing level
 

In the thread about prop de-icing, it was commented that icing can occur anywhere, any time. On that subject, I thought my experiences from today would serve as a caution to others:

The weather was 6km vis in rain, few@2500', Bkn@4000' - distinctly VMC, although not the kind of weather that is fun to fly in.

I was planning an instrument training flight - the flight detail was to fly two instrument approaches. I expected to be cleared for the approach starting at 2000', but knew that if the airfield's overhead was not free I would need to climb to 3000'. There was a small chance of being put in the hold at 4000' if we weren't the only flight doing instrument training. So a relatively small chance of the flight entering IMC.

Bearing in mind the possibility of IMC, we began reviewing the icing conditions. There was a warm front sitting fairly stationary over the airfield. Metform 215 gave the freezing level before the front as 0', and after the front as 6000', so we figured that the actual freezing level was probably somewhere between the two.

Then we reviewed the actual weather. The temperature was +3, the dewpoint +2. I interpreted that to mean that the temperature at the cloudbase would be around about +2 degrees, still warm enough to not be a problem even if we entered IMC. But I knew I would have to be cautious about accepting a clearance to climb too high.

As we were taxying out, a Jet2 airliner was landing. I decided to take the extra precaution of asking the Jet2 pilot whether he encountered any icing on the approach, and he replied Negative. Good - I was now confident that we would be able to carry out the flight without risk of icing.

On departure, we were cleared for a standard missed approach, which consists of a climb to 2000' before making a turn back to the airfield's overhead - a procedure which would mean we maintained VMC the whole time. My student took off and put the foggles on, and then began flying the missed approach procedure.

As we levelled off at 2000' and began the turn, I noticed that the rain on part of the windscreen had stopped moving up the windscreen. I studied it a little closer, and realised that it had turned to ice....... in good(ish) VMC, well below where I expected the freezing level, and despite reports from another pilot who had just landed that he had not encountered any icing.

Needless to say, at this point I aborted the flight and we landed immediately. I am not one to mess around in ice, especially not rain-ice. I have seen ice once or twice before when I've been instructed to climb higher than the freezing level (and of course I have always descended immediately, leaving the holding area if that was necessary to avoid traffic in the descent) - but this is the first time I have ever experienced totally unexpected ice. It took me completely by surprise, and I thought others might be interested in learning from it.

Incidentally, the ice did not begin melting until we were down to 1000'. Fortunately, we were VMC at the time this occured. Had we been IMC, the lowest 10nm MSA for our airfield is 1600' which would still be in icing conditions. The first descent platform on the approach for the runway in use is also 1600', and that is to be maintained for 7nm outbound, through a base turn, and only descending below that after reaching 7nm inbound - so it would have been several minutes in rain-ice before we were able to descend out of it safely. That is quite scary!

As for the cause - it seems there was a sub-zero layer, with base of somewhere around 1000', which obviously caused the aircraft skin temperature to decrease to below freezing. No clouds formed in this layer, though, despite it being several degrees below the dewpoint.

Isn't the aim of this game to fill the bag of experience before you empty the bag of luck?????

Hope this post is useful to others, and maybe even generates some interesting debate...

FFF
---------------

Excellent read, thanks for sharing your experience.:ok:

It doesn't tend to work like that with fronts. The warm air behind is warm and the cold air in front is cold. If you are in the cold sector the freezing level will be as per the cold sector. What you encountered was rain from the warm sector falling into the cold sector and becoming super cooled (or that's what it sounds like anyway). I was flying in the vicinity of the front today and that type of icing was a distinct possibility but I did not encounter any.

Personally I never fly (without de-ice) in freezing conditions where ice is likely unless the temp at MSA will allow removal of said ice, or I know I can become VMC in order to descend further to a positive temp (ie base higher than MSA).

Sounds like you did your homework, got caught out a bit, and have learned from it. The airmet text often gives info on sub zero layers by the way. I think the balooning forecast does also.

Good post FFF, reinforces what a crusty old sort in his 90's told me when I was doing my IR, he had been there in the begining as an air mail pilot in the 1920-30's and a production test pilot during WWII.

What he said about Icing.

You will find it where it is.
You may go flying in a clear sky not expecting it and find it and you may be in the clouds expecting it and not find it.

"You may go flying in a clear sky not expecting it and find it "

Not sure how that would work.

While I don't mean to labour the usual 20:20 hindsight stuff, and your post was well worthwhile, I would have thought that the "ISOL 2000 M FZRA, SEV ICE IN FZ RA. SUB ZERO LAYER SFC to 3000 FT" annotation on the F215 you looked at might have tempered your surprise. ;)

BTW the Jet2 pilot has the benefit of higher speed, and therefore higher skin temperature. As a rule of thumb, never trust an icing pirep from anything faster than you.

Freezing rain under a warm front, a classic ATPL exam scenario. Sounds fairly attention getting!
Not familiar with that correlation, can you expand on it further?

At the standard lapse rate of 2 degree/1000' I would suggest it was at roughly 1500';)

I would have been surprised if you didn't pick up ice:ok:

BTW in jets we put the engine anti ice on at 10 degree C or less in visible moisture.

All good comments - thanks!

I will freely admit that icing is something which I don't know very much about. The reason for this, I think, is that I mainly fly aircraft with no de-icing, so I tend to be very conservative and stay on the ground if I feel there's a chance of icing. (Ironically, in all the time I've spent in IMC in de-iced aircraft - which isn't all that much - I've never once encountered any ice.) And since the best way of learning about something is to experience it, that means I've still got a lot to learn about ice.

A couple of posts on this thread have raised points which I'm interested in learning more about. Dr Eckener said:Indeed, that fits in with my observations of what actually happened. But how does it fit in with a temperature of +3 degrees? This air temperature suggested to me a freezing level of between that quoted for before the front and that quoted for after the front, which seemed to fit in with the front being overhead..... which seemed quite logical at the time, but was obviously wrong.

And HWD picked up on a quote of mine, where I said:Whatever the dewpoint is, if you take a parcel of air from the surface, and reduce the temperature of that air to the dewpoint, it will become fully saturated and start forming clouds. So what I have done (and it seems to have worked for me in the past) is reversed this, and said that, at the base of the clouds, the temperature and the dewpoint must be equal - or, to put it another way, the temperature at the base of the clouds is equal to the dewpoint as given on the ATIS. As I say, this has worked in the past (but maybe it's not actually correct?) It may well even have been correct yesterday - I don't know because we were never as high as the base of the clouds. But there was certainly a sub-zero layer below that - and what I don't quite understand is why no clouds formed in this sub-zero layer despite it being below the dewpoint.

There is always so much more to learn in this game......

(And just to lighten the tone a bit, while we're on the subject of how much there is to learn about the general subject of met, one of my colleagues, who is also a licensed met observer, has his own method of forecasting the weather. He reckons that whatever the weather is doing today, it will do the same tomorrow. He admits that this doesn't work all the time - but he reckons it's at least as reliable as the forecasts from the Met Office!)

FFF
---------------

Whilst it is the case that, at the base of the clouds, the temperature will be the same as the dewpoint, the point you're missing is that dewpoint, like temperature, normally falls with height (in the case of dp, this is the 'hydrolapse rate'). So, normally, the temperature at the cloud base will be lower than the surface dewpoint. Of course, this is weather, so in practice anything's actually possible.:D

Good grief...this is truly trawling the long lost depths of my brain:}

FFF the Saturated Adiabatic Lapse Rate and the Dry Adiabatic Lapse rates are different....thus the wet bulb temp and dry bulb temps on the ground are lapsed at two different rates...when you get to the altitude where those two temps are the same you get the theoretical cloud base. This is why the bigger the split the higher the cloudbase IF any cloud forms at all...which is not guaranteed.

In the case of your flight with only a 1 degree split you could be gauranteed IF there was any cloud the base would be quite low...as it was on the day...2500'

Now from memory dry adiabatic lapse rate is 2 degrees/1000' and saturated adiabatic lapse rate is 1.5 degree/1000'. Remember these are not exact but rounded numbers so dumb pilots can make calculations.

On the day in question;

0'............+3/+2
1000'.......+1/+.5
2000'.......-1/-1

So you can see the cloudbase was always going to be around 2000'

The ice problem is a seperate issue...you can see that the DALR hits 0 at 1500'...so in visible moisture you will experience icing...and by jove you did at 2000'...doesn't matter that you were in VMC...the rain started freezing on your airframe and unlike the jet you were not travelling at speeds sufficiently high for air friction to warm up the skin and alleviate that process.

I can't fecking believe I just pulled that !!!!e out of my head after 25 yrs:( :ok:

What happened to you was as close to a guaranteed outcome as aviation supplies....but a great bit of experience to tuck away:)

Not a bad effort after 25 years, Chimbu, but from my hazy recollection of this esoteric subject, wrong nonetheless.

The Dry Adiabatic Lapse Rate (DALR) is actually 3 degC per 1,000 feet, and is the rate at which a parcel of unsaturated air will change temperature (solely due to change of pressure) when moved vertically through the atmosphere.

It is the Environmental Lapse Rate (ELR) that, in the International Standard Atmosphere, is 2 degC (well, actually 1.98 degC) per 1,000 feet. This is the variation in temperature with height within a non-moving column of air ... and in real life as opposed to the ISA may be greater or less than the DALR. Air forced to rise where the air mass ELR is greater than the DALR will continue to rise (unstable air mass).

The Saturated Adiabatic Lapse Rate (SALR) is the rate at which a parcel of air that has become saturated (i.e. visible moisture) will change temperature if it is forced to move vertically. Its value is half that of the DALR since latent heat is released as condensation occurs. The SALR has no bearing, as far as I'm aware, on the height of the cloud base.

But as with its temperature, the dew point of the air also normally drops with increased height, since its capacity to hold water vapour at a given temperature increases as its pressure is reduced. In a parcel of unsaturated air forced to rise, the dew point decreases by 0.5 degC per 1,000 feet.

So, on the day in question, cloud formed by adiabatic cooling would have a predicted base (from a tephigram) given by the convergence of temperature and dew point, where temperature starts at +3 degC and decreases by 3 degC per 1,000 feet, and dew point starts at +2 degC and decreases by 0.5 degC. This would suggest a low cloud base of circa 400 feet with an OAT of circa 1.8 degC. Markedly different from what was actually observed.

Well, that's as I remember the theory. Hopefully, a meteorologist will be along shortly to set us straight.:hmm:

From what I recall, that is in the PPL Confuser, completely ignored in the Thom book and discredited by the ATPL course which teaches that the dewpoint lapse rate is quite unreliable and difficult to estimate
:confused:

Sounds right to me...should've known not to try such esoteric knowledge dredging after 1/2 a bottle of red:}

I think you will find it is the environment dew point lapse rate (i.e. the figure that corresonds to the ELR) that varies massively and is thus unreliable. The adiabatic dew point lapse rate (i.e. that relating to a rising parcel of air) is a physical constant ... given by the Met Office Handbook of Meteorology as 1.7 degC per km for unsaturated air.

But it's all a mystery to me. Where's that meteorologist, there's never one on hand when they're most needed.:)

I went hunting on google and found a formula for cloudbase;

Dry bulb - wet bulb x 125m

This would give a cloudbase on the day in question of 412' but the cloudbase was 2500'.

While it has been a VERY long time since I studied ATPL Met I have a strong recollection of using the method I used a few posts back. That it seems to give a closer to accurate answer, in this case, is interesting...but the formula espoused on the website I looked at exists too.:confused:

Just goes to show what a bag of worms predicting weather is:}

Meteorology is one of those problem sciences where there is no difficulty at all for a skilled expert to explain what happened or why something was the way it was. Predicting what is going to happen is orders of magnitude harder though!

It all starts going wrong for me when they talk about "parcels of air". Air doesn't come in parcels!

All good theoretical stuff but usually meaningless in the real world.

Just look at today's skew-t chart

http://weather.uwyo.edu/upperair/sounding.html

Select Europe, GIF-Skew-T and your favourite location.

Most days, there is little resemblance to "Trevor Thom weather forecast" conditions.

Precisely what is it about today's skew-t charts for the UK that you think is at odds with the theory?

Not sure what you mean by that IO540. In most cases, the weather behaves remarkably in alignment with the theory. Being "caught out by the weather", is more often than not a thin veil for poor flight planning or poor theoretical knowledge.

2D

The actual lapse rates tend to be wide of the "expected" mark.

We're talking ELR here, for a particular air mass. Pray tell me, where does the theory 'expect' a particular value?

Rather than get bogged down in arguing the minutia of a particular aspect of meteorological theory (which, in any event, in the absence of an expert meteorologist is unlikely to be resolved satisfactorily), let me try and expand the issue.

Like many diligent pilots, I’ve expended vast amounts of effort over the years attempting to assimilate meteorology theory, not only for private and professional pilot qualifications, but also from the pragmatic position that herein lie many issues that can bust my a@se!

Like others, I regularly use my so-called knowledge of the theory to take a view on the particular conditions I’m going to encounter. That my predictions aren’t always borne out in practice is hugely more often than not, I believe, down to my lack of real understanding of the complexities of the subject than it is due to the fallibilities of the theory.

My experience, gained over three+ decades, is that aviation met. forecasting has got progressively better (thanks, presumably, to the enormous computing power increasingly employed), and is now pretty damn good. Where ‘actuals’ vary from forecast, it seems likely to me that this is not usually a question of erroneous theory about what happens given a particular set of conditions, but is more to do with the vagaries of the precise timing of occurrence of those conditions.

Where it is well less than perfect, however, is in the detail dissemination of the weather information. Get to talk to a real, live forecaster (increasingly rare these days), and you may well get a more useful and accurate picture of the weather as it is likely to affect you. Economics (and, perhaps, legal considerations) have determined that, in respect of en-route forecasts in particular, we have to live with more generalised forecasts that don’t give us what we really need, despite the theory being adequate for the task.

Other GA IFR pilots of non-pressurised aircraft that populate the FL100 to FL150 levels will be well familiar with this problem. Where will the cloud tops be? … dunno, there’s much less guidance available for your route than a real forecaster could give you (and, actually, less still now the new Met Form 215 doesn’t even bother with altitudes of 10,000 ft and above). Will I get moderate or severe icing in the cloud if the temperature is below 0 degC? … absolutely always, according to the Met Form 215, despite the fact that, more times than not in stratiform cloud, you won’t. Embedded Cb's? ... TEMPO, oh, that's helpful. And so on.

Now, if only we had real-time weather available in the cockpit – what a massive advance in safety and operational capability that would provide. Sadly, for us in Europe, the fabulous advance that Nexrad has offered in the USA will remain but a dream. Oh well, never mind, we can always look forward to the ultra-expensive benefits of Galileo instead … if anybody can recall exactly what they are.:*

I assume the Jet2 was a 737. Doesn't this aircraft have heated windscreens. Could possibly explain why they hadn't experinced the freezing rain, along with a higher TAT.

Hmmm...how long has a tephigram been called a skew-t?

Probably since America was invented :O

The following hits the nail right on the head:

Other GA IFR pilots of non-pressurised aircraft that populate the FL100 to FL150 levels will be well familiar with this problem. Where will the cloud tops be? … dunno, there’s much less guidance available for your route than a real forecaster could give you (and, actually, less still now the new Met Form 215 doesn’t even bother with altitudes of 10,000 ft and above). Will I get moderate or severe icing in the cloud if the temperature is below 0 degC? … absolutely always, according to the Met Form 215, despite the fact that, more times than not in stratiform cloud, you won’t. Embedded Cb's? ... TEMPO, oh, that's helpful. And so on.

Actually F215 forecasts icing in all cloud, not just below 0C :O Very helpful (not).

Fortunately, there is GFS. One gets TAFs/METARs and MSLP and SIGWX charts from the usual aviation weather services, and then looks at GFS for the more detailed stuff. But it's changing, with more websites doing graphical presentations from GFS.

IO540

You appear to have shifted your ground.

If you are now arguing that we don't have the same richness of weather data in Europe that is available in the US, then few would disagree with you.

Your previous assertion, which appeared to suggest that real-world weather disobeys basic met principles more often than not, has rather less going for it.

2D

"The Dry Adiabatic Lapse Rate (DALR) is actually 3 degC per 1,000 feet, and is the rate at which a parcel of unsaturated air will change temperature (solely due to change of pressure) when moved vertically through the atmosphere.

It is the Environmental Lapse Rate (ELR) that, in the International Standard Atmosphere, is 2 degC (well, actually 1.98 degC) per 1,000 feet. This is the variation in temperature with height within a non-moving column of air ... and in real life as opposed to the ISA may be greater or less than the DALR. Air forced to rise where the air mass ELR is greater than the DALR will continue to rise (unstable air mass).

The Saturated Adiabatic Lapse Rate (SALR) is the rate at which a parcel of air that has become saturated (i.e. visible moisture) will change temperature if it is forced to move vertically. Its value is half that of the DALR since latent heat is released as condensation occurs. The SALR has no bearing, as far as I'm aware, on the height of the cloud base."
---------------------------------

I don't think that's quite correct.....

Surely,

DALR = 3 degrees / 1000 ft
SALR = 2 degrees / 1000 ft

Both of the above relate to the ISA.

ELR = somewhere between the two, what you actually get on the day.

2D in my Trevor Thom statement I was referring to the lapse rates being well off ISA figures, most of the time. I think that's correct.

Of course, the real weather always obeys met principles precisely, doesn't it :O Question is, which ones?

Back to the met books for you, Shy Torque. That's almost complete tosh.

DALR ... correct.

SALR ... almost always incorrect, but unlike DALR it is a variable and depends on the amount of vapour condensed. At low levels in temperate climates it is around 1.5 degC per 1000 feet, which is the figure that tends to be used in the UK. It can be around 1 degC in the warm saturated air of tropical regions and can be (here's where your value may on the odd occasion be correct) 2 degC or more per 1000 feet where the temperature is below freezing.

DALR and SALR relate to the ISA ... no they don't!

ELR is somewhere between DALR and SALR ... not necessarily, indeed if this were the case unsaturated air could never be unstable - and that would transform our weather patterns!

OOPssorry! Yes, Islander, of course you're quite correct. The ELR can be above, below or in between the DALR or the SALR lines.

Sixty lashes with a rolled up Tephigram for me!

ELR>DALR>SALR = Absolute Instability
DALR>ELR>SALR = Conditional Stability
DALR>SALR>ELR = Absolute Stability

DALR = rate of change of T within a vertical current of unsaturated air; this never changes and is reversible on ascent/descent.
SALR = rate of change within a vertical current of saturated air. This does not change from day to day but it does change with height. It is only reversible on ascent /descent if no water (precipitation) has been lost.
ELR = What you see is what you get.

That better?

BTW, reading back to the original question; formation of ice on an aircraft can occur from clear air, if the temperature is below the dew point and below 0 C, i.e. by sublimation.

The question, chaps, is whether all this knowledge helps you in practice, with real flight planning.

Can you point me to a website where I can look up all these values, for different levels along a specified route?

No matter how much one understands met theory, one cannot, IMV, statistically significantly beat the forecasters who have all the data at their fingertips (including data which they don't publish, or do publish but only to fee paying customers) and who do this all day for a living.

I also don't think one can (statistically significantly) even beat totally dumb button pushing on a GFS website like e.g. http://www.arl.noaa.gov/ready/cmet.html.

If somebody with a JAA ATPL tells me they can out-do the pros, I suggest they start up their own weather forecasting company. They would make a lot of money. Or they could offer their services to the UK MO and help all pilots :O

I know this flies in the face of all the "professional" pilot training, which makes you sit through reams of this stuff, so if I have missed some essential point I would be (seriously) grateful if somebody could explain exactly what I am missing. In detail, so I can learn from it.

Interestingly, the FAA doesn't make pilots learn most of this stuff for the IR, but they don't find it a problem. Is the weather really so much more complicated (but predictable) in Europe?

Here's the sounding for Albermarle which is the nearest available. I'd imagine that the sounding for Blackpool, furtehr west, would have looked even more pronounced.

Note the layer between 850 hPa (5000 ft) and 920 hPa (abut 2700 ft) that gives the precip a chance to become liquid before plunging into the freezing air below.

I wonder if anybody has done a comparison between the forecast skew-t data and the (just past) actual skew-t data. It would be a very interesting exercise, for skew-t diagrams contain so much useful info.

One would assume the GFS weather model is calibrated using the actual ascents.

This thread now contains a vast wealth of information - some of which I probably knew once but have forgotten, some of which is new information for me. Without a doubt, I am now significantly more experienced and more educated in this area (but still have loads more to learn too!)

One thing which is very good to see (at least, from a personal point of view it's good; from a professional point of view it's quite disappointing) is that I'm certainly not the only person on this forum whose understanding of this particular aspect of met isn't as thorough as it could be!

FFF
----------------

Just to put something to bed...Total beginner hit the nail on the head with regard to the 737 question..

I drive them, and they aren't particularly prone to icing. (Yesterday was another thing, the ice build up on the wing was quite another thing). Mostly due to the fact that apart from the last moments we drive them about 250 kts to 320 knots, so the the temp rise due to the ram effect will keep the leading edges warm enough to keep most of the ice away (in our temperate climate). We do have heated windscreens, but the heat doesn't do the whole windscreen so you get little patches of ice build up on the edges that are noticeable. Also there is a windscreen wiper nut that we use to observe ice build up, so they would have been aware. I think its just a difference in speed.....

The mach heating should work in any climate.

You just need enough TAS to achieve a temperature rise whose magnitude exceeds the width of the temperature band (about 15C) over which liquid water can hang in stratus cloud. That is about 350kt TAS.

If you can get 15C of temp rise, all over (not actually possible all over due to local pressure drops) then you would "never" ice up.

Just caught up with this post, and also Fuji Abound’s on another thread where he appears to have interpreted this post by IO540 to say: “there are some pilots who think they know more than the professional forecasters who have spent their lives forecasting the weather and have more data available than we.”

I am bemused, IO540 (and Fuji Abound). You seem to be suggesting that learning some meteorology theory is a waste of time since we will never be as competent as a professional forecaster. Sorry, but who was suggesting we would be? It was, after all, you (IO540) that pointed us towards skew-t diagrams … which are raw data diagrams REQUIRING a very good knowledge of meteorological theory to determine the weather that will result (for example in terms of cloud types, bases and tops).

IO540, skew-t diagrams display ALL the data we have been debating on this thread… from ELR, DALR and SALR to hydrolapse rate. Since it was you that provided the link, when you then ask: “can you point me to a website where I can look up all these values”, perhaps I can be forgiven if I question your understanding of skew-t diagram content! My sincere apologies if I’m doing you a disservice in making that observation.

For my part, I think I have made it clear that my knowledge of this complex subject is extremely limited, and that I would always bow to the views of a professional meteorologist. But this leaves me with two problems:

1) in planning, whilst terminal forecasts largely give me what I need, en route forecasts are really lacking (a point, IO540, on which you had previously agreed ); and

2) in flight, especially on a long sector, the situation can change markedly from that forecast, and I tend not to have a met. specialist sat alongside me in the cockpit (neither do I have, regrettably, the NEXRAD resource that is available to our U.S. companions).

In consequence, us pilots are FORCED to make some judgements about the weather … and, let me make it absolutely clear, this has got nothing whatsoever to do with any egotistical belief that we “know better than the professional forecasters.”

I’ve no idea what the FAA requires in terms of met. knowledge, since I went the CAA route. What I can say, though, is that when I was at PPSC in the mid 1990s, the time I wasted studying irrelevant material (including ‘nav. plotting’ and ‘Decca’, for gawd’s sake!) did not extend to meteorology … ‘twas two papers in them days: ‘Theory’ and ‘Practical’, and every single bit of the course material I still view as good, useable stuff, even if I do struggle with the concepts.

So, IO540, what would your proposed ATPL/CPL/IR met. course contain? Would it be sufficient for a student to take a stab at estimating the cloud tops from a skew-t diagram? And would that student be able to recognise the met. conditions on the day of FFF’s flight (as reported at the start of this thread) as classic ‘rain icing’ conditions … one of the truly big ‘avoid’ areas for GA flight?

As an 'umble PPL with an FAA IR, I know very little about weather other than how to read the charts, TAFs and METARs on offer. I fly an Arrow with no deicing equipment of any kind, so if it's close to freezing and cloudy, I stay down here wishing I was up there.

But I did, some years ago, spend a fascinating evening with a senior guru from the British Antarctic Survey, who explained a load of stuff about weather. Most of it has faded into the background, but I do remember 3ºC per 1000 feet as the DALR, and the helpful phrase "drier is higher". I think there was something about the saturated adiabatic lapse rate being constant (might've been 1.5ºC per 1000 feet, I dunno) but the "wet, humidity not defined" one being unpredictable. That may be my brain seizing up.

I also remember on my FAA IR oral being given a METAR with the dewpoint higher than the actual temperature, and being asked what it meant. I don't know what the expected/correct answer was, but I said someone had just taken a leak onto the wet bulb.

"Perhaps pilots place too much reliance on the forecaster and met observer. The rules require that you consult the available information. However the interpretation and use of that information is up to the Captain. The forecaster will never say don't fly today. That decision is up to the aircraft commander.....so don't try to pass the buck if the decision made is incorrect."

Islander2

It was that quote that caught my attention.