TAF Question??

[desmotronic]

BZZZZT times up fogged in! Back to the sheraton for you captain!

[Icarus53]

Desmo,

Either you've got the required vis or you don't. Go or don't go. Both answers then become either 0 or 100%. Simple. If you're trying to pick the difference 10 hours prior, guess what - you're going to struggle.

conservative alternate purposes

Conservative alternates? Really? As opposed to reckless alternates perhaps? Either the valid forecast requires an alternate or it doesn't. If it says 2424 Prob 10 Fog 0500, then you carry the fuel, whether or not your advanced divination of TAF info suggests it will be OK.

[RENURPP]

I suspect this guy is either doing or has done an aviation degree = all theory no common sense.

[Jabawocky]

BZZZZZZZZ

Something tells me your time is up buddy

Kaptin KEG will have a field day!!

Picked the wrong guy dood!

[Mamakim]

Thanks Capt Claret and others

I just wanted clarification on the subject because as Capt Claret says that there will be a new TAF issued by 0630 but it does not clearly state that in the GEN reference that i gave and I wanted confirmation that a new TAF will be reissued by 0630 and then a more accurate update will be given....

Cheers

[desmotronic]

Jaba,
Quit you jibba jabba fool.

[OZBUSDRIVER]

These times are valid for in the case of the first value, ninety minutes after time point HH.

????

Just to clarify something. I always thought those values were valid from HH TO HH+3. There is no overlap similar to the validity for the whole TAF for planning purposes? To clarify if validity is 091200 and temps at 12 13 14 15 then at time 1459 the temp is still 12. At 1500 then temp is forcast to be 13.

Have I got that right?

[ITCZ]

I always thought those values were valid from HH TO HH+3.

You were not alone. Lots of aussie pilots thought so.

This issue came up when a smart cookie (not being sarcastic) ex DHC8 pilot asked some pertinent questions during his 717 line training. ITCZ put question to BoM via website. The email exchange might explain things fully.

Hi, my question relates to the validity of the four three-hourly QNH
figures on Terminal Aerodrome Forecast.

Example.Lets say I have a TAF for YPGV that is valid from 0000 to 1200.
It gives the QNH forecast as 1013 1011 1009 1011, being the QNH
forecast for YPGV as at UTC 0000, 0300, 0600, and 0900.

Here is my question.. I am planning to arrive at Gove at UTC 0430 and I
will need to use the TAF QNH to set my altimeter for an expected
instrument approach.

I interpret AIP MET for the TAF as saying "forecast QNH from 0300 to
0559 UTC is Q1011." and that is what I would set on my altimeter for my
arrival, assuming aerodrome AWIB data unavailable.

Some of my colleagues interpret the TAF as saying "forecast QNH is 1011
at 0300 and 1009 at 0600. Therefore, interpolate and set altimeter to
1010 for an arrival halfway between those two times."

That seems reasonable and practical, but is it what the BoM forecaster
meant?

Reply from BoM

Thanks for your query.
Could you please let me know the AIP reference which you interpret as
saying that the TAF QNH forecast is valid for three hours from the point
forecast time.

Regards

Peter Flint
Aviation Weather Services Program Office
Australian Bureau of Meteorology
Melbourne

Sure, Peter, here is what I referenced:

AIP GEN 12.16.3 (Met services)

"Up to four (4) forecast values of QNH are given, valid at three (3) hourly intervals commencing at the beginning of the validity period of the forecast. The letter Q prefixes the QNH groups."


Here is this morning's TAF for Gove, NT:

GOVE (YPGV)
TAF YPGV 252247Z 260012
04008KT 9999 SCT020
TEMPO 0012 04015G35KT 1000 THUNDERSTORMS WITH RAIN BKN008 SCT020CB
RMK
T 27 29 29 27 Q 1005 1003 1001 1003

The first thing I note is that Temps and QNHs appear as RMK. However this is the only BoM product that gives T and Q forecasts so I treat them as forecast values.

My interpretation:
TAF validity is 0000Z to 1200Z on Feb 26th
Forecast QNH is
Q1005 from 0000z to 0259z
Q1003 from 0300z to 0559z
Q1001 from 0600z to 0859z
Q1003 from 0900z to 1159z


Some pilots think that they can interpolate.

I know that it is most likely that the actual aerodrome QNH at Gove at 0430z and 0730z will be 1002. If I use that for an arrival at 0430z today, that will have me 30' closer to the ground than 1003. If I did the same at 0730 tonight, it would have me 30' higher than the 1001 would allow. So it can cut both ways.

Cheers, ITCZ

ITCZ

Thank you for the information you have supplied.

The TAF QNH forecasts are point forecasts for the indicated times, e.g. 00, 03, 06, 09. However, a point forecast is valid for 90 minutes each side of the time point; and pilots should not interpolate. Unfortunately, the rule doesn't work for a half way point, e.g. 0430, in which case you have to choose one of the two point forecasts.

We will reword GEN 12.16.3 to make this clear.

Regards

Peter

[Creampuff]

The relevant AIP extract (from GEN 3.5 para 12.17.3, with the critical bits bolded and underlined) is:

Up to four forecast values of QNH are given, for the times HH, HH+3 hours, HH+6 hours and HH+9 hours, where HH is the time of commencement of the TAF validity period. These forecasts are point forecasts for these times but are valid for: in the case of the first value, ninety minutes after the time point HH; and, for subsequent values, ninety minutes each side of the time point. The QNH forecasts are prefixed by the letter "Q".

So, in the 'old' days you got (e.g.) a TAF that started (e.g) at 0600 local, with (e.g.) Qs of 1008, 1010, 1011, 1010. The forecast QNH was 1008 from 0600 to 0900, 1010 from 0900 to 1200, 1011 from 1200 to 1500, and 1010 from 1500 to 1800.

Now the interpretation would be: 1008 from 0600 to 0730, 1010 from 0730 to 1030, 1011 from 1030 to 1330, and 1010 from 1330 to 1630.

Hopeless prior education for the change.

[desmotronic]

interpolation

Periods of validity notwithstanding the forecasts are for a point in time. For QNH i would have no problem interpolating as you cant go too far wrong with a linear interpolation.

Temperature change can be much more non linear eg passage of a cold front. Conversely if you are considering temperature change due to celestial radiation then clearly there will be a temperature gradient linear or not, and you may well be able to interpolate.

[Creampuff]

Desmo

You seem to be mixing up 2 different circumstances.

The first is where:

your ETA or ETD is say in 6 hours

In that circumstance, your suggestion that we:

Look at the METAR or listen to the AWIS!

is nonsense, for precisely the reasons you identified. What the METAR or AWIS says now does not help in predicting what the weather will be in 6 hours. That's what TAFs and amended TAFs are for.

The second is where:

the time is 1145UTC.

The 1022 Taf says amongst other things
FM 11 prob 40 fog 500m temps 10 8 6 5

The 1130 metar and current AWIS say viz 8000m temp 9 dew point 9

You want to depart a runway with runway edge lighting spaced 75m:

1. What are your chances at 1300?
2. What are your chances at 1200?

If it's 1145 and you want to determine your 'chances' of departing at 1200, I'd suggest you look out the window. If the AWIS is accurate, I'd say you have a 100% chance of departing (or, more accurately, weather won't be an impediment to you departing) at 1200.

The METAR and AWIS, in the context of the TAF, indicate that the longer you wait, the increase in the likelihood in your being prevented, by fog, from departing. However, I don't see how that helps you, at 1145, determine the 'chances' of departure with any more or less accuracy than '40%'.

But I'm always happy to learn. It's 11.45. In your view, what are the 'chances' of departing at 1300 in your scenario?

[desmotronic]

My aircraft buckley's. Keg's maybe 25% at best.

[Creampuff]

Hmmmm

So through what formula did you determine that "buckley's" and "maybe 25% at best" are any more or less accurate probabilities than the 40% you dismissed?

[OZBUSDRIVER]

Well look at that, amended 7JUN07. Pays to read any changes, doesn't it!

Now this opens a can of worms. Validity periods are going to be out too. A statement of met conditions expected for a specified period in the airspace within 5nm of the centre of the aerodrome runway complex.

So the statement of met conditions for temp and QNH is not valid for 90 minutes prior to the end of a TAF...so to fulfill the requirments of validity of an arrival that a TAF must be valid 60 minutes AFTER arrival then-

Does this mean that arrival now must be 150 minutes prior to the end of validity of the TAF? Certainly doesn't say that in AIP but it stands to reason. The TAF is no longer a valid statement of expected met conditions 90minutes prior to the end of the period. Only a lawyer could come up with this...Not you, creampuff

[OZBUSDRIVER]

Just checked my current version of Command Flight Planner 4.81. It has a new addition of decoding TAFs to realspeak This todays decode for Albury-

ALBURY (YMAY)
TAF YMAY 092305Z 100012
33015G26KT 8000 LIGHT RAIN SCT015 BKN040
FM02 28018G34KT 9999 LIGHT SHOWERS OF RAIN SCT020 BKN045
FM08 30013G27KT 9999 LIGHT SHOWERS OF RAIN SCT015 BKN040
TEMPO 0004 4000 RAIN, SHOWERS OF RAIN BKN010
INTER 0412 5000 SHOWERS OF RAIN BKN015
RMK FM00 MOD TURB BLW 5000FT
T 09 09 10 07 Q 1013 1011 1011 1013

METAR YMAY 100230Z AUTO 32011KT 5000NDV // BKN013 OVC017 08/07 Q1010
RMK RF00.0/003.2


All times are UTC

TAF (Terminal Area Forecast) YMAY (ALBURY) Issued 09-Jul 23:05
Valid from 10-Jul 00:00 to 10-Jul 12:00
Wind 100/01KT. Visability 100012 Metres 100012
33015G26KT 8000 LIGHT RAIN
Cloud: Scattered (3 to 4 OCTAS) at 1500FT
Cloud: Broken (5 to 7 OCTAS) at 4000FT
FROM 09-Jul 02:00, 28018G34KT 9999 LIGHT SHOWERS OF RAIN
Cloud: Scattered (3 to 4 OCTAS) at 2000FT
Cloud: Broken (5 to 7 OCTAS) at 4500FT
FROM 09-Jul 08:00, 30013G27KT 9999 LIGHT SHOWERS OF RAIN
Cloud: Scattered (3 to 4 OCTAS) at 1500FT
Cloud: Broken (5 to 7 OCTAS) at 4000FT
Temporarily between 09-Jul 00:00 and 09-Jul 04:00:
Visability 4000 Metres RAIN, SHOWERS OF RAIN
Cloud: Broken (5 to 7 OCTAS) at 1000FT
Intermitently between 09-Jul 04:00 and 09-Jul 12:00:
Visability 5000 Metres SHOWERS OF RAIN
Cloud: Broken (5 to 7 OCTAS) at 1500FT
FROM 09-Jul 00:00, Moderate Turbulence Below 5000FT

From_____To_______Temp QNH
10-Jul 00:00 10-Jul 03:00 09 1013
10-Jul 03:00 10-Jul 06:00 09 1011
10-Jul 06:00 10-Jul 09:00 10 1011
10-Jul 09:00 10-Jul 12:00 07 1013

METAR (Routine Aerorome Obvservations) YMAY (ALBURY) Issued 10-Jul 02:30
NOTE: Contains only automated observations (excercise caution when interpreting)
Wind 320/11KT. Visability 5000 Metres - No directional variation can be reported
Cloud: Broken (5 to 7 OCTAS) at 1300FT
Cloud: Overcast (8 OCTAS) at 1700FT
Temperature: 08, Dew point: 07, QNH: 1010
Rainfall 00.0mm in the 10 minute period prior to observation time, and 003.2mm since 0900 local time.

normally, do not use this part of the product but interesting to see the decode is to the old method.

[grrowler]

Let me get this straight...
At 1145, not content with the PROB40 on the TAF, or the AWIS info, and after much careful analysis, calculation and debate, all the time carefully avoiding looking out the window, we determine that as the TAF shows the temperature trending down through the dewpoint, based on the information we have, there is actually an increasing probability of fog? Brilliant!

So now what? Captain, will you hold my hand...

[Creampuff]

OZ - the 'old' method lingers in more places than one!!

As with almost every other source of confusion or controversy in the rules, this one's got nothing to do with the lawyers. My understanding is that a person in BOM simply pointed out that what everyone had thought was the correct interpretation of TAFs for the previous few or more decades was wrong. That person better hope they're right!

The really interesting thing is that I don't see how a forecast that it will be, for example, 1013 90 minutes either side of 1900 this evening, is likely to be any more accurate than a forecast that it will be 1013 for 3 hours after 1900 this evening. The weather does not know what time it is.

QNH and temperature changes don't necessarily occur at 3 hourly intervals, and aren't necessarily synchronised to either the commencement time of a TAF or 90 minutes afterwards.

That is, neither method is any more likely than the other to be accurate!

But, in true Aussie-aviation style, some people are using one method and some are using the other. Another masterpiece in change management.

[Keg]

My aircraft buckley's. Keg's maybe 25% at best.

Given that it's a PROB40 then I reckon you've got a 60% chance of departing!

However, I'd be looking outside first and foremost and pilots are authorised observers for the purpose of determining vis for take off. Further, if FG was evident then that triggers a SPECI at most fields with observers and you haven't specified what type of field we're at. If the field doesn't have an observer then we default back to the fact that the professional and trained MET observers reckon that it's only a 40% chance of FG forming at that airfield.

Further, I've seen many occasions where a temp trend appears like that and as the temperature drops, so does the WBD and FG never eventuates or viz doesn't drop below the 8000m. Additionally, I've also seen occasions with nil split in the WBD but due to the lack of wind the FG was never forecast, more important it never materialised and all we ended up with was a heavy dew.

Finally, if the airfield concerned has a take off minima lower than 500m- most places I go to do- then with 500m RVR FG then my chances of departing are probably closer to 90%. The last 10% I'm saving for if I have a return airport. If so then my chances are going are 100%- if the engineers aren't sticking it to the company of course!

[desmotronic]

Well whaddya know forecast temps good for something after all. Good to see we not leading the newbies astray and a few misconceptions resolved.

My minima is 2000m.

[morno]

Errr, I'm still failing to see where Keg has referred to the forecast temperature?