We continue our comparison of the new international codes with their domestic predecessors in this edition of METAR/TAF Training Notes. You will recall that in the last edition differences between the SAO and METAR were highlighted. Our attention now shifts to the FT and how it compares with the TAF.
The approach used in the SAO versus METAR comparison will serve us well in this Training Note. In other words, general format differences will be noted first, followed by a closer inspection of variances in the elements of each forecast group.
As a result of changes in program policy that have occurred within the past two years, the FTs themselves have gone through somewhat of a face-lift. To name just a few of the changes, the six-hour categorical forecast that NWS forecasters in the past tacked onto the end of the FT has been eliminated. The FT is now a full 24-hour forecast of specific weather parameters. Also, use of the probability term "slight chance" is no longer valid, making the minimum threshold for the mention of precipitation probability a "chance". And, amendments (now referred to as updates) no longer are disseminated with numbers.
These and other changes, although primarily made in response to critical comments from aviation users, have actually served another purpose. They have readied our forecasters for implementation of the TAF code. TAFs are full 24-hour forecasts of specfic weather elements. They also do not accommodate a means of denoting extremely low (i.e. slight chance) probability of precipitation. Further, TAFs contain no associated amendment numbers. So, without actually realizing it, NWS forecasters have taken the first step toward preparing for TAF implementation.
Many of the NWS aviation forecasters have some familiarity with TAF composition. We currently prepare, disseminate, and update, as appropriate, 23 separate TAFs in the Southern Region alone. Conversion software (TAFENC), however, has compromised our indepth knowledge of the code by automatically converting from our domestic FT issuances. As a result, much of what will be presented here and in future related editions of Training Notes may be somewhat unfamiliar to a broad cross-section of our forecasters. Add to this the complicating fact that new wrinkles in the existing TAF code are proposed for the 1996 implementation.
So, let's start our comparison by looking at the arrangement of the various parameters in both an FT and TAF. As one might expect, the order of the parameters in the FT follows that of the SAO, while that of the TAF follows the METAR. For an FT:
LOCID FT DYVT CLDS VSBY WX WND [VCNTY] RMKS
where DYVT represents the day of the month and valid time of the forecast, and [VCNTY] represents an optional term. Contrast this with the order of parameters in a TAF:
TAF LOCID DYTM DYVT WND VSBY WX CLDS RMKS
where DYTM and DYVT represent the day and time of forecast issuance and the day and time that the forecast is valid, respectively. As you can see, the WHAT, WHERE, WHEN, and WIND sequence that is characteristic of a METAR observation holds for the TAF as well, with only a minor variation in the WHEN part (to be discussed shortly).
With an understanding of this association, we are now in a position to examine the elements of a TAF in more detail. Let's start at the beginning.
Type of Forecast. This is an easy one. The forecast will begin with TAF for a routine issuance or TAF AMD for an update to the forecast.
Station Identifier. Again, the International Civil Aviation Organization (ICAO) four-letter identifiers will be used for U. S.-prepared TAFs. For most sites, it will be comprised of the normal three-letter identifier found in an FT preceded by a "K". The exception will be in Alaska and Hawaii where identifiers for their sites will be preceded by a "P".
Issuance Date/Time. The next element of the forecast is the day of the month and the time in UTC of issuance. As an example, consider 120920Z. The forecast was issued on the 9th day of the month at 0920 UTC (Z). The current FT does not accommodate such an element.
Valid Date/Time. The valid date/time of the forecast follows. Presently, TAFs, like the FTs, include only the 24-hour valid time of the forecast. A new twist being proposed for the 1996 implementation is to also include the day of the month for which the forecast is first valid. So, 121818 in a TAF means that the forecast is valid from 1800 UTC on the 12th to 1800 UTC on the 13th.
Wind. The wind portion of the TAF forecast, although significantly different from that of an FT, will change little over what is currently being coded. In other words, it will continue to include a three-digit direction from true north followed by a two- or three-digit speed. The speed is in knots and indicated as such by the "KT" at the end. The coding will accommodate wind speeds of 100 knots or more and has some peculiarities in coding variable directions and light speeds. These will be examined in more detail in a later edition of these Training Notes.
One additional wind group that is being considered for the new TAF code is for handling non-convective wind shear. Again, the details are better left for later, but suffice it to say that it will include a height of the shear zone (in feet) and a wind direction and speed (knots) above the zone. The shear group will follow the normal wind group in the forecast.
Forecast Time Groups. There are a number of options for denoting time period changes in the TAF. The first is "FM" (from) designator which is the equivalent of the two- or four-digit time designator in the FT. It serves as an indicator of a significant change in weather conditions and will always be a four-digit (hour and minutes) UTC time, which is a change from the current TAF policy. As an example, FM1630 would indicate that a significant change in one or more weather elements is expected to take place at 1630 UTC.
The term "BECMG" (becoming) can be used in much the same way as the two- or four-digit time indicator in the FT. However, it often denotes a less significant change and one that is expected to occur gradually with time. As such, it will be followed by a four-digit time (UTC) range over which the change is expected to occur. For example, BECMG 1518 denotes a change in one or more elements during the period of 1500 UTC to 1800 UTC.
The "PROB" term in a TAF is the equivalent to the probability term "CHC" in an FT. Unlike the FT though, it will give an actual two-digit percent probability of occurrence followed by a four-digit time group denoting the beginning and ending time that the percent probability is valid. As an example, consider PROB40 1823 which indicates a 40 percent chance of the occurrence of some phenomenon between the hours of 1800 UTC and 2300 UTC.
Finally, the TAF equivalent of "OCNL" in an FT is "TEMPO". This term denotes a temporary change and is followed by a four-digit UTC time range over which the change is expected. For example, TEMPO 0609 would indicate that a temporary change of some weather phenomenon is expected between 0600 UTC (Z) and 0900 UTC (Z). There are other stipulations on the use of the term TEMPO, but these will be addressed in a later edition of these Training Notes.
Once a time period designator is given, look for it to be followed by any or all of the elements in their normal sequence, i.e. wind, visibility, weather, and clouds.
Again, our goal here was to highlight similarities and differences between the TAF and FT. There are a number of other details of the TAF to cover, but these are best left to later editions. In fact, with comparisons of the METAR with the SAO and the TAF with the FT now out of the way, we can direct our attention to the details of the new codes. First things first, though. Next time we'll look more closely at a METAR observation.
FAA, 1995: New Aviation Weather Formats: METAR/TAF.
NWS, 1995: New International Aerodrome Meteorological Code: TAF and FT Comparison.
NWS, August 1993: International Aerodrome Forecasts, WSOM Chapter D-37.