Here is a sample forecast from a (marginal) soaring day, broken into sections, with explanations about each section's contents.
TI Report for Sterling, MA for Tue Apr 23 09:10:06 EDT 1996
ALB upper air data from David J. Knight's server at SUNY Albany.
Forecast max temp from AVN MOS for Fitchburg MA from nws.noaa.gov.
|
Upper-air temperature and wind measurements are taken via balloons at Albany NY at around 1200Z (8 am EDT, 7 am EST). Albany is the closest weather balloon station to Sterling. There are also such stations in Grey ME and Chatham MA, but the Chatham station is typically down-weather from us and thus less a forecast than a history; also, both are greatly affected by the presence of the Atlantic. Since Albany is typically up-weather from us on thermal-soarable days, it tends to have excellent predictive ability on those days.
The forecast high is from the National Weather Service's "AVN" weather model, and is usually +/- 2 degrees of the actual Sterling high...but not always. Historically the forecast high for Concord NH was used for Sterling, but it was not entirely satisfactory. In 2002 forecasts became available for more towns, so the forecast for the town of Fitchburg, less than 10 miles from Sterling - and more importantly, at the same altitude - is now used when available.
I have several friends who are meteorology professors, who have made an effort to make sure the upper-air and temperature data I use is available as early as possible on their university web servers. My program checks each of the sites, as well as the official NWS site, and generates the report as soon as good data appears at one of the sites. It is surprising how often the data is available from one of the university sites before it is available from the NWS!
=== Interpolations (temps in deg. F, altitudes in feet MSL) ===
MSL *TI* Wdir@kts trig VirT 2.1 degrees/division ("`": Dry Adiabatic)
----- ---- -------- ---- . ---- -----------------------------------------
12000 10.3 250 50 93 | 28.0 ` \
11500 10.0 92 | 30.1 ` :
11000 9.7 92 | 32.3 ` :
10500 9.5 91 | 34.5 ` :
10000 9.2 250 51 91 | 36.6 ` :
9500 8.7 90 | 38.5 ` :
9000 8.3 250 51 89 | 40.5 ` :
8500 7.9 89 | 42.4 ` :
8000 7.5 245 47 88 | 44.3 ` :
7500 7.1 87 | 46.3 ` :
7000 6.7 240 43 86 | 48.2 ` :
6500 6.3 86 | 50.2 ` :
6000 5.9 240 41 85 | 52.1 ` :
5500 5.5 84 | 54.0 ` :
5000 5.1 250 31 83 | 56.0 ` :
4500 2.8 79 | 54.6 ` :
4000 -0.3 250 16 74 | 51.6 /
3500 -0.7 73 | 53.5 :`
3000 -1.0 255 10 73 | 55.7 (CB: 2900) :`
2500 -1.4 255 9 72 | 57.6 : `
2000 -2.2 260 7 70 | 58.8 : `
1500 -3.0 69 | 60.1 (High: 69) : `
1000 -3.8 300 5 68 | 61.3 : `
500 -4.6 0 0 67 | 62.5 : `
|
This is a table of the Thermal Index, wind direction and speed, and interpolated trigger and virtual temperatures, every 500 feet. A graph of temperature, with altitude on the X axis and virtual temperature on the Y axis, is aligned with the table rows.
The graph also notes the forecast high temperature (High: 69) on the line corresponding to its matching trigger temperature, and the resulting cloudbase height (CB: 2900) on the line corresponding to its matching altitude. Amount of cloud increases or decreases in proportion to how much the CB: note is below or above the High: note, respectively. If CB is several lines above High, a blue day is likely; if several lines below, overdevelopment is likely.
The cloud base is estimated by the standard formula relating relative humidity and the forecast high temperature... which is reasonably accurate on soarable days. When the actual temperature is higher, the cloud bases will be higher. The cloudbase mentioned in the appended Terminal Aerodrome Forecasts (see below) are frequently more accurate when the cloud bases are below 6000' MSL.
Thermal Index ("TI") is the difference between the bouyancy of air at a given altitude and the bouyancy of air that has risen from the ground to that altitude. If the difference is negative, the rising air will continue to rise. When the TI is 0, it will tend to stop. When the TI is positive, it will tend to sink.
"Virtual" temperature is temperature corrected for relative humidity. It is used to compute the TI, because it more accurately reflects the bouyancy of a body of air; actual temperature is misleading when the air is humid, such as in the summer. There is a rule of thumb that says when the TI is more negative than -3, the air is sufficiently bouyant to raise a glider.
"Trigger" temperature for a given altitude is the ground temperature required to create air that can rise to that altitude. It is very useful when the actual ground temperature is not close to the forecast high; in general, you can usually soar to the altitude at which the trigger temperature matches the ground temperature. For example, if/when ground temperature reached 73 on the day this data describes, you could probably have soared to around 3500 feet. (Of course, this ignores effects such as low clouds, high winds blowing apart the thermals, etc.)
The graph is more useful the more you study it. For example, I have discovered that since it shows the relationship between temperature and wind speed, it predicts not only thermals but also waves. (This is why I sometimes expect waves when no one else does). In general, "stable" air (air in which virtual temperature remains the same or increases with altitude) tends to form waves when it is under horizontal pressure - evidenced by high winds - and is surrounded by unstable air. (There was very likey good wave starting at 4000 feet on this day; note the sharp increase in both temperature and wind at that level. The only question is whether the wave would continue in the unstable air above 5000').
It's no use finding some parts of the forecast that are positive, basing your decisions on them, and then blaming the forecast when you can't stay up. You must look at all the relevant information, and prioritize it properly, if you want to rely on the forecast. In order of decreasing importance:
- Wind
- Cloud base
- Lapse rate
- Predicted high temperature (adjusted based on local forecasts)
- Terminal forecasts
Of course, it should go without saying that you are looking at the right forecast (e.g. the first line says the date the forecast was generated!).
Note the graph part of the forecast is only useful if you display it with a Monospaced Font. If the adiabatic line drawn with "`" characters is not very straight, you are using an inappropriate font. Also, the graph requires a screen or printer 80 characters wide.
=== NWS Regional Summary - 405 AM EDT TUE JUL 7 1996 === High pressure over the Great Lakes will build to the mid Atlantic coast this afternoon, bringing fair weather throughout the northeast. Sunshine will boost afternoon temperatures to nearly 70 across the interior, but afternoon sea breezes will cool readings back into the 60s along the coastline. Another clear and cool night can be expected tonight. Our weather will then become unsettled for the middle of the week as the high slips offshore, and a front becomes stationary from the Ohio valley into the northeast. |
This is the latest NWS synopsis for central New England as of the time the upper-air data became available.
=== The weather observed at WORCESTER at 09:54 AM EDT was: === The skies were clear. The prevailing visibility was 10 miles. Temperature: 55F ( 13C) Dewpoint: 50F ( 10C) Relative Humidity: 82% Winds from the N (005 degs) at 2 mph. Pressure: 1014.8 millibars. Altimeter:29.96 inches of mercury. |
This is the latest official observation from Worcester Airport as of the time the upper-air data became available.
=== ORH Terminal Aerodrome Forecast issued 7-Jul-1996 07:30 EDT === 8am: 30012g20kt P6sm Sct050 7pm: 30007kt P6sm Skc |
The above is the standard Terminal Aerodrome Forecast for the Worcester MA airport. I include it in the report partly because it's a forecast, and includes surface wind and the official data that we are supposed to use to decide between VFR, MVFR, and IFR; but also, because we are supposed to be able to read these silly things, so this will give us practice. This forecast is almost always available by the time the TI forecast is, but my system delivers the TI forecast as soon as it is available, with whatever TAF is available at that time.
Below is the complete original (decoded) data from which the thermal index part of the report was generated.
P(mb) H(ft) Tv(C) T(C) DP(C) wind dir wind spd
1002.0 292 15.8 14.4 10.7
1000.0 348 16.6 15.0 12.1 0 0
996.0 459 17.0 15.2 14.3
920.0 2661 14.0 12.4 11.4 255 9
877.0 3974 10.8 9.4 8.9
850.0 4831 13.7 12.6 4.6 250 31
700.0 10076 2.4 1.6 -2.1 250 51
634.0 12668 -3.8 -4.5 -5.1
601.0 14049 -4.5 -5.1 -8.3
551.0 16270 -9.7 -10.1 -12.6
547.0 16454 -9.9 -10.3 -14.4
538.0 16873 -10.2 -10.3 -29.3
534.0 17062 -10.6 -10.7 -30.7
529.0 17299 -11.1 -11.3 -24.3
523.0 17586 -11.8 -11.9 -30.9
500.0 18710 -14.2 -14.3 -30.3 245 64
483.0 19567 -16.6 -16.7 -29.7
469.0 20289 -18.5 -18.7 -23.6
457.0 20921 -19.9 -20.1 -27.1
442.0 21730 -21.7 -21.9 -26.2
418.0 23073 -23.9 -24.1 -25.6
405.0 23829 -24.6 -24.7 -30.7
400.0 24125 -25.2 -25.3 -32.3 240 61
314.0 29731 -38.9 -38.9 -48.9
306.0 30312 -39.1 -39.1 -49.1
300.0 30756 -40.3 -40.3 -50.3 255 76
250.0 34738 -51.3 -51.3 -60.3 260 76
200.0 39363 -63.5 -63.5 -71.5 265 95
197.0 39667 -64.3 -64.3 -72.3
178.0 41702 -64.3 -64.3 -71.3
150.0 45080 -71.1 -71.1 -78.1 245 58
145.0 45738 -71.3 -71.3 -78.3
138.0 46720 -61.9 -61.9 -69.9
132.0 47626 -60.1 -60.1 -68.1
This is the exact, complete data returned by the weather balloons. It is provided here for academic interest; the summary given by the report is sufficient for all soaring needs. (The complete data does tell how far out of reach the wave is, when you can see lennies but fail to reach the wave...).
P is pressure, H is altitude, Tv is virtual temperature, T is actual temperature, DP is dewpoint. I suspect it may be possible to use the DP and T to do a better job of estimating cloud base... any volunteers willing to do some work in this area? We can get the Albany terminal observations and compare them with the Albany T's and DP's...
I collaborated with Kevin Ford of Texas A&M University on the Thermal Index part of this reporting system. For more information about the program behind the reports, see our "white paper". For more information about the theory of thermal indexes, see the Soaring Handbook and other Aviation Weather sources.