AVN MOS WIND DIRECTION AND SPEED GUIDANCE
February 3, 2000
PREDICTANDS:
Wind speed valid every 3 hours from 6 to 72 hours after initial model time of 0000, 0600,
1200, and 1800 UTC.
- u- and v- wind components valid every 3 hours from 6 to 72 hours after initial model
time of 0000, 0600, 1200, and 1800 UTC.
All predictands are valid on the hour and were taken from TDL's archive of the hourly
METAR data. The u- and v-wind components were computed from the surface wind
direction. These wind observations represent a 2-minute average wind speed and direction at
the 10-m level.
PREDICTORS:
- AVN forecasts of u- and v-wind components and wind speed at 10 m and at isobaric
levels in the lower troposphere; relative vorticity, divergence, and vertical velocity at
isobaric levels in the lower troposphere; thermal stability and relative humidity over layers
in the lower troposphere; model fields valid at the time of the predictands.
- Surface u- and v-wind components and wind speed observed at initial model time + 3
hours; observations are potential predictors only for the 6- through 15-h forecast projections. Note: the use of observed predictors in an operational environment necessitates
development of both primary and secondary equations for forecast projections in which
the observations may be used; the secondary equations are used when observations are
missing; in all tests, both the primary and secondary equations were evaluated.
- Sine and cosine of day of year and twice day of year
FINAL DEVELOPMENTAL SAMPLE:
- warm season (April-September): April 1, 1997 - October 15, 1997; March 16, 1998 -
October 15, 1998; March 16, 1999 - September 30, 1999.
- cool season (October-March): September 16, 1997 - April 15, 1998; September 16,
1998 - April 15, 1999. Note that extra developmental data were added at either end of the
seasons to increase sample size and provide a smoother transition between seasons. The
exact sample dates reflect the time at which the final equation development was done.
TEST SAMPLES: The warm and cool season samples were divided into two parts to assess the
quality of the new guidance, namely, a dependent sample used to develop test forecast equations, and
an independent sample on which to evaluate the forecasts made by the test equations. We chose a
90-day independent test period, that is,
- warm season independent sample: the last 15 days of April, May, June, July, August,
and September 1998;
- cool season independent sample: the last 15 days of October, November, and December
1998 as well as the last 15 days of January, February, and March 1999
VERIFICATION RESULTS:
- The AVN MOS wind speed and direction guidance appeared to improve significantly over
the NGM MOS guidance and the direct model output. The improvement was evident
during both the warm and cool seasons, and both the 0000 and 1200 UTC forecast cycles.
No tests were made on the 0600 and 1800 UTC cycles. Detailed results may be seen on
the TDL home page at: http://www.nws.noaa.gov/tdl/synop/results.htm