RUC/MAPS Frequently Asked Questions
Q: When are the key dates for changes in
horizontal resolution to the NCEP operational RUC?
- Dates
- 13km (50 levels) - 28 June 2005
- 20km (50 levels) - 17 April 2002
- 40km (40 levels) - 6 April 1998
- 60km (25 levels) - April 1994
- See more detail on dates for changes to the operational NCEP RUC here .
- Other major implementations
- 17 Nov 2008 - add assimilation of radar reflectivity, improvements to radiation and convection physics. (TAMDAR assimilation added on 16 Dec 2008, along with new code to select mesonet reports closest to analysis time instead of first report in analysis window, generally 30-40 min prior to analysis time.)
- 11 July 2006 - improved analysis, model, post-processing components in RUC, including changing DFI (digital filter initialization) from adiabatic to diabatic version.
Answer: Precipitation fields only have non-flagged values for forecast output times, not analysis time (f00), since the RUC doesn't analyze precip, so this is as expected. Precip fields should be available from 1h forecasts onward.
Q: What is the actual valid time for RUC fields
(e.g., 3h forecast valid at 03z)?
Answer: Well, the forecasts in this case
are indeed "valid" at 03z, but
since many of the observations assimilated in the RUC
(and other NWP models) are centered 10-30 min before
the hour, the actual valid time is also offset
by 10-30 min.
This is discussed in
Benjamin et al., 2004, Mon.Wea.Rev.
See p. 15 in pdf version and p.503 in MWR version.
Q: What is the difference between the
NCEP and NOAA/ESRL/GSD versions of the RUC?
- 13km RUC versions
- Operational NCEP 13km RUC (operational implementation on 28 June 2005)
- Backup RUC13 at GSD - least frequently changed version of RUC. Runs out to 48h every 12h, out to 24h every 1h (as of 1/5/09). Backup RUC at GSD generally matches code running at NCEP, but may be more advanced in some cases, such as the April 07 - Nov-Dec 08 period during which the backup RUC included assimilation of radar reflectivity and TAMDAR assimilation prior to a 2-part RUC upgrade in Nov-Dec 2008. As of Jan 2009, the backup-RUC13 generally matches code in the NCEP-oper-RUC13, but the backup-RUC uses the ESRL/GSD observation data feed.
- Development RUC13 at GSD
- more experimental version of RUC.
See
http://ruc.noaa.gov/13km_RUC.changelog.html
for latest changes. Runs out to 72h every 12h,
out to 48h every 3h.
- Oct 2008 - added assimilation of NASA Langley cloud products, also includes improved overall cloud assimilation upgrades. Neither of these are yet in the NCEP-oper-RUC.
- 07 May 2008 - no difference at this time, from Oct07-Apr08, dev-RUC13 different from backup-RUC13 in variation of Grell-Devenyi scheme w/ non-local subsidence warming, but this was withdrawn on this date.
- 20km RUC versions
- As of 1/1/2009, 20-km versions of the RUC running at NOAA/ESRL/GSD are discontinued.
- Prior to 1/1/09, 20km dev RUC - Uses same code as operational RUC13, but running at 20km horizontal resolution.
- Prior to 1/1/09, 20km dev-2 RUC at GSD - Same as 20km dev RUC except that dev2 RUC assimilates TAMDAR aircraft observations.
26Jan05 - Q: If the height from the lowest level in the RUC native coordinate
matches the RUC model terrain height exactly, how can the
winds from the lowest native level be non-zero?
The lowest native level is 5m above the surface,
strictly speaking, except for its calculations of
heights. In other words, the winds, virtual potential
temperature, and water vapor mixing ratio can all be considered
as 5m above the surface in regard to the atmospheric/land-water
boundary.
Note Fig. 6 in the Feb 2004
Monthly Weather Review paper about
the RUC model on p.482, which shows 2.5m as the half-point
between the actual surface and the 5m first atmospheric level.
Note also that there are 2-m and 10-m winds in the native
file, reduced in a post-processing via Monin-Obukhov similarity.
Q: (an old one) What is the difference between RUC and MAPS?
The RUC is the operational version run at NCEP.
Versions of the RUC are also run at FSL as a
backup to the operational run and as test versions.
MAPS is a now-ancient name used
in the early-mid 1990s for development versions of the RUC.
Q: What is the history of the operational RUC?
- 1994 - RUC starts running at NCEP, first version with 60-km horizontal resolution and 3-h assimilation cycle.
- 1998 - Implementation of 40-km RUC with 1-h assimilation cycle
- 2002 - Implementation of 20-km RUC with 1-h assimilation cycle
- 2005 - Implementation of 13-km RUC with 1-h assimilation cycle
- History of RUC system with detailed description - written April 2004 - from NWP 50th anniversary symposium
Q: What is the wind direction shown in RUC weather maps?
The wind flags for these maps are the usual convention for
weather maps, showing the direction from which the wind is coming.
If the wind flag is toward the west, that means that the wind
direction is from the west toward the east.
Also, note that the RUC grid map projection is not Mercator
(it's Lambert conformal), so
right-to-left winds are generally not due west (except at 95 deg W).
Finally, wind in RUC grids are grid-relative for direction, not
lat/lon-relative. (see last FAQ item for information on rotation
to lat/lon-relative)
Q: Why do I get forecasts from different runs when
clicking the "All Times" button for RUC or WRFRUC products?
This can sometimes happen because early output times from
new runs are processed into products while the run is still
continuing in order to make available the latest
products. Also, sometimes runs may be missing or products
may be missing because of computer/disk outages.
Q: Could you please run the RUC model out past 12 hours?
Almost every RUC version run at GSD goes beyond 12h,
with the Backup RUC13 out to 48h, and the dev RUC13
out to 72h (for 00z and 12z init times).
Q: Are the mass and velocity values defined at the same grid point
in RUC output grids?
The answer is yes.
The native RUC model horizontal grid is an Arakawa C grid, in
which there is staggering. This is in the model itself. However,
all RUC wind output, even in the hybrid-b native vertical
coordinate, is unstaggered back onto the mass points. So that
is what you are seeing in any output file, unstaggered winds.
Q: Do you have FORTRAN code for converting the u and v wind components aligned to the RUC
grid to components with respect to true north (meteorological standard)?
Here is a code snippet that should do the job.
PARAMETER ( ROTCON_P = 0.422618 ) PARAMETER ( LON_XX_P = -95.0 ) C** ROTCON_P R WIND ROTATION CONSTANT, = 1 FOR POLAR STEREO C** AND SIN(LAT_TAN_P) FOR LAMBERT CONFORMAL C** LON_XX_P R MERIDIAN ALIGNED WITH CARTESIAN X-AXIS(DEG) C** LAT_TAN_P R LATITUDE AT LAMBERT CONFORMAL PROJECTION C** IS TRUE (DEG) PARAMETER ( LAT_TAN_P = 25.0 ) do j=1,ny_p do i=1,nx_p angle2 = rotcon_p*(olon(i,j)-lon_xx_p)*0.017453 sinx2 = sin(angle2) cosx2 = cos(angle2) do k=1,nzp_p ut = u(i,j,k) vt = v(i,j,k) un(i,j,k) = cosx2*ut+sinx2*vt vn(i,j,k) =-sinx2*ut+cosx2*vt end if end do end do