SR/SSD 97-26 6-2-97

Technical Attachment


Tom Schlatter and Stan Benjamin
NOAA/Forecast Systems Laboratory
Boulder, Colorado

The following is a summary of initial information concerning an upcoming field evaluation of the new RUC-II (Rapid Update Cycle) model. The Forecast Systems Lab will coordinate the evaluation, in conjunction with NCEP and the NWS regions. Although only a few NWS offices are specifically involved in the evaluation (those identified below indicated a willingness to participate in response to an earlier request for volunteers), forecasters at all offices will be interested in how the RUC-II performs. We at FSL and NCEP plan to be as responsive as possible to ideas from field offices concerning the test and to questions and comments regarding the performance of the RUC-II itself.

What's new about the RUC-II?

1. Frequency of Assimilating Observations

In the old version of the Rapid Update Cycle (RUC), observations are assimilated once every three hours. In the new version (RUC-II) observations are assimilated hourly.

2. Time of availability

In the 60km RUC, the data cutoff time is +1h, 20 min. For the 40km RUC-II, the data cutoff time is about +0h 15 min, meaning that RUC analyses and forecasts will be available about 65 min earlier than currently. A catch-up cycle will be used to assimilate late-arriving data (running at about +0h 55 min), and forecasts will be run from the catch-up cycle rather than the first cycle at 0000 and 1200 UTC to catch rawinsonde data. At the non-raob times, RUC-II forecasts will be run off the first run since most of the data will be in by that time.

3. Computational grid

RUC-II will cover a geographical domain about 50% larger area-wise than the 60km RUC, extending further in all directions, but especially in the southeast, north, and west, and covering considerably more oceanic area. The old RUC grid has 60-km resolution (81 x 62 grid) and 25 levels. The new grid has 40-km resolution (151 x 113) and 40 levels.

4. Data assimilated

There are several new types of observations assimilated into RUC-II not present in the 60km RUC. These include:

5. Moist physics

Clouds must be inferred from the relative humidity field in the old RUC. In RUC-II, a comprehensive cloud physics package has been imported from the NCAR/Penn State MM5 mesoscale model. Mixing ratios of five types of hydrometeors are explicitly predicted: cloud water, rain water, snow, ice crystals, and graupel. Even the number density of ice crystals is predicted as part of the microphysical processes.

Cloud fields initialize each RUC-II model run, using the most recent hydrometeor forecast to avoid cloud spin-up problems.

6. Surface processes

The currently operational RUC calculates surface fluxes using a simple surface slab with constant soil moisture availability. The RUC-II features a multi-level soil and vegetation model with evolving soil moisture and temperature fields that are far more accurate than climatology. Snow accumulation and melting are accounted for.

The 60-km RUC employs a coarse land use data and derives sea surface temperatures from climatology and does not consider snow cover. The RUC-II employs improved land use data, including vegetation class, monthly vegetation fraction from NESDIS, and soil type, and utilizes daily detailed fields of sea-surface and lake-surface (for the Great Lakes) temperature and snow cover, It also cycles snow cover and canopy water along with the soil fields to further improve short-range forecasts.

7. Turbulence

The old RUC computes turbulent fluxes from a formulation due to Mellor and Yamada (Level 2.0). The RUC-II calculates the kinetic energy associated with turbulence (TKE) explicitly from equations derived by Burk and Thompson (Level 3.0). Boundary fluxes are improved using the explicit turbulence parameterization, and explicit TKE maxima are commonly found in upper-level frontal zones.

8. Radiation

There is no explicit calculation of atmospheric radiative fluxes in the old RUC, partly responsible for its warm bias. Full atmospheric radiation imported from the MM5 model is included in RUC-II. The radiative heating/cooling is influenced by hydrometeors. A more detailed write-up on the RUC-II with figures and verification is available on a special home page:

You may also want to look at the RUC/MAPS home page at

which displays real-time grids from experimental 40-km runs at FSL. The extensive information available on the RUC/MAPS home page may be useful to you.


The installation of RUC II code at NCEP is well under way and continuing at an intense level. Along with Geoff DiMego and Lauren Morone at NCEP, we are planning a field test of this new version of the Rapid Update Cycle during the summer. The new start date for the formal test has been set for Monday July 21. Data from the RUC-II should be flowing reliably at least two weeks before that. The field test will last six weeks until 31 August.


(List current as of 27 May 1997.)

There will be participation from the Storm Prediction Center (SPC), the Aviation Weather Center (AWC), and a number of WFOs from the Eastern, Southern, Central, and Western Regions. A current list of participants/contacts is given below; it is expected to grow and additional participants are welcome.

Aviation Prediction Center
     Focal point: Doug Mathews	(816) 426-3427 x239
     Also interested: Ron Olson		(816) 426-3367 x262
Storm Prediction Center (NCEP/SPC) Focal point: Phil Bothwell (405) 579-0706 Also interested: Paul Janish and Russ Schneider (405) 579-0719 (405) 579-0704

NWS Field Offices
     WFO Coordinator: Bob Reeves at NWS Office of Meteorology	(301) 713-1867 x113

Southern Region Bernard Meisner (SR contact) (817) 978-2671 Dan Petersen Melbourne Steve Amburn Tulsa Jeff Medlin Mobile Rusty Pfost Jackson Mark Jackson Brownsville Richard Wynne Amarillo
Eastern Region Ken Johnson (ER contact) (516) 244-0136 Josh Korotky Pittsburgh Mike Cammarata Columbia Jeffrey Tounge New York City
Western Region Mary Cairns Reno Larry Dunn Salt Lake City Ron Miller Spokane
Central Region Ed Berry (ER Coordinator) (816) 426-5672 Ray Wolf Davenport Kim Schafer Johnston Karl Jungbluth Johnston J. Kwiatkowski Goodland Paul Wolyn Pueblo
If we have listed any incorrect information, please contact us (FSL) and we will correct it:
        Stan Benjamin	(303) 497-6387
        Tom Schlatter	(303) 497-6938
Our mailing address is the same:
        NOAA/ERL/FSL  R/E/FS1
       	325 Broadway
       	Boulder, CO 80303-3328
Access to RUC-II Data

WFOs will gain access to the RUC-II output through their regional headquarters, which will, in turn, gain access through the OSO Server. The SPC and AWC, as part of NCEP, will get their data directly from the Cray C-90. Based upon its experience with a similar evaluation for the Eta model, conducted months ago, NCEP believes that access to grids through the OSO server by evaluators should be relatively straightforward.


Here are some initial thoughts on the key aspects of the evaluation to be made during the field test. We believe that the primary comparison during the field test should be between RUC-II and the current 60-km RUC. Perhaps the key question is:

Does the RUC-II give better overall performance than the first version of the RUC regarding
The evaluation should include both upper-air grids and surface grids, and should include a comparison of the RUC-II hourly surface grids with those from the RUC Surface 60-km hourly analysis system. It should also address the utility of RUC precipitation forecasts. There are several key new explicitly forecast fields from RUC-II that will be of interest to special users, including mixing ratios of cloud water, ice, rain, snow, and graupel; turbulence kinetic energy; and soil moisture and temperature at five levels. Of course, these fields have a strong influence on surface conditions, precipitation, and clouds.

Data available

Four types of files with RUC-II data will be available:

Isobaric (25-mb) grids of 3-D variables plus 2-D fields (including precip, fluxes, CAPE/CIN, tropopause fields, surface fields, fields from special levels - mean layers near surface, freezing, max wind, etc.) (~11-12 MB in GRIB format per output time).

Native (hybrid coordinate) grids of 3-D variables plus 2-D fields (~11-12 MB in GRIB format per output time).

Surface grids (0.2 MB per output time).

Model output soundings at ~489 sites including standard sounding variables plus surface variables (~7 MB for hourly output for a given run and all sites). Model output sounding sites are available at

Following are some of the variables that may be worthy of special attention. A complete list of variables available from RUC-II is in




Relative humidity

Cloud variables

Explicit TKE (turbulence kinetic energy)


Sea-level pressure (MAPS reduction)

Temperature, dew-point, winds

CAPE (convective available potential energy

CIN (convective inhibition)


Derived from 3-D grids of wind and temperature at AWC for AWC use:


Turbulence (diagnostic and explicit)

Just to get the discussion started, we suggest the following potential areas of concentration for different participants. Let us know how you would like to revise these.


Upper-level wind, temperature, moisture


Surface fields


Upper-level winds


Icing, relative humidity


Winds at all levels

Stability - CAPE/CIN


Surface fields


A Web newsgroup has been set up for the RUC-II field test evaluation (which will be continued afterward). It is at

This Web site is similar to a newsgroup, where evaluators can post questions and observations about model performance and developers can post responses as soon as possible, certainly by the next working day. Field test organizers can post other information. It is meant for day-to-day discussion.

This newsgroup is available for use right away. The content of this E-mail will be the first message posted, and we can all post items on it immediately. Note that the newsgroup allows you to request E-mail for responses on any items that you post, or E-mail notification of any posts.

Here is a first discussion item:

What do you think about opening this newsgroup up to anyone looking at experimental RUC-II output from the FSL RUC/MAPS home page?

Other types of feedback:


To further clarify the general direction of the evaluation and to provide a written response medium. We may want to quantify subjective impressions of the RUC-II performance vs. the 60km RUC with 1-5 responses (e.g., much better, slightly better, same, slightly worse, much worse) on specific days for specific fields.

Telephone conversations:

There could be a biweekly (?) teleconference between AWC, SPC, NWS/OM, FSL, and NCEP to discuss the evaluation, problems, etc. Regions or SOOs could also be brought in, but it would not be practical to bring them all in at the same time. There could even be video teleconferences if the number of sites participating is about 3 or smaller, for special situations.

Meeting at end of evaluation -- Friday 12 September in Norman or Kansas City?

Thank you for your help with evaluation. We look forward to getting the significant improvements in RUC-II into operations and this evaluation will certainly result in an even better RUC-II.