|
|
|
By Stan Benjamin
IntroductionA new version of the Rapid Update Cycle (RUC) was implemented into operations at the National Centers for Environmental Prediction (NCEP) in April 2002. This new version includes a doubling of horizontal resolution (40-km to 20-km), an increased number of computational levels (40 to 50), and improvements in the analysis and model physical parameterizations. A primary goal of the 20-km RUC (or RUC20) is improvement in warm-season and cold-season quantitative precipitation forecasts. Improvements in near-surface forecasts and cloud forecasts have also been targeted. The RUC20 provides improved forecasts for these variables, as well as for wind, temperature, and moisture above the surface. The RUC20 continues to produce new analyses and short-range forecasts on an hourly basis, with forecasts out to 12 hours run every 3 hours. This implementation follows previous major implementations of a 60-km 3-hour cycle version of the RUC in 1994 and a 40-km 1-hour cycle version (called RUC-2 or RUC40) in 1998.The RUC forecasts are unique in that they are initialized with very recent data. Thus, the most recent RUC forecast usually has been initialized with more current data than other forecast model runs available. Even at 0000 or 1200 UTC, when other model runs are available, the RUC forecasts are useful for comparison over the next 12 hours. Although the RUC differs significantly from other models at the National Centers for Environmental Prediction, (NCEP), its key unique aspects are its hybrid isentropic vertical coordinate (used in theanalysis and model), hourly data assimilation, and model physics. Hourly analyses are particularly useful when overlaid with hourly satellite and radar images, or hourly observations such as from surface stations or profilers. RUC analyses and forecasts are useful for evaluation of short-term predictions of the Eta and Aviation models. The RUC20 will benefit general public forecasting, severe weather forecasting, and aviation. Here we present an overview of the model's changes and an assessment of their likely impact on aviation applications. More detailed information about all aspects of the RUC20 is available at http://ruc.fsl.noaa.gov. Horizontal and Vertical Resolution ChangesThe RUC20 occupies the same spatial domain as the previous 40-km RUC-2, as shown in Figure1. The RUC20 grid points are still a subset of the AWIPS Lambert conformal grid (AWIPS/GRIB grid 215 for 20-km) used as a distribution grid by the National Weather Service. Direct use of the AWIPS grid reduces the number of distribution grids for the RUC. The AWIPS grid ID for the RUC20 is 252, compared to 236 for the RUC40. The RUC20 domain size is 301 x 225 grid points (compared with 151 x 113 for RUC40). The 20-km grid spacing provides better resolution of variations of terrain elevation, leading to improved forecasts of topographically induced low-level circulations, mountain waves, and orographic precipitation. It also allows better resolution of land-water boundaries and other land-surface discontinuities. While the most significant differences in the terrain resolution of the RUC20 vs. RUC40 (Figure 1) are in the western United States, a number of important differences are also evident in the eastern part of the domain. The true grid spacing is 20.317 km at 35ºN, about 19 km at 43ºN, and about 16 km at the northern boundary.
Figure 1. Terrain elevation for 40-km RUC-2 (RUC 40), above and the RUC-20 (20-km), below. The RUC20 continues to use the hybrid isentropic/sigma coordinate of previous versions, and has 50 vertical levels, compared to 40 levels for the RUC40. Additional levels have been added near the tropopause and lower stratosphere and also in the lower troposphere. Implications for Aviation Forecasts: More detailed near-surface forecasts from higher horizontal resolution, sharper (and more realistic) frontal structures, including areas of potential for turbulence and icing. Improved forecasts of wind and temperature at the tropopause and lower stratosphere. Analysis ChangesThe key enhancements to the RUC20 assimilation of observations are an improved optimal interpolation (OI) analysis and the assimilation of GOES cloud-top pressure, as discussed below. The most important changes in the RUC20 OI analysis are the observation preprocessing and matching to background values. For assimilation of surface data, the RUC20 uses a background for 2-m temperature/moisture and 10-m winds at the station elevation, instead of 5-m values for both. The effect of this is reduced biases in the analysis. Surface observations near coasts now match a background value with the correct land use to avoid inappropriate surface analysis results near coasts. Individual aircraft ascent/descent profiles are used much more completely in the RUC20 by a more intelligent observation search strategy that still avoids ill-conditioning in the OI matrix solution.The RUC20 analysis also includes modifications to the three-dimensional (3-D) hydrometeor fields using GOES cloud-top data (single field-of-view data provided by NESDIS). The purpose of the GOES cloud-top assimilation is to improve short-range RUC forecasts of cloud/hydrometeors and precipitation. With the RUC40, the initial conditions for these fields were simply those carried over from the previous 1-hour RUC forecast. In the RUC20, cloud clearing and building is performed, providing more realism in those 3-D fields. The analysis/prognostic hydrometeor fields that are modified in the RUC20 cloud assimilation are mixing ratios of cloud water, ice, rain, snow, graupel, and mixing ratio, and the number concentration for ice particles. Implementation of a 3-D variational scheme for the RUC20 was deferred until later this year, after testing of recent modifications revealed a need to produce a slight improvement in 3-hour wind forecast accuracy. Implications for Aviation Forecasts: Better matching of surface observations in RUC20 analyses, improved fit to aircraft observations, improved wind and temperature 3 – 12 hour forecasts. Improved cloud, icing potential, and precipitation forecasts out to 12 hours. Forecast Model ChangesThe RUC20 forecast model is also significantly modified from the RUC40 version, including the following changes:
Implications for Aviation Forecasts: Improved forecasts of warm-season and cold-season precipitation, precipitation type, icing potential, clouds, surface variables.
Postprocessing ChangesSeveral of the RUC diagnostic algorithms have been changed for the RUC20, including those for:
Improved BUFR data will be available from RUC20. Hourly BUFR soundings with the same format as used for the Eta model will be available with the RUC20, including individual station files. The station list is the same as that used for the Eta model for stations within the RUC domain. (One small difference in the BUFR data is that the RUC uses 6 soil levels compared with 4 levels with Eta BUFR output.) Implications for Aviation Applications: Decreased bias in RUC20 visibility fields (but too much variability in nighttime RUC20 visibility fields), improved 2-m temperature and dewpoint, 10-m winds, easier access to model forecast soundings with hourly output to 12 hours. Other ImprovementsThe RUC20 uses more frequently updated boundary conditions from the Eta model, with new runs applied every 6 hours instead of every 12 hours (as with the RUC40).Land-use and soil-type fields are specified with much more horizontal detail in the RUC20. These fields are taken from 1-km raw land-use and soil-type information, with the prevailing type within each grid box of the RUC20. Statistical VerificationPrecipitation verification is performed for RUC20 and RUC40 12-hour forecasts by summing two 12-hour forecasts to produce a 24-hour total and comparing that with the NCEP 24-hour gauge-based precipitation analysis. The skill of precipitation forecasts is significantly improved with the RUC20, as shown in Figure 2, the 12-hour forecasts of 3-hour accumulated precipitation from the RUC40 (2a) and RUC20 (2b), and the radar image of the verifying period (2c). In this case, the RUC20 has accurately forecast much more intensity than the RUC40 to the southern end of a convective line, especially in eastern Louisiana and southern Mississippi. Not only is the intensity inproved in the RUC20 forecast, but also the position of the line is more accurately forecast, stretching from central Ohio into northwestern Alabama before bending back to eastern Louisiana.
Figure 2. Precipitation (inches) forecasts initialized at 0000 UTC 26 March 2002 from a) above left, RUC-40 and b) bottom left, RUC20 for 0900 – 1200 UTC (9 – 12-hour forecasts); c) right, Unisys radar summary valid a 1115 UTC (verification). Above-surface forecasts of wind and temperature are generally improved from RUC20 vs. RUC40 at both 3-hour and 12-hour projections. Case studies indicate (not shown here) that 3-hour RUC20 wind forecasts are more accurate than RUC40 counterparts by 0.1 – 0.3 ms-1 from 850 – 250 hPa. For 12-hour wind forecasts, RUC20 forecasts are improved by 0.2 – 0.5 ms-1 at all levels. The value of rapid updating is evident in that 3-hour wind forecasts are substantially more accurate than 12-hour forecasts valid for the same time. The same is true for temperature forecasts. RUC20 temperature forecasts are more accurate than RUC40 forecasts, and the largest improvement is in the lower troposphere (850 – 700 hPA). Verification of RUC20 and RUC40 2-m temperature and 10-m wind forecasts against METAR surface observations has been performed. These results show that the RUC20 provides superior surface forecasts to RUC40 in the daytime with comparable skill at night. ConclusionsThe April 2002 operational implementation of the Rapid Update Cycle with 20-km resolution will provide improved weather guidance for aviation and other operations both at and above the surface. Results of statistical and case study (not shown here but to be published later) inter-comparisons between the RUC20 and previous 40-km RUC-2 (RUC40) indicate that these improvements result in improved forecasts of wind, temperature, and moisture at upper levels and at the surface for wind, temperature, and precipitation. We will continue to improve the RUC20 and look forward to receiving comments from our users regarding its performance.AcknowledgmentsSpecial recognition is due each RUC Team member: John M. Brown, Kevin J. Brundage, Dezso Devenyi, Georg A. Grell, Dongsoo Kim, Tatiana G. Smirnova, Tracy Lorraine Smith, Barry E. Schwartz, Stephen S. Weygandt, and Geoffrey S. Manikin. We also acknowledge the significant effort of many colleagues at NOAA/FSL and NOAA/NWS/NCEP in the develop-ment, testing, and implementation of the RUC20. This work has been sponsored by the FAA Aviation Weather Research Program and NOAA.Note: A complete list of references and more information on this and related topics are available at the main FSL Website www.fsl.noaa.gov, by clicking on "Publications" and "Research Articles." (Dr. Stanley Benjamin is Chief, Regional Analysis and Prediction Branch, and can be reached at benjamin@fsl.noaa.gov. More information on RUC is available at http://ruc.fsl.noaa.gov.)
|
|
