- Количество слайдов: 21
WWRP RDP/FDP FROST-2014 activities in 2011/2012 (FROST: Forecast and Reseach – Olympic Sochi Testbed) Dmitry Kiktev Hydrometcentre of Russia / Roshydromet 16 April 2012, Moscow
Two clusters of «Sochi-2014» Olympic venues Snow sports competitions Ice sports competitions
Goals of RDP/FDP FROST-2014: • To improve and exploit: – high-resolution deterministic mesoscale forecasts of meteorological conditions in winter complex terrain environment, including downscaled modeling; – regional meso-scale EPS (Ensemble Prediction System) forecast products in winter complex terrain environment; – nowcast systems of high impact weather phenomena (snow levels, wind, visibility, precipitation type and intensity) in complex terrain. • To improve the understanding of physics of high impact weather phenomena in the region; • To deliver deterministic and probabilistic forecasts in real time to Olympic weather forecasters and decision makers. • To assess benefits of forecast improvement (verification and societal impacts) As the project evolves these goals might be detailed.
Project components: • Obsevations • Nowcasting • Deterministic NWP • Ensemble NWP • Data assimilation • Information Technologies • Training and understanding • Verification and Impact Assessments
Observational network in the region of the Games None of practically realizable near-surface atmospheric monitoring networks can be representative enough given the complexity of the region and high Olympic demands. In situ observations are mostly concentrated along the coast. Vast sea area from one side and nearby mountains on another side of Krasnaya Polyana are data sparse areas. Nevertheless, today situation is substantially better than 1 -2 years ago, and enhancement of the network will continue. + 48 near-surface automatic stations are installed to enhance the observational network in the region. + Data flow from new Doppler radar, wind and temperature/humidity profilers, and micro-rain radars will be available by the next winter; + More frequent sounding data at the nearest aerological stations will be available
Current Instrumentul Setup: AMSs and Sensors Installed Station Name Air temperature Relative humidity Wind Direction Wind Speed Measured parameters Liquid Solid Snow Precip. Height Amount Snow surface T Snow T Visibility Atm. Pressure Cloud base height + + + + + + + Roshydromet’s AMSs Aviation station «Adler» Automated АМК «Sochi» Krasnaya Polyana Kordon Laura Aibga Alpica-1000 Alpica-1500 Kepsha Solokh-Aul Lazarevskoye Magry Imeretinka + + + + + + Biathlon-1400 Biathlon Stadium Skiing Stadium Biathlon-1500 + + + + Ski Jumping-800 Ski Jumping-650 + + Sledge-830 Sledge-700 11 Sr (Aibga) 12 Sr 13 Sr (Men’s start) 18 Sr (Women’s start) 14 Sr(Super-Giant) 17 Sr(Finish) Snowboard Freestyle + + + + + + + + + + + + + Biathlon + Skiing Venue + + + + + + + + + + + + + + + + + + + + + Ski Jumping + + Sledge + + «Roza-Khutor» + + + + + + +
Current Instrumentul Setup: AMSs and Sensors Installed - II Station Name Measured parameters (Vaisala WXT-520) Air temperature Relative Humidity Wind Direction Wind Speed Liquid Precip. Amount Solid Precip. Amount Atm. Pressure «Meteofon» stations at the towers of mobile communication Sochi-Plastunka Adler-Galitsino Esto-Sadok Matsesta-Chai Akhun mountain Krasnaya Polyana Kalinovo lake Adler-Airport Veseloye Ermolovka Lesnoye Pogranichnik Loo Pikhtinka Zubova shel Adler-Norluis Adler-Moldovka Sochi-Obzornaya Sochi-Verbliud Veseloye-Mirra + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
Current network of AMSs in the region of Sochi and its foreseen enhancement Designations: Red markers – Roshydromet’s AMSs; White – AMSs of sport venue owners; Yellow – AMSs already installed by Megafon company Green – AMSs to be installed by Megafon company
Sensors installed for individual meteorological variables Snow Height Visibility Solid Precipitation Cloud Base
Instrumentul Setup: Remote Sounding • Vaisala Doppler radar WRM 200 was bought 1 year ago, but it is being installed on Akhun mountain in Sochi only now (considerable time has been spent to get endorsements). • Two profilers will be installed on mobile platform: - Temperature/Humidity – HATPRO (RPG Gm. Bh, Germany) - Wind Profiler (Scintec Corporation, U. S. A. ); • Scintec SFAS wind- and MTП-5 temperature profilers; • Micro Rain vertically pointing Radars (MRR-2, METEK Corp. ) • Data flow – autumn 2012.
Nowcasting Many issues should be tackled within this project component, e. g. : - Winter nowcasting of multi-weather elements (wind speed and wind gust, visibility, precipitation intensity and time); - Improvement of blending procedures for NWP and extrapolated observations for winter; - Radar retrieval of precipitation type and intensity; - Assessment and account for observational uncertainty. The project gives a chance to develop mesoscale NWP to fill the gap in 4 -6 hour. Nowcasting potential of participating NWP models should be assessed for direct and post-processed (e. g. Kalman filter, 1 -D model, MOS) model forecasts. Besides the meso-scale models, the specialized nowcasting systems are expected to be used in the project – ABOM, CARDS, INTW, STEPS, INCA, WSDDM, GRAPES.
Numerical weather prediction • Complexity of Sochi region stimulates application of highresolution modeling. • Key areas to be addressed: data assimilation; physics, validation and numerical challenges at high resolution; predictability and uncertainty. • High-resolution data assimilation is a necessary prerequisite for meso-scale forecasting. The remote sounding is the main source of meso-scale structures in the initial data for such a modeling. Potential input for assimilation: Doppler radars; Wind and temperature/humidity profilers; Satellite radiances (AMSU-A, AMSU-B, AVHRR, IASI, SSMIS); Satellite winds (AMV, ASCAT). • Convective-scale multi-model ensemble forecasting might be a new experience of Sochi-2014
NWP component: progress and achivements in 2011/2012 HMC ARPA SIMC DWD Env. Canada Deterministic EPS forecast Operational Experiments with 2 km forecasting over the resolution EPS are in Sochi area with preparation. COSMO-2. 2 km Data Assimilation Comment A regional 3 D-Var was developed by adaptation of the global HMC data assimilation technology (using stretched geometry). - Implementation of the first prototype of 7 -km ensemble, running on a regular basis since 19/12/2011. - Provision of LBC for the 2. 2 km ensemble at HMC. - Provision of EPS-fields to Russian forecasters. Test runs of COSMO-RU with (for assessment 7 km resolution with DA at DWD of DA contribution) Automated system to No work has been done for this Progress have been recently Ready for realbeen achieved with a 10 km downscale GEM from component. It is unlikely that time set-up. 4 DVAR assimilation system. global model (33 km) we will be able to run a Need to This approach will be used for to 1 km has been set mesoscale EPS for FROST. incorporate high 2012/2013 season. up and tested. A -resolution Surface initial conditions single run test was should be provided by a 100 m surface analyses external surface system. This from the 100 m performed over the aspect should be ready for the external system. Sochi region upcoming season. FDP: in blue; RDP: in green
NWP component: progress and achivements in 2011/2012 - II Deterministic forecast NCAR ? CMA EPS ZAMG NOAA TIGGE HIRLAM FDP: in blue; RDP: in green Data Assimilation Comment ? No input till winter 2012/2013 Satellite-based snow cover Accuracy and snow water equivalent estimates are retrievals over the Black Sea performed region have been routinely through evaluated and archived. comparison of First tests of the new 0. 5 km resolution snow cover product derived snow from VIIRS NPP have been properties vs conducted. station observations Some formal steps are needed from Roshydromet GLAMEPS semi operational. Technical work in setting up Harmonie to run in ensemble mode finished. First test with Harmon. EPS for the area of Sochi run sucessfully
NWP component: Milestones for 2012/2013 HMC Deterministic forecast Attempts to introduce data assimilation into Sochi-area COSMO based forecasting. Consolidation of SOCHMEL 7. Generation of new products. Generation of point forecasts (meteograms) with dissemination in xml format. Generation of products from the combination of different LAM-EPS ARPA SIMC DWD ZAMG EPS COSMO-RU (7 km) with data assimilation for winter 2011/2012, 2012/2013 and 2013/2014 in real time at DWD; transfer of selected forecast fields to Russia; same set-up as operational COSMO-RU (which is based on interpolated GMEanalysis). Data Assimilation Regional 3 D-Var based cyclic data assimilation scheme over the Sochi area. Verification Operational mesoscale verification package. Interfacing of radar data for the mesoscale verification package Subject to observation availability, the performance of SOCHMEL 7 will be assessed Operational mesoscale verification package as well as subjective evaluation by forecasters. Main emphasis on nearsurface weather conditions. Setup of Post-processing domain in ALADIN-LAEF for Sochi (Summer 2012) Implementation of ALADIN-LAEF data flow to the project server (Autumn 2012) FDP: in blue; RDP: in green
NWP component: Milestones for 2012/2013 - II Env. Canada Deterministic forecast EPS Items to work on: It is not expected that any • nesting down to 250 progress will be achieved for m (very achievable on this aspect. new EC supercomputers) • install this system in real time NOAA HIRLAM Providing GLAMEPS results routinely. Run Harmon. EPS experiments for the area of Sochi. Calibration of EPS forecasts. Data Assimilation 1: 3 DVAR or 4 DVAR for 10 -km LAM (to replace Global-33 km as driving model) 2: 100 m – from external land surface modeling system 3: high-resolution snow analysis Routine generation of 0. 5 km resolution daily maps of snow cover from VIIRS NPP over the study area. Will work on an improved algorithm for snow depth and snow water equivalent retrievals from satellite observations in the microwave. Will generate snow cover distribution monthly climatology for the Sochi region. Verification If data from the various observation networks over the Sochi area, and if resources are available, objective evaluation will be performed. FDP: in blue; RDP: in green
NWP component: Milestones for 2012/2013 - III Deterministic forecast TIGGE EPS Generation of products from available global EPS. Combination of global EPS and LAM-EPS output. Data Assimilation Comments FMI To be specified NCAR To be specified ? CMA Basel Univers. Ity ? High resolution (1 km) fog forecast for Sochi region, Eventually semioperational (depending on cpu-availability). Extensive postprocessing for selected locations, and point forecast products. ? UKMO ? FDP: in blue; RDP: in green No promises
Project IT support • Project data storage with Internet-access is established; • Project web-site - http: //frost 2014. meteoinfo. ru (MAP D-Phase-like interface is being developed)
Verification • Data storage with Internet-access is already in place; • More advanced data quality control is needed for observations; • Home-made tools, VERSUS, MET …; • As for SNOW-V 10, it is of interest to quantify the added value of forecast refinement between: - Global model; - Regional model without and with its own data assimilation; - High-resolution model with and without data assimilation; - Post-processed model output (Kalman filter, MOS, 1 D-model etc. ); - Nowcasting (based on latest observations and blended with NWP). • Distributed verification activity ? (Roshydromet–FMI (JWG on Verification Research))
“- Forecasters do not always like probabilities (at any scale!) - End-users «hate» probabilities. ” Andrea Montani, Michael Ttsyrulnikov (1 st FROST-2014 meeting) • That’s funny, but the problem behind is very serious. It comes from misunderstanding and lack of training. • Training for Olympic forecasters is held on regular basis and participants are involved into provision of meteorological services for test sport events… Despite this training is getting a crucial issue… • FROST-2014 is RDP/FDP… As the project has an FDP component, it must be an “end-to-end” project and include impact assessment component. First steps in this direction are tried. . .
Thank you! http: //frost 2014. meteoinfo. ru