hm8 Posted October 4, 2012 Share Posted October 4, 2012 FWIW http://dd.weatheroffice.ec.gc.ca/bulletins/alphanumeric/20121002/NO/CWAO/17/NOCN03_CWAO_021644___00044 Link to comment Share on other sites More sharing options...
Amped Posted October 5, 2012 Share Posted October 5, 2012 IN DEVELOPMENT PHASE THE RDPS-300 WAS TESTED EXTENSIVELY. MANY CYCLES, EACH COVERING AT LEAST TWO SUMMER MONTHS AND TWO WINTER MONTHS WERE PERFORMED AND EVALUATED DURING THE DEVELOPMENT PHASE. FOR THE FINAL TESTS, OBJECTIVE VERIFICATION OF UPPER AIR AND SURFACE FIELDS AGAINST RADIOSONDE DATA SHOWED A VERY SIGNIFICANT IMPROVEMENT AT ALL FORECAST LEAD-TIMES. THIS WAS GENERALLY TRUE FOR ALL SEASONS TESTED AND FOR ALL REGIONS AND VARIABLES. THE ONLY - SLIGHT - DETERIORATION SEEN WAS WITH THE WINTERTIME COLD BIAS NEAR THE SURFACE, WHICH WE WERE NOT ABLE TO ELIMINATE. THIS IS NOT CONSIDERED A MAJOR PROBLEM WHEN COMPARED TO THE IMPROVEMENTS OF THE RDPS-300. 3.2 IN PARALLEL RUN THE NEW SYSTEM WAS ALSO RUN IN PARALLEL WITH THE OPERATIONAL SYSTEM FROM MID-JUNE TO THE END OF SEPTEMBER 2012. 3.2.1 OBJECTIVE EVALUATION THE OBJECTIVE EVALUATION OF THIS PARALLEL RUN COMPARED TO THE OPERATIONAL SYSTEM SHOWED IMPROVEMENTS SIMILAR TO THOSE OBSERVED DURING THE DEVELOPMENT PHASE. 3.2.2 SUBJECTIVE EVALUATION A SUBJECTIVE EVALUATION WAS DONE ON A DAILY BASIS BY OPERATIONAL METEOROLOGISTS AT CMC. WHEN A DIFFERENCE IN MASS FIELD FORECASTS WAS FOUND BETWEEN THE TWO SYSTEMS (ABOUT 30% OF THE TIME), A PREFERENCE FOR THE NEW SYSTEM WAS EXPRESSED IN A RATIO OF ABOUT 2:1. DIFFERENCES WERE MORE FREQUENT ALONG THE ATLANTIC COAST THAN THE PACIFIC. HOWEVER, WHEN THE SYSTEMS DIFFERED, PREFERENCE FOR THE NEW SYSTEM WAS STRONGER FOR THE PACIFIC AREA THAN FOR EASTERN NORTH AMERICA AND OVER THE ATLANTIC. FOR PRECIPITATION FORECASTS THE SIGNAL WAS NEUTRAL OR SLIGHTLY IN FAVOUR OF THE NEW SYSTEM, SIMILAR TO THE FINDINGS OF THE OBJECTIVE SCORES. METEOROLOGISTS OBSERVED A NOTICEABLE INCREASE IN THE DETAIL OF PRECIPITATION PATTERNS, ESPECIALLY WITH ORGANIZED MESOSCALE CONVECTION. STORMS OF TROPICAL ORIGIN DID NOT VENTURE OFTEN INTO OUR FORECAST AREAS, THOSE THAT DID WERE BETTER FORECAST BY THE RDPS-300. IT WAS FOUND THAT THEIR CENTRAL PRESSURE WAS LESS DEEP AND THEIR SPATIAL EXTENT WAS BETTER DEFINED THAN WITH THE OPERATIONAL RDPS. IN PARTICULAR, RDPS-300 PROVIDED A BETTER FORECAST OF HURRICANE LESLIE, WHICH AFFECTED CONSIDERABLY NEWFOUNDLAND ON SEPTEMBER 11, 2012. [ 3.3 PERFORMANCE OF DEPENDENT SYSTEMS 3.3.1 UMOS VERIFICATION OF UMOS (UPDATABLE MODEL OUTPUT STATISTICS) CONNECTED TO RDPS-300 DURING THE PARALLEL RUN COMPARED TO UMOS CONNECTED TO THE OPERATIONAL SYSTEM SHOW COMPARABLE RESULTS FOR THE 00, 06, 12 AND 18 UTC RUNS. SCORES DONE ON HISTORICAL CASES FOR WINTER 2011 AND 2012 SHOW SIMILAR RESULTS EXCEPT FOR SOME STATIONS. MORE TRAINING WILL BE PERFORMED DURING THE COMING WINTER SEASON TO KEEP THE UMOS PERFORMANCE AT LEVEL WITH THE OPERATIONAL UMOS SYSTEM. 3.3.2 SCRIBE THE SCRIBE FORECAST SYSTEM CONNECTED TO RDPS-300 IN PARALLEL MODE PERFORMED SIMILARLY TO THE OPERATIONAL SYSTEM, WITH A SLIGHT IMPROVEMENT IN FAVOUR OF THE NEW SYSTEM FOR THE PROBABILITY OF PRECIPITATION. 3.3.3 REGIONAL DETERMINISTIC WAVE PREDICTION SYSTEM (RDWPS) THE SIGNIFICANT WAVE HEIGHT FORECAST IS IMPROVED WHEN CONNECTED TO THE RDPS-300 OVER THE NORTH ATLANTIC DOMAIN. OVER THE GULF ST.LAWRENCE AND THE GREAT LAKES DOMAINS THE SIGNAL IS NEUTRAL. 3.3.4 REGIONAL COUPLED SYSTEM - GOLF OF ST.LAWRENCE (RDPS-CGSL) THIS SYSTEM WAS ALSO BROUGHT TO A SPATIAL RESOLUTION OF 10 KM. THE EVALUATION OF FINAL FORECAST SERIES IN DEVELOPMENT SHOWED A SLIGHT IMPROVEMENT FOR THE 2M AIR TEMPERATURE. 3.3.5 OTHER SYSTEMS THE REGIONAL AIR QUALITY DETERMINISTIC PREDICTION SYSTEM (RAQDPS), THE HIGH RESOLUTION (2.5KM) DETERMINISTIC PREDICTION SYSTEM (HRDPS), AND THE REGIONAL DETERMINISTIC PRECIPITATION ANALYSIS (RDPA) HAVE ALL BEEN ADAPTED TO THE NEW RDPS-300. THEIR PERFORMANCE AND CHANGES TO THESE SYSTEMS ARE DOCUMENTED IN SEPARATE GENOT BULLETINS AND TECHNICAL NOTES WHICH ARE ALSO AVAILABLE VIA THE LINK FOUND AT THE END OF THIS MESSAGE. 4.AVAILABILITY OF PRODUCTS NO NEW ANALYSIS OR FORECAST PRODUCTS ACCOMPANY THIS UPGRADE OF THE RDPS. THE INCREASE IN RESOLUTION AND THE NEW 4D-VAR ASSIMILATION SYSTEM ARE MORE DEMANDING IN COMPUTER RESOURCES, SO TO ENSURE THE AVAILABILITY OF MOST PRODUCTS AT THE SAME TIME THAN WITH THE SYSTEM BEING REPLACED, THE DATA ACQUISITION CUT-OFF TIME HAS BEEN SHORTENED TEMPORARILY BY 10 MINUTES TO 1 HOUR 55 MINUTES. ONGOING CODE OPTIMIZATION COMBINED WITH PROGRAMMED INCREASES IN COMPUTING CAPACITY WILL EVENTUALLY ALLOW US TO GO BACK TO THE PREVIOUS CUT- OFF TIME. THE AVAILABILITY OF THE SCRIBE SYSTEM OUTPUT DESTINED TO THE MSC REGIONAL PREDICTION CENTRES WILL NOT BE DELAYED EITHER. Looks better although I don't see anything quantatative. There should be less bogus TCs. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.