10B.2 Collision Of A Pineapple Express With An Arctic Outbreak . - Confex

1m ago
0 Views
0 Downloads
994.79 KB
12 Pages
Last View : 1m ago
Last Download : n/a
Upload by : Randy Pettway
Transcription

110B.2Collision of a Pineapple Express with an Arctic Outbreak over Complex Terrainsof British Columbia, Canada – Forecast Challenges and Lessons LearnedRuping Mo*, Quanzhen Geng, Mindy Brugman, Greg Pearce,Jim Goosen, and Brad SnyderPacific Storm Prediction Centre, Environment Canada, Vancouver, BC, CanadaAbstractThe interaction of a moisture-laden Pacific southwesterly flow, known as “PineappleExpress”, with very cold air from a strong arctic outbreak during 1 – 5 December 2007 produceda record number of high impact weather events across British Columbia of Canada. Among theweather hazards are bitterly cold wind chills, heavy snow, freezing rain, heavy rain, and strongwinds. The unusual collision of two contrasting air masses caused unprecedented forecastchallenges for both the Global Environmental Multiscale (GEM) models and meteorologists atthe Pacific Storm Prediction Centre (PSPC) of Environment Canada. In this study, the evolutionof the weather systems and the observed severe weather events during this storm cycle areanalyzed. Weather forecasts by the GEM models and PSPC meteorologists are compared withthe observed high impact weather events during the storm. It is demonstrated that meteorologistsat PSPC greatly improved the model forecasts by considering various local effects of thecomplex terrain in British Columbia that the GEM models cannot resolve well. In particular,with warm advection aloft, it is acknowledged that the GEM regional model tends tounderestimate the cold air entrenched in some narrow valleys. The model guidance was thereforedoomed to failure at forecasting the widespread and prolonged freezing rain in the midst of thestorm.*Corresponding author address: R. Mo, Pacific Storm Prediction Centre, Environment Canada, 201-401 BurrardStreet, Vancouver, BC, Canada V6C 3S5; e-mail: ruping.mo@ec.gc.ca.

21. IntroductionForecasting winter weather in BritishColumbia (BC) Canada is very challenging, dueto the complex terrain of the province (Fig. 1)and the lack of observed data over the NorthPacific (e.g., Hacker et al. 2003). During thewinter, an area of high pressure often forms inthe very cold air over Alaska and northernCanada. The arctic air pushes southwards intothe northern and central interior of BC beforecoming to rest; at least once or twice each year,the advance of the cold arctic air is so strong inBC that it spreads into the southern interior andflows through the mountain valleys to the coastalregions (Jackson and Steyn 1994; Johnson andMullock 1996) Meanwhile, numerous marinestorms are generated in the North Pacific eachwinter. Some of them follow the Pacific stormtrack toward BC, ensuring mild, wet, and windyconditions along the coast. On a few occasions,an area of low pressure over the NortheastPacific is capable of maintaining an influx ofwarm, moist air stretching from near theHawaiian Islands to the west coast of NorthAmerica (e.g., Fig. 2). This moist air current,frequently called the “Pineapple Express”(hereafter PE; Heidorn 2004, Mass 2008), is themost important ingredient for producing heavyprecipitation and flooding along the coast fromnorthern California to central BC (Lyons 1901;Loukas and Quick 1996; Lackmann andGyakum 1999). When a PE collides with anarctic outbreak over the complex terrain of BC,widespread weather hazards are almostinevitable, and that is just what transpired inearly December 2007.(Fig. 4) also play an important role in producingsnow across the BC South Coast on 1December and the following day (see furtherdiscussion in the next section).2. Evolution of the weather systemsAn area of high pressure was building overthe Yukon Territory toward the end of November2007 (Fig. 3). This high had pushed the coldarctic air southward into the BC Interior in themorning of 28 November. In the afternoon,northerly outflow winds began to develop alongthe coastal valleys and inlets as the cold airspilled out across the Coast Mountains. Thearctic high over Yukon reached its peak strengthwith a 1055 hPa center at 0000 UTC 1December (not shown). Locally heavy snowbegan to develop along the east coast ofVancouver Island, where the northerly outflowthrough the mainland inlets served as anonshore flow (see Jackson and Steyn 1994). Arapidly deepening low off Vancouver IslandFIG. 1. Physiographic and publicregions of British Columbia, Canada.forecast

3On 2 December, a low pressure systembegan to develop rapidly in the middle of thePacific (Fig. 5). Animation of satellite imagesfurther reveals that this low was fed by theremnants of at least one tropical depression andtwo named typhoons (Mitag and Hagibis). It wasthis low that set a powerful PE in motion, asclearly visible in the weather-satellite imageshown in Fig. 2. At 1200 UTC 3 December, thecenter of the low was about 1500 km southwestoff Vancouver Island (Fig. 5b). However, a warmfront accompanying the PE had already spreadmoist subtropical air flows over southern BC(Fig. 2). The PE arrival forced the cold arcticairmass to retreat from the South Coast and theSouthern Interior of BC, where, as expected,heavy rain and strong winds began to developon 3 December. The low itself eventually arrivedand moved onshore from the Central Coast on 4December (Fig. 5c); its collision with the arcticairmass put an end to this dramatic period ofearly-winter storminess in BC (Fig. 5d).FIG. 2. The “Pineapple Express” phenomenon, illustrated by the satellite (GOES-West) water vaporimage with derived winds at 1200 UTC, 3 December 2007. The black letter L marks a low center of sealevel pressure. Note that Hawaii (where the pineapple reference comes from) is located at the southwestend of the cold front.

4FIG. 3. The 1200 UTC mean sea level pressure (contour interval 4 hPa, solid) and 1000-500 hPathickness (contour interval 6 dam, dashed) for the period of 28 Nov. – 1 Dec., 2007, from the GEMregional model operational analysis. The surface front drawn in (a) was kept from reaching the BC coastby the strong northeasterly outflow; it moved onshore from the Washington coast instead in the eveningof 28 November.

5FIG. 4. PSPC surface analysis showing the rapid development of an outflow-induced lee low to the westof Vancouver Island. The central pressure of the low fell by 23 hPa during the 24-h period from 0600 UTC1 December (1011 hPa) to 0600 UTC 2 December (988 hPa).

6FIG. 5. The 1200 UTC mean sea level pressure (contour interval 4 hPa, solid) and 1000-500 hPathickness (contour interval 6 dam, dashed) for the period of 2 – 5 Dec. 2007, from the GEM regionalmodel operational analysis.3. Observed severe weather eventsFig. 6 shows the observed severe weatherevents that exceeded the public weatherwarning criteria in BC during 1 – 5 December2007. Cold wind chills, strong winds, heavyprecipitation, and extended periods of freezingrain affected most parts of the province.3.1 Cold wind chills and strong windsAs shown in Fig. 6a, strong arctic outflowwinds combined with cold temperaturesproduced wind chill values lower than -20 C fora long period over the North and Central Coast.Extreme wind chill values (lower than -40 C)were observed in the BC Peace River District.Meanwhile, strong southerly winds spreadacross the BC South Coast with the arrival of thePE system at about 1200 UTC 3 December.Southeasterly winds were observed over most ofthe exposed areas of BC South Coast.3.2 Heavy snowThe first major snowfall event occurred on 1December. Dry outflow winds, soaking upmoisture over the Strait of Georgia, deposited asmuch as 46 cm of snow over the east side ofVancouver Island. Meanwhile the lee lowoffshore helped spread snow across much of theremainder of the South Coast. On 2 December,

7the PE system began to spread snow across awider swath of BC. East Vancouver Island,Howe Sound, and Whistler were hit hard,receiving between 35 and 50 cm of snow in 24hours. Heavy snowfall was also observed insome interior regions such as Fraser Canyon,West Columbia and East Columbia.3.3 Freezing rainWarm air associated with the PE systemmoving over the arctic air entrenched in theboundary layer produced widespread andprolonged freezing rain in the southern andcentral sections of the province. Fig. 6c showsthat Howe Sound, Whistler and Fraser Canyonreceived up to 50 mm of freezing rain that lasted10 to 20 hours. Some other regions in southernBC also reported 5 to 10 mm of freezing rainlasting up to 10 hours.3.4 Heavy rainAs the warm and moist air accompanyingthe PE storm pushed further inland, the arcticairmass retreated to northern BC. Most of thesnow or freezing rain across southern BCchanged to heavy rain. More than 100 mm ofrain fell over most areas of the South Coast on 3December . West Vancouver Island and HoweSound reported 218 mm and 170 mm of rainfallin 24 hours, respectively.4. Forecast guidance from the GEM modelsGenerally speaking, the Canadian GEMmodels provided reasonably good forecastguidance for this storm cycle. The GEM globalmodel initialized at 0000 UTC 28 Novemberforecast heavy rain over the BC South Coast on3 December, with 116 mm at Squamish and 105mm at Vancouver International Airport (Table 1).Theseexceptionallylargeamountsofprecipitation, which continued to show up in thesubsequent model runs, served as an earlywake-up call (144-hour lead time) to theoncoming PE storm, and caught the forecaster’sattention immediately. Snow was forecast 72hours ahead for many regions of the SouthCoast. The 1 December global model runinitialized at 0000 UTC 29 November (notshown), was in good agreement with theobservations (Fig. 6b). The same model run alsopredicted 10 to 20 mm of precipitation on 2December, and over 100 mm on December 3 formost areas of the South Coast.Fig. 7 shows the GEM regional modelforecasts (up to 48-hour lead time) of the meansea level pressure and 1000-500 hPa thicknessfor 1200 UTC 3 December. As compared withthe analysis shown in Fig. 5b, positions of thePacific low and the warm front across southernBC were well forecast. The strength of the low,however, was somewhat over-forecast. Notethat both the strength and position of the arctichigh over northern Canada were well forecast.Both the strength and position of theoffshore low at 1200 UTC 2 December (Fig. 5a)was also well forecast by the GEM regionalmodel (Fig. 8). However, both of the arctic highover northern Canada and the outflow-inducedlee low along the BC coast were over-forecast;the deeper lee low in the model brought instronger southerly winds to the Inner SouthCoast and the Lower Mainland, as shown inFig. 9. The stronger warm advection resulted inmisleading forecast guidance for the area. Forexample, the guidance for Metro Vancouver,based on the model run initialized at 1200 UTCof Saturday 1 December (Fig. 8c), was given asfollows:“ Tonight.Snow changing to rain nearmidnight. Snowfall amount 5 cm. Becomingwindy. Low zero with temperature rising to 6 bymorning. Sunday.Rain. Temperature steadynear 5.”As it turned out, the warm advection brought inby the lee low was much weaker. Temperaturesobserved at Vancouver Airport were steady near-2 C through Saturday night and the maximumtemperature on the following day was only2.4 C. Snow in Metro Vancouver did not changeto rain until Sunday evening when strongerwarm advection from the PE system arrived.Our verification reveals that the GEMregional model failed to forecast all but one ofthe 14 freezing rain events in BC during 1 – 5December 2007. This poor performance is dueto the poor model resolution that cannot resolvethe complex terrain in BC (i.e., the modeltopography is much smoother than the reality,with shallower valleys and lower mountains).This leads to the cold air at low levels being“scoured out” too quickly by the warm advectionin the model.Model performance for rain and snow eventswas generally better. The GEM regional modelcorrectly forecast 63% of the snowfall warningevents and 60% of the rainfall warning eventsduring 1 – 5 December 2007. Still, the model did

8struggle in certain areas. Virtually allprecipitation warning events over EastVancouver Island and the eastern sides of theCoast and Columbia Mountains were missed bythe model. Meanwhile, the model over-forecastthe precipitation amounts on the windward(western) sides of the mountains. Again, the lackof resolution in the model topography is believedto have contributed to these errors in orographicprecipitation.FIG. 6. The observed severe weather events that exceeded public weather forecast warning criteriaduring 1 – 5 December 2007 over British Columbia. (a) Coldest wind chills and strongest winds, (b)maximum snowfall in 24 hours, (c) storm total freezing rain, and (d) maximum rainfall in 24 hours.

9Table 1: 24-h precipitation amounts (00Z-00Z water equivalent in mm) for selected cities on 3 December2007, forecast by the GEM global model 0000 UTC runs of 28-30 November 2007 and observed on 3December 2007.th00Z 2800Z 29th00Z 30thObservedPort 109869FIG. 7. Mean sea level pressure (contour interval 4 hPa, solid) and 1000-500 hPa thickness (contourinterval 6 dam, dashed), valid at 1200 UTC of 3 December 2007, forecast by the GEM regional model.Fig. 5b is the corresponding analysis chart.

10FIG. 8. Same as Fig. 7, except valid at 1200 UTC of 2 December 2007. Fig. 5a is the correspondinganalysis chart.FIG. 9. Analysis (left panel) and forecast (right panel) surface winds over the BC South Coast at 1200UTC of 2 December 2007. The thin lines are elevation contours. Fig. 8c shows the correspondingforecast mean sea level pressure.

115. Forecasts by PSPCThrough the period of 1 – 5 December 2007,PSPC meteorologists issued a total of 79 severeweather warnings across BC. The criticalsuccess index for all warning events for thisstorm cycle is 0.68 (see Table 2), which can beconsidered as a very successful skill score.Specifically, all the wind chill events over the BCPeace River region and the North and CentralCoast were successfully forecast. PSPCachieved a POD (probability of detection) of 0.77for the snowfall warning events across theprovince, as compared with the model POD of0.63. Out of the 14 freezing rain events, PSPCsuccessfully forecast 7, and the model picked uponly one of them. The PSPC forecast alsoachieved a higher score than the model forecastfor the heavy rainfall events, with a POD of 0.8as compared to the model POD of 0.6.6. Concluding remarksA strong arctic outbreak at the end ofNovember 2007 and its collision with a warm,moist subtropical southwesterly flow produced alarge number of high impact weather eventsacross BC during the period of 1 – 5 December.Given the complex terrain of the province, thisgreatstormposedsignificantforecastchallenges for both numerical models andoperational meteorologists.Generally speaking, the Canadian GEMmodels provided reasonable forecast guidancefor this storm. In particular, the arrival of thewarm Pacific moisture conveyor (PineappleExpress) with heavy precipitation acrosssouthern BC on 3 December was forecast by theGEM global model with a 6-day lead time.However, the models have some difficultydealing with the complex terrain of BC. It isnoticed that the outflow-induced lee low alongcoastal BC on 2 December (Fig. 5a) was overforecast by the GEM regional model (Fig. 8).This deeper low in the model produced strongerwarm advection across southern BC. As aresult, the model provided misleading guidancecalling for snow changing to rain for somecoastal areas on 2 December. The reason forthis model error is not clear. It could be relatedto the occasional over-forecast of the arctic ridgeover northern Canada by the GEM models.It was further pointed out that the modelforecast for BC has always been too warm insituations where there is warm advection. Webelieve the root of the problem is the model’shandling of the topography. The actualmountains in BC are quite steep with deepnarrow valleys that cannot be resolved at 15 kmof the GEM regional model resolution. Themodel topography, therefore, is smoother thanthe real one, leading to the cold air at low levelsbeing scoured out too quickly in the model. Coldair in the real world has more staying power thanthe model indicates. This has severalconsequences: (1) the inland sea levelpressures fall too quickly in the model becausecold air in the valleys is not resolved, leading toweaker northeasterly outflow winds or strongersouthwesterly inflow winds; (2) warmer surfacetemperatures are forecast; (3) snow changes torain too quickly, leaving little room for freezingrain to develop; (4) snow is under-forecast andrain is over-forecast in the valleys.TABLE 2. Weather warning verifications for the period of 1-5 December 2007.Number of HitsNumber MissedNumber of False AlarmsNumber UnverifiableTotalEvents5822516101Probability of DetectionFalse Alarm RatioFrequency of HitsBiasCritical Success IndexScores0.730.080.920.790.68

12It was noticed that precipitation was underforecast on the leeward sides and over-forecaston the windward sides of the mountains by theGEM regional model. A possible explanation forsuch precipitation bias is that the smoothertopography in the model encourages more air toflow over as opposed to around the mountains.On the windward side, stronger upward airmotion and weaker low-level horizontaldivergence lead to heavier precipitation, and theopposite applies to the leeward side. Anotherpossibility is that the model underestimates the‘spillover’ of precipitation to the lee sides of themountains. Browning et al. (1975) pointed outthat the spillover precipitation depends mainlyon the contribution of mid-level convection. Thisis because orographically induced enhancementmid-level convection gives rise to precipitation offairly high intensity which, owing to its high levelof origin and slow fall-speed above the meltinglevel, is carried over the mountain tops by thestrong winds. The smoother topography in themodel results in weaker mid-level pitation.Acknowledgments.We would like to thank all PSPC meteorologistswho were involved in forecasting the severewinter events through the period of 1 – 5December 2007. Special thanks go to FordDoherty and Don Tatar for their assistance withdata collection and weather warning verification.Michel Gélinas, Barry Brisebois and Ian Okabeprovided valuable discussions.ReferencesBrowning, K. A., C. W. Pardoe, and F. F. Hill,1975: The nature of orographic rain atwintertime cold fronts. Quart. J. Roy. Meteor.Soc., 101, 333-352.Hacker, J. P., E. S. Krayenhoff, and R. B. Stull,2003: Ensemble experiments on numericalweather prediction error and uncertainty for aNorth Pacific forecast failure. Wea. Forecasting,18, 12-31.Heidorn, K.C., 2004: The BC Weather Book:From the Sunshine Coast to Storm Mountain,Fifth House Ltd, 300 pp.Jackson, P. L. and D. G. Steyn, 1994: Gapwinds in a fjord. Part I: Observations andnumerical simulation. Mon. Wea. Rev., 122,2645-2665.Johnson, K. and J. Mullock, 1996: AviationWeather Hazards of British Columbia and theYukon. Environment Canada, 136 pp.Lackmann, G.M. and J.R. Gyakum, 1999: Heavycold-season precipitation in the northwesternUnited States: Synoptic climatology and ananalysis of the flood of 17–18 January 1986.Wea. Forecasting, 14, 687–700.Loukas, A. and M. C. Quick, 1996: Spatial andtemporal distribution of storm precipitation insouthwestern British Columbia. J. Hydrol., 174,37-56.Lyons, C. J., 1901: The storms of the HawaiianIslands. Mon. Wea. Rev., 29, 68-68.Mass, C., 2008: The Weather of the PacificNorthwest. University of Washington Press, 180pp.

2.4 C. Snow in Metro Vancouver did not change to rain until Sunday evening when stronger warm advection from the PE system arrived. Our verification reveals that the GEM regional model failed to forecast all but one of the 14 freezing rain events in BC during 1 - 5

Related Documents:

5 1. Collision Physics – An Overview. 1.1 Outline. Collision physics includes ANY collision of a quantum particle with a target. Collision Particles may be: PHOTONS (eg from a Laser, Synchrotron source or FEL) ELECTRONS (usually of well defined momentum from an electron gun) IONS (usually from an ion source of well defin

The Hewlett-Packard 10B-II (HP 10B-II), or Hewlett-Packard 10-BII (HP 10B-II Plus) calculator is strongly recommended. All examples shown on lecture notes will use this calculator. If you choose to use another, it is your re

instructions for completing California Highway Patrol (CHP) Traffic Collision Report forms (CHP 555, Traffic Collision Report, CHP 555D, Truck/Bus Collision Supplemental Report, CHP 556, Narrative/Supplemental, and CHP 555-03, Traffic Collision Report - Property Damage Only), and is available to all law enforcement agencies. An additional

Tyson Anaka Transtar Autobody Technologies Bradford ON CA Supplier/Wholesaler Enzo Anania CSN - Hwy 27 Auto Collision Vaughan ON CA Collision Repairer Maurice Anderson Canavans Central Appraisal Dartmouth NS CA Insurer Shellie Andrews CSN - DANA'S Collision Centre Fredericton NB CA Collision Repairer

California via TIMS, such as the distribution of type of collision in the past three years. 3. System Design Figure 1 shows the end-to-end pipeline of the whole collision analysis system based on DMV AV collision reports and the TIMS datase

If opened, the collision diagram file will like look similar to one of the following: If created using “Save Crash ID” list (preferred method for State network users) OR If created using “Save Collision Diagram Data File” or the “Generate Collision Diagram File” method Step 1

Toyota Collision Repair & Refinish Training Toyota Certified Collision Centers Toyota Genuine Parts Toyota Technical Education Network Toyota Wholesale Parts & Certified Collision Department REFER ALL CORRESPONDENCE TO: Collision Pros Toyota Motor North America, Inc. 6565 He

Each TR-31 0 consists of an Original Collision Report and three Financial Responsibility forms. The Original Electronic report is submitted through the South Carolina Collision and Ticket Tracking System (SCCATTS) to the Office of Highway Safety (OHS). The existing collision reports (paper) are submitted to the Office of Financial Responsibility.

Answers may vary. Sample answer: Elastic collision: No, it is not possible for two moving masses to undergo an elastic head-on collision and both be at rest immediately after the collision.

collision avoidance and compliance with the rules of COL-REGS, using velocity and line-of-sight vectors to express the COLREGS rules. The constraints are implemented as penalties in order to ensure that the best possible control behavior can be chosen also when collision with at least one obstacle seems unavoidable. II. SYSTEM OVERVIEW

In the Avensis, you’re better protected than ever. Toyota Safety Sense adds advanced active safety systems for an even safer and more relaxed drive. 3. Pre-Collision System* Toyota Safety Sense’s Pre-Collision System uses a camera and laser to detect other vehicles in the road ahead. When the possibility of collision is detected

these classes of crashes may be prevented or their effects lessened by using a forward collision warning system (Dogan Ibrahim, 2006). From this we can infer that rear end collision is the

Montclair Auto Collision Center is a centrally located complete auto body and paint shop dedicated to never send-ing out a repair that is not 100% to our high standards. We are committed to making our client’s collision repair experience as positive as possible, and we survey e

How to Resolve File Sync Warnings. 114. 114. Common Types of File Sync Warnings. 115. Resolving a Collision. 115. Resolving a Locked Change Warning. 118. Other Types of File Sync Warnings. 120. File Collision. 120. Folder Exists Upstream \(Folder collision\) 120. Invalid Credentials for Syn

3/10/18 Herkimer Collision w/Tree 3/17/18 Wyoming Collision w/Vehicle (Pickup) 3/19/18 Jefferson Collision w/Tree *Information being collected Historically, alcohol is a factor in 33% of fatal accidents and 10% of all accidents. Three accidents occurred ON the state funded trail system, four occurred OFF funded trails.

the time of the collision, the angle of blow by or to the other vessel, the speed of each vessel at the time of the collision, any alteration of speed and course prior to the collision Update if necessary vessel's position in radio room, satellite terminal and other automatic distress transmitters (GMDSS)

Fitchburg, MA 20-21 Lusid Technologies Roland’s Auto Body White Plains, NY 22-23 CCC Information Services Big Sky Collision Billings, MT 14-15 Celette Bill Ussery Motors Collision Center Miami, FL 16-17 Matrix Wand Carrillo & Sons Collision Center San Diego, CA 24-25 Sherwin-Williams

Corporation) train P91, operating on a track warrant, was diverted from the main track through a . Amtrak Silver Star route from New York to Miami showing collision location. . 5 Figure 4. . New York Air Brake . POC. point of collision . PSAP. Public Safety Answ

Dedicated Quantum Collision Attacks. Recently, it has been shown that collision attacks on hash functions with quantum machines can break more rounds than the attacks with classical machines [17]. Whether a hash function is attacked or not is judged by comparing the complexity of the

236 Study Guide for An Introduction to Chemistry Section Goals and Introductions Section 14.1 Collision Theory: A Model for the Reaction Process Goals To describe a model, called collision theory, that helps us to visualize the process of many chemical reactions. To use collision theory to explain why not