Current Case Display machine:SREF
If cases do not appear click here for alternate server
For Previous years and events:
|Years of data |
Years of data
Years of data
Years of data
Years of data
Years of data
Years of data
Years of data
Years of data
Years of data
Years of data
Years of data
Years of data
Years of data
Years of data
Years of data
Years of data
Years of data
Years of data
Years of data
Years of data
Years of data
Years of data
Years of data
Years of data
Years of data
Years of data
|Years of data |
Years of data
Years of data
Years of data
Years of data
Years of data
Years of data
Years of data
Years of data
eyewall.met.psu.edu web 1
A strong trough and surface anticyclone brought an early season cold episode to much of the United States east of the Rocky Mountains. The deep trough and deep cold air were accompanied by a strong surface anticyclone. The strong anticyclone produced mean sea-level pressure values near 1050 hPa and based on the CFSR, implied many regions of the northern plains likely experienced both sea-level pressure values and pressure anomalies which occur about once every 30 years.
The pressure data from the CFSR (Table 2) and the plots of the PDF and CDF imply that the anticyclone which tracked across eastern ND and MN was a record event for the 21-day window centered on 24 November. The NAEFS forecast data indicate that the NAEFS was able to predict a record event in close proximity the region where such a record event occurred. In addition to the massive anticyclone and record surface pressure, the NAEFS also correctly predicting the intrusion of cold to near record cold air over the region. In the core of the cold air mass, many locations saw a once in 30 year event.
The data for Richmond showed that Richmond and likely other parts of the Mid-Atlantic region had the lowest 850 hPa temperatures on record in the past 30 years. The -16.85C reading was the coldest record in this window and was associated with the most negative standardized anomaly. Though not shown, shorter range NAEFS 850 hPa temperatures forecast the -16C contour to reach the Richmond area. The more general forecasts of -12 to -14C at 48 to 72 hours in advance present successful forecasts of a cold episode
This paper will document the strong severe weather event of 17-18 November 2013. The focus on the impacts is related to the traditional use of standardized anomalies and some of the limitations in using these data to identify extremely anomalous stations. The concept of using return periods and percentiles to compliment standardized anomalies is introduces as a more advanced means to assess the potential of a High Impact Weather Events (HIWE). With over 580 reports of severe weather and over 400 events in the Midwest, this was by far one of the largest November severe events in recent history.
A relatively well predicted strong cyclone developed over the Great Lakes on 31 October and into 1 November 2013. The pattern as analyzed (Figs. 1-4), showed a strong cyclone develop over the Great Lakes with a surge of strong southerly winds and a surge of deep moisture ahead of a progressive 500 hPa trough. The strong southerly flow head of the implied frontal zone in the PW field is a relatively well-known pattern associated with convection typically along or just ahead of the front. Not surprising the system developed several lines of convection and produced severe weather where convection brought the strong winds to the surface (Fig. 5).
Forecasts from the NAEFS showed that the NAEFS clearly predicted the strong southerly winds ahead of the front and the deep cyclone over the Great Lakes (Figs. 7-13). The return period on the wind anomalies indicated an extreme event in the tails of the 30-year climatology (Fig. 10) and the anomalies on the 850 hPa pressure surface were the largest anomalies relative to the other levels where wind anomalies were forecast by the NAEFS.
The resulting convection over Pennsylvania showed a textbook example of an NCFRB which zipped across the State producing gusty winds and reports of down trees along its path. Due to the speed of movement, the line of convection and the trailing WCFRB did not produce heavy rain and heavy rain was not considered a threat by forecasters.
This case shows the value in assessing the potential for a high impact weather event using standardized anomalies. But the key to using them effectively is the enhanced knowledge of the character of the standardized anomaly relative to the return period and where along the PDF the event lies. The higher end synoptically driven events are often observed when key parameters are in the tails of the PDF. The system produced 190 reports of severe weather from Ohio, across West Virginia, Pennsylvania, New York and New Jersey. The reports for central Pennsylvania are listed in Table 1.
Though not shown here, the NAEFS anomalies were analyzed in real-time using a simple algorithm based on the wind anomalies, PW anomalies and the temperature anomalies. The algorithm correctly forecast a Synoptic type event, dominated by the extreme values of the low-level 850 hPa V-wind anomalies. This was not a synoptic heavy rainfall event in the East (Fig. 18) likely due to the speed of movement of the system. Though not shown, model QPF values also did not indicated significant rainfall threat. Most of the rain in Pennsylvania fell in a short period between 01/0600 and 01/1200 UTC.
Keywords:anomalies high impact weather HIWE.
Mid-Atlantic Heavy Rainfall event of 11 October 2013: The remnant circulation and moisture associated with tropical storm Karen and a slow moving frontal system produced heavy rainfall in the Mid-Atlantic region. The heaviest rain was observed over southeastern Pennsylvania where many locations received between 5 and 10.5 inches of rainfall . The rain fell on the cool side a low-level boundary with strong easterly flow over the boundary, a classic larger scale “frontal type” heavy precipitation pattern (Stuart and Grumm 2006; Grumm 2011b) in the eastern United States.
The heavy rain produced flooding at 5 points along secondary rivers and streams (Table 1). Likely due to relatively dry conditions prior to the rain, minor flooding was observed at 4 of the 5 points and only 1 point experienced moderate flooding. The event of 11 October was relatively unique as the rain fell over about 36 hours and there were 5 6-hour period of rain over 25 to 50mm. Most heavy rain events typically span 4-12 hours, this was an extraordinarily long event and 4 of the 6 hour periods ending at 11/0000 UTC through 12/0000 UTC are shown in Figure 13. The 4 times shown only span 11/0000 through 11/1800 UTC. Each image shows a broad region of 25mm of QPE and each image has at least a small region of over 50mm of QPE. The long period of rainfall likely limited the flash flood aspects of the event.
All analyses presented here imply that the heavy rainfall was associated with a strong 850 hPa LLJ with significant u-wind anomalies. The models and SREF had some difficulty predicting the exact location and intensity of this feature and the associated surge of high PW. This may have contributed to the displacement issues associated with QPF. However, the overall pattern was relatively well predicted by all forecasts systems, which alerted human forecasters to the potential for heavy rainfall in the Mid-Atlantic region. The details as to where the heavy rain would fall were more difficult to ferret out.
A slow moving frontal system and a plume of deep moisture brought heavy rain to the Ohio Valley on 5-6 October 2013 (Fig. 1). The City of Louisville, KY had a record 24-hour and storm total rainfall during the period of heavy rain. The total rainfall on 5 October was 5.91 inches which broke the previous record of 3.07 inches set in 1910 . This also broke the monthly 24-hour record of 5.07 set in 2004. The heavy rainfall produced significant flooding in and around the City of Louisville resulting in numerous water rescues. The rain near Louisville peaked after 0000 UTC 6 October 2013 with the heaviest rainfall falling in the 6-hour period ending at 1200 UTC. The gridded rainfall plots imply that the heaviest rain occurred west of the City of Louisville.
The heavy rain fell in a pattern often associated with heavy rain and the pattern was relatively well predicted. Despite the relatively predictable pattern, the NCEP SREF under estimated the total QPF.
Historic Northern Colorado rainfall and floods of 12-13 September 2013
Historic flooding affected the front range of the Rocky Mountains of Colorado on 12-13 September 2013. The flooding was the result of heavy rainfall over the region. The heaviest rainfall amounts, perhaps in excess of 15 inches fell in the mountains to the west. Water drained down the streams in the canyons to include Big Thompson River and Boulder Creek. The flooding along Boulder Creek causes flooding and damage to the city of Boulder, Colorado. The flooding along the Big Thompson River caused flooding caused more damage than the damage reported during the Big Thompson Canyon flood of July 1976.
The heavy rainfall fell in a well-documented pattern associated with heavy rainfall. Pontrelli et al (1999) examined the pattern associated with the Big Thompson, Rapid City, Fort Collins, and Madison County flood. The pattern associated with floods of 2013 September 2013 was similar to the pattern associated with the previously mentioned historic flood events. This included the quasi-stationary east-west frontal boundary with strong low-level easterly flow into the terrain feature. This produced a prolonged period of heavy rain with inflow of +2 to +4 above normal PW air.
The NCEP models and ensemble forecast system correctly predicted the large scale pattern. The standardized anomalies of key fields, such as the strong low-level easterly jet and the above normal precipitable water on the warm side of the frontal boundary were all well predicted by the NCEP forecasts systems. Not surprisingly, these systems produced QPF amounts in excess of 75 to 150 mm. The location of the higher QPF amounts varied by forecast system. The probability of high QPF amounts combined the forecast of the pattern when used together may have provided clues to a potentially high end flood event.
Keywords: Flood, Boulder, Ensembles, standardized anomalies.
Heavy rain event of 28 August 2013
A large 500 hPa ridge over the central United States (Fig. 1) with a closed 5940 m contour placed Ohio and Pennsylvania in the strong northwesterly flow at 250 hPa over the ridge (Fig. 2). An embedded short-wave in the flow combined with deep low-level moisture (Fig. 3) led to the development of a mesoscale convective complex (MCC:Maddox 1980) over Michigan and Ohio and a smaller series of mesoscale convective systems (MCS) over western Pennsylvania. The precipitable water (PW) values were over 50mm in Michigan which was about 3 above normal for the time of year. Over western Pennsylvania the PW values peaked between 45 and 50mm after 0000 UTC 28 August 2013 (Fig. 3d-f).
The larger scale NCEP NAM showed the strong flow at 850 hPa (Fig. 4) with a westerly low-level wind maximum moving into Indiana, southern Michigan, and northwestern Ohio around 0000 UTC 28 August (Fig. 4d), the jet peaked at over +3 above normal around 0600 UTC (Fig. 4e) as it moved to the southeast. This larger feature was associated with the main MCC (Fig. 5 upper) which produced the heavy rains in Michigan and Ohio (Fig. 5). A smaller scale MCS (Fig. 5 lower) produced the locally heavy rainfall over Pennsylvania (Fig. 6 inset). It will be shown that the heavy rainfall in southwestern Pennsylvania fell between about 0600 and 0800 UTC.
Most of the ingredients for locally heavy rainfall and flooding were in place (Doswell et al. 1996). Additionally, the flow about the strong ridge and strong low-level jet were indicative of conditions associated with MCS (Maddox 1983).
This paper will document the event of 28 August 2013 to include the larger scale pattern conducive for the heavy rainfall the key components of the heavy rain and flooding in southwestern Pennsylvania.
A strong tropospheric ridge to the east and a trough over the eastern Pacific produced a strong 250 hPa jet and deep south-southwesterly flow over the southwest deserts of the United States. This strong flow brought a plume of deep moisture and above normal PW air into southeastern California, southern Nevada, and southwestern Arizona resulting in locally heavy rainfall. Rainfall amounts west of Las Vegas exceed the estimated return periods for a 1000 year event (Table 2) and similar number likely were achieved in the higher terrain of the Mojave Desert.
The synoptic pattern was favorable for locally heavy rainfall where the plume of deep moisture intersected with terrain features. Most of the heavy rainfall occurred along and east of several prominent terrain features in the southwestern desert. Comparing heavy rainfall to terrain indicated that the Spring Mountains, west of Las Vegas (Dark Oval), the Dead Mountain Wilderness area west of Needles and Bullhead City (Yellow oval), the Providence Mountains (red oval); where Mitchell Caverns is located; were regions of focused heavy rainfall. The topography suggest that some of the heavier rainfall likely fell east of the topographic features implying heavy rains on the eastern slopes and near the valley floors close to the terrain features. In the Providence Mountains many peaks and ridge are over 5000 and in places 6000 feet above sea level. Ridges in the dead mountain are over 3000 ft above sea-level and in the Spring Mountains west of Las Vegas, Mt Charleston (Charleston Peak) peaks out at 11,916 (3632m) above sea-level. The local radar indicated over 12 inches of rainfall in the higher terrain. These numbers were not the same as those in the NMQ Q2 or Stage-IV data examined for the event.
The heavy rainfall produced significant flooding in valley locations. News accounts, social media posts, and videos of flooding and damage to roads due to flooding were abundant especially in western Las Vegas toward the Spring Mountains.
A surge of high PW air and instability aligned with a strong low-level jet produced locally heavy rain and severe weather across northern Maryland, southeastern Pennsylvania, Delaware and New Jersey a few hours either side of 1200 UTC 13 August 2013. Most of the severe weather was in the form of wind damage with at least 1 EF0 tornado in New Jersey. As shown in Figure 1, the severe weather was confined to a relatively small region of the Mid-Atlantic region and much of it was associated with a few isolated thunderstorms.
The analyzed data suggest that monitoring high PW plumes and surges of strong low-level winds may be of value in short-term forecasts of regions of heavy rainfall. The addition of the high CAPE added the threat of strong cores capable of locally heavy rainfall and severe weather.
The forecasts from the SREF were shown here. The 6 most recent forecasts prior to the onset of the rain were presented. These data indicated increased skill as the forecast length decreased, an implied convective response for the locally heavy rain (Fig. 11), the random areas where convection and thus 50 mm or more was produced, and the stronger mesoscale signal in the SREF as forecast length decreased. The clear signal here is that the SREF forecasts can and do improve markedly as forecast length decreases and it can produce useful guidance as to where to monitor the potential for heavy rainfall as forecast length decreases. Though not shown, these data imply convection allowing models and rapid updates of these models need to be monitored to further improve forecasts of mesoscale heavy rainfall events. The advent of convective ensembles will likely improve forecast of mesoscale rainfall events.
A fast moving short-wave (Fig. 1) with -1s 500 hPa height anomalies and a modest 250 hPa jet (Fig. 2) produced a severe weather event over eastern Ohio and western Pennsylvania. It will be shown that a bow echo developed in northwestern Pennsylvania and moved southeastward across the State and produced the majority of the severe wind reports. To the southwest, supercells produced a mix of wind and hail reports.
These distinct systems and their ability to persist, producing significant number of severe reports and semi-linear features in SPC storm reports (Grumm and Colbert 2013). The one short-lived EF0 tornado was observed in Quemahoning, Jenner and Lincoln Townships in Somerset County, Pennsylvania, south of the bow echo track.
Convective Heavy rainfall event of 28 July 2013:A very mesoscale and convectively driven heavy rainfall event produced a record event at the Philadelphia International airport on 28 July 2013. The event total was 8.26 inches of rainfall on 28-29 July 2013 with 8.02 inches of rain observed on Sunday 28 July 2013 breaking the old record for the date of 3.28 inches set in 1969. The 8.02 inches also established a new record rainfall for the site eclipsing the record of 6.63 inches set on 16 September 1999 during Hurricane Floyd.
This rain fell in a plume of moisture with relatively high CAPE and modestly strong southerly flow ahead of a slow moving cold front. There was an unseasonably strong 500 hPa trough to the west associated with the frontal system. The along front flow indicated that the front would be slow moving and thus the potential existed for some locally heavy rainfall. Despite the potential for training and the relatively high CAPE, exceeding 1200JKg-1 at times on 28 July, the numerical guidance indicated a relatively modest rainfall event. Additionally, the NCEP SREF showed that rainfall in excess of 25mm, let along locally over 200 mm, was a very low probability event.
The inability of the larger scale guidance to predict this event is not too surprising as the locally heavy rain was observed on a mesoscale. An examination of the RUC and 4km NAM implied (not shown) high resolution models offered little assistance with the extremely heavy rainfall over southeastern Pennsylvania. We may be decades off before models can simulate events of this nature with lead-times in excess of a few hours. Thus, the NCEP NAM, GFS and SREF got the overall pattern and areas of rainfall relatively well, but they were not able to or designed to capture the extreme rainfall observed on 28 July 2013.
The key to events like of this nature is knowing the pattern and the potential and when the convection evolves, monitor the radar and observations to provide as much information regarding the locally extreme events as possible. The radar and spotters are still one of the best short-range forecast tools available for local scale forecasting.
A low-topped shower produced a low-level mesocyclone (Fig. 1) which produced a narrow 3-mile long tornado in Potter County around 450 PM (2044 UTC) local time on 27 July 2013. Lightning data indicated little lightning with the storms in northern Pennsylvania on 27 July with only one observed strike in Potter County during the event. This tornadic storm was one of two low-topped rotating storms to produced winds in the lower volume scan in excess of 45kts (Fig. 2). The later storm, despite the mini-supercell appearance and hook echo with rotation produced no reports of severe weather.
The EF1 tornado near Coudersport, PA was one of 2 EF1 tornadoes in the eastern United States, a second tornado developed from the same storm which produced the tornado near Coudersport as it moved out of Pennsylvania, producing an EF1 tornado near Troupsburg, NY (Fig. 3). In total there were 3 reports of severe weather in NY and PA 2 of which were tornadoes with low-topped convection and both tornadoes came out of the same long-lived low-topped supercell storm.
Convective Heavy rainfall event of 23 July 2013
A record rain event affected southern Lebanon County on 23 July 2013. Most of the rain fell in a 3-hour period and the estimated 3-hour rates exceed the value for a 1000 year return period. The 6-hourly amounts were slightly lower than those for a 1000 year return period. The heavy rain produced flooding in the affected area.
The pattern over much of the Mid-Atlantic region contained all the ingredients for a heavy rainfall event including a trough to the west, above normal precipitable water, and strong low-level southerly flow. Early in the event the atmosphere was unstable and late afternoon CAPE values were above 1200JKg-1supporting deep moist convection. Once the thunderstorms developed, the surface based CAPE rapidly decreased.
The numerical guidance indicated the potential areas of heavy rainfall. The focus of the locally heavy rainfall was related to model terrain features and locations where model convective parameterization schemes worked to reduce instability in the model atmospheres. Thus some of the guidance was of value in anticipating the potential for heavy rain; the models were unable to focus on the correct regions. The rapidly updated RAP guidance produced short-range forecasts which provided better clues in both the mass and precipitation fields.
Mesoscale Convective System of 10 July 2013:
An MCS developed over eastern Illinois on 10 July racing across Indiana and Ohio (Fig. 2) producing hundreds of reports of severe weather along its path from Illinois to west-central Pennsylvania (Fig. 1). The MCS died in west-central Pennsylvania close to the eastern limb of the MCS climatology produced by Bentley and Mote (1998). Thus, the severe weather in the State College warning area was limited to a few of the westernmost counties.
The MCS developed in a plume of deep moisture (Fig. 5) as a strong mid-summer 500 hPa trough was moving into a strong subtropical ridge to the east (Fig. 4). The high PW air (Fig. 5) and cooling aloft likely contributed to the high CAPE, over 3000JKg-1 in the region where the MCS originated (Fig. 6). This MCS classified as a derecho as it produced continuous swath wind damage or winds over 50kts for well over 400 km (250 miles: Johns and Hirt 1987). As shown in Figures 9-11, the bow echo in the north south line had winds of 60kts and farther north a cyclonic bookend vortex or mesocyclone was present. The damage and structure show this MCS classified as a derecho
A short-wave moving out of West Virginia triggered the development of convection in West Virginia and western Maryland in the early afternoon hours of 7 July 2013. The convection developed west of a weak frontal and dry-line like boundary (Figs. 2 & 3) where the CAPE ranged from 1200 to over 3000JKg-1 in the warm moist air. The winds at 850 hPa were relatively weak from the south-southwest and the flow from the surface to 250 hPa was from the southwest, a very linear shear profile.
The satellite data implied that at least 2 MCS’s (Maddox 1980) developed, the northern one (Fig. 6) was associated with the long-lived and persistent QLCS which had a strong bow-echo within it (Figs. 7-10). These systems where persistent, not comprised of rotating supercells, produced severe weather over several hours and a long track frequently occur (Grumm and Colbert 2012) in the warm season. They produce a distinct pattern in the Storm Prediction Storm (SPC) data. How to identify in advance when persistent long-lived QLCS systems with embedded intense bow echoes will develop would be value in the forecast process and may be an area where storm scale ensembles may be of value in the near future.
Understanding the model simulations and studies (Weisman 1992; Weisman and Rotunno 2004; Weisman et al 2008) showing how these systems develop and the maintain themselves is important when predicting these systems. An examination of RAP 13km Skew-T data in the affected region revealed a linear shear profile with general south-southwest flow from the surface to over 250 hPa (Fig. 10), a linear hodograph as indicated by prior cited research. The thermodynamic profile also showed a weak dry area and an “inverted-V” at low-levels in the sounding. The profiles at sites examined showed this from southwest Pennsylvania into northeastern Pennsylvania along with CAPE values in the 1800 to 2400JKG-1 range. Unfortunately downdraft CAPE (DCAPE) data was not archived.
This case demonstrates the value of base Doppler products, particularly base winds in following and warning on distinct bow-echoes. Though examined, dual-pole products added little value to diagnosing the evolution of the system. At best CC and HCA data showed that there was some small hail mixed with rain which may have contributed to the cold-pool production and maintenance. The ability to produce cold pools is a key ingredient to predicting these long lived convective systems.
2 July 2013 Flash Flood EventA retrograding 500 hPa cyclone and anticyclone (Fig. 1) set up deep southerly flow over the eastern United States from 1 to 5 July 2013. The deep easterly flow brought a plume of deep moisture (Fig. 2) with 2-3 above normal precipitable water (PW) values into the eastern United States. The result was a series of severe weather events and locally heavy rainfall events over the eastern United States. This paper will focus on the heavy rain and flash flood event over New England on 2-3 July 2013.
The impressive features at 500 hPa at this time included the close 5940 m high over the western United States, the deep 500 hPa low over the southern United States, and the 5940 m anticyclone over the western Atlantic. Each of these feature contributed to significant weather to include the western United States heat wave and fires; the cool period over the southern United States from North Carolina to Oklahoma, and the warm and wet period over the eastern United States.
The pattern, with the deep southerly flow between the anomalous trough to the west and the anomalous anticyclone to the east is a well-known and documented heavy rainfall event pattern (Junker et al. 2008; Bodner et al 2011). The key to the pattern is the amplified flow and the anomalies aid in identifying systems which have higher potential to produce higher end events (Junker et al 2009).
This paper will document the heavy rainfall event of 2-3 July 2013. The focus is on the pattern and key features which produced the favorable synoptic scale environment. This case is one of many which shows the power of standardized anomalies and subtropical anticyclones in producing high impact weather.
Early Summer Heat Waves of 2013
From late June into mid-July large ridges brought periods of hot weather to the western and eastern United States. At least twice during the peak of the ridges the 500 hPa heights peaked at or over 6000m. During the third and final phase of the heat waves, the NCEP GEFS predicted a closed 6000m contour over Midwest at 1800 UTC 16 July 2013. Meteorologically, 500 hPa heights in excess of 6000 m are rare events. Despite the successful forecasts of 6000 m heights, the heat events of July in the eastern United States did not set many high temperature records.
The ridges and warm episodes were all meteorologically and significant events. In the West, the late June ridge and dry air beneath it created ideal conditions for fires. Several fires and one deadly fire made National News as the heat wave in the west peaked. From a forecast perspective, the western ridge and high fire weather threat was well predicted by the larger scale models and the NCEP GEFS.
As the ridge become pronounced in the East, it brought warm and relatively wet weather. Both surges of high heights in the East lacked deep low-level warm air. Thus, in the east as the ridge retrograded a plume of deep moisture brought on the western flanks of the system brought heavy rainfall to regions from the East Coast westward to the eastern plains. During the event, the 500 hPa heights peaked at around 6000m. Despite the large ridge in the mid-troposphere, the low-level temperature anomalies were generally on the order of +1s above normal. Thus, the event did not break a significant number of high temperature records.
All three phases of the event showed how the use of R-Climate data in the forecast process can be of value. They also how these data, when a high impact event is being forecast, provide some confidence information when applied to ensemble forecasts systems. Large anomalies in ensemble forecast require that the system have good agreement on the intensity and location of the event and that these forecast are of a potential significant weather event.
Flooding and Severe weather of 27 June 2013:
Flooding and severe weather affected Pennsylvania on 27 June 2013. The severe weather included at least 2 confirmed tornadoes one of which occurred in close proximity of the National Weather Service Office in State College. The event occurred on the northeastern edge of a strong subtropical ridge which was centered over the southwestern United States. The 500 hPa heights implied a strong mid- and upper-tropospheric jet stream moving over the ridge which produced strong low-level southerly flow allowing a surge of deep moisture into the Mid-Atlantic region. The 850 hPa winds indicate that the 850 hPa cyclone centers moved across Pennsylvania between 1200 UTC 27 June through about 0600 UTC 28 June 2013. It will be shown that the severe weather and tornadoes were observed in close proximity to the deep lower tropospheric cyclone.
Alaskan heat episode of 16-19 June 2013: A strong 500 hPa ridge developed over Alaska from 15-20 June 2013. The 500 hPa heights peaked around +4s above normal 17-18 June 2013. Despite these relative values, data from the climate forecast system indicated that the 500 hPa heights were higher than 500 hPa heights observed in the climate forecast system since 1979. A similar signature was observed in the 700 hPa temperature field. The forecast and observed 700 hPa temperatures were at the extreme tail of the probability distribution function relative to the climate forecast.
The resulting large ridge with above normal heights and temperatures lead to a multi-day heat episode over Alaska. Temperatures at several climate sites exceed 90F on at least 1 day and several locations had high temperatures in excess of 80F for three consecutive days, a true Alaskan heat wave.
The results shown here suggest that for extreme events and heat episodes that the traditional standardized anomaly provides insights into extreme events. However, the skewed distribution of the height and temperature fields reveals that where a forecast lies relative to the probability distribution can provide a clearer signal when extreme weather events approach historic levels.
Lancaster Severe Thunderstorm 17 June 2013: A cluster of showers and thunderstorms developed shortly before 1900 UTC in the Harrisburg area of southeastern Pennsylvania. From this cluster one large and relatively long-lived storm traversed Lancaster County from northwest to southeast. Along its path is produced hail stones reported to be 0.70 to 1.00 inches in diameter. The storm likely produced several wet microbursts along its path which led to reports of sporadic areas of downed trees, and wires. Strong winds from a wet microburst blew down a tobacco barn in Paradise, PA.
This storm developed in a modest low-level equivalent potential ridge, on the warm side of a weak boundary The CAPE supported large updrafts as forecast and analyzed CAPE was in the 1400 to 2200JKg-1 range. The -20 level was near 25kft and this storm produced 50 dBZ cores to around 25kft as it traversed Lancaster County. Few 50 dBZ cores reached much above 25kft. Most of the hail was associated with TBSS signature on KCCX radar.
The large scale pattern over the region (Fig. 6) was relatively quiet with weak winds, weak vertical shear and lacking a source of deep moisture. The analyzed CAPE in the 13km RAP showed some potential for CAPE values to reach as high as 2400 JKg-1 in southeastern Pennsylvania in close proximity to where the isolated storms developed.
Several images depicting the type of damage observed are included in Figures 8-10.
Eastern Derecho 12-13 June 2013 A strong mid-tropospheric short-wave coming over a sharp 500 hPa ridge produced a series of mesoscale convective complexes on 12-13 June 2013. One MCS classified as a derecho producing widespread wind damage from the western Great Lakes into southeastern Pennsylvania. The severe weather in Pennsylvania was observed mainly between 0600 and 1200 UTC 13 June 2013 with the long-lived derecho producing MCS.
This system included a deep surface cyclone for so late in the season and NCEP models produced significant amounts of QPF with the system. Overall, most forecast systems grossly over forecast the QPF relative to observed rainfall data. These over predictions of QPF lead to the issuance of flood watches and concerns about flooding that did not materialize due to the general lack of heavy rainfall. Probabilistic QPF threshold values from the NCEP SREF showed high end rainfall amounts were a low probability outcome.
This paper shows the pattern in which the derecho producing MCS developed. The pattern and thus the pattern of the precipitation was relatively well predicted, though as stated earlier the NCEP models grossly over predicted the rainfall.
Heavy Rainfall Event of 10-11 June 2013: A fast moving short-wave (Fig. 1) produced rainfall and areas of over 48mm of heavy rainfall from Maryland, across Pennsylvania and southern New England (Fig. 2). The heaviest rain, over 64mm, fell over portions of New Jersey and southern New York. A significant portion of the rainfall and the areas of heavier rainfall over Pennsylvania fell late in the event (Fig. 6) mainly between about 2300 UTC 10 June and 0300 UTC 11 June 2013. This led to some local flooding in central Pennsylvania where over a small area rainfall exceeded 75 mm (3 inches).
The pattern was a pattern associated with precipitation and most of the higher precipitation amounts fell in the region affected by the 1-2.5 above normal precipitable water plume (Fig. 2c). The PW field also showed a more north-south boundary from south-central Pennsylvania into the Carolinas which lined up well with the more north-south band of convection which affected northern Maryland and Pennsylvania between about 10/2200 and 11/0300 UTC. This sharper line with limited instability in Pennsylvania but 1200 to 2400 JKg-1 to the south produced severe weather and tornadoes from Delaware southward into the Carolinas (Fig. 16). The area affected by the severe weather was also experienced the passage of a strong low-level jet with +3 850 hPa v-wind anomalies (Fig. 7).
The rainfall, after the main area of rain moved to the east (not shown) over central Pennsylvania, locally exceed 75mm. The SREF, GFS, and NAM clearly showed that after 2100 UTC 10 June the area where this rain fell was not an area the model atmosphere’s expected heavy rainfall. Not a single SREF member had predicted much more and 25mm in the affected region. The shorter range RAP showed some clues for increased rainfall potential.
Memorial Day Weekend 2013: Snow and Cold
A slow moving 500 hPa low and associated unseasonably cold air in the lower troposphere brought unseasonably cold weather and a record late season fall to the higher elevations of northern New York and New England. Over 30 inches of snow fell on the slopes of Whiteface Mountain in northern New York. Many of the higher elevation locations, where ski resorts are often located had heavy snow. Trace amounts of snow were reported as far south as Binghamton, New York and Scranton, Pennsylvania.
The deep trough and cold air delayed the start of summer for a few days on the normally busy Memorial Day weekend. Despite the time of year, this storm shared many of the characteristics often associated with East Coast Winter Storms including a strong low-level northeasterly jet. The heaviest rainfall and heavy snow fell in and near the axis of this feature. The 850 hPa temperatures over the northeastern United States were below 0C during the precipitation and many locations were -2C, despite the deep cold air, accumulating snows were generally limited to elevations above about 2000ft.
The overall pattern and the potential for cold and rain was well predicted by the NCEP models and ensemble forecasts systems.
Mid-Atlantic Severe Event of 22 May 2013:strong low-level and a surge of high CAPE produced a severe weather event over Pennsylvania during the late afternoon and evening hours of 22 May 2013. Most of the reports of severe weather were due to strong and damaging winds. In central Pennsylvania a long-lived bow echo developed which accounted for a significant portion of the wind damage reports from Westmoreland County northeastward to Luzerne County, Pennsylvania.
This was the second event in May of 2013 where a strong bow echo developed out of a cluster of storms. The bow echo then went on to produce a significant number of the reported severe weather.
Multi-day severe event of 18-22 May 2013
A relatively slow moving Trough over the western United States and a ridge over the eastern
United States setup a relatively persistent pattern from 18-22 May 2013. This pattern produced a
period of enhanced severe weather over the United States from 18-22 May 2013. Relative to the
month of April 2013 this was an extremely active 4 day period which in fact produced more
severe weather reports than were reported during the entire month of April 2013.
The key features associated with the active severe weather of 18-22 May 2013 included a slow
moving Trough over the western United States and a ridge over the eastern United States. The
resulting enhanced southerly flow, the evolution of a strong LLJ, between these two systems
allowed warm moist air from the Gulf of Mexico to move into the central United States from
Texas to the Great Lakes. This led to increased values of CAPE in close proximity to strong
shear. The result was a multi-day period of enhanced severe weather with three successive days
of 300 or more reports of severe weather from 19-21 May 2013 (Table 1). These data show that
despite previous conditions, a rapid change in the pattern can rapidly produce significantly
A very dry air mass over the southwestern United States showed up as a region of below normal
precipitable water from 18-22 May 2013. This dry air, as it moved off the higher terrain of New
Mexico and west Texas produced an elevated mixed layer and a loaded gun sounding over
portions of Oklahoma including the period of the deadly New Castle and Moore, Oklahoma
The quick warm up of 14-15 May 2013 A retreating late season cold front and an advancing early season warm air mass led to rapid temperature changes in the Midwest on 14 May and in the Mid-Atlantic region on 15 May 2013. The elevated mixed layer of very warm air mixed down as the cool air mass retreated producing 283 near or tied record high temperatures in the afternoon hours of 14 May 2013 after a relatively chilly period for early May.
High temperatures in the Midwest reached well into the 80s and 90s on 14 May. Several dozen station reached high temperatures of 100F or greater to include locations in the State of Minnesota.
In the Mid-Atlantic region, the rapid erosion of the low-level cold air during the later morning and afternoon hours of 15 May 2013 led to many rapid temperature rises in the Mid-Atlantic region. Despite the rapid temperature rises, few high temperature records were set in the Mid-Atlantic region.
This meteorologically curious event was also relatively well predicted.
Multi-day severe event of 18-22 May 2013:A relatively slow moving Trough over the western United States and a ridge over the eastern United States setup a relatively persistent pattern from 18-22 May 2013. This pattern produced a period of enhanced severe weather over the United States from 18-22 May 2013. Relative to the month of April 2013 this was an extremely active 4 day period which in fact produced more severe weather reports than were reported during the entire month of April 2013.
The key features associated with the active severe weather of 18-22 May 2013 included a slow moving Trough over the western United States and a ridge over the eastern United States. The resulting enhanced southerly flow, the evolution of a strong LLJ, between these two systems allowed warm moist air from the Gulf of Mexico to move into the central United States from Texas to the Great Lakes. This led to increased values of CAPE in close proximity to strong shear. The result was a multi-day period of enhanced severe weather with three successive days of 300 or more reports of severe weather from 19-21 May 2013 (Table 1). These data show that despite previous conditions, a rapid change in the pattern can rapidly produce significantly different weather.
A very dry air mass over the southwestern United States showed up as a region of below normal precipitable water from 18-22 May 2013. This dry air, as it moved off the higher terrain of New Mexico and west Texas produced an elevated mixed layer and a loaded gun sounding over portions of Oklahoma including the period of the deadly New Castle and Moore, Oklahoma tornado.
Pennsylvania Severe Weather Event of 10 May 2013 The combination of instability and a frontal system brought severe weather to Pennsylvania and southwestern New York on 10 May 2013. Vertical profiles implied that there was a modest inverted-V sounding implying the potential for evaporative cooling and the production of cold pools should convection develop.
Most of the severe weather in western Pennsylvania was associated with a quasi-linear convective system. Over northwestern Pennsylvania (5) and southwestern New York (5) several modestly rotating storms developed which producing hail around 25mm (inch) to 37.5 mm in diameter. All other reports were associated with strong surface winds.
In central Pennsylvania the larger QLCS produced only 1 report of damage. All other wind reports came from a single storm which developed in Somerset County and raced northeastward producing damaging winds from Altoona to Lock Haven, Pennsylvania. This storm, in its early evolutions had dual-pol characteristics which implied that melting hail and large drops may have contributed to the initial evolution of the cold pool and strong outflow boundary which raced up Bald Eagle Valley.
Early May Cut-off low and Mid-Atlantic rains:A deep 500 hPa cutoff developed in the southern Plains on 3 May 2013. It produced a prolonged period of unseasonably cold weather in the Plains to the Gulf States from 2-6 May 2013. There were some reports of late season snow from this system in places were spring time observations of snow are rare.
As this relatively long-lived cutoff moved northeastward, it produced bands of heavy rainfall in the Mid-Atlantic region and southern New York. The rain fell in the northeast quadrant of the cutoff as it slowly moved northward and merged with westerlies.
The initial rain bands in the northeast quadrant of the cutoff were relatively well predicted by the NCEP SREF. The second surge of heavy rainfall across New Jersey and New York, though predicted by the SREF was a relatively low probability outcome event.
A slow moving frontal system and a plume of deep moisture brought heavy rain to the Ohio Valley on 5-6 October 2013. The City of Louisville, KY had a record 24-hour and storm total rainfall during the period of heavy rain. The total rainfall on 5 October was 5.91 inches which broke the previous record of 3.07 inches set in 1910 . This also broke the monthly 24-hour record of 5.07 set in 2004. The heavy rainfall produced significant flooding in and around the City of Louisville resulting in numerous water rescues. The rain near Louisville peaked after 0000 UTC 6 October 2013 with the heaviest rainfall falling in the 6-hour period ending at 1200 UTC. The gridded rainfall plots imply that the heaviest rain occurred west of the City of Louisville.
The pattern was well predicted and was a pattern often associated with heavy rainfall. Despite the generally well predicted pattern, the NCEP models and the SREF place the axis of heavy rainfall to far north and west of where it was observed and under predicted the high end rainfall amounts.
Impacts of the April 2013 Mean trough over central North America weather
The mean 500 hPa flow over North America featured a trough over the continent and ridges along the East and West Coasts. The mean trough was associated with several surges of cold dry air which penetrated into the Gulf of Mexico. This produced several cold episodes and a 10 day period where over 100 record low high-temperatures were set or tied. The relatively cool air led to 3 significant late season snow falls from the Rockies to the Great Lakes.
There were two surges of deep moisture and high PW air. The first surge produced 3 of the 6 severe events which had 100 more severe reports between 9-11 April 2013. The second surge around the 18th of April was associated with a second round of severe weather and a heavy rainfall event in the Mid-West which produced river flooding in Illinois and Michigan.
The mean trough limited the surges of high PW which in turn limited the severe weather and tornado activity in April 2013.
A fast moving cold front with limited CAPE, strong low-level winds and strong shear produced a minor convective event across Pennsylvania during the afternoon hours of 19 April 2013. Due to the low CAPE storms had limited vertical extent. In Pennsylvania wind damage was the primary means to verify severe weather. No METAR sites observed winds over 45kts and in central Pennsylvania, only KUNV reported thunder. Most of the lightning with this event was south of the Mason-Dixon line.
Farther south higher CAPE over Virginia and the Carolinas produced deeper convection and stronger more persistent thunderstorms. This led to more widespread reports of severe weather in the Mid-Atlantic States south of the Mason-Dixon Line.
There were a few unique bowing segments in Pennsylvania that lined up well with some of the areas of wind damage.
Mid-West Heavy rains 18 April 2013: The relatively wet conditions during the first 16 days of April 2013 set the stage for potential flooding over the Midwest. A strong frontal system and a strong ridge to the east pushed a plume of deep moisture and high values of precipitable water into the Mid-Mississippi Valley and Great Lakes region. This plume of moisture produced heavy rains, several areas received 50 to 100 mm in 12 hours and 75 to 175 mm in about 36 hours.
The heavy rains produced flooding and disrupted transportation in towns and cities. The intense rainfall disrupted autos, rail, and air transportation in Chicago. As the water flowed into rivers and streams, it produced flooding along many rivers. The Grand River in western Michigan and the Illinois river in Illinois both had major flood and near record flooding.
From a prediction perspective, both the pattern and the probability of heavy rainfall were relatively well predicted by the NCEP guidance systems. The GFS produced over 100m of QPF and the SREF showed a broad region to receive in excess of 100 mm of QPF.
A large subtropical ridge over the western Atlantic and adjacent southern United States pumped deep moisture, strong winds, and instability into the Mississippi and Ohio Valleys then eastward across Ohio and Pennsylvania. This produced a series of early season MCS which brought strong and damaging winds to the region.
High CAPE was observed in the warm moist air over the ridge and south of a strong low-level frontal boundary. The conditions including the strong low westerly winds, deep moisture, and unseasonably high CAPE produced long lived quasi-linear convective systems. Embedded within these systems were bow echoes which accounted for most of the severe weather, from damaging winds, which affected the Mid-Atlantic region.
A persistent high latitude block over northeastern North America produced relatively cold conditions over much of the eastern United States during March 2013. A series of short-waves moved beneath this block produced several late winter snow events in the eastern United States and one significant early spring Mid-Atlantic snow event on 25-26 March 2013. In the west, several large ridges developed keeping most of the southwestern United States relatively warm. Several surge of Pacific moisture moving over the ridge brought moisture and heavy precipitation to the Pacific Northwest (Fig. 2).
These daily data show contribution and the impact of the series of transient ridges in western North America and short-waves moving beneath the high latitude ridge which contributed significantly the monthly pattern. The only persistent feature though most of the month was the persistent high latitude ridge over northeastern North America. The high latitude ridge was associated with blocking and a period of strongly negative values of the AO.
Late winter storm beneath high latitude block.Produced heavy snow in portions of the East. Over forecast snow and QPF in Pennsylvania. Rain and severe weather event in Gulf States. Focus is on the pattern and anomalies and the over prediction of snow in the Mid-Atlantic region.
These data show that a fast moving short-wave brought a brief surge of above normal PW into
the Mid-Atlantic region. The PW values were only about 1σ above normal (Fig. 2) indicative
of a good but not extreme rainfall event. The pattern was a well-known pattern associated with
many moderate and heavy rainfall events, a Maddox-Synoptic pattern with a surge of strong
southerlies and moisture ahead of an advancing frontal system.
As shown here, the NCEP SREF correctly predicted the relative timing, orientation, and pattern
of the precipitation shield. In areas of heaviest rainfall it may have slightly underestimated the
precipitation, though at least 1 member did predict in excess of 50 mm of QPF in close proximity
to the close 48 mm contour in Figure 3. The SREF forecasts indicated the axis of heavier QPF
quite well and as forecast length decreased, the forecast region of heavier rainfall converged
toward where most of the heavier precipitation was observed.
The pattern and the SREF QPF probabilities correctly indicated that this was not likely a high
flood threat event. The two points in New York which reached flood stage likely had some
contribution due to snowmelt. Most locations remained well below flood levels. Without a
contribution from snow melt or frozen ground, it is often difficult to get serious flooding impacts
when rainfall is below 75mm over much of Pennsylvania.
A strong mid-tropospheric wave moved across the central United States and produced heavy rainfall in the mid-Mississippi Valley (MMV) and a band of moderate to heavy snow across the central Plains. The NCEP GEFS correctly predicted the potential for 25 to 50 mm of rainfall in MMV with ~6 to 7 days lead-time. The forecasts of the precipitation band on the cold side of the wave had a predictability horizon on the order of ~1 day.
The NCEP SREF showed large uncertainty with the 500 hPa trough and surface pressure fields. As the forecast length decreased, uncertainty decreased and the SREF and GEFS converged on a band of precipitation north and west of the track a deeper cyclone than longer range forecasts had indicated. The uncertainty contributed to large differences in the character and evolution of the 250 hPa and 850 hPa upper and lower level jets respectively.
Despite the uncertainty with the trough and significantly different evolution of the system in shorter-range forecasts, the area of 25 to 50 mm of rainfall was relatively well predicted with ~3days lead-time in the SREF and about ~6-7 days lead-time in the GEFS. The predictability horizon of the band of precipitation, which produced heavy snow, was on the order of only ~1 day. This case shows that within a similar region and when affected by the same synoptic weather system, predictability horizons can vary considerably.
A late season winter storm brought snow, rain, and strong winds to a broad swath of the eastern United States on 5-7 March 2013. The estimated precipitation field (Fig. 1) provides a good overview of the strong northern stream system (Fig. 2) which pulled moisture form the Gulf of Mexico, to produce the deep cyclone along the East Coast (Fig. 3). The Clipper-like (Thomas and Martin 2007). short-wave with -1 height anomalies (Figs. 2a-e) produced a swath of moderate to heavy snow (Fig. 4) as it moved across Wisconsin, Illinois, Indiana, Ohio, West Virginia, and western Pennsylvania. As the energy transferred to the coast system, beneath the deep 500 hPa low heavy snow fell in southern Pennsylvania, West Virginia, Virginia, and western Maryland.
This paper will document the 5-6 March 2013. The focus is on the pattern and standardized anomalies to show how strong the storm was and which features in the may have played a role in event. The paper also examines forecasts from the NCEP GEFS and SREF to provide some insights into predictability of this storm and perhaps aid in better prediction of similar storms in the future. In the Mid-Atlantic region, many lower elevations locations facing strong low-level easterly flow received rain despite below freezing temperatures above the PBL. This caused scenarios where snow fell with warm temperatures with little accumulation
The second storm within a week brought heavy snow from the eastern slopes of the Rocky Mountains and southern plains into New England. Heavy snow and blizzard conditions affected portions of the New Mexico, Texas and Oklahoma panhandle where wind gusts reached 50 to 84 mph. Portions of Missouri and Iowa saw the second large snowfall in less than a week.
Similar to the storm of 21-22 February 2013, the heavy snow fell in a region of strong easterly winds. The 850 hPa u-wind anomalies during the storm reached -5 below normal.. The strong low-level winds were the result of the gradient between a modest anticyclone to the north and a deept surface cyclone to the south. Unlike the previous storm, this storm lacked a strong anticyclone which and had a deep cyclone, which may have limited the amount and extent of sleet and freezing rain.
The storm was relatively well predicted by the NCEP forecast systems to include both the Global Ensemble forecast system and the short-range ensemble forecast system. Both showed the potential for heavy snow, a deep cyclone, anomalous easterly flow on the cold side of the boundary and sufficient QPF to produce heavy snowfall. During this event model and ensemble precipitation types and quantitative precipitation amounts correctly highlighted the areas likely to receive heavy snowfall. There were some precipitation type issues in the transition zone from snow to rain.
Central United States Winter Storm of 20-22 February 2013:
A strong winter storm brought heavy snow, sleet, and freezing rain to the central plains and lower Missouri Valley. Heavy snow affected Kansas, Nebraska, and Missouri with sleet and freezing rain along the southern edge of the snow shield across southern Kansas, Missouri and northern Arkansas. Snowfall amount between 12 and 20 inches were observed in Kansas to Nebraska and a more east-west band of heavy snow fell across Missouri.
The heavy snow fell in a region of strong easterly winds. The 850 hPa u-wind anomalies during the storm reached -4 below normal. The strong low-level easterly winds implied a strong frontal circulation which kept the low-level cold air in place and enhanced the lift. The strong low-level winds were the result of the gradient between a strong anchoring anticyclone to the north and a modest surface cyclone to the south. Not all major winter storms require a strong surface cyclone and often strong anticyclones are key players in mixed precipitation and heavy snow events.
The storm was relatively well predicted by the NCEP forecast systems to include both the Global Ensemble forecast system and the short-range ensemble forecast system. Both showed the potential for heavy snow, an anomalous anticyclone to the north, anomalous easterly flow on the cold side of the boundary and sufficient QPF to produce heavy snowfall. During this event model precipitation types and areas to be affected by heavy snow and mixed precipitation were relatively well forecast with 3-5 days lead-time.
Key words: ensembles standardized anomalies winter storms.
A fast moving short-wave (Fig.1) and surface cyclone (Fig. 2) brought rain and snow from the Ohio Valley into the Mid-Atlantic region (Fig. 3). Despite what appeared to be marginal conditions for snow central Pennsylvania, most areas received snow. Snowfall in central Pennsylvania briefly fell at rates of up to 2” per hour with 2-4 inches reported in Altoona and State College communities. Short range model guidance struggled with the precipitation type forecasts, with the NCEP SREF and NAM precipitation types were a mix of rain and snow, primarily due to a warm boundary layer. The 850 hPa temperatures over most the region was -2 to -4C and was forecast toward cool to the wet-bulb temperature during the event.
The fast moving 500 hPa short-wave (Fig. 1) produced a strong area of lift well north of the surface cyclone (Fig. 2). The 500 hPa heights and IR satellite image (Fig. 4) show this enhanced area well north of the surface circulation. The central Pennsylvania radar (KCCX:Fig. 5-6) showed enhanced displaced south the coldest IR clouds tops, a northward moving band of snow. The maximum reflectivity in the band was in the 30 to 40dBZ range. The 500 hPa wave and the IR imagery show a similar structure to the banded snowfall conceptual model (Novak et. al 2004), though the wave in these case had an open wave structure.
This paper will document the event of 13 February 2013. Focus is on the pattern and use of short-term forecasts to forecast low predictability horizon events of similar nature in the future.
A fast moving short-wave raced across the southwestern United States and into the Mid-Mississippi Valley from 8-10 February 2013 (Fig. 4). This wave pulled a plume of deep moisture with PW anomalies of 2-4 (Fig. 5) above normal with strong low-level (Fig.6). The convection and severe weather developed in this plume of deep moisture and strong shear.
Radar imagery (Figs. 3 &4) and RAP simulated radar (Figs. 8-10) showed a strong line of convection moved across Texas then into the Gulf States. After 1800 UTC 10 February the line broke into more discrete storms and these discrete storms produced tornadoes and at least one EF4 tornado near Hattiesburg, MS. The storm damaged in excess of 800 homes and caused 10s of millions of dollars of damage to the University of Mississippi (Huffington Post 2013).
The strong frontal system and convection produced 25 to 50 mm of rainfall across western Texas and across the lower Mississippi Valley. Due to the strong forcing the NCEP SREF was able to predicted a line of enhanced rainfall moving across the region at about the correct time. The SREF also predicted over 25 mm of rainfall over the correct region. The NCEP 13km RAP did reasonably well showing the convective evolution of the system over the Gulf States. It lacked resolution and had both intensity and timing issues. However, these data show the emergent power of high resolution and rapidly updated forecasts to aid in anticipating the timing and mode of convection for strongly forced high-impact weather events.
A high impact winter storm brought rain, wind, and record heavy snow to the eastern United States on 8-9 February 2013. The heavy snow from southeastern New York and Long Island, across Connecticut and into Maine was the result of a strong cyclone which tracked up the East Coast on 8 February before interacting with a northern stream wave. The two systems merged during the evening hours of 8 February 2013. During this merger period intense snowfall affect central Long Island and Connecticut producing areas of 30 to 40 inches of snowfall. From a historic perspective this storm was compared to the February 1978 Storm. This was a top 5 snowfall event in many locations across southern and eastern New England and for many sites in Long Island and Connecticut is the new snowfall of record.
The storm was relatively well predicted 1 to 3 days in advance and the European Centers high resolution model provide some insight into storm potential about 6 days in advance. As the storm approached and the forecast length decreased the models and ensemble prediction systems produced a deep cyclone with an anchoring intense anticyclone to the north, resulting in forecasts of 850 hPa winds -5 to 6 below normal, implying a near record if not historic event.
With the strong frontal forcing implied by the strong winds, both model and ensemble forecast system quantitative precipitation forecasts were on the order of 25 to 50 mm in areas where the predicted precipitation type was forecast to fall mainly as snow. Forecasts of 18 to 38 inches were common in SREF forecasts at least 48 hour prior to the onset of precipitation.
This paper documents the event of 8-9 February 2013 providing reanalysis of the event using standardized anomalies to but the event in a climatological context. Supporting data on the predictability of the event is provided focused on the European Centers longer range deterministic model forecasts and the National Centers for Environmental Prediction Centers forecasts of the event.
A strong Pacific wave coming over a ridge produced a strong trough over western North America. This trough pulled a plume of deep moisture from the eastern Pacific into the southwestern United States and northern Mexico. Ahead of this plume of deep moisture and the developing wave, a large 500 hPa ridge developed over much of the eastern United States.
The large ridge produced a period of unseasonably warm weather from 27-30 January 2013. At the peak of the intrusion of warm air ahead the trough, over 300 high temperature records were set on 28 January. Many sites also set new all-time record high minimum temperature records. The warm air eventually flooded most of the eastern United States.
As the wave moved eastward a strong surface front developed and a strong surface cyclone formed along the front. In the plume of deep and anomalous moisture, where precipitable water anomalies peaked in the +3 to +6 above normal range a heavy rainfall and severe weather event developed. In the course of two days there were over 600 reports of severe weather and over 20 tornadoes. This was one of the largest and most widespread cold season severe events. Though it produced fewer tornadoes than the February 2008 “Super Tuesday” outbreak. It shared many of the characteristics of the 3 winter season severe outbreaks of January and February 2012.
In addition to the severe weather, the event produced heavy rainfall and flooding.
The overall pattern favoring heavy rainfall and severe weather was generally well predicted by the NCEP models and ensemble forecast systems.
Key words anomalies ensembles winter severe and tornadoes.
The period of 15-27 January saw the incursion of arctic air into eastern North America and much of the north-central and northeastern United States. During the peak of the cold episode (Fig. 1) a deep trough with -2 to -3s height anomalies (Fig. 1a) and a pool of deep cold air (Figs. 1b&1c) were present over much of eastern Canada and the northern tier of the United States. Much of the North America was dry with large areas where the precipitable water (Fig. 1d) was near or below normal. The strong and persistent ridge over the southwestern Atlantic limited the penetration of the cold air into the southern and southeastern United States.
A sudden stratospheric warming (SSWE:Smith and Kushner 2012; Baldin et al. 2012) event was observed in long range forecasts in late December 2012 and early January 2013. SSWE events are typically monitored above 50 km and temperatures on model pressure surfaces of 70 to 10 hPa are often examined to monitor these events. Baldwin and Dunkerton (2001 hereafter BD2001) noted that stratospheric events often follow the arctic oscillation (AO). Observational studies suggest that SSWE events be used to predict changes in weather regimes. Large warm ups over the Polar Regions often lead to arctic outbreaks over North America. BD2001 entitled their paper “Stratospheric Harbingers of Anomalous weather Regimes” due to the apparent observational predictability component of such events. During most winters, in the January to February time frame there is typically 1 major stratospheric warming event (Kuttippuarth and Nikulin 2012)
During the onset of the SSWE event, conditions had been relatively warm over most the eastern United States. Through December through about 6 January a cold pocket was present over the pole at 10 hPa which was replaced by a ridge and above normal temperatures after 6 January 2013 (not shown). The strong ridge over the southwestern Atlantic (Fig. 1a) was dominant feature through first half of January 2013, producing generally warm weather over most the eastern United States. High temperature records were tied or broken during a prolonged period in the southeastern United States (Table 1) through 18 January 2013. A surge of warm air ahead of the first blast of cold air tied or broke over 100 daily maximum temperatures records from 12-13 January 2013 in the eastern United States. The warmth then emerged over the southwestern United States (Table 1).
This paper examines the pattern of January 2013 with a focus on the evolution of the big chill of mid-January 2013. The persistent ridge over the southwestern Atlantic precluded the intrusion of the cold air into the citrus growing regions of the southern United States. Forecasts of the event are presented showing that the event and pattern change were relatively well predicted. Finally, this event and its associated pattern are compared to the arctic outbreaks of January 1985 and 1994.
Key words: cold anomalies ensembles
A large and persistent subtropical ridge (Fig. 1) provided unseasonal warm conditions to much
of Australia during early January 2013. The large subtropical ridge had +1 to +2σ above normal
250 hPa height anomalies associated with it. The ridge appeared to peak on 15 January 2013
when the 250 hPa height anomalies exceeded +2σ beneath the closed anticyclone. News
accounts indicated that the heat wave persisted most of the first 2 weeks of the month with high
temperatures over 48C common place and 49.6C reading at Moomba in southern Australia.
Beneath the ridge the 850 hPa temperatures were above normal over most of the continent (Fig.
2). The subsidence beneath the ridge over the eastern side of the continent maintained a hot dry
air mass (Fig. 4). Plume of high precipitable water air and deep moisture (Fig. 3) was present
on the southwestern edges and ocean regimes south of the continent. The strong flow about the
subtropical ridge blocked the region from the deep moisture plume.
Beneath the ridge surface temperatures were warm with extended heat on 7 and 8 January (Fig.
5) the heat beneath the ridge shown in Figure 1 is depicted in Figure 6. The ridge earlier in the
month which produced the extreme heat on 7-8 January was also associated with a strong ridge
(Fig. 7). The ridge earlier in the month was displaced farther south and the air was considerably
drier (not shown) over much of the continent beneath the strong ridge in early January.
This paper will document the pattern and standard anomalies associated with the eastern United
States precipitation event of 15-16 January 2013. The focus is on the standardized anomalies to
describe the pattern and on using guidance to include ensemble guidance to aid in the prediction
of this and similar events.
This case shows the value of standardized anomalies in characterizing high impact weather
events and the clear association of strong ridges with enduring heat episodes and droughts.
A weak wave and a strong anticyclone (Fig. 1a-e) produced a precipitation event from the Mid-Mississippi Valley into southern New England on 15-16 January 1996 (Fig. 1f). A strong frontal boundary was present over the region with cold air to the north associated with the surface anticyclone and warm air to the south in the region of the surface trough and to the south and east (Fig. 2). The 850 hPa temperatures were above normal on the warm side of the boundary. The heaviest precipitation fell over portions of the Ohio Valley and western Virginia, along and mainly on the warm side of the 850 hPa frontal boundary. A moderate snowfall was observed on the cold side of the storm. Central Pennsylvania received 1-4 inches of snowfall early on the 16th of January.
There was a surge of moisture ahead of the frontal boundary (Fig. 3) with precipitable water (PW) values in the 25 to 35mm range in the warm air. The anomalies were in the +2 to +3 range on the warm side of the southwest-to-northeast oriented frontal boundary. The 850 hPa winds were relatively weak and out of the south-southwest during the period of precipitation (Fig. 4). The southwesterly flow and modest 850 hPa winds produced higher values of moisture flux and at times 3 to 4s above normal moisture flux anomalies in the warm air, in close proximity to the region of the higher precipitation amounts (Fig. 5).
The larger scale pattern showed a deep trough over the western United States and strong ridge over both the eastern Pacific and western Atlantic (not shown). The flow between the deep trough and the western Atlantic ridge produced a strong 250 hPa jet (Fig. 6) with 250 hPa winds near 100kts in the jet axis going over the implied strong Atlantic ridge. The 250 hPa wind anomalies. Were +4 to +5s above normal near the ridge and implied a strong jet entrance region over the eastern United States with the strong jet core on the cold side of the 850 hPa boundary (Fig. 2).
The synoptic pattern suggested a strong frontal boundary and potential jet entrance moving over the region. There were indications in the 700-500 hPa layer of a short-wave which was part of the jet entrance circulation. The pattern was well suited to produce a precipitation event and had the potential, with the sub-zero C air to the north to produce some snow. The winds at lower levels were not indicative of a widespread high impact weather event (HIWE). Lacking strong forcing, the event was not well predicted with significant lead-time, especially the northern edge of the precipitation shield.
This paper will document the pattern and standard anomalies associated with the eastern United States precipitation event of 15-16 January 2013. The focus is on the standardized anomalies to describe the pattern and on using guidance to include ensemble guidance to aid in the prediction of this and similar events.
A slow moving and deep 500 hPa cyclone moved over the eastern Mediterranean basin from 5-10 January 2013. As the low deepened the strong frontal system pulled warm moist air into the eastern Mediterranean leading to locally heavy rainfall on the 5th and 6th of January 2013. As the low deepened it both generated and advected cold air into the Middle East bringing with it sub-zero temperatures through a most of the atmospheric column. Once the cold air moved over the region, this supported snow inland and at higher elevations. Jerusalem received in excess of 20 cm of snow. The resulting multi-day event produced heavy precipitation, flooding, cold, damaging winds, and heavy snowfall.
The heavier rainfall in the event occurred as a surge of high PW air and strong winds moved into the eastern Mediterranean. This produced high moisture flux and moisture flux anomalies on the order of 6s above normal. The result was heavy rainfall which was relatively well predicted by the NCEP GEFS, likely due to the strong forcing within the model atmosphere.
The snow and cold phase of the event occurred when the deep 500 hPa cyclone moved into Israel, Lebanon, and Syria. Beneath this deep cold 500 hPa cyclone the 850 hPa temperatures were in the -2 to -5C range. This was sufficiently cold to allow the precipitation to fall as snow in the higher terrain and away from the warm boundary layer air from the Mediterranean to the west. The deep cyclone and cold air was relatively well predicted, with long lead-time by the NCEP GEFS.
Synoptically, a deep slow moving cyclone, with a close mid-level cyclone brought a surge of strong winds and above normal moisture into the eastern Mediterranean basin. The result was a multi-day precipitation event, with locally heavy rainfall, inland and elevation dependent snowfall, and strong winds. The precipitation resulted in regional flooding and some areas of heavy snow fall. The heavy snow portion of the event involved a deep 500 hPa closed cyclone and a core of cold air at 850 hPa. This produced the largest snowfall in Jerusalem since February 1992 when a similar deep cut-off low and deep cyclone moved over the region.
key words: Ensembles anomalies snow Israel January 2013
An examination was made of lake ice out data over the eastern United States. Ice out data was obtained from Minnesota to Maine. Every lake examined showed a general trend toward earlier ice out dates. In addition to ice out dates, a few lakes had both ice-in and ice-out data allowing the examination of the changes in total ice days. Similar to ice out dates, the length of time lakes are iced over shows a trend toward a later total time of ice cover. The few lakes examined showed a trend toward a later beginning of the ice season.
National Weather Service Cooperative Observing site data in close proximity to several of the lakes were examined. These data showed a basic trend toward warmer late winter and early season mean temperatures similar to the trend in earlier ice out dates. In general warm March and April weather was associated with the earlier ice out dates. Conversely cold late winter and early springs were associated with the later ice out dates, which often extended into late April and May during the coldest years.