Second Northeast Regional Operational Workshop

Albany, New York

Agenda and Abstracts

 

Tuesday, November 7, 2000

 

1:00 pm     Welcoming Remarks

Eugene P. Auciello, Meteorologist in Charge, NWS, Albany, NY
Warren R. Snyder, Science & Operations Officer, NWS, Albany, NY

 

Session 1. Cold Season Events and Phenomena

 

1:15 pm:  Predictability of the 24-25 January 2000 Snow Storm with Respect to Model Grid Resolution and Initial State

Fuqing Zhang, National Center for Atmospheric Research, Boulder, CO


1:45 pm:  The 24-25 January 2000 East Coast Snowstorm: WFO Sterling Perspective

Steve Zubrick, Weather Forecast Office, NOAA/NWS, Sterling, VA.


2:15 pm:  Evaluation of RUC2 Snow Forecasts for the Northeast U.S. During Winter 1999-2000.

Greg Carbin, National Centers for Environmental Prediction, Storm Prediction Center, Norman, OK.

 

2:45 pm:    Break



3:15 pm:  A Foot of Snow from a 3000 Foot Thick Cloud: The Ocean Effect Snowstorm 14 January 1999

Jeff S. Waldstreicher, NOAA/NWS Binghamton, New York

 

3:45 pm:  Oceanic Bombs: What makes them "Tick"?

Lance F. Bosart, Department of Earth and Atmospheric Sciences, University At Albany, State University of New York, Albany, New York

 

4:15 pm:  Measurement of Snow over the Lake Champlain Basin

Thomas Econopouly and Michael Winchell Northeast River Forecast Center, NOAA/NWS, Taunton, Massachusetts


4:45 pm Adjourn

 

Evening Session

 

7:00 pm:   The East Coast Snowstorm of 24-25 January 2000

Geoff Manikin, General Sciences Corp., Washington, District of Columbia.



Wednesday, November 8, 2000


Session 2. Warm Season Events and Phenomena

 

8:30 am:   The Development of Severe Thunderstorm Warning Criteria for Pulse Severe Thunderstorms in the Northeastern United States

Carl S. Cerniglia Jr. and Warren R. Snyder   Weather Forecast Office, NOAA/NWS, Albany, New York

 

9:00 am:  Northeast United States Severe Weather Distribution as a Function of Flow Regime

Alicia C. Cacciola, Sheryl Honikman, Thomas J. Galarneau, Lance F. Bosart Department of Earth and Atmospheric Sciences, University at Albany State University of New York, Albany, New York;  Kenneth D. LaPenta, John S. Quinlan, Weather Forecast Office, NOAA/NWS, Albany, New York


9:30 am:   The Dynamics of Heavy Rainfall Associated with Hurricane Floyd

 Eyad Atallah and Lance Bosart, Department of Earth and Atmospheric Sciences, University At Albany, State University of New York, Albany, New York 

10:00 am: Break

 

10:30 am:   Radar Observations of Northeastern United States Tornadoes

Kenneth D. LaPenta, George J. Maglaras, John S. Quinlan and Hugh W. Johnson Weather Forecast Office, NOAA/NWS, Albany, New York,  Lance F. Bosart and Thomas J. Galarneau  Department of Earth and Atmospheric Sciences, The University at Albany State University of New York, Albany, New York


11:00 am:   Planetary- and Synoptic-Scale Environments of Strong Tornadic Events over the Northeastern United States

Kelly Lombardo, Eyad Atallah, and Daniel Keyser. Department of Earth and Atmospheric Sciences, University At Albany State University of New York, Albany, New York

 

11:30 am: Forecasting Synoptic and Mesoscale Environments for Tornadoes and Derechos in the Northeast United States

Sheryl F. Honikman, Alicia C. Cacciola , Tom J. Garlarneau Jr. and Lance F. Bosart Department of Earth and Atmospheric Sciences, University at Albany State University of New York, Albany, New York Kenneth D. LaPenta, Weather Forecast Office, NOAA/NWS, Albany, New York


Noon:   Lightning In Tornadic Thunderstorms Over the Northeastern United States

Thomas J. Galarneau, Sheryl F. Honikman, Alicia C. Cacciola, and Lance F. Bosart Department of Earth and Atmospheric Sciences, The University at Albany State University of New York, Albany, New York  Kenneth D. LaPenta, John S. Quinlan, and Glenn E. Wiley Weather Forecast Office, NOAA/NWS, Albany, New York 

12:30 pm:   Lunch

 

Session 3. Operational Techniques

 

2:00 pm:   AWIPS Display Strategies for Enhancing the Visualization of Radar Imagery and NWP Model Output Fields

Josh Korotky, Weather Forecast Office, NOAA/NWS, Pittsburgh Pennsylvania


2:30 pm:  New Present Weather Imaging Technology for the Operational Forecaster

Mark Beaubien, Yankee Environmental Systems, Inc., Turners Falls, MA  Jeff Freedman, Atmospheric Sciences Research Center, University At Albany State University of New York, Albany, New York


3:00 pm:   COOP Station Forecast and Verification Studies

Joseph P. Villani, Lance F. Bosart, and Daniel Keyser Department of Earth and Atmospheric Sciences, University at Albany State University of New York, Albany, New York.  Michael J. Cempa, Thomas A. Wasula, and Stephen R. Pertgen.  Weather Forecast Office, NOAA/NWS, Albany, New York

3:30 pm Evaluation of the SUNYA ReCM for the Winter Climate of Northeastern U.S.

Michael Notaro, Wei-Chyung Wang, and Wei Gong Atmospheric Sciences Research Center, University At Albany State University of New York, Albany, New York


4:00 pm:   A Dynamical-Statistical Atmospheric Modeling System Designed To Simulate Wind Generated Power Production Rates

John W. Zack *, Joseph J Nocera, Kenneth T. Waight, Glenn E. Van Knowe, and Mary D. Bousquet, MESO, Inc., Troy, New York

 

4:30 pm:  Adjourn




 




Predictability of the 24-25 January 2000 Snow Storm with Respect to Model Grid Resolution and Initial State

Fuqing Zhang, Chris Snyder and Rich Rotunno

National Center for Atmospheric Research, Boulder, Colorado

On 24-25 January 2000, an intense winter storm off the southeastern coast of the United States brought heavy snowfalls from the Carolinas, through the Washington D.C. area, and into New England.

Record snowfall amounts fell across North Carolina, with the Raleigh-Durham area reporting a snowfall total from the storm of over 20 inches, breaking not only the snowfall record of a single storm, but also the highest total in one month. The intensity and position of the storm was ill-handled by the operational models running at the National Center for Environmental Prediction (NCEP), and posed a serious challenge for forecasters in the affected region.

The purpose of this study is to explore possible reasons for the failure of the operational models to simulate the explosive coastal cyclogenesis an its associated onshore precipitation and to investigate various factors which could have affected the predictability of the snow storm with mesoscale numerical models. In particular, the forecast sensitivity with respect to model grid resolution and initial state will be emphasized. Through numerous experiments with a state-of-the-art mesoscale model, we are trying to search answers to the following questions: Could the storm have been forecasted more accurately? What would it take to make more accurate forecasts? Are we approaching the limit of mesoscale predictability? And what is the nature of the error growth in the mesoscale model given this storm environment, is it nonlinear and how fast can small scale errors influence larger scales?

The triple-nested MM5 which is running as operational with the highest grid resolution of 3.3km simulated well the explosive coastal cyclogenesis in terms of the cyclone strength and location as well as the heavy onshore precipitation band. The success of the high-resolution control simulation shows that the storm could have been well forecasted with conventional data in real-time. Various sensitivity experiments suggest that the difficulty of the real-time operational numerical forecast come from insufficient model grid resolution, errors in the model initial state, and the strong nonlinearity inherent in the dynamic system.

 

 


 

 

The 24-25 January 2000 East Coast Snowstorm: WFO Sterling Perspective

Steve Zubrick

Weather Forecast Office, NOAA/NWS, Sterling, Virginia

On 24-25 January 2000, an major East Coast snowstorm blanketed much of the Eastern seaboard region. Many locales experienced upwards of nearly two feet of snow. The major metropolitan centers of Raleigh, Washington DC, Baltimore, Philadelphia and others were impacted by the rapid spread of heavy snow during this event. Record snowfall was recorded was recorded in central North Carolina as a result of this storm.

This storm presented a particular challenge to weather forecasters up and down the East Coast beginning five days prior and even until the afternoon before the storm explosively deepened. This talk will focus on the perspective of the forecasters at the National Weather Service (NWS) Forecast Office in Sterling, Virginia (LWX) as they dealt with the challenges of forecasting this storm, as well as other snow events leading up to this event. Discussion will include clues provided by operational data sets (satellite, radar, surface/upper air data, etc.) as the storm began to undergo rapid surface cyclogenesis during the day on Monday 24 January. Discussion will also focus on the rather broad range of model solutions for this event. While forecasting this storm was a challenge prior to 1200 UTC Monday 24 January due to model uncertainty, acute attention to observational data sets after that time, coupled with a knowledge of conceptual models for rapid cyclogenesis, may have lead to improved public NWS forecasts of this event. Lessons learned at the local NWS office level for this event will be discussed.






 

 

 

 

 

 

 

 

 



Evaluation of RUC2 Snow Forecasts for the Northeast U.S. During Winter 1999-2000

Greg Carbin

National Centers for Environmental Prediction, Storm Prediction Center Norman, Oklahoma

 

The accuracy of the precipitation-type variable for snow from the Rapid Update Cycle Model (RUC2) was evaluated at 22 northeast U.S. ASOS sites from December 1999 to March 2000. The 6- and 12-h snow forecasts were verified by interpolating the precipitation-type output to each ASOS site, and then recording the observed weather at each site at the forecast valid time. Over 19,000 forecast/observation pairs were used to produce contingency tables for all categorical snow forecasts.

Web based comparison graphics of model precipitation-type to observed weather were produced daily during the winter of 1999-2000 at: www.spc.noaa.gov/staff/carbin/rucwx/. A monthly surface file containing the model forecasts for snow, as well as the observed weather at each ASOS site, was constructed from the daily data at three hourly intervals beginning at 0000 UTC. The output from the monthly file was then analyzed and processed to produce verification statistics including critical success index (CSI), probability of detection (POD), false alarm rate (FAR), and bias at each site. Sites were also combined to produce regional 6- and 12- hour forecast statistics.

Preliminary evaluation of verification statistics for individual locations shows large geographic variability exists across the selected domain. Exceptionally high forecast skill appears at a few stations. Mostly small changes occur in CSI between 6- and 12-h forecasts. However, 12-h forecasts at a few sites actually exhibit greater skill than 6-h forecasts A high false alarm rate for snow appears at several stations where the RUC2 model terrain is over 200 m above the actual station elevation. A correlation between model terrain and forecast skill will be shown. Overall, forecast bias is high at 12-h, becoming nearly unbiased at 6-h. These findings will be further illuminated with graphs and plots of selected statistics. The model evaluation technique used in this study should eventually lead to analysis of larger sets of data for more robust point forecast verification.









A Foot of Snow from a 3000 Foot Thick Cloud:  The Ocean Effect Snowstorm 14 January 1999

Jeff S. Waldstreicher

NOAA/NWS Binghamton, New York


During the afternoon of 13 January 1999, a cold front swept through southern New England, ushering in a arctic airmass. A 1050 mb surface high building across Quebec resulted in a north northeast flow of very cold air across the New England coastline. The heat and moisture flux between this bitterly cold airmass, and the much warmer (relatively) waters of the Atlantic Ocean, in combination with frictional wind convergence associated with the land-water interface, resulted in the formation of snow bands along the coast of Massachusetts south of Boston during the evening of the 13th. Despite a capping subsidence inversion at 900 mb, these snow bands not only persisted for nearly 24 hours, but strengthened and expanded northward across the Cape Ann region north of Boston. These snow bands produced snowfall totals of 8-15 inches, with a maximum of 16.5 inches measured in South Weymouth, MA, before synoptic scale warm advection- induced snow (and subsequently sleet, freezing rain and then heavy rain), overspread southern New England during the evening of January 14th into the 15th.

This paper will examine the evolution and structure of this ocean effect snow storm. Comparisons will be made to conceptual models of lake effect snow events over the Great Lakes. A simulation of this event using the NCAR/Penn. St. Mesoscale Model - Version 5 (MM5) will be presented, and the ability of a mesoscale model like MM5 to predict small scale events like will be examined. Particular emphasis will be placed on the important role of precipitation micro physical processes and resultant snowfall efficiency mechanisms had in producing such substantial snow accumulations despite apparent strong limiting factors.





 




Oceanic Bombs: What makes them "Tick"?

Lance F. Bosart

Department of Earth and Atmospheric Sciences, University At Albany State University of New York, Albany, New York

 

Explosively deepening oceanic cyclones (aka "bombs") occur preferentially over the western margins of major oceanic basins during the cool season in middle and high latitudes (30-60°N) of both hemispheres. In the case of the North Atlantic Ocean, evidence is accumulating that the phase of the North Atlantic Oscillation (NAO) can play an important role in providing an environmental envelope of opportunity for bombs. When the NAO is in its negative phase (anomalously weak Atlantic jet) the meteorological bombing range tends to lie close to the eastern coast of North America. When the NAO is in its positive phase (anomalously strong Atlantic jet) more bombs tend to occur over the central and eastern Atlantic Ocean (e.g., December 1999)

This talk will present a multiscale overview of oceanic bombs. Emphasis will be placed on how bombs are influenced by atmospheric dynamics, atmospheric thermodynamics, atmospheric stability, and atmospheric moisture by means of selected examples. Forecast issues will also be addressed.



 

 


 

Measurement of Snow over the Lake Champlain Basin

Thomas Econopouly and Michael Winchell

Northeast River Forecast Center, NOAA/NWS, Taunton, Massachusetts

 

Runoff from snow melt is a significant component of spring runoff across the Northeastern United States. Measurements of snow water equivalent are made to enhance the National Weather Service=s (NWS) ability to model the snow melt process. Presently, the NWS snow observation network includes: point measurements, snow courses and flight lines. In this study, the various methods used by the NWS to measure snow water equivalent over the Lake Champlain Basin are compared and analyzed to determine their utility.

 

 

 

 

 

 

 

 

 

 


 

 

 

Development of Severe Thunderstorm Warning Criteria for Pulse Severe Thunderstorms in the Northeastern United States.

Carl S. Cerniglia Jr. and Warren R. Snyder

Weather Forecast Office, NOAA/NWS, Albany, New York

 

Although pulse severe thunderstorms account for a small percentage of severe storm related deaths, injuries and damage, they are generally considered more difficult to effectively warn for due to their short life cycle and isolated nature. This difficulty is evident when looking at warning verification scores for severe thunderstorms. Scores for pulse severe storms are generally lower than those for large, well organized convective events. Earlier work to develop warning criteria (Lemon 1980) for pulse severe thunderstorms was successful in improving these scores when the NWS was using the WSR-57, WSR-74C and WSR-74S radars. However, many of these techniques can not be directly translated to the WSR-88D, due to differences in scan strategies, beam resolution, system sensitivity and display properties (dBZ vs VIP levels).

The focus of this study is to try to develop criteria that will allow the warning meteorologist to distinguish between pulse thunderstorms that may become severe, verses those that will not. Due to the short lived nature of these events, many parameters will be examined with an eye toward gaining the maximum amount of lead time possible. To accomplish this, a data-set is being assembled of pulse thunderstorms from across the Northeast for the years 1995 through 1998. WSR-88D, Level II archive data was used in order to have the ability to create cross-sections and view the highest resolution data possible. Future investigation will include the examination of high resolution satellite data and cloud to ground lightning information in an attempt to increase warning lead time.

Storms must meet fairly strict criteria to be considered for this study. Storms displaying any organized structure, such as being aligned along fronts or squall lines, having supercell characteristics, producing tornadoes or having a mesocyclone present, were not included. Control cases also need to complete their life cycle over a region with a high enough population density to minimize the chance of having an undetected severe event, included as a control case. The intent behind these strict criteria is to capture the storms that are the most difficult to warn for. However, these criteria have limited the number of storms available for this study.

The presentation will show some preliminary results of the radar derived differences between severe and non-severe pulse storms. Average lead times for calculated parameters indicates that some of the currently used WSR-88D products offer little if any useful lead time for many of these events. Preliminary analysis of storm cross sections does indicate some potentially useful differences between storms that become severe and those that do not. Severe pulse storms generally develop relatively higher reflectivities, at higher altitudes, earlier in the life of the storm when compared to non-severe storms.


 


 

 

Northeast United States Severe Weather Distribution as a Function of Flow Regime

Alicia C. Cacciola, Sheryl Honikman, Thomas J. Galarneau, Lance F. Bosart 

Department of Earth and Atmospheric Sciences, University at Albany State University of New York, Albany, New York

Kenneth D. LaPenta, John S. Quinlan

Weather Forecast Office, NOAA/NWS, Albany, New York


This talk will report on the results of a COMET-funded research project between the University at Albany and the National Weather Service in Albany, New York. The focus of the research is a study of how the occurrence and location of severe weather (tornadoes, hail, high winds) in eastern New York and western New England is modulated by the underlying terrain as a function of the prevailing large-scale flow patterns. It is hoped that the results of this study will help forecasters isolate preferred regions for severe weather (more specifically, tornadoes) based on the large scale flow pattern.

The study has been divided up into two main areas: 1) a climatology of severe weather events (focusing on tornado events) in eastern New York and western New England using Storm Data reports from 1955 to 1998, and 2) a synoptic analysis of tornado events in this region from the period 1980 to 1998 using NCEP Reanalysis Data..

All available severe weather reports were stratified based on the 700mb flow direction as determined from the Albany radiosonde reports. The 700mb level was chosen in hopes that the winds would be high enough to reflect some of the large scale flow, yet low enough to be affected by the topography. Preliminary observations indicate that there is a tendency for severe weather to occur in preferred regions based on the mid-level flow direction. This suggests that funneling of wind by the river valleys may be an important factor in creating preferred regions for severe weather. Currently, work is being done to quantify these suggested results. The region has been divided into 0.5 degree latitude/longitude boxes, and in each box the average number of events per severe weather day with northwest flow is being compared to the average number of events per severe weather day with southwest flow. In addition, an attempt is being made to account for some of the known biases in the Storm Data reports, such as population.

The goal of the second area of this project is to determine the evolution and structure of synoptic-scale flow features during and preceding tornadoes in eastern New York and western New England. We examined 47 reported tornado events from 1980 to 1998. For each tornado event, we are documenting the large scale flow pattern (from the surface to the tropopause), as well as parameters important to severe weather such as vertical motion, location of jet streaks, moisture, and instability (CAPE). Composites were created of events with similar large scale features to illustrate representative signatures. The composites were run from 48 hours before the event to 24 hours after the event to try and determine what the significant features were leading up to, during, and after the tornado.

In addition, detailed case studies of selected events will be done. It is hoped that results of this part of the project will assist forecasters in identifying particular flow patterns that might indicate a preference for severe weather to occur in a particular area of the forecast region.




 








The Dynamics of Heavy Rainfall Associated with Hurricane Floyd

Eyad Atallah and Lance Bosart

Department of Earth and Atmospheric Sciences, University At Albany State University of New York, Albany, New York

 

Currently precipitation forecasts associated with land falling tropical cyclones are based on a simple algorithm where the maximum 24-h precipitation (in inches) is forecast by 100/v, where v (in m.p.h.) is the translational speed of the cyclone. This algorithm, however, provides little insight as to the precipitation distribution and intensity that can be expected in a landfalling tropical cyclone. Furthermore, several recent cases (Danny 1997, Dennis, Floyd, and Irene 1999) show that precipitation distribution and intensity can be drastically altered by interactions with mid-latitude troughs and jet streaks which often result in extratropical transitions. Occasionally, these interactions produce catastrophic rainfalls as illustrated by hurricane Floyd in September 1999. This talk will present diagnostics, from a potential vorticity (PV) perspective, of results from a case study of Hurricane Floyd, designed to elucidate the important dynamical and thermodynamical processes responsible for the modulation of the precipitation distribution and intensity.


Preliminary results suggest that regions of strong baroclinicity are often created as the cold-core anomaly associated with the mid-latitude trough approaches the warm-core anomaly associated with the tropical system. The resulting baroclinic zone can resemble a coastal front, with the significant exception that this baroclinic zone is a very deep feature, often extending throughout the troposphere. As the extratropical transition takes place, cool dry air is wrapped around the southern extent of the cyclone, resulting in a marked decrease in the precipitation on the east side of the cyclone. Meanwhile, a significant increase in both the aerial extent and rate of the precipitations occurs in the northwest quadrant of the cyclone in response to deep ascent associated with warm-air advection and differential cyclonic vorticity advection (positive PV advection).






 


Radar Observations of Northeastern United States Tornadoes



Kenneth D. LaPenta, George J. Maglaras, John S. Quinlan and Hugh W. Johnson 

Weather Forecast Office, NOAA/NWS, Albany, New York

Lance F. Bosart and Thomas J. Galarneau

Department of Earth and Atmospheric Sciences, The University at Albany State University of New York, Albany, New York


    During recent years, there have been several major tornado outbreaks in the northeastern United States. The introduction of the WSR-88D radar produced significant improvement in the National Weather Service's ability to warn the public of tornadoes during these events. Radar analysis of a large number of northeastern tornadoes will provide a better understanding of tornadic thunderstorm morphology and the radar characteristics that best identify impending tornadogenesis, leading to additional improvement in warning accuracy and lead time.

   While supercells produced greater than half of the more than 80 tornadoes examined, a variety of storm structures were associated with the tornadic thunderstorms. More than a third of the tornadoes developed with bowing lines or cells, with most of these tornadoes forming on the bulging portion of the bow. Occasionally, supercells were observed in lines of thunderstorms, or bowing line segments evolved into supercells. Appendages or hooks were observed in less than half the tornadoes, and boundary interactions may have played a role in the evolution of about one fifth of the tornadic storms. Gate to gate rotating velocity couplets of varying intensities were identified with a majority of the tornadoes. However, more than a third of the tornadoes (mostly F0 and F1) were associated with non-rotating wind maxima and may have been indicative of gustnadoes. A number of tornadoes produced no discernable radar signature due to time and spatial sampling limitations of the radar.

Preliminary analysis showed that the WSR-88D calculated shear of the gate to gate rotating velocity couplets associated with many tornadoes may be useful in identifying tornadic storms. The effects of beam spreading with range in both the actual shear calculation and the ability of the radar to resolve storm features required that the observed shear values be adjusted for range. The adjusted shear values, which provide evidence of the strength of small scale (1 km or less), low-level mesocyclones were compared to the larger scale1 km to 10 km), midlevel mesocyclone strength in order to assess a storm's tornadic potential.


 

 


 

 

 

Planetary- and Synoptic-Scale Environments of Strong Tornadic Events over the Northeastern United States

Kelly Lombardo, Eyad Atallah, and Daniel Keyser

Department of Earth and Atmospheric Sciences, University At Albany

State University of New York, Albany, New York


Although a number of studies have been conducted concerning tornadic severe weather events over the United States, relatively few studies have focused on tornadic events over the northeastern United States. Furthermore, of those studies addressing tornadic events over the Northeast, the emphasis has been on their synoptic- and mesoscale characteristics. Recently, there also has been a trend in the literature to document the modulation of significant weather events by planetary- and synoptic-scale flow anomalies, such as those associated with El Niņo/Southern Oscillation and the North Atlantic Oscillation. Given these considerations, it seems reasonable to determine whether there exists a relationship between the distribution and intensity of tornadic severe weather events in the Northeast and their planetary- and synoptic-scale flow environments. Toward this end, the distribution of tornadic events in the Northeast spanning the period from 1950 through 1995 has been mapped using Storm Data obtained through the Storm Prediction Center. The strong tornadic events (defined as F3 or greater) belonging to this distribution are divided according to season, and then further subdivided into three classes according to background flow direction (defined at the 500 hPa level over the individual event): northwesterly, westerly, and southwesterly. Composite analyses are then created for summer events belonging to these three classes using reanalyses produced by the National Centers for Environmental Prediction. The composite analyses are used to relate strong summer-season tornadic events over the Northeast to their planetary- and synoptic-scale flow environments.

Preliminary results suggest that strong tornadoes (as defined above) occur preferentially in the spring and summer, with very few events occurring in the fall and none during the winter. Furthermore, strong tornadic events are favored near the Great Lakes during spring, and near the confluence of the Hudson and Mohawk Rivers as well as over central and western Pennsylvania during summer. Examination of daily weather maps for each strong tornadic event indicates that summer events can be associated with northwesterly (NW) flow or southwesterly (SW) flow, while spring events are predominantly associated with SW flow. Broadly speaking, summer NW flow events occur in conjunction with broad troughs over the northwestern and northeastern United States flanking a strong ridge over the Plains. The NW flow composite also features a northwest-southeast-oriented upper-level jet streak centered over the eastern Great Lakes, such that the poleward-exit region overlies the northeastern United States. Summer SW flow events occur in conjunction with a broad ridge over the Intermountain West and a positively tilted trough stretching from Hudson Bay to the Great Lakes Region. In this composite, a southwest-northeast-oriented upper-level jet streak is centered near the New England coast, such that the equatorward-entrance region is located over the northeastern United States. Additional results to be presented at the Workshop will emphasize characteristic dynamical and thermodynamical signatures identified in the NW and SW flow composites that are conducive to the outbreak of severe tornadic convection over the Northeast.




 








Forecasting Synoptic and Mesoscale Environments for Tornadoes and Derechos in the Northeast United States

 

Sheryl F. Honikman, Alicia C. Cacciola , Tom J. Garlarneau Jr. and Lance F. Bosart

Department of Earth and Atmospheric Sciences, University at Albany State University of New York, Albany, New York


Kenneth D. LaPenta

Weather Forecast Office, NOAA/NWS, Albany, New York

 

The short lead time associated with issuing a tornado and/or severe thunderstorm warning limits the average forecaster to a quick subjective examination in identifying synoptic and mesoscale conditions associated with tornadoes and severe thunderstorms. This analysis is usually based on assessing a few parameters based on the convective nature of the event. Previous modeling studies of tornadic and severe storms (e.g. Weisman and Klemp 1982,1984; Stensrud etal., 1997) and theoretical studies (e.g. Davies-Jones 1984; Rotunno and Klemp 1985) have shown relationships between low-level vertical wind shear and mid-level mesocyclones in supercell thunderstorms. This talk will focus on comparing the synoptic and mesoscale environments that produce significant tornado outbreaks and derechos in the northeastern sector of the United States. It will be shown that such an analysis allows for the identification of common signatures that can help to distinguish favorable patterns for tornadogenesis and the formation of derechos.

Our analysis is based on eleven tornado outbreaks and ten derecho events for the last thirty years in the northeast. Preliminary results, using the NCEP/NCAR reanalyis grids from 1973-1999, show that the majority of the tornadoes developed in the equatorward exit region of a strong upper-level jet streak. The majority of these cases occurred beneath an upper-level soutwesterly flow. The derechos developed on the leading edge of higher theta-e and ascent regions and formed under an upper-level west or northwesterly flow. All cases featured a combination of synoptic-scale ascent, upper-level divergence and lower-level instability (CAPE >0). Composite analyses that illustrate the life cycle of tornado and derecho events will be presented. We will also show the results from an individual case study on May 31, 1998.





 





Lightning In Tornadic Thunderstorms Over the Northeastern United States

Thomas J. Galarneau, Sheryl F. Honikman, Alicia C. Cacciola, and Lance F. Bosart

Department of Earth and Atmospheric Sciences, The University at Albany State University of New York, Albany, New York

and

Kenneth D. LaPenta, John S. Quinlan, and Glenn E. Wiley 

Weather Forecast Office, NOAA/NWS, Albany, New York

Over the 1993-1998 period, there have been 73 tornado days over New York, New England, New Jersey, and Pennsylvania. Although tornado warning accuracy and lead time has improved greatly (lead times have improved by as much as 20 minutes) with the use of the WSR-88D, tornado warning false alarm rates are still too high (on 31 May 1998, only 13 of the 51 tornado warnings issued were actual tornadoes). The purpose of this talk is to report on an attempt to use cloud-to-ground (CG) lightning data collected by the National Lightning Detection Network (NLDN) to determine if signatures are apparent in the NLDN CG data that could be used to provide additional guidance to forecasters in tornado situations.


A computer program was developed to isolate single tornadic thunderstorms and monitor storm CG lightning temporal and spatial characteristics in relation to tornado touchdown time. Tornado touchdown time is defined as the reported time the tornado touches the ground as noted in storm data. The reported touchdown times can be in error for many reasons.  Archived WSR-88D data as available was used to verify tornado touchdown times. The single tornadic cells were isolated by using the lat/lon of the tornado touchdown as a reference.

Results for the temporal distribution of CG lightning show a weak relative maximum (approximately 15-20 min^-1) at tornado touchdown amidst decreasing frequency. There is also an increase in percent positive (3% to 6%) centered on tornado touchdown. These signals are relatively more robust for the F2/F3 storms.  Results for the spatial distribution of CG lightning show a concentration of activity to the left of the storm track, with a vault and forward/rear flank downdraft signatures seen in some cases. A classic supercell signature was seen in the Mechanicville tornadic thunderstorm of 31 May 1998. In terms of possible forecasting applications, no robust precursor CG lightning signatures were apparent in the tornado cases examined.


 

 




AWIPS Display Strategies for Enhancing the Visualization of Radar Imagery and NWP Model Output Fields

Josh Korotky

Weather Forecast Office, NOAA/NWS, Pittsburgh Pennsylvania



Advanced Weather Interactive Processing System (AWIPS) workstations can be configured to enhance the display characteristics of Weather Surveillance Radar 1988-Doppler (WSR-88D) imagery and NWP model output fields. Given the growing volume of available data, it is becoming increasingly important to extract relevant information effectively. Enhanced visualization of NWP model data allows forecasters to better understand model solutions and guidance; effective visualization of WSR-88D radar products sets the stage for more effective warning operations.


This presentation will illustrate visualization techniques for using AWIPS in severe weather forecast and warning operations. Techniques will illustrate how to use images, colors and contours to create composite fields that reveal important NWP processes. Radar techniques will focus on configuring AWIPS effectively for severe weather, and using color effectively to reveal the important features contained in radar products. Case study examples will be used to illustrate these techniques.

Composites of model output fields will demonstrate the ingredients that promote severe storm development, convective organization, and storm mode. Composite strategies should reveal 1) measures of instability and vertical wind shear, 2) three dimensional moisture distribution and content, and 3) large scale and mesoscale forcing mechanisms.

Effective use of color will reveal the important features contained in WSR-88D radar products. Strategies will be presented for limiting Avisual noise@ with color gradients, while highlighting relevant physical and flow structures. Color schemes will be presented for enhancing WSR- 88D base and derived products, including Base Reflectivity (Z), Base Velocity (V), Storm Relative Mean Radial Velocity Map (SRM), Storm Relative Mean Radial Velocity Region (SRR), Composite Reflectivity (CZ), Vertically Integrated Liquid (VIL), Layer Composite Reflectivity Maximum (LRM), Reflectivity Cross Sections (RCS), and Velocity Cross Sections (VCS).












New Present Weather Imaging Technology for the Operational Forecaster

Mark Beaubien

Yankee Environmental Systems, Inc., Turners Falls, MA

Jeff Freedman

Atmospheric Sciences Research Center, University At Albany State University of New York, Albany, New York


Recent developments in instrumentation have made it possible to make remote ground-based unattended observations of environmental conditions such as present weather. Two sky image processing algorithms are compared to derive winds aloft from sequential images. One algorithm uses a single sky image and a single ceilometer; the second algorithm uses two separate imagers spaced approximately one mile apart. The goal of this work is to develop new approaches for the extraction of accurate environmental parameters such as cloud type and winds for general meteorology and airport safety applications.



 

 

 

 

 

 

 

 


  

 

COOP Station Forecast and Verification Studies

Joseph P. Villani, Lance F. Bosart, and Daniel Keyser

Department of Earth and Atmospheric Sciences, University at Albany State University of New York, Albany, New York

Michael J. Cempa, Thomas A. Wasula, and Stephen R. Pertgen

Weather Forecast Office, NOAA/NWS, Albany, New York

 

This talk will report on statistical procedures to create operational maximum and minimum temperature forecasts for selected Cooperative Observer Program (COOP) sites within the forecast zones of responsibility of the Albany National Weather Service Forecast Office (WFO). This procedure will become necessary as future NWS forecasting improvements will include forecast verification by zone. The operational problem is how to forecast for specific COOP stations, given that existing guidance products are only generally available for selected National Weather Service/Federal Aviation Administration (NWS/FAA) locations. COOP stations exist in each zone and will be used as the main source for maximum and minimum temperature data specific to each zone. Linear regression equations are in the process of being developed and tested for the individual COOP sites based on observed and forecast conditions at Albany. Eventually, forecast regression equations will be developed for maximum and minimum temperature stratified by weather regime and elevation.

Initially we have selected several COOP site locations within the Albany WFO zones for a three-month test period in developing linear regression equations. The COOP sites include Cairo (Greene County), East Jewett (Greene County), and Gloversville (Fulton County), and the three-month period is from 1 May 1999 through 31 July 1999. Eventually, the dataset will be expanded to include temperature data for the entire 1995-1999 period. To date, four different linear regression equations have been developed. The first, using Albany observed maximum temperatures versus the individual COOP sites' maximum temperatures; the second, using Albany observed minimum temperatures versus the individual COOP sites' minimum temperatures; the third, using Albany Model Output Statistics (MOS) maximum temperatures versus the individual COOP sites' maximum temperatures; and the fourth, using Albany MOS minimum temperatures versus the individual COOP sites' minimum temperatures.

Preliminary results have indicated that in general, maximum temperatures have shown a better correlation than the minimum temperatures when comparing the COOP sites to Albany. For example, in comparing Cairo's temperatures with Albany's observed temperatures, the maximum temperatures for the three-month test period correlated very well, with a value of 0.9507 for the square of the correlation coefficient (R2). However, the minimum temperatures yielded a value of only 0.8284 for R2. It is hypothesized that stratification by weather regime and elevation will help account for the variability in the minimum temperatures. Future research will test this hypothesis.


 

 


 

 

 

Evaluation of the SUNYA ReCM for the Winter Climate of Northeastern U.S.

 

Michael Notaro, Wei-Chyung Wang, and Wei Gong

Atmospheric Sciences Research Center, University At Albany State University of New York, Albany, New York

 

The ability of the SUNY A ReCM to reproduce features of the winter climate across the Northeast is evaluated here. SUNY A ReCM, based on PSU/NCAR MM5, is run for November 1998 through March 1999 with updates every 12 hours from the driving fields of TOGA atmospheric and NCEP SST data. The domain is centered on New York with a horizontal grid spacing of 20 km. Previously, this regional model had been evaluated for the summer monsoon over East Asia (Wang et al., 2000; Gong and Wang, 2000).

The model does a reasonable job at capturing synoptic and mesoscale features, although generally its greatest errors are found within the boundary layer. It successfully captures the spatial and temporal patterns of maximum and minimum temperature, including terrain impacts, although the model exhibits a 3-4°C cold bias, most notably at night. Model precipitation tends to be limited, particularly in the western domain, although the occurrence of events is typically captured by the model. A change from Grell to Kuo convective scheme (Grell, 1993; Kuo, 1974) produces more precipitation in one event and generally cooler temperatures, although the average climate for November 1998 using Grell, Kuo, and Kain Fritsch (Kain and Fritsch, 1993) is not very different. A sounding comparison showed that the model is typically too cool and moist near the surface and too warm and dry aloft. When compared against these soundings and NCEP Reanalysis, the model circulation is very good, particularly in the mid- to upper levels. Radiation variables from GCIP (GEWEX Continental-Scale International Project) (Pinker et al.,1999) were also compared to the model output, showing that the model's surface albedo is too low while surface irradiance is too large. This is most notable on the western and eastern sides of the inner domain, where cloud cover is too little, allowing for too much irradiance to reach the surface.

A six-member ensemble was also run for December 1998 with initial dates varying from November 25 to 30. After about ten days into December, the results for the six runs were very similar, indicating that the initial conditions are only important for about one to two weeks. Overall, the SUNY A ReCM captures the major winter climate features across the Northeast, both spatially and temporally. Future work includes applying a different precipitation scheme, comparing snow depth to observations, and updating the boundary conditions every six hours. Once the model is adjusted to produce the best results, it can eventually be coupled to a general circulation model.


 


 



 

 

 

A Dynamical-Statistical Atmospheric Modeling System Designed To Simulate Wind Generated Power Production Rates


John W. Zack *, Joseph J Nocera, Kenneth T. Waight, Glenn E. Van Knowe, and Mary D. Bousquet

MESO, Inc., Troy, New York


The generation of electrical power from the energy of the wind has been the fastest growing source of energy during the last ten years. This has been a result of: (1) the continuing efforts to reduce the emission of greenhouse gases from traditional electric generation plants fueled by oil, coal or natural gas; (2) major changes in the economic and organizational structure of the electric utility industry which allow customers a choice in the supplier of their electricity and how it is generated; (3) improvements in wind power technology that have reduced the cost of generation to a level that is competitive with the traditional methods of generating electricity; and (4) the desire to minimize the dependence on external economic or political events which affect the cost and availability of fuel as recent events have demonstrated. As a result, the total U.S. wind-generated power capacity has more than doubled during the last 2 years.

In order to achieve optimal wind power generation and use, the wind power industry requires two types of meteorological information. (1) High-resolution spatial maps of the wind climatology to assist in the optimally siting of wind power generation facilities; reduce the time needed to site a plant; and provide a reliable method of estimating the future plant output and (2) short-term (0-48 hr) forecasts of the hourly wind-power output. These forecasts are needed to improve scheduling and dispatching of generation and transmission resources, to optimize wind energy buying and selling strategies in deregulated markets and to increase the perception of wind energy as a reliable source of power within electric power-producing organizations.

Both the wind mapping and wind forecasting have obstacles making it difficult to meet customer requirements. There are three significant obstacles to reliable wind mapping in most areas of the world, including the U.S. First, the observational density required to create reliable maps in the detail required for the wind maps is not available. Second, the data from observational sites are often unrepresentative of the actual wind turbine site. Finally, in many parts of the world the observational data is of questionable quality. The 0-48 hour forecasts of wind power generation need to be within 10-20% accuracy. This requirement in turn demands highly accurate localized forecasts of the wind speed and generation at the wind turbine height (usually about 50 m above the surface) not typically available.

In order to address these needs of the wind power industry, MESO, Inc., under contract to TrueWind Solutions, LLC is currently generation windmaps and wind energy forecasts using a 3-D mesoscale model called the Mesoscale Atmospheric Simulation System (MASS) and has developed a unique dynamical-statistical modeling system designed for wind simulation called FOREWIND. There are three components to the FOREWIND modeling system: (1) a numerical-dynamical model, (2) an adaptive statistical model, and (3) a wind plant power generation model. The FOREWIND numerical model is a high-resolution boundary-layer model that is designed to run on a fine horizontal grid (< 5 km). The adaptive statistical model generates site-specific wind forecasts based upon statistical relationships between the dynamical model output and observational data at a site. The wind plant power generation model converts the wind information produced by the other two components of the FOREWIND modeling system to estimate the hourly generation of power.

FOREWIND is designed to simulate the lowest 2-3 km of the atmosphere over a limited horizontal domain with a horizontal resolution between 1 and 5 km. The dynamical model includes a Turbulent Kinetic Energy (TKE) boundary layer submodel for the simulation of atmospheric boundary layer processes. The boundary layer submodel includes a full surface energy budget and a detailed representation of surface properties (roughness, etc.) However, the initial version of the model contains no representation of moist convection and no condensation physics. An external source of data from a larger scale model must be used to supply boundary conditions at lateral and top boundaries of model.

The FOREWIND modeling system is designed to operate in two modes: a climatological mode and a forecast mode. In the forecast mode the model utilizes output from an external forecast model as a first guess for its initialization and for the specification of the lateral and top boundary conditions. The external model could be one of the operational models run at the National Centers for Environmental Prediction (NCEP), such as the Eta or AVN models, or a mesoscale model such as MM5 or MESO's MASS model run on a workstation at a local site. FOREWIND also has the capability to ingest a variety of raw data types such as rawinsonde, METAR-format surface or wind profiler data to update the first-guess initialization dataset.

In the climatological mode, the model utilizes a grid point analysis of observational data for both the initialization and boundary condition specification. The modeling system has the capability to ingest external analysis datasets such as those in the NCAR/NCEP reanalysis database, or to generate its own grid point analysis from raw observational data. Climatological wind maps are constructed by running a large sample of historical cases with FOREWIND and then blending the simulated data with the quality-controlled observational data to construct the final wind climatology.

At the 1999 Northeast Regional Operational Workshop, early results of the FOREWIND wind mapping and wind forecasting for the northeast U. S.were presented. The primary factors identified limiting the performance of the FOREWIND system were identified as: (1) The need to compensate for the difference in model versus actual elevation (resolution). The largest errors were found to be at sites with the largest discrepancy between the model elevation and actual elevation. (2) Improving the specification of roughness height. Currently roughness height is tied to land use/cover class. There is considerable variation within land use classes such as "mixed woodland" so it is hard to estimate the correct roughness height in all cases. Research had top be done to help compensate for this problem. (3) Boundary condition update frequency. The Eta is available at only 6-hr intervals in real-time. Running a local mesoscale model over a large area for more frequent updates of the boundary conditions was looked at as a possible solution. (4) Local-scale site characteristics on the order of 10-500 m will not be captured in the model so very localized wind effects may not be simulated correctly. A post-run statistical adjustment (i.e. FOREWIND-MOS) has been created to fully address this problem.

During the past year since the last workshop, the FORWIND system has been tested in many areas around the world ranging from the polar to equatorial regions. The system is now in full production producing wind products for a world wide array of customers include several in the Northeast U. S. and southern Canada. The presentation at the 2000 workshop will discuss the solutions to the four limiting factors as well as update the performance of the FOREWIND system as compared to a full 3-D mesoscale modeling system.