The paper below discusses the lightning program at National Weather Service Pueblo. Thank you for visiting.
The following paper was presented at the 20th Local Severe Storms Conference in Orlando Florida, September 2000.
THE LIGHTNING PROGRAM AT NATIONAL WEATHER SERVICE PUEBLO
The lightning program at National Weather Service (NWS) Pueblo is a three tiered program which incorporates 1.) The operational use of Cloud to Ground (CG) lightning activity, 2.) Applied lightning research, and 3.) Lighting safety. This paper will concentrate on the operational use of CG lightning activity, mainly with respect to severe storms forecasting. Applied lightning research will briefly be discussed. All three aspects of the lightning program will be presented in detail during the poster session.
2. USE OF CG ACTIVITY TO ASSIST IN SHORT TERM SEVERE STORM FORECASTING
The lightning program was instituted to observe if real time CG lightning activity could assist forecast and warning operations across the county warning and forecast area (CWFA). NWS Pueblo's area of responsibility covers a large, diverse geographical region of southern and southeastern Colorado, ranging from the plains of southeastern Colorado to the high mountainous regions of central and southern Colorado. The primary tool for convective storm discernment used by the NWS Pueblo staff is the KPUX WSR-88D, which is located 60 km southeast of Colorado Springs, Colorado. The umbrella of coverage of KPUX encompasses a large majority of the CWFA, however, low level base data is severely restricted over most of the western half due to beam blockage by the higher terrain.
With the advent of the Advanced Weather Interactive Processing System (AWIPS) within the NWS, real time lightning data, along with meteorological variables associated with the convective process, are available to the operational forecaster. The CG lighting data can be displayed in a variety of temporal resolutions, including 1 hour increments, 15 minute increments, 5 minute increments and a 1 minute lightning sequence. Parameters associated with the convective process are available on an hourly basis, including, but not limited to, CAPE, CIN, LI, helicity, and advection of moisture and temperature fields. These variables are updated on an hourly basis and are processed by the Local Analysis and Prediction System (LAPS, McGinley, 1989). All the data above can be time lapsed and overlaid with a variety of other meteorological data.
Unlike radar data, CG activity detected by the National Lightning Detection Network (NLDN) does not suffer detection deficiencies over diverse terrain, and thus can be used as a proxy for radar data. Having real-time knowledge of the spatial and temporal resolution of this CG activity, and access to meteorological parameters associated with convective processes, the forecaster now has the capability to better comprehend the convective threat in geographical diverse regions.
3. EXAMPLES USING CG DATA IN SEVERE LOCAL STORMS WARNING OPERATIONS
3.1 17 June 1999
During the early afternoon of 17 June, AWIPS was indicating a large amount of CG activity (~1500 flashes per hour) over the San Juan mountains of southwest Colorado. This is believed to be a very high amount of lighting activity for an one hour time period for the state of Colorado (see section 4). This lightning was associated with a short wave entering the state of Colorado from the Four Corners region. Based on the lightning information on AWIPS, the following short term forecast (NOW) was issued:
LIGHTNING DETECTION EQUIPMENT AT THE NATIONAL WEATHER SERVICE WAS INDICATING AN EXCESSIVE AMOUNT OF CLOUD TO GROUND LIGHTNING MOVING TOWARDS THE EASTERN SAN JUAN MOUNTAINS. BETWEEN 1 AND 2 PM THIS AFTERNOON...OVER 1500 LIGHTNING STRIKES OCCURRED IN THE WESTERN SAN JUAN MOUNTAINS. WITH THE ATMOSPHERE CONTINUING TO DESTABILIZE...THE LIGHTNING ACTIVITY WILL CONTINUE TO INCREASE. RESIDENTS AND VISITORS IN THE SAN JUAN MOUNTAINS SHOULD SEEK SHELTER FROM THESE DANGEROUS LIGHTNING STORMS.
As time progressed, the amount of CG activity continued to remain very high with these storms. In addition to the lightning threat in the mountains, it was becoming apparent that the severe weather threat was increasing for the San Luis Valley (SLV), located to the east of the San Juan mountains. LAPS analysis over the valley was indicating CAPE values were in excess of 1000 J/Kg, which is significant as the SLV elevation is generally 2.2 km above sea level. Coordination between NWS Grand Junction and NWS Pueblo indicated small hail (< 2.54 cm) was accumulating with storms in the San Juan mountains. Based on the high observed CAPE values, hail accumulations and high CG activity, the forecaster believed the storms would intensify and become severe as they entered the SLV, and the following NOW was issued for the valley region:
STRONG THUNDERSTORMS WITH SIGNIFICANT AMOUNTS OF HAIL WILL BE POSSIBLE OVER THE SAN LUIS VALLEY LATER THIS AFTERNOON. AT 3 PM THIS AFTERNOON...DOPPLER RADAR AND LIGHTNING DATA WAS INDICATING A RATHER INTENSE LINE OF THUNDERSTORMS MOVING ACROSS THE SAN JUAN MOUNTAINS. WHEN THESE STORMS MOVE INTO THE SAN LUIS VALLEY...THEY WILL ENCOUNTER AN UNSTABLE ATMOSPHERE AND LIKELY INTENSIFY...AND HAVE THE CAPABILITY TO BECOME SEVERE AND PRODUCE A SIGNIFICANT AMOUNT OF HAIL. THE STORMS SHOULD MOVE INTO THE VALLEY AFTER 4 PM.
The thunderstorms did intensify as they moved into the SLV around 4 p.m., and a warning was issued. The storms did produce large hail (2.54 - 3.81 cm) as they moved across the valley between 4:20 and 4:30 p.m.
3.2 7 October 1999
On 7 October 1999, a late season severe thunderstorm producing a considerable amount of marginally severe hail affected the Colorado Springs area. Similar to example 3.1, this storm also began as an increasing area of CG lightning activity in the Pikes Peak mountain region, located southwest of the Colorado Springs area. LAPS CAPE analysis indicated the small cluster of storms would be moving into an area of higher CAPE, and the following short term forecast was issued for the Colorado Springs/Pikes Peak region at 1:25 p.m.:
THUNDERSTORMS...SOME CAPABLE OF PRODUCING HAIL UP TO 1/2 INCH IN DIAMETER...WIND GUSTS TO 50 MPH, AND DEADLY LIGHTNING AND BRIEF HEAVY RAINFALL ARE EXPECTED TO MOVE ACROSS PORTIONS OF EL PASO COUNTY...TELLER AND PUEBLO AND EASTERN FREMONT COUNTIES THROUGH 245 PM MDT. AT 1:24 PM MDT PUEBLO DOPPLER RADAR INDICATED STRONG THUNDERSTORMS 5 MILES SOUTH OF WOODLAND PARK...2 MILES WEST OF FALCON AND 30 MILES SOUTHEAST OF PUEBLO. THESE STORMS WERE MOVING GENERALLY TO THE NORTH AT 10 MPH.
The storms continued to intensify as they moved into the higher CAPE air over the Colorado Springs area. Based on this intensification (VILS reaching 40 G/Kg, 50 dBz echoes reaching 9 km, and radar cross sections indicating a weak echo region), a severe thunderstorm warning was issued at 2:03 p.m. for El Paso county. At 2:15 p.m., HAM radio reports of severe hail in the Colorado Springs area were received. The storm remained severe for about 10 minutes as it moved across the Colorado Springs area.
3.3 30 June 1999
The use of CG lightning data to assist in severe thunderstorm development is not restricted necessarily to the mountainous regions. This example will show how lighting assisted in severe thunderstorm detection over the far eastern Plains of Colorado.
During the late afternoon and evening of 30 June1999, a dryline was located over southeastern Colorado. LAPS analysis was indicating over 3000 J/Kg of CAPE along and to the east of the dryline. To the west of this boundary, the atmosphere was more stable, however a few weak thunderstorms persisted in the airmass. These weak thunderstorms were typically short lived, but would leave outflow boundaries as they dissipated. As the evening progressed, two of these boundaries collided near the dryline.
In addition, other weak shallow thunderstorms were noted moving east-southeast across northeast Colorado. Although this convection was shallow (and did not show colder cloud tops in the IR imagery), NLDN was detecting persistent CG activity with these storms. The airmass was rather stable over this region, but a weak short wave was believed to moving across northeastern Colorado. This disturbance was strong enough to force convection, but since the airmass was stable, the convection remained shallow. Although these thunderstorms in northeast Colorado did not directly affect the weather in the NWS Pueblo CWFA, it made the forecaster on duty aware that a weak upper level weather system was likely moving towards the region, which could assist in deeper convective development in the vicinity of the dryline later that evening.
Although weak convective activity was occurring over the region, it was unknown if the cap would be too strong to allow for deep convection. Based on these events, the following NOW was issued at 7:15 p.m.:
ISOLATED THUNDERSTORMS WILL BE POSSIBLE OVER KIOWA COUNTY THROUGH 8 PM THIS EVENING. DOPPLER RADAR DATA INDICATED TWO OUTFLOW BOUNDARIES COLLIDED OVER KIOWA COUNTY AT 7 PM. THIS COLLISION MIGHT BE ENOUGH TO INITIATE THUNDERSTORMS OVER THE AREA. THE ATMOSPHERE OVER KIOWA COUNTY IS VERY UNSTABLE AND THERE IS ADEQUATE SHEAR IN THE ATMOSPHERE TO ALLOW THE STORMS TO INTENSIFY AND BECOME SEVERE. HOWEVER...A WARM LAYER OF AIR OVER THE REGION MIGHT BE ENOUGH TO SUPPRESS THUNDERSTORM ACTIVITY. RESIDENTS ACROSS KIOWA COUNTY SHOULD BE ALERT TO RAPIDLY CHANGING WEATHER CONDITIONS.
Thunderstorm initiation did occur due to the collision, and became better organized. With the persistent convective activity moving east towards the dryline in eastern Kiowa county, a second short term forecast was issued at 8:54 p.m.:
A STRONG THUNDERSTORM WILL MOVE ACROSS NORTHEASTERN KIOWA COUNTY THROUGH 925 PM. THIS STORM IS MOVING INTO AN UNSTABLE ATMOSPHERE AND HAS THE POTENTIAL TO BECOME SEVERE...AND PRODUCE LARGE DAMAGING HAIL AND STRONG WINDS. IF YOU LIVE IN NORTHEAST KIOWA COUNTY...BE PREPARED FOR RAPIDLY CHANGING WEATHER CONDITIONS.
This storm did intensify to severe limits based on radar signatures and a warning was issued for the cell at 9:08 p.m. The storm did produce large hail in eastern Kiowa county at 9:17 and 9:18 p.m. (4.5 cm hail).
4. APPLIED RESEARCH
During the month of August 1999, the staff at NWS Pueblo assisted in documenting CG lightning data for a study of cloud to ground lightning over the state of Colorado. A procedure was created on the AWIPS workstation which saved 32 frames of hourly CG lightning data, with the data being plotted over the state of Colorado and immediate adjacent areas. The amount of hourly CG data, both positive and negative, was then logged.
The primary reason for this study was to observe the amount and variability of CG lightning over the state of Colorado. A few days had missing data.
A total of 143,499 negative CG flashes and 10,103 positive flashes were documented over the state of Colorado, for a total of 153,602 CG flashes. This breaks down to a percentage of 93% negative flashes and 7% positive flashes. The maximum hourly flash rate occurred on 11 August during the 1 hour period ending at 01 UTC (10 August, 7 p.m. MDT) when a total of 3,038 flashes occurred across the state.
On average, just under 5000 flashes (4990) occurred on any given day. However, some days were much more lightning active than others. Lopez and Holle (1986) also observed this day to day pattern of lighting activity over parts of northeast Colorado.
In order to observe if the total amount of lightning was related to significant weather activity over the region, the total amount of CG activity for each day was compared to the number of significant weather events for that corresponding day (Fig 1). A significant weather event is defined as any "event" which was logged in STORM DATA (NOAA 1999) for that date. An event can be either hydrological in nature ("flood") or a severe weather report.
Maximum in lightning activity did correspond to significant weather events on quite a few occasions during the month of August. This relationship occurred on the 4th, 5th, 10th, 17th and during the end of the month. The 10th particularly stands out as a maximum in lightning activity and significant weather reports. However, there were a few active lighting days in which not many significant weather reports occurred.
Figure 1. Comparison between the daily amount of cloud to ground lightning
(/100), severe weather reports, and significant hydrological events ("flood")
during the month of August 1999 across the state of Colorado.
This paper examined how CG lightning data, in conjunction with other data sources, can be used operationally to anticipate severe convection across the NWS PUB CWFA. The combination of new convective forecast parameters on the AWIPS platform, CG lightning activity from NLDN, and a high situational awareness of the environment can allow the forecaster to anticipate the threat of strong to severe convection across both the diverse terrain over the western sections of the forecast area, and over the plains located to the east.
Additionally, a brief 1 month lightning study was also examined. It was observed that on days when lighting activity was high, their was a corresponding increase in significant weather events.
More information about the lightning program at NWS Pueblo can be found the Colorado Lightning Resource Page.
NOAA, 1999. STORM DATA. National Climatic Data Center, Asheville, NC.
McGinley, J. A., 1989. The Local Analysis and Prediction System. Preprints, 12th conf. on Weather Analysis and Forecasting, Monterey, CA. Amer. Meteor. Soc., 15-20