Presentation is loading. Please wait.

Presentation is loading. Please wait.

A Real-Time Automated Method to Determine Forecast Confidence Associated with Tornado Warnings Using Spring 2008 NWS Tornado Warnings John Cintineo Cornell.

Published byClifton Golden Modified over 6 years ago

Similar presentations

Presentation on theme: "A Real-Time Automated Method to Determine Forecast Confidence Associated with Tornado Warnings Using Spring 2008 NWS Tornado Warnings John Cintineo Cornell."— Presentation transcript:

1 A Real-Time Automated Method to Determine Forecast Confidence Associated with Tornado Warnings Using Spring 2008 NWS Tornado Warnings John Cintineo Cornell University Travis Smith Valliappa Lakshmanan Kiel Ortega NOAA - NSSL 20 September 2018

2 Background Warning Decision Support System – Integrated Information (WDSS-II) Uses merged, multi-sensor CONUS radar network combines model, lightning, and GOES satellite data Short-term severe weather forecasting products Objective: To examine how WDSS-II products can be used as predictors for issuing NWS tornado warnings. Assign objective probabilities to warnings based on varying the attribute threshold. 20 September 2018

3 Radar-derived products Storm Environment Data
Maximum Expected Size of Hail (MESH) Probability of Severe Hail (POSH) Severe Hail Index (SHI) Vertically Integrated Liquid (VIL) Area of VIL +30 Echo Tops of 50, 30, & 18dBZ 3-6 km & 0-2 km Azimuthal Shear Lowest level max dBZ Reflectivity at 0C, -10C, & -20C Overall max reflectivity Height of 50dBZ above 253K isotherm Environmental Shear Storm Relative Flow 9-11km AGL Storm Relative Helicity 0-3km CAPE, CIN LCL min height SATELLITE: IR band-4 min temp. (cloud tops) Total of 23 products 20 September 2018

4 Methodology: Investigated archived NWS spring 2008 CONUS tornado warnings with WDSS-II radar-derived products Each storm attribute maximum (or minimum) values computed every 1 minute of the warning Compared attribute values from the issuance of the warning (initial values) and the expiration of the warning (lifetime max/min). Composite time series of each attribute Warnings broken down by verified vs. unverified Verification data obtained from the Storm Prediction Center’s storm data (preliminary). storm environment data provided by 20-km RUC model. 20 September 2018

5 Dataset: 2 May – 1 July for 0-2 km Azimuthal shear, VIL, Area of VIL +30, and reflectivity products 15 May to 1 July for 3-6 km Azimuthal shear 20 random days for Storm Environment attributes NB: for 1 May – 10 May 0-2 km Azimuthal shear is replaced by 0-3km Azimuthal shear 1,617 Tornado Warnings Frequency of Hits = (414 verified warnings) False Alarm Ratio = (1,203 unverified warnings) Average Warning Duration: 38.6 mins 20 September 2018

6 Initial 0-2 km Azimuthal Shear
UNVERIFIED VERIFIED Mean: s^ Mean: s^-1 SD: s^ SD: s^-1 20 September 2018

7 Lifetime Max 0-2km Azimuthal Shear
UNVERIFIED VERIFIED Mean: s^ Mean: s^-1 SD: s^ SD: s^-1 20 September 2018

8 20 September 2018

9 Probability that a warning verified, given an initial 0-2 km Az. Shear
20 September 2018

10 Initial Vertically Integrated Liquid (VIL)
UNVERIFIED VERIFIED Mean: kg/m^2 Mean: kg/m^2 SD: kg/m^2 SD: kg/m^2 20 September 2018

11 Lifetime Maximum VIL UNVERIFIED VERIFIED
Mean: kg/m^2 Mean: kg/m^2 SD: kg/m^2 SD: kg/m^2 20 September 2018

12 20 September 2018

13 Probability that a warning verified, given an initial Vertically Integrated Liquid
20 September 2018

14 Initial 0-2 km Az. Shear (s^-1)
Initial Vertically Integrated Liquid (kg/m^2) y >= > y >= > y >= 20 y < 20 CONDITIONAL PROBABILITY CONTINGENCY TABLE x < <= x < <= x < x >= 0.012 Ver: 41 Unv: 238 PROB: 0.147 Ver: 10 Unv: 61 PROB: 0.141 Ver: 5 Unv: 34 PROB: 0.128 Unv: 19 PROB: 0.208 Ver: 29 Unv: 70 PROB: 0.293 Ver: 32 Unv: 49 PROB: 0.395 Ver: 18 Unv: 42 PROB: 0.300 Ver: 24 Unv: 26 PROB: 0.480 Ver: 15 Unv: 41 PROB: 0.268 Ver: 23 Unv: 69 PROB: 0.250 Ver: 26 Unv: 35 PROB: 0.426 Ver: 25 Unv: 27 PROB: 0.481 Ver: 0 Unv: 14 PROB: 0.000 Ver: 9 Unv: 36 PROB: 0.200 Unv: 15 PROB: 0.400 Ver: 8 Unv: 9 PROB: 0.471 20 September 2018

15 Summary Provide warning guidance for the NWS
Once a NWS tornado warning is issued, WDSS-II can automatically assign a probability that it will verify, in real-time More years of warning data will lead to a better climatology of warning probabilities With more warning data, create a contingency table based on 3 or 4 of the best predictors Forecasters can use such probability data to reduce their FAR 20 September 2018

16 Future avenues of research
Extend the data set to include past springs Examine environment just outside the warning polygons (to capture the entire storm) Compare spring and fall tornado warnings Compare attributes in tornado and severe T-storm warnings Compare warning data based on region Investigate warnings issued in watches, and those outside of watches 20 September 2018

17 Summary Provide warning guidance for the NWS
Once a NWS tornado warning is issued, WDSS-II can automatically assign a probability that it will verify, in real-time More years of warning data will lead to a better climatology of warning probabilities With more warning data, create a contingency table based on 3 or 4 of the best predictors Forecasters can use such probability data to reduce their FAR 20 September 2018

18 Acknowledgements Travis Smith Lak Kiel Ortega Owen Shieh
This research was supported by an appointment to the National Oceanic and Atmospheric Administration Research Participation Program through a grant award to Oak Ridge Institute for Science and Education. 20 September 2018

19 References Erickson, S. A., Brooks, H., 2006: Lead time and time under tornado warnings: rd Conference on Severe Local Storms Guillot, E., T. M. Smith, Lakshmanan, V., Elmore, K. L., Burgess, D. W., Stumpf, G. J., 2007: Tornado and Severe Thunderstorm Warning Forecast Skill and its Relationship to Storm Type. Lakshmanan, V., T. M. Smith, K. Cooper, J. J. Levit, G. J. Stumpf, and D. R. Bright, 2006: High- resolution radar data and products over the Continental United States. 22nd Conference on Interactive Information Processing Systems, Atlanta, Amer. Meteor. Soc. Lakshmanan, V., T. Smith, K. Hondl, G. J. Stumpf, and A. Witt, 2006: A real-time, three dimensional, rapidly updating, heterogeneous radar merger technique for reflectivity, velocity and derived products. Weather and Forecasting 21, Lakshmanan, V., T. Smith, G. J. Stumpf, and K. Hondl, 2007: The warning decision support system - integrated information (WDSS-II). Weather and Forecasting 22, Ortega, K. L, and T. M. Smith, 2006: Verification of multi-sensor, multi-radar hail diagnosis techniques. 1st Severe Local Storms Special Symposium, Atlanta, GA, Amer. Meteo. Soc. Ortega, K. L., T. M. Smith, G. J. Stumpf, J. Hocker, and L. López, 2005: A comparison of multi- sensor hail diagnosis techniques. 21st Conference on Interactive Information Processing Systems (IIPS) for Meteorology, Oceanography, and Hydrology, Amer. Meteo. Soc., P CD preprints. Witt, A., Eilts, M., Stumpf, G. J., Johnson, J. T., Mitchell, D. E., Thomas, K. W., 1998: An Enhanced Hail Detection Algorithm for the WSR-88D. 20 September 2018

20 20 September 2018

21 Initial 3-6 km Azimuthal Shear
UNVERIFIED VERIFIED Mean: s^ Mean: s^-1 SD: s^ SD: s^-1 20 September 2018

22 Lifetime Max 3-6 km Azimuthal Shear
UNVERIFIED VERIFIED Mean: s^ Mean: s^-1 SD: s^ SD: s^-1 20 September 2018

23 20 September 2018

24 Probability that a warning verified, given an initial 3-6 km Az. Shear
20 September 2018

25 Initial Max LL Reflectivity
Mean: dBZ Mean: dBZ SD: dBZ SD: dBZ 20 September 2018

26 Lifetime Max LL Reflectivity
Mean: dBZ Mean: dBZ SD: dBZ SD: 7.11 dBZ 20 September 2018

27 20 September 2018

28 Initial dBZ @ -20C Mean: 46.58 dBZ Mean: 53.08 dBZ
SD: dBZ SD: dBZ 20 September 2018

29 Lifetime Max dBZ @ 20C Mean: 52.75 dBZ Mean: 57.86 dBZ
SD: dBZ SD: dBZ 20 September 2018

30 20 September 2018

31 Probability a warning verified, given a certain Az. shear
0-2km Az. Shear km Az. Shear 20 September 2018

Download ppt "A Real-Time Automated Method to Determine Forecast Confidence Associated with Tornado Warnings Using Spring 2008 NWS Tornado Warnings John Cintineo Cornell."

Similar presentations

Multiple Sensor Precipitation Estimation over Complex Terrain AGENDA I. Paperwork A. Committee member signatures B. Advisory conference requirements II.

Multiple Sensor Precipitation Estimation over Complex Terrain AGENDA I. Paperwork A. Committee member signatures B. Advisory conference requirements II.
Build 10 Tornado Detection Algorithm The Build 9 Tornado Vortex Signature ( TVS ) Detection Algorithm was not very robust, and was designed to be a place.

Build 10 Tornado Detection Algorithm The Build 9 Tornado Vortex Signature ( TVS ) Detection Algorithm was not very robust, and was designed to be a place.
SNPP VIIRS green vegetation fraction products and application in numerical weather prediction Zhangyan Jiang 1,2, Weizhong Zheng 3,4, Junchang Ju 1,2,

SNPP VIIRS green vegetation fraction products and application in numerical weather prediction Zhangyan Jiang 1,2, Weizhong Zheng 3,4, Junchang Ju 1,2,
Radar Climatology of Tornadoes in High Shear, Low CAPE Environments in the Mid-Atlantic and Southeast Jason Davis Matthew Parker North Carolina State University.

Radar Climatology of Tornadoes in High Shear, Low CAPE Environments in the Mid-Atlantic and Southeast Jason Davis Matthew Parker North Carolina State University.
5 th International Conference of Mesoscale Meteor. And Typhoons, Boulder, CO 31 October 2006 National Scale Probabilistic Storm Forecasting for Aviation.

5 th International Conference of Mesoscale Meteor. And Typhoons, Boulder, CO 31 October 2006 National Scale Probabilistic Storm Forecasting for Aviation.
WSR-88D Signatures Associated with One Inch Hail in the Southern Plains Dennis E. Cavanaugh Jessica A. Schultz National Weather Service Weather Forecast.

WSR-88D Signatures Associated with One Inch Hail in the Southern Plains Dennis E. Cavanaugh Jessica A. Schultz National Weather Service Weather Forecast.
ounding nalog etrieval ystem Ryan Jewell Storm Prediction Center Norman, OK SARS Sounding Analog Retrieval System.

ounding nalog etrieval ystem Ryan Jewell Storm Prediction Center Norman, OK SARS Sounding Analog Retrieval System.
TRMM Tropical Rainfall Measurement (Mission). Why TRMM? n Tropical Rainfall Measuring Mission (TRMM) is a joint US-Japan study initiated in 1997 to study.

TRMM Tropical Rainfall Measurement (Mission). Why TRMM? n Tropical Rainfall Measuring Mission (TRMM) is a joint US-Japan study initiated in 1997 to study.
A Spatial Climatology of Convection in the Northeast U.S. John Murray and Brian A. Colle National Weather Service, WFO New York NY Stony Brook University,

A Spatial Climatology of Convection in the Northeast U.S. John Murray and Brian A. Colle National Weather Service, WFO New York NY Stony Brook University,
1 GOES Users’ Conference October 1, 2002 GOES Users’ Conference October 1, 2002 John (Jack) J. Kelly, Jr. National Weather Service Infusion of Satellite.

1 GOES Users’ Conference October 1, 2002 GOES Users’ Conference October 1, 2002 John (Jack) J. Kelly, Jr. National Weather Service Infusion of Satellite.
A Study on Convective Modes Associated with Tornadoes in Central New York and Northeast Pennsylvania Timothy W. Humphrey 1 Michael Evans 2 1 Department.

A Study on Convective Modes Associated with Tornadoes in Central New York and Northeast Pennsylvania Timothy W. Humphrey 1 Michael Evans 2 1 Department.
Preparing for a “Change” in Severe Hail Warning Criteria in 2010 Brian J. Frugis NWS WFO Albany, NY NROW XI November 4-5, 2009.

Preparing for a “Change” in Severe Hail Warning Criteria in 2010 Brian J. Frugis NWS WFO Albany, NY NROW XI November 4-5, 2009.
Tornado Detection Algorithm (TDA) By: Jeffrey Curtis and Jessica McLaughlin.

Tornado Detection Algorithm (TDA) By: Jeffrey Curtis and Jessica McLaughlin.
RECENT PROGRESS IN CONVECTIVE PHENOMENA MONITORING AND FORECASTING AT THE INM F. Martín, F. Elizaga, I. San Ambrosio and J. M. Fernández Servicio de Técnicas.

RECENT PROGRESS IN CONVECTIVE PHENOMENA MONITORING AND FORECASTING AT THE INM F. Martín, F. Elizaga, I. San Ambrosio and J. M. Fernández Servicio de Técnicas.
Using WSR-88D Reflectivity for the Prediction of Cloud-to- Ground Lightning Nowcasting a storm’s first strike Brandon Vincent NWS-Newport Larry Carey Texas.

Using WSR-88D Reflectivity for the Prediction of Cloud-to- Ground Lightning Nowcasting a storm’s first strike Brandon Vincent NWS-Newport Larry Carey Texas.
Travis Smith (OU/CIMMS) February 25–27, 2015 National Weather Center

Travis Smith (OU/CIMMS) February 25–27, 2015 National Weather Center
A Spatial Climatology of Convection in the Northeast U.S. John Murray and Brian A. Colle Stony Brook University Northeast Regional Operational Workshop.

A Spatial Climatology of Convection in the Northeast U.S. John Murray and Brian A. Colle Stony Brook University Northeast Regional Operational Workshop.
Local Severe Weather Climatologies for WSR-88D Radar Areas across the United States John A. Hart NOAA/NWS/NCEP Storm Prediction Center Norman, Oklahoma.

Local Severe Weather Climatologies for WSR-88D Radar Areas across the United States John A. Hart NOAA/NWS/NCEP Storm Prediction Center Norman, Oklahoma.
The National Severe Storms Laboratory Jeff Kimpel, Director NSSL NOAA Science Advisory Board Norman, Oklahoma November 5, 2002.

The National Severe Storms Laboratory Jeff Kimpel, Director NSSL NOAA Science Advisory Board Norman, Oklahoma November 5, 2002.
Federal Department of Home Affairs FDHA Federal Office of Meteorology and Climatology MeteoSwiss Operational thunderstorm nowcasting in the Alpine region.

Federal Department of Home Affairs FDHA Federal Office of Meteorology and Climatology MeteoSwiss Operational thunderstorm nowcasting in the Alpine region.

Similar presentations

Ads by Google