1 / 21

Hazardous Cell Tracking

Hazardous Cell Tracking. Robert Boldi 29 September 2008 NSSTC/UAH. What is Hazardous Cell Tracking?. Hazardous Severe Lightning, Winds, Hail Tornadoes Cell Compact region embedded within a larger-scale convective system Tracking Compute time-rate-of-change of Cell’s attributes. .

ima
Download Presentation

Hazardous Cell Tracking

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Hazardous Cell Tracking Robert Boldi 29 September 2008 NSSTC/UAH R. Boldi NSSTC/UAH

  2. What is Hazardous Cell Tracking? • Hazardous • Severe Lightning, Winds, Hail • Tornadoes • Cell • Compact region embedded within a larger-scale convective system • Tracking • Compute time-rate-of-change of Cell’s attributes. R. Boldi NSSTC/UAH

  3. Methodology • Construct an “Interest Image”, each pixel of which represents the likelihood that a hazardous cell is present [ 0 .. 1 ] • Track this image using Multi-Scale XCT • Locate “Cells” by thresholding this image • Construct Lagrangian history of properties of the cells • Predict future location of cells ( and properties ) R. Boldi NSSTC/UAH

  4. Constructing “Interest” Image • Functional Template Correlation • A template (small image), convolved against a larger input image, producing an “interest image”, each pixel of which is the probability that the feature represented by the template is present at the given pixel. • Templates map onto simple features ( lines, circular cells, low variance regions, etc. • Fuzzy Logic • Groups of “Images” are combined to produce meteorologically-understandable phenomena, e.g. Stratiform, Convective, Decaying, Growing … • These phenomena are combined to produce the final “Interest Image” R. Boldi NSSTC/UAH

  5. Cell Detection and Tracking Lightning Cell Identification Cell-Scale Motions Interest Image Generator Multi-Scale XCT Tracker Satellite Data Storm-Scale Motions FTC  WxTypes Fuzzy LogicFusion Radar Data R. Boldi NSSTC/UAH

  6. Cell Detection and Tracking Lightning Events Interest Image 2008_04_08 @ 01:48Z Radar Data R. Boldi NSSTC/UAH

  7. Cell Detection R. Boldi NSSTC/UAH

  8. Multi-Scale Tracking Envelope Motion Interest Image 2008_04_08 @ 01:48Z Cell Motion R. Boldi NSSTC/UAH

  9. Multi-Scale Tracking R. Boldi NSSTC/UAH

  10. Cell Identification = Thresholding Interest Image Interest Image 2008_04_08 @ 01:48Z Cell ID’s 2008_04_08 @ 01:48Z R. Boldi NSSTC/UAH

  11. Cell ID’s NOT persistent ( movie ) R. Boldi NSSTC/UAH

  12. Lagrangian History Re-Construction ….. ….. T(-2 dt) T(- dt) Cell 1 History T(0) Multi-Scale Vectors Cell 2 History Current Cell Map T(0) Past Data Advector Cell Integrator Cell 3 History ….. ….. : T(-2 dt) T(- dt) T(0) • Input Fields • Lightning • Radar • Satellite R. Boldi NSSTC/UAH

  13. Advector: Forward Advection of Inputs to Current Time In the absence of convergence, divergence, growth or decay, the alignment would be perfect.  = Past Time Advected to Current Time Current Time Past Time R. Boldi NSSTC/UAH

  14. Advector: Forward Advection of Inputs to Current Time R. Boldi NSSTC/UAH

  15. Lagrangian History Re-Construction R. Boldi NSSTC/UAH

  16. Lagrangian History Re-Construction R. Boldi NSSTC/UAH

  17. Lagrangian History Re-Construction R. Boldi NSSTC/UAH

  18. Cell Position Prediction ….. ….. T(-2 dt) T(- dt) Future Cell Map T(+dt) T(0) Multi-Scale Vectors Current Cell Map T(0) Advector Future Cell Map T(+2.dt) : R. Boldi NSSTC/UAH

  19. Input Field Prediction T(-2 dt) T(- dt) Forecast Input Field T(+dt) T(0) Multi-Scale Vectors Advector Forecast Input Field T(+2.dt) Current Input Field T(0) : Trends, WxType R. Boldi NSSTC/UAH

  20. Summary • Tracker running • Creating Histories for Lightning Jump ALG • Creating Predicted Cell Positions • Platform for Lightning Prediction Work In Progress • Obtaining proxy ABI data for : • WxType • Storm intensity image R. Boldi NSSTC/UAH

  21. THE END R. Boldi NSSTC/UAH

More Related