X hits on this document

PDF document

web-61st-IHC-Booklet.pdf - page 92 / 122

344 views

0 shares

0 downloads

0 comments

92 / 122

Verification of the Monte Carlo Tropical Cyclone Wind Speed Probabilities: A Joint Hurricane Testbed Project Update

John A. Knaff, NOAA - NESDIS - Office of Research and Applications, Fort Collins, CO Mark DeMaria, NOAA - NESDIS - Office of Research and Applications, Fort Collins, CO Chris Lauer, NOAA - NWS – Tropical Prediction Center, Miami, FL

A Monte Carlo tropical cyclone wind speed probability estimation algorithm was developed at CIRA/NESDIS under previous Joint Hurricane Testbed (JHT) funding. The Monte Carlo Probability (MCP) program estimates the likelihood of 34-, 50-, and 64-kt winds out to five days. This new probability program was used to produce several National Hurricane Center (NHC) operational products during the Atlantic and east Pacific 2006 hurricane season. Versions were also developed for the Central and Western Pacific using forecast information from the Central Pacific Hurricane Center (CPHC) and the Joint Typhoon Warning Center (JTWC). The purpose of the current JHT project at CIRA/NESDIS is to develop a verification system for the new probability program. This presentation will provide a brief summary of the MCP algorithm and a status report on the verification system development.

The MCP program utilizes the past history (5 years) of the official track and intensity forecast errors from NHC, CPHC, and JTWC in combination with the errors from a simple climatology and persistence wind radii model (Knaff et al. 2007). The error fields are randomly sampled in a manner that includes serial correlation to provide wind speed probabilities that take into account the uncertainty in the track, intensity and structure forecasts. The verification program includes metrics that help to answer the following questions: 1.) Does the MCP program create skillful 34-, 50-, and 64 –kt wind speed probabilities? 2.) Are the probabilities unbiased? 3.) Do the MCPs tend to over/under predict the frequency of observed events? 4.) Are the MCPs an improvement over simply using the deterministic official forecast in a

probabilistic way (100% probability if a point comes within the forecast wind radii, 0% probability if it does not)? These questions can be answered using standard statistical techniques such as the expectation value, the Brier Skill Score (BSS; Brier, 1950; Murphy 1973), Relative Operating Characteristics (Mason and Graham, 1999), and Reliability Diagrams (Wilkes, 2006). The statistics associated with seasonal verification of the MCPs and the Official (i.e., deterministic) forecasts are compared and presented.

The views, opinions, and findings in this report are those of the authors and should not be construed as an official NOAA and or U.S. Government position, policy, or decision.

Brier, G. W., 1950: Verification of forecasts expressed in terms of probability, Mon. Wea. Rev., 78, 1-3. Knaff, J. A., C. R. Sampson, M. DeMaria, T. P. Marchok, J. M. Gross, and C. J. McAdie, 2007: Statistical tropical cyclone wind radii prediction using climatology and persistence. Wea. Forecasting, (in press). Murphy,A. H., 1973: Hedging and skill scores for probability forecasts. J. App. Met., 12, 215- 223. Mason, S. J. and N. E. Graham, 1999: Conditional probabilities, relative operating characteristics and relative operating levels. Wea. Forecasting, 14, 713-725. Wilkes, D. S., 2006: Statistical Methods in the Atmospheric Sciences, second edition, Elsevier Inc., pp. 627.

Document info
Document views344
Page views344
Page last viewedFri Dec 09 21:16:12 UTC 2016
Pages122
Paragraphs1288
Words31463

Comments