Probabilistic Hurricane Storm Surge psurge NOAAs National Weather
Probabilistic Hurricane Storm Surge (psurge) NOAA’s National Weather Service Arthur Taylor MDL / OST April 13, 2006
Introduction NHC begins operational SLOSH runs 24 hours before landfall. • Provides a storm surge estimate for non-evacuation applications. Problem: Surges are based on a single NHC forecast track and associated parameters. • When provided accurate input, SLOSH results are within 20% of high water marks. • Track and intensity prediction errors are the largest cause of errors in SLOSH surge forecasts and can overwhelm the SLOSH results. Probabilistic Storm Surge 2006
Example: Katrina Advisory 23 Probabilistic Storm Surge 2006
Probabilistic Storm Surge Methodology Create an ensemble of SLOSH runs based on NHC’s official advisory and historic forecast errors. • Creates a probability of storm surge for this one forecast of this particular threatening hurricane. Not to be confused with FEMA’s 100 -year surge levels. Which hurricane forecast errors most impact storm surge? • • Cross track error (impacts landfall location) Along track error (impacts the timing of the storm) Intensity errors Structure of the storm errors. Probabilistic Storm Surge 2006
Varying Katrina’s Tracks • The cone of error is 50% of possible cross track error. • Include 90% of possible cross track error (roughly 3 times the size of the cone of error). • Spacing based on size of the storm Probabilistic Storm Surge 2006
Varying the Other Parameters: Size: Small (30%), Medium (40%), Large (30%) Forward Speed: Fast (30%), Medium (40%), Slow (30%) Intensity: Strong (30%), Medium (40%), Weak (30%) Probabilistic Storm Surge 2006
Calculate probability of exceeding X feet Look at each cell in each SLOSH run. • If the surge exceeds X, add the weight associated with that SLOSH run to the total. • The weight of a run is: cross track weight * along track weight * intensity weight * size weight • The total weight is the probability of exceeding X feet. Probabilistic Storm Surge 2006
Katrina Adv 23: Probability > 5 feet of storm surge Probabilistic Storm Surge 2006
Calculate height exceeded by X percent of storms. Determine what height to choose so that in any cell there is a specified probability of exceeding it. How? • • For each cell, sort the heights of each SLOSH run. From the tallest height downward, add up the weights associated with each SLOSH run until the given probability is exceeded. The height associated with the last weight added is the value for that cell. Probabilistic Storm Surge 2006
Katrina Adv 23: 10% of storms can exceed this height Probabilistic Storm Surge 2006
Arlene Adv 10 Probability > 5 feet of storm surge Probabilistic Storm Surge 2006
Arlene Adv 10: 10% of storms can exceed this height Probabilistic Storm Surge 2006
Expected Products in 2006 What: • Probability of storm surge exceeding 5 feet. • Height exceeded by 10% of storms. • Suggestions? (What could you use? ) Where: • http: //www. weather. gov/mdl/psurge • The National Digital Guidance Database (NDGD). When: • Beginning with the first NHC advisory forecasting landfall of a hurricane in 48 hours. • Available approximately 1 hour after the advisory release time. Probabilistic Storm Surge 2006
Is It Reliable? (from a statistical perspective) If we forecast 20% chance of exceeding 5 feet, does it actually exceed 5 feet 20% of the time? Combined: Bonnie 98, Earl 98, Georges 98, Bret 99, Floyd 99, Lili 02, Claudette, Isabel 03, Charley 04, Frances 04, Gaston 04, Ivan 04, Jeanne 04, Katrina 05 Probabilistic Storm Surge 2006
Implications How can you use this? • Please let us know how you plan to use it. Thoughts as to how one might use it: • The “probability of storm surge > 5 feet” product can show where it is likely to flood, and how susceptible an area is to storm surge. • The “height exceeded by 10% of storms” product can show the extent of the potential flooding, approximating “how bad it might get”. Feedback: • psurge@noaa. gov (currently arthur. taylor@noaa. gov) • http: //www. weather. gov/survey/nws-survey. php? code=phss Probabilistic Storm Surge 2006
- Slides: 15