Show simple item record

dc.contributor.authorGombos, Daniel
dc.contributor.authorHoffman, Ross N.
dc.date.accessioned2017-06-09T17:36:07Z
dc.date.available2017-06-09T17:36:07Z
dc.date.copyright2013/06/01
dc.date.issued2013
dc.identifier.issn0882-8156
dc.identifier.otherams-87887.pdf
dc.identifier.urihttp://onlinelibrary.yabesh.ir/handle/yetl/4231605
dc.description.abstractxigent analysis supplements an ensemble forecast of weather-related damage with a map of the worst-case scenario (WCS), a multivariate confidence bound of the damage. For multivariate Gaussian ensembles, ensemble-based exigent analysis uses a Lagrange multiplier technique to identify the unique maximizing damage map at a given uncertainty level based on the ensemble-estimated covariance of the damage. Exigent analysis is applied to two case studies. First, using ensemble forecasts of 2-m temperature and estimates of the number of inhabitants at each location, exigent analysis is applied to forecast the worst-case heating demand for a large portion of the United States on 8?9 January 2010. The WCS at the 90th percentile results in only 1.26% more heating demand than the ensemble mean. Second, using ensemble forecasts of 2-m temperature and estimates of the number of citrus trees at each location, exigent analysis is applied to forecast the worst-case freeze damage to Florida citrus trees on 11 January 2010. For this case study, the WCS at the 90th percentile damages about 14.2 million trees, about 4.3 times more than the ensemble mean.
dc.publisherAmerican Meteorological Society
dc.titleEnsemble-Based Exigent Analysis. Part I: Estimating Worst-Case Weather-Related Forecast Damage Scenarios
dc.typeJournal Paper
dc.journal.volume28
dc.journal.issue3
dc.journal.titleWeather and Forecasting
dc.identifier.doi10.1175/WAF-D-12-00080.1
dc.journal.fristpage537
dc.journal.lastpage556
dc.treeWeather and Forecasting:;2013:;volume( 028 ):;issue: 003
dc.contenttypeFulltext


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record