A probabilistic framework for assessing drought recovery

Ming Pan, Xing Yuan, Eric F. Wood

Research output: Contribution to journalArticlepeer-review

72 Scopus citations

Abstract

A probabilistic framework is proposed to explore drought recovery and its uncertainty based on any ensemble forecast. First, the joint distribution between precipitation and drought index is established from the forecast ensemble using the copula method, which allows arbitrary marginal distributions and fine-tuned correlation structure. Then, questions like "how much precipitation is needed for recovery and its uncertainty?" and "what is the likelihood that a specified drought index threshold be surpassed given specified cumulative precipitation over a fixed period?" are studied. The application investigates how the 2012-2013 drought over central United States may recover during the forecast period from February to July 2013. The ensemble streamflow prediction method is used to create the ensemble forecast, with soil moisture percentile against climatology as the drought index. We find significant uncertainty in drought recovery given the same precipitation, suggesting that a probabilistic analysis will offer important additional information about drought risk.

Original languageEnglish (US)
Pages (from-to)3637-3642
Number of pages6
JournalGeophysical Research Letters
Volume40
Issue number14
DOIs
StatePublished - Jul 28 2013

All Science Journal Classification (ASJC) codes

  • Geophysics
  • General Earth and Planetary Sciences

Keywords

  • drought
  • forecast
  • recovery
  • risk

Fingerprint

Dive into the research topics of 'A probabilistic framework for assessing drought recovery'. Together they form a unique fingerprint.

Cite this