Geophysical Research Letters · 2015 · 18 citations · 23 references
Storm SurgeEngineeringHidden Markov ChainWeather ForecastingClimate ModelingEarth ScienceData ScienceDrought Risk ManagementDrought BustersDrought ForecastingHydroclimate ModelingMarkov ChainHydrometeorologyMeteorologySouth KoreaDrought AnalysisGeographyWeather DisasterForecastingStochastic ModelingDroughtDrought ManagementDisaster Risk ReductionFlood Risk Management
Abstract This study proposed a hidden Markov chain model‐based drought analysis (HMM‐DA) tool to understand the beginning and ending of meteorological drought and to further characterize typhoon‐induced drought busters (TDB) by exploring spatiotemporal drought patterns in South Korea. It was found that typhoons have played a dominant role in ending drought events (EDE) during the typhoon season (July–September) over the last four decades (1974–2013). The percentage of EDEs terminated by TDBs was about 43–90% mainly along coastal regions in South Korea. Furthermore, the TDBs, mainly during summer, have a positive role in managing extreme droughts during the subsequent autumn and spring seasons. The HMM‐DA models the temporal dependencies between drought states using Markov chain, consequently capturing the dependencies between droughts and typhoons well, thus, enabling a better performance in modeling spatiotemporal drought attributes compared to traditional methods.
23