Statistical adjoint sensitivity distributions of meteorological forecast errors of Asian dust transport events in Korea

Hyun Mee Kim, Jun Kyung Kay, Eun Gyeong Yang, Sehyun Kim, Meehye Lee

Research output: Contribution to journalArticlepeer-review

2 Citations (Scopus)


The sensitivities of meteorological forecast errors associated with Asian dust transport events to changes in the initial state were evaluated for 46 occurrences that affected the Korean Peninsula from 2005 to 2010. Adjoint-based sensitivities were used to determine these sensitivities. Sensitive regions were located primarily over two regions upstream from the Korean Peninsula: the northern source region, including areas of Mongolia and northern China, and the Tibetan Plateau. Depending on transport trajectories, month, and year, the sensitive regions were located over either the northern source regions or the Tibetan Plateau. That is, the Asian dust forecast in Korea was found to be sensitive to the meteorological fields over the northern source regions, but also those over the Tibetan Plateau even though the latter is not a dust source region or an upstream area according to the transport trajectories. Employing additional observations at existing instrumentation sites or developing new observational sites in both sensitive regions could be beneficial in reducing the atmospheric circulation forecast errors in East Asia, thus improving the accuracy of transport forecasts of Asian dust events affecting the Korean Peninsula.

Original languageEnglish
Article number20554
JournalTellus, Series B: Chemical and Physical Meteorology
Issue number1
Publication statusPublished - 2013


  • Adjoint sensitivity
  • Asian dust transport
  • Meteorological forecast error
  • Northern source region
  • Sensitive regions
  • Tibetan Plateau

ASJC Scopus subject areas

  • Atmospheric Science


Dive into the research topics of 'Statistical adjoint sensitivity distributions of meteorological forecast errors of Asian dust transport events in Korea'. Together they form a unique fingerprint.

Cite this