Open AccessUnderstanding of Interdecadal Changes in Variability and Predictability of the Northern Hemisphere Summer Tropical–Extratropical Teleconnection
Author(s) -
JuneYi Lee,
KyungJa Ha
Publication year - 2015
Publication title -
journal of climate
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 3.315
H-Index - 287
eISSN - 1520-0442
pISSN - 0894-8755
DOI - 10.1175/jcli-d-15-0154.1
Subject(s) - teleconnection , climatology , predictability , extratropical cyclone , northern hemisphere , environmental science , el niño southern oscillation , southern hemisphere , monsoon , madden–julian oscillation , boreal , la niña , sea surface temperature , atmospheric sciences , geology , geography , meteorology , mathematics , paleontology , statistics , convection
Two dominant global-scale teleconnections—namely, western North Pacific–North American (WPNA) and circumglobal teleconnection (CGT)—in the Northern Hemisphere (NH) extratropics during boreal summer (June–August) have been identified as important sources for NH summer climate variability and predictability. An interdecadal shift in interannual variability and predictability of the WPNA and CGT that occurred around the late 1970s was investigated using reanalysis data and six coupled models’ retrospective forecast with a 1 May initial condition for the period 1960–79 (P1) and 1980–2005 (P2). The WPNA had a tight relationship with the decaying phase of El Niño–Southern Oscillation (ENSO) in P1, whereas it had a remarkably enhanced linkage with western North Pacific (WNP) summer monsoon rainfall in P2. The correlation coefficient between the WPNA and preceding ENSO (WNP monsoon rainfall) was reduced (increased) from −0.69 (0.1) in P1 to −0.60 (0.5) in P2. The CGT had a considerable connection with Indian summer monsoon rainfall (ISMR) in P1, whereas it had a strengthened relationship with the developing ENSO in P2. The correlation coefficient between the CGT and simultaneous ENSO (ISMR) was increased (decreased) from −0.41 (0.47) in P1 to −0.59 (0.24) in P2. Although dynamical models have difficulties in capturing the observed interdecadal changes, they are able to predict the interannual variation of the WPNA and CGT one month ahead, to some extent. The prediction skill of six models’ multimodel ensemble (MME) decreased (increased) from 0.78 (0.23) to 0.67 (0.67) for the WPNA (CGT) interannual variation. It is also noted that the spatial distribution of predictability and MME skill for 200-hPa geopotential height has been changed in relation to the changes in the WPNA and CGT.