Impacts of mid-latitude circulation on winter temperature variability in the Arabian Peninsula: the explicit role of NAO

Abstract

We investigated the impacts of mid-latitude circulation on winter temperature variability in the Arabian Peninsula by using NCEP reanalysis data and a historical Coupled Model Intercomparison Project phase-6 (CMIP6) simulation performed with the MPI-ESM1-2-HR coupled global climate model. Special emphasis is given to the North Atlantic Oscillation (NAO), an important mode of winter climate variability in the northern hemisphere. We first decomposed the winter temperature into dominant modes by applying an Empirical Orthogonal Function (EOF) analysis. The leading first and second EOF modes explaining 28.6% and 21.0% of the total variance of the winter temperature, reveal positive anomalies in the northern and negative anomalies in the southern Arabian Peninsula. The principal components associated with leading EOF modes reveal significant correlations with upper level circulation in the mid-latitude and display a Circumglobal wave-like pattern (CGT) extending from the North Atlantic to the East Asia region. We further defined winter temperature indices for the northern (hereafter WTNAP) and southern (hereafter WTSAP) Arabian Peninsula. The correlation maps between WTNAP and upper-level circulation exhibits a wave-like pattern in mid-latitudes similar to the CGT. At lower levels, the WTNAP reveals significant correlations with sea surface temperature and mean sea level pressure over the North Atlantic Ocean, depicting an NAO-like pattern. We further carried out an inverse analysis using the winter NAO index. The NAO impacts the winter temperature over the Arabian Peninsula via mid-latitude circulation. At lower levels, the positive (negative) NAO phases are associated with anomalous anticyclonic (cyclonic) circulation over Mediterranean. The anomalous circulation associated with NAO favors cold (warm) temperature advection to the Arabian Peninsula and hence modulates the winter temperature. The ERA5 reanalysis and a historical CMIP6 model simulation performed with the global climate model MPI-ESM1-2-HR also underlines the proposed findings and mechanism.