A 350,000 year summer-monsoon multi-proxy stack from the Owen Ridge, Northern Arabian Sea

Five summer-monsoon proxies from the Northern Arabian Sea are combined using stacking and principal components analysis (PCA) to create two very similar multi-proxy records of summer-monsoon variability. The five individual proxies all respond to monsoon variability but are largely independent in terms of the processes that complicate their interpretation as summer-monsoon indicators (e.g. preservation, dissolution, diagenesis, sediment reworking). As such, stacking and PCA average out non-monsoon variance, yielding a more pure monsoon signal. These stacked and PCA records (hereafter summer-monsoon stack and summer-monsoon factor) allow evaluation of relative monsoon strength through time as well as the relative concentration of variance within orbital bands; these two parameters are less reliable when estimated from individual proxy records. In fact, the summer-monsoon factor (SMF) accounts for only 33% of the total variance in the five records, suggesting that relative amplitude variations in each individual proxy time series are influenced by non-monsoon processes. The summer-monsoon stack (SMS) and SMF are spectrally very similar, dominated by variance in the 41-k.y. (obliquity) and 23-k.y. (precession) bands; there is very little variance at the 100-k.y. (eccentricity) band associated with large-scale changes in global ice-volume. Indeed, equally strong monsoons occur in both glacial and interglacial intervals. Within the 23-k.y. precession cycle, monsoon maxima fall at 3121‡ relative to precession minima (June 21 perihelion, maximum Northern Hemisphere (NH) summer insolation). This phase falls midway between N18O minima (378‡) and December 21 perihelion (3180‡) indicating that two mechanisms exert equal influence in determining the timing of strong summer monsoons within the precession band: (1) sensible heating of the Asian Plateau which is maximized at times of ice-volume minima (378‡), and (2) latent heat export from the southern subtropical Indian Ocean which is maximized at times of December 21 perihelion (3180‡). The seasonal cycle at December 21 perihelion is characterized by warm Southern Hemisphere (SH) summers followed by cold SH winters, a combination that preconditions the ocean to export latent heat during the boreal summer-monsoon season. Summer-monsoon winds transport this latent heat into Asia where it is released during precipitation, enhancing the Asian monsoon low. Within the 41-k.y. obliquity cycle, monsoon maxima are in phase with obliquity maxima. This indicates that two mechanisms, quite similar to those in the precession band, influence the timing of strong summer-monsoons in the obliquity band: (1) sensible heating of the Asian Plateau but with no ice-volume delay, and (2) latent heat export from the southern subtropical Indian Ocean which is maximized at times of obliquity maxima. Again, the seasonal cycle at obliquity maximum is characterized by warm SH summers followed by cold SH winters, ideal for maximizing latent heat export during the boreal summer monsoon. 0025-3227 / 03 / $ ^ see front matter ? 2003 Elsevier B.V. All rights reserved. doi:10.1016/S0025-3227(03)00207-X * Corresponding author. Tel. : +1-401-863-1964; Fax: +1-401-863-2058. E-mail address: [email protected] (S.C. Clemens). MARGO 3366 12-9-03 Marine Geology 201 (2003) 35^51 R Available online at www.sciencedirect.com www.elsevier.com/locate/margeo ? 2003 Elsevier B.V. All rights reserved.