Holocene changes in the position and intensity of the southern westerly wind belt

Abstract

The position and intensity of the southern westerly wind belt varies seasonally as a consequence of changes in sea surface temperature. During the austral winter, the belt expands northward and the wind intensity in the core decreases. Conversely, during the summer, the belt contracts, and the intensity within the core is strengthened. Reconstructions of the westerly winds since the last glacial maximum, however, have suggested that changes at a single site reflected shifts throughout the entire southern wind belt 1-4 . Here we use sedimentological and pollen records to reconstruct precipitation patterns over the past 12,500 yr from sites along the windward side of the Andes. Precipitation at the sites, located in the present core and northern margin of the westerlies, is driven almost entirely by the wind belt 5 , and can be used to reconstruct its intensity. Rather than varying coherently throughout the Holocene epoch, we find a distinct anti-phasing of wind strength between the core and northern margin over multi-millennial timescales. During the early Holocene, the core westerlies were strong whereas the northern margin westerlies were weak. We observe the opposite pattern in the late Holocene. As this variation resembles modern seasonal variability, we suggest that our observed changes in westerly wind strength can best be explained by variations in sea surface temperature in the eastern South Pacific Ocean.Chile is ideally located to reconstruct past variability of the southern westerly wind belt (SWW) as the SWW almost entirely controls precipitation on the western side of the Andes in southern South America with an extreme north-south rainfall gradient from the semiarid, winter-rain climate in central Chile to yearround hyper-humid conditions in the fjord region of southern Chile 5 (Supplementary Fig. S1). Therefore, any paleoclimatic proxy record primarily controlled by rainfall changes is suitable for reconstructing past changes in the SWW in this region. In present-day austral winters, the SWW extends northward, providing rainfall to central Chile (33 • -40 • S), but zonal winds are reduced in its core zone in southernmost Chile (50 • -55 • S; Fig. 1a). During austral summer, the zonal wind pattern shows a latitudinally more confined and intensified SWW with maxima over southernmost Chile (Fig. 1b). Previous reconstructions of the SWW were primarily based on single sites and generally suggested a northward migration and intensification of the SWW during colder periods 1,2,4 . Intensity variations across the wind belt have only recently been addressed and interpreted in terms of co-varying

The Last Termination in the South Indian Ocean: A unique terrestrial record from Kerguelen Islands (49°S) situated within the Southern Hemisphere westerly belt

“…Taking into account that the archipelago is situated in the core of the SHW today we suggest that the SHW shifted south of Kerguelen Islands during Northern Hemisphere cold events (HS1 and YD) and shifted equator-ward during the ACR-OCR. Following Lamy et al (2010), Kerguelen Islands should be less influenced by the SHW during the ACR-OCR as the belt was more expanded (Austral winter today), than during the warm SH events during HS1 and YD, which contradicts our results. However, the difference between summer and winter conditions of the wind belt is much less pronounced in the Indian Ocean than in South America and we want to stress that the hypothesis was brought forward based on Holocene records and has not been tested for the Last Termination.…”

“…An alternative hypothesis of SHW changes was brought forward by Lamy et al (2010) for Holocene records of South America but the mechanism may also apply to the Last Termination (Kilian and Lamy, 2012). It was proposed, due to opposing precipitation records from the core and the northern boundary of the SHW, that the SHW belt did not shift latitudinally as a whole but rather that the wind belt expanded during colder periods (as observed during Austral winter today) and became more confined during warmer conditions (Austral summer).…”

“…Denton et al, 2010), although its northern limit is uncertain. A second research topic that has received extensive attention is the role of the Southern Ocean (SO) and related to this, shifts of the Southern Hemisphere westerly belt (SHW), in releasing CO 2 from the deep ocean during warming phases in Antarctica (Toggweiler et al, 2006;Anderson et al, 2009;Denton et al, 2010;Lamy et al, 2010;Siani et al, 2013). A conceptual model has been presented by Toggweiler et al (2006) suggesting a pole-ward shift of the SHW from its equator-ward Last Glacial Maximum (LGM) position, aligning the SHW with the Antarctic Circumpolar Current (ACR), favouring ventilation of CO 2 to the atmosphere around Antarctica at the Antarctic divergence.…”

Glacial discharge along the west Antarctic Peninsula during the Holocene

“…S1) at 9.6 ka 16 and increased influence of glacial melt and precipitation in Maxwell Bay 13 . This warmth was concomitant with the last influence of orbitally-forced deglaciation 5,15 and ENSO-sensitive proxy records from the Peru margin and Ecuador 5,17 that indicate a persistent warm, mean La Niña-like state (relative to mean El Niño-like conditions).…”

“…S1). Increased upwelling of upper Circumpolar Deepwater (UCDW) onto the continental shelf at this time 14 associated with strong winds in the southern region of the Southern Hemisphere westerlies 15 (Fig. 1g) would have supplied additional warmth to the upper water column making frontal melting the dominant mechanism driving glacial discharge/δ 18 O diatom .…”