13.07.2015 Views

Have the Southern Westerlies changed in a zonally symmetric ...

Have the Southern Westerlies changed in a zonally symmetric ...

Have the Southern Westerlies changed in a zonally symmetric ...

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

38M.-S. Fletcher, P.I. Moreno / Quaternary International 253 (2012) 32e46with<strong>in</strong> <strong>the</strong> westerly zone of <strong>in</strong>fluence north of w50 S <strong>in</strong> westernPatagonia (e.g. Villagran, 1988; Jenny et al., 2002; Abarzúa et al.,2004; Haberle and Bennett, 2004; Moreno, 2004; Latorre et al.,2007; Markgraf et al., 2007; Abarzúa and Moreno, 2008). Thewell dated (18 AMS radiocarbon dates) pollen record from LagoCondorito (41 S; Fig. 1a) can be used to represent changes <strong>in</strong> SWWderivedmoisture <strong>in</strong> this region over <strong>the</strong> last 14,000 years. LagoCondorito is a small closed-bas<strong>in</strong> lake, located <strong>in</strong> <strong>the</strong> transitionalzone between <strong>the</strong> warm and seasonal Valdivian ra<strong>in</strong>forests and <strong>the</strong>cool, perennially wet North Patagonian ra<strong>in</strong>forests (Moreno, 2004).A pollen-based normalized Eucryphia þ Caldcluvia (Valdivian taxa)/Podocarpaceae (North Patagonian taxa) <strong>in</strong>dex calculated for thissite proxies <strong>in</strong>tensity variations and movements <strong>in</strong> <strong>the</strong> nor<strong>the</strong>rnedge of <strong>the</strong> westerlies and reveals clear multi-millennial trends <strong>in</strong>moisture balance (Fig. 4e; Moreno, 2004). Initially, above averagemoisture between w14 and 12.5 ka is evident, followed bya decl<strong>in</strong><strong>in</strong>g moisture trend with negative anomalies betweenw10.5 and 7.8 ka and characterised by strongly negative valuesbetween 9.5 and 8.5 ka, reveal<strong>in</strong>g a decrease <strong>in</strong> westerly derivedmoisture north of <strong>the</strong> modern zone of strongest westerly flowdur<strong>in</strong>g this time (Fig. 4e). The <strong>in</strong>dex suggests that moisture<strong>in</strong>creased until a maximum at 6e5 ka, reflect<strong>in</strong>g an <strong>in</strong>crease <strong>in</strong>westerly derived precipitation north of <strong>the</strong> core of <strong>the</strong> westerlies,after which a multi-millennial decrease is evident (Fig. 4e).Fig. 4. Palaeoenvironmental data from sou<strong>the</strong>rn South America plotted on a calendarage scale: (a) Lago Cardiel lake-level reconstruction (dashed l<strong>in</strong>e <strong>in</strong>dicates <strong>the</strong> modernlake-level) (Site 12, Fig. 1a; Ariztegui et al., 2009); (b) Gran Campo-2 hygrophyte curveand hydrological phase (Site 14, Fig. 1a; Fesq-Mart<strong>in</strong> et al., 2004); (c) Lago TamarMisodendron curve (Site 15, Fig. 1a; Lamy et al., 2010); (d) sou<strong>the</strong>rn South Americancharcoal curve (dashed l<strong>in</strong>e <strong>in</strong>dicates <strong>the</strong> 14 ka mean) (Power et al., 2008); (e) LagoCondorito palaeovegetation <strong>in</strong>dex (dashed l<strong>in</strong>e <strong>in</strong>dicates <strong>the</strong> 14 ka mean) (Site 11,Fig. 1a; Moreno, 2004); (f) Lago Guanaco palaeovegetation <strong>in</strong>dex (dashed l<strong>in</strong>e <strong>in</strong>dicates<strong>the</strong> 12 ka mean) (Site 13, Fig. 1a; Moreno et al., 2010); (g) Laguna Potrok Aike lake-levelreconstruction (Site 16, Fig. 1a; Anselmetti et al., 2009); (h) Laguna Aculeo precipitationreconstruction (Site 10, Fig. 1a; Jenny et al., 2003). See Fig. 1a for <strong>the</strong> location ofsites. Grey shad<strong>in</strong>g <strong>in</strong>dicates <strong>the</strong> early Holocene (11e8 ka) period of weak westerlyflow <strong>in</strong> <strong>the</strong> Sou<strong>the</strong>rn Hemisphere.<strong>in</strong> <strong>the</strong> delivery of atmospheric moisture <strong>in</strong>to <strong>the</strong> extra-Andeanregion of Central Patagonia. After 8 ka, lake-levels fell, fluctuat<strong>in</strong>grepeatedly between þ15 m and <strong>the</strong> present-day level (Markgrafet al., 2003; Ariztegui et al., 2009).Ano<strong>the</strong>r much studied lake bas<strong>in</strong> <strong>in</strong> Argent<strong>in</strong>e Patagonia,Laguna Potrok Aike (52 S; Fig. 1a), lies south of <strong>the</strong> modern core of<strong>the</strong> SWW <strong>in</strong> a region where modern precipitation displays nostatistically significant relationship with SWW speed (Fig. 1b). Apossible reason for <strong>the</strong> lack of correlation of precipitation at this sitewith SWW speed is <strong>the</strong> proximity of both <strong>the</strong> Andes and <strong>the</strong>Atlantic Ocean, provid<strong>in</strong>g precipitation spill<strong>in</strong>g eastward with <strong>the</strong>westerly w<strong>in</strong>ds and <strong>in</strong>creased humidity from <strong>the</strong> South Atlanticunder weak westerly flow, respectively. A lake-level reconstructionfrom Potrok Aike (Anselmetti et al., 2009) reveals a lake regressivephase from >8.5 ka to a lake-level m<strong>in</strong>imum at w7 ka, after whicha multi-millennial lake-level transgression started and has persisteduntil <strong>the</strong> present (Fig. 4e).4.2.2. Western Patagonian vegetation4.2.2.1. North of w50 S. A consensus exists between virtually allvegetation-based multi-millennial moisture reconstructions from4.2.2.2. South of w50 S. Recent palaeovegetation-based <strong>in</strong>terpretationsof westerly derived moisture variations s<strong>in</strong>ce 14 ka fromsouth of 50 S <strong>in</strong> western Patagonia are divided between twooppos<strong>in</strong>g models of westerly change (Lamy et al., 2010; Morenoet al., 2010; Fletcher and Moreno, 2011). Important for bothmodels is <strong>the</strong> <strong>in</strong>terpretation of a normalized Nothofagus/Poaceaepollen <strong>in</strong>dex (NPI) from Lago Guanaco, a small closed-bas<strong>in</strong> lakesituated near <strong>the</strong> modern forest-steppe ecotone (52 S; Morenoet al., 2010) <strong>in</strong> an area where local precipitation is strongly andpositively correlated with zonal w<strong>in</strong>d speed (Fig. 1b). The NPI fromLago Guanaco was <strong>in</strong>terpreted by Moreno et al. (2010) as a localsensor of past shifts of <strong>the</strong> forest-steppe ecotone and <strong>the</strong> amount ofSWW-derived precipitation spill<strong>in</strong>g eastward over <strong>the</strong> Andes. TheLago Guanaco <strong>in</strong>dex displays <strong>in</strong>creas<strong>in</strong>g woody vegetation/moistureat <strong>the</strong> site between 13 and 11 ka, steppe/decreas<strong>in</strong>g moisturebetween 11 and 8 ka and an <strong>in</strong>creas<strong>in</strong>g, albeit variable, forestexpansion/moisture <strong>in</strong>creas<strong>in</strong>g trend <strong>the</strong>reafter (Fig. 4f). Ina subsequent study, Lamy et al. (2010) offer an alternative <strong>in</strong>terpretationof <strong>the</strong> NPI, with negative/steppe (positive/forest) excursionsconsidered as evidence for <strong>in</strong>creased (decreased) evaporationat <strong>the</strong> site under enhanced (attenuated) westerly flow.The location of <strong>the</strong> forest-steppe ecotone <strong>in</strong> Patagonia isdependent on moisture balance (Paruelo et al., 1998) and is <strong>in</strong>fluencedby <strong>the</strong> amount and seasonality of westerly precipitationspill<strong>in</strong>g eastward over <strong>the</strong> Andes and <strong>the</strong> amount of evaporationresult<strong>in</strong>g from desiccat<strong>in</strong>g (westerly) foehn w<strong>in</strong>ds. Lago Guanacopresently lies with<strong>in</strong> steppe vegetation, despite an average annualra<strong>in</strong>fall (720 mm p/a; Moy et al., 2008) above <strong>the</strong> suggestedthreshold for steppe development <strong>in</strong> Sou<strong>the</strong>rn Patagonia and Tierradel Fuego (500e350 mm p/a; Tuhkanen, 1992; Tuhkanen et al.,1992), which Lamy et al. (2010) <strong>in</strong>terpret as a reflection of <strong>the</strong>dom<strong>in</strong>ant <strong>in</strong>fluence of evaporation (strongest <strong>in</strong> <strong>the</strong> summer highw<strong>in</strong>d period) over <strong>the</strong> local moisture regime and vegetation type.They consider trends toward steppe (forest) <strong>in</strong> <strong>the</strong> <strong>in</strong>dex to bedriven by stronger (weaker) westerly flow and <strong>in</strong>creased(decreased) evaporation. An important factor neglected by thoseauthors is that <strong>the</strong> modern forest-steppe ecotone <strong>in</strong> Sou<strong>the</strong>rnPatagonia is not <strong>in</strong> equilibrium with climate, but is <strong>the</strong> result ofEuropean deforestation (Veblen and Markgraf, 1988; Huber andMarkgraf, 2003). Forest covered <strong>the</strong> landscape more than more

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!