U.S. Earth System History Science

Home , East Iceland Current, Heinrich event

PAGES International Project Ofﬁ ce Sulgeneckstrasse 38 3007 Bern Switzerland Tel: +41 31 312 31 33 Fax: +41 31 312 31 68 [email protected] Text Editing: Leah Christen News Layout: Christoph Kull Julie Brigham-Grette, Christoph Kull and Circulation: 3700 Thorsten Kiefer, Editors

VOL.14, N°2 – AUGUST 2006 U.S. Earth System History Science

An unusual debris ladened iceberg photographed off the coast of East Greenland close to the site of JM96- 1205 (see Fig. 1 on page 8). It appears that the iceberg has turned over thus revealing what was probably the basal debris zone of a tidewater outlet glacier (photographer: Anne E. Jennings, Sept. 1996).

Contents

2 Announcements

www.pages-igbp.org - Editorial: U.S. Earth System History Program - Models and abrupt climate change - Inside PAGES - Late Holocene hydrologic variability in - PAGES-LOTRED-SA meeting in Argentina North American peatlands

3 Program News 29 Science Highlights (open section) - DAYACLIM: A Sahara climate program - San Valentin Glacier ice core (Chile) - Holocene climate change (Yemen) 4 National Page - Climate and human evolution (Africa) - Indonesia - Surface temperatures over the past two millennia 5 ESH Science Highlights - East Pacifi c production during the 37 Workshop Reports last deglaciation - HOLIVAR open science meeting - UK - Holocene IRD in the North Atlantic - PAGES/CLIVAR workshop - Switzerland - Dendrochronology and the Asian Monsoon - ICDP workshop PASADO - Argentina - Tropical Pacifi c climate, solar forcing and - Climate change challenge - Sweden the last millennium - LUCIFS workshop - Germany - The mystery interval 17.5 - 14.5 kyr ago - Ocean-Land INQUA workshop - Spain - South American Monsoon and the LGM - Indonesian throughfl ow - Germany - U.S. lake drilling projects - Post glacial oaks in North America 44 Last Page - Consolidating high and low frequency - Calendar information for climate reconstructions - New on the PAGES bookshelf

ISSN 1563–0803 The PAGES International Project Ofﬁ ce and its publications are supported by the Swiss and US National Science Foundations and NOAA. 2 Announcements

Editorial: U.S. Earth System History Program in Transition

A significant portion of paleoclimate science conducted by the United States over the past 14 years has been carried out through proposals to the Earth System History (ESH) Program of the U.S. National Science Founda- tion (NSF). This cross-disciplinary program has been jointly supported under the Directorate for Geosciences with contributions from its divisions, including Earth Sciences, Ocean Sciences and Atmospheric Sciences. Initially developed under the U.S. Global Change Research Program involving a number of federal agencies, NSF acutely acknowledged the need for a program to promote pioneering research on the temporal and spatial scales of natural climate variability, climate proxies, and modeling in order to better understand the forcing mechanisms, systemic sensitivities, and feedbacks responsible for past climate change on regional and global scales. Partnerships for shared support of selected proposals were also fostered in the earlier years with the NSF Office of Polar Programs and the National Oceanic and Atmospheric Administration’s (NOAA) former Paleoclimate element in the Office of Global programs (now known as the Climate Change Data and Detection element in the Climate Program Office). Since its inception, the ESH program has been guided by an adhoc advisory group composed of scientists on rotation from a range of disciplines who periodically sought community input to guide program manag- ers in developing targeted areas of emphasis for annual solicitations as the science evolved. The program has supported a vast range of projects world-wide, primarily but not exclusively, aimed at high-resolution climate records using a variety of sedimentary, geochemical, and biogeochemical paleo-proxies from diverse environments to develop both qualitative and quantitative estimates of past change. The program has particularly encouraged over time the integration of empirical and observational data with a variety of model simulations of regional and global climate change. Recent initiatives of emphasis within the ESH program have focused on topics including Holocene change, abrupt change on a variety of timescales, and regional vs. global scale change. Because the solicitation for the ESH program expires September 30, 2006, NSF and the science community are now poised to determine how this successful program might evolve in the coming years to address ongoing and future scientific needs (see www.nsf.gov/pubs/2006/nsf06029/nsf06029.jsp). Plans are now being formulated for an official review, during which time the Directorate for Geosciences will continue to seek community input during the decision making process. An initial workshop held in November 2005 focused on a few relevant cross-cutting science questions and the workshop recommendations are available at www. mesh.usc.edu/workshop.htm for community comment. The short articles presented in this issue were volunteered at large from ESH-funded researchers and represent only a small cross-section of the range of science address by the ESH initiative. To learn more about past ESH initiatives, go to the ESH website at www.nsf.gov/funding/pgm_summ.jsp?pims_id=5750&org=ATM.

JULIE BRIGHAM-GRETTE PAGES SSC Chair [email protected]

Tales from the ﬁeld Do you have an interesting and humorous story from your paleoenvironmental ﬁeldwork? Send us your tale (max. 500 words) so that we can publish it in PAGES News!

��

PAGES NEWS, VOL.14, N°2, AUGUST 2006 Announcements 3

In early June, the PAGES-CLIVAR tions will close 15 September 2006. workshop on Past Millennia Variabil- We particularly encourage nomina- Inside PAGES ity, organized by the PAGES office, tions of paleoscientists from eastern was held in Wengen, Switzerland. Asia, and from Russia and former This very successful event will al- Soviet Union countries, since these This spring, the PAGES office was so have some follow-up activities. geographic regions will be under- busy keeping different PAGES proj- Read more in the workshop report represented when current members ects on track: on page 38. leave at the end of this year. For de- Several guest scientists spent Finally, the PAGES office has tails, please go to: www.pages-igbp. a few days to a few weeks at the been engaged in organizing several org/people/sscleaders/nominations. PAGES office. Bert Rein compiled an upcoming fall activities. The IGBP- html. extensive National PAGES Germany SCOR Fast Track Initiative on Ocean website and Sri Yudawati Cahyarini Acidification will hold its first work- Next issue of PAGES News: set up the Indonesian PAGES (page shop at the end of September in the The next issue of the PAGES news- 5, this issue). All 18 of the currently USA (http://igbp-scor.pages.unibe. letter will be published in December. available national PAGES websites ch/firstworkshop.html). Regional It will contain an open section on are online at: www.pages-igbp.org/ Variability workshops will be held general paleoscience themes and a about/national/. A few more Nation- for the Mediterranean area, together special section on radiocarbon chro- al PAGES are in the pipeline and will with the MEDCLIVAR program, in nometry. Please send your contribu- be ready in autumn this year. Spain in November, and for South tions of science highlights, program In May, the IPO, with the help America, alongside the annual news, workshop reports and humor- of Isabelle Larocque, organized the PAGES SSC meeting, in Argentina ous tales from the field to Christoph second Databoard meeting, at the in October. Information is available Kull ([email protected]) by 31 Oc- PAGES office in Bern. This project, on our calendar webpage (www. tober 2006. Guidelines can be found which aims to develop a better da- pages-igbp.org/calendar/). at www.pages-igbp.org/products/ ta network in the near future, will newsletters/instructions.html. be of special benefit to the paleo- SSC nominations community. Further information is At the upcoming SSC meeting, two available at www.pages-igbp.org/ new SSC members will be elected. services/data/. The related Open Call for Nomina- Regional Climate Variations in South America over the late Holocene October 4-7, 2006; Malargüe, Mendoza, Argentina

The aim of this meeting is to deepen the academic interactions among paleoclimatologists of South America, without neglecting the presence of notable investigators from outside the region. During the four day meeting, the conference will be organized around 8 thematic sessions related to climatic, high-resolution paleoclimatic and numerical modeling studies in South America. The preliminary sessions are related to:

- Documentary data and early instrumental data - Tree rings - Lake and marine sediments - Ice cores - Glaciers - Pollen and vegetation records - Multi-proxy reconstruction methods - Climate and modeling

Abstract Deadline: 8 September, 2006

Further Infromation: www.pages-igbp.org/science/initiatives/lotred-sa Claudia Bottero, IANIGLA – CRICYT, Mendoza, Argentina, [email protected]

PAGES NEWS, VOL.14, N°2, AUGUST 2006 4 Program News

DAYACLIM: A northern Sahara desert climate change project

BELHADJ HAMDI-AÏSSA1,2 AND NACER MESSEN2 1Dép. Sci. Agronomiques, Laboratoire de Biogéochimie des Milieux Désertiques and ECO-SYS, Université de Ouargla, Algeria; [email protected] 2Centre de Recherche Scientiﬁque et Technique sur les Région Arides, Biskra, Algeria; [email protected]

In order to address the consequenc- 3. Optically Stimulated Lumines- es of global change in the Northern cence induced by laser (OSL) Sahara Desert (NSD) region, an ini- A particular stratigraphic challenge tiative for collaborative research on for dating is to quantify hiatuses, i.e. global change in this region, the to evaluate the duration correspond- DAYACLIM project (N°17/03/01/02/ ing to the absence of deposits, which MN), was officially launched in Feb- usually occur at the transitions from ruary 2003. This initiative formulates arid to humid periods and inverse- and coordinates an integrated mul- ly. tidisciplinary approach to observing, The 1st interdisciplinary DAYA- reconstructing, and modeling global CLIM Workshop “Dating of climate change phenomena and processes oscillations, hydrologic and thermic in the NSD region. events in North Africa”, was sponsored by the EUR-OPA Risques-Ma- Objectives jeurs and organized by CRSTRA and The DAYACLIM objectives are: Ouargla University, Algeria, from 3-4 - Understanding past climate March 2004, followed by a fieldtrip changes and variability to im- to the DAYACLIM investigation field prove the prediction of future site. global environments Figure 1: Pedo-sediment layers in the inves- The DAYACLIM research group - Contributing to the understand- tigated area. welcomes international scientific or ing of the origin and causes of et al., 2004). Sediments in depres- financial collaboration. climate changes and variability, sions fall dry temporarily during arid and collecting archived informa- phases and get covered with a sand NOTES DAYACLIM is named for “Dayas” (Daïas), tion useful for paleoclimatic re- layer and eolian sand deposits (Fig. a North African and Middle Eastern word construction of the NSD region 1). Thereafter, the fluvial and detritic meaning desert non-saline endorheic - Understanding the interaction be- sediments transported by water dur- depressions on limestone areas. tween groundwater and hydro- ing pluvial phases are interstratified ACKNOWLEDGEMENTS logical changes. with sandy accumulations. DAYACLIM is formally endorsed by the Arid - Establishing a database for pa- Our approach relies on the theory Regions Research Centre (CRSTRA) (Prof. leoenvironmental data from the of biorhexistasy, which describes a A. Gaouar, N. Ferhat), Ouargla University NSD region complex succession of sedimenta- (Prof. B. Hamdi-Aïssa, M. Hacini, B. Djili, F. Youcef), Tebessa University (Prof. A. - Providing a coordinated contri- tion/pedological phases alternating Djerrab), USTHB University (Prof. F. Youcef- bution of Algerian research to with erosional episodes, both being Ettoum, S. Benzinah) and Algerian Nuclear multidisciplinary international controlled by climate forcing (Fe- Research Centre (CNRB-CNRA) (Prof. N. Messen, D. Imatoukéne), INA Alger (H. projects, and exploring the pos- doroff and Courty, 1999). The concept Aouam). sibilities of future paleoenviron- of hierarchy of soil attributes (i.e. pa- mental research. leosol) in terms of a chronology of REFERENCES events and their paleoclimatic sig- Fedoroff, N. and Courty, M.A., 1999: Soil and soil forming processes under increasing aridity. In: Strategy nificance (Fedoroff and Courty, 1999, Singhvi, A.K., and Derbyshire, E. (Eds). Paleoenvi- DAYACLIM paleoclimatologists have Hamdi-Aïssa, 2002) is adopted in the ronmental reconstruction in arid lands. Oxford & two major concerns: (i) to select sites DAYACLIM project. IBH, New Delhi, 73-108. Hamdi-Aïssa, B. 2002: Paleogeochemical interpreta- for investigation with high-resolution For dating the pedo-sediment tion of some gypsic microfabrics in hyper-desert signals of climate changes and vari- layers, we will have recourse to ab- soils. In: ISSS ed, 17th World Congress of Soil ability; (ii) to make sure that these solute and relative dating. According Science, Bangkok, (CD-Rom publication), paper n° 1861:1-9. investigated sites offer continuous to the time interval and the nature Messen, N., Gaouar, A., Hamdi-Aïssa, B., Hacini, signals of past climate oscillations. of the selected material, useful chro- M. and Youcef-Ettoumi, F., 2004: Etude du climat Flooded base-level plains and en- nostratigraphic methods are limited paléolithique dans un bassin sédimentaire du sud dorheic (closed) depressions where to: algérien: sélection du site. Journal Algerien pour les Zones Arides, 3: 38-44. water is periodically stored at the sur- 1. Uranium series absolute dating Rognon, P., 1980: Pluvial and arid phases in the face could function as paleoclimate 2. Electronic Spin Resonance (RSE) Sahara: the role of non climatic factors. In: geo-archives of hydrological chang- or Electronic Paramagnetic Sarnthein, M., Seibold, E. and Rognon, P., (Eds.), “Sahara and Surrounding Seas”, Palaeoecology of es during interglacial phases in the Resonance (RPE) Africa, 12: 45-61. NSD region (Rognon, 1980; Messen

PAGES NEWS, VOL.14, N°2, AUGUST 2006 National 5

Indonesia NATIONAL CONTACT: Dr. Wahyoe Soepri Hantoro Indonesian Institute of Sciences Bandung, 40135 [email protected]

Welcome Indonesia consists of a cluster of about 18,000 islands near the equator, located between the Pacific and Indian Oceans to the east and west, and the Asian and Australian land masses to the north and south. Accordingly, Indonesia’s seasonal climate is controlled by the migrating Intertropical Convergence Zone (ITCZ) and the interaction of the tropical Indopacific with the large land masses in both hemispheres, that is by the Asian and Australian Monsoons. On interannual timescales, Indonesia’s climate and environment is affected by large scale ocean-atmosphere processes in the adjacent tropical seas, namely the El Niño Southern Oscillation (ENSO) and the Indian Ocean Dipole (IOD). With roughly three quarters of Indonesia’s area being ocean, studying the interplay between components in the climate-environment system, such as land ocean linkages, Pacific-Indian Ocean exchange, ENSO-IOD coupling, is a major target. It is not surprising that two oceanic features are the main targets of international research. These features are the Western Pacific Warm Pool (WPWP) and the Indonesian Throughflow (ITF), both being key to the distribution of heat and salt/freshwater on a regional and global scale and hence of crucial importance for the global climate system.

National Science Highlight: Coral Proxy Records - The Indian Ocean Dipole Event Abram et al., 2003 Geochemical records from living and fossil corals from off western Sumatra document repeated events of anomalously strong upwelling in the eastern Indian Ocean during the last two centuries and the mid-Holocene. The upwelling events and resulting cooler surface waters were associated with the Indian Ocean Dipole (IOD) oscillation. During the last of these IOD anomalies in 1997, coral reefs in the area died, whereas they survived all the earlier anomalies recorded. Detailed analysis reveals that iron fertilization by wildfires in Indonesia in 1997 produced an unusual red tide that led to the death of the corals. While wildfires in Indonesia can result from drought in response to IOD (and ENSO) oscillations, they seem to become more severe with ongoing land-use changes, posing an increasing threat also to coastal ecosystems.

Figure 1: Coral Sr/Ca (red), δ18O (blue) records, and grid sea surface temperature anomaly (black) from Mentawai, western Sumatra shows cooling during IOD events (figure modified from Abram et al., 2003).

Reference: Abram, N.J., M.K. Gagan, M.T. McCulloch, J. Chappell and W.S. Hantoro (2003), Coral reef death during the 1997 Indian Ocean Dipole linked to Indonesian wildfires, Science, 301, 952-955.

Research Institutions: - Earth Dynamics and Geological Disasters Group, Research Center for Geotechnology, Indonesian Institute of Sciences (LIPI), Bandung www.geotek.lipi.go.id/ Funding Agencies: - Marine Geological Institute, Bandung www.mgi.esdm.go.id/ - State Ministry of Research and Technology www.ristek.go.id/english/incentive.html - Geology and Paleontology Group, Department of Geology, Institute of Technology, Bandung - Indonesian Institute of Sciences (LIPI) www.fiktm.itb.ac.id/%7eweb-baru/ www.lipi.go.id/ - Earth Physics Group, Department of Physics, Institute of - Directorate General of Higher Education Technology, Bandung www.dikti.org www.fi.itb.ac.id/?p=rearth

Number of Indonesian subscribers in the PAGES database (www.pages-igbp.org/people/) on July 12th: 16

Background photo: Sri Yudawaiti Cahyarini

PAGES NEWS, VOL.14, N°2, AUGUST 2006 6 Science Highlights: U.S. ESH Program

Climate-related changes in export production off the Paciﬁc coast of Mexico during the last deglaciation

ROBERT F. ANDERSON1, MARTIN Q. FLEISHER1 AND PETER W. KUBIK2 1Lamont-Doherty Earth Observatory, Columbia University, Palisades, USA; [email protected] 2Paul Scherrer Institute, c/o Institute of Particle Physics, ETH Hönggerberg, Zurich, Switzerland

Naturally occurring radionuclides such as 230Th, 231Pa and 10Be are � �� removed from the ocean by scav- � �� enging onto particles. Due to their �� differing sensitivities to particle �� 231 �� composition and particle ﬂux, Pa/ �� 230Th and 10Be/230Th ratios measured �� in marine sediments can be used �� to trace past changes in biological ��

productivity of the ocean. �� Climate-related changes in the �� ocean’s oxygen minimum zones �� A striking feature of abrupt climate �� change during the last glacial period is the correlation between tempera- ��

ture proxy records from Greenland �� ice cores and the intensity of the ox- �� ygen minimum zone (OMZ) in the ��

��

eastern North Paciﬁc Ocean (Ortiz ��

et al., 2004) and the Arabian Sea ��

�� (Altabet et al., 2002). Oxygen con- ��

��

centrations decreased during warm ��

�� interstadial periods, much like con- �� ditions today, whereas oxygen �� concentrations were much greater �� during cold stadials and the Last Glacial Maximum. Intensification Figure 1: Records from piston core MV99-PC08 (23.5°N, 110.6°W, 700 m) covering the period of the OMZ is manifested in several from ~6 to ~16 ka BP. (A) Diffuse spectral reflectance (DSR) factor 3, which is representative of proxies, including concentrations in the organic carbon content of the sediments (from Ortiz et al., 2004). (B&C) Preserved fluxes sediments of organic carbon and of (g cm-2 kyr- -1) of opal and organic carbon evaluated using the 230Th normalization method (Fran- cois et al., 2004). (D) Decay-corrected unsupported 231Pa / 230Th (activity ratio; diamonds) and redox-sensitive trace elements, as 10Be / 230Th (109 atoms/dpm, divided by 2 to put them on the same scale as Pa/Th; circles). well as in nitrogen isotope indicators of denitrification. distinguishable from that of high particles or the ratio of carbonate Various investigators have at- biological productivity and associ- to opal (Chase et al., 2002)) as well tributed these features to either 1) ated organic rain rates. as to changes in particle flux (pro- changes in the supply of oxygen ductivity). Therefore, although these associated with the ventilation of New geochemical proxies link ratios do not provide a unique and intermediate water mass, or 2) lo- OMZ changes to biological unambiguous paleoproductivity cal changes in productivity, as a productivity tracer, they are a valuable compo- result of ocean-atmosphere reor- Natural radionuclides can help re- nent of a multiproxy study because ganizations, such as is seen during solve this. Due to the differential their sensitivities to changing envi- El Niño events today in the eastern solubilities of the radionuclides, ronmental parameters are orthog- North Pacific. Whereas evidence for 231Pa/230Th and 10Be/230Th ratios in- onal to the sensitivities of other changes in the intensity of the OMZ crease with increasing particle flux paleoproductivity proxies. is unequivocal, most sedimentary (Kumar et al., 1995), which, in turn, Concentrations of 231Pa, 230Th, proxies cannot discriminate be- is regulated by biological produc- 10Be and biogenic constituents were tween the two proposed causes tivity. These ratios have the advan- measured in sediments from a pis- because the impact of low bottom- tage that they are insensitive to the ton core recovered from within the water oxygen concentrations on the concentration of dissolved oxy- OMZ off the southern tip of Baja composition of marine sediments gen. However, they are sensitive California (Mexico). Across the last (e.g., carbon concentration; trace to changes in particle composition deglaciation, we find that 230Th- metal enrichment) is generally in- (e.g., the ratio of clay to biogenic normalized fluxes (Francois et al.,

PAGES NEWS, VOL.14, N°2, AUGUST 2006 Science Highlights: U.S. ESH Program 7

scavenging of Th, Pa and Be in the ocean. Earth 2004) of opal and of organic carbon logical productivity and the export and Planetary Science Letters, 204: 215-229. were positively correlated with the fl ux of biogenic particles. Enhanced Francois, R., Frank, M., Rutgers Van Der Loeff, M.M., high organic carbon content of productivity during warm events re- and Bacon, M.P., 2004: Th-230 normalization: An essential tool for interpreting sedimentary fl uxes sediments deposited during warm fl ects a shoaling of the nutricline, during the late Quaternary. Paleoceanography, periods (Bølling and Holocene; Fig. an increase in upwelling-favor- 19(1): PA1018, doi:10.1029/2003PA000939. 231 1). We fi nd elevated Pa/230Th and able winds, or some combination Kumar, N., Anderson, R.F., Mortlock, R.A., Froelich, 10Be/230Th ratios during warm peri- thereof. P.N., Kubik, P., Dittrich-Hannen, B., and Suter, M., 1995: Increased biological productivity and export ods as well (Fig. 1), indicating that production in the glacial Southern Ocean. Nature, increased productivity must have NOTE 378(6558): 675-680. contributed to the higher preserved Data used in this study will be available in Ortiz, J.D., O’Connell, S.B., Delviscio, J., Dean, W., the NOAA Paleoclimatology data base at: Carriquiry, J.D., Marchitto, T., Zheng Y., and Van fl uxes of opal and carbon at those www.ncdc.noaa.gov/paleo/paleo.html Geen, A., 2004: Enhanced marine productivity off times. western North America during warm climate inter- REFERENCES vals of the past 52 ky. Geology, 32(6): 521-524. Conclusions Altabet, M.A., Higginson, M.J., and Murray, D.W., 2002: The effect of millennial-scale changes in While these results do not rule out Arabian Sea denitrifi cation on atmospheric CO2. changes in the ventilation of inter- Nature, 415(6868): 159-162. mediate waters, they do provide Chase, Z., Anderson, R.F., Fleisher, M.Q., and Kubik, P., 2002: The infl uence of particle composition on clear evidence for changes in bio-

Is there a pervasive Holocene ice-rafted debris (IRD) signal in the northern North Atlantic? The answer appears to be either no, or it depends on the proxy!

JOHN T. ANDREWS1, ANNE E. JENNINGS1, MATTHIAS MOROS2, CLAUDE HILLAIRE-MARCEL3 AND DENNIS EBERL4 1Institute of Arctic and Alpine Research and Department of Geological Sciences, University of Colorado, Boulder, USA; [email protected] 2Baltic Sea Research Institute Warnemünde, Rostock, Germany and Bjerknes Centre for Climate Research, Bergen, Norway 3GEOTOP, Université du Québec à Montréal and McGill University, Canada 4Dennis D. Eberl, U.S. Geological Survey, Boulder, USA

Introduction �� One of the most cited papers (271 �� citations by June 2006) on Holocene �� climate change is the 2001 paper by �� Bond and colleagues (2001), which �� argued for a pervasive ~1500 yr �� signal in the delivery of hematite- �� stained quartz (HSQ) sands to sites �� at or beyond the historic limits of �� observed drift ice (Fig. 1) (we use � � the term “drift ice” to denote any �� mixture of glacier icebergs or vari- �� ous forms of sea ice). To critically �� examine the hypotheses set forth �� in Bond et al. (2001), we selected �� a series of cores for analysis of �� ice-rafted debris (IRD). These cores �� were collected from sites that are �� much more within the historically �� known limits of drift ice (Fig. 1), �� hence lie to the north of the sites �� examined for HSQ. Two of the cores are exceptionally well dated (Moros �� et al., 2006) and the other two have Figure 1: Location of cores used in this study (red circles) and cores reported by Bond et al. (2001). Historic maximum limits of drift ice are shown based on data from the International suffi cient age control to examine Ice Patrol (www.uscg.mil/LANTAREA/IIP/home.html) and the Norsk Polar Institute (acsys. the Holocene trends in IRD. npolar.no/ahica/quicklooks/looks.htm). Arrow labels: EGC = East Greenland Current; B/LC = Baffi nland and Labrador Current; EIC = East Iceland Current. Methods variations in sediment carried by of provenance (i.e. mineralogy) There is no single working defi ni- drift ice are often defi ned on the ba- (Ruddiman, 1977; Andrews et al., tion of IRD (Andrews, 2000); the sis of grain-size or on some aspects 1989; Andrews et al., 1997; Bischof,

PAGES NEWS, VOL.14, N°2, AUGUST 2006 8 Science Highlights: U.S. ESH Program

2000). Icebergs and sea ice carry sediments that vary in size from pebbles to fi ne clay and the largest contribution by weight is generally in the silt fraction. We use quantitative X-ray diffraction identifi cation on the <2000 µm size fraction (Eberl, 2004; Moros et al., 2006), i.e. the sand plus silt plus clay, to identify minerals that are exotic to particular sites, and which can be associated with the transport and delivery of sediments in drift ice (Moros et al., 2006). Between 1-3 g of sediment is milled and spiked with a known weight of corundum or zincite. X-ray scanning is conducted between 5 and 65° two-the- ta (Eberl, 2004) and the weight% of non-clay and clay minerals is carried out automatically on the digital output fi les. Results are reported as weight% of the non-clay and clay minerals. Replication experiments indicate that standard errors on the major minerals of interest are of the order of ± 0.5 weight%. We typically identify 15-18 non-clay and 7 clay minerals. The total volume of icebergs calved into Baffi n Bay is of the order of 200 km3/yr. Some 40,000 icebergs a year are estimated to calve into Baffi n Bay from the large outlet glaciers of W/NW Greenland and the Canadian High Arctic; these icebergs are carried southward in the Baffi nland and Labrador currents. The primary sediment signature for glacial erosion and transport in icebergs from NW Greenland is dolomite (Marlowe, 1966). We hypothesize that the major agent of transport in the western North Atlantic, including Baffi n Bay and the Labrador Sea, is in icebergs, as the only major source for sediment loading of sea ice is the western area of Hudson Bay. The bedrock geology of Iceland and the East Greenland area in the vicinity of JM96-1205 is composed of Cenozoic fl ood basalts—hence there is virtually no normative Figure 2: Data from northern North Atlantic cores (Fig. 1) compared with hematite-stained quartz data from V28-14 (Bond et al., 2001). Data in plots A-D are based on weight% of minerals quartz in the bedrock. However, on in the <2000 µm size fraction. A) MD99-2237—Orphan Knoll; B) JM96-1205—Grivel Basin; C) East Greenland, the basalts lie on MD99-2269 and B997-327—North Iceland; D) MD95-2011—Vorring Plateau; E) V28-14—South Precambrian granites and gneisses, Denmark Strait. and large outlet glaciers extend sea- lie to the north of Kangerlussuaq sediment transport by drift ice in ward along several fjords. Early Ter- Fjord and Scorsby Sund, respec- the eastern North Atlantic (through tiary and older sedimentary rocks tively (Jennings et al., 2002). The Fram Strait) is dominated by fi ne-

PAGES NEWS, VOL.14, N°2, AUGUST 2006 Science Highlights: U.S. ESH Program 9

grained silt and clay incorporated shows a decrease from high values Our results should make re- into the sea ice on the extensive in the early Holocene to the mid Ho- searchers question the underlying shelves fronting the Arctic Ocean locene, and then a sharp and pro- forcing behind the HSQ data (Fig. (Hebbeln, 2000). nounced increase in quartz over the 2E). The cycles in this and the oth- last 5-6 cal ka BP. Quartz data from er records may be linked to solar Results core B997-327PC, which was taken forcing but the specific link to ice We discuss the cores along a west within a few 100 m of MD99-2269, rafting appears ambiguous. The to east transect (Fig. 1 & 2A-D) and are nearly identical. varying trends in our data suggest compare these results with the The most easterly core, MD99- we need to consider the Holocene HSQ data from V28-14 just south 2011, was taken from the Vorring oceanographic/climate evolution of of the Denmark Strait. (Bond et al., Plateau (Fig. 1). Here, the Quartz/ each region individually (de Vernal 2001) (Fig. 2E). MD99-2237 (Fig. 1) Plagioclase (Q/P) ratio is used as and Hillaire-Marcel, in press). was sampled at ~100 yr intervals. a measure of ice rafting. The long- On the East Greenland shelf we term trend in these data (Fig. 2E) ACKNOWLEDGEMENTS use the results from JM96-1205, is for the Q/P ratio to decrease to- Andrews, Jennings and Eberl acknowledge the assistance of the NAS-ESH program in which we also sampled at 100 yr/ ward the present. Hence, it does not funding their research on the Holocene ice sample intervals. MD99-2269 from show the late Holocene increase in rafting records. the North Iceland shelf is from an IRD proxies that characterize the area where drift ice from the East sites on either side of the Denmark NOTE Iceland Current frequently traverses Strait (Figs. 2B & C). Data used in this study will be available in the NOAA Paleoclimatology data base at: the site. The core was sampled at www.ncdc.noaa.gov/paleo/paleo.html ~30 yr/sample and covers the last Discussion and conclusions For detailed information, contact the 12 cal ka BP (Moros et al., 2006). Our four records (Fig. 2A-D) bear lit- authors. The easternmost core is MD95-2011, tle resemblance to the %HSQ from REFERENCES located on the Vorring Plateau, and south of Denmark Strait or from Andrews, J. T., 2000: Icebergs and Iceberg Rafted has an average sampling interval of the other sites (Fig. 1 & Fig. 2E), Detritus (IRD) in the North Atlantic: Facts and 20 yrs/sample. although our data are from sites Assumptions. Oceanography, 13: 100-108. The Orphan Knoll data (Fig. 2A) where the deposition of ice-rafted Bond, G., Kromer, B., Beer, J., Muscheler, R., Evans, M.N., Showers, W., Hoffman, S., Lotti-Bond, R., presently covers the last 10 cal ka sediments should be much more Hajdas, I., and Bonani, G., 2001: Persistent solar BP. Quartz and dolomite track each obvious because of their more ice- influence on North Atlantic climate during the other over this interval. Both miner- proximal locations. Our data do not Holocene. Science, 294: 2130-2136. Darby, D. A., 2003: Sources of sediment found in als show a progressive decrease in posses the amplitude of variation the sea ice from the western Arctic Ocean, new their weight% toward the present. noted in the HSQ data (although insights into processes of entrainment and drift There are no large-scale changes this is partly a function of the plot- patterns. Journal of Geophysical Research, 108: 13-11 to 13-10. doi: 10, 1111029/1112002JC1001 in either species and, surprisingly, ting scale and whether the data are 350, 1112003. there is no clear indication of an in- detrended or not) and they do not De Vernal, A., and Hillaire-Marcel, C., in press: Pro- crease in delivery associated with show a clear ~1.5 ka periodicity. In- vincialism in trends and high frequency changes in the northwest North Atlantic during the Holocene: the neoglacial readvance of the deed, the sites differ significantly Global Planetary Change. Greenland Ice Sheet. in their long-term trends (compare Eberl, D. D., 2004: Quantitative mineralogy of the The quartz weight% data from Figs. 2A & 2C). In each area, we Yukon River system: Variations with reach and the East Greenland margin, south of have results from other cores (e.g. season, and sediment source unmixing. American Mineralogist, 89: 1784-1794. Scorsby Sund, (Figs. 1 & 2B) shows Fig. 2C) that we are using to devel- Jennings, A. E., Knudsen, K. L., Hald, M., Hansen, C. a very slight trend to increase over op robust regional IRD signals. Of V., and Andrews, J. T., 2002: A mid-Holocene shift the last 4 cal ka BP but the aver- special interest on the Labrador and in Arctic sea ice variability on the East Greenland shelf. The Holocene, 12: 49-58. age input of quartz has remained East Greenland margins is whether Moros, M., Andrews, J. T., Eberl, D. D., and Jansen, relatively constant over the last 10 or not there is a coupling between E., 2006: Holocene history of drift ice in the cal ka BP. The counts of clasts >2 the IRD proxies associated with northern North Atlantic: Evidence for different spatial and temporal modes, Paleoceanography, mm on X-radiographs shows vari- grain-size and mineralogy (e.g. Fig. 21, PA2017, doi:10.1029/2005PA001214. able IRD inputs between 10 and 8 2B). This begs the question as to cal ka BP followed by a period with whether we can construct a proxy For full references please consult: virtually no IRD and then a marked for Holocene IRD that incorporates www.pages-igbp.org/products/newsletters/ref2006_2.html increase in IRD ca. 4 cal ka BP, con- all the necessary elements of both sistent with previous data from the grain-size and provenance. Given East Greenland shelf (Jennings et that the bulk of glacial and sea ice al., 2002). transported sediments is much finer Drift ice entrained in the East than 63 µm, then the XRD method Iceland Current historically im- offers significant advantages, espe- pacts the North Iceland shelf where cially if we can distinguish mixtures MD99-2269 was retrieved (Moros et of minerals that define source areas al., 2006). The quartz weight% data (Darby, 2003).

PAGES NEWS, VOL.14, N°2, AUGUST 2006 10 Science Highlights: U.S. ESH Program

Progress in the study of Asian Monsoon climate dynamics using dendrochronology

BRENDAN M. BUCKLEY*, ROSANNE D. D’ARRIGO, EDWARD R. COOK, GORDON C. JACOBY, WILLIAM E. WRIGHT Tree Ring Laboratory, Lamont-Doherty Earth Observatory, Columbia University, Palisades, USA; *[email protected]

Introduction tree-ring scientists from Lanzhou Scientists at the Tree-Ring Laborato- University in China, the Birbal ry of the Lamont-Doherty Earth Ob- Sahni Institute of Palaeobotany in servatory of Columbia University in Lucknow, India, and the Indian In- New York are entering the third year stitute of Tropical Meteorology in of a fi ve-year project on Tree-Ring Pune. In all of these locations, we Reconstructions of Asian Monsoon have defi ned our top priorities as Climate Dynamics (see www.ldeo. true collaboration, education, ca- columbia.edu/res/fac/trl/). The scope pacity building and technology of the project spans the entire Asian transfer, in order to insure that our region linked to or affected by the research will have a lasting impact monsoon, from Siberia, Mongolia in Asia. In turn, our Asian collabo- and North Asia, to Australasia and rators have had much to teach us Oceania, mainland Southeast Asia about the climate and environ- and the Indian subcontinent in the Figure 1: Map showing the Asian monsoon ment of their respective regions, south. The first two years have region that is the project focus. Red dots: and have been instrumental in fa- tree-ring data locations that were available heavily emphasized the building of at the start. Yellow dots: data that have been cilitating our ability to collect tree- tree-ring data networks for recon- contributed from various sources. Light blue ring samples. This project has also structing features of Asian monsoon dots: new data and chronologies that have benefi ted from contributors to the been developed over the past ca. two years climate and the factors infl uencing of fi eld work in monsoon Asia. NOAA Paleoclimatology Data Bank its variability (e.g., Indo-Pacifi c sea (www.ncdc.noaa.gov/paleo/paleo. surface temperature (SST) fi elds), Renewable Natural Resources Re- html) and from many individuals in part for the purpose of driving search Center in Jakar, Bhutan, are who have contributed data directly General Circulation Model (GCM) all fully functional. New programs to our project. estimates of monsoon climate are being developed at the Univer- variability with paleoclimate data sity of Los Baños in the Philippines, Developing a climate story spanning the past several centu- and at the University of Peradeniya Our efforts have already begun to ries. A dearth of tree-ring data from in Kandy, Sri Lanka. We have also yield results, as we have substan- the tropical regions prompted our developed close ties to regional tially increased our tree-ring data commitment to obtaining such data from the Asian tropics fi rst; hence our initial focus on fi eld investigations in tropical Asia.

Regional collaboration A project of this magnitude requires a strong commitment from regional collaborators, and the list of foreign participants that are integral to this project is far too exhaustive to present here. Importantly, we have helped to develop tree-ring laboratories at several institutions in Asia, and we have consistently worked to supply formal training in den- drochronological techniques. For example, the Laboratory of Tropi- cal Dendrochronology at Kasetsart University in Bangkok, the Tree Ring Laboratory at Mongolian National Figure 2: Indonesian warm pool SST reconstruction (D’Arrigo et al., in press). A shows actual (black) and reconstructed (red) September-November Indonesian warm pool SSTs. Several University in Ulan Baatar, a facility volcanic episodes during the 19th century are indicated with arrows. B shows running 31-year at the Institut Teknologi Bandung in correlation plots between reconstruction and instrumental data. Correlations are calculated for Java, Indonesia, and another at the both unﬁ ltered data (green) and 1st differenced transforms (blue) (Fig. by Robert J. Wilson).

PAGES NEWS, VOL.14, N°2, AUGUST 2006 Science Highlights: U.S. ESH Program 11

network for the Asian monsoon 2.00 region, particularly in the Asian tropics (see Fig. 1). It is critically important to the goals of our project 1.50 that we define the role of the tropics in driving or modulating notable cli- 1.00 mate changes over the past 500 to Indices 1000 years. We framed our project around analyzing three primary fea- 0.50 tures of climate variability that were identified by Meehl and Arblaster 0.00 (2002) as having links to variabil- 1600 1700 1800 1900 2000 Year ity in the strength of the monsoon; Indian Ocean SSTs, tropical Pacific Figure 3: Plot of the MHS teak chronology from northwestern Thailand, developed from living SSTs, and land temperatures over trees and 57 stumps that remain from logging at the beginning of the 20th century. We made individual tree chronologies using 4-7 radii from each tree. The growth reductions in the early the Tibetan Plateau and northern and mid-1700s mark periods of reduced monsoon strength that are likely related to warm SST Eurasia. We are sampling tree-ring anomalies in the eastern tropical Pacific, consistent with evidence from Dunbar et al (1994). sites that are influenced by one or more of these monsoon-related fea- Cannariatto and Lowell Stott, pers. monsoon through regional recon- tures. For example, Buckley et al. comm.) that indicates an unusually structions and modeling. (2005) demonstrated links between weak monsoon in central India at pine sites from Bhutan, India and this time. The speleothem record ACKNOWLEDGEMENTS Thailand, and all three of the above- also suggests enhanced monsoon This five-year project is supported by the National Science Foundation’s Paleoclimate mentioned features, while D’Arrigo activity throughout much of the Me- Program (Grant OCE 04-02474), and runs et al. (2005) modeled relationships dieval Climate Anomaly (MCA) pe- until the end of July 2009. between the North Pacific Index and riod that would be consistent with the tropics, and D’Arrigo et al. (in persistent La Niña-like cold SST REFERENCES press) developed a tree-ring and anomalies in the eastern tropical Buckley, B.M., Cook, B.I., Bhattacharyya, A., Dukpa, D., and Chaudhary, V., 2005: Global surface coral-based reconstruction of In- Pacific. We have yet to build con- temperature signals in pine ring width chronolo- donesian warm pool SSTs (Fig. 2a). tinuous tropical tree-ring chronologies from southern monsoon Asia. Geophysical Many old teak logs buried in allu- gies that extend back far enough to Research Letters, 32: L20704, doi:10.1029/ 2005GL023745, 2005. vial sediments may also yield ex- analyze the MCA, although we have D’Arrigo, R., Wilson, R., Deser, C., Wiles, G., Cook, E., tensions of our existing Indonesian collected sub-fossil teak river logs Villalba, R., Tudhope, A., Cole, J., and Linsley, B., chronologies by several centuries. and log coffin wood that date back 2005: Tropical-North Pacific climate linkages over the past four centuries. Journal of Climate, 18: Other efforts are resulting in new to this period. We are collaborating 5253-5265. climate-sensitive records from sites with Drs. Cannariatto and Stott to D’Arrigo, R., Wilson, R., Palmer, J., Krusic, P., Curtis, across our study region, in places compare isotopic records of these A., Sakulich, J., Bijaksana, S., Zulaikah, S., Ngko- imani, O., and Tudhope, S. in press: Reconstructed as diverse as Mongolia, the Philip- teak samples with their speleothem Indonesian warm pool SSTs from tree rings pines, and Laos. results, and see this as an important and corals: linkages with ENSO and the Asian Our studies are beginning to new avenue of research. monsoon. Paleoceanography, 21:, doi:10.1029/ clarify periods of reduced monsoon 2005PA001256, 2006. Dunbar, R.B., G.M. Wellington, M.W. Colgan and strength that appear to be linked to Summary Glynn, P.W., 1994: Eastern Pacific sea surface SST variability in the eastern tropi- These and other recent studies temperature since 1600 A.D.: The δ18O record of cal Pacific. For example, a 450-year demonstrate our ability to model climate variability in Galapagos corals. Paleocean- ography 9(2): 291-315. moisture-sensitive teak chronology past climate in the Asian monsoon Meehl, G. and Arblaster, J., 2002: The tropospheric that we have developed from north- region, and highlight the important biennial oscillation and Asian-Australian monsoon western Thailand (Fig. 3) shows role of tropical forcing on global rainfall. J. Climate, 15: 722-744. marked reductions in inferred mon- climate. Combined with our very soon strength in the early to middle long tree-ring records from extra- 1700s. This finding is consistent with tropical Asia, we anticipate millen- Dunbar et al. (1994), who showed nial-length coverage at several key persistent warm SST anomalies in locations across the range of our the eastern tropical Pacific during developing Asian network. This re- this time, based on coral records search is ongoing but gives us hope from the Galapagos Islands. It is for some very exciting results in the also consistent with the expected near future. Over the remaining effects of persistent El Niño condi- three years of this grant, we an- tions on rainfall in this area of Thai- ticipate that our efforts will result land, and agrees with a speleothem in an increased understanding of record from central India (Kevin the mechanisms behind the Asian

PAGES NEWS, VOL.14, N°2, AUGUST 2006 12 Science Highlights: U.S. ESH Program

Solar forcing of the tropical Paciﬁc climate and impacts over North America for the last millennium

AMY C. CLEMENT1, JULIEN EMILE-GEAY2, RICHARD SEAGER2, MARK CANE2 AND MICHAEL N. EVANS3 1Rosenstiel School of Marine and Atmospheric Sciences, University of Miami, USA; [email protected] 2Lamont-Doherty Earth Observatory, Columbia University, Palisades, USA; [email protected], [email protected], [email protected] 3Laboratory of Tree-Ring Research, University of Arizona, Tucson, USA; [email protected]

Introduction �� It is now well known that the state �� of the tropical Paciﬁc climate has � �� a large inﬂuence on the global climate, as evidenced by the inter- ��

annual variations associated with � the El Niño/Southern Oscillation (ENSO) and its global teleconnec- �� tions. If the mean state of the tropi- ��

cal Paciﬁc were to undergo shifts �� on longer timescales, it is likely

that such changes would also have � global impacts. This possibility has �� prompted debate about whether � � the tropical Paciﬁc has undergone � � � changes over the 20th century that � � are akin to the warm phase (El Niño) �� or cold phase (La Niña) of ENSO, � � and also whether such shifts will be �� part of the climate response to in- � creased greenhouse gases (GHGs) �� in the future. At present, state-of- the-art coupled ocean-atmosphere � general circulation models (GCMs) � simulate different responses in the � tropical Paciﬁc climate to increased � GHGs, with some models predict- �� ing a decreased east/west sea sur- ��

face temperature (SST) gradient �� (akin to El Niño), others predicting �� an increased gradient (akin to La �� Niña), and still others predicting no �� change (Collins et al., 2005). Given �� this uncertainty, it is reasonable to turn to the past to learn about how Figure 1: Top Panel: Solar forcing over the last millennium, from Crowley (2000). Units are Wm-2. Middle Panel: NINO3 SST index simulated by the ZC model in response to the solar forc- the tropical Pacific responded to cli- ing shown in the top panel (blue curve). The green curve shows the NINO3 value reconstructed mate forcing, and whether the influ- using the methodology of Evans et al. (2002) applied to a network of 65 corals (reported by ence of the tropical Pacific can be Evans et al., 2001). Both curves are normalized and smoothed with a 40-year low-pass filter. Bottom panel: Palmer Drought Severity index for the western U.S. as reconstructed from the seen in past regional climate prox- Cook et al. (2004) gridded data. Negative values indicate drought conditions and positive values ies around the globe. wet conditions. The data are normalized and smoothed with a 40-year low-pass filter.

Modeling results the solar forcing as reconstructed member) are shown in Figure 1b In this paper, we present results by Crowley (2000). This forcing is (blue line). There is a clear anti-cor- from several recent studies, many imposed on the Zebiak-Cane (ZC) relation between the solar forcing of which have been supported coupled ocean-atmosphere model, and the value of NINO3 (average under the National Science Foun- which simulates anomalies about SST anomaly over the region 5°S- dation’s Earth System History pro- the current climate in the tropical 5°N and 150-90°W) such that when gram, in order to illustrate how Paciﬁc. Results from a 100-member the solar irradiance is higher, the the tropical Paciﬁc climate has re- ensemble (in which the forcing is mean NINO3 value is negative (akin sponded to solar forcing over the identical but initial conditions are to a La Niña with an increased east- last 1000 years. Figure 1a shows varied slightly from member to west gradient), and when irradiance

PAGES NEWS, VOL.14, N°2, AUGUST 2006 Science Highlights: U.S. ESH Program 13

the 20th century, while the data do not show such a trend, perhaps because changes in greenhouse gas concentrations are not modeled. There are other coral records that support the picture of a warm eastern equatorial Pacific (or reduced east-west gradient) in the mid-17th to mid-18th century. Cobb et al. (2003) found relatively light values for δ18O (warm, fresh conditions) in corals from Palmyra Island during the 17th century. They also found relatively heavy values (cold, salty) during the 12th to 13th centuries, a time when the solar forcing was high. If the modern behavior of δ18O on interannual timescales is Figure 2: Schematic representation of the ‘thermostat’ mechanism (see text) proposed by extended to the past, this would in- Clement et al. (2006). Figure courtesy of Lowell Stott. dicate a reduced east-west gradient for the 17th to 18th centuries, and is lower, NINO3 is higher (akin to the Basin. Is there any evidence of an increased gradient for the earlier an El Niño with a decreased gradi- such behavior in the paleoclimate medieval period, consistent with the ent). It should be noted that there record? model results. Damassa et al. (2006) is currently considerable debate as also showed that coral records from to the magnitude of the centennial Paleoclimate observations the western Indian ocean indicate timescale variations in the solar ir- An extremely important archive that SSTs were warmer during radiance, so these results should be of past behavior of ENSO and the the mid-17th to mid-18th century. interpreted only qualitatively. mean state of the tropical Pacifi c Again, if the modern ENSO analogy The model results can be ex- comes from both living and fos- is extended to the past, this would plained by a mechanism that has sil corals. Evans et al. (2002) have imply a reduced east-west gradient been dubbed the ocean dynami- used a network of coral records to at that time, also consistent with the cal thermostat (Clement et al., reconstruct SST from the tropical model results. 1996), in which a heating of the Pacifi c over the last several centu- This collection of model results tropical Pacific ocean leads to a ries. The reduced space objective and coral data from the tropical cooling of the NINO3 region (il- analysis methodology of Evans et Pacifi c appears to suggest that so- lustrated schematically in Fig. 2). al. (2002) was applied to a larger lar forcing has an infl uence on the This result arises from the differ- number of coral records (Evans et tropical Pacifi c climate in a manner ent surface temperature response al., 2001), and the resulting NINO3 consistent with the ocean dynami- in the eastern and western Pacifi c: reconstruction is plotted in Figure cal thermostat. The implications of In the west, where the thermocline 1b (green line). Agreement between this may extend beyond the Pacifi c is deep, the response to surface independently-reconstructed NINO3 Basin. Seager et al. (2005) and Her- heating is largely thermodynamic and model-simulated NINO3 is en- weijer et al. (2006a) have used ob- and the mixed layer adjusts with an couraging; both lines of evidence servations and model simulations increased temperature. In the east suggest a maximum value in the to argue that decadal timescale La where the thermocline is shallow, mid- to late-17th century, coincid- Niña-like SST changes were respon- cooling by vertical advection offsets ing with the Maunder Minimum, sible for the six severe multi-year the surface heating, producing a and a shift towards lower values of North American droughts since the smaller temperature response. The NINO3 from that time until through mid-19th century. Herweijer et al. increased zonal sea surface tem- the 20th century. There are, howev- (submitted) used a gridded drought perature (SST) gradient acceler- er, notable differences between the reconstruction based on tree-ring ates the trade winds, which leads model and data on multi-decadal data from Cook et al. (2004) to show to further thermocline shoaling and timescales which may be due to the existence of ‘mega-droughts’ cooling by vertical advection in the defi ciencies in both products. We during the medieval period that east, further accelerating the winds. must assume that the interpreta- had a spatial pattern very similar This ‘climatological’ Bjerknes feed- tion of coral-reconstructed SST to modern droughts. They argued back, akin to that which operates has fi delity on timescales longer that the temporal drought variabil- on interannual timescales, leads to than are available for statistical ity was similar to that of the 20th a cooling of the eastern equatorial calibration. The model simulates century but that it occurred around Pacifi c in response to a heating of a trend of decreasing NINO3 over a drier mean state, such that peri-

PAGES NEWS, VOL.14, N°2, AUGUST 2006 14 Science Highlights: U.S. ESH Program

ods of, for example, Dust Bowl level precipitation (Schubert et al., 2004; be so large. Perhaps the tropical aridity, lasted for 20-40 years. Gra- Seager et al., submitted; Sutton Pacific is part of the answer but it ham et al. (submitted) have used and Hodson, 2005), so there is no may not be the whole story. proxies to estimate medieval SSTs reason to believe that conditions and demonstrated that when they in the tropical Pacific are a perfect NOTE are used to force an atmospheric predictor of central U.S. rainfall. It Data used in the study can be found in the ‘data’ section of the authors research page GCM, drought results over North should also be noted that the ob- at http://www.ldeo.columbia.edu/~julieneg/ America. The altered mean state servations of tropical Pacific SSTs research.html. of the medieval period, not only in are quite sparse prior to the 20th North America but also in other EN- century, and coral data for the last REFERENCES Clement, A.C., Seager, R., Cane, M.A., and Zebiak, S.E., SO sensitive regions (e.g., East Af- 1000 years are critical to filling in 1996: An ocean dynamical thermostat. J. Climate, rica, northern South America, South these gaps. Finally, there is the is- 9: 2190-2196. Africa, the Sahel and the Indian sue of the magnitude of solar ir- Emile-Geay, J., Cane, M., Seager, R., Kaplan, A., and Almasi, P., accepted: ENSO as a mediator of the ocean; Herweijer et al., submitted), radiance changes. While the ZC solar influence on climate, Paleoceanography. is consistent with a solar-induced model is fairly sensitive to even Evans, M.N., Kaplan, A. and Cane, M.A., 2002: Pacific La Niña-like state in the tropical Pa- small solar forcing (Emile-Geay sea surface temperature field reconstruction from cific. The tree ring data also show et al., accepted), if the mid-range coral delta O-18 data using reduced space objective analysis. Paleoceanography, 17, Art. No. 1007. -2 that the Little Ice Age was a rela- value of 0.5 Wm is accepted, the Herweijer, C., Seager, R. and Cook, E.R., 2006a: North tively wet period in the central U.S. model response is quite small American droughts of the mid to late nineteenth compared with the earlier medieval (~0.15°C). Thus, while the link- century: a history, simulation and implication for Mediaeval drought, The Holocene, 16(2): 159-171. period (Fig. 1c), consistent with a ages explored here might have Seager, R., submitted: The turn of the century drought shift towards a reduced east-west qualitative support from models across North America: Dynamics, global context gradient, as shown in Figure 1b. and paleoclimate observations, it and prior analogs. J. Climate. is difficult to argue for such large For full references please consult: Final points quantitative changes in the climate www.pages-igbp.org/products/newsletters/ref2006_2.html Of course, a number of caveats system from solar forcing alone. apply here. First, model studies Thus remains the centuries-old have shown that Atlantic SSTs problem of why the climate re- exert an influence on central U.S. sponse to solar forcing appears to

The mystery interval 17.5 to 14.5 kyrs ago

GEORGE H. DENTON1, WALLACE S. BROECKER2 AND RICHARD B. ALLEY3 1Department of Earth Sciences and Climate Change Institute, University of Maine, Orono, USA; [email protected] 2Lamont-Doherty Earth Observatory, Columbia University, Palisades, USA 3Pennsylvania State University, University Park, USA

The time period between the be- Ocean (Bond et al., 1992; Hem- • The Asian monsoons were no- ginning of Heinrich event #1 (H-1) ming, 2004). tably weakened (Wang et al., and the onset of the Bølling/Allerød • Export of 231Pa from the Atlantic 2001). rivals the Younger Dryas in im- to the Southern Ocean ap- • The Atlantic Intertropical Conver- portance to our understanding of pears to have come to a halt, gence Zone (ITCZ) was locked how the planet responds to abrupt signaling a shutdown of North into a southerly position (Pe- mode switches. This interval also Atlantic meridional overturn- terson et al., 2000; Lea et al., constitutes the onset of the most ing circulation (McManus et 2003). recent termination, arguably the al., 2004). • Temperate mountain glaciers in most fundamental climate shift of • The northern Atlantic Ocean and both hemispheres underwent the last 100-kyr glacial cycle. As the Mediterranean Sea be- major recession that began some of the responses during this came unusually cold (Bard et about 17.5 kyr (Denton et al., time appear to be mutually con- al., 2000; Cacho et al., 2001). 1999; Schaefer et al., 2006). By tradictory, we term it the “Mystery • Conditions in Greenland were 14.5 kyr at the end of the Mys- Interval”. heavily weighted toward hy- tery Interval, Swiss mountain percold winters (stadial GS-2a; glaciers had withdrawn to the Key observations Denton et al., 2005). inner Alps (Schlüchter, 1988). Notable events during the 3000- • Northwestern Europe was simi- • Closed-basin lakes in North year-long Mystery Interval are as larly dominated by hypercold America’s Great Basin reached follows (Fig. 1). winters (Late-Pleniglacial; their greatest size (Benson, • The H-1 iceberg armada was dis- Renssen and Isarin, 2001). 1993; Garcia and Stokes, charged into the North Atlantic 2000).

PAGES NEWS, VOL.14, N°2, AUGUST 2006 Science Highlights: U.S. ESH Program 15

• The African rift valley suffered maximum aridity (Johnson et al., 1996). • The 14C to C ratio in the atmosphere dropped by 190 per mil (Broecker and Barker, in press). • Deglacial warming of Antarctica and the rise of atmospheric

CO2 content became pronounced (Monnin et al., 2001).

Possible explanation How these events fi t together is a major puzzle. One seeming contradiction may afford an important clue; namely, extensive deglaciation of the European Alps occurred during a time of maximum cold in the adjacent Atlantic Ocean and Mediterranean Sea, along with some indications of continued cold conditions, at least during winter, on the continent itself. This conclusion comes from oxygen-isotope curves of carbonate from lakes and mires that allow European climate oscillations to be tied directly to the Greenland template (e.g., von Grafenstein et al., 1999). Such records from the Swiss Alps show that extensive glacier recession had already occurred prior to the distinctive earliest Bølling isotope switch, which was accompanied by a “juniper jump” in pollen profi les (Eicher and Siegenthaler, 1976). It is important to note from isotope chronologies that both Andean and New Zealand mountain glaciers abandoned their innermost Last Glacial Maximum (LGM) positions and also suffered extensive retreat during the Mystery Interval (Den- ton et al., 1999; Strelin and Denton, 2005; Schaefer et al., 2006). Assuming that our read of the evidence is correct, we can think of only one explanation for this contradiction; namely, that the cold oceanic and terrestrial condi- Figure 1: The Mystery Interval is marked by yellow background. A shows deuterium and CO2 records from the EPICA Dome C ice core on the East Antarctic plateau (Monnin et al., 2001). tions refl ect extensive winter sea- Chronology is based on Schwander et al. (2001). B gives the GISP2 isotope record (Stuiver ice cover in the northern Atlantic and Grootes, 2000). C illustrates ice-rafted grains and magnetic susceptibility in core SU-8118 and that the retreat of temperate off Portugal (Bard et al., 2000). IRD is expressed as the number of grains per gram for the size fraction greater than 150 µm. The H-1a IRD peak features quartz and feldspars, with mountain glaciers refl ects warm- minor hematite-coated grains, glauconite, and volcanic shards. The H-1b IRD peak is largely ing summer conditions brought detrital carbonate, with minor amounts of quartz and feldspar (Bard et al., 2000). D is SST on by rising atmospheric CO2. If from two calibrations for alkenone unsaturation ratios from core SU-8118 (Bard et al., 2000). E is a 231Pa/230Th profi le from core GGC5 from the Bermuda Rise (McManus et al., 2004). YD: this is indeed the explanation, then Younger Dryas, B/A: Bølling/Allerød, MI: Mystery Interval, LGM: Last Glacial Maximum. Figure we suggest the following scenario: adapted in part from McManus et al. (2004). The Mystery Interval was initiated

PAGES NEWS, VOL.14, N°2, AUGUST 2006 16 Science Highlights: U.S. ESH Program

by the melting of the H-1 iceberg perature and of atmospheric CO2. tent of the closed-basin lakes in the armada. This melting reduced the A consequence of this shutdown Great Basin and the dessication of salinity of the northern Atlantic to of northern overturning appears to equatorial Africa’s Lake Victoria. the point where deep water could have been an overall reorganiza- Hostetler et al. (1999) found such no longer form. The resulting tion of ocean circulation, probably a signature for modeled water- shutdown of meridional overturn- through the operation of a bipolar balance anomalies in response to ing circulation was so severe that thermohaline seesaw (Broecker, speciﬁed changes in North Atlan- the export of 231Pa from the North 1988), that led to shrinkage of the tic sea-surface temperatures at the Atlantic was reduced by an even sea-ice fringe around Antarctica. As time of Heinrich event #2 (H-2). larger extent during the Mystery a result, Antarctica warmed while One last occurrence merits

Interval than during either the LGM atmospheric CO2 slowly increased, mention. As shown by Adkins and or the Younger Dryas (McManus et perhaps due to more efficient de- Schrag (2003), during peak glacial al., 2004). As a consequence of this gassing from the Southern Ocean time, an abyssal ocean reservoir shutdown, the northern North At- (Stephens and Keeling, 2000). with one gram per liter higher salt lantic cooled, allowing the spread These events were curtailed when content than the rest of the deep of winter sea ice. Such expansion northern overturning resumed dur- ocean appears to have existed. of sea ice, perhaps as far south ing the Bølling/Allerød. They were Their idea is that the reservoir as 48°N, is required for the se- then renewed during the Younger owed its existence to the input vere winter cooling that produced Dryas, which in many ways was a of dollops of brine released from widespread periglacial features in shorter repetition of the Mystery Antarctica’s greatly expanded gla- northwestern Europe (Renssen and Interval. cial sea-ice apron. If, as the result Isarin, 2001). In addition, winter sea It is far from clear why the shut- of its isolation from the rest of the ice is almost certainly the primary down from the H-1 iceberg armada ocean, this reservoir were greatly cause of depressed mean annual could have had a greater impact depleted in radiocarbon, it could temperatures in Greenland, which than those associated with the five help to explain why the 14C to C ra- are highly skewed toward winter earlier Heinrich armadas. A possi- tio of the glacial atmosphere was (Denton et al., 2005). As shown by ble clue is that terminations begin so much higher than that for the Chiang and Bitz (2005), widespread at what would intuitively seem to late Holocene. Even more impor- winter sea ice in the northern At- be unlikely times; namely, just as tantly, the demise of this reservoir lantic not only cools the northern ice sheets achieve maximum vol- caused by the reduction in brine high-latitude region, but pushes umes at the culmination of the supply related to the retreat of the southward the Atlantic sector of long buildup leg of an asymmetric ice margin could explain the large the ITCZ. Furthermore, according 100-kyr cycle. In fact, Raymo (1997) drop in the atmosphere’s 14C to C to Barnett et al. (1988), the conse- suggested from marine benthic ratio during the course of the Mys- quent increase in the duration of oxygen-isotope records that the tery Interval (Broecker and Barker, Asian snow cover lessens the ef- prerequisite for a sharp and com- in press). fectiveness of the summer warm plete termination was “excess” ice season in heating the continent, stored in northern ice sheets. Once REFERENCES hence weakening summer mon- excess ice volume is achieved, ma- Bard, E., Rostek, F., Turon, J.-L., and Gendreau, S., 2000: Hydrological impact of Heinrich events in the soons. It thus appears likely that jor deglaciation awaits only the northeast Atlantic. Science, 289: 1321-1323. severe winter conditions in the next rise of northern summer in- Denton, G.H., Alley, R.B., Comer, G.C., and Broecker, northern latitudes during the Mys- solation. Does this suggest that the W.S., 2005: The role of seasonality in abrupt climate change. Quaternary Science Reviews, 24: tery Interval, originating from the collapse of such excess ice into the 1159-1182. shutdown of North Atlantic over- North Atlantic is what caused the McManus, J.F., Franciois, R., Gherardi, J.-M., turning and spread of winter sea H-1 ice armada to be so effective in Keigwin, L.D., and Brown-Ledger, S., 2004: Col- lapse and rapid resumption of Atlantic meridional ice, afforded the physical linkage triggering the last termination? circulation linked to deglacial climate change. between millennial-scale Green- Princeton GFDL’s Isaac Held Nature, 428: 834-837. land, European, and North Atlantic (2006) makes the case that the Monnin, E., Indermühle, A., Dällenbach, A., Flückiger, temperature fluctuations; the ITCZ; warmer climate to be generated by J., Stauffer, B., Stocker, T.F., Raynaud, D., and Barnola, J.-M., 2001: Atmospheric CO2 concentra- and Asian monsoons (Denton et the ongoing buildup of greenhouse tions over the last glacial termination. Science, al., 2005). gases will lead to even wetter con- 297: 112-114. The relative timing of northern ditions in the tropics and even drier Stuiver, M., and Grootes, P., 2000: GISP2 oxygen-isotope ratios. Quaternary Research, 53: 277-284. and southern events shown in Fig- conditions in the deserts of the ex- ure 1 suggests that the shutdown tra tropics. The converse of this is For full references please consult: of meridional overturning in the that in the colder glacial world, the www.pages-igbp.org/products/newsletters/ref2006_2.html North Atlantic during the Mystery opposite should have been true. If Interval may have been the major so, then the cold ocean conditions factor in driving the last termina- and the enhanced sea ice cover tion, which in panel A of Figure 1 is of the Mystery Interval might ac- marked by the rise of Antarctic tem- count for the unusually large ex-

PAGES NEWS, VOL.14, N°2, AUGUST 2006 Science Highlights: U.S. ESH Program 17

Delayed onset of the South American Summer Monsoon during the Last Glacial Maximum

KERRY H. COOK AND ED K. VIZY Department of Earth and Atmospheric Sciences, Cornell University, Ithaca, USA; [email protected]

Introduction Climate conditions in the Amazon basin during the last glacial maximum (LGM, about 21,000 years ago) are actively discussed in the literature. Faunal evidence, noble gas analyses, lake status reconstructions, pollen counts, atmospheric dust, and charcoal records have all been interpreted to provide information about LGM surface moisture and temperature. Much of the evidence suggests that large-scale climate conditions were drier and cooler in the Amazon basin but there is great uncertainty and little quantitative guidance. Proxy data from the high Andes and the Nordeste region of Brazil, e.g., indicate that some regions may have been wetter during the LGM. Differences in Amazon climate during the LGM are of great interest in the light of current pressing climate change issues. Fundamental questions remain about the stability of tropical climate, and the infl uence of climate and deforestation on the diversity of tropical vegetation. drier during LGM wetter during LGM Methodology There are two ways of learning about past climates, namely, inference Figure 1: Model domain and total annual precipitation difference between the present day -1 based on physical evidence, and and the LGM simulations. Shading interval is 1 mm day , and the 0 line is contoured. Marks denote LGM precipitation differences from present day interpreted from various proxy recon- computer simulation. Each method struction sites has its own strengths and weakness- es. Physical evidence is undeniably evidence for a more direct compari- well, providing a defi nite improve- real but there may be uncertainties son. ment over global model simula- associated with interpretation and Here we use an RCM with 60 km tions, which typically have 200 km dating. Climate modeling provides horizontal resolution and 24 verti- resolution and diffi culties resolving complete information about the cal levels, coupled with a potential the steep Andean topography (Vizy climate state but there are always vegetation model (PVM; Oyama and Cook, 2005). concerns about accuracy. Both ap- and Nobre, 2004). The model do- Paleoclimate application of re- proaches to studying paleoclimate main, which is unusually large, is gional models is complicated by are needed and the most powerful shown in Figure 1. A realization of the need to specify lateral boundary way of reconstructing past climate the present day annual climatology conditions for wind velocity, tem- is to use the two methods in a com- was produced by initializing with perature and moisture throughout plementary way, so that they test, the observed climatology for 00Z 15 the depth of the troposphere; such challenge and correct each other. October and running through 382 conditions are not available from The most effective way of doing this days, with a 1-minute time step and the geological record. Here, pres- is to use a regional climate model vegetation types held fi xed at today’s ent day lateral boundary conditions (RCM), which is capable of fi ner res- values. A second year was run to based on modern observations are olution than a global model, to bring verify that the climatology is stable. used for both the present day and the paleoclimate simulation closer to The regional model replicates the LGM simulations. This is justifi ed by the spatial scales of the geological observed climate of South America extensive testing of the sensitivity of

PAGES NEWS, VOL.14, N°2, AUGUST 2006 18 Science Highlights: U.S. ESH Program

the simulation to the lateral bound- rainfall rates are similar to those of ary conditions (Vizy and Cook, 2005), the present-day climate. and is likely successful because of Figures 2a and b show the veg- the large domain and the tropical etation distributions from the pres- location. But remote influences on ent-day and LGM simulations. The climate, e.g., in association with the good agreement between the mod- North American ice sheets or sea ice eled and observed vegetation (Fig. in the North Atlantic, are not includ- 2a) is a powerful validation of the ed. Such influences can be repre- RCM’s ability to capture the pres- sented by imposing lateral boundary ent-day climate accurately. In the conditions from global model simu- LGM simulation (Fig. 2b), rainforest lations but we find that our present vegetation in the central and west- day simulation is seriously degrad- ern Amazon is maintained despite ed by this lateral boundary forcing, the shorter rainy season and longer since an accurate simulation of this dry season. Rainforest in the eastern region’s climate is a particular chal- Amazon, however, cannot be main- lenge for the lower-resolution global tained under LGM conditions, and models. In any case, remote forcing the overall extent of the rainforest is may be somewhat influential and smaller than in the present day. warrant future consideration but it is not a primary determinant of South Conclusions American climate (Lenters and Cook, Simulations of South America dur- 1995). ing the LGM with a regional climate The LGM climate is simulated by model coupled with a potential veg- making minor changes in the Earth’s etation model produce a quantitative orbital parameters, reducing the at- picture of LGM climate and vegeta-

mospheric CO2 concentration from tion distributions that is consistent 330 to 200 ppmv, and by specifying with the geological proxy data. The LGM sea surface temperatures fol- model results associate drier condi- lowing Paul and Schäfer-Neth (2003). tions in LGM South America with Use of the PVM provides indepen- a 2-3 month delay in the onset of dence from having to specify LGM the monsoon. Such a shorter, but vegetation from geological recon- equally intense, South American struction. The output of the coupled Figure 2: Potential vegetation model vegeta- monsoon maintains a smaller Ama- RCM/PVM integration consists of a tion distributions from (A) present day and (B) zon rainforest through the glacial pe- LGM simulations. mutually-adjusted solution for both riod, perhaps with modified species the LGM climate and the LGM veg- soon onset dries the surface and populations. At the same time, this etation. decreases evaporation. scenario is consistent with evidence The monsoon onset delay is primar- of higher charcoal amounts (i.e., in- Results ily associated with cool conditions in creased fires) and atmospheric dust Figure 1 shows simulated differ- the tropical Atlantic Ocean. Spring- loading due to an extended dry sea- ences in annual precipitation for time inflow from the tropical Atlantic son. the “LGM” minus “present day” into the Amazon basin is much drier simulations. In most parts of the in the LGM simulation because of re- REFERENCES Amazon basin, total annual rainfall duced evaporation from the cooler Cook, K. H., and Vizy E.K., 2006: South American climate during the Last Glacial Maximum: in the LGM simulation is 25-35% surface waters, so the latent heat Delayed onset of the South American monsoon, lower than in the present-day simu- content and, therefore, the moist J. Geophys. Res., 111: D02110, doi:10.1029/ lation, but there are some regions static energy of the low-level air, 2005JD005980. Lenters, J. D., and Cook, K.H., 1995: Simulation and with precipitation increases. A de- is lower than in the present day in diagnosis of the regional South American precipi- layed onset of the South American August, September, and October. tation climatology, J. Climate, 8: 2988-3005. monsoon by 2-3 months is primarily In addition, moist static energy lev- Oyama, M. D., and Nobre, C.A., 2004: A simple potential vegetation model for coupling with the responsible for the simulated LGM els during the winter months of the simple biosphere model (SiB). Revista Brasileira de aridity (Cook and Vizy, 2006). The LGM are lower than in the present Meteorologia, 19: 203-216. annual mean surface temperature day because land surface tempera- Paul, A. and Schäfer-Neth, C., 2003: Modeling the over subtropical South America is tures are colder. The result is a delay water masses of the Atlantic Ocean at the Last Glacial Maximum, Paleoceanography, 18(3), 1058 only slightly cooler than in the pres- in the build-up of the low-level moist doi: 10.1029/2002PA000783. ent day simulation. The land surface static energy in the spring, keeping Vizy, E.K., and Cook, K.H., 2005: Evaluation of LGM is 1-2 K cooler for most of the year the vertical column stable and inhib- SST reconstructions through their influence on South American climate. J. Geophys. Res., 110: but warms significantly during the iting convection. When the monsoon D11105, doi:10.1029/2004JD005415. spring, since the delay in the mon- finally begins in the LGM simulation,

PAGES NEWS, VOL.14, N°2, AUGUST 2006 Science Highlights: U.S. ESH Program 19

Large lake drilling projects supported by U.S. National Science Foundation Earth Systems History Program

SHERI FRITZ1, THOMAS JOHNSON2, PAUL BAKER3, STEVE COLMAN2, WALTER DEAN4 AND JOHN PECK5 1University of Nebraska, Lincoln, USA; [email protected] 2University of Minnesota, Duluth, USA; [email protected]; [email protected] 3Duke University, Durham, USA; [email protected] 4U.S. Geological Survey, Denver, USA; [email protected] 5University of Akron, USA; [email protected]

dating to the last summer solar mini- The U.S. National Science Founda- mum (~32 ka BP), is coincident with tion’s Earth Systems History (ESH) MIS5e, the penultimate interglacial Program, in cooperation with the stage (~125 ka BP). This suggests that International Continental Drilling the water balance of the lake is as Program (ICDP) and the national strongly influenced by global-scale science foundations of international (and tropical) temperature changes collaborators, has funded the drill- and boundary conditions as by pre- ing of several lacustrine basins that cession forcing of the South Ameri- contain key continental archives of Figure 1: The GLAD800 raft on Lake Titi- can summer monsoon. climate and tectonic history, includ- caca, Bolivia/Peru ing Great Salt Lake and Bear Lake elevation (3812 m) lake in the tropi- (USA 2000), Lake Titicaca (Bolivia/ cal Andes of Bolivia and Peru, and Lake Bosumtwi drilling project John Peck (University of Akron, USA), Chris- Peru 2001), Hvitarvatn/Hestvatn/ was drilled in April/May 2001 using tian Koeberl (University of Vienna, Austria), Huakadalsvaten (Iceland 2003), Lake the GLAD 800 drilling platform and John King (University of Rhode Island, USA), Bosumtwi (Ghana 2004), Lake Ma- coring system (Fig. 1). This site is Bernd Milkereit (University of Toronto, Canada), Jonathan Overpeck (University of Arizona, lawi (Malawi 2005), Qinghai (China critical for reconstructing both the USA), Christopher Scholz (Syracuse Univer- 2005), and Lake Péten Itzá (Guate- timing of regional glaciations and sity, USA). mala 2006). These lake drilling proj- the history of the South American Lake Bosumtwi (6°N, 1°W) occupies ects mark the onset of what could summer monsoon system, and for a 1.07 Ma impact crater located in become a systematic acquisition of evaluating how long-term variation Ghana, West Africa. Bosumtwi is one continuous high-resolution records in glaciation and effective moisture of only four craters with tektites and of continental climate change ex- in the tropical Andes is affected by contains a thick (300 m) post-impact tending through multiple glacial- global-scale glacial boundary condi- lacustrine sediment sequence. This interglacial cycles. The projects are tions, orbitally produced changes in hydrologically closed lake is located generating multi-proxy records of seasonal insolation, and changes in beneath the path of the seasonal past temperature and hydrological tropical Pacific and/or Atlantic sea- migration of the ITCZ and is ideally change that can be placed in the surface temperature. The longest situated to provide a long record context of the rich climatic history recovered sequence spans 136 m and of North African monsoon variabil- derived from the highly successful consists of alternations between two ity. In addition, the anoxic bottom ocean and ice drilling programs. The primary lithologic units, indicative of water allows for the preservation of ultimate goal of the lake drilling pro- four major regional stages of glacial finely laminated sediments and the grams is not to confirm the influence advance and retreat. Generally, the potential for high-resolution (annual) of large-scale boundary conditions intervals of increased glaciation were paleoclimate reconstruction. A com- on global climate, but rather to be- periods when Lake Titicaca was deep bined sediment and rock-drilling pro- gin to delineate spatial and temporal and fresh (based on calcium carbon- gram was undertaken in summer deviations from the global mean on ate concentrations, diatom stratig- 2004. Using the GLAD800 lake drill- the continents where we live, provid- raphy, and δ13C isotopic measure- ing system, five sites were drilled ing important paleoclimate time se- ments on bulk organic carbon), thus, along a water-depth transect, yielding ries to test and improve our models regional glacial periods were times a total sediment recovery of 1833 m. and understanding of global climate of high effective moisture and likely Drilling in the crater’s annular moat dynamics. Below, we highlight sci- high precipitation. A chronology recovered 291 m of sediment. The ence objectives and results to date based on radiocarbon, U-series ages lowermost lacustrine sediment over- from a few of these projects, many on aragonite laminae, and tuning to lying impact breccia and lapilli is a of which are still in the early stages the Vostok CO2 record suggests that bioturbated carbonate mud with gas- of analysis. the drilled sequence extends over tropod shells, indicating that a shal- approx. the last 370,000 years. The low-water oxic lake was established Lake Titicaca drilling project most recent period of ice expansion in the crater. Much of the overlying Paul Baker (Duke University, USA), Sheri Fritz in the cordillera surrounding the lake 291 m of mud is laminated. Initial (University of Nebraska, USA), Geoff Seltzer began approx. 60 14C ka BP following study has identified two contrast- (Syracuse University, USA), Mark Bush (Florida Institute of Technology, USA). a major dry interval. The penultimate ing lithologies that delineate glacial/ Lake Titicaca (17°S, 69°W) is a high- low stand of Lake Titicaca, rather than interglacial variability and are read-

PAGES NEWS, VOL.14, N°2, AUGUST 2006 20 Science Highlights: U.S. ESH Program

ily correlated to marine sedimentary tions with oxygenated bottom waters records. Increased boreal summer that apparently existed for extended insolation and the resulting intensi- times in the past. One of the surpris- fied summer monsoon produce a ing results to date was to find that positive moisture balance. Organic- much drier conditions existed prior rich sediment with a low-coercivity to 75,000 years ago than during the magnetic mineralogy accumulated Last Glacial Maximum. during these wet intervals. At times of decreased summer insolation and Lake Qinghai drilling project the accompanying weaker summer Zhisheng An (Chinese Academy of Sciences, monsoon, lake levels fell and a min- Figure 2: The drilling barge Viphya on Lake China), Steve Colman (University of Minnesota Malawi Duluth, USA), Gerald Haug (GeoForschun- eral-rich sediment with lower organic δ18O values (average = -5.8‰), indica- gsZentrum Potsdam, Germany), Peter Molnar content accumulated. During these (University of Colorado, USA), and Takayoshi tive of strongly evaporitic conditions dry periods, aerosol dust export from Kawai (Nagoya University, Japan). when the lake disconnected from Lake Qinghai (36°N, 100°E), in the Sahel increased greatly and con- the river and retracted into a topo- the People’s Republic of China is tributed to a high-coercivity magnetic graphically closed basin. The chang- extremely sensitive to changes in mineral assemblage. Ongoing sedi- ing paleoenvironments inferred climate because it lies in a critical mentary studies continue to address from these observations are gener- transitional zone between the humid paleoclimatic questions regarding ally consistent with other records of climate region to the southeast and variability in the North African mon- glacial-interglacial fluctuations in the dry inland region of the Tibetan soon and teleconnections between the western U.S., with an overprint Plateau. Three major atmospheric high and low latitudes. of catchment-scale processes. circulation systems affect its climate: (i) the winter monsoon, induced by Bear Lake, Utah and Idaho Darrell Kaufman (Northern Arizona University, Lake Malawi drilling project Siberian high pressure, (ii) tropical USA), Walter Dean (U.S. Geological Survey, Christopher Scholz (Syracuse University, USA), moisture from low latitudes, carried Thomas Johnson (University of Minnesota USA), Jordon Bright (Northern Arizona Uni- by the Asian summer monsoon, and versity, USA), and Joseph Rosenbaum (U.S. Duluth, USA), Andrew Cohen (University of Geological Survey, USA). Arizona, USA), John King (University of Rhode (iii) the westerlies, with their telecon- Island, USA), Michael Talbot (University of nection to the North Atlantic region. A continuous 120-m-long core col- Bergen, Norway). The lake basin occupies a closed lected from Bear Lake, Utah and Lake Malawi (10-14°S, 35°E), at the tectonic depression, or piggy-back Idaho (41°N 111°W) in September southern end of the East African Rift basin, on the upper plate of a major, 2000 contains evidence of hydro- Valley, provides perhaps the best active thrust fault. Seismic-reflection logical and environmental change record of Southern Hemisphere trop- data show that the lake sediments are over the last two glacial-interglacial ical climate dynamics available in tectonically deformed in some parts cycles (ca. 250,000 years). Chrono- Africa. The lake is 580 km long, 35 km 14 of the basin and largely undeformed logical control is provided by C wide and 700 m deep, with wind and in other parts, where they are more dates, paleomagnetic excursion, U- sea conditions that required a more than 700 m thick. Drilling operations series age, and correlation of carbon- substantial drilling platform than the in 2005 were conducted with the ate abundance and mineralogy with GLAD 800, as well as dynamic posi- 18 GLAD800 coring system, including δ O in Devils Hole calcite. Analyses tioning capability (Fig. 2). Two sites its enlarged barge. In total, 324 core completed at multi-centennial scale were drilled in February and March runs for 548 m of drilling acquired include rock-magnetic properties, O 2005: Site 1 in the central basin (water 323 m of core at an average recov- and C isotopes on bulk-sediment car- depth = 592 m) recovered sediment ery rate of 59%. The upper few tens bonate and ostracodes, organic- and to 385 m below lake floor (blf), span- of meters of sediment were mainly inorganic-C contents, mineralogy ning the past 1.5 million years (pre- gray clay and silty clay, in which (XRD), Sr isotopes on bulk carbonate, liminary chronology based on dated core recovery was excellent. The and pollen and diatom assemblages. magnetic reversals and excursions sediments below the clay-rich upper Most of the core comprises massive supplemented with radiocarbon, section were mainly fine-grained to banded silty clay and marl, with OSL and Be-10 dates), and; Site 2 sand, in which penetration and variable carbonate content and O- (361 m) in the northern basin of recovery were poor. The principal isotopic values. The abundance of the lake was triple cored to a major investigators thus decided to post- siliciclastic sediment indicates that unconformity at 38 m blf, recovering pone the planned 700-m drilling Bear River was usually connected to a record spanning the past 75,000 for a future campaign and to focus Bear Lake. Calcite is the dominant years. Two major sediment litholo- on obtaining high-quality, overlap- carbonate mineral but three arago- gies dominate the sequences: diato- ping cores of the upper 30–50 m of nitic marl intervals were deposited maceous varved silty mud similar to relatively fine-grained sediment at during the Holocene and the previ- what is accumulating in anoxic deep several sites, in order to obtain a ous two interglacial intervals, equiv- basins of the lake today, and biotur- high-resolution record of most of alent to oxygen isotope stages (OIS) bated, ostracode-diatom calcareous the last glacial cycle. 5e and 7e. These intervals have high silty clay, reflecting more arid condi-

PAGES NEWS, VOL.14, N°2, AUGUST 2006 Science Highlights: U.S. ESH Program 21

Oak growth in Midwestern North America linked with post glacial climate epochs in the North Atlantic

RICHARD P. GUYETTE1, MICHAEL C. STAMBAUGH1, ANTHONY LUPO1, ROSE-MARIE MUZIKA1 AND DANIEL C. DEY2 1School of Natural Resources, University of Missouri, Columbia, USA; [email protected] 2North Central Research Station, U.S. Forest Service, Columbia, USA; [email protected]

Early Holocene tree-ring climate proxy development from sub-fossil oak wood (buried wood not miner- alized) has been limited to Europe. Sub-fossil oak that has been collected from the rivers of Missouri and Iowa is abundant, well preserved, and has been determined to be at least 13.82 cal ka BP in age. This ancient oak wood represents riparian forest trees (Quercus bicolor Mich. and Quercus macrocarpa Mich.) that have been recruited into sediments by fl ooding and the erosion of stream banks. Large and heavy oak trees become buried rapidly by stream sediments and are then stored in the broad fl oodplains of Figure 1: Oak growth (BAI - basal area increment) in mid-continental North America (93°20’ these low-gradient streams for W, 40°10’ N) during the glacial-Holocene transition. Mean annual BAI of individual oak trees shows signifi cantly (probability p < 0.01) reduced growth and variance during the Younger thousands of years. Although re- Dryas (blue band). Dashed line: mean BAI before, during and after the Younger Dryas. Red cruitment, reburial, excavation, and arrows: hypothesized range in growth between tree populations limited by endogenous factors movement of wood is common, (competition) and exogenous factors (climate). the relatively short length of these and isotopic evidence that growth able endogenous factors (forest streams ensures that sampled oak and climate change were synchro- competition) to regional exogenous is rarely, if ever, more than 50 km nous between the North Atlantic factors (climate). Five-point moving from where it originally grew, thus, and mid-continental North Ameri- averages of the mean ring width of spatially defi ning the climate region can sites that lie 12,000 km to the the oak trees and temporally paired of origin. About 7 ancient oak trees southwest. δ18O values (Fig. 2) from Greenland (>500 yrs BP) are found per km of Oaks that were dated between ice (GISP2) were correlated (r = stream reach that have a suffi cient 9.5 and 14 cal ka BP using 14C and 0.57) during the post glacial climate number of rings (90 to over 280) for density dating (Guyette and Stam- period. Whole tree mean ring width use in the construction of tree-ring baugh, 2003) showed decreased was signifi cantly (p <0.01) reduced chronologies. The 14C and tree-ring growth (Fig. 1) and depleted δ13C during the Younger Dryas compared dating of the 230 oak trees collected (Fig. 2) coincident with North At- to adjacent post glacial climates. In to date indicate that the average lantic climate proxies. The mean addition to growth evidence, the tree grew about 2000 years ago. annual basal area increment (BAI; low abundance of oaks in the early One of several ongoing projects an area-based metric of tree stems) part of the Younger Dryas suggests utilizing ancient oak wood is the of oak trees was significantly (p a climate that was limiting to oak documentation of growth and cli- <0.01) less (6.5 cm2/yr) during the reproduction. The longest temporal mate during the last post glacial cli- Younger Dryas (Fig. 1) compared to gap between sample dates in the mate epoch, in what is now part of growth during the rest of the post 14,000-year-record occurs during the mid-continental North American glacial climate period (11.2 cm2/yr). the fi rst part of the Younger Dryas agricultural region. One of our goals Mean maximum annual BAI (17.0 ca. 12.48 to 12.99 cal ka BP, and cois to link well-documented climate cm2/yr) was reduced by 54% during incides with gaps in the European change in maritime environments the Younger Dryas. Between-tree record (Friedrich et al., 2004). This with the continental climate of what growth variance in BAI was about suggests unfavorable conditions are now important agricultural re- 75% less during the Younger Dryas may have existed for regeneration gions. For example, the Younger than during the whole post glacial and growth of trees across both Dryas event has been documented period (9.5-11.5 and 13-14 cal ka continents during the fi rst half of in the North Atlantic and Europe as BP). A decrease in the variance of the Younger Dryas. a period of time marked by abrupt growth rates within a tree popula- Values of δ13C (Fig. 2) in oak climate change and cooling (Alley, tion would be expected if limits on wood are found to be positively and 2000). Here, we provide tree-ring growth switched from highly vari- significantly correlated (r = 0.74)

PAGES NEWS, VOL.14, N°2, AUGUST 2006 22 Science Highlights: U.S. ESH Program

in the North Atlantic (~4,500 km northeast of the study site). In addition, serial correlation among tree growth and isotope chronologies suggests that post glacial climate variations had marked effects on the growth rates of trees. The construction of tree-ring chronologies for the Younger Dryas will eventually provide an annual-resolution climate proxy for an under-represented continental region.

NOTE Radiocarbon dates are on ﬁ le at commercial laboratories (Beta Analytic Inc. Miami, FL) as well as the Missouri Tree Ring Laboratory (http://www.missouri.edu/~guyetter/) where the growth data is also stored. Future 14C dated ‘ﬂ oating’ ring-width chronologies will be posted at International Tree-Ring Data Bank (http://www.ncdc.noaa.gov/paleo/treer- ing.html). Figure 2: Stable isotope data for oak (δ13C) in mid-continental North America (93°20’ W, 40°10’ N) paired with δ18O from Greenland ice (GRIP and GISP2 data, Grootes et al., 1993). Paired values REFERENCES of δ13C and δ18O infer a climate link during the post glacial-Holocene transition between maritime Alley, R.B., 2000: The Younger Dryas cold interval as climate of the North Atlantic and continental climate of Midwestern North America. viewed from central Greenland. Quaternary Sci- ence Reviews, 19: 213-226. with temporally paired 18O values Atlantic. Additionally, an increase Guyette, R.P., and Stambaugh, M.C., 2003: The age δ and density of ancient and modern oak wood in 13 (GRIP, GISP2) during the post gla- in oak δ C ca. 12.0 cal ka BP is co- streams and sediments. IAWA, 24: 345-353. cial period. This association occurs incident with intra Younger Dryas Loader, N.J., Roberston, I., and McCarroll, D., 2003: despite the complexities of possible variability, as documented by δ18O Comparison of stable carbon isotope ratios in 13 the whole wood, cellulose, and lignin of oak C signals in tree rings (Loader et from Greenland ice cores (Grootes tree-rings. Palaeogeography, Palaeoclimatology, al., 2003). The termination of the et al., 1993). Increasing δ13C in oak Palaeoecology, 196: 395-407. Younger Dryas (about 11.5 cal ka BP) during the post Younger Dryas tran- Grootes, P.M., Stuiver, M., White, J.W.C., Johnsen, S., and Jouzel, J., 1993: Comparison of oxygen iso- in mid-continental North America sition to the Holocene is found to be tope records from the GISP2 and GRIP Greenland oak is marked by an expected and consistent (r = 0.64) with changes in ice cores. Nature, 366: 552-554. abrupt enrichment in the climate- Greenland δ18O (GISP2). Friedrich, M., Lucke, A., Schwalb, A., and Hansich, sensitive stable isotope 13C. This Our preliminary results indicate S., 2004: Late glacial environmental and climatic δ changes from synchronized terrestrial archives of abrupt termination is synchronous that despite the low resolution of Central Europe: The Network PROSIMUL. PAGES with dates of Younger Dryas termi- our data, climate change during the News, 12(2): 27-29. nation based on pollen, ice core glacial-Holocene transition in mid- and tree-ring studies from marine continental North America was ap- and maritime regions of the North proximately synchronous with that Consolidating high- and low-resolution information from different sources into a Northern Hemisphere climate reconstruction

SHAOPENG HUANG Department of Geological Sciences, University of Michigan, Ann Arbor, USA; [email protected]

Our understanding of climate his- sions in the more remote past. But It would be desirable to develop tory in the pre-industrial era relies the compensation for the loss of a technique to integrate comple- principally on climate proxies, each resolution is a better determination mentary high- and low-resolution of which has strengths and limita- of the mean surface temperature in climate information preserved in tions in representing past climate the interval of time for which the different data sources for a more variability. details cannot be resolved. Con- complete picture of the past climate A paleoclimate reconstruction versely, reconstructions from many change. Huang (2004) attempted from borehole temperature data traditional proxy approaches offer such an integrated reconstruction has its foundation in the theory higher temporal resolution of rela- of the Northern Hemisphere sur- of thermal diffusion and, as such, tive changes but with greater uncer- face temperature history over the is characterized by a progressive tainties in the long-term trends. past ﬁ ve centuries through an in- inability to resolve climatic excur- tegrated analysis of the global da-

PAGES NEWS, VOL.14, N°2, AUGUST 2006 Science Highlights: U.S. ESH Program 23

of which 697 were located in the used to calculate a representative Northern Hemisphere. Northern Hemisphere subsurface A standardized borehole data signal (Fig. 1b), with the tempera- inversion technique independent ture at year 1500 taken as the initial of any other proxy or meteorologi- steady state. cal records is employed for the re- The last step of the integrated construction of century-long trends analysis is to bring into a coherent (Huang et al., 2000) from each of reconstruction process the subsur- the 697 Northern Hemisphere bore- face temperature signal and the an- hole data sets. The individual bore- nually resolved multi-proxy model hole-based reconstructions are then of Mann et al. (1999). Such an in- aggregated for a hemispheric rep- tegrated reconstruction is accom- resentation. plished through a functional space From a Northern Hemisphere Bayesian inversion, which allows perspective, the 697 boreholes sug- for explicit incorporation of an a pri- gest a cumulative change of about ori model, i.e., a hypothesis of the 1°C over the past five centuries, climate history under examination. with an accelerated warming trend For the integrated reconstruction of towards the present day (Fig. 1a). Northern Hemisphere surface tem- With approx. 50% more data, this perature history over the past fi ve new borehole-based reconstruction centuries, the temporal domain is is consistent with that derived ear- parameterized at an annual interval lier from a smaller data set of 453 to enable the incorporation of the boreholes employed by Huang et Mann et al. (1999) reconstruction as Figure 1: A) Extended surface temperature al. (2000). the a priori model. history combining the 1900-1980 surface air temperature (SAT) record (blue) and the 5°×5° The independent borehole- area weighted average of the borehole-based based estimate of 20th century Results reconstruction over the 16th-19th century warming is remarkably consistent The integrated reconstruction (Fig. period (red). The SAT trend and the borehole- based estimate of the 20th century trend are with the meteorological record 2b) shows that the late 16th century used for extension (dashed lines). B) Subsur- (Jones et al., 1999). Given that the to the early 17th century was the face temperature anomaly corresponding to 20th century meteorological record coldest, and the 20th century the the extended surface temperature history. is the most reliable climate record, warmest, over the past fi ve centu- tabase of borehole temperatures, Huang (2004) extends the SAT re- ries. Despite some fl uctuations at the annually resolved multi-proxy cord with the 5°×5° area weighted annual to decadal scales, the gen- reconstruction of Mann et al. (1999), 16th-19th century-long trends de- eral trend over the past fi ve centu- and the 20th century meteorologi- rived from the 697 boreholes. This ries is a progressive warming. The cal record. borehole-extended five-century integrated reconstruction suggests surface temperature history is then that the 20th century warming is a Methods and data The integrated reconstruction involves several steps: reconstructing century-long trends from hundreds of borehole temperature profi les, extending the surface air temperature (SAT) record with the borehole- based reconstruction, determining a subsurface temperature anomaly with the extended temperature record, and inverting the subsurface anomaly with the Mann et al. (1999) reconstruction as the a priori model. With support from the international heat-flow community, a global database of borehole tem- Figure 2: Comparison of the reconstructed surface temperature anomaly (red) versus radiative perature logs has been constructed forcing (blue) before (A) and after (B) the integrated reconstruction (Huang, 2004). The radiative as an archive of geothermal signals forcing series shown by the thin curve comprises the effects of solar irradiance, anthropogenic aerosols, greenhouse gases, and volcanism; the thick curve shows volcanism excluded. Both of climate change. At the time of radiative forcing series are based on Crowley (2000). Surface temperature anomaly is shown this study, the database contained with reference to the 1961-1980 mean. Radiative forcing is relative to the 1961-1980 mean of 862 borehole temperature profi les, the forcing series with volcanism excluded.

PAGES NEWS, VOL.14, N°2, AUGUST 2006 24 Science Highlights: U.S. ESH Program

borehole temperatures is hosted by the continuation of a long-term warm- the a priori model. For the 20th cen- University of Michigan (www.geo.lsa. ing. However, the warming appears tury, where the a priori multi-proxy umich.edu/climate) and accessible at World to have been accelerated towards model is well trained by the mete- Data Center-A for Paleoclimatology Data the present day. orological record, little alternation (www.ncdc.noaa.gov/paleo/borehole).

Integration of the complementa- is made through the inversion. At NOTE ry information greatly improves the very long periods, the subsurface The time series of this integrated recon- comparability of the reconstructed temperature data provide informa- struction can be downloaded from www. temperatures and the Crowley tion that is weak or absent in the ncdc.noaa.gov/paleo/pubs/huang2004. (2000) radiative forcing history (Fig. multi-proxy reconstruction. At in- REFERENCES 2). Analysis of the reconstructed termediate periods both the a priori Crowley T.J., 2000: Causes of climate change over temperature and radiative forcing model and subsurface data provide the past 1000 years. Science, 289: 270-277. series offers an independent esti- important constraints. Huang, S., Pollack, H. N., and Shen, P.Y., 2000: Temperature trends over the past five centuries mate of the transient climate-forc- The improvement in the com- reconstructed from borehole temperatures. Nature, ing response rate of 0.4 - 0.7 K per parability between climate recon- 403: 756-758. Wm-2 and predicts a temperature struction and the radiative forcing Huang, S., 2004: Merging Information from Different Resources for New Insights into Climate Change increase of 1.0-1.7 K in 50 years series is a validation of the climate in the Past and Future. Geophys. Res, Lett., 31: (Huang, 2004). information integration strategy. L13205, doi:10.1029/2004GL019781. The good agreement between the Jones, P.D., New, M., Parker, D.E., Martin, S., Rigor, Discussion integrated reconstruction and the I.G., 1999: Surface air temperature and its changes over the past 150 years. Rev. Geophys., 37:173- It is worth pointing out that the forcing model confirms that there 199. integrated reconstruction is not a are both natural and anthropogenic Mann, M.E., Bradley, R.S., Hughes, M.K., 1999: simple superposition of the high factors influencing the recent warm- Northern hemisphere temperatures during the past millennium: inferences, uncertainties, and limita- frequencies of the multi-proxy re- ing. tions. Geophys. Res. Lett., 26:759-762. construction on the lower frequencies of the borehole reconstruction. ACKNOWLEDGEMENTS For full references please consult: The integrated reconstruction con- Support for this study comes from NSF www.pages-igbp.org/products/newsletters/ref2006_2.html Grant ATM-0317572. The author thanks solidates information given in the Henry Pollack and Po-Yu Shen for stimulat- subsurface temperature data and in ing discussions. The global database of

Can models of abrupt climate change be tested from sea level reconstructions?

CHARLES JACKSON1, YURUN LIU2 AND OLIVIER MARCHAL3 1Institute for Geophysics, University of Texas at Austin, USA; [email protected] 2Department of Physics, University of Texas at Austin, USA 3Woods Hole Oceanographic Institution, USA

Introduction (inferences). Most notably, it can Freshwater forcing hypothesis As hypotheses concerning the ori- re-express the problem of abrupt The freshwater forcing hypothesis gin of abrupt climate change accu- climate change in terms of its im- to explain abrupt climate change mulate, there is an increasing need plications for what we should see involves changes in the freshwater to synthesize data and to develop in observations if models provide supply to the North Atlantic Ocean methods that may help assess adequate representation of the pro- through instabilities of the large their strengths and inconsisten- cesses that exist in nature. Here, continental ice sheets that covered cies. Understanding the mecha- we demonstrate the potential of the high northern latitudes during nisms of abrupt climate change is this approach (i) by estimating the the last glacial period. A key part particularly challenging because of changes in global sea level implied of this hypothesis is the capability the uncertainties in paleoclimate by the combination of a Greenland of the ocean meridional circulation observations and in our knowl- ice core paleotemperature record and of the attendant poleward heat edge of the physics of the climate during the last glacial period, with ﬂux to suddenly switch state and system. Among the techniques that a highly simpliﬁed model of the hold the climate system in a new can be used to piece together the buoyancy-driven ocean circula- state for hundreds to thousands of common elements of noisy ob- tion, and (ii) by comparing these years. This possibility is suggested servations given knowledge of a changes to reconstructions of by a hierarchy of ocean models system’s physics are so-called “in- global sea level during this period. that exhibit multiple equilibria of verse methods”. In particular, sto- The comparison provides a test of the meridional circulation and that chastic inverse modeling provides the consistency between the vari- demonstrate the potential of sur- a way to navigate the uncertainties ous datasets and a test of the hy- face freshwater forcing to initiate in data and models, and deal with pothesis of a freshwater forcing of transitions between the equilibria. the possibility of multiple solutions abrupt climate change. Thus, a hysteresis exists within

PAGES NEWS, VOL.14, N°2, AUGUST 2006 Science Highlights: U.S. ESH Program 25

(a) (b) 25 0.4 Northern Box Sinking Strength Fresh Water Forcing Times Series 20 vs ] ] −1 s

-1 Northern Fresh Water Forcing

15 3

m 0.2 6

6 3 10 0 m s

[1 5

0 0

−5

−10 −0.2 −15 Constant mixing energy Stability dependent diffusivity Fresh Water Forcing [ 10

Circulation Strength −20 Constant mixing diffusivity

−25 −0.4 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 1.2 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 Fresh water forcing to North Atlantic (~39000 ka) Time [years] (~30000 ka) [10 6 m 3 sec-1 ] (c) (d) 4 10 Greenland Temperature Anomaly Relative Sea Level Anomaly 3 5 ]

2 0

1 −5

0 −10

−1 −15

−2 −20

−3 −25 Sea level anomaly [meters] −4 −30 coral data (Thompson and Goldstein, in press) Observations

Temperature Anomaly [degrees deep sea oxygen isotopes (Waelbroeck et al., 2002) −5 −35 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 (~39000 ka) Time [years] (~30000 ka) (~39000 ka) Time [years] (~30000 ka)

Figure 1: Inverse modeling of abrupt climate change: A) equilibrium sensitivity of three models to freshwater forcing of the North Atlantic, B) freshwater forcing history required to match Greenland data, C) model reconstructions of Greenland data, and D) effect of freshwater forcing on sea level as compared to observations with estimates of uncertainty. these models, such that multiple following a claim that this mixing “Bayesian Stochastic Inversion” circulation strengths occur for the may actually force the meridional (Jackson et al., 2004) to estimate same rate of freshwater supply to circulation and its attendant heat the time evolution of freshwater the high latitudes. The freshwater flux (Munk and Wunsch, 1998). forcing (Fig. 1b) that permits each hypothesis is apparently support- According to a scaling theory of model to reproduce the Greenland ed by the observation of ice rafted the oceanic thermocline based ice core record between 39-30 ka debris in North Atlantic sediments on various approximations in the BP (Fig. 1c). The method works by during the cold phases of abrupt equations of motion, different as- randomly guessing the freshwa- climate change. Records of global sumptions about vertical mixing ter forcing for 81 evenly spaced sea level, which provide an inte- leads to a qualitatively different points along the 9 kyr interval and grated measure of the freshwater response of the meridional circu- using information about the effects forcing history, would provide an- lation to the surface freshwater of this guess on model-data mis- other test for this hypothesis. forcing (Fig. 1a). The assumptions match to guide subsequent forcing considered are representative of histories. This process is repeated Uncertain physics the range of what is physically multiple times and the statistics Among the major uncertainties justiﬁable, including mixing rep- of the results are collected. Thus, in numerical models of the ocean resented with constant vertical solutions may be represented by circulation is the representation diffusivity, stability-dependent dif- a multi-dimensional probability of sub-grid scale phenomena. The fusivity, and ﬁxed mixing energy. distribution indicating the relative most common assumption is that The solutions are generated from likelihood, the uncertainties, and the transport of properties such as a 4-box model of the global meridi- the dependencies of the different momentum, heat, and salt at small onal circulation, thereby extending quantities that are inferred from scales acts as a diffusion of these the single-hemisphere model study the original data. Illustrated in variables at large scales that is ef- of Nilsson and Walin (2001). Figure 1b-c are the optimal (most fectively resolved by the models. likely) solutions that provide the Most notably, the issue of the rep- Bayesian stochastic inversion minimum mismatch between the resentation of small-scale vertical For each of the three mixing rep- data and the model. mixing is gaining a lot of attention, resentations, we use a method of

PAGES NEWS, VOL.14, N°2, AUGUST 2006 26 Science Highlights: U.S. ESH Program

Inversion results and phases being directly related to the of temperature changes in the high implications amplitude of the sea level fall. northern latitudes is correct, and (ii) The global sea level variations The striking similarities among the highly simplified model we have implied by the freshwater forcing the three mixing formulations may used captures the relevant physics. histories for the three different be surprising considering the very This consistency does not depend mixing formulations are nearly different steady state response of on the formulation of vertical mix- identical; each showing a drop of each model to freshwater forcing ing in the model. Whether more ~15 m during the prolonged ~3 kyr (Fig. 1a). We find that the processes complete models are sensitive to warm phase starting around 38 ka governing the model’s transient re- the formulation of vertical mixing BP or model year 1000. The sub- sponse to freshwater forcing to be in their transient response to high sequent sequence of warm-cold initially very similar and only be- latitude freshening warrants further phases (between 35-30 ka BP) is gin to diverge after a few hundred investigation. associated with ~5 m variations in years. In all cases, the imposition of sea level. The timing and amplitude a high latitude freshening reduces REFERENCES of these sea level variations are not meridional density gradients and, Jackson, C., M. Sen, and Stoffa, P., 2004: An Efficient Stochastic Bayesian Approach to Optimal Param- inconsistent with two independent hence, the strength of the meridi- eter and Uncertainty Estimation for Climate Model reconstructions, given their uncer- onal circulation with changes in Predictions, Journal of Climate, 17(14), 2828-2841. tainties (Thompson and Goldstein, vertical mixing lagging. A more Munk, W., and Wunsch, C., 1998: Abyssal recipes II: Energetics of tidal and wind mixing, Deep Sea in press; Waelbroeck et al., 2002). thorough examination of the causes Research Part I: Oceanographic Research Papers, Although the relatively large uncer- of this similarity is being investi- 45(12), 1977-2010. tainties preclude a solid conclusion, gated. Nilsson, J., and Walin, G., 2001: Freshwater forcing as a booster of thermohaline circulation. Tellus, the comparison illustrates the level 53A, 629-641. of accuracy that is necessary for a Conclusion Thompson, W. G. and Goldstein, S.L., in press: A more stringent test of the freshwa- This stochastic inverse model ex- radiometric calibration of the SPECMAP timescale, ter hypothesis from the sea level ercise suggests that the freshwater Quaternary Science Reviews. Waelbroeck, C., Labeyrie, L., Michel, E., Duplessy, record. Furthermore, these results forcing hypothesis of abrupt climate J.C., McManus, J.F., Lambeck, K., Balbon, E., and highlight the potential importance change is consistent with the most Labracherie M., 2002: Sea-level and deep water of ice sheet growth in maintaining direct estimates of past sea level, temperature changes derived from benthic fora- minifera isotopic records, Quaternary Science a negative high latitude freshwater assuming that (i) our interpretation Reviews, 21(1-3), 295-305. balance for the phases of sustained of the Greenland record in terms warmth, with the length of the warm

Late Holocene hydrological variability in ombrotrophic peatlands of eastern North America

STEPHEN T. JACKSON1, ROBERT K. BOOTH2, YONGSONG HUANG3, ELISE G. PENDALL1, JONATHAN E. NICHOLS3, THOMAS A. MINCKLEY1 AND MEGHAN TAYLOR1 1Department of Botany, University of Wyoming, Laramie, USA; [email protected] 2Department of Earth and Environmental Sciences, Lehigh University, Bethlehem, USA 3Department of Geological Sciences, Brown University, Providence, USA

Introduction tation and temperature variations St. Lawrence corridor, from Min- Ombrotrophic peatlands, particu- across a range of temporal scales, nesota to Maine. We are coupling larly “raised bogs”, comprise a rich from seasonal to millennial. these paleoclimate reconstructions but underutilized source of Holo- Sedimentary records from with extensive modern calibration cene paleoenvironmental records ombrotrophic peats can span studies and investigations of his- for North America. These peatlands, 1000-10,000 years, with temporal torical climate variability. These which are scattered from the Atlan- resolution ranging from sub-cen- studies are leading to detailed mul- tic to the Paciﬁc coasts at latitudes tennial to sub-decadal depending tivariate climatic reconstructions between 42° and 65°N, are domi- on accumulation rates. A variety and development and testing of nated by Sphagnum moss and a of paleoenvironmental proxies, hypotheses regarding the under- few vascular plants, have elevated including testate-amoebae, peat lying climate dynamics. surfaces, and receive all surface humiﬁcation, pollen, plant macro- moisture directly from the atmo- fossils, charcoal, stable isotopes Peatland records of drought sphere. Water tables of these bogs (H, C, O), and biomolecular mark- synchrony in the Central are perched above the groundwa- ers, are preserved in these peats. United States ter table, and most water loss is We are conducting a study of all Peat records from sites 1000 km through evapotranspiration. Ac- of these proxies in late Holocene apart in north-central Minnesota cordingly, ombrotrophic bogs are peats from raised bogs along a (Hole Bog) and southeastern Mich- hydrologically sensitive to precipi- transect spanning the Great Lakes/ igan (Minden Bog) span the past

PAGES NEWS, VOL.14, N°2, AUGUST 2006 Science Highlights: U.S. ESH Program 27

�� Emerging proxies: Molecular biomarkers and compound-

�� specific stable isotopes We are currently developing and applying other proxies from ombrotrophic peatlands of mid-continental and eastern North America. One new approach is the use of molecular biomarkers. For example, Sphagnum differs from vascular plants and terrestrial mosses in the frequency distribution of n-alkane chain-lengths (Nichols et al., in press); because water retention is less critical for Sphagnum, cuti- cle waxes tend to be constructed of shorter (and presumably more eco- nomical) alkane chains. Ecological and paleoecological studies (based on plant macrofossils) indicate that Sphagnum moss tends to be espe- Figure 1. Paleohydrological records from ombrotrophic peatlands 1000 km apart in southeastern Michigan and north-central Minnesota show spatial coherence over the past 2000 years. High- cially productive during extended frequency patterns between 600-1000 years ago show good correspondence with tree-ring wet periods, while droughts favor records from the Great Plains and Rocky Mountains to the West. The blue trace represents expansion of ericads, sedges, and the Minnesota record and the red trace the Michigan record, both expressed in terms of depth to water table based on multivariate analysis of testate-amoeba assemblages (see Booth & other vascular plants. Stratigraphic Jackson, 2003 for details). Modified from Booth et al. (2006). records of n-alkane distributions from Minden Bog generally cor- several thousand years (Fig. 1). Period droughts recorded in tree- respond to patterns inferred from Water-table depths inferred from ring chronologies from the western testate amoebae and humification testate-amoeba assemblages re- United States (Cook et al., 2004). (Fig. 2). However, humification and veal substantial hydrological vari- The occurrence of these drought n-alkane spectra show an apparent ability at decadal to sub-millennial events as far east as eastern Mich- dry anomaly ca. 1300 yr BP that is scales, with much of this variabil- igan suggests that North Atlantic unrecorded by the testate amoe- ity coherent between the two sites. sea-surface temperatures played a bae. This underscores the need for Coherent fluctuations are particu- major role in modulating effects of multiple proxies that record differ- larly evident during time periods SST variability in the tropical Pa- ent kinds of hydrological respons- of high-magnitude moisture fluc- cific (e.g., McCabe et al., 2004). es to different aspects of climate tuations, suggesting that large The Minnesota and Michigan variation. moisture fluctuations are often bog records also show synchro- Another promising avenue is widespread—an observation con- nous drought events at 1850, 1800, the identification and extraction sistent with documented patterns and 1650 yr BP, during a low-fre- of organic compounds specific to of widespread drought during the quency transition from dry to wet Sphagnum. Preliminary results in- past century (McCabe et al., 2004; conditions. Few other high-resolu- dicate that 2-heptacosanone can Booth et al., 2006). tion records (e.g., tree-rings) are be extracted from at least seven The western Great Lakes re- available to corroborate these find- different species of Sphagnum gion is susceptible to multi-year ings but the synchroneity across but not from any vascular plants droughts centered in the Great 1000 km suggests these were re- or other mosses found in peat- Plains (e.g., the 1930s droughts gional-scale events. Hydrological lands. Such Sphagnum-specific penetrated eastward into the re- variability was dampened during biomarkers can be used not only gion). The high-frequency pa- the wet period between 1600-1100 as an index for quantifying Sphag- leohydrological patterns in our yr BP, and spatial coherence di- num productivity but also for com- bog records indicate that prehis- minished (Fig. 1). In fact, the sites pound-specific analysis of δD and toric multi-year to multi-decadal become modestly antiphased at δ13C, which can provide further in- droughts of the Great Plains were the height of the wet period. Simi- formation on moisture balance and felt at least as far east as Michigan. lar antiphasing occured between temperature. The Michigan and Minnesota peat- these regions during times of low Sphagnum macrofossils are lands both record severe, multi- variability in the last century’s in- well preserved in all but the most decadal droughts centered on strumental record (Booth et al., humified peats. Analysis of com- 700, 800, and 1000 yr BP (Fig. 1), 2006). pound-specific and taxon-specific corresponding to Medieval Warm δ18O (from cellulose in Sphagnum)

PAGES NEWS, VOL.14, N°2, AUGUST 2006 28 Science Highlights: U.S. ESH Program

can provide information on moisture balance. δ18O from modern Sphagnum plants is correlated with that of bog waters and with water-table depth. Ongoing analyses of δ18O from Sphagnum macrofossils will provide an additional proxy. Downcore analyses of compound-speciﬁc and taxon-speciﬁc δD, δ18O, and δ13C should allow us to tease apart moisture-balance, temperature, and humidity signals in the bog records (e.g., Pendall et al., 2001).

Pollen liberation: Assessing vegetational responses to climate change Pollen assemblages have tradition- Figure 2. Different proxies from the Minden Bog (MI) record show general correspondence, ally been the primary paleoclimato- with some potentially revealing contrasts. Water-table depth inferred from testate-amoeba logical tool for continental regions assemblages and peat humification measured colorimetrically represent two conventional at timespans longer than the past approaches to peatland sediments (see Booth and Jackson, 2003). Different expressions of n-alkane ratios from the sediments represent indices of relative biomass of Sphagnum and few centuries. Development of in- other plants. In all five panels, moisture increases along the y-axis. Negative exponential dependent paleoclimate records smoothing has been applied to each record to emphasize broad scale correlation. Modified from testate amoebae, stable iso- from Nichols, et al. (in press). topes, and biomarkers preserved in peatlands and other archives be assessed at timescales rang- REFERENCES Booth, R.K., and Jackson, S.T., 2003: A high-resolu- now allows us to use pollen data ing from multi-decadal to millen- tion record of Late Holocene moisture variability to assess vegetational responses nial. Spatiotemporal patterns of from a Michigan raised bog. The Holocene, to climate change and variability. variability within the network, and 13:865-878. Booth, R.K., Notaro, M., Jackson, S.T., and Kutzbach, The Minden record, for instance, between bog records and other J.E., 2006: Widespread drought episodes in demonstrates that a regional shift archives (tree-ring, lake-level, the western Great Lakes region during the past from beech to white pine domi- aeolian, glacial) should provide a 2000 years: geographic extent and potential mechanisms. Earth and Planetary Science Letters, nance 800-1000 years ago, and basis for assessing mechanisms 242:415-427. an accompanying increase in fire governing climate change and Cook, E.R., Woodhouse, C., Meko, D.M., and Stahle, incidence was driven by the on- variability. Bog-based records of D.W., 2004: Long-term aridity changes in the set of drier conditions (Booth and climate change can also be uti- western United States. Science, 306:1015-1018. McCabe, G.J., Palecki, M., and Betancourt, J.L., Jackson, 2003). This vegetational lized to identify climate drivers of 2004: Pacific and Atlantic Ocean influences on transition had been controversial, forest and fire dynamics, inferred multidecadal drought frequency in the United ascribed variously to decreas- from pollen, macrofossil and char- States. Proceedings of the National Academy of Sciences, 101:4136-4141. ing temperatures or activities of coal data. European scientists are Nichols, J.E., Booth, R.K., Jackson, S.T., Pendall, indigenous peoples. We are cur- currently developing a network of E.G., and Huang, Y., in press: Paleohydrologic rently developing high-resolution bog records across Eurasia. Thus, reconstruction based on n-alkane distributions in ombrotrophic peat. Organic Geochemistry. pollen records for selected time the potential exists for a circumbo- Pendall, E., Markgraf, V., White, J.W.C., Dreier, M., intervals to assess regional veg- real paleoclimate network for the and Kenny, R., 2001: Multiproxy record of late etational responses to transient entire Northern Hemisphere, based Pleistocene-Holocene climate and vegetation climate events, and to gauge the on multiple proxies from peatland changes from a peat bog in Patagonia. Quater- nary Research, 55: 168-178. limits of pollen sensitivity to cli- archives. Such a network could mate change. generate significant advances in understanding the patterns and Outlook mechanisms of Holocene climate Our ultimate goal is to develop a evolution. multiproxy, synoptic-scale network of paleohydrological and paleoeco- NOTE logical records from ombrotrophic Data used in this study will be available in the NOAA Paleoclimatology data base at: peatlands. Eventually, this network www.ncdc.noaa.gov/paleo/paleo.html could stretch from coast to coast, For detailled information, contact the and from ca. 45°-60°N in North authors. America. With such a network, Holocene climate variability can

PAGES NEWS, VOL.14, N°2, AUGUST 2006 Science Highlights: Open Section 29

San Valentin glacier ice core (Chilean Patagonia) – Filling the gap between Central Andes and Antarctica

PATRICK GINOT1, FRANÇOISE VIMEUX2, MARTINE DE ANGELIS1 AND OLIVIER MAGAND1 1Laboratoire de Glaciologie et Géophysique de l’Environnement, St Martin d’Hères, France; [email protected] 2IRD Great Ice, Laboratoire des Sciences du Climat et de l’Environnement, Gif-sur-Yvette, France: [email protected]

Introduction plumes). Further investigations are For the last 30 years, ice core inves- needed to deconvoluate the sulfur tigations along the Andes have been signal, and so, to locate the main mainly focused in the Central An- volcanic eruptions over this time des, between the equator and 30°S. period (e.g. Cerro Hudson 1991). Major paleoclimate reconstructions The water stable isotope content covering the last 25,000 years were (δD, δ18O) of the ice was measured obtained from Quelccaya, Huas- with a twice-lower sampling reso- caran, Sajama (Thompson, et al., lution, which does not allow us to 1998), Illimani (Ramirez, et al., 2003) clearly distinguish the seasonal and Cerro Tapado (Ginot, et al., variations (Fig. 2). However, the 2006). South of 60°S, from the Ant- dry/wet season alternation, as arctic Peninsula to the South Pole, revealed by the aerosol profiles, the international ice core commu- matches the narrow peaks in the nity is investigating numerous and isotopic profile. Moreover, the iso- widespread paleoclimate records, tope profile exhibits an interesting providing longer timescales. In or- interannual signal that should be der to understand the connection further discussed. Deuterium ex- between tropical areas and high cess has also been calculated (not latitudes, we have set up a new ice shown). The δD/δ18O slope is 8.64, core deep drilling program in the confirming that no melting at the Chilean/Argentinean Patagonia, surface occurs. Deuterium excess Figure 1: Top: South American map showing at 47°S, on San Valentin and San the location of San Valentin and San Lorenzo. exhibits a large range from about 4 Lorenzo summit glaciers (see proj- Bottom: San Valentin summit glacier (photo to 18‰, strongly suggesting large ect description). taken from plane) showing the location of the fluctuations in airmass origins and March 2005 shallow ice core drilling site. trajectories. This is in agreement San Valentin ice core drilling feature illustrates that the chemical with what has been inferred from project profile is well preserved from wa- chemical profiles. A first 16-m-long shallow ice core ter percolation alteration. The sea- The promising results from this was extracted in March 2005 from sonal variations of some chemical preliminary study have lead us to San Valentin glacier (Fig. 1) where concentrations (e.g. nss-Ca2+) allow plan deeper drilling operations in radar sounding indicates a gla- annual layer counting, in agree- this area in April/May 2007. An ice cier thickness of about 170 m (G. ment with the radio-nuclide dating. Casassa, pers. com.). The firn tem- Some specific events are recorded: perature measured at 16 m depth (i) Oceanic aerosol deposition, Project facts: is -12°C, confirming the excellent marked by sea salt peaks associ- The Sanvallor project deals paleoclimate signal preservation ated with biogenic sulfate and MSA with two deep ice cores at San at this high altitude site. The ice (Methanesulfonate), were found at Valentin glacier (Chile, 3747 m, core stratigraphy reveals no surface several depth levels (e.g. surface, 46°35’19”S, 73°19’39”W) and San melting/refreezing ice layer. No ash 3.8 m, 7.6 m); the Cl-/Na+ ratio pro- Lorenzo glacier (Chile/Argentina, or dust layer has been observed. file, generally close to the oceanic ~3500 m, ~47°32’S, 72°20’W). It The ice has been analyzed for ionic sea salt ratio, indicates rapid aero- started in 2005, with the prelimi- contents, water stable isotopes (δD sol transport; (ii) Biomass burning nary study on the shallow core, and δ18O) and some radio-nuclides. events characterized by high nitrate and should run until 2009. The 210Pb decay and 137Cs peaks indicate concentrations associated with nss- preliminary study was funded by a time range of about 1965 to 2005, Sulfate and/or nss-chloride increase IRD-Great Ice, LSCE, LGGE and which corresponds to a mean an- were observed. The strongest event CECS Valdivia (Centro de Estu- nual accumulation rate of about 33 is located at 11.1 m. These two dif- dios Cientificos). The first deep cm of snow per year (180 mm water ferent types of events may be re- drilling operations attempt was equivalent). The chemical concen- lated to the two main atmospheric funded by the Chilean Fondecyt trations measured along the pro- circulation regimes: westerlies (oce- and IRD. A proposal has been file are extremely low, comparable anic input) or atmospheric transport submitted to fund the analyses. to polar levels (Fig. 2). Their main from Argentinean Patagonia (fire

PAGES NEWS, VOL.14, N°2, AUGUST 2006 30 Science Highlights: Open Section

core will be recovered down to the bedrock by IRD Great Ice and its partners on the San Valentin site, and some preliminary investigations performed on the San Lorenzo glacier, about 50 km farther southeast. Ice thickness and estimated accumulation rate on San Valentin allow us to expect paleoclimate reconstructions at high resolution over, at least, the last 2000 years. The deepest part of the ice cores might include the Last Glacial Maxi- mum. Ice will be dispatched to all the participating laboratories. The ice core diameter will permit us to provide enough material to estab- lish all the proﬁ les with the required resolution.

REFERENCES Ginot, P., Kull, Ch., Schotterer, U., Schwikowski, M. and Gäggeler, H.W., 2006: Glacier masse balance reconstruction by sublimation induced enrichment of chemical species on Cerro Tapado (Chilean Andes), Climate of the Past, 2: 21-30. Ramirez, E., Hoffmann, G., Taupin, J.D., Francou, B., Ribstein, P., Caillon, N., Ferron, F.A., Landais, A., Petit, J.R., Pouyaud, B., Schotterer, U., Simões, J.C. and Stievenard, M., 2003: A new Andean deep ice core from Nevado Illimani (6350 m), Bolivia, Earth and Planetary Science Letters, 212: 337-350. Thompson, L. G. Davis, M.E., Mosley-Thompson, E., Sowers, T.A., Henderson, K.A., Zagorodnov, V.S., Lin, P-N, Mikhalenko, V.N., Campen, R.K., Bolzan, J.F., Cole-Dai, J. and Francou, B., 1998: A 25000- year tropical climate history from Bolivian ice cores, Science, 282: 1858-1864.

Figure 2: Selected chemical concentration (µEq.L-1), ratio and δD (‰ versus SMOW) proﬁ les measured along the San Valentin 16-m-long shallow ice core extracted in March 2005.

Sediment, pollen and isotope evidence for an Early to Mid-Holocene humid period in the desert of Yemen

ANNE-MARIE LÉZINE1, CHRISTIAN ROBERT2, JEAN-FRANÇOIS SALIÈGE3 AND JEAN-JACQUES TIERCELIN4 1Laboratoire des Sciences du Climat et de l’Environnement, UMR 1572 CNRS-CEA, Gif-sur-Yvette, France; [email protected] 2CEREGE, Aix-en-Provence, France; [email protected] 3Laboratoire d’Océanographie et du Climat, UMR 7159 CNRS-IRD-UPMC, Paris, France; [email protected] 4IUEM, UMR 6538 CNRS-UBO, Plouzané, France; [email protected]

Introduction vestigating atmosphere-vegetation inland and its impact on hydrology Environmental reconstructions feedback in the climate system. By and vegetation are only poorly un- of tropical deserts during the Ho- contrast, very little is known about derstood because of the scarcity of locene have mainly focused on the environmental and climate his- continuous continental sedimentary Northern Africa. There, numerous tory of Arabia, even though recent archives. Here, we present the ﬁ rst lake-level and pollen data have discoveries of speleothems in Oman continuous record of environment been extensively used for regional (Fleitmann et al., 2003) have yield- and climate in Southern Arabia, reconstruction of past hydrological ed high-resolution oxygen isotope based on sedimentological, miner- or vegetation changes, as well as records reﬂ ecting variations in the alogical, pollen and isotope stud- for paleoclimatic models connect- Indian monsoon rainfall during the ies of al-Hawa (Yemen) paleolake, ing environmental changes with Holocene. However, the northern which covers the time interval from variations in the Earth’s orbit or in- penetration of the Indian monsoon 12 to 7.5 ka BP.

PAGES NEWS, VOL.14, N°2, AUGUST 2006 Science Highlights: Open Section 31

rain falling over the core site. The increase in percentages of fern spores and Typha conﬁ rms the humid conditions and points to the extension of reed swamp populations at the lake shore during the same time interval. The uppermost levels of the al-Hawa sequence indicate the return to dry conditions at ca. 7.5 ka BP responsible for the subsequent drying out of the al-Hawa lake.

Regional vegetation Up to 91 pollen and fern spore taxa have been recorded in the sediment. They belong to three clearly distinct phytogeographical groupings: Tree taxa are mainly of Somalia-Masai (Acacia, Commiphora, Maerua, Sal- vadora persica), and Afro-Montane Figure 1: The lacustrine sequence of al-Hawa, Yemen: lithostratigraphy and chronology. (Olea, Juniperus, Myrica, Podocar- pus, Ephedra, Acacia, Dodonaea The lacustrine sequence of al- the Holocene (Fig. 1). The lacustrine viscosa, Rhus, Celtis) origin and be- Hawa sequence is mainly composed of long to sclerophilous woodlands and The sample site (15°52N, 46°53E, 710 massive or fi nely laminated siltstone, scrubs. All these taxa are found in m asl) is a large fl at basin located silty mudstone and mudstone with modern Yemen, except Podocarpus, in the inland desert of Yemen, the abundant biogenic components, for which the nearest pollen source Ramlat as-Sabatayn, which is the such as molluscs, ostracods, spong- is the East African highlands. Herba- southernmost extension of the Rub es and diatoms. It is interrupted by ceous pollen taxa are of Saharo-Sin- al-Khali hyperarid sand sea. The ar- quartz-rich fi ne sandstone layers, the dian origin, with modern distribution ea is surrounded by dunes and fed most important of which, from 1.75 in the deserts of Rub al’Khali and by wadis, which drain the adjacent to 0.75 m, being dated from ca. 8 to Ramlat as-Sabatayn. Pollen spectra Yemen highlands and Hadramawt 7.7 ka BP. are dominated by herbaceous taxa, plateau (Fig. 1). The modern climate Concentrations of both authigen- mainly Poaceae, Cyperaceae, Cichori- is of desert type with annual rainfall ic and biologic (molluscs and ostra- ae, Amaranthaceae/Chenopodiaceae, of Indian monsoon origin averag- cods) carbonates suggest phases of Typha and ferns. Tree pollen grains ing 28 mm, and temperature 27°C at major lake development peaking be- are scarce, never reaching percent- Marib (15°26N, 45°20E). Vegetation tween 11-10.5 ka BP (lacustrine phase ages higher than 13%. This points to is scarce, with trees (Acacia, Com- 1), 10.1-9.1 ka BP (phase 2), 8.4-8 ka the permanency of a semi-arid land- miphora) confi ned to wadi channels BP (phase 3), and 7.8-7.5 ka BP (phase scape in the al-Hawa basin during and highland foothills. Herbaceous 4) (Fig. 2). Pollen and isotope data the early to mid-Holocene. plants occur on sand, with Cyperus from freshwater shells show that the spp., Calligonum spp. and Dipterigy- paleolake of al-Hawa was formed in Hydrological changes and um glaucum mainly growing on sta- a very arid environment: high per- rainfall variability bilized dunes. centages of Amaranthaceae/Cheno- The al-Hawa record shows that, dur- The sediments were discovered podiaceae and high values of δ18O ing the Holocene, Southern Arabian in 1993. Following preliminary stud- (+4.18‰) indicate saline and highly hydrology was very sensitive to both ies carried out on selected outcrops evaporated conditions. Such high orbitally induced monsoon variations (Lézine et al., 1998), a pit was dug δ18O values are observed today in and other, superimposed, centenni- in 2000 into the lake bed down to dry areas of North Africa, at Lake al- to millennial-scale variations of a depth of 7.45 m below the top of Chad, Lake Abhé or Lake Awassa, the global climate system. It is con- the sequence. The al-Hawa lacus- while Amaranthaceae-Chenopodia- sistent with the nearby speleothem trine sequence lies on a wadi-type ceae pollen types dominate in salty record from Oman (Fleitmann et al., microconglomerate layer with abun- places in the desert. Conversely, the 2003), which shows that the summer dant well-rounded centimetric quartz intermediate levels record negative monsoon infl uence increased signifi - gravels in a sandy matrix. Such fa- δ18O values, averaging -4.6‰, close cantly after 10.3 ka BP and decreased cies confirm that a fluvial system to the characteristically low δ18O of after 7 ka BP in response to the north- connected wadi al-Jawf and wadi monsoonal rainfall, suggesting that south movements of the Intertropical Hadramawt before the al-Hawa de- the isotopic composition of precipita- Convergence Zone (ITCZ) over the pression became endoreic during tion was controlled by the amount of Arabian peninsula. Our data sug-

PAGES NEWS, VOL.14, N°2, AUGUST 2006 32 Science Highlights: Open Section

Figure 2: The “Arabian Humid Period” from marine and continental archives. Left panel: Globigerina bulloides percentages from ODP site 723 (Gupta et al., 2003) with highest percentages reﬂ ecting phases of strong upwelling in the Arabian sea. Numbers refer to Bond events (Bond et al., 1997). Right panel: the lacustrine record of al-Hawa (Yemen) with three “dry” events punctuating the humid period (arrows). Pollen data and the oxygen-isotope record on fresh-water shells shows the dramatic increase in summer monsoon rains from 9000 yrs B.P. onwards. Arid, evaporated, conditions increased toward the end of the “Arabian Humid Period”, and the al-Hawa lake completely dried out at 7500 yrs B.P.

gest that the ITCZ reached the core benefi ted from surfi cial runoff from REFERENCES site (16°N) from 9 ka BP onwards. In the adjacent highlands located to the Bond, G., Showers, W., Cheseby, M., Lotti, R., Almasi, P., deMenocal, P., Priore, P., Cullen, H., Hajdas, I., and addition, the abrupt climate varia- south (the Hadramawt plateaus) and Bonani, G., 1997: A Pervasive Millennial-Scale Cycle tions recorded from ice and ocean to the west (the Yemen highlands). in North Atlantic Holocene and Glacial Climates. cores during the Holocene (Bond et The fl at morphology of its wide basin Science, 278: 1257-1266. al., 1997) have also been recorded in made it very sensitive to low ampli- Fleitman, D., Burns, S.J., Mudelsee, M., Neff, U., Kramers, J., Mangini, A., and Matter, A., 2003: Southern Arabia, by a succession of tude variations in rainfall. The pres- Holocene forcing of the Indian monsoon recorded dry phases, between 10.5-10.1, 9.1- ence of the Hadramawt plateaus and in a stalagmite from Southern Oman. Science, 300: 8.4, and 8-7.8 ka BP, thus confi rming the Yemen highlands, which acted as 1737-1739. Gupta, A.K., Anderson, D.M., and Overpeck, J.T., 2003: previous observations on the Indian topographical and ecological barriers Abrupt changes in the Asian Southwest monsoon monsoon system from the adjacent against the northward penetration of during the Holocene and their links to the North ocean (Gupta et al., 2003). tropical plants (as observed in the Atlantic Ocean. Nature, 421: 354-357. Lézine, A.-M., Saliège, J.-F., Robert, C., Werthz, F., and Saharan desert at that time), played Inizan, M.-L., 1998: Holocene lakes from Ramlat-as Conclusion a determinant role in the vegetation Sab’atayn (Yemen) illustrate the impact of monsoon The al-Hawa sedimentary sequence composition, which remained of des- activity in Southern Arabia. Quaternary Research, 50: 290-299. provides an exceptional, and rare, ert, semi-desert type throughout the Lézine A.-M., Tiercelin J.-J., Robert C., Saliège J.-F., continuous record of environment Holocene humid period. Cleuziou S., Braemer F., and Inizan M.-L., in press: and climate variations in the South- Centennial to millennial-scale variability of the ern Arabian Desert. Paleolake records NOTE Indian monsoon during the early Holocene from a sediment, pollen and isotope record from the are surprisingly scarce in this region The presented data is available in the MEDIAS database at: desert of Yemen. Palaeogeography, Palaeoecology, Palaeoclimatology. (Lézine et al., in press). Al-Hawa, as medias.obs-mip.fr/eclipse with many of the other freshwater extensions in the Rub al’Khali desert,

A better climate for human evolution

MARTIN H. TRAUTH1, MARK A. MASLIN2, ALAN DEINO3 AND MANFRED R. STRECKER1 1Department of Geosciences, University of Potsdam, Germany; [email protected] 2Environmental Change Research Center, University College London, UK 3Berkeley Geochronology Center, USA Investigations of both terrestrial particular, hominin evolution (de- vironmental changes inferred from and marine paleoclimate archives Menocal, 1995; Trauth et al., 2003, marine records. For example, the have resulted in an ongoing debate 2005). Because of the unique tec- tectonic and magmatic evolution concerning global vs. local climate tonic and magmatic evolution of the EARS has resulted in highly forcing and associated environ- of the East African Rift System variable sedimentary environments mental changes in East Africa. A (EARS) and resulting changes in in close proximity to each other, better knowledge of shifts between orography and drainage patterns, whose depositional record may be dry and humid conditions is key terrestrial paleoclimate records fundamentally inﬂ uenced by local to our understanding of processes from these environments may not conditions rather than global sig- influencing mammalian and, in always automatically reﬂ ect the en- nals. Thus, it is important to strive

PAGES NEWS, VOL.14, N°2, AUGUST 2006 Science Highlights: Open Section 33

AFARAR Central Afar for a better understanding of rift- BASIN Bogoria basins in the north started related processes and indicators in the Middle and Upper Miocene, of regional climate change, with Ethiopian Rift whereas the age of the oldest lacus- the aim of characterizing hydro- ETHIOPIAN RIFT trine sequences in the central and WESTERN 6 BRANCH logical changes in East Africa and, Areas > 1000 m southern segments of the rift (i.e., OF THE EARS Omo above sea level

Normal Fault ultimately, of comparing them to Lake Turkana Kenya and Tanzania) is Early Plio- 4 EASTERN global processes. Turkana BRANCH cene (Tiercelin and Lezzar, 2002). OF THE EARS 2°N Suguta Valley However, if tectonics were the sole Lake Mt. Elgon KENYA RIFTRIFT Albert Baringo-Bogoria History of the East African rift Nakuru- E q u a t o r control over the appearance and Lake Elmenteita Mt. Kenya Edward Turasha lakes Lake Naivasha TANZANIANANZANIAN RIFTRIFT disappearance of lakes, then either Victoria Gicheru Lake Magadi- Indian Ocean The EARS is characterized by a Kivu Natron Olorgesailie a N-S or NW-SE temporal pattern Olduvai Mt. Kilimanjaro southward propagation of rifting, 30°E 32 34 38 40 42 44 would be expected. In contrast, including the formation of major Figure 1: Map of East Africa showing topog- what is observed is the appearance faults, subsequent densifi cation and raphy, faults and lake basins. of large, deep lakes synchronous tectonic differentiation of exten- across a large geographical area sional basins and coeval magmatic and Southern Kenya Rift between at 2.7-2.5 Ma, 1.9-1.7 Ma and 1.1-0.9 activity. The earliest formation of 9 and 6 Ma. These early halfgraben Ma (Trauth et al., 2005), suggest- lake basins in the northern parts of formed during approx. E-W oriented ing regional climatic control. For the rift is an integral part of these extension and were subsequently the reconstruction of these lakes, processes. In the Ethiopian Rift, vol- faulted antithetically between about we sampled several tens of meters canism started between 45 and 33 5.5 and 3.7 Ma, which generated a thick lacustrine sequences at meter Ma, in northern Kenya at around 33 full-graben morphology. Major nor- to half-meter intervals. The diato- Ma, and continued to about 25 Ma, mal faulting in the Magadi-Natron mite and pyroclastic material-bear- whereas the magmatic activity of and Olduvai basins occurred at 1.2 ing sediments were investigated the central and southern segments Ma and produced the present-day for sediment characteristics, authi- of the rift in Kenya and Tanzania rift escarpments (Tiercelin and Lez- genic minerals, and diatom fl ora. started between 15 and 8 Ma. Major zar, 2002). Where possible, these units were faulting in Ethiopia between 20 to The southward propagation of radiometrically dated using anor- 14 Ma was followed by the genera- extensional processes in the EARS thoclase and sanidine phenocryst tion of major east-dipping faults in is also refl ected in the earliest for- concentrates from several tuff beds northern Kenya between 12 and 7 mation of lake basins in East Africa. and lava fl ows. Ma, superseded by normal faulting The fl uvio-lacustrine history of the on the western side of the Central Afar, Omo-Turkana and Baringo-

��

0.00 �� 0.02 0.04 0.06

0.0 H. flo Ma 14C, U/Th 14C 14C 14C At 160 to 154 kyr �� Oldest Members of U/Th U/Th extrapolated r Homo sapiens esiensis Homo sapiens �� 0.235 �� 0.284 �� H. neanderthalensis 0.509 0.5 0.493 0.552 �� P Paranth

0.545 H. heidelbe 0.601 �� .

0.610 r

0.662 obustus 0.72 0.746 �� 0.780 At ~1.0 Ma r 0.8-1.0 opus. boisei 0.946 H. antecessor Second major

0.974 r migration of Homo 1.0 0.992 0.977 gensis 0.987 erectus out of Africa, �� extinction of the �� genus Paranthropus ��

�� 1.38 �� 1.38 1.44 �� 1.5 �� 1.749 �� At ~1.8 Ma 1.761 1.6546 1.66 �� 1.764 Earliest Homo 1.798 1.71 �� erectus, Out Australopithecus garhi 1.806 1.85 of Africa, P

�� . aethiopicus 1.859 1.96 1.88 1.87 Acheulean H. e

1.865 H. e 1.976 Technology

2.0 2.0 A. africanus

2.018 r r ectus 2.029 2.0-2.1 gaster Praeanth �� A. bah

2.3 Homo habilis 2.35 r elghazali r

Age data �� opus afa 2.51 At ~2.6 Ma

2.5 Olduvai Technology, H. rudolfensis Lava, tuff 2.66-2.56 2.58 Earliest Homo, Deep freshwater lake 2.68 2.71 Oldest "robust" r

Australopithecines ensis Shallow alkaline lake �� No lake 3.0 3.01

Figure 2: Insolation variation, wet periods in East Africa and human evolution. The compilation of lake records is based on a large volume of literature (Olduvai, Magadi-Natron-Olgesailie, Baringo-Bogoria, Omo-Turkana-Suguta, Ethiopian Rift and Afar), as well as from our own work (Central Kenya Rift) (see Trauth et al. (2005) and references therein for details).

PAGES NEWS, VOL.14, N°2, AUGUST 2006 34 Science Highlights: Open Section

Paleoclimatic implications climatic changes, as well as peri- alyst for human evolution. In this The alternating wet and dry phas- ods of maximum eccentricity. The respect, the close link between la- es during the last 3 Ma correlate period between 2.7 and 2.5 Ma cor- custrine environments and sites of with numerous other humidity responds to the onset of Northern early hominin occupation is intrigu- paleoclimate indicators in East Af- Hemisphere Glaciation, the period ing. Based on these observations, rica, such as mammal assemblages, between 1.9 and 1.7 Ma corresponds we identify two important issues in fossil forest leaves, wood and pol- to a significant shift in the Walker mammalian and hominin evolution len, as well as with Mediterranean Circulation, and the interval be- in East Africa. First, sustained peri- sapropels and dust. The most fas- tween 1.0 and 0.8 Ma corresponds to ods of amenable climatic and en- cinating phenomenon in this con- the initiation of the Mid-Pleistocene vironmental conditions lasting for text is the anti-correlation between Revolution. The phases of high lake- protracted periods on the order of the East African lake levels and levels in the Ethiopian and Kenyan 104–105 years, promoting synpatric paleoclimate records from sub- rifts also correlate with periods of evolution. Second, lakes may have tropical West and Northeast Africa maximum eccentricity, suggesting rapidly appeared and disappeared (deMenocal, 1995). In this context, it that changes in local solar radiation with precessional cyclicities, pro- has been proposed that eolian dust are equally important in the mois- ducing a rapidly changing mosaic in ocean sediment cores off west- ture history on longer time scales. of environments, which ultimately ern and eastern subtropical Africa Since the main effect of eccentric- helped promote allopatric evolu- record environmental changes in ity is to modulate the amplitude of tion. Africa that were also thought to rep- Earth’s precession cycles, at times resent drier conditions in regions of maximum eccentricity, the sea- REFERENCES that we studied. However, records sonality in insolation and, hence, Clement, A.C., Hall., A. and Broccoli, A.J., 2004: The importance of precessional signals in the tropical from sediment cores off West Africa the intensity of the rainy seasons climate. Climate Dynamics, 22: 327-341. and Oman may not automatically in the tropics increases (Clement et deMenocal, P., 1995: Plio-Pleistocene African reflect the nature of East African al., 2004). The combination of sig- Climate. Science, 270: 53-59. Tiercelin, J.J., and Lezzar, K.E., 2002: A 300 million climate change, since the source nificant global climate changes with years history of rift lakes in Central and East areas of these terrigenous depos- maxima in precessional amplitudes Africa: an updated broad review. In: Odada, E.O., its are in the Sahara/Sahel regions may explain why these humid peri- Olago, D.O. (Eds.), The East African great lakes: limnology, paleolimnology and biodiversity, off West Africa, Arabian peninsula ods recur only after 0.4 Ma, whereas Kluwer, Dordrecht, pp. 3-60. (more than 80%) and northeast maximum eccentricity and, hence, Trauth, M.H., Deino, A., Bergner, A.G.N., and Strecker, Africa (less than 20%), rather than maximum amplitudes in precession M.R., 2003: East African climate change and equatorial East Africa where the occur at higher periodicities. orbital forcing during the last 175 kyr BP. Earth and Planetary Science Letters, 206: 297-313. hominins evolved. These periods of extreme cli- Trauth, M.H., Maslin, M.A., Deino, A., and Strecker, Importantly, the periods of in- mate variability coincide with key M.R., 2005: Late Cenozoic Moisture History of East creased humidity, as reconstructed periods of speciation and dispersal Africa. Science, 309: 2051-2053. from East African lake levels, cor- of mammals and hominins in East relate both with significant global Africa and may have been the cat-

Surface temperature reconstructions for the last 2000 years

GERALD R. NORTH1 AND IAN KRAUCUNAS2 on behalf of the Committee on Surface Temperature Reconstructions for the Last 2,000 Years, U.S. National Research Council* 1 Department of Atmospheric Sciences, Texas A&M University, College Station, USA; [email protected] 2Board on Atmospheric Sciences and Climate, National Research Council, Washington, USA; [email protected]

In response to a request from the Third Assessment Report (2001), gies and data that were used. The U.S. Congress, the U.S. National attracted considerable attention NRC committee’s report (2006) at- Research Council (NRC) assem- because the authors concluded tempts to diffuse this controversy bled a committee to describe and that the Northern Hemisphere was by providing a comprehensive as- assess scientiﬁc efforts to recon- likely warmer during the late 20th sessment of surface temperature struct surface temperature records century than at any other time dur- reconstructions for the last 2000 for the Earth over the last 2000 ing the past millennium. Contro- years, as well as the implications years. This research, and especially versy arose because many people of these efforts for our understand- the reconstructions originally pub- interpreted this result as deﬁnitive ing of global climate change. lished by Mann et al. (1998, 1999) evidence of anthropogenic causes and featured in the Intergovern- of recent climate change, while mental Panel on Climate Change others criticized the methodolo-

*Gerald R. North (Chair), Franco Biondi, Peter Bloomﬁeld, John R. Christy, Kurt M. Cuffey, Robert E. Dickinson, Ellen R.M. Druffel, Douglas Nychka, Bette Otto-Bliesner, Neil Roberts, Karl K. Turekian, John M. Wallace, Ian Kraucunas (Study Director)

PAGES NEWS, VOL.14, N°2, AUGUST 2006 Science Highlights: Open Section 35

Figure 1: Smoothed reconstructions of large-scale (Northern Hemisphere mean or global mean) surface temperature variations from six different research teams, each using its own methodology and selection of proxies, are shown along with the instrumental record of global mean surface temperature. Each curve portrays a somewhat different history of temperature variations, and is subject to a somewhat different set of uncertainties that generally increase going backward in time (as indicated by the gray shading). This set of reconstructions conveys a qualitatively consistent picture of temperature changes over the last 1100 years, and especially the last 400. Findings including ice cores, tree rings, data increase substantially After considering all of the avail- marine sediments, and his- backward in time through this able evidence, including the curves torical sources from Europe period and are not yet fully shown in Figure 1, the committee and Asia, but the exact timing quantifi ed. reached the following conclusions: and duration of warm periods (v) Very little confi dence can be as- (i)The instrumentally measured may have varied from region signed to statements concern- warming of about 0.6°C dur- to region, and the magnitude ing the hemispheric mean or ing the 20th century is also and geographic extent of the global mean surface tempera- refl ected in borehole tempera- warmth are uncertain. ture prior to about A.D. 900 be- ture measurements, the retreat (iii) It can be said with a high level cause of sparse data coverage, of glaciers, and other obser- of confi dence that global mean and because the uncertainties vational evidence, and can be surface temperature was high- associated with proxy data and simulated with climate models. er during the last few decades the methods used to analyze (ii)Large-scale surface tempera- of the 20th century than during and combine them are larger ture reconstructions yield a any comparable period during than during more recent time generally consistent picture of the preceding four centuries. periods. temperature trends during the This statement is justifi ed by preceding millennium, includ- the consistency of the evidence Factors limiting confi dence ing relatively warm conditions from a wide variety of geo- The main reason that our confi- centered around A.D. 1000 graphically diverse proxies. dence in large-scale surface tem- (identifi ed by some as the “Me- (iv) Less confi dence can be perature reconstructions is lower dieval Warm Period”) and a placed in large-scale surface before A.D. 1600, and especially be- relatively cold period (or “Little temperature reconstructions fore A.D. 900, is the relative scarcity Ice Age”) centered around for the period from A.D. 900 of precisely dated proxy evidence. 1700. The existence and extent to 1600. Presently available Other factors limiting our confi- of a Little Ice Age from roughly proxy evidence indicates that dence in surface temperature re- 1500 to 1850 is supported by temperatures at many, but constructions include the relatively a wide variety of evidence, not all, individual locations short length of the instrumental re- including ice cores, tree rings, were higher during the past 25 cord (which is used to calibrate and borehole temperatures, glacier years than during any period validate the reconstructions); the length records, and historical of comparable length since fact that all proxies are infl uenced documents. Evidence for re- A.D. 900. The uncertainties as- by a variety of climate variables; gional warmth during medieval sociated with reconstructing the possibility that the relationship times can be found in a diverse hemispheric mean or global between proxy data and local sur- but more limited set of records, mean temperatures from these face temperatures may have varied

PAGES NEWS, VOL.14, N°2, AUGUST 2006 36 Science Highlights: Open Section

over time; the lack of agreement as Analysis of previous to further improve our knowledge to which methods are most appro- conclusions of temperature variations over the priate for calibrating and validating The basic conclusion of Mann et al. last 2000 years, particularly if ad- large-scale reconstructions and for (1998, 1999) was that the late 20th ditional proxy evidence can be selecting the proxy data to include; century warmth in the Northern identified and obtained from areas and the difficulties associated with Hemisphere was unprecedented where the coverage is relatively constructing a global or hemispher- during at least the last 1000 years. sparse, and for time periods be- ic mean temperature estimate using This conclusion has subsequently fore A.D. 1600 and especially before data from a limited number of sites been supported by an array of evi- A.D. 900. Furthermore, it would be and with varying chronological pre- dence that includes both additional helpful to update proxy records cision. All of these considerations large-scale surface temperature that were collected decades ago, introduce uncertainties that are dif- reconstructions and pronounced in order to develop more reliable ficult to quantify. changes in a variety of local proxy calibrations with the instrumental indicators, such as melting on ice- record. Improving access to data Role of reconstructions in caps and the retreat of glaciers used in publications would also climate science around the world, which in many increase confidence in the results Overall, the committee found that cases appear to be unprecedented of large-scale surface temperature efforts to reconstruct temperature during at least the last 2000 years. reconstructions both inside and histories for broad geographic re- Not all individual proxy records outside the scientific community. gions using multiproxy methods indicate that the recent warmth is New analytical methods, or more were an important contribution to unprecedented, although a larger careful use of existing ones, may climate research and that these fraction of geographically diverse also help circumvent some of the large-scale surface temperature sites experienced exceptional existing limitations associated with reconstructions contain meaning- warmth during the late 20th century surface temperature reconstruc- ful climatic signals. The individual than during any other extended pe- tions based on multiple proxies. proxy series used to create these riod from A.D. 900 onward. Finally, because some of the most reconstructions generally exhibit Based on the analyses present- important potential consequences strong correlations with local envi- ed in the original papers by Mann of climate change are linked to ronmental conditions and, in most et al. and this newer supporting changes in regional circulation pat- cases, there is a physical, chemi- evidence, the committee finds it terns, hurricane activity, and the cal, or physiological reason why plausible that the Northern Hemi- frequency and intensity of droughts the proxy reflects local tempera- sphere was warmer during the last and floods, regional and large-scale ture variations. Our confidence in few decades of the 20th century reconstructions of changes in other the results of these reconstructions than during any comparable period climatic variables, such as precipita- becomes stronger when multiple over the preceding millennium. The tion, over the last 2000 years would independent lines of evidence point substantial uncertainties currently provide a valuable complement to to the same general result, as in the present in the quantitative assess- those made for temperature. case of the Little Ice Age cooling ment of large-scale surface temper- and the 20th century warming. ature changes prior to about A.D. REFERENCES Surface temperature reconstruc- 1600 lower our confidence in this Intergovernmental Panel on Climate Change, 2001: Climate Change 2001: The Scientific Basis. Cam- tions for periods prior to the indus- conclusion compared to the high bridge University Press, Cambridge, UK. trial era are only one of multiple level of confidence we place in the Mann, M.E., Bradley, R.S., and Hughes, M.K., 1998: lines of evidence supporting the Little Ice Age cooling and 20th cen- Global-scale temperature patterns and climate forcing over the past 6 six centuries. Nature, conclusion that climatic warming tury warming. Even less confidence 392:779-787. is occurring in response to human can be placed in the original con- Mann, M.E., Bradley, R.S., and Hughes, M.K., 1999: activities, and they are not the pri- clusions by Mann et al. (1999) that Northern Hemisphere Temperatures During the Past Millennium: Inferences, Uncertainties, mary evidence. “the 1990s are likely the warmest and Limitations. Geophysical Research Letters, Surface temperature recon- decade, and 1998 the warmest year, 26(6):759-762. structions also provide a useful in at least a millennium” because National Research Council, 2006: Surface Tempera- source of information about the the uncertainties inherent in tem- ture Reconstructions for the Last 2,000 Years. National Academies Press, Washington, D.C. variability and sensitivity of the perature reconstructions for indi- climate system. To within existing vidual years and decades are larger uncertainties, climate model simu- than those for longer time periods, lations show that the estimated and because not all of the available temperature variations during the proxies record temperature infor- two millennia prior to the Industrial mation on such short timescales. Revolution can be explained plausi- bly by estimated variations in solar Next steps radiation and volcanic activity dur- Large-scale surface temperature ing the same period. reconstructions have the potential

PAGES NEWS, VOL.14, N°2, AUGUST 2006 Workshop Reports 37

The HOLIVAR Open Science Meeting

LONDON, UK; 12-15 JUNE 2006 After just over fi ve years, six workshops, three summer schools and one conference, the HOLIVAR program came to an end with a fi nal Open Science Meeting (OSM) at University College London from 12-15 June 2006. HOLIVAR was designed to bring climate change scientists together to promote interaction and collaboration between climate modelers Figure 1: Fossil pine stumps from the Cooran Lane, Galloway, South-west Scotland (photo and paleoclimatologists. At the courtesy of John and Hilary Birks) OSM, both communities combined Theme 1—Millennial time scales Theme 3—Climate variability in to present a state-of-the-art synthe- Eystein Jansen (Bergen, Norway) ar- the last 2000 years sis of Holocene climate variability gued that the Holocene “climate op- Mike Mann (Pennsylvania, USA) not- research via both keynote lectures timum” was a regional phenomenon ed that that the basic hockey-stick na- and poster sessions. driven by the northern hemisphere ture of Northern Hemisphere mean In opening the meeting, John summer insolation maximum and temperature trends for the last 1000 Birks (Bergen, Norway and UCL, amplifi ed by sea-ice feedback in the years was supported by many inde- UK) reviewed the history of Ho- polar regions. He also presented evi- pendent proxy- and model-based locene research, tracing its roots dence that sea surface temperatures analyses covering the same period. back to a competition by the Dan- were higher in the Medieval Warm However, he also emphasized the ish Academy of Sciences in Den- Period and lower in the Little Ice Age importance of moving beyond the mark in the 1830s to explain the but that these were out of phase with hockey stick. He explained how the occurrence of pine stumps in Dan- SSTs in the Caribbean. use of data-model comparisons, in- ish peat bogs (Figure 1). The prize Michel Crucifi x (Hadley Centre, cluding natural (especially volcanic) was won by Japetus Steenstrup UK) described the use of 2.5D and 3D forcing could explain many features (Figure 2) who argued in 1841 that models, explaining the importance of pre-19th century climate variabil- the plant and animal remains pre- of, but diffi culty in implementing, ity, including the El Niño-like nature served in bogs might refl ect past transient simulations in the Holo- of the Little Ice Age, and the regional changes in moisture and tempera- cene. He stressed the need for a pattern of change more generally as- ture. hierarchy of different modeling ap- sociated with El Niño and the NAO. The second keynote by Paul proaches, including modeling paleo- Hugues Goosse (Louvain-la- Valdes (Bristol, UK) reviewed the records themselves. Neuve, Belgium) presented model somewhat shorter history of Ho- results that showed how tempera- locene climate modeling from Theme 2—Decadal to centennial ture differences between regions COHMAP through to the pres- time scales ent day, focusing on PMIP1 and Jürg Beer (Zurich, Switzerland) de- 2 simulations and the continuing scribed the forcings that were im- evolution towards complete Earth portant on a decadal to centennial system models—an evolution driv- time-scale, focusing especially on en by comparisons between model solar forcing. He argued that there simulations and the paleorecord. was clear evidence for solar forcing In the third keynote, Frank Old- from the Holocene record but great fi eld (Liverpool, UK) urged future diffi culty in quantifying it. Earth system research to include Bas van Geel (Amsterdam, the the additional complexity intro- Netherlands) was concerned with duced by human-climate-environ- the potential drivers of decadal to ment interactions through time. He century scale climate variability. He pointed to the need for a greater pointed out that the role of solar understanding of how human so- forcing was poorly understood but cieties in the past were affected by could have caused past changes in or adapted to climate change. moisture balance suffi cient to affect Following these three intro- human societies. He noted that its Figure 2: Japetus Steenstrup (1813-1897), a pioneer of Holocene research. Steensrup ductory presentations, the pro- role may be underestimated in the won the Danish Academy of Sciences prize gram was organized around four present global warming debate. for “solving the problem” of how pine trees themes: once grew on Danish bogs.

PAGES NEWS, VOL.14, N°2, AUGUST 2006 38 Workshop Reports

could be due to spatial responses to rica during the Holocene, and that support. There were 150 posters particular forcings and/or to internal the climate at 5500 BP was especially covering the four themes, and book variability. He also argued that data unstable. He maintained that Earth prizes were awarded to the ﬁve best assimilation of paleo-proxy data in system models are now capable of student posters: Rixt de Jong (Lund, models could improve our under- simulating rapid swings between Sweden, Jun Inoue (Osaka, Japan), standing of past changes. arid and wet phases in the past but Bettina Stefanini (Dublin, Ireland), may not yet be able to predict future Auriel Per�oiu (Romania), and Peter Theme 4—Rapid hydrological rapid transitions reliably. De Geest (Brussels, Belgium). change In the final keynote, Ray Brad- Dirk Verschuren (Gent, Belgium), us- ley (Amherst, USA) presented an FURTHER INFORMATION ing proxy evidence from Europe and array of evidence to demonstrate The HOLIVAR2006 website (www.holi- var2006.org) will remain active for at least Africa, argued that a number of peri- the relevance of understanding past 12 months and is being updated with poster ods of cooler and wetter conditions, climate to provide insights for the pdfs. It will also provide video-streaming inferred from peatland and lake-level future. He stressed the importance of the keynote presentations. The ofﬁcial journal for papers is Climate of the Past, and changes in Europe, corresponded to of understanding the cause and the conference book (Global warming and periods of reduced solar activity. In consequences of rapid changes, natural variability: a Holocene perspective, Africa, especially in Eastern Equato- especially abrupt, unprecedented Eds. Battarbee, R.W. and Binney, H.) will be rial Africa, there was some evidence and persistent climate anomalies published by Blackwell in 2007.

for a similar inverse relationship be- (AUPs), for which there were many RICK BATTARBEE, HEATHER BINNEY AND MIKE tween solar activity and moisture. examples, mainly droughts, in the HUGHES Martin Claussen (Hamburg, Ger- paleo-record. Environmental Change Research Centre, many) argued that rapid climate The meeting was attended by 250 University College London, UK; [email protected] change, capable of affecting early people from 30 countries, including civilizations, occurred in North Af- 83 students, 18 of whom received PAGES/CLIVAR workshop on past millennia climate variability – synthesis and outlook

WENGEN, SWITZERLAND; 7-10 JUNE 2006

This workshop followed the spirit of • Late 20th century warming is lead to multidecadal-scale ef- the reorganized PAGES/CLIVAR work- likely anomalous in the context fects. ing group (see PAGES News, 2005/1). of the past 1000 years at hemi- • Usefully constraining estimates of It was organized by the PAGES/CLI- spheric scales. There is evidence global climate sensitivity from VAR Intersection Working Group (P. for periods of cooling and paleoclimate data will require a Jones, M. Mann, H. Wanner, K. Briffa) warming that occur on all tim- better knowledge of past radia- in concert with the PAGES office in escales and on all spatial scales. tive forcing and the amplitude Bern. Twenty-four participants (Fig. For pre-instrumental periods, of internal, as well as forced, 1), representing various different it is vital to consider both the natural variability. climate data and modeling sub-dis- spatial extent and duration of The discussions emphasized the ciplines were invited to discuss the regional climate anomalies. importance of distinguishing past state of the art and future needs in • Comparisons of model-predicted hemispheric or global-scale vari- the study of late Holocene climate and reconstructed climate varia- ability from regional variations. variability. Two days were devoted tions over the past several cen- For example, it was shown that the to short presentations and extend- turies are generally favorable, widely used term “Medieval Warm ed discussions on hot topics and taking into account the currently Period” is simply not an appropri- issues, covering the areas of proxy available data and their uncer- ate description of Medieval climate data, climate reconstructions and tainties. in many regions of the world. Coral paleoclimate modeling. The issue of • Natural radiative forcing appears data from Kim Cobb, for example, how to deal with uncertainties in as- to play an important role on the suggest instead a “Medieval Cool sessing climate variability over the relevant timescales. Solar forc- Period” for the tropical Pacific. Such past one-to-two millennia was dis- ing may account for variability considerations reinforce the principle cussed in detail. Questions posed on decadal through millennial that a better regional documentation and addressed included: What do timescales. Individual volcanic of past climate is necessary to better current uncertainty estimates take in- eruptions impact climate gen- understand the past. The group that to account? How relevant are current erally for only a few years but was assembled took note of the im- uncertainties for the general findings longer-term episodes of closely portance of focusing not simply on regarding past climate variability in spaced large eruptions (e.g. as the often emphasized hemispheric the past? The following primary con- in the early 19th century) can mean temperature variations of clusions were reached: past centuries, but also on spatial

PAGES NEWS, VOL.14, N°2, AUGUST 2006 Workshop Reports 39

construction challenge” in the near future (see Program News, page xx, this issue). This challenge will build on the theme of model intercom- parison, using synthetically derived “pseudoproxy” test datasets derived from climate models as a test of methods used in paleoclimate reconstruction, similar in many respects to last year’s “EPICA Challenge” initiative (see e.g. EOS, No. 38, 2005). A second project addressing proxy data uncertainties and data archiving is also being planned. The four-day workshop in the unique setting of the snow-covered Figure 1: The group of participants in front of the Wengen mountains. Bernese Alps provided a stimulating setting for informal discussions, set- patterns of surface temperature, at- importance of encouraging scientists ting the stage for productive activi- mospheric circulation, precipitation to provide not just the proxy data ties in the months and years ahead. and drought. The group also noted and climate reconstructions, but as A synthesis publication on the state the importance of considering dy- much information as possible about of the art of late Holocene climate namical mechanisms such as the El random and systematic error and variability will be published by the Niño/Southern Oscillation (ENSO) in uncertainty. The latter information is workshop participants. Follow-up ac- interpreting past climate changes. crucial, for example, in comparisons tivities are currently being planned. The recent PAGES and CLIVAR ini- of paleoclimate evidence with model tiatives on past regional variability simulation results. The participants ACKNOWLEDGEMENTS in South America and the Mediter- agreed that progress in climate mod- The workshop was funded by the Electric Power Research Institute (EPRI), PAGES, ranean region follow this principle eling approaches is also crucial to a Swiss NCCR-Climate and CLIVAR. (see PAGES News 2005/1, 2005/2 better understanding of past climate and 2005/3). variability, and to the assessment of CHRISTOPH KULL1, KEITH BRIFFA, PHIL JONES, The group recognized that addi- current and future anthropogenic MICHAEL MANN AND HEINZ WANNER 1PAGES Ofﬁce, Bern, Switzerland tional effort is needed in the archiving impacts on climate. To this end, the [email protected] of paleoclimate data and associated workshop participants agreed to plan metadata. They also agreed on the a community-wide “paleoclimate re- ICDP workshop PASADO: Deep drilling at Laguna Potrok Aike, a maar lake in southern Argentina

RÍO GALLEGOS, ARGENTINA; 15-19 MARCH 2006 Patagonia is subjected to shifts in po- the German Climate Research Pro- not only inter-site but also inter-ar- lar and mid-latitude winds, pressure gram (DEKLIM) and carried out in chive correlations available. Such ﬁelds and precipitation regimes, as close cooperation with Argentine data will advance hemispheric recon- well as to variations related to El Ni- scientists (PAGES News 12/2, 9-11). structions of climate variability and ño Southern Oscillation (ENSO) and On-site monitoring and multidisci- proxy-model comparisons to better the Antarctic Oscillation (AO). There- plinary reconstructions based on understand climate forcing factors. fore, environmental and climatic re- sediment cores from the maar lake First results are promising but coring constructions have the potential to Laguna Potrok Aike have formed the ended around 16 cal ka BP, due to provide unique records of variations basis for environmental and climat- technical restrictions of the available in the hydrological cycle, changes in ic reconstructions on decadal up to coring device. aeolian dust deposition, and conse- millennial time scales. Located 52°S, With additional funding from quences and frequencies of volcanic the lake received dust and volcanic the Priority Programme “ICDP” activities. Since 2001, crater lakes of ash from Patagonia and the Andean of the German Research Foun- the Pali Aike Volcanic Field (south- volcanic chain, which are distributed dation (DFG), it was possible to ernmost Patagonia, Argentina) have not only across South America but carry out deep seismic surveys, been studied in the framework of also blown over the South Atlantic which revealed >150 m of lacus- the interdisciplinary project “South and onto Antarctica. Thus, isotopic trine sediments underlain by >350 American Lake Sediment Archives and aerosol tracers could be used m of volcano-clastic deposits. As and Modelling” (SALSA) funded by to link these different areas, making the maar formation occurred at

PAGES NEWS, VOL.14, N°2, AUGUST 2006 40 Workshop Reports

770±220 ka, seismic data suggest that lacustrine deposits may cover up to seven glacial cycles. This calls for a much larger scientific endeav- or that can only be accomplished within the framework of the Interna- tional Continental Scientific Drilling Program (ICDP). Thus, the research initiative “Potrok Aike Maar Lake Sediment Archive Drilling Project” (PASADO) was developed, with the goal to get the GLAD800 coring system to Patagonia. Figure 1: Participants of the PASADO workshop in Río Gallegos. The four-day ICDP PASADO work- environmental change in the field. and several proposals have been shop, consisting of 46 talks organized The day ended with an “asado”, a fa- developed and submitted to national into 10 sessions, was convened in mous Argentine-style barbeque, on funding agencies in support of PAS- the Conference Hall at the Complejo the premises of the INTA field station ADO. However, although the first Cultural Santa Cruz in Río Gallegos, Potrok Aike. milestone has passed, there are still Argentina from 15-19 March 2006, The concluding day was devoted plenty of opportunities to expand bringing together 52 participants to break-out groups on four major the scientific community involved from 11 countries (Fig. 1). topics related to technical aspects through international collaboration. The first two sessions introduced of scientific drilling and on volca- The scientific program, list of par- participants to the study area. At- nological, inorganic and organic ticipants, abstracts of presentations tendees were impressed with the aspects of sediment core analysis. and excursion guide were published data presented and felt that Laguna During a final plenary discussion, as Terra Nostra 1/2006. Further infor- Potrok Aike should be recommended all the ideas and suggestions were mation can be obtained from www. as an appropriate target for future pooled. salsa.uni-bremen.de. ICDP deep drilling because it is not The ICDP PASADO workshop only a unique site for terrestrial cli- proved to be very successful, both ACKNOWLEDGEMENTS mate reconstruction in the southern in terms of scientific collaboration, The PASADO workshop was organized by the team of GEOPOLAR (Institute of Geogra- hemisphere but also for investiga- as well as technical and logistical phy, University of Bremen) and hosted by the tion into the formation of a relatively issues. Participants actively fine- Universidad Nacional de la Patagonia Austral young mid-Pleistocene phreatomag- tuned analytical strategies, focused (Río Gallegos). It was supported logistically by the Government of Santa Cruz and the matic crater. A fieldtrip during the on improving cooperation and coor- Instituto Nacional de Tecnologia Agropecuaria workshop presented an overview of dination between scientific groups, (INTA), and financially by ICDP. the lake and the surrounding catch- and developed the timeframe and 1 ment area from a prominent basalt outline for the full ICDP deep drill- BERND ZOLITSCHKA , HUGO CORBELLA, NORA MAIDANA AND CHRISTIAN OHLENDORF escarpment, and focused on techni- ing proposal. The science plan that 1University of Bremen, Germany; cal and logistical issues, as well as was established already serves as a [email protected] on volcanic evidence and records of basis for international cooperation, Climate change: A multi-dimensional challenge

LUND, SWEDEN; 26-28 OCTOBER 2005 As part of the British Council’s pro- work with instrumental time series was a founding Chair of the Intergov- gram International Networking for and those who work on climate ernmental Panel on Climate Change Young Scientists (INYS) a British- modeling. Six keynote lectures were (IPCC) and was instrumental in the Swedish-Danish workshop was given and these included topics such initial formulation of the World Cli- organized by the Swedish Meteo- as projections of climate change over mate Research Programme (WCRP). rological and Hydrological Institute, the next few decades, the usefulness In his 80th year, he captivated a full the Geological Survey of Denmark of global climate models, the added audience with his personal and often and Greenland, the Climate Research value of regional modeling, and the revealing insight into how the scien- Unit at the University of East Anglia, nature of the geological data that tiﬁc community must relay their con- and the Hadley Centre (part of the form the basis of a reconstruction of cerns about climate change to the Meteorological Ofﬁce). A principal a widespread cooling event at ap- public and politicians. His message aim of the workshop, which brought prox. 8200 years ago. was clear; there is a great deal of together 19 young scientists, was to A highlight of the workshop was knowledge on climate and, notwith- promote interaction between people Professor Bert Bolin’s keynote lec- standing the remaining uncertain- who generate reconstructions of past ture “To use scientific knowledge ties (Fig. 1), urgent measures must climates from paleo-data, those who on climate change in politics.” Bert

PAGES NEWS, VOL.14, N°2, AUGUST 2006 Workshop Reports 41

road-rail link to Copenhagen. It was, therefore, appropriate that the workshop ended with a visit to the Niels Bohr Institute to see the ice cores recovered from the Greenland ice sheet.

ACKNOWLEDGEMENTS The organizers are grateful to the British Council for their financial support and for the efficient management of the workshop by Ki Figure 1: A) Annual mean temperature change from 1971-2000 to 2081-2090, as averaged Andersson (Sweden) and Michael Sorensen- over 21 models forced with the SRES A1B emissions scenario. B) Estimated probability that Jones (Denmark). We thank the participants temperature change exceeds zero. The relatively low probability in the northern North Atlantic and keynote lecturers for an intellectually is a result of models that simulate cooling due to changes in ocean circulation (Räisänen and intense and stimulating few days, and Sune Ruokolainen, 2006, Tellus A, 58). Olander Rasmussen for his guided tour of be taken to curb human influence on of peat, lake and marine sediments. the Niels Bohr Institute. the properties of the atmosphere. This fieldtrip was followed by a “cli- A parallel program was also mate panel debate”, in which the stu- IAN SNOWBALL1, SVANTE BJÖRCK, RICHARD scheduled for approx. 50 high school dents were given the opportunity to BRADSHAW, KEITH BRIFFA, RICHARD JONES AND MARKKU RUMMUKAINEN students. These students were taken ask scientists at the GeoBiosphere 1Lund University, Sweden; to a local peat bog where they were Science Centre about past, present [email protected] given an introduction to palaeocli- and future climates. Lund is located matology, which included the coring only 20 km from the recently opened New trends in Geomorphology – systems-based understanding of long term man-landscape interactions

LUCIFS WORKSHOP, MUENZENBERG, GERMANY; 11-14 MAY 2006

LUCIFS is a constitutive part of PAGES Focus 5 “Past Ecosystem Processes and Human-Environment Interactions”. The LUCIFS research explores past to present responses of fluvial systems to climate change and human activities. After 2000, the LUCIFS group organized a second major workshop open to all interested members of the research community. The workshop focused on man-environment interactions, Figure 1: Participants of the LUCIFS 2006 workshop in Muenzenberg, Germany. particularly on geomorphological which 5 keynote talks considered On Saturday afternoon, small and sedimentological perspectives the state of the art in various fields working groups elaborated on se- on mid- to long-term quantitative of man-environment research. All lected topics of Holocene human- sediment fluxes. contributions were of outstanding environment interactions. The 4 The meeting was convened in quality and sparked animated dis- thematic foci were: Muenzenberg near Frankfurt, Ger- cussions. - Systems-based understanding many from 11-14 May 2006. The The workshop program was of earth surface change with LUCIFS leader, Richard Dikau (Univ. supplemented by a pre-workshop respect to Holocene sediment Bonn, Germany), and the local orga- fieldtrip led by Peter Houben. Dur- flux. nizers, Peter Houben (Univ. Frank- ing the fieldtrip, the methodical ap- - Evaluating quantitatively human furt, Germany), Lothar Schrott (Univ. proach and results of a Holocene impact on earth surface sys- Salzburg, Austria) and Jürgen Wun- sediment budget study for Rock- tems. derlich (Univ. Frankfurt, Germany) enberg catchment were presented. - Exploring the sedimentary record: were very pleased to welcome 41 One of the main messages there was cross-cutting approaches to participants from 5 continents and that in this area, the spatial pattern of catchment-scale sediment flux. 14 countries (Fig. 1). The workshop erosion and redeposition, the rout- - Coupling changing pressures of provided a stimulating working at- ing of sediments and the changes in anthropogenic and natural driv- mosphere with high-profile talks. rates of flux and delivery along the ers with modeling approaches. The tight schedule consisted of 19 sediment cascade have been chiefly Structured discussions helped to oral and 11 poster presentations, of controlled by human actions. identify key issues for further re-

PAGES NEWS, VOL.14, N°2, AUGUST 2006 42 Workshop Reports

search, which were presented to the addressed future LUCIFS research ACKNOWLEDGMENTS The Deutsche Forschungsgemeinschaft whole group of participants in a suc- strategies against the backdrop of the (DFG) gave considerable support by ceeding plenary session. At least one workshop contributions and current covering all workshop costs including staff of the small groups decided to edit a approaches to Earth System Science appropriations for the large number of student helpers. PAGES co-sponsored the joint review paper based on the top- and will result in a revised LUCIFS workshop by funding the travel costs of a ics discussed during that afternoon. research plan. Workshop contribu- number of participants from Eastern Europe Richard Dikau chaired the ﬁnal tions and results will be published and South America. plenary discussion on Sunday morn- in a special issue of Geomorphology PETER HOUBEN ing. His introduction dealt with the (Elsevier). People interested in LU- Wolfgang Goethe University, Frankfurt, development of LUCIFS objectives CIFS activities are invited to visit the Germany; [email protected] and approaches over the past 6 PAGES Focus 5 homepage (www. years and included a critical review pages-igbp.org/science/research/fo- of past deﬁciencies. The discussions cus5.html) for further information. Linkage between marine and terrestrial processes during past rapid climatic changes

BARCELONA, SPAIN; 13-15 MARCH 2006

This workshop was organized by a Alboran Sea were discussed new working group of the INQUA for their potential in obtaining Commission on Coastal and Marine ultra-high resolution records. Processes. The group was inspired New data from the Gulf of Ca- by our sadly missed colleague, diz provided the basis for an Nick Shackleton, with the aim of IODP drilling proposal under analyzing paleoclimatic records evaluation. from an integrated marine and ter- Zone 5: The Western Iberian restrial perspective. The objective Margin is a unique area for of the workshop was to identify marine-terrestrial linkage at key scientiﬁc questions that could centennial-millennial scale. be addressed through future drill- New Holocene records off the ing/coring programs. New data sets Tagus River resolve decadal from various regions (Fig. 1) were changes linked to the North reviewed, and stimulated discus- Atlantic Oscillation (NAO). sions on atmospheric and oceanic Holocene sediments from the processes active in propagating Galician Rias record changes and modulating past rapid climatic in the Meridional Overturning variability: Circulation (MOC). Marine pol- Zone 1: Sediment cores from the len records suggest different delta fan systems of the Congo Figure 1: Map illustrating the regions from timing in the evolution of veg- which data were presented and discussed Ogooué, Nyong and Sanaga during the workshop: (1) F. Marret and S.-Y. etation and sea level, providing rivers record changes in river Kim; (2) S. Weldeab; (3) L. Dupont and O. an intriguing insight into the discharge associated with Romero; (4) I. Cacho, F. Martinez, F. Lobo, feedback mechanisms during L. M. Fernández-Salas, E. Llave and J. Her- latitudinal migration of the In- nandez-Molina; (5) F. Abrantes, T. Rodrigues, glacial interceptions. tertropical Convergence Zone L. Pena, C. Tzedakis and M. F. Sanchez-Goñi; Zone 6: Sediments from the south- (ITCZ) over Africa during the (6) J. Scourse; (7) W. Austin and I. Hall; (8) western British Margin record I. Hall. Map Source: IHO-UNESCO, General last deglaciation and Holocene Bathymetric Chart of the Oceans, Digital Edi- the millennial-scale evolution periods. tion, 2003, www.ngdc.noaa.gov/mgg/gebco of the British ice sheets (BIS). Zone 2: The evolution of the East (thanks to B. de Mol for data map process- The role of the British Shelf ing). African ITCZ was discussed on in emitting or trapping atmo-

the basis of sediment cores means of new high-resolution spheric CO2 was discussed on from the Nile delta fan, which records from around Cape the basis of model and proxy records changes in the hy- Blanc and off Morocco. data of sea level, tidal dynam- drological regime of the Nile Zone 4: Multi-proxy analyses of ics and primary production. catchment regions. Alboran Sea sediments provide Zone 7: The North British Margins Zone 3: The humidity evolution of records of both marine and provide sequences with ex- North Africa and the interplay terrestrial processes at Dan- traordinary potential to study between low- and high-latitude sgaard-Oeschger frequency. BIS sensitivity, and the inter- processes during rapid climatic Geophysical data from shallow play between surface oceanog- changes were discussed by areas in the Gulf of Cadiz and raphy and MOC. New records

PAGES NEWS, VOL.14, N°2, AUGUST 2006 Workshop Reports 43

from the Scottish Fjords show ciﬁc regions such as the N. African will be held in Oban, UK from 15-16 the potential for monitoring and W. Mediterranean regions and September 2006. past decadal changes linked to the British Margin, to work on fu- the NAO. ture drilling proposals. We also ACKNOWLEDGMENTS Zone 8: The Atlantic-Indian Ocean plan a thematic workshop on the This workshop was co-sponsored by INQUA, PAGES, the Spanish Ministry of connection in relation to South geographical distribution of Hein- Education and Science (Grant: CGL2005- African climate was discussed rich event signatures across the N. 23763-E-CLI) and the Faculty of Geology, in the context of new sediment Atlantic region by September 2007. University of Barcelona. records from the Agulhas Cur- An associated workshop on “Fjord ISABEL CACHO rent and a submitted IODP drill- Environments: Past, Present and Fu- University of Barcelona, Spain; ing proposal. ture” (post Challenger 2006 meet- [email protected] Future group activities will involve ing; www.sams.ac.uk/challenger/) new workshops focusing on spe- Monitoring Indonesian throughﬂow variability: Challenges and perspectives

KIEL, GERMANY; 19-22 JULY, 2006

The transport and thermohaline North Pacific Thermocline • Little Ice Age (1500-1900) includ- South Pacific Thermocline Pacific Ocean stratification of the Indonesian ing Tambora eruption (1815) as Celebes Sea Throughﬂow (ITF, Fig. 1) inﬂuence Borneo a marker event Halmahera Molucco Sea the heat and freshwater budgets 1.5 • Medieval Warm Period (1000 si e ~ a w in assar Straitla N u a of both the Pacific and Indian S L AD±200) Mak 8

Oceans, and alter patterns of heat Banda Sea • Early Holocene (9 ka) and water vapor exchange with Java • Last Glacial Maximum (19-23 ka)

k r o o 4.3 b 4.5 im m T o 1.7 the atmosphere on a global scale. L • Specific Marine Isotope Stage 10 ITF The ITF, which is tightly linked to Indian Ocean Australia (MIS) 3 events regional climate systems, such as Figure 1: Main pathways of present-day Indo- • MIS 5e (122 ka) the El Niño Southern Oscillation nesian Throughflow at thermocline depth and • MIS 11 (400 ka) estimates of total volume transport (in Sver- (ENSO) and the Australasian mon- drups). Modified from Gordon (2005) Longer time series will provide in- soon, also plays a central role in sight into (1) the phasing between the global “conveyor” circulation overview of the latest research ini- ITF variations and other factors: thereby exerting a critical control tiatives on oceanography and pa- warm pool changes, monsoons, on Earth’s climate. It is likely that leoceanography in the ITF region, atmospheric CO2 and sea level, the ITF was substantially modu- and to outline urgent research and; (2) the ITF response to the lated by changes in the geometry goals for the next few years. “mean state of the tropical Pacif- of the Indonesian pathways due Proxy studies that target the ic” (“ENSO” thermocline tilt, SST to sea level changes during the reconstruction of inter-ocean gra- distribution). New IMAGES coring Pleistocene and tectonic reorgani- dients and the transfer of heat, salt initiatives will focus on regions zations of the archipelago during and nutrients are crucial. These re- that still have poor core coverage: the Cenozoic. Because of its close constructions can be accomplished Makassar Strait, the southern Ce- link with regional climate systems, by providing reliable estimates of lebes Sea, east and west of Halma- it may have also experienced more (1) thermocline structure and tem- hera, Molucca Sea, Ninetyeast subtle changes during the Holo- perature in the ITF region, and; Ridge and West of Sumatra. Since cene. The processes driving ITF (2) ITF circulation changes (inten- only a few piston cores in this high variability and its magnitude are, sity, vertical structure of flow and accumulation rate region will reach however, not very well constrained temperature weighted flow). Ulti- back to MIS 12, the drilling of new nor is the effect of ITF variability on mately, time slice reconstructions IODP Sites in the Timor Sea and tropical and extra-tropical climate should be integrated with modeling Makassar Strait will be crucial for fully understood. studies to evaluate the influence of evaluating the impact of tectonic To address these issues a the- changing boundary conditions on and orbital forcing on regional and matic workshop was held from the ITF (i.e., insolation, volcanic ep- global climate. 19-22 July 2006 at the Institute of isodes, geometry of passages, sea Geosciences, Kiel University, Ger- level), and the linkages between WOLFGANG KUHNT many, attended by 22 scientists ITF variability and fluctuations in Christian-Albrechts-Universität, Kiel, Germany; from the U.S., Switzerland, France, regional and global climate. Time [email protected] Germany, China, India, Indonesia, slices of specific interest are: Timor Leste and Australia. The goal • Last 50 years of the meeting was to present an

PAGES NEWS, VOL.14, N°2, AUGUST 2006 44 Last

CALENDAR 2006/2007

October 24 - 27, 2006, Birmingham, UK November 20 - 23, 2006, Istanbul, Turkey Rapid Climate Change - International Science Climate Change and the Middle East - Past, Present, Conference Future

Further Information: Further Information: www.noc.soton.ac.uk/rapid/rapid2006/ www.climatechange_middleeast.itu.edu.tr/

November 9 - 11, 2006, Carmona, Seville, Spain March 25 - 30, 2007, Ventura, USA First MedCLIVAR workshop: Reconstruction of past Gordon Research Conference on Polar Mediterranean climate Marine Science

Further Information: Further Information: www.medclivar.eu/ www.grc.org/programs/2007/polar.htm

November 9 - 12, 2007, Beijing, China April 11 - 14, 2007, Nanjing, China ESSP Open Science Conference - Global Environmental 2nd LIMPACS Salinity, Climate Change and Change: regional Challenges Salinisation Workshop

Further Information: Further Information: www.geog.ucl.ac.uk/ecrc/limpacs/events.htm www.essp.org/essp/ESSP2006/

www.pages-igbp.org/calendar

New on the PAGES bookshelf ��

��

PAGES NEWS, VOL.14, N°2, AUGUST 2006 Printed on wood- and chlorine-free paper