Eva Interglaciation Forest Bed, Unglaciated East-Central Alaska: Global Warming 125,000 Years Ago bJ TroJ L. Ptwt • Glenn W. Berger • John A. Westgate • Peter M. Brown • Steyen W. LeaYitt --- - • - .l' - --· San 319 Copyright © 1997, The Geological Society of America, Inc. (GSA). All rights reserved. GSA grants permission to individual scientists to make unlimited photocopies of one or more items from this volume for noncommercial purposes advancing science or education, including classroom use. Permission is granted to individuals to make photocopies of any item in this volume for other noncommercial, nonprofit purposes provided that the appropriate fee ($0.25 per page) is paid directly to the Copyright Clearance Center, 27 Congress Streett Salem, Massachusetts 01970, phone (508) 744- 3350 (include title and ISBN when paying). Written permission is required from GSA for all other forms of capture or reproduction of any item in the volume including, but not limited to, all types of electronic or digital scanning or other digital or manual transformation of articles or any portion thereof, such as abstracts, into computer readable and/or transmittable form for personal or corporate use, either noncommercial or commercial, for-profit or otherwise. Send permission requests to GSA Copyrights. Copyright is not claimed on any material prepared wholly by government employees within the scope of their employment. Published by The Geological Society of America, Inc. 3300 Penrose Place . P.O. Box 9140 Boulder, Colorado 80301 Printed in U.S.A. . GSA Books Science Edjtor Abhijit Basu Library of Congress Cataloging-in-Publication Data Eva lnterglaciation Forest Bed, unglaciated east-central Alaska : global warming 125,000 years ago I by Troy L. Pewe .. . [et al.]. p. em. -- (Special paper; 319) Summaries in French and German. Includes bibliographical references (p. - ). ISBN 0-8137-2319-1 1. Trees, Fossil--Alaska. 2. Geochronometry--Alaska. 3. Paleobotany--Pleistocene. 4. Glacial climates--Alaska. 5. Geology, Stratigraphic---Pleistocene. l Pewe, Troy Lewis, 1918- . IT. Series: Special papers (Geological Society of America) ; 319. QE991.E93 J9 97 551.69798--dc21 97-25874 CIP Cover photograph: Stem djsk of spruce (Picea sp.) cut from log preserved in permafrost in Eva Forest Bed, 125 000 years old. Tree was 144 years old and grew at elevation of 250m on south-facing slope. Collected by Pewe and Westgate in 1987, from mining exposure at Eva Creek, 14 km west of Fairbanks, Alaska, through the courtesy of Walter Wigger and Mike Wigger, mine owners. Stem disk prepared and tree rings counted by Peter Brown Laboratory for Tree .Ring Research, University of Arizona, Tucson. Pewe sample No. 187 {Table 1). Photograph No. 30,373 by S. M. Selkirk, Arizona State University, Tempe, February 20, 1997. 10 9 8 7 6 5 4 3 2 1 • VI ~Contents Abstract. . ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . 1 Introduction ......................................... .. ........ ........................ 5 ·~ Ackn.ow led.gments .. . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Physical setting . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Topography and geology .......................... ................. .................... 8 Summary of late Cenozoic stratigraphy . . . . . . . . . . . . . . . . . . . . . . . . . . .................. .... 8 ~ Modem cli_mate ............ __ . .. . . . . . . . . . . . . . . . . . .. . . . . . . . ....... 10 ~ 8 •••••••••••••••. •• •. Present per 111afrost . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Modern vegetation .............................. .. ...... ................. ... ... ... .... 12 Eva Forest Bed ............................................................... . ..... .. 14 . . D escnpt1on ...................................................... .. .. ...... . ..... . 14 Eva Forest Bed as a stratigraphic unit ................................................ 14 Trees of the Eva Forest Bed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ., . . . .. . . . . . . . . . . . . . . 14 ~ Poll·en . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , . . 21 Distribution ........... " ..... ........ ....................................... ....... 21 Stratigraphy ................. .. .................... . ................ ............... 23 Genera) statement .......... .. .. .. .... .... ............................ ... .............. 23 Relation of tephra layers in upper Gold Hill Loess Eva Forest Bed ............... . . . ,. .. 25 to ~ ~ Age . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . .. . . . .. . . . .. . . . 35 Early stratigraphic interpretation .............. . ... ............... ....... ........... 35 Radiocarbon dating ......................... .. ...................... .. . .... .. ..... . . 35 Tephrochronology .................. . ............................................. 37 Therrnolumine cence ......... ....... . .... ........................................ 37 Summary of age discussion ...................... . ..................... ...... ..... 38 ~ Paleoenvironmental interpretation ........................................ . ....... . ..... 40 Preliminary statement . . . . . . . . . . . . . . . . . ........................... ................. .. .. 40 Environment of the loess formations bracketing the Eva Forest Bed ....... . ... .. ...... .. ..... 40 Early history of investigation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . 40 Modern. systematic studies ....... . .................. . ..... . .. . ..... ...... ........ ... 41 •• Vll ••• Contents Vlll Dendrochronology of Eva Forest Bed trees and comparisons with modem trees ................ 42 Introduction ...................................................................... 42 Ring width and density in Eva Forest Bed and modem trees .............................. 42 General ring characteristics of the Eva Forest Bed samples ............................... 43 Statistics of ring width and density time series. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 Variability of past climate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ....... 45 13C/12C isotopic ratios of Eva Forest Bed trees and comparison with modem and Holocene trees ... 47 ot3C comparison of modem, Holocene, and Eva Forest Bed trees .......................... 47 o Environmental implications of 48 13C ................................................. Summary of the environment of the Eva interglaciation Forest Bed .......................... 48 Introductory statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 Botanical evidence Yukon-Tanana Upland .......................................... 49 Physical evidence Yukon-Tanana Upland ........................................... 49 Adjacent Yukon Territory, Canada .................................................. 50 Astronomical climatic inferences ................................................... 50 Concluding remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... 50 References cited . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 Geological Society of America Special Paper 319 1997 Eva Interglaciation Forest Bed, Unglaciated East-Central Alaska: Global Warming 125,000 Years Ago ABSTRACT The Eva Interglaciation Forest Bed represents a frozen,buried,ancient boreal for- est in the Yukon-Tanana Upland of east-central Alaska. It consists of excellently pre- served peat lenses,sticks,roots,and logs as well as rooted and unrooted stumps of trees, mainly spruce and birch. Consistent with the modern boreal forest, the largest and most common tree in the fossil forest is spruce, mainly white spruce (Picea glauca). Remains of birch trees are common,mostly Betula papyrifera. The forest remains were buried by loess that became frozen and so are well preserved. None of the wood is min- eralized. Many of the fragments are black from buring, suggesting forest fires were widespread in the Yukon-Tanana Upland during the interglaciation. Also,evidence is presented for the first time of the existence of spruce bark beetles (Scolytidae) during the last interglaciation in Alaska. Efforts to determine the age of the Eva Forest Bed in this study have covered the past 50 years. Methods applied have varied from the use of stratigraphic interpreta- tion of sedimentological events and preserved evidence of climatic changes to the use of modern geochronometry. Several methods of dating have come to fruition in the 1990s. New radiocarbon dating by liquid scintillation (LS) detectors indicates the forest wood to be older than 70,000 years. Perhaps the greatest breakthrough is the development of the isothermal-plateau fission-track method of dating geologically young volcanic glass shards. The Old Crow tephra closely underlying the Eva Forest Bed has been dated at 140±10ka and strongly supports the original interpretation of the forest bed as of last interglaciation. In the early 1990s,highly improved thermoluminescence (TL) sediment dating techniques were utilized for dating loess above and below the forest bed indicat- ing the age of the Eva Forest Bed is probably 125,000 years with a duration of the Eva Interglaciation of probably only a few thousand years (Sangamon, Oxygen Isotope Substage5e). Stratigraphically,the Eva Forest Bed lies at the prominent unconformity between the underlying massive, green Gold Hill Loess (pre-Sangamon) and the overlying blackish,ice-wedge-rich retransported loess of the Goldstream Formation (Wisconsin). Studies of the frozen Gold Hill Loess indicate that the warm interglacial interval was characterized by deep and rapid thawing of permafrost and erosion of loess accompa- nied by gullying and block slumping of frozen loess. After extensive slumping, the topography became smooth and the forest became extensive. Tilting of enclosed tephra layers outline the slump blocks. Evidence for deep permafrost thawing is supported by the absence today of ice wedges,buried pingos,and mammal carcasses in the presently refrozen loess of pre-Wisconsin age. Deep thawing is also indicated by reduction of iron on loess grains from ferric to ferrous turning the traditional tan color of loess to green- ish in the buried Gold Hill Loess. It is the unique sequence of refreezing in Wisconsin time that has preserved the remarkable evidence for deep thawing in earlier Sangamon Péwé,T.L.,Berger,G.W.,Westgate,J.A.,Brown,P.M.,and Leavitt,S.W.,1997,Eva Interglaciation Forest Bed,Unglaciated East-Central Alaska:Global Warming 125,000 Years Ago:Boulder,Colorado,Geological Society of America Special Paper 319. 1 2 T. L. Péwé and Others time—the green color. The forest bed formed after much of the thawing,erosion,and slumping activity had ceased,and it overlies the angular unconformity. More than half- a-dozen distinct tephra layers have been identified,characterized,and correlated in the upper part of the Gold Hill Loess,aiding in the reconstruction of the sequence of events leading to the erosion,thawing,and emplacement of the Eva Forest Bed. Dendrochronology studies of trees and 13C/12C isotopic ratios of wood from the Eva Forest Bed,and comparisons with wood from the modern boreal forest,strongly sug- gest environmental conditions at least similar to those of today. Some plant remains and ground beetle taxa of Eva Forest time in Canada represent species that extended farther north than they do today. Also,buried spruce macrofossils suggest that the boreal forest may have extended north of the Brooks Range in Alaska. These botanical and physical data indicate an environment warmer than the pres- ent interglaciation with the mean annual air temperature warmer than 0°C,perhaps +1 or +2°C or warmer to permit the ice to melt and permafrost to thaw from the sur- face downward. Supporting this concept are astronomical inferences that during the last interglacial (Oxygen Isotope Substage5e) the July insolation anomaly at 65°N. lat- itude reached values of almost 50% higher than 10,000 years ago,the beginning of the Holocene Interglaciation. ZUSAMMENFASSUNG est Bed unterlagernden Old Grow Tephra wurden auf 140000 Jahre ±10000 datiert,was sehr stark die ursprüngliche Einord- Das “Eva Interglaciation Forest Bed”stellt einen gefrorenen, nung der Forest Bed in das letzte Interglazial unterstützt. In den sedimentüberdeckten, ehemaligen borealen Wald im Yukon frühen neunziger Jahren wurden die stark verbesserten Tanana-Hochland von Ost-Zentralalaska dar. Es besteht aus her- Thermolumineszenz-Sedimentdatierungsmethoden eingesetzt, vorragend erhaltenen Torflinsen,Ästen,Wurzeln und Baumstäm- um den Löß über und unter den Forest Bed zu datieren. Sie men sowie aus bewurzelten und wurzellosen Baumstümpfen, deuten auf ein Alter des Eva Forest Bed von wahrscheinlich hauptsächlich von Fichten und Birken. Fichten,insbesondere Picea 125000 Jahren mit einer Dauer des Eva Interglazials von glauca (Weißfichte),sind die größten und häufigsten Bäume dieses wahrscheinlich nur wenigen tausend Jahren (Sangamon, Oxy- fossilen Waldes,der somit mit den heutigen borealen Wäldern ver- gen Isotope Substage5e). gleichbar ist. Überreste von Birken, hauptsächlich von Betula Die Eva Forest Bed liegen an der bekannten stratigraphischen papyrifera, sind häufig. Die Waldüberreste wurden durch Löß Grenze zwischen dem unterlagernden, grünen Gold Hill Loess überschüttet,das anschließende Gefrieren des Materials sorgte für (pre-Sangamon) und den darüber liegenden schwarzen,eiskeilre- gute Erhaltung. Es fand keine Mineralisierung des Holzes statt. ichen,umgelagerten Lössen der Goldstream Formation (Wiscon- Viele der Bruchstücke zeigen Brandspuren,was die Vermutung sin). Studien des gefrorenen Gold Hill Loess zeigen an,daß die nahelegt, daß Waldbrände während des Interglazials im Yukon warme interglaziale Phase durch ein tiefes und rasches Auftauen Tanana-Hochland weit verbreitet waren. Erstmals wird auch der des Permafrostes und durch Erosion von Löß charakterisiert ist, Beweis für das Vorkommen des Borkenkäfers (Scolytidae) begleitet durch Gullybildung und Sackungen von gefrorenem Löß. während des letzten Interglazials in Alaska erbracht. Infolge der ausgedehnten Materialverlagerungen wurde die Die Bemühungen zur Festlegung des Alters der Eva Forest Topographie ausgeglichen und der Wald konnte sich stark ausbre- Bed reichen fünfzig Jahre zurück. Die angewendeten iten. Die Rutschungen werden durch die schräg gestellten Arbeitsmethoden reichen von der stratigraphischen Interpreta- Tephralagen charakterisiert. Das tiefe Auftauen des Permafrostes tion von Sedimenterfolgen und den darin enthaltenen Hin- wird durch das Fehlen von Eiskeilen, überschütteten Pingos, weisen auf Klimaänderungen bis zum Einsatz moderner Kadavern von Säugetieren in der heute wiederum gefroreren Löß Geochronometrie. Verschiedene Datierungsmethoden brachten präwisconsinzeitlichen Alters belegt. Auch die Reduktion von erst in den neunziger Jahren Erfolg. Neue Radiocarbon- Eisen auf Lößpartikel belegt das tiefe Auftauen des Permafrostes, Kartierungen mittels LS-Detektoren (Liquid Scintillation) der überschüttete Gold Hill-Löss wird dadurch grünlich. Die grüne deuten darauf hin, daß der Wald älter als 70000 Jahre ist. Der Farbe als bemerkenswerter Beweis für das tiefe Auftauen in der vielleicht größte Durchbruch ist die Datierung der geologisch frühen Sangamonzeit verdankt ihre Erhaltung dem einzigartigen jungen vulkanischen Glasbruchstücke mittels der IPFT- Wiedergefrieren im Wisconsin. Die Forest Bed entwickelten sich Methode (Isothermal-Plateau Fisson-Track). Die das Eva For- nach Beendigung der Auftau-,Erosions- und Rutschaktivität. Über Eva Interglaciation Forest Bed 3 ein halbes Dutzend verschiedener Tephralagen im oberen Teil des Le tephra “Old Crow,”un peu inférieur à la couche Eva étudiée, Gold Hill Lösses wurden identifiziert, charakterisiert und a été daté par cette méthode de 140 ± 10 KA et étaye très fort miteinander verglichen und halfen bei der Rekonstruktion der l’interprétation interglaciaire de cette couche. Au début des Ereignisabfolge,die zu Erosion,Auftauen und Ablagerung der Eva années 1990,des techniques de thermoluminescence hautement Forest Bed führten. améliorées ont été utilisées pour dater le loess au-dessus et en- Die dendrochronologischen Studien und die C13, C12- dessous de la couche Eva. Elles ont indiqué que l’âge probable Isotopenverhältnisse im Holz der Eva Forest Bed deuten stark de cette couche est de 125000 ans et que la durée de formation de darauf hin,daß die Umweltbedingungen zumindest ähnlich den la couche elle-même a été probablement seulement de quelques heutigen waren. Einige pflanzliche Überreste und Laufkäfer milliers d’années (Sangamon:sous-stage isotopique5e). (Taxa) aus der Eva Forest-Zeit belegen die Existenz von Arten, Stratigraphiquement, la couche forestière Eva se trouve die heute soweit nördlich nicht mehr vorkommen. So lassen z.B. immédiatement au-dessus de la grande discordance qui existe fossile Fichtenteile vermuten, daß sich der boreale Wald bis entre le loess massif sous-jacent dénomme “Gold Hill” (pre- nördlich der Brooks Range in Alaska ausgedehnt hat. Diese Sangamon) et le loess noirâtre riche en glace de fentes de gel botanischen und physikalischen Daten belegen wärmere Kli- des formations de “Goldstream” (Wisconsin). Des études du maverhältnisse als im gegenwärtigen Interglazial,die mit einer loess gelé “Gold Hill” indiquent que l’intervalle chaud inter- mittleren jährlichen Lufttemperatur von über 0 °C (möglicher- glaciaire était caractérisé par un profond et rapide dégel du weise 1 oder 2°C oder mehr) das Auftauen des Eises und des Per- pergélisol et une érosion du loess accompagnée par des ravine- mafrostes erlaubten. Astronomische Schlußfolgerungen stützen ments et des glissements de blocs de loess gelés. Après ces ebenfalls diese Annahme, in dem während des letzten Inter- glissements très nombreux, la topographie devint douce et la glazials die Einstrahlungsanomalien im Juli auf 65°N Werte erre- forêt a couvert le territoire. L’inclinaison de couches de tephra ichten, die um 50% über den zu Beginn der Nacheiszeit (vor démontre la réalité des glissements rotationnels des blocs de 10000 Jahren) herrschenden Werte lagen. loess. Le dégel profond du pergélisol est aussi indiqué par la réduction du fer du loess,le transformant de fer ferrique en fer RÉSUMÉ ferreux et remplaçant la couleur jaune traditionnelle des loess en la couleur grisâtre du loess enterré “Gold Hill.”C’est parce La couche forestière interglaciaire Eva représente une anci- qu’il n’y a eu qu’un seul gel au Wisconsinien que la trace (la enne forêt boréale enterrée et gelée dans les hautes terres couleur verdâtre) d’un dégel profond au début de la période drainées par le Yukon et la Tanana au centre est de l’Alaska. Elle Sangamon a été conservée. La couche Eva formée après que le consiste dans des lentilles de tourbe,des branches,des racines et dégel,l’érosion et les activités de glissement de terrain se soient des troncs extrêmement bien préservés, ainsi que des souches terminées, est au-dessus de la discordance. Plus d’une demi- enracinées et non enracinées, principalement de sapins et de douzaine de couches distinctes de tephra ont été identifiées, bouleaux. Comme dans les forêts boréales actuelles, les arbres caractérisées et corrélées au sein de la partie supérieure du loess les plus grands et les plus communs de cette forêt fossile étaient “Gold Hill.” Elles ont aidé à la reconstruction de la séquence le sapin, et principalement le sapin blanc (picea glauca). Des d’événements conduisant à l’érosion, au dégel et à l’emplace- restes de troncs de bouleaux sont communs, principalement de ment de la couche forestière Eva. Betula papyrifera. Les restes de cette forêt ont été enterrés sous Les études dendrochronologiques des arbres et les pourcent- du loess et gelés, c’est pourquoi ils sont aussi bien préservés. ages isotopiques 13C/12C de bois de cette couche ainsi que des Aucun des restes de bois n’est minéralisé. De nombreux frag- comparaisons avec du bois de la forêt boréale actuelle,suggèrent ments sont noircis par le feu, suggérant que les feux de forêts fortement des conditions environnementales semblables à celles étaient largement répandus dans cette région pendant cet inter- que nous connaissons aujourd’hui. Quelques groupes de plantes glaciaire. En outre,pour la première fois,on peut démontrer ici et de coléoptères du sol de cette couche forestière représentent l’existence de coléoptères attaquant l’écorce du sapin (scolyti- des espèces qui s’étendaient alors plus au nord que maintenant. dae) en Alaska,pendant le dernier interglaciaire. Par exemple des macrofossiles de sapins suggèrent que la forêt De nombreux efforts pour déterminer l’âge de ce lit forestier boréale s’est étendue au nord de la chaîne de Brooks en Alaska. Eva ont été poursuivis pendant les 50 dernières années. Les Ces données botaniques et physiques indiquent un environ- méthodes utilisées ont couvert une large gamme,allant de l’in- nement plus chaud que le périglaciaire actuel avec une tempéra- terprétation systématique des événements sédimentologiques et ture moyenne annuelle plus chaude que 0°C,et peut-être même d’évidences de changements climatiques, jusqu’à l’emploi de atteignant +1 à +2°C,ce qui a permis à la glace de fondre et au géochronométres modernes. Plusieurs méthodes de datation ont pergélisol de dégeler depuis la surface vers la profondeur. Les porté des fruits dans les années 1990. Des datations par C14 déductions provenant des observations astronomiques parlent en effectuées par scintillation liquide indiquent que les bois trouvés faveur de ce interprétation, car pendant le dernier interglaciaire sont plus vieux que 70000 ans. Mais la plus grande avancée (sous-stage isotopique 5e), l’anomalie d’insolation de juillet, à résulte d’une méthode de détermination des traces de fission util- 65° de latitude nord, atteignait des valeurs presque 50% plus isée habituellement pour dater des éclats de verre volcaniques. élevées qu’il y a 10000 ans,au début de l’Holocène. 4 T. L. Péwé and Others Eva Interglaciation Forest Bed 5 INTRODUCTION The Eva Forest Bed lies in the upper part of the thick,perenni- ally frozen late Cenozoic loess deposits that occur throughout unglaciated east-central Alaska (Fig. 1). We show here that the forest bed provides clear evidence for a great, widespread cli- matic warming of 125,000 years ago, here formally named the Eva Interglaciation. In addition to the forest bed,evidence for this conclusion consists of prominent breaks in the sedimentary,cli- matic,floral,permafrost,and geomorphic records. Rapid thaw- ing of the permafrost occurred with formation of widespread thermokarst terrain accompanied by a major erosional interval with gullying of loess on hillsides and removal of loess in creek valley bottoms. The thawing and erosion were accompanied and/or followed by a return of the boreal forest,the taiga,after an earlier colder,treeless glacial time. This interglacial forest is now represented by the frozen buried Eva Forest Bed of well-preserved rooted stumps and prostrate logs. This prominant stratigraphic break in the geological record was recognized in central Alaska at Eva Creek near Fairbanks by Péwé on June 28,1949,and reported1 informally as the Eva Intergla- ciation:the latest interglaciation—the Sangamon (Oxygen Isotope Substage 5e). However, only now can this interglacial event recorded in the loess of the subarctic be more quantitatively demonstrated to be equivalent to the last interglaciation. Westgate and his associates (Westgate,1988,Westgate et al.,1990) charac- terized and dated tephra in the loess and determined the paleo- magnetism. In addition,Berger dated the loess near the forest bed by TL (Berger and Péwé,1994), and Long and other scientists dated wood from the Eva Forest Bed by specialized radiocarbon dating methods (Long and Kalin,1992a,1992b and McCormac et al.,1993). Also,scientists at the University of Arizona Laboratory of Tree Ring Research have recently demonstrated by dendro- chronological studies (Brown) and by 13C/12C isotopic-ratio stud- ies (Leavitt) the similarity of wood from the Eva Forest Bed to the wood from modern trees. The data summarized here include results from examination of hundreds of loess exposures exhibited when gold mining exca- vations were made and the loess faces were still frozen. These observations are combined with old and new mechanical,miner- alogical,and chemical analyses of the loess along with data on permafrost,and with the new information mentioned in the pre- ceding paragraph. Therefore,we present here in a historical man- ner,with the latest quantitative results,the study of the Eva Forest Bed over the past 50 years. We take this approach because of (1)the long-developing knowledge of both the age and paleoen- vironmental character of the Eva Forest Bed; (2)the importance of all continental sedimentological records of the last interglacia- tion; (3) the early creation and subsequent loss by placer gold 1Péwé,1952a,1957,1958,1965b,1965c,1975a,1975b,1989,1992,in press; Péwé and Hopkins,1967; Péwé and Sellmann,1973; Péwé and Reger, 1983,1989; Péwé et al.,1989; Berger and Péwé,1994; Berger et al.,1996; Matthews,1970; Wintle and Westgate,1986; Westgate,1988. 6 T. L. Péwé and Others central Alaska as the climate warmed from the earlier treeless MammothSteppeenvironment(Guthrie,1990).Theforestexisted for perhaps a few thousand years and then disappeared with increased loess deposition and the onset of the colder,drier gla- cial climate of Wisconsin time. Remnants of the perennially frozen forest bed are today sandwiched between the overlying loess of Wisconsin age and the underlying older loess. ACKNOWLEDGMENTS In collecting the field data and assembling this manuscript, Troy L. Péwé has drawn heavily on his experience with col- leagues of the Branch of Alaskan Geology,U.S. Geological Sur- vey and the Departments of Geology at the University of Alaska Fairbanks and at Arizona State University over the past 50 years. In 1987 Malcolm K. Hughes,Director,Laboratory of Tree-Ring Research, University of Arizona,Tucson, initiated informative discussions with personnel of the laboratory concerning study of the Alaskan wood. In the 1940s and 1950s Otto Wm Geist of the University of Alaska introduced Péwé and his wife and field associate,Mary Jean Péwé,to the great number of “mining cuts,”large and small, and to the more than a hundred industrious gold miners from one end of the Yukon-Tanana Upland to the other,as well as to those in the Dawson area in adjacent Yukon Territory,Canada. These Figure 1. Extent of late Cenozoic glaciations (shaded) in Alaska and northwestern Canada and location of Yukon-Tanana Upland. From cooperative miners generously provided access to mining expo- Péwé (1975a) and Tarnocai and Schweger (1991). CB indicates location sures and frozen ground data to the Péwé’s and their associates of Ch’ijee’s Bluff in northern Yukon Territory. throughout the last half of the twentieth century. Especially coop- erative were Roy Earling,John E. Metcalfe,and James D. Craw- ford,administrators of the U.S. Smelting,Refining,and Mining mining of the most dramatic exposures of the Eva Forest Bed, Company; Pete Eagan,Manager,Fairbanks Department,Alaska and the consequent unique documentation of descriptions of Gold Company; and Walter and Mike Wigger,owners of proper- these exposures in the decades-long field notes of Péwé; and ties in the Eva Creek area. (4) the consequent importance for posterity of this otherwise In the field Péwé was aided by many assistants of the U.S. inaccessible information and of this historical development of Geological Survey and University of Alaska,notably E.S. King, knowledge of this major Alaskan feature. This summary,honed Jr.,A. M. Gooding, D. R. Loftus, George Herman, E.W. Mar- to mesh with modern and conclusive quantitative data,will prob- shall,D.D. Smith,R.A. Paige,L.R. Mayo,R.D. Reger,P.V. ably become increasingly valuable to Quaternary scientists (and Sellmann,N.W. Rutter,J.M. Blackwell,N.R. Rivard,J.W. Bell, present and future students of this region and topic) as the per- and E.J. Bell. The following provided valuable field assistance to mafrost continues to thaw, and as the sedimentological record J.A. Westgate:S.J. Preece,B. Stemper,Qiang Hu,and A. West- thereby continues to fragment and disperse spatially. We hope gate. Berger thanks his co-authors and S. J. Preece for field that our summary is both timely and useful to future Quaternar- assistance. ists who wish to reconstruct the geological record of the last Special thanks go to former Alaska State Senators Bettye interglaciation in central Alaska. Fahrenkamp and Shirley Craft for obtaining financial support in These long-term studies reveal that a unique set of physical 1989 to create the Gold Hill Heritage Loess Preserve at Gold Hill events took place during this prominent interglaciation in a region near Fairbanks. Loess stratigraphy at this preserve was studied characterized by a perennially frozen ice-rich blanket of easily for this report. From the early 1950s,E.H. Beistline,former dean erodible loess overlying a hilly bedrock topography of the Yukon- of the School of Mines at the University of Alaska, provided Tanana Upland (Fig.2) (Péwé,in press). Extensive piping,gully- information on mining activities and personnel. G.E. Weller and ing,and block slumping of frozen loess took place over a large G.D. Guthrie,also of the University of Alaska,kindly provided region in the warmer-than-present erosional episode,and most per- climatic and vertebrate paleontological information respectively. mafrost disappeared. F.R. Weber and J.D. Townshend of the U.S. Geological Survey A boreal forest of mainly spruce, birch, and poplar species provided both Péwé and Westgate logistic aid in the field over developed over much of the landscape in the lower elevations of several summers and James and Sally Murphy of Fairbanks pro-