• Save
Amazonia during the last glacial
Upcoming SlideShare
Loading in...5

Amazonia during the last glacial






Total Views
Slideshare-icon Views on SlideShare
Embed Views



0 Embeds 0

No embeds



Upload Details

Uploaded via as Adobe PDF

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
Post Comment
Edit your comment

    Amazonia during the last glacial Amazonia during the last glacial Document Transcript

    • PAIAF/Ittc0t0G~ELSEVIER Palaeogeography,Palaeoclimatology,Palaeoecology109(1994)247-261Amazonia during the last glacialThomas van der Hammen a, Maria Lucia Absy ba Projecto Tropenbos, Corporation Araracuara, Calle 20 No 5-44 Apartado Aereo 034 174, Bogotd D.E., Colombia~Hugode Vries-laboratory, Department of Palynology and Paleo/Actuo-Ecology, University of Amsterdam, Kruislaan 318,1098 SM Amsterdam, The NetherlandsbCaixa Postal 478, 69000 Manaus, Brazil(Received June 21, 1993;revised and accepted July 26, 1993)AbstractNew data on the vegetational history and Quaternary geology of Amazonia permit an improved reconstruction ofpast environments in Amazonia during the last glacial period. Although limited, data from Rondonia, Carajas andGuyana show that, in certain areas, savanna-type vegetation and savanna forest had replaced the rain forest duringthe late Pleniglacial (ca. 22,000-13,000 yr B.P.). The Amazonian forest may have been split up into one major westAmazonian and several other medium-size forest areas. This suggests a decline in rainfall of 500 to 1000mm (areduction of 25 to 40%). Temperatures may also have been 2° to 6°C (4 + 2°C) lower than today, possibly substanciallyinfluencing Amazonian vegetation. During the humid middle Pleniglacial (55,000-26,000 yr B.P.), rivers carried a lotof water and sediment, resulting in the deposition of lower terrace sediments with one dry interruption around 40,000yr B.P. (Carajas and Katira). In Carajas, there is evidence of dry periods ocurring at about 40,000 yr B.P., and duringthe early Pleniglacial (ca. 60,000 yr B.P.). Rivers carried little water and incised into the low terrace sediments duringthe dry late Pleniglacial. Water levels rose during the late glacial (13,000-10,000 yr B.P.) or at the beginning of theHolocene (10,000 yr B.P.). Sedimentation in (and of) the present inundation valleys commenced after that.1. IntroductionDiscussion on the possible replacement of (rain)forest by savanna vegetation in Amazonia, thatmay have resulted in forest "refugia" during glacialtimes, as originally proposed by biogeographersand geomorfologists (Haffer, 1969; Vanzolini andWilliams, 1970; Vanzolini, 1973; Prance, 1973, etc.;see also Prance, 1982) is still going on. The answerto what really happened will depend very muchon concrete historical vegetational data (especiallypalynological data).There is extensive evidence on Late Pleistoceneclimatic changes to the west, north and south ofAmazonia.In the Andes (see overviews in e.g. Van der0031-0182/94/$7.00© 1994ElsevierScienceB.V.All rightsreservedSSDI 0031-0182(94)E0175-SHammen, 1974, 1981, 1982, 1986a,b, 1988, 1991;Schubert, 1988; Markgraf, 1989), lake levels aregenerally high during the middle Pleniglacialinterval (ca. 60,000 to ca. 26,000 yr B.P.), whileglacier extension was maximal. The climate wasmoderately cold and relatively wet, although thereare a few short colder and drier intervals duringthis period. Considerable amounts of fluvioglacialand fluvial gravel were deposited: e.g. in the highplains of the Eastern Cordillera of Colombia andin the Venezuelan Cordillera. During the latePleniglacial (ca. 26,000-21,000 to 14,000-13,000yr B.P.), lake levels became very low, indicatingprecipitation values much lower than today; andglaciers started to retreat from their maximumextent. The climate was very cold; annual temper-
    • 248 T. van der Hammen, M.L. Absy/Palaeogeography, Palaeoclimatology, Palaeoecology 109 (1994) 247-261atures were 8°C lower than today. During the lateglacial (14,000-10,000 yr B.P.), temperaturesbegan to rise and lake levels started to increase,indicating an increase in rainfall. During theHolocene (ca. 10,000 yr B.P.-present), currentconditions were established but with minoralthough marked fluctuations in rainfall. MiddleHolocene annual temperatures were somewhathigher than today.Evidence from the lower elevations of the Andes(1000-1500 m altitude; Monsalve, 1985; Bakker,1990) suggests a lowering of vegetation zones,implying that (late) Pleniglacial temperatures pos-sibly were 6°C lower during the (late) Pleniglacialthan nowadays. As the lowering of temperatureamounted to ca. 8°C at 2500 m, the temperaturegradient may have been steeper than at present(probably related to generally drier air). Takingthis into account, and an estimated 4°C drop intemperature during the late middle Pleniglacial atca. 1100 m on the Amazonian Andean slopes ofEcuador (Liu and Colinvaux, 1985; Colinvaux andLiu, 1987), an late Pleniglacial lowering of temper-ature in the South American tropical lowlandsbetween 2° and 6°C (4_2°C) seems probable.While this may not have been a problem forAmazonian species, more montane elements couldhave have invaded the area (as suggested by somewest Amazonian pollen data now available;Espejo, unpublished data). It is interesting to notethat CLIMAP (1976) had already predicted alowering of temperature of at least that much forAmazonia, based on a computer model that usedthe glacial sea-surface temperatures as a majorinput (Caribbean glacial temperatures were only1° or 2°C lower than today; CLIMAP, 1976;Gates, 1976).During glacial-time low sea-level periods, north-ern Guyana (Georgetown) and Surinam were cov,ered with dry grass savanna (Van der Hammen,1963; Wijmstra, 1969, 1971) and Late Pleistocenelow levels of Lake Valencia indicate a considerablydrier climate in northern Venezuela (Salgado-Labouriau, 1986). On the present-day savannas ofthe Colombian and Venezuelan Llanos, large areaswere covered with sanddunes during the LatePleistocene, being more like a desert than savanna,and indicating a climate considerably drier thanat present (Roa Morales, 1979). On the otherhand, lowland Panamfi continued, at least partly,forested (Bush and Colinvaux, 1990), as did thelower eastern slopes of the Andes in Ecuador(Bush et al., 1990).South of Amazonia, there is important newinformation from Salitre (19°S, Brazil; Ledru,1992), showing wet climatic conditions during themiddle Pleniglacial and late glacial, and dry condi-tion during the late Pleniglacial. Southwest ofAmazonia, there is information on a drier latePleniglacial climate from Lake Titicaca (Ybert,1988). All this shows clearly that tropical SouthAmerica, at least north, west and south of theAmazon Basin, had a relatively dry (and colder orcooler) climate during the world maximum of thelast glacial period (ca. 21,000-14,000 yr B.P.).Evidence of a former drier Amazonian climate hasbeen presented by geomorphologists, but has beendebated by others (see e.g. Prance, 1982). It istherefore most important to obtain informationon the history of vegetation from Amazoniaproper, and to consider and interprete it indepen-dent of the forest refugia theory that was initiallybased on biogeographical data.Therefore, we will discuss next the palynologicaland Quaternary geological information thatrecently became available, and present some cru-cial new information. Finally, we will see to whatconclusions all this may lead.2. Recent data from Amazonia2.1. Rondonia (Brazil)In this section, palynological and sedimentologi-cal information (Van der Hammen, 1972; Absyand Van der Hammen, 1976) and additionalrecently obtained •13C data and AMS dates fromKatira (Rondonia; approximately 9°S, 63°W) willbe reviewed. The palynological data (Fig. 1) showreplacement of Amazonian wet forest by grasssavanna. At the time of publication no dateswere available, as sample size and carbon contentwere insufficient for dating with conventionalradiocarbon methods; only a young Cenozoic age(and supposedly a Late Pleistocene age could beestablished. However, using accelerator mass spec-trometry, 14C dating of very small samples is now
    • T van der Hammen, M.L. Absy/Palaeogeography, Palaeoclimatology, Palaeoecology 109 (1994) 247-261 249possible and has permitted to date two levels ofthe original Katira section and pollen diagram(samples C and E; Figs. 1 and 2). Dating wascarried out at the Van der Graaff Laboratory ofthe University of Utrecht (The Netherlands), withthe following results.Sample Katira C, depth ca. 7 m. Black clayey material;pollen content indicatinggrass-savannavegetation.Two frac-tions of the organicmaterialweredated:Carbon Analysedfraction UtC-Nr. 613C Age(wt%) (%0) (yr B.P.)44,000_20007 org. C. residue 1852 -16.2 +30oo50 alk. sol.fraction 1853 -15.6 18,500_+150Sample Katira E, depthca. 13m. Dark greyclay;pollencontentindicatinghumid forest.Two fractionsof the organicmaterialweredated:Carbon Analysedfraction UtC-Nr. 613C Age(wt%) (%,0 (yr B.P.)38 org. C. residue 1854 -28.6 49,000+s°oo3ooo38 alk. sol. fraction 1855 -29.4 41,300+17°°_14ooThe samples are from one of the boreholes madeby Ferusa in Katira Creek (an affluent of the RioPreto do Jamari), about 120 km southeast of PortoVelho (Figs. 1 and 2). Potential vegetation in thispart of the Amazon Basin is a dense tropical rainforest. The nearest patches of more open naturalvegetation are found 150-200 km to the south, onhigher elevation tablelands with sandstone sub-strata. Several boreholes were made in a transectcrossing the valley. Below the sedimentary valleyfill, at a depth of approximately 25 m, a hardlateritic layer was found. The sediment from thebase upward is represented by an older valley fillof brown coloured clay, passing at 15 m depthinto dark grey clays with plant remains, at ca. 11m to grey silty clays which are topped by a blacksilty clayey layer between ca. 5.5 and 8 m in oursection. In the more central part of the originalvalley, this dark coloured layer is thicker and liesup to 8 m deeper in the section. At this place, thebase should represent an erosion level (see Fig. 2).On top of the black layer, there are yellow tored coloured silty clays locally containing oldersoil fragments. These sediments represent colluvialdeposits that filled the valley and covered the olderdeposits (Fig. 2). The uppermost 2 m are a yellowsandy clay. Thus, there seems to be an oldersequence of brown clay topped by dark grey clayswith plant remains and grey silty clays that waspartly eroded at a later phase. Then, a youngersequence of sedimentation started with black siltyclay, covered by yellowish to red colluvium. Whenthe upper grey to dark grey part of the lowersequence was deposited, there was first a densemarshy (rain) forest vegetation in the valley (thedark grey part with plant remains), followed bygrass savannah (the lighter grey upper part). Theorganic material of the dark grey layer, withabundant wet (rain) forest pollen, has 613C valuesof -28 to -29%0 in both the organic residue andthe alkaline soluble fraction, indicating that it wasproduced by C3 plants in full agreement with theinterpretation of the pollen spectrum, representedby arboreal taxa. The organic-rich layer probablyformed as wet soil under forest. Stable organicremains that do not (or only slowly) decomposeare known to accumulate in soils mixed withminerogenic materials; the more soluble part ofthe humus, however, easily disappears in time.When such a soil is "fossilized" (covered withsediment), the relatively old part of stable humusthat accumulated over long periods of time ispresent together with the more soluble fractionfrom the later phase of soil development. In ourcase with sample Katira E, organic residue showedan finite age of 49,000 yr B.P. and, the alkalinesoluble fraction about 41,000 yr B.P., both with arather large standard deviation. This may be inter-preted as a marshy forest soil that most likelyformed between roughly >50,000 and 41,000 yrB.P. (with the end somewhere between 43,000 and40,000 yr B.P.). The light grey clay that wasdeposited on top of the darker layer, apparentlywithout a hiatus or erosional phase in between,according to the pollen spectrum, was depositedin a grass savanna, practically devoid of trees. Itsage might be estimated as somewhere between43,000 and 40,000 yr B.P., assuming that theorganic residue of sample Katira C was formed inplace, which seems quite probable.The savanna phase is followed by an erosionalphase in the central part of the valley (some 8 m
    • 250 T. van"derHammen, M.L. Absy/Palaeogeography,Palaeoclimatology,Palaeoecology109 (1994) 247-261of incision). At this higher part in the valley, thebasal part of the black silty sandy clay layer(sample C), a black soil, contains abundantorganic residue dated between 47,000 and 42,000yr B.P., while the alkaline soluble fraction showed18,500 yr B.P. Both fractions have a ~13Cvalue of- 15 to - 16%0indicating a substantial componentof C4 plants such as tropical grasses. This is inagreement with the pollen spectrum, which indi-cates grass-savanna vegetation with very few trees.The black layer apparently represents a savannasoil that was formed in the entire valley. In viewof the age range of organic residue, it should havebeen deposited during the same savanna period asthe underlying light grey clay; that is, shortly afterthe end of the wet forest period. The dates of thealkali soluble fraction of Katira E and the organicresidue of Katira C are almost the same, and havean overlap of only 1000 yr (43,000-42,000 yrB.P.). Perhaps part of the organic residue wasredeposited via erosion of layer E higher up in thevalley. In the light of the 613C values, this isunlikely because they indicate provenance fromsavanna vegetation. Whatever the case may be,data indicate the occurrence of a pronouncedsavanna phase at about 42,500(+2500) yr B.P.The alkaline soluble fraction of Katira C showedan age of 18,500(+ 150) yr B.P. According to the~13C value, savanna vegetation continued to pre-vail, and the top of the black layer must have beenunder savanna vegetation at that time. The blacklayer is covered by a yellow to red colouredcolluvium with fragments of older eroded soils. Itmust have been deposited under extremely drycircumstances. This is also evidenced by the pollencontent, which represents a grass savanna practi-cally without trees. It seems probable that thiscolluvium and the savanna are from approximatelythis date or shortly thereafter. We have no pollendata on what may have happened between ca.40,000 and 20,000 yr B.P. Perhaps the older soilwas partly eroded before being covered again bysavanna vegetation (see Fig. 2). In the deepercentral part of the valley, where black savanna soillies some 8 m deeper, a 3.5 m thick layer of greyclay with plant remains is intercalated between theblack soil and the overlying 15 m thick yellow redcolluvium that fills the valley at that site. Thisintercalated layer is similar to the grey clay withplant remains from our section that correspondsto the earlier wet (rain) forest period. It thereforeseems possible that it represents a wet (rain) forestinterval after the 42,000(___2500) yr B.P. and beforethe 18,500 yr B.P. savanna periods.In summary, the savanna period(s) of Rondonia,according pollen data, ~13C data and colluvialsediments, are of the Late Pleistocene, of the lastglacial age. The dates are from 42,500(+2500) to18,500 yr B.P. The savanna period (s) are preceededby a wet (rain) forest period. From the geologicalcontext, it seems probable that the dates corre-spond to two savanna periods, separated by awetter forest period. The more recent savannaperiod, at and after 18,500 yr B.P., seems to havehad the more extremely dry climate, judging fromthe thick colluvia deposited in the valley. We haveno data from this site on when the rain forestinvaded the area again but, on the basis of datafrom other sites in Rondonia (Absy and Van derHammen, 1976) and from Carajas (Absy et al.,1991), this probably happened at the beginning ofthe late glacial or Holocene.2.2. Carajas (Brazil)Recently, paleoenvironmental vegetational datahave become available from the Carajas area(southeastern Amazonia, Brazil). They include awell-dated pollen record covering most of the lastca. 60,000 yr, representing the first longAmazonian palynological record (Absy et al.,1991). The diagram is from one of a number oflakes on top of table mountains, at 700-800 melevation, completely surrounded by Amazonianforest. On the plateau surfaces, patches of forestalternate with different types of more or lessedaphic savanna vegetation. The area is located atthe southeastern end of the drier corridor thatcrosses Amazonia from northwest to southeast,with annual rainfall values between 1500 and2000 mm. The pollen diagram shows changes inthe surrounding forest and in the vegetation ontop of the hills. Wet intervals, when forest domi-nated (both on the hills and in the surroundinglowlands) and lake levels were high, alternatedwith intervals of extension of savanna vegetation
    • T. van der Hammen, M.L. Absy/Palaeogeography, Palaeoclimatology, Palaeoecology 109 (1994) 247-261KATIRARondonio(Brosd)oO,Kotiro CUtr. 1853-olk. sol. fraction |18.500:1:150 years B.P.Utr, 1852-org. C. residue44.000 + soooyearsB.P.- 2000Del-t3C. rail resp.15.6and 16,27 ~8-9-10-Kotiro E 12-Utr. 18,55-olk. sol. fraction II41.300 ÷,roo I- ~4ooyears B.P ~3-Utr 1854-org. C.residue m.49.000 +~ years B-P.Oel-43C. mil resp.-29.4 and 28.6~ Yellow- grey sandy cloyYellow-brown and red cloyi Block and dark grey clay~z:~ Grey cloy"7"-77";2-. ----2~----i3 _~-4- 7.--~.Nir11-@<~o % 50% ~00%~oo% ~% 0% ~A m" 11E ~ _ _ . . , J l I 121~I/i//F~I~ Forestelements(trees)z ~ Gromineoe (grosses)3 ~ Compositoe+ Cupheo 4- others herbsScale for BI~curve0 IC 20Fig. 1. Section, simplified pollen diagram, radiocarbon dates and 813Cvalues from Katira Creek, Rondonia (Brazil). For the compollen diagram, see Absy and Van der Hammen (1976).
    • T. van der Hammen, M.L. Absy/Palaeogeography, Palaeoclimatology, Palaeoecology 109 (1994)247-261 255Vert. scole(m)O-"-A_ .7=:7--""~=~~- m. . . . . ,~, , =..=.= 0s- ~ - - , A-z-- -- -- - -- - ~ Yellow sandy-clayey sediment(colluvium)-A1o- -A. -- -- -El 5 I~ Red-yellow to yellow-red sondy.cla~y colluvium~D ~ Froqments of older eroded soil in colluviumt5- . .Io m Block silty sandy cloy soilGrey sandy clayey sediment:® ~ (dark) grey cloy withplant- remains20- .15 ~ Brown sandy clayey sediment. . . . . . . .~--- ---"~-~=-:- . . . . . . . . . . . . . . . . . . . ~ Hard laterite~---.--_--._--_-_-_-_. . . . . . . . . . . . . . . . . . ] ~ Probable erosion level and/or hiatus25, G--_--_---_---_--~_----~ _- _- ZZZ- - ZL-~._c_-_-_ _ --_ ~zo. . . . . ]~......_ _ _ -.~,.~30 [25Fig. 2. Reconstructionof partial section throughthe Katira Creekvalley. Based on threedrilling logs by Feusa. Location of thestudysection is indicated.(both on the hills and in the surrounding forest and interesting information from last glacial times,area) and low lake levels. Periods of drier climate from the Andean eastern slopes towards Amazoniaand extension of savanna were dated approxi- (Amazonian headwater areas), at an altitude ofmately between 60,000 and 40,000 yr B.P., and ca. 1100 m. River sediments and peaty intercal-between 22,000 and 11,000 yr B.P. During this ations were dated at 33,500 and 26,500 yr B.P.later period of savanna extension, climatic condi- (late middle Pleniglacial, a cold and wet period intions must have been extremely dry. The lake must the northern Andes). Palynological data indicatehave dried up almost completely. This interval a possible lowering of vegetation zones of montanecorresponds to the late Pleniglacial and to the forest by some 700 m, which might be interpretedinterval of very (cold and) dry climate west (the as a lowering of average annual temperature byAndes), north and south of Amazonia. The other approximately 4°C.two intervals seem to correspond well with coldphases known from the late and middle Pleniglacialin the Andes (see above). 2.4. The Middle Caquet(t area (Colombia) andeastern Peru2.3. Mera (Ecuador)Recently, a field and laboratory study of theLiu and Colinvaux (1985), Colinvaux and Liu Late Quaternary of the Middle Caquetfi River area(1987) and Bush et al. (1990) published some new in Colombian Amazonia revealed interesting new
    • 256 T. van der Hammen, M.L. Absy/Palaeogeography, Palaeoclimatology, Palaeoecology 109 (1994) 247-261data on the history of the west Amazonian region(Van der Hammen et al., 1992a,b). Sediments onthe present river plain were 14C dated between ca.13,000 yr B.P. and the present (30 dates). The lowterrace, the lower part of which consists of sandand coarse gravel of Cordilleran origin, was datedby seven finite radiocarbon dates between 55,000and 30,000 yr B.P. (and six infinite dates of>40,000 and >53,000 yr B.P.). There are still 5m of sediment younger than 30,000 yr B.P. Takinginto account the sedimentation rates calculated forthe Holocene, this might very well represent about4000 yr. This suggests that the lower terrace of theCaquet~i River was deposited between (>) 55,000and 30,000(-26,000) yr B.P., i.e. of the middlePleniglacial age. In view of this age, the coarsesand and gravel deposits, the fact that the headwa-ters of the Caquet~i River are in the high easternAndes, and the fact that the extensive fluvioglacial-fluvial deposits and terraces in the eastern Andesare also of middle Pleniglacial age, it seems thatformation of the lower terrace is related to arelatively colder and wetter climate in the Andesthat probably was also cooler and wetter in westernAmazonia. A few dates from Peruvian Amazonia,seem to indicate that a middle Pleniglacial lowterrace is present over a wide area in westernAmazonia (Rasanen et al., 1982; Campbell andRomero Pittman, 1989). The presence of sedimentin the headwaters on the east Andean slopes ofEcuador, equally of middle Pleniglacial age, wasalready mentioned (Liu and Colinvaux, 1985, etc.).Incision of the lower terrace should have takenplace between 30,000(-26,000) and 13,000 yr B.P.,during the late Pleniglacial which was a relativelycold and dry period. At around 13,000 yr B.P.sedimentation started in the newly formed valley.With the data now available it is impossible toconclude whether the incision took place at thebeginning, during the entire time or at the end ofthe late Pleniglacial. In the Middle and LowerValley of the Solimoes-Amazonas River, strongincision must have ocurred during the same period,but was clearly related to the low "glacial" sea level(at least 80 m lower than at present; Irion, 1976).From Colombian western Amazonia weobtained pollen data, partly published (Van derHammen et al., 1992a,b) and partly unpublished,showing predominantly Amazonian forest ele-ments in river-valley deposits of the middlePleniglacial low terrace near the junction of theCaquet~i and Cahuinari rivers; however, there isan interval (possibly ca. 40,000 yr B.P.) of somelocal extension of savanna-type (Caatinga) vegeta-tion (with grasses, Byrsonimaand Dydimopanax),possibly on the edaphically drier sites. Apparently,Amazonian rain forest dominated in westernAmazonia during much of the middle Pleniglacial,but during drier intervals of that period, locallysavanna-caatinga-type vegetation could developor extend somewhat locally. At the base of asequence of marsh sediment on top of the lowerterrace of the Caquet~i River (northwest ofMarifiame Island), a pollen diagram shows vegeta-tion dominated by Ilex and accompanied by somepollen of more montane affinity. It might representa late Upper Pleniglacial or early late glacialcaatinga-type vegetation. In another section of thesame area, a similar vegetation type dominated byIlex was found, and radiocarbon dated to 30,000yr B.P. (Espejo, unpublished data).3. Discussion and conclusionsThe data that have been presented and discussedpoint to the fact that most or all of warm andcold tropical South America had a cooler or colderclimate and relatively high values for rainfallduring the middle Pleniglacial (ca. 60,000-26,000yr B.P.) and a considerably cooler or colder cli-mate, becoming markedly drier between ca. 21,000and ca. 14,000 yr B.P., during the late Pleniglacial(ca. 26,000 to 14,000-13,000 yr B.P.). For LakeFuquene, in the northern Andes, the drop inrainfall during the dry late Pleniglacial, comparedto the present, was estimated as at least 50%, andprobably more (Van Geel and Van der Hammen,1973). In the same area, the maximum drop intemperature at an altitude of ca. 2500 m, wasestimated at ca. 8°C.The available data on maximum temperaturedrops at different elevations in the (tropical)Andes, seem to indicate a somewhat steeper lapserate during late Pleniglacial times (ca. 0.6°C/100m today, ca. 0.7°C/100 m during the late
    • 7". van der Hammen, M.L. Absy/Palaeogeography, Palaeoclimatology, Palaeoecology 109 (1994) 247-261 257Pleniglacial; Bakker, 1990). This difference maybe explained by the drier air of that time. If thesedata are right, the maximum lowering of temper-ature in the tropical lowland might have been 6°C.However, the sea surface temperatures in theCaribbean (today corresponding to the annualaverage temperature of the air), were only 1° or2°C lower than today during the late Pleniglacial,according to the CLIMAP reconstruction (1976).As we do not have direct temperature data for latePleniglacial Amazonia, it seems prudent to accept,for the time being, that the glacial-time loweringin that area may have been between 2° and 6°C(4 _+2°C). Values on that order were already gener-ated by the CLIMAP computer reconstruction ofthe Ice Age world (Gates, 1976).During the relatively cold and dry part of thelate Pleniglacial (eventually including part of thelate glacial), savanna vegetation extended in theCarajas area (6°S, at the southeastern end ofthe dry Amazonian NE-SW corridor with present-day rainfall between 1500 and 2000 mm) and inKatira, Rondonia (9°S, with a present-day rainfallof approximately 2000-2500 mm), replacing therain forest. Moreover, savanna replaced wet forestat Georgetown (and probably in the entire presentcoastal area of Guyana and Surinam) during (partof) the last glacial (Van der Hammen, 1963;Wijmstra, 1971).A comparision of todays vegetation and rainfallin northern South America shows that, to reachclimatic conditions permitting the natural exten-sion of savanna woodland with open spaces ofgrass savanna, annual rainfall below 1500mm(1000-1500 mm) seems necessary. To produceextensive natural grass savanna with very littleforest (as seems to have been the case in Rondoniaand Georgetown), a lowering to values between500 and 1000 mm seems necessary. "Accordingly, late Pleniglacial data from Carajasand Rondonia indicate a decline in rainfall (com-pared with the present) between 500 and 1000 mm(a decline of 25-40%).If we accept, as a first approximation, theseciphers of rainfall decline for Amazonia (respec-tively tropical South America as a whole) duringthe maximum of the last glacial (the dry latePleniglacial) and apply them to the rainfall mapof the area (Fig. 3) we obtain an interesting picture(Fig. 4). With 500mm (25%) decrease, the2000 mm curve becomes a 1500 mm curve, as theapproximate savanna-forest boundary, onlyexplaining the fact that Carajas is in the savannaarea (Fig. 4a). With the 1000 mm (40%) decrease,the 2500 mm Curvebecomes 1500 mm curve, whichwould be the approximate glacial-time boundaryof savanna-savanna woodland and forest, explain-ing that not only Carajas, but also Katira andGeorgetown are under savanna. In this way,extrapolation of the paleodata from Carajas,Katira and Georgetown leads to a picture of onemajor west Amazonian, and to several other medi-um-size forest "refugia".Data now available on lower and upperAmazonia also indicate that the river systemsuffered great changes during the last glacial (Vander Hammen et. at., 1992a,b). The rivers musthave carried a great deal of water and (partly verycoarse) sediment during much of the middlePleniglacial (related to relatively high rainfall inthe Andes and Amazonia, and maximum glacierextension in the Andes) and deposited the lowterrace sediments know at least from westernAmazonia. During the late Pleniglacial, water car-ried by the rivers must have been seriously reduced,while the rivers incised into their earlier sediments.This incision was reduced on the middle course ofthe rivers (on the order of 10 m), but was veryhigh on the lower course, especially on theAmazone River itself where incision may havebeen 100 m or more, directly related to glacial-time low sea level (Irion, 1976). This importanterosion phase during a dry climatic interval isapparently even registered in the sedimentaryrecord of Atlantic deep-sea sediments west andnorth of South America (Daimuth andFairbridge0 1970).During the late glacial (13,000-10,000 yr B.P.),the climate of northern South America becamewetter, a!though appar,ent.ly with marked localdifferences in degree and timing. The rivers carriedan increasing quantity of water and a new cycleof sedimentary deposition began and continued inthe Holocene. The increase in water level seems tohave been considerable, locally leading to temporalpermanent inundation of the (upper to) middle
    • 258 T. van der Hammen, M.L. Absy/Palaeogeography, Palaeoclimatology,Palaeoecology 109 (1994) 247-261so-ooo--. I IL. Ft~quenle / ~-~/ ,, : .:..""" " "? "1~"" :"--i.....: - .:~:,::f :. ,,,,... , , . . . " ..~i~/" Georgetown_8 o.4 o0 o12°i• .... ]i t"~..~. eL Titlcocoi~0 ° 70 o 60 °~traJasl ""8oIf,,. =6°El Solitre9!50 °I I Present rainfall ( 2000 m (---1500 mm)Present day rainfall 2000-2500 mmPresent day rainfall ) 2500mm (---3000 mm)Fig. 3. Rainfall map of Amazonia s.l. (northern South America, not including the andean and pacificareas). Basedon the rainfallmap of Reinke (1962), Nimer (1977), Simpson and HalTer (1978) andFigueroa and Nobre (1990).river valleys (Urrego, 1991; Van der Hammenet al., 1992,a,b). In the lower valley of the AmazonRiver, large-scale inundation of the glacial-timebroad eroded valley took place because of therapid post-glacial rise of the sea level, convertingit well inland into an inmense estuary or river lake.It seems to have taken thousands of years beforethis "lake" was filled with sediment and the pre-sent-day broad, seasonally inundated "varzeas"were formed (Absy, 1979).During the Holocene, there were rather markedrainfall changes resulting in periods of extensiveand frequent river inundations, alternating withdrier periods with reduced inundations (Absy,1979; Van der Hammen et al., 1992b). Thesechanges may have influenced considerably humanriverine settlements and migrations. However, theyare very small in comparision with the environmen-tal changes that took place during the last glacial.The conclusions on the vegetational history ofAmazonia based on extrapolation of currentlyavailable paleodata, are independent of any theory
    • oo.JcaX-./o.4oq,/.//}"//KatiraL,<"~oo/o°~o°~o,;o°,~°~oo~ooFig.4.(a)Areaof>1500ramrainfallandaproximate(rain)forestarea,incaseofageneral500mm(25%)reductionofrainfall.(Carajasinsavanna-savannawoodlandarea,aswasthecaseduringdryphasesofthelastglacial).BasedontherainfallmapofFig.3.(b)Areaof>1500mmrainfallandapproximate(rain)forestarea,incaseofageneral1000mm(40%)reductionofrainfall(Carajas,KatiraandGeorgetowninsavanna-savannawoodlandarea,aswasthecaseduringdryphase(s)ofthelastglacial.AraracuaraandMeraarestillthe(rain)forestarea.BasedontherainfallmapofFig.3.-ng~t.,a
    • 260 T. van der Hammen, M.L. Absy/Palaeogeography, Palaeoclimatology, Palaeoecology 109 (1994) 247-261to explain existing centers of species diversity orendemism. For an explanation of these centers,like those presented by Haffer (1969) and Prance(1973, 1982), the existence or non-existence offorest "refugia" is of course of considerable impor-tance, but other factors should also be taken intoaccount, like changes in temperature, present-dayclimatic-ecological diversity, broad rivers, etc.A glacial Amazonian type of situation may haverepeated itself several times during the last glacialperiod (e.g. ca. 20,000, ca. 40,000 and ca. 65,000yr ago), and one or several times during each ofthe many glacial periods of the Pleistocene [thelast 2.5 million years; see Hooghiemstra (1984)and Hooghiemstra and Ran (1994) for the Andesand Wijmstra (1971) for the Guianas].When discussing the high Amazonian alpha-and beta-diversity, however, we should also con-sider the 60 million year history of the Tertiary.In this respect it is interesting to note that thediversity of pollen types in Lower Miocene flood-plain sediment in the Caquet~i area (C. Hoorn,pers. commun., 1993) is about twice that inHolocene floodplain sediment of the same westAmazonian Caquet~i River area (L.E. Urrego,pers. commun., 1993), 275 and 140, respectively.The cipher for Holocene sediment, is very near theaverage number of species found on 0.1 ha plotsof present-day vegetation in the same area: 135(Duivenvoorden, in prep; Urrego, unpublisheddata). These data suggest the possibillity thatdiversity was considerably higher during theMiocene than during the Holocene and today, andhence, Plio-Pleistocene (glacial) extinction mayhave been a factor of importance in the explanationof present-day diversity. High Amazonian diversitymay be inherited mainly from the Tertiary. Whilespeciation may have occurred in the wet Plio-Pleistocene forest refugia, extinction due to thesevere impact of dry (and cooler) climatic episodesmay have dominated, and the now relatively highspecies diversity of western Amazonia may be theresult of a relatively low extinction rate attributedto the comparatively low impact of dry climaticepisodes in this large wet forest "refuge" area.There is little doubt that the Pleistocene glacial--interglacial climatic history of Amazonia musthave had a great impact on both flora and vegeta-tion and it gives us a highly dynamic picture ofAmazonia and its (rain) forests.ReferencesAbsy, M.L., 1979. A palynological study of Holocene sedimentsin the Amazon basin. Thesis, Univ. Amsterdam, 84 pp.Absy, M.L. and Van der Hammen, T., 1976. Some paleo-ecological data from Rondonia, southern part of the AmazonBasin. Acta Amazonica, 6(3): 293-299.Absy, M.L., Cleef, A.M., Fournier, M., Martin, L., Servant,M., Sfeddine, A., Ferreira da Silva, M.F., Soubies, F.,Suguio, K., Turcq, B. and Van der Hammen, T., 1991. Miseen 6vidence de quatre phases douverture de la forrt densedarts le sud-est de lAmazonie au cours des 60.000 dernirresannres. C. R. Acad. Sci., Paris, Ser. II, 312: 673-678.Bakker, J.. 1990. Tectonic and climatic controls on LateQuaternary sedimentary processes in a neotectonic intramon-tane basin (The Pitalito Basin, South Colombia). Thesis,Univ. Wageningen. [Also in: T. van der Hammen (Editor),The Quaternary of Colombia, 16. Amsterdam, 169 pp.]Bush, M.B. and Colinvaux, P.A., 1990. A long record ofclimatic and vegetation change in lowland Panama. Veget.Sci., 1: 105-118.Bush, M.B., Colinvaux, P.A., Wiemann, M.C., Piperno, D.R.and Liu, K., 1990. Late Pleistocene temperature depressionand vegetation change in Ecuadorian Amazonia. Quat. Res.,34: 330-345.Campbell, K.E. and Romero Pittman, L., 1989. The Quaternarygeology of Departamento de Madre de Dios, Peril. Proc.Cong. Peruano de Geologia, 6. (Lima.) Soc. Geol delPeril, Lima.CLIMAP, 1976. The surface of the Ice-Age Earth. Science,191: 1131-1137.Colinvaux, P.A., 1987. Amazon diversity in light of thepaleoecological record. Quat. Sci. Rev., 6:93-114.Colinvaux, P.A., 1991. A commentary on "Paleoecologicalbackground: Neotropics". Climatic Change, 19( 1/2): 49-51.Colinvaux, P.A. and Liu, K., 1987. The late Quaternaryclimate of the western Amazon basin. In: W. Berger and L.Laberey (Editors), Abrupt Climatic Change. pp. 113-122.Daimuth, J.E. and Fairbridge, R.W., 1970. Ecuatorial Atlanticdeep-sea arkosic sands and Ice-age aridity in tropical SouthAmerica. Bull. Geol. Soc. Am., 81: 189-206.Duivenvoorden, J.F., in prep. Rain forest plant diversity inthe Middle Caquet~i firea, Colombia, western Amazonia.Figueroa, N. and Nobre, C.A., 1990. Precipitation distributionover central and western tropical South America.Climanfilise, 5(6): 36-42.Gates, W.H., 1976. Modeling the Ice-Age climate. Science,191: 1138-1144.Haffer, J., 1969. Speciation in Amazonian forest birds. Science,165: 131-137.Hooghiemstra, H., 1984. Vegetational and climatic history ofthe high plain of Bogotfi, Colombia: A contiuous record ofthe last 3.5 million years. Dissertationes Botanicae, 78.Cramer, Vaduz, 368 pp. [Also in: T. van der Hammen(Editor), The Quaternary of Colombia, 10. Amsterdam.]
    • T. van der Hammen, M.L. Absy/Palaeogeography, Palaeoclimatology, Palaeoecology 109 (1994) 247-261 261Hooghiemstra, H. and Ran, E.T.H., 1994. Late Pliocene-Pleistocene high resolution pollen sequence of Colombia: anoverview of climatic change. Quat. Int., 20.Irion, G., 1976. Quaternary sediments of the Upper Amazonlowlands of Brazil. Biogeographica, 7: 163-167.Ledru, P.M., 1992. Modifications de la vegetation du Br6silCentral entre la derni6re epoque glaciare et 1 interglacialactuel. C.R. Acad. Sci. Paris, S6r. II, 314:117-123.Liu, K. and Colinvaux, P.A., 1985. Forest changes in theAmazon basin during the last glacial maximum. Nature,318: 556-557.Livingstone, D.A. and Van der Hammen, T., 1979.Palaeogeography and palaeoclimatology. In: Tropical ForestEcosystems: A State of Knowledge Report Prepared byUnesco/Unep/FAO/. Unesco, Paris, pp. 61-90.Markgraf, V., 1989. Palaeoclimates in Central and SouthAmerica since 18,000 BP based on pollen and lake-levelrecords. Quat. Sci. Rev., 8: 1-24.Monsalve, J.G., 1985. A pollen core from the HaciendaLusitania. Pro Calima, 4: 40-44.Nimer, E., 1977. Clima. In: Geografia do Brasil: Regifio Norte.IBGE, Rio de Janeiro, 1, pp. 39 58.Prance, G.T., 1973. Phytogeographic support for the theory ofPleistocene forest refuges in the Amazon Basin, based onevidence from distribution patterns in Caryocaraceae,Chrysobalanaceae, Dichapetalaceae and Lecythidaceae. ActaAmazonica, 3(3): 5 26.Prance, G.T. (Editor), 1982. Biological Diversification in theTropics. Columbia Univ. Press, New York, NY.Rasanen, M. C., Salo, J.S. and Kalliola, R.J., 1982. Fluvialperturbance in the western Amazon basin: regulation bylong-term sub-Andean tectonics. Science, 238: 1398-1401.Reinke, R., 1962. Das Klima Amazoniens. Thesis, TObingen,116 pp.Roa Morales, P., 1979. Estudio de los m6danos de los llanoscentrales de Venezuela. Evidencias de un clima des6rtico.Acta Biol. Venez., 10(1): 19 49.Salgado-Labouriau, M.L., 1986. A pollen diagram of thePleistocene-Holocene boundary of Lake Valencia, Venezuela.Rev. Palaeobot. Palynol., 30: 297-312.Salo, J., 1987. Pleistocene forest refuges in the Amazon:evaluation of the biosteratigraphical, lithostratigraphical andgeomorphological data. Ann. Zool. Fenn., 24:203-211.Schubert, C., 1988. Climatic changes during the last glacialmaximum in Northern South America and the Caribbean:A review. Interciencia, 13(3): 128-137.Servant, M., Fontes, T.C., Rieu, M. and Saliege, J.F., 1981.Phases climatiques arides holocenes dans le sud-ouest delAmazonie (Bolivie). C. R. Acad. Sci. Paris, S6r. II, 292:1295-1297.Simpson, B.B. and HalTer, J., 1978. Speciation patterns in theAmazonian forest biota. Ann. Rev. Ecol. Syst., 9:497 518.Urrego, L.E., 1991. Sucesi6n Holoc6nica de un bosque deMauritiaflexuosa L.f. en el vaUe del Rio Caquet~i (Amazoniacolombiana). Colomb. Amaz6nica 5(2): 99-118.Van der Hammen, T., 1963. A palynological study on theQuaternary of British Guiana. Leidse Geol. Meded., 29:125-180.Van der Hammen, T., 1972. Changes in vegetation and climatein the Amazon basin and surrounding areas during thePleistocene. Geol. Mijnbouw, 51(6): 641-643.Van der Hammen, T., 1974. The Pleistocene changes ofvegetation and climate in Tropical South America.J. Biogeogr., 1:3 26.Van der Hammen, T., 1981. Environmental changes in theNorthern Andes and the extinction of Mastodon. Geol.Mijnbouw, 60(3): 369-372.Van der Hammen, T., 1982. Palaeoecology of tropical SouthAmerica. In: G.T. Prance (Editor), Biological Diversificationin the Tropics. Colombia Univ. Press, New York, NY,pp. 60-66.Van der Hammen, T., 1986a. La Sabana de Bogotfi y su lagoen el Pleniglacial Medio. Caldasia, 15(71/75): 249-262.Van der Hammen, T., 1986b. The paleoecology of TropicalSouth America. Thirty years of research into the history ofvegetation and environment. Neth. Found. Adv. Trop. Res.Rep., 1982: 35-91.Van der Hammen, T., 1988. South America. In: B. Huntleyand T. Webb (Editors), Vegetation History. Kluwer,Dordrecht, pp. 307 337.Van der Hammen, T., 1991. Paleoecology of the neotropics:an overview of the state of affairs. Proc. Global Changes inSouth America during the Quaternary. Bol. 1G-USP, Publ.Esp., 8; 35 55.Van der Hammen, T., Duivenvoorden, J.F., Lips, J.M., Urrego,L.E. and Espejo, N., 1992a. The Late Quaternary of theMiddle Caquet~i River area (Colombian Amazonia). J. Quat.Sci., 7(1): 45-55.Van der Hammen, T., Urrego, L.E., Espejo, N.,Duivenvoorden, J.F. and Lips, J.M., 1992b. Late-glacial andHolocene sedimentation and fluctuations of river water levelin the Caquet/t-area (Colombian Amazonia). J. Quat. Sci.,7(2): 57-67.Van Geel, B. and Van der Hammen, T., 1973. UpperQuaternary vegetational and climatic sequence of theFuquene area (Eastern Cordillera, Colombia). Palaeogeogr.,Palaeoclimatol., Palaeoecol., 14: 9-92. [Also in: T. van derHammen (Editor), The Quaternary of Colombia, I.Amsterdam.]Vanzolini, P.E., 1973. Paleoclimates, relief and species multi-plication in Equatorial forest. In: B.J. Meggers et al.(Editors), Tropical Forest Ecosystems in Africa and SouthAmerica. Smithsonian Press, Washington, DC, p. 255.Vanzolini, P.E. and Williams, E.E., 1970, South Americananoles: the geographic differentiation and evolution of theAnolis chrysolepis species group (Sauria, Yguanidae). Arg.Zool. S. Paulo, 19: 1-240.Wijmstra, T.A., 1969. Palynology of the Alliance well. Geol.Mijnbouw, 48: 125-133.Wijmstra, T.A., 1971. The palynology of the Guiana coastalbasin. Thesis Univ. Amsterdam, 72 pp.Ybert, J.P., 1988. Apports de la palyologie fi la connaissancede lhistoire du lac Titicaca (Bolivie-Perou) an cours duQuaternaire r6cent. Inst. Fr. Pondichery, Trav. Sect. Sci.Tech., 25: 139-150.