UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl)

UvA-DARE (Digital Academic Repository)

Late Pliocene-Pleistocene high resolution pollen sequence of Colombia: an overview ofclimatic change

Hooghiemstra, H.; Ran, E.TH.H.

Published in:Quaternary International

Link to publication

Citation for published version (APA):Hooghiemstra, H., & Ran, E. TH. H. (1994). Late Pliocene-Pleistocene high resolution pollen sequence ofColombia: an overview of climatic change. Quaternary International, 21, 63-80.

General rightsIt is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s),other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulationsIf you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, statingyour reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Askthe Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam,The Netherlands. You will be contacted as soon as possible.

Download date: 18 Dec 2020

rerpmou

1 0 4 0 - 6 1 9 2 ( 9 3 ) ~ B

Qua:ernary l m e ~ , Vol. 21, pp. 63-80, 1994. Copyright O 1994 INQUA/Elseviet S¢ieme Ltd.

Printed in Great Britain. All tights reserved. 1040-6182/94 $26.00

L A T E P L I O C E N E - P L E I S T O C E N E H I G H R E S O L U T I O N P O L L E N S E Q U E N C E O F C O L O M B I A : A N O V E R V I E W O F C L I M A T I C C H A N G E

Henry Hooghiemstra and Eva T. H. Ran Hugo de Vries Laboratory, Department of Palynology and PaleolActuo-Ecology, University of Amsterdam, Kruislaan

318, 1098 SM Amsterdam, The Netherlands

Two long continental pollen records, FunzJ I (357 m) and Funza II (586 m) from the high plain of Bogott (Colombia) at 2550 m elevation, have been studied palynologically. Fusion track dates on zircon from intercJlat~l volcanic ashes provided a chronological framework and showed the pollen records, which conelate with high ~ to be continuous over the interval from ca. 3.2 Ma to ca. 27 ka. The Late Pfiocene-Pleistocene histot7 of the montane forests am/opm alpine pm'amo vegetation is documented with a temporal resolution of ca. 6-5 ka, and for the upper 1.1 Ma with a resolutim of ca. 1.2 kit. 'I'ne immigration of the northern hemisphere elements Alnus (at ca. 1 Ma) and Quercus (at c& 0.33 Ma), which travelled along the Panamanian land~dge, caused significant changes in the composition of the Andean montane forests. The mcceseive Pleistocene glac~iom fmced the Andean vegetation belts to an almost continuous altitudinal movement; the upper foreat line (at present at ¢g 3200 m) shifted between 1800 m (ghtchds) and ca. 3500 m (interglaci~), conesponding to a variation in temperature between ca. 5 and 15°C at 2550 m altitude. A provisional land-sea correlation (Funza pollen-ODP Site 6778~'O) is shown for the upper 1.2 Ma (Stages 3-35). l~equemcy analym of several time series showed timnifu:ant periods of the eccentricity (100 ka) and incession (23 and 19 ka) bands, showing orb/tal forcing and a slnmg change in climatic variability around 800 ka- At ca. 2.7 Ma, a significant cooling of ca. 5°C is documented, reflecting the classical terrestrial Pliocone-Pleistocene boundary, which correlates to the Reuverian-Ih-aefiglian boundary of the NW European stratigraphical climatic subdivision.

INTRODUCTION

Long continental records of climatic change are scarce but are of great importance in facih'tating the comperison of land-based and ocean-based climatic histories. In the Eastern Cordillera of Colombia, the high plain of Bogotl (ca. 25 x 40 km in size) represents the bottom of a former lake that occupied a subsiding intermontane basin (Fig. 1). After the final upheaval of the northern Andes between 5 and 3 Ma (van der Hammen et al., 1973; Helmens, 1990, this volume), the development of a basin environment in the area of the present high plain of Bogotl started some 3.5 Ma (Helmens, 1990). Subsidence of the floor of the tectonic basin was more or less in equilibrium with sediment accumulation during most of the time and resulted in a sequence of almost 600 m of mainly lake sediments with an early fluviatile influx. Pollen records have been retrieved from deep bore.holes in these se~!ments. During periods with low lake levels in the central part of the basin, sediment accumulation ceased in the outer parts. This caused the presence of hiatuses in pollen records from the outer parts of the basin (van der Hammen and Gonz~dez, 1960, 1964).

The objective of this paper is to present an overview of the continuous history of vegetation development and climatic change during the last ca. 3 Ma, based on several research papers which have recently been published or axe in press. Data are based on the deep boreholes Funza I (357 m) and Funza H (586 m) in the centre of the sedimentary basin, where sediments reach the greatest depth. The Funza II core, recovered in 1988, reached bedrock, indicating that the complete basin infill, representing Late Pliocene to latest Pleistocene, has been recovered.

Changes in the composition of the vegetation, which reflect climatic change, are documented by the pollen rain that is conserved in slowly accumulating lake sediments. Tropical mountains especially seem to be in a favourable position because climatic change results mainly in a vertical shift of vegetation belts over the mountain slopes (Fig. 2). The different vegetation belts stay in the vicinity of the lake and are registered continuously by their intercepted pollen. The sediments in the Bogot~ basin, at 2550 m elevation, accumulated at an altitude that ties halfway between the highest position of the upp~ forest line (during interglacial conditions ca. 3500 m) and the lowest position of the upper forest line (during glacial conditions ca. 1800 m), rendering the Bogota sediments a sensitive recorder of palaeocfimafic change.

SETTING OF PRESENT VEGETATION AND CLIMATE

The present-day altim0inal zonation of the vegetation in the Eastern Cordillera of Colombia (Fig. 2) is summarized in order to understand the changes documented by the pollen record. More complete accounts of the modern montane forest and paramo vegetation of the Colombian Andes are given in e.g. Cleef (1981), Cleef et aL (1983) and Cleef and Hooghiemstra (1984). The following vegetation zones can be recognized:

D tropical lowland rain forest from 0 to 1000 m: main taxa are Byrsonima, Iriartea and Mauritia

- - subandean forest belt (lower montane forest) from 1000 to 2300 m: main taxa are Acalypha, Alchornea and Cecropia

63

_('v4 H. Hooghiemstra and E.'F.H. Rail

~.f. ~.~ ~ C EcuaOor \ "~

=.,n o, [

o Guuca

Subachoque

e ~ 0

~ FUflZai :

N / g ~ T equendarna

/ Fails

0 1 5 10km . ~

3500-4000 m

U 3000-3500 m

r - - ' l 3ooo m

X~h rain of Bogotd (2550-2600 m)

FIG. 1. Map of Colombia and the area of the high plain of Bogot~ in the Eastern Cordillera, indicating the topography of the surrounding mountains. (Redrawn after Andriessen et al., 1993.)

- - Andean forest belt (upper montane forest) from 2300 to 3200-3500 m: main taxa are Podocarpus, Hedyosmum, Weinmannia, Quercus, Alnus, Vallea, Myrsine (formerly Rapanea), Symplocos, llex, Juglans, Miconia, Eugenia and Myrica

- - subparamo belt from 3200-3500 m to 3400-3600 m: main taxa are Ericaceae, Hypericum, Compositae and Polylepis-Acaena

- - grassparamo belt from 3400--3600 m to 4000-4200 m: main taxa are Gramineae, Valeriana, Caryophyllaceae, Gentianaecae, Plantago, Aragoa, Geranium, Ranunculus and Lycopodium (species with foveolate spores)

- - superpararno belt extending from 4000-4200 m upward: main taxa are Draba, mosses and blue algae

- - nival zone proper, practically devoid of vegetation, extending from 4500--4800 m upward.

The highest areas of the Eastern Cordillera in the Sierra Nevada de Cocuy, some 200 km north of Bogota, extending up to 5500 m, may be pennatmntly covered by snow.

During glacial periods, lower temperatures caused a depression of the Andean vegetation belts, and a lowering of the position of the upper forest line of ca. 1200-1500 m has been seen (van der Hammen, 1974). The modem upper forest

line in the area of Bogotfi at ca. 3200 m altitude corresponds with the ca. 9.5°C annual isotherm. Thus, temperature changes at the level of the high- plain of Bogotat (2550 m; present-day average annual temperature 13-14°C) can be calculated when changes of the altitudinal position of the upper forest line are estimated on the basis of the pollen record, using a lapse rate of 0.66°C per 100 m displacement of the upper forest line.

ABSOLUTE T I M E C O N T R O L OF THE S E D ~ T S OF T H E BOGOTA BASIN

A revised geochronological framework for the sequence of unconsolidated sediments present in the Bogotat area was published recently (Andriessen et al., 1993) to replace the original time frame of the Funza I pollen record (Hooghiemstra, 1984, 1989). This is based on 11 fission track datings on zircons that were obtained both from exposed ash layers and from a series of ashes from the Funza II core. The data 5.33 + 1.02 Ma, 3.67 + 0.50 Ma and 2.77 _+ 0.50 Ma (Andriessen et al., 1993) for sediments that are considered to have been deposited before, at the beginning

High Resolution Pollen Sequence of Colombia 65

5000

4000

3OOO

2000 A E v ~l i o o o

m 0

PRESENT 20,000-14,000 yr BP I

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.......... L~ue~..a~. . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

• h6gtl plain of Bogo~ , l~gaslana Rlver ,4v/xemplMlc vegelatlon ~ upper forest line

FIG. 2. Altitudinal distribution of vegetation belts in the Eastern Cordillera of Colombia at present and during the last glacial maximum. Vertical shifts of the vegetation belts are mainly related to changes in temperature.

(Redrawn after Van der Hammen, 1974.)

and shortly after the final major upheaval of the Eastern Cordillera provide absolute chronological control for the

older part of the sequence (6-2.5 Ma). Fission track data on

zircon from the Funza II core (Fig. 3 and listed below) provide geochronological contzol for the younger part of the

sequence (3-0 Ma) and these are coherent with the fission track dates of the older part of the sediment sequence:

67.7 m: 0.20 ± 0.12 Ma;

298-307 m: 1.02 ± 0.23 Ma; 317 m: 1.44 ± 0.33 Ma;

322 m: 1.01 ± 0.21 Ma; 506 m: 2.74 ± 0.63 Ma.

Nevertheless, three fission track dates on zircon, at 239 m

0 ' E

e~

100

Funza 1 and I1 Pollen Records

• . . z - - Z i r c o n f i s s i o n track ages

• Control points from the con'elation betw~m Funza I Arboreal Pollen % record and ODP Site 677 dlSO record

• Estimated ages of the Pollen Zone boundaries of Funza II

2OO

40{

500

' 0 ~ 5 " 1~0 Age (Ma)

l :s 21o 3'.0

FIG. 3. Time control of the sediments of the Funza I and Funza II cores from the high plain of Bogot~ (Colombia) based on fission track dates (after Andriessen et aL, 1993) on zircon from intercalated volcanic ash horizons of the Funza 11 co~. Open ~ indicate the conlrol points of the graphical correlation between Funza I Arboreal Pollen% record and ODP Site 677 8'~O record (Fig. 5). Solid circles indicate the estimated ages of the boundaries of the Funza II pollen zones (see Figs 11 and 12). For the upper part of the Funza I1 core, age estimations are based on a provisional correlation with the Specmap time scale (Imbrie et al., 1984). For the lower part of the Funza II core, age estimations are based on a provisional visual correlation with the ODP

Site 677 61~D record (Shackleton et aL, 1990).

66 H. Hooghiemstra and E.T.H. Ran

0 lOO f "

g- so

0 o

100 200 300 400 500 600 700 800 900 100o 1100 1200 1.]O0 1400 1500 . . . . . . . . . . . . , . . . . . . . . | I 0 0

J

A r b o r e a l p o l l e n I i pollen record) 80

i 70

i . 60

4O

30

20

10

' o" ' ' ' ' ' ' ' ' . . . . ' ' ' ' - o ' 1 o o . . . . . . . . o 1 0 200 300 400 500 600 700 8oo 900 1 1oo 12oo 13o0 14-oo ~5oo

TIME {KYR)

HG. 4. Graph of Funza I Adxa~ Polka lmaXmtages (A/ma included) v ~ time baaed on the revised time frame. Note the develepnamt tg a distinct 100 ka climatic cycle (claui~ IIlaeial-iatel11~al cycles) in the

last ca. 0.8 Ma of the Quaternary. (Al~er H ~ et al., 1993.)

(0.26 ¢ 0.18 Ma), 250 m (0.27 ± 0.I1 Ma) and 270-277 m (0.53 ± O. 15 Ma), are considered as too young (Fig. 3; see the discussion in Andriessen et al., 1993).

LAND-SEA COl tRRL~TION OF RF.£ORI}S OF CLIMATIC CHANGE

A challenging similarity between the climatic change, registered in Andean Colombian pollen records (Fig. 4), and the climate records in marine sediments is apparent. Graphical correlation of the Funza I Arboreal Pollen record with the 81q3 record of ODP Site 677 from the Eastern Pacific, by Shacklemn and Hooghiemstr~ is shown in Fig. 5. A set of 36 control points was established (Fig. 5; Hooghiemstra et aL, 1993). At present this 36-point model is considered the best estimate for a geochronological

framework for the upper part of the Funza sediment sequence. These control points, as well as the estimated ages of the pollen zone boundaries, based on provisional land-sea correlation, are graphed in Fig. 3. The intervals represented by Stages 3-25 show the best correlation between climatic oscillations.

The correlation procedure will be repeated when the temporal resolution of the Funza II core (at present ca. 5-6 ka; already for several intervals ca. 1 ka) is as high as that of the Fonza I core (ca. 1 ka).

VARIABILITY OF QUATERNARY CLIMATIC CHANGE AND ORBITAL FORCING

The evolution of climatic variability has been studied in the interval of the Funza I pollen record that represents

100

o ~

o o p s ~ a ~ s o !

6

m o 1 ~ ~ o o

Aal (kl)

FIG. 5. ~ cenelation of the Ftmza I ,*atxmml Pol~ mcm~ ~ ~ttt immdza ~ 1984, ]989; tim omaet brad eaAadeiema a a L t ~ J ' ~ ) e a d l P ' O m a m l ~ m i 6 ~ ~ a a L ]990).

minimally ~ and ~ over 35 d~x imm~ab Imwe~ 36¢mmoi l m i m m ~ as ~ ) . The parallel Rcords show the provisJoul c¢~lmloa for ~C) Slalp 3-25. (After Hoogldmm~ et aL, 1993; see

also Hooghic~nstra and 5arndeato, 1991.)

High Resolution Pollen Sequence of Colombia 67

Arboreal pollen (Furtza I pollen record)

1010-1410

,0-1310

.210

.0

1130 50 33.3 25 20 16.7 14.3 12.5 11.1 10

FIG. 6. Frequency analysis (Maximum Entropy Spectrum Analysis) using the 'Arboreal Pollen' data set (A/nus included) of the Funza I core interval 2-340 m (period 30-1450 ka) by a 400-ka-wide window moving in 20 ka steps. This pollen data set represents the altitudinally shifting upper forest fine, mainly as a response to temperature change. The first spectrum covers the interval 30-430 ka. Every fifth spectrum is shown bold, with spectral peaks identified by their corresponding period in ka. Some peaks are truncated for readabifity. Note the presence in the upper 29 spectra of the record periods of the eccentricity band (ca. 100 ka, representing the classical Quaternary glacial-interglacial cycles). A band which is related to the precession is continuously present in the record with periods centred at 21-24 and 17-19 ka. (After Mefice in Hooghiemstra et al., 1993.)

(using the revised time frame; Andriessen et al., 1993) the period from 30 to 1450 ka (Hooghiemstra and Melice, 1991, 1993; Hooghiemstra et aL, 1993). The 1178 analysed pollen samples between 2.90 and 340.00 m core depth generate five time series with an average time resolution of ca. 1200 years. For this purpose a time frame is used, based on absolute datings (fission track datings on zircon; Andriessen et aL, 1993) in combination with correlation of the pollen record

• with the ODP Site 677 5~K) record of the Eastern Pacific (Hooghiemstra et al., 1993). This last procedure resulted in small downcore adjustments of the pollen record; in fact, the best possible corrections to compensate for changes in sediment accumulation rates. A set of 36 control points was generated (Hooghiemstra and Sarmiento, 1991; Hooghiemstra et aL, 1993).

The five different data sets (Arboreal Pollen, Arboreal Pollen excludingAlnus, Alnus, marsh elements and Quercus) represent different variables of the palaeoclimate (Hooghiemstra et aL, 1993). These time series were then analyzed by Melice in the frequency domain with the help of Maximum Entropy Spectrum Analysis and Thomson Multi- Taper Spectrum Analysis using a moving 400 ka window with steps of 20 ka. Frequencies closely related to the Milankovitch theory were detected in the five data sets. The presence of ca. 100,000 year periods (eccentricity band) was

found in time series with Arboreal Pollen and appeared to be restricted to the last ca. 800 ka (Fig. 6). Periodicities close to 23 ka (precession band) are present throughout the pollen record and are strongest in the data set of subparamo elements (Fig. 7). The five selected data sets reveal an evolution of orbital frequencies in the Middle and Upper Quaternary which is in general agreement with other marine (e.g: Imbrie et al., 1984) and continental (e.g. Kukla et al., 1990) records.

DEVELOPMENT OF FLORA AND VEGETATION OF ANDEAN MONTANE FORESTS AND PARAMO

Around 4-3 Ma, the principal upheaval of the area had ceased (van der Hammen et al., 1973; Helmens, 1990; Andriessen et al., 1993). The high plain of Bogot~ was by then an extensive lake at approximately 2500 m altitude. In this lake basin, ca. 600 m of fluvial lacustrine and later pure lacustrine sediments were formed (Fig. 8), providing a long and continuous pollen record of the development of the montane vegetation belts and climatic change in northern South America during the Late Pliocene and Pleistocene time (Figs 9 and 10). Figure 9 shows the 357-m-long Funza I pollen record with a temporal resolution of ca. 1200 years (based on Hooghiemstra, 1984, 1989) representing the

68 H. Hooghiemstra and E.T.H. Ran

subparamo elements (Funza I pollen record)

100 50 33.3 25 20 16.7 14.3 12.5 11.1 10

FIG. 7. Frequency analysis (Maximum Entropy Spectrum Analysis) using the data set of 'subperamo elements' of the Funza I core interval 2-340 m (pexiod 30-1450 ka) by a 400-ka-wide window moving in 20 ka steps. The subparumo vegetation forms a belt with open (alpine) scrub vegetation above the upper forest line and is at present located from ca. 3200 to ca. 3500 m elevation (Fig. 2). The first spectrum covers the interval 30-430 ka. Every fifth spectrum is shown bold, with spectral peaks identified by their corresponding period in ka. Note the continuous presence of periods of the precession band (21-24 and 18-19 ka). (After Melice in Hooghiemstra

et al., 19933

interval from ca. 1.5 to 0.03 Ma. Figure 10 shows the upper 565 m of the Funza II pollen record with a temporal resolution of ca. 5000--6000 years (based on Hooghiemstra and Ran, submitted and Hooghiemstra and Cleef, submitted). Several phases in the development of montane forests and paramo vegetation of the Colombian Eastern Cordillera and distinct phases in the climate history can be recognized. In the next part, a concise account, based on the Funza II pollen record (depth intervals) and corroborated by the Funza I record, is given. Summary diagrams, upper forest line oscillations (temperature change) and provisional correlation with the marine ~'80 record is given in Fig. 11 for the interval 2-158 m (based on Hooghiemstra and Ran, submitted) and for the interval 205-540 m (based on Hooghiemstra and Cleef, submitted).

The interval 540-465 m core depth (3.2-2.7 Ma) shows that warm climatic conditions and pollen spectra have no Upper Quaternary analogues. The basin had just started to accumulate lacustrine and river sediments, after a period in which sediment only accumulated in the perifere valleys. The upper limit of the subandean forest belt was situated at some 500 m lower elevation than today. In the Andean forest belt Podocarpus-rich forest, Hedyosmum-Weinmannia forest (a precursor of the modern Weinmannietum) and

Vallea-Miconia forest, respectively, were the main constituents with increasing elevation. Hypericum and Myrica played an important part in the timberline dwarf forests, which possibly constituted a substantial transitional zone from the early Andean forest belt (upper montane forest belt) to the open grassparamo belt. The contribution of herbs to the paramo vegetation, dominated by Gram_ineae and Compositae, seems less diverse than during the Upper Quaternary. The Late Pliocene (upper) Andean forests were more open than during the Middle and Upper Quaternary, as heliophytic elements, such as Borreria, were abundant, The composition of forests on the high plain was subject to considerable change: arboreal taxa with pioneer qualities (Dodonaea, Eugenia) and other taxa (Syrnplocos, llex) constituted, seemingly at irregular intervals, azonal forests in the basin. The upper forest line oscillated most of the time from 2800 to 3600 m elevation. The average annual temperature on the high plain was 11.5-16.5°C.

The interval 470-460 m, dated around 2.7 Ma, shows a significant decrease in the contribution of Arboreal Pollen, suggesting that the average temperature level was lowered by some 4-5°C (provisional estimation). This e p i s ~ of rapid cooling is followed by a period of gradually lowering temperatures, i.e. from 2.7 to 2.2 Ma (460-405 m core

High Resolution Pollen Sequence of Colombia 69

(A)

A

E V

e, alma Q. O

= ! -' ~ a . o

10 -

[ I , . a t

Z O " i ] - - ~ - - ~ . !

3 0 ~ ~ _

:2-2

:2-3

F2~

F2-5

F2-e

F2-'~

!.0

3.0

17.5

!9.5

33.5

40.5

• - _ - . - - _ - - . _ - - - ,

. . . . - - - - - - -

60-~ _

g . . . . . .

7 0 -

w

b . . . .

8 0 -

9O ' * -_ --_--_ --...~ ~

r - -

| 0 0 - " ~ . . . .

F2 -7

F2-8

F2-9

F2-1"

F2-1:

F2-1:

F2-1'

;1.5

;5.5

1.5

87.5

93.5

4 0 "

1 1 0 . Z . . . . .

. ~ - - - - - -

1 2 0 ~ _ - _ - - _ - .

. .=.~=----_

1 4 0 - _ . . . .

?Z=',_t~.Lrt.Lm.

- 3,,m..._--

• ~ .~ . :S , - -_ - -_

- ~we

. . : 4 . . . .

1~. . . . . . .

100.5

F2-15

105.5

F2-16

110.5

F2-17

116.5

F2-181

131.5

F2-19

144.5

F2-20 1146.5

F2-21

FIG. 8. Lithology of the sediments from the Funza II bore hole (2-586 m core interval) in the basin of Bogota. Descriptions are made by S.R. Arevalo-Gamboa and I.D. Pinzon-Villazon. (After Sarmiento-Perez, G.,

Ingeominas, Bogota; internal document.)

70 H Hooghiemstra and E.T.H. Ran

(B)

¢I.

~60~

I?Oq

2

"t

I 1

0 0 e-

..J

w

0 N

no o= .el

~ 152.5

155.5 158.0

FIG. 8. cont.

interval). This drop in temperature falls within the global period of transition; from the average warm Pliocene climates to the cold climatic conditions during the Lower Pleistocene. Indeed, the classical terrestrial Pliocene- Pleistocene boundary (Zagwijn, 1975, 1985, 1992) falls within the period which is characterized in the Funza record by cooling.

The interval 465-415 m core depth (2.7-2.2 Ma) shows colder climatic conditions. The upper forest line oscillated most of the time from 2600 to 2800 m in the first half of this period, and from 2400 to 2800 m in the second half. Average annual temperatures were 10.5-11.5 and 8.8-11.5"C, respectively. The Podocarpus-rich forest type occurred until the end of this period. Weinmannia was almost absent and Hedyosmum completely dominated, for the first time, the Hedyosraum-Weinmannia forest type. Miconia dominated in the VaUea-Miconia forest type, in which flex, Myrsine and Daphnopsis were probably associated elements. For the first time in the Pleistocene, paramo vegetation became

widespread in the Eastern Cordillera near Bogota_ Caryophyllaceae and Valeriana were the most dominant paramo herbs at that time, whereas the contribution of Plantago and Aragoa increased. During intervals with low water levels, marsh vegetation, including Cyperaceae. Polygonum, Hydroco~le, Ludwigia, Myriophyllum Sphagnum andAzolla, was abundant on the high plato

The interval 415-337 m core depth (2.2-1.42 Ma) shows, for the first time in the record, a rather persistent cold climate. The upper forest line oscillated most of the time from 1900 to 2500 m, corresponding to an average annual temperature of 5.5-9.5°C on the high plain. Podocarpus had lost its dominant part in the (lower) Andean forest belt. Hedyosmum-Weinmannm forest and Miconia-Vallea forest were most important. Daphnopsis had almost disappeared from the Andean forest belt and Borreria became less common during this period, suggesting that forests became denser. Hypericum was, for the first time in the record, the most important element in the dw',wf forest, but at the end of this period Polylepis dwarf forest started to increase near the upper forest line and lower paramo. Juglans appeared for the first time regularly with low frequency and S~loceras also became a more regular component of the Andean forest belt. On a local scale, Plantago became very abundant in the basin and probably replaced a great part of the local grassparamo.

The interval 337-257 m core depth (1.42-1.0 Ma) shows a long period with mainly cold climatic conditions. The upper forest line oscillated most of the time from 2200 to 2600 m, at the end of this period slightly increasing to 2400-2800 m. This corresponds to average annual temperatures on the high plain of 7.5-10 and 9-! i.5°C, respectively. Weinmannia was almost absent in the Andean forest belt and a Hedyosmum forest, possibly with important contribution of Eugenia, Myrsine and Ericaceae, constituted a precursor of the present-day Weinmannia forest. Vallea contributed, for the first time in the record, of the same order as Miconia to the Vallea-Miconia forest type. Borreria occurred at low frequency and disappeared almost at the end of this period, indicating that the forest structure was more dense and unsuitable for heliophytic elements. Polylepis dwarf forest was important at the upper forest line, and in the (lower) paramo Myrica (M. pubescens) probably contributed substantially to the upper part of the Andean forest belt. The upper limit of subandean forest during this period reached higher elevations than during the lower part of the record (but only reached modern conditions after the immigration of Quercus, later on in the record). The lake on the high plain was shallow and extensive marsh prevailed during the first half of this period. Possibly due to tectonic adjustments of the basin, the lake became deeper from ca. 3130 m core depth onward, and lso~tes vegetation and algae became abundant in short time.

The interval 257-205 m core depth (1.0--0.85 Ma) shows the first major glacial-interglacial cycles, characteristic of the Middle and Upper Quaternary. Comparison with the fi'80 record of the deep sea indicates that climatic change in the interval 240-205 m correlates with Stages 25-2I . The interglacial (Stage 25) to glacial (Stage 22) lowering of the upper forest line is ca. 800 m (from 3000 to 1800 m, maximally), corresponding to a temperature decrease ,H

High Resoludon Pollen Sequence of Colombia '71

(c)

i ! I i ' F2-27

F2-28

F2-29

F2-30

207.5

216.5

232.5

I I I l I l l i l IH I

- - i l I I J

i i J l i i

I I i i J I

250. ' -- '~-._'- '_~,

. . . . . . q

. . . . . . i

. . . . . ~ 9

' I | | | | l l | l l . | l l J

. . . . . q

~_-_-----~

ZT0 . . . . . . a l i l l I I l lO l , i l l n l l l l l l I

i i l l l l l l l lU

I i i i i i i i i i i i

280 .------------------ m ~ m i ~

. - - m l m .

:-i--'---"----"

I | | | | | | | | 1 ~

I l I I I I I i l l l l l

I I I I I I I I I I I I I I

--%--__--_.-.

Di l l l l l l l l l l l

F2-30

F2-X1

F2-X2

F2-X~

257.0

265.0

297.5

336.5

346.5

HG. 8. com.

..,

.,

~,

,

,,

, ~

,,

,

, ~,

, ,

, ~

, ..

:..

,,,,

,,

,,,

,,,,

~,.

,,,.,i

,,.~,

:,,,',

~,,,~

,, ,,,

,.i.,,

,,-,,,

,,,.,,

,,,,,,

,,~-,,

, =,

,,,,,,

"

' ..,

;,:,,,

., ,,

.,','

,',',"

','.'.

',',"

', '1

" ',,

,,; '

-,.,'.

,'.,',

' ~'

"",',

',~,,,

',,':,

:~','

,-,',

,'.="

,'

=','.

',":"

" '";

;",:

','.',',

::!=i

Ii

, I'~

tll :

Ii1::

'~ I

• :h

;InV,

l',l

tll%

ll ~1

,1

I!1',;

' ;I

~ I

Ihhl

,U'.

'.11

i'. :~

'~t,

1.1'

~ '~

lll.

ll:l.I

i %

1,I,

it!i!1

'. '

=l'lq

, I

: I'.

l'l'~

'i'.ll

lll,

',11'

.1'!1

;1',.

'1:1

1 r"

lrl

I:"

it h

l:n.!.

:r,l

~-

.Ih~

.ll,l:

l:lil'

,ltll]

l~

* ',l

'.nl'.

ftlV

l!l:*

~ll:l

'.l,~

ti,l*

Ii1

.: I

',i;!',i',!i

!:: in

].

ll I,

l',lll

,llll'

,,,ll

~;l~

l~l'(

i,,I

I Ijl

ld,!

ll;I,

ii'l',

il.;;

I.i;,

~ll

&".liLlq'l,,ll

I!il'

~llli

tl:l|l

llllv

!l.lu

ll.lll

l Ji

m

I,~

;~ .,

, I.~

:~

,,:~,

;,;,,,

,.,

,,;,

:.,;

:,=

,: =,

,,,,,.

;,:,.,

:,;,~

,:,:~

:,,:,.

,~

-!,.,:

,,,!,

:,.,,;,:

.,,,,;,:

,,,,=.:

.:,=,,!.

~,,:~,

,m!

Dep

th (

m)

Lith

olog

y

v

" ii

" ~

x

L "1

1 PO

b~

I P

olle

n zo

nes

Dep

th

Pol

len

zone

bo

unda

ry

• ?

I ,

I ¢~

I

I I

? ,I,

, I

# *

' '

m

, *

m

• Ill

ll!lll

l I'"

"!

~# I

'l'i~

illlll

~Jill

" '~

1""

"1" :

I:

ll'*l

'illd

',

• Ilt

lli"

im!lm

m~'

" 'l

viiir

l'li"-

' "i1

' "1",

"|"

""

.'l'lll

. , I

~. I;

Iliim

l rl

J, •

, ,.1

"* ,-".,

..

eill,

l:l'l

I -

.I •

i ,ll

llll.l

lllll.

i .~

l...i.

l'.Tl

l'rl'l

,

I.'l

i',l:

:l',l;

l:l]l;

I i'.

ll;I.;

l~

;i;1:

,'[-I

i"~.:1

.~1"

111

IiIih

ll'.ll

:!i'l:

iil'=

l"z.l.

I"

I--

"11

I I,

, '

--I

I ii

I

I I

,.~

.'-

'.1.1

" " i

I

' :

* Il

l i

Imll

lSll

',

"''.

II

, '1

I1'1

" :.

1~"

I |

I s

I ."

,*

i"''

lil,, " ....,

"", , - ..

:I,.

"

- , '

,.,' ,, ...,

" , ,,[,

" ,,, ' .,,..

" -..

-"'

""

...., ...,, , .... ,, II,:I,,,.

,, ,, ,

,,', !,:,:,,,!,,,!,,:,I,!.,~,.,I,,,,-,-!.,

-_.:,.,.!,.,,I,,,!,,I,.,=:, ,,:,:,-!,

• I !

~l:

*;*:

hlll

°ll!

l!ll

ll'*

':l;

':l,

ll-I

'!'l

"'1 I

:l~i

?ll

~l~

l.lt

llil

~*l

I!

1 ,l'

l '1"

~"

rO

CO

L.

- ]

o

i i

i I

l

~1[:~

1:1"

1;I

I:lll

~'ll

I ~

~1.1

.1'

.I I

,,

i

~ •

~;

.: ~:

lll'i;

!lllll

~ll

-I:ll

'l:.ll

I

II,I

I I I

III;

,.11

Il

lli'.

l:l;

I

" ['1

x

b

Dep

th (

m)

Lith

olog

y

Pol

len

zone

s

Dep

th

Pol

len

zone

bo

unda

ry

r~

Q

O =. Y

I;ll,

l'l

:l:lil

l;llll

I

Illill

llll

l'll

l !

II!li

lll;I

;l:l

I I

I I

I It

.lli

.I

,I,.

It

hlll

ll~

!.1

,l

it

i:1t

;i;1

11

1"i

! o

0 *

• !

i s

•

llllll

illlll

llllll

illlil

llilll

!IH

le

~ll

I'I'I

lll'll

lllnl

llil~

i I

I,I,

I,II

I:IE

I I

f I

I I

..

..

i

I I

I Iii

Iii

ii Iii

l:l

~,',.

t, ,l,

!,l,!H

d,,,l

,,,,,l

'

""'

, ..

....

,,

.I

d,iH

d!H

Jtl~

Il

l,ll

l'll!

l!ll

lJl!

l!,

I1''1

'1,

li*'

, II,

.',,'*

I a'

ll~l'

:;<

' 1

I I,

I !

P ~

q [i'

." ,

I li

t i;

( 3

".;.

'lli

IIIl

ill'

I .

"",'

"i

- "

"1

t" "

""

""

~lll

i- I

l=

• ,

' I'1

•

., ,.

I,

",'

• '

[I..'J

I

Ill.

.'1 I

t"

"LII

IIII

III,

t.

I I

, I,

" .

.'1 ,

,It

I t'l'

t .

.: tl

i!-.-lil

LH-

i=

18t,~

I,.

i,,,

. .

|1

i s

, ,

|1.1

",'

l¢

/.

III

1,1'

1 I.

- i

- '

'"l

,l'l

i -]

!:--.r,i'"1'l

,,h:..,,,...i.

;:--,:!.Iih!,',i,.~ :-:*-*.,

,.. ,i.

:~::-"|1.

i I

"n

p,

p O

On

Dep

th (

m)

Lith

olog

y

Pol

len

zone

s

Dep

th

Pol

len

zone

bo

unda

ry

High Resolution Pollen Sequence of Colombia 73

(E)

510 -

~20-1. . . . ...I i . ' . : . ". .t

- : , , . .:.

@ . . U .

,w v =

m~..-.-.. - =.. n . . . . . , i

, i . . . . . . ,

9 - . . . . . .

I . . . . . . I

~40. 7-~.~_-~ :,r_7._-~, -'7_-L-~

~---.J---'Z 7..r_-~;

550 7-"~---"r:-*

¢ . c : ¢ ~ _o Q Q

0 Q O 0

l 5 ~ .0

m~m

509.0

:2"X11

! 517.0

:2-X 15

:2-X1E

nO I le l~

529.0

539.0

i

nomm

LEGEND:

Clay, plastic near the surface and compact at greater depth, mostly dark grey, varying depending the organic matter content. Mica and vivimute present with maximal I%.

Lenses of medium sized sand

Volcanic ash horizons, white m grey of colour.

Lignite horizons with variable percentages of organic matter

Sandy clay and clayey sand.

Not determined sediment (most possibly sands i nc lud ing d i f f e r en t minerals).

Sand of fine to medium grain size (often difficult to recover).

Clay, olive m grey-olive and very compact with occasionally surfaces indicative of friction and bioturbation (occurs below 410 m core depth).

Dark mud with some lamination affected by b io turba t ion . Local wesetw~ of vivianite.

Very fine sand with biogenetic structures, undulating fragments of organic material and occasionally presence of gravel up to 5 mm diameter.

FIG. 8. Cont.

74

(A)

40 55

H. Hooghiemstra and E.T.H. Ran

Pollen Dia0tlm FUNZA I (2-357 m). Ea l l tm ~ r l , Cok3m¢~ (25,f~0 m illlttL~l) maoluUon ¢. 1200 years .~' ,~y~: H .Ho~ '~m l lm ~ O.K.Hua¢~3040

75 t00~

80

subandcan Andean / grass- forest forest / paramo

subparamo

3040 60 ' - " 30 40 60 20 30 50

FIG. 9. Summary diagram of pollen record Funza I showing the altitudinally shifting vegetational belts in response to climatic change during the last ca. 1.5 Ma with an average time resolution of ca. 1200 years. Core depth (m) and age (ka) are indicated at the left hand side of the diagram. Indicated ages correspond to the control points of the correlation between Fuaza I Arboreal Pollen record and ODP Site 677 8'nO record (see Fig. 5). For convenience, ~ 'O Stage numbers 3-23 are indicated in the diagram based on a provisional correlation. Downcore oscillations in the retm~'sentation of vegetation belts are shown for subandean forest belt, Andean forest belt, subparamo belt and grassparamo belt. The graph of downcore changes of the percentage of total Arhoral Pollen (AP; Alnus included) shows oscillations that, in fact, represent vertical shifts of the upper forest line over the mountain slopes as a response to mainly temperature change. Three percentage levels of AP are indicated that correspond to altitudinal positions of the upper forest line at 2000, 2550 and 3000 m. These levels, on which the boundary between interglacial and glacial periods is based, change to other percentages when new immigrants from the northern hemisphere, after crossing the Panamanian landbridge, changed the composition of the Andean montane forests This happened at 257 m core depth (ca. 1 Ma), when Alnus immigrated into the area of Bogot~t, and between 77 and 45 m when the contribution of zonal Quercus forest increased rapidly as a part of the Andean forest belt (Quercus immigrated into the area of Bogot~ around 340 ka and appeared at ca. 94 m core depth in the Funza records). These changes in percentage levels are approximations to account for distinct changes in the composition of the Andean forest belt (percentage levels are not indicated in a short interval around 255 m core depth, which is characterized by a hiatus). Estimations are based on improved understanding of the Bogo~ pollen records (Hooghiemstra, Ran, Van't Veer and Mommersteeg, unpublished data). The inferred changes in mean annual temperature, at the elevation of Bogota, ate from about 6 to 15°C. The former lake of Bogot~ drained ca. 27 ka and this last part of the Quaternary is missing in the Fanza records. The top samples are of Holocene age (Pollen data

after Hooghiemstra, 1984, 1989; time control afterb Andriessen et al.. 1993.)

High Resolution Pollen Sequence of Colombia 75

(B)

1 2 3 4

1 subandean forest 2 Andean forest 3 subparemo 4 grassparamo

FIG. 9. Cont.

76 H. Hooghiemstra and E.I'.H. Ran

"-" t0 E

v

_c 20 ~o "o

30

40

50

6O

7O

80

g0

100

110

120

t30

t40

t50

Pollen diagram FUNZA 11 (2-540 m), Eastern Cordillera, Colombia (2550 m altitude) Time resolution c. 6000 years Analysis: H.Hooghiemstra and E.T.H.Ran

240 470

250

250

270

280

290

300

310

320

330

340

350

360

370

380

390

480

1000 490

1050 5OO

5t0

52O

1260 530

1420

t490

1580

540

550~ no poilen 1 recovery i

I 2 3 4

lg10 1960

1 subandean forest 2 Andean forest 3 subparamo 4 gruapsramo

400

no core 410 r e c o v e r y

420

430

210 ~ 8 5 0 440

220 ~ 890 450

230 ~ ~ - i l ~ -" .............. 940 460

2200

2360

2530

2700

FIG. 10. Summary. diagram of pollen record Funza II (565-2 m core interval) for the period of ca. 3.2 Ma Io ca. 27 ka, with a time resolution of ca. 5000-6000 years (sample distance 100 cm). Core depth (m) and estimated ages (ka) of pollen zone boundaries, based on absolute time control and land-sea correlnfion ( ~ n et al., 1993), are indicated. In the interval 565-586 m, no pollen was recovered. For convenience, 8"O StN~ numbers 5-25 are indicated in the diagram based on a provisional correlation. Downeore oscillations in the t~resenl~ion of vegetation belts are shown for subandean forest belt, Andean forest belt, subp;u'amo belt and grassla~anm ~eit. The graph of downcore changes of the percentage of total Arboreal Pollen (AP; A/nus incladed) shows oscillations that, in fact, represent vertical shifts of the upper forest line over the mountain slopes as a response to mainly temperature change. (After Hooghiemstra and Cleef, submitted, and Hooghiemstra and Ran,

submitted, for lower and upper parts of the pollen record, respectively.)

2700

2820

2850

2910

3000

3060

3120

3200

4.8°C. Af te r tbe i m m i g r a t i o n of Alnus, a charac ter i s t ic nor thern hemisphere genus, large areas o fca r r (swamp forest vegetat ion) developed on the wet fiats around the lake, but Alnus probably also occurred incidental ly as an e lement of the zonal forests. The contr ibut ion of Myrica was reduced considerably, indicat ing that Myrica contr ibuted before to

the azonal vegetat ion (M. parvifolia) as well as to the zonal forests (M. pubescens). The large a l t i tudinal shifts o f all m o n t a n e vege ta t i on bel ts , in r e sponse to the m a i n glacia l - in terglacia l cl imatic cycles, in the remain ing part o f the record places the high pla in al ternately in the Andean forest belt and in the grassparamo belt.

High Resolution Pollen Sequence of Colombia 77

:: | - ~ Estimated altitudinal

' I+! il ;-_- ,on.., ,in.

| " | 0 % 1 0 0 % . . . . . .

85?

877

101?

107?

1117

117?

22 O.87

23 0.9

24 25 O.95

27 O.99

29 .1.02

37? 1.24

39? 1.25

43? 1.35

45? 1.35

47? 1.44

57? 1.63

73? 1.94 757 . 1.99

2.23

2.25

.2.47

2.60

,2.73

2.88

| tO

me

I m

140

l eo

17o

18o

Iio

3oo

31to

380

~ 0

IdO

~ 0

38O

310

4OO

410

130

44O

45O

410

470

410

49O

i l i a .

S I 0 -

5 3 0 -

5dlO- | !:++!.+. , . _R

1 I : I 4

1 m l k a d u n b m m I ~ka l l l en f m l m I wmm, m 4 Immmmam

F2-25

F2-29

P2-30

F2-X1

I~-X2

F3-X3

F2-X4

~-xs

I ~ -X8

F2-X8

F2-X9

F2-Xl 0

~2-X11

F2:X12

F2-X13

i

F2-X16

2 0 5 . 0 •

2O7.5 | 216.5 ,I.

!

257.0

25s.o + I t t t

i

2 ~ . 5 I" i

335.5 'I"

348.5 .],

354.5 'F

380.5 ,L 382.5 T

405.0.4'

430.0 +

443.0 .I.

465.0 +

486.0 4

494.0 I'. 500.0 .I.

509.0 .I.

517.0 4.

527.5 ,t

539.0 x

1217

I~, kll .

l ¢ m

I J

:7

I

/

( :

m ~

i

I % 1 1 m ~ q ,

¢ : " -

1

1

,w

C

3 , r

q

i .

, i !

. . . . . . . . "F1.2"r" F1-28

F1-29 . . . . . . . .

F1 4 0

. . - " ° F141

",.. :F1-t,r

F1-60

l

. . . . . . . . IFI-S4:

208.00. 216.00

233.00

251.00

259.50

~ . ~ 310.00

325.00 331.00

347.00 350.00

Estimated temperature at the level of the high plain el Bogota

FIG. 11. Correlation of the pollen records Funza I (Hooghiemstra, 1984, 1989) and Funza II (Hooghiemstra and Ran, submitted) and inferred climatic change for the interval of ca. 24-735 ka. Correlating pollen zones in both records have the same numbers, with a prefix FI (Funza I) or F2 (Funza II). Depth and age of pollen zone boundaries are indicated. A provisional correlation with the marine 8~q3 stratigraphy is indicated. Ages of 8'*0

stages after hnbrie et al. (1984). (After Hooghiemstra and Ran, submitted.)

7~ H. Hooghiemstra and E.T.H. Ran

The interval 205-158 m is not recovered in Funza II due to technical problems (Fig. 10). This interval of the sediment sequence, however, is well documented in the Funza I core (Fig. 9). It corresponds to the interval of ca. 170-207 m of the Funza I core (Funza I pollen zones 27, 26 and 25A), representing the period of ca. 735-845 ka. This interval represents two distinct interglacial periods with a marked glacial in between and corresponds to the ~i~80 Stages 21, 20 and 19 (Fig. 9). More precise fitting of this Funza II hiatus in the Funza I record is deferred until high temporal resolution pollen data are available. The upper forest line oscillated in this period, mainly from 3000 to 1900 m, for most of the time. The corresponding average annual temperature on the high plain is 13.6°C

The interval 158-131 m core depth (estimated age 735-569 ka) shows warm climatic conditions most of the time and is tentatively correlated with the 8'"O Stages 19.1-15.1. The pollen spectra have no direct modern analogues because of the absence of Quercus and related conditions. The upper forest line oscillated mainly from 2100 to 2700 m for most of the time. The corresponding average annual temperature on the high plain is 6.5-11°C. The high plain was situated in the Andean forest belt most of the time. The upper limit of the subandean forest belt (Acalypha, Alchornea) was situated some hundreds of metres below the modern elevation. Podocarpus was most important in the lower part of the Andean forest belt. Weinmannia forest, the precursor of the modern Weinmannietum, included a substantial contribution of Hedyosmum, with lower frequency Myrsine and Eugenia. A type of Vallea-Miconia forest, including low presence of llex and Myrsine, could have occurred on the drier parts of the high plain, the lake was shallow most of the time, with local marsh vegetation of cyperaceous reed swamp and Hydrocotyte. Myrica thickets (M. parvifolia) and Alnus carr covered the wet fiats around the lake. Myrica (M. pubescens) and AInus possibly also contributed with low frequency to the zonal Andean forest belt. Dwarf forest of Polylepis, Myrica and Compositae scrub occurred at the upper forest line

The interval 131-100 m core depth (estimated age 569-350 ka) shows cold climatic conditions most of the time and is tentatively correlated with the 8'sO Stages 14.4-11.1. The upper forest line oscillated mainly from 1800 to 2500 m. The corresponding average temperature on the high plain is 5-9.5°C. The high plain was situated in the grassparamo belt most of the time. Apart from Gramineae (e.g. Calamagrostis, Chusquea) and woody stem rosettes of Espeletia (Compositae), a variety of paramo herbs (Valeriana, Caryophyllaceae, Geranium, Aragoa, Lycopodium fov.) were present with substantial frequencies, and abundant cushion bogs ofPlantago (P. rigida) were present. The water level in the lake was high and marsh vegetation limited. Polylepis dwarf forest occurred in the subparamo belt, along with shrub of Compositae, Hypericum and Ericaceae. In the Andean forest belt, Vallea-Miconia forest and Weinmannia-Hedyosmum forest were most im- portant.

The interval 100-57 m core depth (estimated age 350--186 ka) shows warm climatic conditions most of the

time and is tentatively correlated with the ~sO Stages 10.2-7.1. The upper forest line oscillated from 2000 to 2600 m in the first part and from 2600 to 2900 m in the last part of this interval. The corresponding average annual temperatures are 6-10 and 10--12°C, respectively. The high plain was, in the first part of this interval, mostly situated in the paramo and in the last part of this interval in the Andean forest belt. During this interval Quercus immigrated into the area of the high plain. Quercus forest occurred in a wide altitudinal range (1000-2800 m) and initially constituted local patches of forest, but at the end of this interval, zonal Quercus forests were a major part of the Andean forest belt. Acalypha and Alchornea reached higher elevations m the Quercus forests, and the upper limit of subandean forest rose to modern elevations. Weinmannia dominated in the Weinmannia-Hedyosmum forest type. At the end of this interval, the contribution of Vallea-Miconia forest increased markedly and replaced Weinmannia forest. Podocarpus-rich forest occurred in the lower part of the Andean forest bek Alnus cart and vegetation of Myrica thickets were abundant around the lake, which was of a shallow type. Algae (Botry'ococcus) became very abundant from the beginning of this interval to the top of the record.

The interval 57-2 m core depth (estimated age 186-24 ka) shows, for the first time in the record, abundant presence of zonal Quercus forests. The composition of the Andean forest belt had changed dramatically. Based on Arboreal Percentages, climatic conditions seem warm most of the time, but the high frequency of Quercus, a wind-pollinated tree that produces large amounts of pollen, exaggerates natural conditions. This interval is provisionally correlated with the 5~O Stages 6-3.0. The upper forest line oscillated from 2000 to 3000 m most of the time. The corresponding average annual temperature is 6--12.5°C. Quercu.~ forests. resembling the modern Saurauia-Quercus humboldtii forest, and Weinmannia-Hedyosmum forest, resembling the modern Weinmannietum, dominated in the Andean forest belt. Vallea-Miconia forest probably resembled the modern Xvlosma-Duranta-Vallea forest, but the latter is palynologically difficult to recognize. Eugenia, llex and Myrsine contributed substantially to this rather dry forest type of low stature. Polylepis dwarf forest was frequent at the forest line and possibly also in the paramo belt up to 4000 m. Alnus carr completely dominated the flat parts of the high plain. Myrica thickets and marsh vegetation were reduced in the last part of this period. Sediment accumulation was v e ~ high (up to 60 cm per 1000 years). Supposedly, erosion of the Tequendama Falls in the Rio Bogota, the only outlet of the high plain, led to the final drainage of the lake, depending on the location between ca. 28 and 22 ka.

The Andean biozones IV-VII (van der Hammen et al., 1973) are represented in the Funza records. Biozone IV (540-415 m core interval; estimated age 3.2-2.2 Ma) is mainly characterized by high percentages of Borreria. Alnus and Quercus are absent. In biozone V (415-257 m core interval; 2.2-1.0 Ma), Polylepis replaced Hypericum as the major element in the dwarf forest zone, Weinmannia replaced Hedyosmum as the most important element in the Weinmannia-Hedyosmum forests (precursor of the modern Weinmannietum), and the upper limit of the subandean forest

High Resolution Pollen Sequence of Colombia 79

+: i ++i{ i + "'t ++ '

+i- -+ ++ +'- _ - . , - , o , , + +m,.~ ,.co., F,.,=, ,, = : ~ + ~ +

• : ~ . ~ _ _ _ - g , , ,;t t~ 3.0

3.1

3.3

4.2

5.1

5.2 5.3 5.5

6

7+1

7.2 7.3 7.4 7.6

6.2 9.1

g.3

10.2 11.1 11.3 12 13.11 13.13 14,2 14.3 14.4

15,1

24

28

53

65

8O

87 99

122

163

205 ~216 P2~6

2 ~

249

310 231

Pd 405

45t

535

574

617 16,5 16 17,1 17.3 16 18.1

100% .F2-2--

F2-3

F2.,4

F2-5

F2-6

F2-7

F2-8 F2-9

F2.11

F2-12

F2-13

F2-14

F,-i5 F2-16

F2-17

F2-18

F2-19

F2-21

1677

1016

1142

32~

2269

6"6i 51661

z~3s I

120001 J, J 58331

1

39331

4078

IgO0 4 5 3 3

I : J r ' ~

. ~ . . ~ ; ; . _ _ i ¢ , i [ I " ~ - - , -

I ,

L ~= --L- n ~ ,,m~l E

, . . i , i

! c : I

.3",

i ~. .~ , i s r

F1-2 .°.

. . , ' " F1-4

, - ' " - F1-5 .*" . - ° ' 7

..-" F1-6

" . . . . F~-7.

.- '", Fi'-'e ¢• . - - p •

" ~ N "

I •oo°

I " ° ° " ' " 1 '

• o - - . , . . . . ,.

,oo.

i °jp~.

,..o

I • - . . : : o . o --.

o. 0¢.~ W lq l~ ~. ~ m m. 0 C •-/lID i~. lID @D w t~l t~ ~ ca,/ ~-:~:.-..:::

1 2 3 4 E s t i m a t e d t e m p e r a t u r e

1 subandxn ior~t a t t h e level of t h e 2 Andun b r u t high plain of Bogota a s u m m u m 4 grmmrmo

F1-11

F1-12

F1-13

F1-14

F1~15

F1-16

F1~17 F1-18

F1-19

FI-20

FI-21

F1-22

820 9.50

22.50

28.00

38.00

43.60 48.60

58.50 62.00

74.00

50.00

92.00

lOT.00

101.00

114.00 116.00

123.00

137.00

143.20

151.40

1~.00

FIG. 12. Correlation of the pollen records Funza I (Hooghiemstra, 1984, 1989) and Funza II (Hooghiemstra and Clenf, submitted) and inferred climatic change for the interval 0.8 to ca. 3.2 Ma. Correlating pollen zones in both records have the same numbers, with a prefix F1 (Funza I) or F2 (Funza II). Depth and age of pollen zone boundaries are indicated. A provisional correlation with the marine 8LtO stratigraphy is indicated. Ages of 8'"O stages after Imbrie et al. (1984) up to Stage 22 and after time control of ODP Site 677 (Shacldeton et aL, 1990)

for the lower part of the pollen record. (After Hooghiemstra and Cleef, submitted.)

belt had increased several hundreds of metres. Alnus and Quercus are still absent. Biozone VI (257-94 m core interval; estimated age 1.0--0.33 Ma) is characterized by the immigration of Alnus (first appearance date for the study area 1.0 Ma). Biozone VII (94-0 m core interval; estimated age 0,33 Ma to recent) is characterized by the immigration of Quercus (first appearance date for the study area 0.33 Ma, but present as an element that formed relevant zonal forests'

since 0.2 Ma). The upper limit of the subandean forest belt increased and reached modern elevations.

CONCLUSIONS

The extremely thick sediment sequence of the high plain of Bogot~l, which accumulated continuously during the last 3 Ma, forms a unique and important source of data for

80 H. Hooghiemstra and ET.H Ran

palaeoclimatological, palaeoecological and biogeographical studies with high temporal resolution. The first studies were initiated by van der Hammen and the present research is characterized by interdisciplinary studies. Several research projects concerning different aspects of the Quaternary history of the high plain of Bogot~i are in progress. They are carried out within the research tasks of the PAGES Project of IGBP (International Geosphere-Biosphere Programme); see IGBP Global Change Reports 12 (1990) and 16 (1991).

ACKNOWLEDGEMENTS

We thank the following researchers for their stimulating cooperation: P.A.M. Andriessen (Free University, Amsterdam), A. Berger (Institut d'Astronomie et de Geophysique, Universit~ Catholique de Louvain, Louvain-la-Neuve), A.M. Cleef (Hugo de Vries Laboratory, University of Amsterdam), K.F. Helmens (Hugo de Vries Laboratory, University of Amsterdam; at present University of Alberta, Edmonton), J.L. Melice (Institut d'Astronomie et de Geophysique, Louvaln-ta-Neuve; at present Observatorio Nacional, Rio de Janeiro), D. Mosquera (Ingeominas, Bogota), P.A. Riezebos (Institute of Physical Geography and Soil Science, University of Amsterdam), G. Sarrmento-Perez (Ingeominas, Bogota), N.J. Shackleton (Sub-department of Quaternary Research, Cambridge University) and T. van der Hammen (Hugo de Vries Laboratory, University of Amsterdam; at present Tropenbos, Bogot~i). We thank Colciencias (Bogota), the Netherlands Foundation for Scientific Research (NWO, grant H 75-284) and Ingeominas (Instituto Nacional de Investigaciones Geologico-Mineras, Bogot,5) for generous support.

REFERENCES

Andriessen, P.A.M., Helmens, K.F., Hooghiemstra, H., Riezebos, P.A. and van der Hammen, T. (1993). Absolute chronology of the Pliocene--Quaternary sediment sequence of the Bogoui area, Colombia. Quaternary Science Reviews (in press).

Cleef0 A.M. (1981). The vegetation of the paramos of the Colombian Cordillera Oriental. Dissertaciones Botanicae, Vol. 61,321 pp. [Also published as The Quaterna O' of Colombia (van der Hammen, T., ed.), Vol..9.]

Cleef, A.M. and Hc~ghiemstra, H. (1984). Present vegetation of the area of the high plain of Bogot& In: Hooghiemstra, H. (ed.), Vegetational and Climatic Histor) of the High Plain of Bogotd, Colombia: A Continuous Record of the Last 3.5 Million Years. Dissertaciones Botanicae, Vol. 79, pp. 42--66. J. Cramer, Vaduz. [Also published as The Quaternao' of Colombia, van der Hammen. T. (ed.), Vol. 10.]

Cleef, A.M., Rangel-Ch., O. and Salamanca-V.. S. (1983). Reconocinuento de la vegetacion de la parte alta del transecto Parque los Nevados. In: Van der Hammen, T., Perez-Preciado, A. and Pinto-E., P. (eds), La Cordillera Cet, tral Colombiana transecto Parque los Nevados (Introduccion y datos iniciales). Studies on Tropical Andean Ecosystems, Vol. 1, pp. 150-173. J. Cramer. Vaduz.

Helmens, K.F. (1990). Neogene-Quaternar3 ~ Geology of the High Plain of Bogotti, Eastern Cordillera, Colombia (Stratigraphy, Paleoenvironments and Landscape Evolution). Dissertaciones Botanicae, Vol. 163, 202 pp. J. Cramer, Berlin.

Hooghiemstra, H. (ed.) (1984). Vegetational and Climatic History of the High Plain of Bogotd, Colombia: A Continuous Record of the Last 3.5 Million Years. Dissertaciones Botanicae, Vol. 79. 368 pp. J. Cramer, Vaduz. [Also published as The Quaternary of Colombia, van der Hammen, T., ed.), Vol. I0 I

Hooghiemstra, H. (1989). Quaternary and Upper-Pliocene glaciations ant* forest development in the tropical Andes: Evidence from a long high- resolution pollen record from the sedimentary basin of Bogota. Colombia. Palaeogeography. Palaeoclimatologv Palaeoecolo~y 72 1t-26

Hooghiemstra, H. and CleeL A.M. (submitted). Lower Pleistocene and Upper Pliocene climatic change and tbrest development in Colombia: Pollen record Funza II (205-540 m core interval). Palaeogeograph)~ Palaeoctimatology, Palaeoecology

Hooghiemstra. H. and Melice, J.L. (1991). Evolution of the orbitaJ periodicities in the high-resolution 30-1450 ka pollen record Funza L Eastern Cordillera, Colombia. Volume of Abstracts, S),mposium Y)rbital Forcing and Cyclic Sedimentary Sequences ', 22-23 February t 99 I. Utrecht University, The Netherlands, pp. 15-21.

Hooghiemstra, H. and Melice, J.L (1993). Pleistocene evolution of orbital periodicities in the high-resolution pollen record Funza 1, Eastern Cordillera, Colombia. In: de Boer, P. and Smith, D.G. (eds), Orbital Forcing and Cyclic Sequences, Vol. 19, pp. 117-126. Special Publications, International Association of Sedimentologists. Blackwell, London.

Hooghiemstra, H. and Ran, E.T.H. (submitted). Upper and Middle Pleistocene climatic change and forest development in Colombia: Pollen record Funza II (2-158 m core interval). Palaeogeography. Palaeoclimatology, Palaeoecology~

Hooghiemstra, H. and Sarmiento, G. (1991). New long continental pollen record from a tropical intermontane basin: Late Pliocene and Pleistocene history from a 540 m-core. Episodes, 14, 107-115.

Hooghiemstra, H., Melice, J.L., Berger, A. and Shackleton, N.J. t 1993). Frequency spectra and paleoclimatic variability of the high-resolution 30-1450 kyr Funza I pollen record (Eastern Cordillera, Colombia). Quaternary Science Reviews (in press).

Imbrie, J., Hays, J.D., Martinson, D.G., Mclntyre, A., Mix, A.C, Morley, J.J., Pisias, N.G., PreU, W.L. and Shackleton, N.J. (1984). The orbital theory of Pleistocene climate: support from a revised chronology of the marine delta ~O record, In: Berg©r, A. et al. (eds), Milankovitch and Climate. Part I, pp. 269-305. NATO ASI Series, C 126. Reidel, Dordrecht.

Kukla, G.. An, Z.S., Malice, J.L., Gavin, J. and Xiao, J.L (1990). Magnetic susceptibility record of Chinese loess. Transactions of the Royal Society Edinburgh, Earth Sciences, 81,263-288.

Shackleton, N.J., Berger, A. and Pettier, W.R. (1990). An alternative astronomical calibration of the lower Pleistocene timescale based on ODP Site 677. Transactions of the Royal Society of Edinburgh: Earth Sclences, 81, 251-261.

van tier Hammen, T. (1974). The Pleistocene changes of vegetation and climate in tropical South America. Journal of Biogeography, 1, 3-26.

van der Hammen, T. and Gon~lez, E. (1960). Upper Pleistocene and Holocene climate and vegetation of the 'Sabana de Bogota' (Colombia, South America). Leidse Geologische Mededelingen, 25, 261-315.

van der Hammen, T. and Gonr/dez, E. (1964). A pollen diagram from the Quaternary of the Sabana de Bogoui (Colombia) and its significance for the geology of the northern Andes. Geologie en Mijnbouw, 43, I 13-117

van der Hammen, T., Werner, J.H. and van Dommelen. H. (1973). Palynologicat record of the upheaval of the northern Andes: a study of the Pliocene and Lower Quaternary of the Colombian Eastern Cordillera and the early evolution of its high-Andean biota. Review of Palaeobotany and Palynology, 16, 1--122.

Zagwijn, W.H. (1975). Variations in climate as shown by pollen analysis, especially in the Lower Pleistocene of Europe. In: Wright. A.E. and Moseley. F. (eds), Ice Ages: Ancient and Modern, pp. 137-152.

Zagwijn, W.H. (1985). An outline of the Quaternary stratigraphy of The. Netherlands. Geologie en Mijnbouw, 64, 17-24.

Zagwijn, WH. (1992). The beginning of the ice age in Europe and its major subdivisions. Quaternary Science Reviews, 11,583-591.