Late Pleistocene and Holocene environmental change inferred from the Cocha
Caranga sediment and soil records in the southeastern Ecuadorian Andes

Holger Niemann, Hermann Behling

PII: S0031-0182(09)00046-7
DOI: doi: 10.1016/j.palaeo.2009.02.018
Reference: PALAEO 4943

To appear in: Palaeogeography

Received date: 24 August 2008

Revised date: 30 January 2009
Accepted date: 4 February 2009

Please cite this article as: Niemann, Holger, Behling, Hermann, Late Pleistocene
and Holocene environmental change inferred from the Cocha Caranga sediment and
soil records in the southeastern Ecuadorian Andes, Palaeogeography (2009), doi:
10.1016/j.palaeo.2009.02.018

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
 ACCEPTED MANUSCRIPT

1 Late Pleistocene and Holocene environmental change inferred

2 from the Cocha Caranga sediment and soil records in the

3 southeastern Ecuadorian Andes

PT
4

RI
5 Holger Niemann*, Hermann Behling

SC
6 *corresponding author

NU
7 Department of Palynology and Climate Dynamics, Albrecht-von-Haller-Institute

8 for Plant Sciences,University of Göttingen, Untere Karspüle 2, 37073 Göttingen,

MA
9 Germany

10 E-mail: holnie@web.de
ED

11
PT

12 Abstract

13
CE

14 Late Pleistocene and Holocene vegetation, climate and fire dynamics of

15 mountain forest and paramo ecosystems, as well as human impact, are presented
AC

16 from the Cocha Caranga area, at 2710 m elevation in the Podocarpus National

17 Park, southeastern Ecuadorian Andes. Palaeoenvironmental changes, inferred

18 from two sediment cores and a soil core were investigated by pollen, spore, algae

19 and charcoal analyses.

20 During the transition from late Pleistocene to early Holocene between ca.

21 14,500 to 9700 cal yr BP upper mountain forest vegetation expanded, suggesting

22 increasing temperature and moisture. This expansion abruptly stopped with

23 increasing fires at ca. 9700 cal yr BP and open grassy vegetation became

24 established. The period from ca. 9700 to 1300 cal yr BP of strong fire intensity

1
 ACCEPTED MANUSCRIPT

1 indicate that vegetation components, mainly Weinmannia and Myrica, react

2 sensitively to past, probably human caused fires. During the last few centuries

3 modern vegetation established, characterised by open grassy areas with forest

PT
4 patches and small mires.

5 The green algae Botryococcus braunii, Isoetes and Cyperaceae were used to

RI
6 identify lake level fluctuation to reconstruct Holocene wet/dry phases. Drier

SC
7 climatic conditions occurred from ca. 9700 to 6900 cal yr BP and from ca. 4200 to

8 1300 cal yr BP. From ca. 6900 to 4200 cal yr BP and from ca. 1300 cal yr BP to

9
NU
modern times, wetter climatic conditions occurred.
MA
10

11 Keywords: Ecuador, pollen analysis, mountain forest, fire history, human impact,

12 Botryococcus
ED

13
PT

14 1. Introduction
CE

15 The Ecuadorian Andes harbour the most species rich ecosystems on earth

16 (Barthlott et al., 2005). To study the highly diverse mountain forest and paramo
AC

17 ecosystems in the Podocarpus National Park region, currently more than 25

18 different research groups are working in this area. Extended research has been

19 carried out in the Podocarpus National Park region (Beck et al., 2008). More than

20 130 families, 420 genera and 1200 species of spermatophytes has been identified

21 (Homeier and Werner, 2005). Despite its high biodiversity huge areas have been

22 strongly affected by human activity (e.g., industrial deforestation) during the last

23 decades. Natural vegetation regeneration and sustainable management, as well as

24 conservation of less degraded areas is urgently needed. Palaeoecological work

25 analysing more than 10 different lake, peat and soil cores in the Podocarpus

2
 ACCEPTED MANUSCRIPT

1 National Park region started in 2005. The study of the palaeoenvironment change

2 is needed to understand the natural composition and dynamics of modern

3 ecosystems for proper management and conservation (Froyd and Willis, 2008).

PT
4 Among the few palaeoenvironmental records are available from Podocarpus

5 National Park region, are those from the El Tiro-Pass (2810 m), ca. 10 km

RI
6 northeast of the core site (Niemann and Behling, 2008a), from Laguna Rabadilla

SC
7 de Vaca (3310 m), ca. 25 km south of the core site (Niemann et al., 2009) and

8 from the upper Rio San Francisco valley (ca. 2000-3200 m), ca. 15 km northeast

9
NU
of the core site (Niemann and Behling, 2008b).
MA
10 Neighbouring regions in South America offer archives, for example from the

11 Colombian Cauca Valley (Berrio et al., 2002), from Laguna Loma Linda located

12 in the transition zone between the savannah of the Llanos Orientales and the
ED

13 Amazon rainforest in Colombia (Behling and Hooghiemstra, 2000), from Lake

PT

14 Ayauch in the southeastern part of Ecuadorian Amazonia (Bush and Colinvaux,

15 1988), from Rio Napo in the northwestern Amazon basin of Ecuador (Colinvaux
CE

16 et al., 1988a), from Lake Yambo at the Inter-Andean Plateau of Ecuador

AC

17 (Colinvaux et al., 1988b), from Yasuni National Park in the Ecuadorian part of

18 Amazonia (Weng et al., 2002), from Lake Surucucho in the Las Cajas Nationa

19 Park at the eastern flank of the western Ecuadorian Andes (Colinvaux, 1997),

20 from Laguna Chochos in the eastern Peruvian Andes (Bush et al., 2005), from

21 Huascaran (ice cores) in the central Peruvian Andes (Thompson et al., 2003) and

22 from different lakes of the central Peruvian Andes (Hansen et al., 1994).

23 Locations of greater interest for this study are shown in Figure 1.

24 In this study we want to address following main questions: (1) How did the

25 vegetation develop over the recorded period? (2) Did fire provide an important

26 control over the vegetation and was it human-induced? (3) Is the grassy vegetation
3
 ACCEPTED MANUSCRIPT

1 with forest patches natural or caused by humans? (4) Do the pollen, spore, algae

2 and charcoal records provide a broader regional climate signal than previous

3 records from the area?

PT
4

5 2. Site description

RI
6

SC
7 2.1. Location

9
NU
The Andes of southern Ecuador and northern Peru include the so-called Andean
MA
10 depression (Depression de Giron-Cuenca in Ecuador and Huancabamba in Peru).

11 The highest peaks reach up to about 4000 m elevation; and there are no active

12 volcanoes in the region (Richter and Moreira-Munoz, 2005).

ED

13 The southern Cordillera Real is mainly built up by the "Zamora series",

PT

14 consisting of Palaeozoic metamorphic rocks of widely varying metamorphic

15 grade. Local bedrock is dominated by semipelites, quartzites and black phylites

CE

16 with some granitic intrusions (Litherland et al., 1994).

AC

17 The study area (Fig. 1) is located on the western slope of the eastern Cordillera

18 (Cordillera Real), 8 km southeast of the city of Loja (2200 m), in the Podocarpus

19 National Park.

20 The ca. 2000 m2 in size and 2 m deep Laguna Cocha Caranga (4° 02’ 45.1“S,

21 79° 09’ 34.5“W, 2710 m) is exposed just below a NW-SE orientated mountain

22 ridge, in a small depression. The core site of Cocha Caranga mire is a small bog

23 from ca. 50 m2, directly at the northern shore line of the lake. The core site of

24 Cocha Caranga forest is a forest patch of about 500 m2, southeast and just 10 m

25 below the lake. The three core locations are shown in Figure 2.

26
4
 ACCEPTED MANUSCRIPT

1 2.2. Modern vegetation

3 The core sites from Cocha Caranga are located in the upper mountain forest

PT
4 vegetation zone, below the modern tree line. The surrounding landscape is

5 characterised by grassy vegetation with forest patches. The valleys and slopes

RI
6 below the lake are strongly degraded by pastures. Prehistoric terraces at the lower

SC
7 slopes point to agricultural activity during the past. Following the mountain ridge

8 to the southeast (Fig. 2, background), closed forest vegetation occurs after 500 m.

9
NU
Two different types of paramo ecosystems are identified in the Podocarpus
MA
10 National Park region. The herb-paramo (Paramos herbaceos), located between ca.

11 3200 to 3400 m, is rich in Calamagrostis macrophylla, Neurolepis nana

12 (Poaceae) and Niphogeton dissecta (Apiaceae). The herbs and shrubs grow up
ED

13 from 0.2 to 1 m height. The shrub paramo (Paramos arbustivos bajos), located
PT

14 between ca. 2900 to 3400 m, is rich in Brachyotum andreanum

15 (Melastomataceae), Ilex andicola (Aquifoliaceae), Oxalis spiralis (Oxaliaceae)

CE

16 and Weinmannia rollottii (Cunoniaceae). The shrubs and herbs grow from 0.5 to
AC

17 1.2 m in height (Lozano et al., 2003).

18 At lower elevation between ca. 2750-3100 m, subparamo is present,

19 characterised by Brachyotum rotundifolium (Melastomataceae), Oritrophium

20 peruvianum (Asteraceae) and Puya nitida (Bromeliaceae). The shrubs and herbs

21 grow up to 2-3 m height (Lozano et al., 2003).

22 The modern tree line in northern and central Ecuador is at about 3400-3600 m.

23 At the El Tiro-Pass, ca. 10 km northeast of the core site, the modern tree line is at

24 ca. 2800 m, which is about 600-800 m lower. The shift of the tree line, as well as

25 the vegetation zones to lower elevation is probably a result of the so-called

26 Andean depression (Bader, 2007).

5
 ACCEPTED MANUSCRIPT

1 Two different mountain forest ecosystems occurs in the Podocarpus National

2 Park region. The upper mountain forest (UMF), located between ca. 2100 to 2750

3 m consists of low, monotypic formation, with only one tree stratum between 5-10

PT
4 m; rarely up to 15 m. Characteristic trees include Myrsine andina (Myrsinaceae)

5 Purdiaea nutans (Cyrillaceae) and Weinmannia pinnata (Cunoniaceae). The

RI
6 lower mountain forest (LMF) is located between ca. 1300 to 2150 m with an

SC
7 extremely diverse, two-storied tree stratum, composed of numerous 20-35 m tall

8 tree species. Characteristic species include Alzetea verticillata (Alzataceae),

9
NU
Graffenrieda miconioides (Melastomataceae) and Morus insignis (Moraceae);
MA
10 (Bussmann, 2001, 2005; Lozano et al., 2003).

11

12 2.3. Climate
ED

13
PT

14 The climate in the southeastern Ecuadorian Andes is influenced by warm

15 moisture-laden air from the Amazon lowland, which collides with cold mountain
CE

16 air masses. This produces much of the rainfall in the eastern Andean mountains.
AC

17 The climate of the paramo is classified as humid tropical diurnal with cold nights

18 and cool days, there is a drier season that lasts from December until March

19 (Bosman et al., 1994). As part of the Andean depression, all summits in the

20 southern Ecuadorian Andes are below the snowline.

21 The eastern Andean mountains form a division that separates the moist eastern

22 slopes of the Andes from the dry inner-Andean basins (e.g., the Loja and

23 Catamayo Basin). Between the eastern slopes of the eastern Cordillera and the dry

24 valley of Catamayo, which are only 70 km apart, rainfall rates drop from over
-1
25 4000 to 300 mm yr (Bendix et al., 2004). The mean annual temperature

6
 ACCEPTED MANUSCRIPT

1 increases from Cordillera Real, crest region (<9°) to Loja city (15-19°) (Beck et

2 al., 2008b).

3 The ECSF research area, located in the upper Rio San Francisco valley, ca. 15

PT
4 km northeast of the core site show a strong altitudinal gradient of increasing

5 precipitation, from the valley bottom (1800 m, 2297 mm yr -1) to the crest region

RI
6 (3200 m, 6701 mm yr -1) (Bendix et al., 2008). The average precipitation rate near

SC
7 the El Tiro-Pass (2880 m), ca. 10 km northeast of the core site, is about 3500 mm

8 yr -1. In Cajanuma at the western slope of the eastern Cordillera (3400 m), 20 km

9
NU
south of the core site; it is about 5700 mm yr -1 (Emck, 2007).
MA
10 Laguna Cocha Caranga is located in the transition zone, between the wet

11 western slopes of the eastern Cordillera and the dry Loja basin.

12
ED

13 3. Material and methods

PT

14

15 At Laguna Cocha Caranga a 58 cm long sediment core was recovered with a

CE

16 Livingstone piston-corer near the centre of the lake (ca. 2 m water depth), from an
AC

17 inflatable rubber raft. The core was left in the tube, and stored under dark and cold

18 (+4°C) conditions before processing. The uppermost 26 cm of the soft sediment

19 was recovered with a Kajak-Corer. The peat deposits (Cocha Caranga mire) near

20 the shore line of the lake and forest patch (Cocha Caranga forest) were cored

21 using a Russian Corer. For both core sides central positions were chosen. Sections

22 of 50 cm length were extruded on-site with split PVC tubes and wrapped with

23 plastic film. The location of the three core sites are shown in Figure 2, site

24 specific data are given in Table 1.

25 Six subsamples (bulk, charcoal and soil organic matter) were taken for

26 Accelerator Mass Spectrometer (AMS) radiocarbon dating, performed at the

7
 ACCEPTED MANUSCRIPT

1 University of Erlangen/Nürnberg, Germany. Radiocarbon ages were calibrated

2 with CalPal (Cal Curve 50 ka cal BP to modern) (Weninger et al., 2004).

3 From the sediment core and from the contents of the Kajak-Corer of Laguna

PT
4 Cocha Caranga a total of 34 subsamples (0.25 cm3) were taken for palynological

5 and charcoal analyses, at 2 cm (1 cm) intervals. The overlapping section was fixed

RI
6 at the marked decrease of the algae Botryococcus between 22 and 24 cm core

SC
7 depth. From the sediment core of Cocha Caranga mire and from the soil core of

8 Cocha Caranga forest a total of 31 subsamples (0.25 cm3) were taken at 4 cm

9
NU
intervals. All samples were processed with standard analytical methods (Faegri
MA
10 and Iverson, 1989). Exotic Lycopodium spores were added to each sample before

11 treatment for calculation of pollen, algae and charcoal concentration (grains,

12 particles or individuals/cm3) and for calculation of pollen and charcoal influx

ED

13 (grains or particles/cm2/yr). Approximately 300 pollen grains were counted from

PT

14 each sample.

15 The pollen sum includes trees, shrubs and herbs and excludes Cyperaceae, ferns,
CE

16 ,Isoetes and Sphagnum and the algae Botryococcus. Pollen identification was
AC

17 based on the reference collection of H. Behling, which includes about 3000

18 neotropical species, and on the literature (Behling, 1993; Hooghiemstra, 1984). A

19 reference collection with about 300 species, collected during fieldwork and at the

20 herbarium of the (ECSF) research station is available too.

21 The ecological grouping of the identified pollen taxa into LMF, UMF,

22 subparamo and paramo has been carried out according to available data from

23 literature (Bussmann, 2001, 2005; Homeier and Werner, 2005; Lozano et al.,

24 2003).

25 Pollen and spore data are presented in pollen diagrams as percentages of the

26 pollen sum. Carbonized particles (10-150 µm) and the algae individuals were
8
 ACCEPTED MANUSCRIPT

1 counted on slides and presented as concentration and influx rates. Pollen diagrams

2 were prepared with TILIA, TILIAGRAPH and CONISS (Grimm, 1987). The total

3 number of identified different pollen and spore types is given in Table 1.

PT
4

5 4. Results

RI
6

SC
7 4.1. Stratigraphy

9
NU
The 65 cm long sediment section (the sediment core and the contents of the
MA
10 Kajak-Corer) of Laguna Cocha Caranga consists of light brown to grey silty clay

11 with organic contents (65-28 cm) and decomposed brown organic material with a

12 decreasing content of clay (28-22 cm). The uppermost part of the core consists of
ED

13 dark brown fine detrital mud (22-0 cm). The 69 cm long sediment core of Cocha
PT

14 Caranga mire consists of dark brown decomposed organic rich material (69-38

15 cm) and brown organic material with roots (38-10 cm). The uppermost part of the
CE

16 core consists of light brown peat (10-0 cm). The 49 cm long sediment core of
AC

17 Cocha Caranga forest consists of dark brown decomposed organic rich material

18 (49-26 cm) and brown organic material with roots (26-0 cm).

19

20 4.2. Chronology

21

22 Six AMS radiocarbon dates, performed at the University of Erlangen/Nürnberg,

23 Germany (Table 2) provide the chronological control for the sediment cores of

24 Cocha Caranga. From the soil core of Cocha Caranga forest a sample of soil

25 organic matter was dated. The material was contaminated with recent organic

26 material (decomposed roots and rootlets) and cannot be used for the chronology.
9
 ACCEPTED MANUSCRIPT

1 Dating of fractions of soil organic matter obtained by alkali-acid extraction is

2 promising, but which fraction renders the most accurate 14C dates is still subject to

3 debate (Tonneijck et al., 2006).

PT
4 The AMS date close to the base of the sediment core of Laguna Cocha Caranga

5 (61 cm core depth) indicates late Pleistocene deposits. Extrapolation shows an age

RI
6 of ca. 14,500 cal yr BP. The AMS date close to the base of the sediment core of

SC
7 Cocha Caranga mire (67 cm core depth) indicates early Holocene deposits. The

8 extrapolation gives an age of ca. 1550 cal yr BP. The AMS date of the soil core of

9
NU
Cocha Caranga forest reflects early Holocene deposits, older than ca. 150 cal yr
MA
10 BP.

11 The series of AMS dates, shown in Figure 3, reflects a consistent age-depth-

12 model of the sediment records of Laguna Cocha Caranga and Cocha Caranga
ED

13 mire. The abrupt environmental change at ca. 1300 cal yr BP, increasing
PT

14 Botryococcus and Melastomataceae and decreasing ferns and fire intensity,

15 observed in the record of Laguna Cocha Caranga conducts similar to the same
CE

16 change at ca. 250 cal yr BP in the record of Cocha Caranga mire. A gap in the
AC

17 record of Laguna Cocha Caranga cannot be excluded finally.

18 The accumulation rates (years/cm) are given for the sediment records of Laguna

19 Cocha Caranga (Fig. 4) and Cocha Caranga mire (Fig. 6), indicating, that

20 sediments accumulated continuously. The average sediment accumulation rate of

21 the record of Laguna Cocha Caranga is 0.0484 mm yr -1, in detail it is 0.1917 mm

-1 -1
22 yr (modern to 1149 cal yr BP), 0.0451 mm yr (1149 to 8256 cal yr BP) and
-1
23 0.0140 mm yr (8256 to 12569 cal yr BP). The average sediment accumulation

24 rate of the record Cocha Caranga mire is 0.4268 mm yr -1, in detail is 1.2286 mm

25 yr -1 (modern to 301 cal yr BP) and 0.1967 mm yr -1 (301 to 1523 cal yr BP).

26
10
 ACCEPTED MANUSCRIPT

1 4.3. Description of the pollen diagram of Laguna Cocha Caranga (CL) (Figs. 4

2 and 5)

PT
4 Zone CL-1 (65-57 cm, ca. 14,500-9700 cal yr BP, 4 subsamples) shows highest

5 percentages of upper mountain forest (UMF) pollen taxa, such as Weinmannia

RI
6 (15-48%), Hedyosmum (4-15%), Ilex and Myrsine (both 3-6%) and lowest

SC
7 percentages of paramo pollen taxa, especially Poaceae (5-15%). Subparamo

8 pollen taxa, mainly Melastomataceae (5-9%) are stable represented. Pollen of

9
NU
Cyperaceae (3-5%) are relatively rare. Isoetes show highest values (up to 100%)
MA
10 and fern spores decreases from 37 to 6% in this zone. Botryococcus concentration

11 (50,000-500,000 individuals/cm3) is low; pollen concentration (200,000-7.5

12 million grains/cm3) is highest and charcoal concentration (0.1-2.5 million

ED

13 particles/cm3) increase during this period. Pollen influx (1000-10,000

PT

14 grains/cm2/yr) increase and charcoal influx (<5000 particles/cm2/yr) is lowest in

15 this zone.
CE

16 Zone CL-2 (57-49 cm, ca. 9700-6900 cal yr BP, 4 subsamples) is characterised
AC

17 by a decrease of UMF pollen taxa, represented by Weinmannia (11-15%), Myrica,

18 Hedyosmum, Myrsine and Ilex (all 3-8%). Paramo pollen taxa, especially Poaceae

19 rise from 10 to 38%, as well as pollen from Cyperaceae (from 10 to 32%). Isoetes

20 is well represented (6-15%) in this zone and fern spores (15-25%) are relatively

21 frequent. Botryococcus concentration (50,000-500,000 individuals/cm3) and

22 pollen concentration (0.3-1.5 million grains/cm3) is relatively low, charcoal

23 concentration (3.3-7 million particles/cm3) is high during this period, as well as

24 pollen influx (2000-45000 grains/cm2/yr) and charcoal influx (8000-32,000

25 particles/cm2/yr).

11
 ACCEPTED MANUSCRIPT

1 Zone CL-3 (49-37 cm, ca. 6900-4200 cal yr BP, 6 subsamples) is marked by

2 relatively stable amounts of mountain forest and (sub)-paramo pollen taxa,

3 represented by Poaceae (25-30%), Weinmannia (9-18%), Myrica (5-25%),

PT
4 Melastomataceae, Moraceae/Urticaceae, Myrsine, Ilex (all 3-6%) and

5 Podocarpaceae (1-6%). Pollen of Cyperaceae (20-40%) are frequent. Isoetes (4-

RI
6 12%) show lower amounts as the preceding zone and fern spores (25-45%) shows

SC
7 highest representation during this period. Botryococcus concentration (0.5-2.5

8 million individuals/cm3) is highest, pollen concentration (<500,000 grains/cm3) is

9
NU
low and charcoal concentration (1.8-6.5 million particles/cm3) is very high in this

zone. Pollen influx (1000-2500 grains/cm2/yr) is low and charcoal influx (7000-
MA
10

11 30,000 particles/cm2/yr) is high.

12 Zone CL-4 (37-22.5 cm, ca. 4200-1300 cal yr BP, 8 subsamples) shows an
ED

13 increase of mountain forest pollen taxa, such as Weinmannia (from 9 to 25%),

PT

14 Moraceae/Urticaceae (3-6%), Hedyosmum, Myrsine, Alnus and Podocarpaceae

15 (all 2-5%). Myrica (6-28%) reach highest amounts during this period. Poaceae
CE

16 pollen decrease from 40 to 15%. Pollen of Cyperaceae (25-68%) show highest

AC

17 values in this zone. Isoetes (2-9%) and fern spores (8-27%) shows lower amounts

18 as the preceding zone. Botryococcus concentration (0.05-1 million

19 individuals/cm3) decrease, pollen concentration (<500,000 grains/cm3) is low and

20 charcoal concentration (1.2-5 million particles/cm3) is very high. Pollen influx

21 (1000-2500 grains/cm2/yr) is low and charcoal influx (9000-25,000

22 particles/cm2/yr) is high during this period.

23 Zone CL-5 (22.5-0 cm, ca. 1300 cal yr BP to modern times, 12 subsamples) is

24 marked by an increase of subparamo taxa, represented by Melastomataceae (7-

25 17%), Clethra (3-6%) and Asteraceae (2-5%). Mountain forest pollen taxa, such

26 as Weinmannia (20-30%), Alnus (3-10%), Moraceae/Urticaceae (6-8%),

12
 ACCEPTED MANUSCRIPT

1 Hedyosmum (1-7%), Myrica and Myrsine (both 3-6%) shows relatively stable

2 values. Poaceae pollen (12-20%) are well represented. Pollen of Cyperaceae (5-

3 25%) decrease and Sphagnum spores (3-10%) show highest amounts in this zone.

PT
4 Isoetes is nearly absent now and fern spores (3-10%) shows low amounts.

5 Botryococcus concentration (250,000-900,000 individuals/cm3) is high, pollen

RI
6 concentration (<100,000 grains/cm3) and charcoal concentration (<300,000

SC
7 particles/cm3) is lowest during this period. Pollen influx (ca. 1000 grains/cm2/yr)

8 and charcoal influx (ca. 5000 particles/cm2/yr) is very low.

9
NU
MA
10 4.4. Description of the pollen diagram of Cocha Caranga mire (CM) (Figs. 6 and

11 7)

12
ED

13 Zone CM-1 (68-33 cm, ca. 1550-260 cal yr BP, 10 subsamples) shows a
PT

14 relatively stable percentages of mountain forest and (sub)-paramo pollen taxa,

15 such as Poaceae (28-33%), Weinmannia (12-15%), Asteraceae (5-10%),

CE

16 Melastomataceae, Moraceae/Urticaceae, Myrica (all 4-8%), Hedyosmum (5-7%),

AC

17 Alnus (1-5%) and Bomarea (0-5%). Cyperaceae pollen (8-12%) and Sphagnum

18 spores (3-6%) show low amounts. Fern spores decreases from 32 to 8%. Pollen

19 concentration (40,000-90,000 grains/cm3) and charcoal concentration (2-4.8

20 million particles/cm3) is highest during this period. Pollen influx (600-6000

21 grains/cm2/yr) is lowest and charcoal influx (50,000 220,000 particles/cm2/yr) is

22 relatively high.

23 Zone CM-2 (33-0 cm, ca. 260 cal yr BP to modern times, 8 subsamples) is also

24 marked by relatively stable representation of mountain forest and subparamo taxa,

25 such as Weinmannia (12-25%), Melastomataceae (7-10%), Moraceae/Urticaceae

26 (5-8%), Myrica (4-10%), Asteraceae (1-10%), Hedyosmum, Alnus (both 3-6%),

13
 ACCEPTED MANUSCRIPT

1 Clethra and Myrsine (both 2-5%). Paramo pollen taxa, especially Poaceae (20-

2 38%) show high amounts in the lowermost part of the zone. Pollen of Cyperaceae

3 (7-25%) and Sphagnum spores (4-12%) show highest amounts. Fern spores

PT
4 increases in the uppermost part of the zone, especially Lycopodium curvatum (up

5 to 15%). Botryococcus concentration (<10,000-1.2 million individuals/cm3) and

RI
6 pollen concentration (5000-60,000 grains/cm3) decrease and charcoal

SC
7 concentration (<30,000 particles/cm3) is lowest during this period. Pollen influx

8 (1000-9000 grains/cm2/yr) decrease and charcoal influx (<30,000

9 particles/cm2/yr) is very low.

NU
MA
10

11 4.5. Description of the pollen diagram of Cocha Coranga forest (CF) (Fig. 8)

12
ED

13 Zone CF-1 (26-48 cm, older ca. 150 cal yr BP, 6 subsamples) is marked by
PT

14 increasing values of UMF pollen taxa, especially Weinmannia rise from 20 to

15 35%. Other mountain forest and subparamo taxa, mainly Hedyosmum (7-10%),
CE

16 Melastomataceae, Asteraceae (both 5-10%), Myrica (3-7%) and

AC

17 Moraceae/Urticaceae (3-6%) are relatively frequent during this period. Pollen of

18 Poaceae (12-26%) show high amounts. Fern spores (8-25%) show highest values

19 in this zone. Pollen concentration is relatively high (100,000-250,000 grains/cm3)

20 and charcoal concentration (2-7.5 million particles/cm3) is highest during this

21 period.

22 Zone CF-2 (26-0 cm, younger ca. 150 cal yr BP, 7 subsamples) is characterised

23 by an increase of UMF pollen taxa, Hedyosmum strong increase from 8 to 80%,

24 Weinmannia strong decrease from 55 to 1%. Myrica (5-28%), Ilex (1-13%) and

25 Myrsine (1-7%) are relatively frequent during this period, as well as subparamo

26 pollen taxa of Melastomataceae (5-20%), Clethra (1-7%) and Asteraceae (1-5%).

14
 ACCEPTED MANUSCRIPT

1 Poaceae (4-10%). Fern spores (1-13%) show lowest values in this zone. Pollen

2 concentration is high (100,000-350,000 grains/cm3) and charcoal concentration

3 (<0.01-1.2 million particles/cm3) is low during this period.

PT
4

5 5. Interpretation and discussion

RI
6

SC
7 5.1. Late Pleistocene and Holocene vegetation and climate development

9
NU
Palaeoenvironmental changes were investigated from three pollen, spore, algae
MA
10 and charcoal records, cored in a small circle from less than 100 m at the Cocha

11 Caranga area. The 65 cm long sediment record of Laguna Cocha Caranga at 2710

12 m elevation (Figs. 4 and 5) shows an extrapolated age of ca. 14,500 cal yr BP,
ED

13 which probably reflects the beginning of sediment accumulation. High

PT

14 percentages of Isoetes indicate that the small basin was filled with shallow water.

15 The reason for the establishment of the lake may be the result of a local landslide,
CE

16 a remnant of a glacier can be excluded. Glaciers in the Podocarpus National Park

AC

17 region reach up to ca. 2750-2800 m (Rozsypal, 2000). At the El Tiro-Pass (2810

18 m), ca. 10 km northeast of the core site, the bottom of the sediment core shows an

19 age of ca. 20,100 cal yr BP, indicating that glaciers were not present at this

20 elevation since that time (Niemann and Behling, 2008a).

21 During the late Pleistocene to early Holocene period (Zone CL-1, ca. 14,500-

22 9700 cal yr BP) upper mountain forest taxa, mainly Weinmannia, strongly

23 increased, reflecting a rise in temperature and a shift of vegetation zones, as

24 well as the tree line into higher elevations. The strong decrease of Isoetes

25 indicates a lake level rise, coupled with higher precipitation at Laguna Cocha

26 Caranga. At the El Tiro-Pass (2810 m) mountain forest taxa, as well as ferns

15
 ACCEPTED MANUSCRIPT

1 became frequent during the period from ca. 11,200 to 8900 cal yr BP,

2 reflecting higher temperatures and moisture (Niemann and Behling, 2008a).

3 Fossil pollen data from the central Peruvian Andes (4000 m) point to

PT
4 increasing moisture, as well as higher temperatures from about 12,910 to 7850

5 cal yr BP (Hansen et al., 1994). The pollen record of Lake Surucucho in the

RI
6 Las Cajas National Park (3200 m) indicates an establishment of mountain

SC
7 rainforest vegetation, due to an increase of Weinmannia and Hedyosmum at ca.

8 12,260 cal yr BP (Colinvaux et al., 1997). At Laguna Chochos (3285 m), in the

9
NU
eastern Peruvian Andes, a warm and wet early Holocene between ca. 11.500 to
MA
10 9500 cal yr BP was estimated (Bush et al., 2005).

11 The period from early to mid Holocene (Zone CL-2, ca. 9700-6900 cal yr BP)

12 is characterised by a decrease of arboreal taxa, mainly Weinmannia and

ED

13 Melastomataceae. The expansion of Poaceae and ferns reflects a response of the

PT

14 vegetation to a marked increase of fires. At the El Tiro-Pass (2810 m), fires

15 increased at ca. 7500 cal yr BP (Niemann and Behling, 2008a). At Laguna

CE

16 Chochos (3285 m) fires were rare before ca. 11,500 cal yr BP (Bush et al., 2005).
AC

17 Cyperaceae, an indicator for marshy lake shores and Isoetes, an indicator for

18 shallow water reflects a low water level during this period. The absence of roots

19 in the sediment indicate that a minimum water body probably left. The low

20 concentration of the green algae Botryococcus supports this interpretation.

21 Botryococcus braunii usually thrives in freshwater bogs, temporary pools, ponds,

22 and lakes (Batten and Grenfell, 1996). Palynomorph studies from Itapeva Lake,

23 southern Brazil show a marked decrease of algae (e.g. Botryococcus) and higher

24 amounts of Cyperaceae and Poaceae probably indicating a drop of the water

25 column (Bohns-Meyer, 2005). High percentages of lake shore vegetation (e.g.

26 Cyperaceae) reflect low water depths at Tulare Lake in California, relative to the
16
 ACCEPTED MANUSCRIPT

1 littoral indicators. The colonial green algae Botryococcus is used as an indicator

2 of open water at the coring site (Davis, 1999).

3 The early to mid Holocene period from ca. 9700 to 6900 cal yr BP is interpreted

PT
4 as a drier period (Fig. 9). High values of Cyperaceae and Isoetes coupled with low

5 concentration of Botryococcus reflecting a lower lake level, caused by a reduced

RI
6 precipitation. XRF data from Laguna Rabadilla de Vaca (3310 m), ca. 25 km

SC
7 south of the core site, indicates a drier period lasting from ca. 8990 to 6380 cal yr

8 BP (Niemann et al., 2009). At the El Tiro-Pass (2810 m) an upper mountain forest

9
NU
established during the mid Holocene period from ca. 8900 to 3300 cal yr BP,
MA
10 reflecting warmer climatic conditions (Niemann and Behling, 2008a). At Laguna

11 Chochos (3285 m) a warm and wet early Holocene was interrupted by a warm-dry

12 event that lasted from ca. 9500 to 7300 cal yr BP (Bush et al., 2005). The pollen
ED

13 record from Laguna Loma Linda (310 m elevation) in the Colombian savannah,
PT

14 shows a significantly lower but seasonal stronger precipitation during the period

15 from ca. 9650 to 6850 cal yr BP than present-day (Behling and Hooghiemstra,
CE

16 2000). Data from Colombian Cauca Valley (1020 m) indicates a dry maximum at
AC

17 ca. 8300 cal yr BP, when dry forest reached maximum expansion (Berrio et al.,

18 2002).

19 During the mid Holocene (Zone CL-3, ca. 6900-4200 cal yr BP) upper

20 mountain forest vegetation expanded little and Poaceae decreased, as a reaction of

21 lower fire intensity during this period. The mid Holocene period from ca. 6900 to

22 4200 cal yr BP is interpreted as a wetter period (Fig. 9). An increase of ferns and

23 the high concentration of Botryococcus coupled with lower values of Cyperaceae

24 and Isoetes indicates a lake level rise, caused by enhanced precipitation. The

25 increasing concentration of Botryococcus suggests an enlarging of the water body;

26 the marshy lake shores are flooded now. XRF data from Laguna Rabadilla de
17
 ACCEPTED MANUSCRIPT

1 Vaca (3310 m) indicate a warm and wetter period between ca. 6380 to 3680 cal yr

2 BP (Niemann et al., 2009). At Laguna Loma Linda rainforest expanded markedly

3 during the period from ca. 6850 to 3900 cal yr BP, reflecting higher precipitation

PT
4 (Behling and Hooghiemstra, 2000). The pollen record of the Yasuni National Park

5 (220 m elevation) in Ecuadorian Amazonia, shows wet climatic conditions from

RI
6 ca. 5800-4900 cal yr BP (Weng et al., 2002).

SC
7 During the mid to late Holocene period (Zone CL-4, ca. 4200-1300 cal yr BP),

8 upper mountain forest vegetation, especially Myrica, expanded. The strong

9
NU
increase of Myrica may be a result of human influence (discussed later). The mid
MA
10 to late Holocene period from ca. 4200 to 1300 cal yr BP is interpreted as a drier

11 period (Fig. 9). The decrease of ferns and the lower concentration of Botryococcus

12 coupled with a strong increase of Cyperaceae indicates a lake level drop, due to a
ED

13 reduced precipitation. The water body declined, but marshy lake shores prevailed.
PT

14 At Laguna Rabadilla de Vaca (3310 m), a Poaceae dominated herb-paramo

15 occurred from ca. 3680 cal yr BP until modern times, reflecting cooler climatic
CE

16 conditions relative to the mid Holocene period. XRF- and charcoal data indicates
AC

17 a decline in precipitation after ca. 3680 cal yr BP (Niemann et al., 2009). At the El

18 Tiro-Pass (2810 m) modern subparamo vegetation became established after ca.

19 3300 cal yr BP, indicating a shift of vegetation zones into lower elevation, due to

20 cooler climatic conditions during the late Holocene period (Niemann and Behling,

21 2008a). At Lake Aricota (2800 m), in the central Peruvian Andes, maximum

22 Holocene lake level was attained before ca. 2800 cal yr BP. Moderately high lake

23 levels occurred at ca. 1600 and ca. 1230 cal yr BP (Placzek and Quade, 2001). In

24 opposite to a dryer late Holocene, the pollen record from Laguna Loma Linda

25 shows a continued expansion of rainforest from ca. 3900-2300 cal yr BP.

26 Precipitation was still increasing and the length of the annual dry period possibly
18
 ACCEPTED MANUSCRIPT

1 shortened (Behling and Hooghiemstra, 2000). At the Yasuni National Park wetter

2 climatic conditions occurred from ca. 3700-1000 cal yr BP (Weng et al., 2002).

3 During the late Holocene period (Zone CL-5, ca. 1300 cal yr BP to recent) the

PT
4 vegetation is characterised by open grassy areas with forest patches. Peat bogs

5 established around Laguna Cocha Caranga. Weinmannia, Melastomataceae,

RI
6 Moraceae/Urticaceae, Myrsine and Clethra became more frequent. Myrica and

SC
7 Poaceae were less frequent, probably as a result of the lower fire intensity. This

8 may indicate a reduction of human activity and/or wetter climatic conditions at

9
NU
this elevation, confirmed by an increase of Alnus. Alnus occurs particularly on
MA
10 marshy grounds, covers large wetland areas, grows along river beds and follows

11 landslides as a pioneer (Marchant et al., 2002). In the upper Rio San Francisco

12 valley (ca. 2000-3200 m), about 15 km northeast of the core site, the forest started
ED

13 to regenerate after the decrease of fires between ca. 970 to 400 cal yr BP
PT

14 (Niemann and Behling, 2008b).

15 The late Holocene period from ca. 1300 cal yr BP to recent is interpreted as a
CE

16 wetter period (Fig. 9), modern vegetation, as well as modern climatic conditions
AC

17 become established. The occurrence of the moss Sphagnum, reflects the formation

18 of a mire near the shore line of Laguna Cocha Caranga. An abrupt increase of

19 Botryococcus concentration and a strong decrease of Cyperaceae and Isoetes

20 reflect a rapid flooding of marshy lake shores. Laguna Cocha Caranga reached a

21 lake level close to modern time at ca. 1300 cal yr BP. Palaeoenviremental data

22 from Laguna Zurita (2590 m elevation) in the upper Rio San Francisco valley

23 show a decrease of Isoetes and Cypreaceae, indicating a flooding of marshy lake

24 shores between ca. 1350 and 790 cal yr BP. After ca. 790 cal yr BP the modern

25 lake level was reached (Niemann and Behling, 2008b). Pollen data from three old

26 river channel lakes (Rio Napo, ca. 300 m elevation) in the western Amazon basin
19
 ACCEPTED MANUSCRIPT

1 of Ecuador shows a regional flooding from ca. 700-1200 AD (Colinvaux et al.,

2 1988a). Pollen data from the endorheic Lake Ayauch (500 m elevation) in the

3 Inter-Andean Plateau of Ecuador reveal local evidence for a moist episode from

PT
4 ca. 400 to 1200 AD (Colinvaux et al., 1988b).

RI
6 The 69 cm long sediment core of Cocha Caranga mire at 2710 m elevation

SC
7 (Figs. 6 and 7) represents late Holocene peat bog deposits with an extrapolated

8 age of ca. 1550 cal yr BP. The lower part of the record (Zone CM-1, ca. 1550-

9
NU
250 cal yr BP) in general shows no marked changes except higher occurrence of
MA
10 (sub)-paramo taxa, especially Asteraceae and Bomarea, as well as ferns. In the

11 upper part of the record (Zone CM-2, ca. 250 cal yr BP to recent) Weinmannia

12 and Melastomataceae increases and Poaceae decrease, probably as a result of the

ED

13 lower fire intensity. The abrupt increase of Botryococcus at ca. 250 cal yr BP
PT

14 suggests that the lake level raised, the mire became flooded, probably episodic,

15 during months with higher precipitation. However, it has to be considered that

CE

16 Botryococcus probably occurs in freshwater bogs; otherwise Botryococcus

AC

17 flushed into the mire due to the higher lake level.

18

19 The 49 cm long soil core of Cocha Caranga forest at 2700 m elevation (Fig. 8)

20 represents late Holocene deposits of a forest patch. The lower part of the record

21 (Zone CF-1, probably older than 150 cal yr BP) show a higher occurrence of

22 Poaceae and fern taxa. The higher fire frequency may be responsible for that.

23 After the decrease of fires the forest patch started to expand (Zone CF-2,

24 probably younger than 150 cal yr BP) with varying frequencies of Weinmannia,

25 Melastomataceae, Hedyosmum and Myrica. Weinmannia reacts first and is highly

26 sensitive to changing fire intensity. This is also observed in the pollen record of
20
 ACCEPTED MANUSCRIPT

1 Laguna Cocha Caranga during the transition from late Pleistocene to early

2 Holocene. The pollen records of the upper Rio San Francisco valley at ca. 2000-

3 3200 m (Niemann and Behling, 2008b) and the record of Laguna Rabadilla de

PT
4 Vaca at 3310 m (Niemann et al., 2009) show an early reaction of Weinmannia to

5 decreasing, as well as increasing fires. A natural development of the environment

RI
6 around Laguna Cocha Caranga seems to be improbable, due to human activities,

SC
7 pastures in modern times and agricultural activities during the past (terraces) at

8 the slopes below. In particular, at an elevation of ca. 2710 m a closed forest

9 ecosystem has to be considered.

NU
MA
10 At the El Tiro-Pass, ca. 10 km northeast of the core site, the modern tree line is

11 at ca. 2800 m (Bader, 2007). The question, at which elevation the natural tree line

12 in the Podocarpus National Park area was during the past cannot be answered.
ED

13 Pollen analysis from Pantano de Pecho (3870 m) in the central Ecuadorian paramo
PT

14 documents altitudinal migrations of the upper forest line prior to deforestation.

15 The upper forest line moved upslope from a minimum between 3400 to 3500 m
CE

16 during the period from 1290 to 1315 AD, to the highest level of 3750 m (1640-
AC

17 1765 AD). Palynological and ecological evidence do not support the hypothesis

18 that the natural forest line in central Ecuador lies between 4100 and 4350 m

19 elevation, but has been lowered as a result of frequent burning and other human

20 impact (Wille et al., 2002). According to Bakker et al. (2008) human-induced fires

21 and deforestation during the long occupation history of the Central Valley in

22 Ecuador have caused a downslope shift of the upper forest line and have given

23 way to a downslope expansion of páramo vegetation. More recently, montane

24 forests and lower areas of paramo have been replaced to a large extent by

25 agricultural land.

26
21
 ACCEPTED MANUSCRIPT

1 5.2. Human impact

3 The charcoal record of Laguna Cocha Caranga shows a marked increase of fire

PT
4 intensity during the period from ca. 9700 to 1300 cal yr BP. An abrupt decrease of

5 arboreal taxa, coupled with high values of Poaceae and ferns taxa reflects the

RI
6 reaction of the vegetation of increasing fires. The charcoal record in comparison

SC
7 with the wet/dry phases, illustrated in Figure 9, show nearly no reaction of higher

8 or lower precipitation during this more than 8000 years lasting period. This

9
NU
probably explains, that human impact is responsible for the strong fires. It is
MA
10 possible that fires at the slopes below Laguna Cocha Caranga originate from

11 anthropogenic activities, probably caused by hunting activities in the savannah or

12 dry forest areas of the nearby dry Loja basin. At the El Tiro-Pass (2810 m) fire
ED

13 intensity increased at ca. 7500 cal yr BP (Niemann and Behling, 2008a). In the
PT

14 Sabana de Bogota (Colombia) the presence of Amerindians could be established

15 from ca. 14,800 cal yr BP onward and possibly even before that time (Van der
CE

16 Hammen, 1978).
AC

17 Between ca. 4800 and 1300 cal yr BP Myrica expanded, probably indicating

18 settling activities in the valleys below Laguna Cocha Caranga. Myrica occurs in

19 degenerated areas and along trails, reflecting that this plant accompanies settling

20 activities. Due to the low mean growth height; Myrica is not common in closed

21 forests (Homeier, personal communication). According to Marchant et al. (2002),

22 Myrica also occurs on wet soils (e.g. swamp areas), but it has to be considered

23 that the increase of Myrica coincides with a drier period. First human activity in

24 the region of Loja, evident by ceramic fragments, has been dated around 4000 cal

25 yr BP. Since then the native Palta culture established around Loja and Zamorra

22
 ACCEPTED MANUSCRIPT

1 (Guffroy, 2004). At the El Tiro-Pass (2810 m), fires strongly increased at ca. 3500

2 cal yr BP (Niemann and Behling, 2008a).

3 The decrease of Myrica and Poaceae, coupled with an increase of e.g.

PT
4 Weinmannia and Melastomataceae, as well as a lower fire intensity after ca. 1300

5 cal yr BP probably reflects a reduction of human activity around Laguna Cocha

RI
6 Caranga. The reduction of human activities may be a result of civil conflicts

SC
7 during pre Inca times, the resettling politics of the Inca and the arrival of the

8 Spanish Conquest. The pollen records from the upper Rio San Francisco valley

9
NU
(2000-3200 m elevation) show a decrease of Poaceae and the crop taxa Zea mays
MA
10 from ca. 970 cal yr BP onwards. The forest started to regenerate since that time.

11 Coupled with a decrease in fire intensity, a reduction of human activities could be

12 assumed (Niemann and Behling, 2008b). Data from the Colombian Cauca Valley
ED

13 (1020 m elevation) show a reduction of human activity after ca. 870 cal yr BP.
PT

14 The expansion of secondary forest taxa may indicate the depopulation and

15 abandonment of previously cultivated land (Berrio et al., 2002). The pollen record
CE

16 from Lake Ayauch (500 m) shows Zea mays cultivation between ca. 2960 and ca.
AC

17 720 cal yr BP. The abandonment of corn cultivation may be a result of social

18 unrest (Bush and Colinvaux, 1988). In the 8th century, indigenous population

19 came to conflict in the La Toma valley, about 10 km north of Loja (Guffroy,

20 2006). In the middle of the 15th century the Inca occupied Ecuador, they defeated

21 the indigenous chiefdom of Palta. It was the strategy of the Inca to settle defeated

22 populations in other regions of their empire, which occurred as well with the Palta

23 (Alvaredo, 2002). After the almost 70 year rule (1463–1531 A.D) in southern

24 Ecuador the Inca were defeated by the Spanish Conquest (Pohle, 2008). The

25 Europeans brought Old World diseases to the New World, including Ecuador;

26 responsible for the strong reduction of the indigenous population (Alchon, 1991).
23
 ACCEPTED MANUSCRIPT

2 5. Conclusions

PT
4 The late Pleistocene to early Holocene expansion of mountain forest vegetation

5 at Laguna Cocha Caranga was interrupted by increasing fires at ca. 9700 cal yr

RI
6 BP. The high fire intensity from ca. 9700 to 1300 cal yr BP strongly influenced

SC
7 past vegetation development. During this more than 8000 years lasting period,

8 open grassy vegetation occurred. This was probably a period of stronger human

9
NU
impact, suggested due to a limited reaction of the charcoal record to the estimated
MA
10 wet/dry phases.

11 The late Holocene development of forest patches probably depends on

12 decreasing fire intensity and or higher precipitation rates. Due to the strong human
ED

13 activities at the slopes below Laguna Cocha Caranga, a natural development

PT

14 seems to be unlikely. In particular, at an elevation of ca. 2710 m a closed forest

15 ecosystem should be considered.

CE

16 The identified lake level fluctuations were used to reconstruct Holocene wet/dry
AC

17 phases at Laguna Cocha Caranga, The late Pleistocene and Holocene climate

18 development in general corresponds to the previous studies of the Podocarpus

19 National Park region (Laguna Rabadilla de Vaca, El Tiro-Pass and upper Rio San

20 Francisco valley). A broader regional climate signal can be assumed in

21 comparison to other palaeoenviremental records of tropical South America.

22

23 Acknowledgements

24 Corinna Brunschön and Fernando Rodriguez are thanked for help during the

25 field work, as well as Felix Matt (research station leader) for his logistical support

24
 ACCEPTED MANUSCRIPT

1 and for his information about the study region. The project FOR 402/D1

2 (Vegetation-, climate- and fire dynamics in the Podocarpus National Park region)

3 is kindly funded by the Deutsche Forschungsgemeinschaft (DFG).

PT
4

5 References

RI
6

SC
7 Alvaredo, P.J., 2002. Historia de Loja y su Provincia. Municipio de Loja.

8 Industria grafica Senefelder.

9
NU
Alchon, S.A., 1991. Native society and disease in colonial Ecuador. Cambridge
MA
10 University Press, Cambridge, England.

11 Bader, M.Y., 2007. Tropical alpine treelines; how ecological processes control

12 vegetation patterning and dynamics. PhD Thesis, Wageningen University,

ED

13 Wageningen, the Netherlands.

PT

14 Bakker, C., Moscol Olivera, M., Hooghiemstra, H., 2008. Holocene

15 environmental change at the upper forest line in northern Ecuador. The

CE

16 Holocene 18 (6), 1-17

AC

17 Barthlott, W., Mutke, J., Rafiqpoor, M.D., Kier, G., Kreft, H., 2005. Global

18 centres of vascular plant diversity. Nova Acta Leopoldina 92, 61–83.

19 Batten, D.J., Grenfel, H.R., 1996. Botryococccus. In: Jansonius, J., MacGregor,

20 D.C. (Eds.), Palynology: Principles and Applications, vol. 1. American

21 Association of Stratigraphic Palynologists Foundation, pp. 205–214.

22 Beck, E., Bendix, J., Kottk,e I., Makeschin, F., Mosandl, R., 2008a. Gradients in a

23 Tropical Mountain Ecosystem of Ecuador. Ecological Studies 198, Springer

24 Verlag, Berlin, Heidelberg.

25 Beck, E., Makeschin, F., Haubrich, F., Richter, M., Bendix, J., Valerezo, C.,

26 2008b. The Ecosystem (Reserva Biológica San Francisco). In: Beck, E.,
25
 ACCEPTED MANUSCRIPT

1 Bendix, J., Kottke, I., Makeschin, F., Mosandl, R. (Eds.), Gradients in a

2 Tropical Mountain Ecosystem of Ecuador. Ecological Studies 198, Springer

3 Verlag, Berlin, Heidelberg, pp. 1-14.

PT
4 Behling, H., 1993. Untersuchungen zur spätpleistozänen und holozänen

5 Vegetations- und Klimageschichte der tropischen Küstenwälder und der

RI
6 Araukarienwälder in Santa Catarina (Südbrasilien). Dissertationes Botanicae

SC
7 206, Cramer, Berlin, Stuttgart.

8 Behling, H., Hooghiemstra, H., 2000. Holocene Amazon rain forest - savanna

9
NU
dynamics and climatic implications: high resolution pollen record Laguna
MA
10 Loma Linda in eastern Colombia. Journal of Quaternary Science 15, 687-695.

11 Bendix, J., Fabian, P., Rollenbeck, R., 2004. Gradients of fog and rain in a

12 tropical montane cloud forest of southern Ecuador and its chemical

ED

13 composition. Proceedings 3rd Int. Conference on Fog, Fog Collection and

PT

14 Dew, 11-15 Oct. 2004, Cape Town, South Africa.

15 Bendix, J., Rollenbeck, R., Richter, M., Fabian, P., Emck, P., 2008. Climate. In:
CE

16 Beck, E., Bendix, J., Kottke, I., Makeschi, F., Mosand, R., (Eds.), Gradients in
AC

17 a Tropical Mountain Ecosystem of Ecuador. Ecological Studies 198, Springer

18 Verlag, Berlin, Heidelberg, pp 63-74.

19 Berrio, J.C., Hooghiemstra, H., Marchant, R., Rangel, O., 2002. Late-glacial and

20 Holocene history of the dry forest area in the south Colombian Cauca Valley.

21 Journal of Quaternary Science 17, 667-682.

22 Bohns-Meyer, K.E., Reichhart, K., Ashraf, A.R., Marques-Toigod, M.,

23 Mosbrugger, V., 2005. Holocene evolution of Itapeva Lake, Rio Grande do

24 Sul, Brazil: Palynomorphs Corg, N, and S records. Journal of South American

25 Earth Sciences 19, 181–192.

26
 ACCEPTED MANUSCRIPT

1 Bosmann, A.F., Hooghiemstra, H., Cleef, A.M., 1994. Holocene mire

2 development and climatic change from a high Andean Plantago rigida cushion

3 mire. The Holocene 43, 233-243.

PT
4 Bush, M.B., Hansen, B.C.S., Rodbell, D.T., Seltzer, G.O., Young, K.R., Leon, B.,

5 Abbott, M.B., Silman, M.R., Gosling, W.D., 2005. A 17 000-year history of

RI
6 Andean climate and vegetation change from Laguna de Chochos, Peru.

SC
7 Journal of Quaternary Science 20, 703-714.

8 Bush, M.B., Colinvaux, P.A., 1988. A 7000-year pollen record from the Amazon

9
NU
lowlands, Ecuador. Vegetatio 76, 141-154.
MA
10 Bussmann, R.W., 2001. The montane forests of Reserva Biologica San Francisco

11 (Zamora-Chinchipe, Ecuador). Vegetation zonation and natural regeneration.

12 Erde 132, 9-25.

ED

13 Bussmann, R.W., 2005. Bosques andinos del sur de Ecuador, clasificación,

PT

14 regeneración y uso. Revistas Peruviana Biologica. 12, 203-216.

15 Colinvaux, P.A., Bush, M.B., Steinitz- Kannan, M., Miller, M.C., 1997. Glacial
CE

16 and Postglacial Pollen Records from the Ecuadorian Andes and Amazon.
AC

17 Quaternary Research 48, 69-78.

18 Colinvaux, P.A., Frost, M., Frost, I., Liu, K.-B., Steinitz-Kannan, M., 1988a.

19 Three pollen diagrams of forest disturbance in the western Amazon basin.

20 Review of Palaeobotany and Palynology 55, 73-81.

21 Colinvaux, P.A., Olson, K., Liu, K.-B., 1988b. Late-glacial and Holocene pollen

22 diagrams from two endorheic lakes of the Inter-Andean plateau of Ecuador.

23 Review of Palaeobotany and Palynology 55, 83-99.

24 Davis, K., 1999. Pollen analysis of Tulare Lake, California: Great Basin-like

25 vegetation in Central California during the full-glacial and early Holocene.

26 Review of Palaeobotany and Palynology 107, 249–257.

27
 ACCEPTED MANUSCRIPT

1 Emck, P., 2007. A climatology of south ecuador - with special focus on the major

2 andean ridge as atlantic-pacific climate divide. Dissertation, Universität

3 Erlangen-Nürnberg.

PT
4 Faegri, K., Iversen, J., 1989. Textbook of Pollen Analysis. Wiley, New York.

5 Froyd, C.A., Wilis, K.J., 2008. Emerging issues in biodiversity & conservation

RI
6 management: The need for a palaeoecological perspective. Quaternary

SC
7 Science Reviews 27, 1723-1732.

8 Grimm, E.C., 1987. CONISS: A Fortran 77 program for stratigraphically

9
NU
contrained cluster analysis by the method of the incremental sum of squares.
MA
10 Computer and Geosciences 13, 13-35.

11 Guffroy, J., 2004. Catamayo precolombino. Investigaciones arqueologicas en la

12 provincia de Loja (Ecuador). IRD editions, Paris.

ED

13 Guffroy, J., 2006. El Horizonte corrugado: correlations estilisticas y culturales.

PT

14 Bulletin de l'Institut Francais d'Etudes Andines 35 (3), 1-13.

15 Hansen, B.C.S., Seltzer, G.O., Wright, H.E., 1994. Late Quaternary vegetation
CE

16 change in the central Peruvian Andes. Palaeogeography, Palaeoclimatology,

AC

17 Palaeoecology 109, 263-285.

18 Homeier, J., Werner, F.A., 2005. Preliminary Checklist of the Spermatophytes of

19 the Reserva San Francisco (Province Zamora-Chinchipe, Ecuador).

20 Department of Plant Ecology, Albrecht-von-Haller-Institute for Plant

21 Sciences, University of Göttingen, Göttingen.

22 Hooghiemstra, H., 1984. Vegetation and climatic history of the High Plain of

23 Bogota, Colombia. Dissertationes Botanicae 79, 368 pp. Cramer,Vaduz.

24 Litherland, M., Aspen, J.A., Jemielita, R.A., 1994. The metamorphic belts of

25 Ecuador. Overseas Memoir of the British Geological Survey 11, 1–147.

28
 ACCEPTED MANUSCRIPT

1 Lozano, P., Delgado, T., Aguirre, Z., 2003. Estado actual de la flora endemica

2 exclusive y su distribucion en el Occidente del Parque Nacional Podocarpus.

3 Funbotanica y Herbario y Jardin Botanico. Loja, Ecuador.

PT
4 Marchant, R., Almeida, L., Behling, H., Berrio, J.C., Bush, M., Cleef, A.,

5 Duivenvoorden, J., Kappelle, M., De Oliveira, P., Teixeira de Oliveira-Filho,

RI
6 A., Lozano-Garcia, S., Hooghiemstra, H., Ledru, M.P., Ludlow-Wiechers, B.,

SC
7 Markgraf, V., Mancini, V., Paez, M., Prieto, A., Rangel, O., Salgado-

8 Labouriau, M.L., 2002. Distribution and ecology of parent taxa of pollen

9
NU
lodged within the Latin American Pollen Database. Review of Palaeobotany
MA
10 and Palynology 121, 1-75.

11 Niemann, H., Behling, H., 2008a. Late Quaternary vegetation, climate and fire

12 dynamics inferred from the El Tiro record in the southeastern Ecuadorian

ED

13 Andes. Journal of Quaternary Science 23, 203-212.

PT

14 Niemann, H., Behling, H., 2008b. Late Holocene environmental change and

15 human impact inferred from three soil monoliths and the Laguna Zurita multi-
CE

16 proxi record in the southeastern Ecuadorian Andes. Vegetation History and

AC

17 Archaeobotany (in review).

18 Niemann, H., Haberzettl, T., Behling, H., 2009. Holocene climate variability and

19 vegetation dynamics inferred from the (11,700 cal yr BP) Laguna Rabadilla de

20 Vaca sediment record in the southeastern Ecuadorian Andes. The Holocene 19

21 (2), 000-000.

22 Placzek, C., Quade, J., 2001. Holocene Lake-Level Fluctuations of Lake Aricota,

23 Southern Peru. Quaternary Research 56, 181–190.

24 Pohle, P., 2008. The People Settled Around Podocarpus National Park. In: Beck,

25 E., Bendix, J., Kottke, I., Makeschin, F., Mosandl, R., (Eds.), Gradients in a

29
 ACCEPTED MANUSCRIPT

1 Tropical Mountain Ecosystem of Ecuador. Ecological Studies 198, Springer

2 Verlag, Berlin, Heidelberg, pp 25-36.

3 Richter, M., Moreira-Munoz, A., 2005. Climatic heterogeneity and plant diversity

PT
4 in southern Ecuador experienced by phytoindication. Review of Peruvian

5 Biology 12, 217-238.

RI
6 Rozsypal, A.A., 2000. Die pleistozäne Glazialmorphologie in Ecuador und

SC
7 Nordperu unter besonderer Betrachtung der Cordillera Oriental bei Loja.

8 Diplomarbeit, Institut für Geographie, Universität Erlangen-Nürnberg.

9 (unpublished).
NU
MA
10 Thompson, L.G., Mosley-Thompson, E., Davis, M.E., Lin, P.-N., Henderson, K.,

11 Mashiotta, T.A., 2003. Tropical glacier and ice core evidence of climate

12 change on annual to millennial time scales. Climatic Change 59, 137-155.

ED

13 Tonneijck, FH., Van der Plicht, J., Jansen, B., Verstraten, J.M., Hooghiemstra, H.,
PT

14 2006. Radiocarbon dating of soil organic matter fractions in Andesols in

15 northern Ecuador. Radiocarbon 48, 337–353.

CE

16 Van der Hammen, T., Urrego, C.G., 1978. Prehistoric man on the Sabana de
AC

17 Bogota. Data for an ecological prehistory. 25, 179-190.

18 Weng, C., Bush, M.B., Athens, J.S., 2002. Holocene climate change and hydrarch

19 succession in lowland Amazonian Ecuador. Review of Palaeobotany and

20 Palynology 120, 73-90.

21 Weninger, B., Jöris, O., Danzeglocke, U., 2004. Calpal - The Cologne

22 radiocarbon CALibration and PALaeoclimate research package.

23 http://www.calpal.de

24 Wille, M., Hooghiemstra, H., Hofstede, R., Fehse, J., Sevink, J., 2002. Upper forest line

25 reconstruction in a deforested area in northern Ecuador based on pollen and

26 vegetation analysis. Journal of Tropical Ecology 18(3), 409–40.

30
 ACCEPTED MANUSCRIPT

2 Figure and table captions

3 Fig. 1: Map of central northwestern tropical South America, showing the study

PT
4 site (star) and other locations of greater interest for this study (circles). Source:

5 OMC (www.aquarius.geomar.de).

RI
6 Fig. 2: Photograph of Cocha Caranga area (view from the north), the three core

SC
7 sites (arrows) are shown.

NU
8 Fig. 3: Age-depth model of Laguna Cocha Caranga (left) and Cocha Caranga mire

9 (right) sediment cores.

MA
10 Fig. 4: Pollen percentage diagram of the sediment core of Laguna Cocha Caranga

11 (2710 m) showing the sample ages (cal yr BP), selected pollen and spore taxa
ED

12 grouped into lower mountain forest (LMF), upper mountain forest (UMF),

13 subparamo, and paramo and the pollen zones.

PT

14 Fig. 5: Pollen summary diagram of the sediment core of Laguna Cocha Caranga
CE

15 (2710 m) showing the AMS radiocarbon dates, sample ages (cal yr BP), the

16 lithology, sums of ecological groups, the pollen sum, the deposition time,
AC

17 algae, pollen and charcoal concentration, pollen and charcoal influx, the

18 pollen zones and the CONISS dendrogram.

19 Fig. 6: Pollen percentage diagram of the sediment core of Cocha Caranga mire

20 (2710 m) showing the sample ages (cal yr BP), selected pollen and spore taxa

21 grouped into lower mountain forest (LMF), upper mountain forest (UMF),

22 subparamo, and paramo and the pollen zones.

23 Fig. 7: Pollen summary diagram of the sediment core of Cocha Caranga mire

24 (2710 m) showing the AMS radiocarbon dates, sample ages (cal yr BP), the

25 lithology, sums of ecological groups, the pollen sum, the deposition time,

31
 ACCEPTED MANUSCRIPT

1 algae, pollen and charcoal concentration, pollen and charcoal influx, the

2 pollen zones and the CONISS dendrogram.

3 Fig. 8: Pollen percentage and summary diagram of the soil core of Cocha Caranga

PT
4 forest (2700 m) showing the AMS radiocarbon dates, the lithology, selected

5 pollen and spore taxa grouped into lower mountain forest (LMF), upper

RI
6 mountain forest (UMF), subparamo, and paramo, sums of ecological groups,

SC
7 the pollen sum, pollen and charcoal concentration, the pollen zones and the

8 CONISS dendrogram.

9
NU
Fig. 9: Holocene wet/dry phases assumed from the Laguna Cocha Caranga record
MA
10 in comparison with two other core sites (Laguna Rabadilla de Vaca and the El

11 Tiro-Pass) of the Podocarpus National Park region.

12
ED

13 Tab. 1: Site specific data of the three cores from Cocha Caranga area.
PT

14 Tab. 2: AMS-radiocarbon dates and calibrated ages of the three cores from Cocha

15 Caranga area.
CE
AC

32
 ACCEPTED MANUSCRIPT

Figure 1

PT
C RI
US
N
MA
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

Figure 2 (not coloured in publication)

PT
C RI
US
N
MA
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

Figure 3 (not coloured in publication)

PT
C RI
US
N
MA
ED
PT
CE
AC
 Sa

9000
8000
7000
6000
5000
4000
3000
2000
1000

14000
13000
12000
11000
10000
0
m
pl
e
D A

8
6
4
2
0

64
62
60
58
56
54
52
50
48
46
44
42
40
38
36
34
32
30
28
26
24
22
20
18
16
14
12
10
ep ge
th (c
Figure 4
C (c al
m yr
ec
ro ) B
p P
C ia )
el
tis
cf
.F
ab
M ac
ea
or
ac e
ea
Tr e/
em U
a r ti
ca
Al ce
nu a e
s
H
ed
LMF
yo

20
sm
Il e um
x
M
yr
ic
a

20
40
M
yr
s
M ine
yr
Po ta ce
d a
W oca e
e i rp
UMF

n m ac

20
an e a
ni e
a

40
As
te
C rac
le e
t a
AC
Er hra e
ic ce
M a c e ae
el a
as e

20
to
CE
Br m
at
as ac
H si c ea
yd ac e
ro e
PT
P o c ae
a c ty o
ea le

20
e
ED

40
Bl
ec
M h nu
on m
MA
M o le im
Subparamo Paramo

o n te p
N e
C o le p s ria
ya te il a l e
t
C t h e v er e <
ya a ru 6
c
H the on cate 5um
US
up a ju
v g
J a er z err ata C
m ia uc
e at
Ly so e
co ni
T r po a
i le diu
RI
te m
20
ps c l
ACCEPTED MANUSCRIPT

Tr i la a v
PT
i le te a t
te um
Ferns

20
ve
C r r
yp uc
er at
e
20

ac
ea
e
40
60

Sp
h
Is agn
oe u
te m
20

s
40
60
80 100

Bo
t ry
oc
oc
cu
s
1000
2000
Zone

CL - 1
CL - 2
CL - 3
CL - 4
CL - 5
 ACCEPTED MANUSCRIPT

Figure 5

PT
Pollen Charcoal

00
00

0
RI
x1

00
3)

10
m

x
/c

3)
s

00
al

m
du

x1
/c
00

es
C
vi

r)
di

cl
00

² /y
( in

ti

cm
x1
F)

ar
F)

n
M

(p
M

3)

/
tio

es
(U
(L

um
US
cm

10
tr a
at
)

)
cl
BP

m
st

st

in

ti
x
s/
en

/c
0

ar
re

re

rm

r)
in

15
yr

rs
(p
c
Fo

Fo

²/ y
P)

ra
te

on

(y
-
al

10
rB

um
(g

m
de
n

n
(c

e
ai

ai

/c

tim
(y

In

n
us
ge

50
nt

nt

s
tio

tio
)

in
es

um

cc
ou

ou
A

-1

n
am

tra

tra
ra
(c

tio
at

o
M

10
S
ow
o

(g
y

en

en
pl

oc
D

th

ar

i
og

os
N
n
er

er
m

s
S-

tr y

x
ep

bp

le
ra

rn

flu

flu
w
ol

pp

on

on

ep
nk
Sa

l
AM

Su

Pa

Po

Fe

Bo
Lo
th

Zone CONISS

In

In
U

D
Li

0 296
0

MA
2 299
4 297
6 296
8 300
10 298
CL - 5

ED
12 303
14 298
16 297
18 301

PT
1000 20 300
1208±48 22 304
298
24 302
26 302
2000

3000
28
30
32
302
298
302
CE CL - 4

34 301
AC
4000 36 301
38 297
5000 40 298
42 300
CL - 3
44 301
6000 46 303
48 301
7000 50 302
52 302
CL - 2
8000 54 303
8256±58
9000 56 299
10000 58 300
11000
60 300 CL - 1
12569±140 12000
13000 62 300
14000 64 302
20 20 40 60 20 20 40 20 20 40 100 200 300 200 400 600 800 500 1000 1500 200 400 600 800 100 200 300 400 200 400 600 800 2 4 6 8
Pollen + Spores

Fine Detritus Mud Decomposed Organic Material with Roots Silty Clay with Organic Material
 Sa

1500
1400
1300
1200
1100
1000
900
800
700
600
500
400
300
200
100
m
pl
e
D Ag

68
64
60
56
52
48
44
40
36
32
28
24
20
16
12
8
4
0
Figure 6
ep
th e
(c
C (c al
ec m yr
ro ) BP
C
pi
a )
el
t is
M
or
ac
Tr ea
e/
em U
a r ti
ca
Al ce
nu
s a e
LMF

H
ed
yo
sm
Il e
x um

M
yr
ic
a

20
M
yr
P o s in e
d
W oc a
e i rp
nm a
UMF

20
an c e a
ni e
a
AC As
te
ra

20
ce
ae
CE Bo
m
C ar e
le a
t
PT
E r hr a
ic c e
M ac e a e
ED
el ae
as

20
to
H m
yd at
ac
MA
P o r oc ea
N a c o ty e
l
Subparamo

20
ea e
e US

40
Va
le
C
M r ia
o n na
RI
C ol e
ya te
ACCEPTED MANUSCRIPT

Ly t h e p s
PT
Paramo

co a v ila
Ly p o e r r t e
co d iu uc
20

po m a t e
Tr d i cl
um av
i le
Tr e t cu atum
i le p s rv
te ila at
um
20

ve te
C r ru
yp
Ferns

er ca
te
20

ac
ea
e
Sp
ha
20

gn
Bo um
t ry
oc
oc
cu
s
1000
2000
Zone

CM - 1
CM - 2
 ACCEPTED MANUSCRIPT

Figure 7

PT
P o lle n C ha rco a l

00
10

00
x

RI

00
3)

x1
cm

3)
ls/

cm

0
ua

00
id

s/

x1
C
v

le
di

t ic

r)
10
(in

²/y
F)

ar
F)

x
n

(p
M

cm
Fo (LM

3)
t io
(U

US

um
m

s/
)

ra
at
BP

/c

)
in
st

st

nt
in

m
50
ns

ra
re

re

rm

r/ c
r)
yr

(g
nc

i
P)

Fo

-1
²/ y
ra
te

(y
al

10
Co
rB

um
(g

cm
de
in

n
(c

e
ai

t im
ta
(y

n
In

s/
s
e

nt

50
tio

tio
pp oun

cu
)
Ag

in
Sa t es

m
m

+
ou

on
ra

ra

-1
am

oc

ra
Su
(c

n
e

M
Da

10
nt

nt

it i
N
w
y

(g
oc
pl

ar
th

og

os
no

ce

ce
er

er

s
S-

ry

x
ep

bp

lle
ra

rn
ol

on

flu

on

flu

ep
w

nk

t
AM

Su

Pa

Po

Fe

Bo
th
D

Lo

Zone CONISS

In

In
U

D
Li

MA
0 304

4 304
0
8 302

ED
12 304

16 304
100
20 306

PT
24 305
CM - 2
28 304

200 32 304 CE
36 305
AC
40 302
300
259±44 44 303
400
500 48 305
600
700 52 301
800 CM - 1
900 56 299
1000
1100
60 300
1200
1300
64 299
1400
1636±46 1500
68 301
20 20 40 20 20 40 20 20 40 500 1000 1500 200 400 600 8001000 5000 10000 200 400 600 100 200 300 20 40 60 2 4
Pollen + Spores

Peat Organic Material with Roots Decomposed Organic Rich Material

 130±43
-160±35
AM
S-
Da
D t e

8
4
0

48
44
40
36
32
28
24
20
16
12
ep (y
th r
Figure 8
(c B P
m )
Li )
th
ol
og
C y
el
tis
M
or
ac
Tr ea
em e/
Ur
a ti ca
Al
nu c
s ea
e
H
ed
yo
s

20
m
um

40

Decom pos ed Organic Rich Material

60
80
LMF

I le
x

20
M
yr
ic
a

20
M
yr
P o s ine
do
W ca
ei r p
m ac

20
an e
ni ae
a

40
U MF

60
As

Organic Material with Root s

te
B o ra c
m ea
C a re e
le a
th
E r ra c
ic e
AC
M ace ae
el ae
as

20
to
m
at
ac
CE
Br ea
as e
P o sic
ac ac
ea ea

20
e e
PT
Va
le
M r i an
on a
ED
Ly ole
co te
Subparam o

T r po ps
ile diu ila
t
T r te p m e
ile si c la
MA
C te lat e va
yp ve tu
N e r rru m
ac c
Lo ea ate
w e
U er M
pp o
Param o

er u n

20
M ta
US
o u in
nt Fo

40
a i re C
n
Fo s t (
re L M
60 st F
(U )
M
Ferns

F)
RI
ACCEPTED MANUSCRIPT

80 100

PT
Su
bp
ar
20

am
o
Pa
ra
m
20

o
U
nk
no
20

w
n
Po +
lle In
298
302
302
306
298
304
302
304
302
305
301
304
301

n d et
Su er
Fe m m
rn in
s at
e
20

Po
lle
n
Co
nc
en
tra
t io
n
(g
ra
in
100 200 300 400

s/
cm
C 3)
ha
rc x1
oa 0 00
lC
on
c.
10
-1
50
um
200 400 600 800

(p
ar
t ./
cm
3)
Zone

CF - 1
CF - 2

x1
00
00
1
CON ISS

2
Pollen + Spores
3
 ACCEPTED MANUSCRIPT

Figure 9

PT
C RI
US
N
MA
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT

Table 1

PT
Core name Material Elevation Core Number of
length identified taxa

RI
Laguna Cocha Caranga (CL) Lake sediment 2710 m 65 cm 109

C
Cocha Caranga Mire (CM) Peat 2710 m 69 cm 86

US
Cocha Caranga Forest (CF) Soil deposit 2700 m 49 cm 54

N
Table 2

MA
ED
14
Core name Labor code Dated material Core depth C yr BP cal yr BP

PT
Laguna Cocha Caranga Erl-11035 Bulk sample 23 cm 1208 ± 48 1149 ± 70
Laguna Cocha Caranga Erl-11395 Bulk sample 55 cm 7413 ± 52 8256 ± 58
Laguna Cocha Caranga
Cocha Caranga Mire
Erl-11036
Erl-11393
CE
Bulk sample
Bulk sample
61 cm
43 cm
10636 ± 38
259 ± 44
12569 ± 140
301 ± 114
Cocha Caranga Mire Erl-11394 Bulk sample 67 cm 1636 ± 46 1523 ± 71
AC

Cocha Caranga Forest Erl-11392 Soil organic matter 26 cm -160 ± 35

Cocha Caranga Forest Erl-12106 Charcoal 30 cm 130 ± 43 140 ± 101