Late-Holocene pollen-based paleoenvironmental reconstruction of the El Triunfo wetland,Los Nevados National Park (Central Cordillera of Colombia)

Abstract

We present a palynologic study of El Triunfo wetland (4°58′54.5″N and 75°19′55.8″W), located at 3834 m a.s.l. in the Central Cordillera of Colombia. Pollen and spores record, which spans for the past ~1930 yr BP, shows that Páramo vegetation has been dominated by Poaceae, Asteraceae, Lycopodiaceae, and Ericaceae. The sub-Andean forest is characterized by Urticaceae and Acalypha (Euphorbiaceae), while the Andean forest by Quercus (Fagaceae), Hedyosmum (Chloranthaceae), and Weinmannia (Cunoniaceae). We observed that aquatic vegetation pollen such as Cyperaceae were abundant, indicating marsh conditions in most of the record. Cold and warm periods, inferred from Páramo and Andean forest elements, respectively, reveal that the warmest periods were registered at ~1000 cal. yr BP (~AD 1016) (‘Medieval Climatic Optimum’) and in the last century, whereas the coldest ones around 1200 cal. yr BP (AD 831) and 650–150 cal. yr BP (AD 1368–1750) (‘Little Ice Age’). The pollen signal suggests that warm periods were dry and cold periods were wet. A comparison of these results with some localities of northwestern Colombia, the Caribbean, and South America was made, displaying variations that reflect regional and/or global climatic phenomena such as Intertropical Convergence Zone and the El Niño Southern Oscillation. In the past 50 years, the decrease in Quercus, Weinmannia and Clusia recovery and the increase in Rumex, Plantago, and Gunnera may reflect human impact derived from livestock, agriculture, and/or firewood. Volcanic events of variable magnitude have been identified with an apparent minimal effect on vegetation, reaching a maximum variation of about 10–15% between the beginning and the end of a volcanic event.

Keywords

Andean forest climate change Colombia late Holocene Páramo pollen tropical Andes

Introduction

Understanding and adapting to global environmental change is one of the major challenges for the humanity in this century. The past 3000 years in particular, rapid and dramatic natural changes linked to climate dynamics such as the sub-Boreal/sub-Atlantic transition, ‘Little Ice Age’ (LIA), ‘Medieval Climate Optimum’, and the recent warming have produced an effect on humans (IPCC, 2007; Velásquez, 2005; Wanner et al., 2008), increasing the interest in working on paleoclimate archives. Pollen records, especially those of tropical areas like Colombia, have been useful in terms of reconstructing vegetation dynamics and environmental changes associated with climatic changes in the geological record (Bogotá et al., 2011; Boom et al., 2001; Bosman et al., 1994; Bush et al., 2007; Gómez et al., 2007; Groot et al., 2011; Hooghiemstra, 1989; Hooghiemstra et al., 2006; Marchant et al., 2001a, 2001b, 2002, 2004; Mommersteeg et al., 2005; Van der Hammen and Hooghiemstra, 1995, 2003; Velásquez and Hooghiemstra, 2013; Weng et al., 2007). Also, this signal probably will be helpful to understand recent and future effects of global warming on ecosystems and human population; particularly, the great human pressure on the Páramo vegetation linked to agricultural activities promote the destruction of wetlands and certain forest taxa and can explain the most recent tendency to warming. In this sense, a region of particular importance due to its high sensitivity to climate changes is the Los Nevados National Natural Park (Los Nevados) located in the central part of the Central Cordillera of Colombia (Figure 1). At this place, many lakes and peat bogs serve as a water reservoir for more than 2 million people (Parques Nacionales Naturales de Colombia, 2007), as well as high biodiversity sites and beautiful touristic landscapes. This region is a great archive not only of the climatic and environmental history of the area but also of volcanic activity. In the past 40 years, a notorious decrease in the ice cover of surrounding peaks has been detected along with intensive anthropogenic activity (cattle, corps, and firewood) that has significantly altered the original vegetation (Álvarez et al., 2007; Hofstede, 1995; Verweij et al., 2003). In the 1980s, some paleoecologists studied climate and vegetation changes during the late Glacial and Holocene with a millennial or centennial resolution, mainly in the southern part of the Los Nevados (Santa Isabel and Santa Rosa ice caps) (Kuhry, 1988; Melief, 1989b; Salamanca and Noldus, 2003; Thouret and Van der Hammen, 1983). The poor resolution was related to insufficient chronologic and sampling control and technical aspects of core recovery. Nevertheless, in order to obtain a more detailed understanding of Holocene climate changes, processes operating at decadal scales (e.g. El Niño Southern Oscillation (ENSO)), their causes, and interactions (Wanner et al., 2008) must be taken into account.

Figure 1.

(a) Map of South America and Colombia, which includes some selected localities with high-resolution paleoecological information: (1) Altiplano Perú-Bolivia, (2) La Cocha lake, (3) Lagunares de Santa Isabel, (4) El Triunfo wetland (this study), (5) Páramo de Frontino, and (6) Cariaco basin. EC: Eastern Cordillera; CC: Central Cordillera; WC: Western Cordillera. The dotted line represents the position of the Intertropical Convergence Zone (ITCZ) in July and January. Elevation model taken from DWH DIVER (2015). (b) Elevation model of Los Nevados and its main hydrographic basins (3 and 4 in Figure 1a) and (c) general view of the El Triunfo wetland (3834 m a.s.l.). The arrow indicates the location of the drill site. The elevation model is taken from ASF DAAC (2015). m a.s.l.: meters above sea level.

Other studies carried out on Los Nevados have been aimed at understanding the response of vegetation to volcanism. Melief (1989a) and Salomons (1986) found that volcanic eruptions did not have a strong effect on the vegetation cover, and if they did, regeneration occurred at a fast rate. Similarly, Salamanca (2003) suggested volcanic activity and a localized effect, reflecting structural and compositional vegetation changes in some areas.

With the aim to reconstruct the late-Holocene vegetation and interpret climate and environmental changes, the authors collected a 4-m core from the El Triunfo wetland, which is located 3834 meters above sea level (m a.s.l.) (Figure 1). This core was studied using pollen and spores and sampled with a decadal time resolution. Nine radiocarbon dates indicated a continuous record of the last 2000 years, providing one of the most complete and continuous high-resolution records for the neotropics. A correlation with other results from the Caribbean and South America was achieved with the purpose of finding possible regional climate links.

Study area

El Triunfo wetland (4°58′54.5″N and 75°19′55.8″W) is located in the Central Cordillera of Colombia, in the headwaters of the Gualí river basin (Tolima department, Figure 1), around Los Nevados at 3834 m a.s.l. (Parques Nacionales Naturales de Colombia, 2007). It is situated in a valley associated with glacial activity and surrounded by steep hills composed of andesites and basalts which originated as lava flows or volcanic necks during the Neogene (Herd, 1982; Thouret et al., 1995). Nowadays, the wetland remains partially filled with water from rainfall and runoff that is conducted into a creek, which runs in W-E direction and makes part of the tributaries of the Gualí River (Figure 1). It is enclosed by the Andean forest (2400–3800 m a.s.l.) and Páramo vegetation (3800–4000 m a.s.l.) (Cuatrecasas, 1958; Figure 1), and the aquatic vegetation is composed of the same elements mentioned in the pollen diagram. The wetland is surrounded by a human-modified ecosystem that includes agriculture (grasses and potato) and livestock. Despite the strong anthropic activity, these ecosystems still conserve some species of the original communities, particularly in the gallery forest, than flank the watercourses and on the top of their adjacent mountains (Álvarez et al., 2007).

The modern Páramo community contains Poaceae, Lycopodiaceae, Asteraceae, and Ericaceae. In the Andean forest, Quercus (Fagaceae), Hedyosmum (Chloranthaceae), Viburnum (Adoxaceae), and Cyatheaceae are the most abundant species (Álvarez et al., 2007). Cyperaceae, Plantago (Plantaginaceae), Senecio sp. (Asteraceae), Apiaceae (Hydrocotyle), Lachemilla orbiculata (Rosaceae), and Gunnera magellanica (Gunneraceae) are currently growing in the wetland.

The annual average precipitation in this zone is 2100 mm, corresponding to a bimodal pattern with four seasons: two humid seasons of high precipitation in April–May and September–October, when the Intertropical Convergence Zone (ITCZ) reaches the center of the country and two dry seasons in July–August and November–December when the ITCZ moves to the north of the country (Poveda et al., 2001). The temperature in the zone has an annual average of 6.8°C (Brisas Meteorological Station) (Velásquez and Henao, 2007) and decreases at a rate of ~0.6°C/100 m in altitude change (Flórez, 1992; Mesa et al., 1997).

Another important factor that controls rainfall in the area is ENSO. The periodicity of this phenomenon is around 3 years: In the middle of the first year, ENSO brings a decline in precipitation to the Andes and the Caribbean (Marchant et al., 2006; Poveda et al., 2011). The second year is generally dry, and finally, at the end of the period, there is more rain than average (Poveda et al., 2011). Nowadays, an increase in the phreatic level of the wetland can be observed in response to the wet periods of ENSO; nevertheless, there are no systematic hydrologic studies in this area. It is important to point out that since ENSO forms a pattern of vegetation changes in several places in South America (Chepstow-Lusty et al., 2003), it is feasible to detect this pattern in the past and refer to periods with little or much ENSO activity, not individual phenomena.

Methodology

The coring site was chosen near to the Upper Forest Limit (UFL) after careful analysis of aerial photographs and field work. Two transects crossing the wetland north to south and east to west were performed in order to find the depocenter of the basin. In December 2012, two cores TRIU 1 and 2 (4°58′54.5″N to 75°19′55.8″W) were obtained 50 cm apart using a Russian Corer-Eijkelkamp peat sampler 50 cm long with a diameter of 5 cm. The extracted cores were described in the field, stored in PVC pipes, and transferred to the Instituto de Investigaciones en Estratigrafía (IIES, Universidad de Caldas), where they are stored at 4°C.

Lithological description, sampling, and palynological analysis

The selected core, TRIU 1, was described lithologically and stratigraphically. In total, 72 samples were collected and prepared by the standard acetolysis process. We used pyrophosphate (10%) to disaggregate sediments and 10 µm sieves to separate pollen and spores. After counting 4000 pollen grains and spores in a surface sediment sample until obtaining an asymptotic curve and achieving a Monte Carlo simulation, it was decided that 500 pollen grains and pteridophyte–bryophyte spores should be counted on 7 µL aliquots. These methods allow us to calculate influx rate and give a good estimation of pollen diversity in the samples (Velásquez et al., 1999, 2004). For the palynomorph identification, the atlas of Hooghiemstra (1984), Velásquez (1999), and Murillo and Bless (1974) and the palynologic reference collection of the National University Paleoecology Laboratory (Medellín, Colombia) were employed. Pollen diagrams and ecological groups were produced using TILIA and TILIAGRAPH, while the cluster analysis of the pollen spectra utilized CONISS (Grimm, 1987). Pollen and spores coming from regional vegetation (sub-Andean forest, Andean forest, and Páramo) constituted the pollen sum used for making summary diagrams. The percentages of other taxa, not included in the pollen sum and growing in situ, were estimated based on pollen recovery, and in some cases they could reflect more than 100%. Tephras and ash layers are shown in the pollen diagram as lines in order to obtain a continuous representation of vegetation changes in the organic intervals and to facilitate ecological interpretations.

Dating

Nine samples of bulk organic matter of sediment were collected for AMS (accelerator mass spectrometry) radiocarbon dating at 54, 100, 119, 154, 189, 209, 284, 354 and 385 cm and then dated at Beta Analytic laboratory (Florida, US). Radiocarbon dates were calibrated following CALIB 7.00 (Stuiver and Reimer, 2015). The age model was developed using the software Bacon (Blaauw and Christen, 2011).

Ecological framework

For the paleovegetation reconstruction of the El Triunfo wetland, we considered the main taxa in modern vegetation belts (Cuatrecasas, 1958; Grabandt, 1980; Marchant et al., 2002; Rangel-Ch et al., 1995; Salomons, 1986; Velásquez and Hooghiemstra, 2013). Páramo elements Poaceae, Asteraceae, Ericaceae, Lycopodiaceae, Valeriana, Gentiana, Hypericum, Geranium, Aragoa, Lycopodium sp., and Huperzia (Lycopodiaceae) are major pollen producers. These elements which grow in the Páramo, including in the emerged areas of the wetland, were included in the pollen sums because they are good indicators of cold conditions. Andean Forest is dominated by Hedyosmum, Podocarpus, Quercus, Alnus, Cyatheaceae, Viburnum, Melastomataceae, Rosaceae (Hesperomeles type), Morella, Juglans, Weinmannia, Clusia, and Rubiaceae and the sub-Andean Forest by Pilea, Urticaceae, Acalypha, Alchornea, and Cecropia.

The vegetation growing inside the wetland, such as Lachemilla orbiculata (Rosaceae), Gunnera magellanica (Gunneraceae), Senecio formosus (Asteraceae), and Apiaceae (Hydrocotyle), reflects high precipitation due to the fact that all of them grow along small and slow streams. Cyperaceae is indicative of marsh conditions. Plantago (Plantaginaceae) normally grows in a cushion way, in the driest areas of the wetland or in totally closed peatlands in the Los Nevados area and other places of Colombia, like the Páramo de Frontino (Velásquez and Hooghiemstra, 2013), and therefore is indicative of relative dryness.

Fluctuations in relative percentages of forest and Páramo vegetation can be used as temperature indicators. Vegetation measurements in the current tropical Andes showed that between 2000 and 2550 m of elevation, the arboreal pollen is 25%; at 2700 m, 35%; at 2900 m, 45%; at 3100 m, 55%, and at 3300 m, 65% (Hooghiemstra, 1984). Temperature decreases 0.6°C each 100 m up in Eastern Cordillera and it is equivalent to 5% of arboreal pollen change; it is possible to calculate past temperatures and also the upper forest position (Hooghiemstra, 1984).

Results

Lithology

The core is composed mainly of degraded vegetal remains (leaves and roots) with different colors (Table 1). Some of them show disseminated volcanic tephra (ash and/or lapilli). At least seven discrete volcanic layers composed of ash and lapilli can be identified (Figure 2). A depth/lithology relation is depicted in Table 1.

Table 1.

Lithological description of El Triunfo core, Los Nevados, Colombia. Letters indicate the most important volcanic layers. Peat color based on the Munsell rock color book (2014).

Depth (cm)	Lithology
400–351	Peat brownish black (5YR 2/1) and peat brownish black (5YR 2/1) with lapilli and mixed volcanic ash
350–301	Peat brownish black (5YR 2/1) with mixed volcanic ash
300–281	Peat brownish black (5YR 2/1) with mixed lapilli
280–264	Volcanic layer grading from lapilli to ash (A)
263–261	Peat brownish black (5YR 2/1)
260–251	Volcanic layer grading from lapilli to ash (B)
250–239	Mixture of peat and lapilli
238–211	Volcanic layer grading from lapilli to ash (C)
210–205	Peat brownish black (5YR 2/1)
204–201	Volcanic ash layer (D)
200–183	A blend of peat dusky yellowish brown (10YR 2/2) and volcanic ash
182–170	Volcanic layer grading from lapilli to ash (E)
170–141	Mixture of peat brownish black (5YR 2/1) with lapilli.
140–126	Volcanic layer grading from lapilli to ash (F).
125–111	Mixture of peat dusky yellowish brown (10YR 2/2)
110–51	Peat brownish black (5YR 2/1
50–47	Volcanic lapilli layer (G)
46–0	Peat brownish black (5YR 2/1)

Figure 2.

Stratigraphic log and age model for El Triunfo wetland core. The model used Bacon, version 2.2 (Blaauw and Christen, 2011). The cal. yr BP age distributions are shown in black. In the age–depth model, the dark gray color indicates the most probable calendar ages. External dotted lines show 95% confidence intervals; the central dotted line is the best unique model based on weighed mean age for each depth. Letters indicate the most important volcanic layers.

Radiocarbon dating

The radiocarbon ages obtained are shown in the Table 2 and Figure 2, revealing that the record is continuous, without significant hiatuses. The mean accumulation rate is about 6 yr/cm. Samples were collected every 3 cm, corresponding, thus, to a resolution of approximately 20 years.

Table 2.

AMS radiocarbon dating from the El Triunfo wetland.

Code laboratory	Depth (cm)	Material	Conventional radiocarbon age	Best estimated age (cal. yr BP/AD)
Beta – 368996	54	Organic sediment	90±30	82/1868
Beta – 330145	100	Organic sediment	410±30	474/1476
Beta – 368997	119	Organic sediment	480±30	520/1429
Beta – 368998	154	Organic sediment	570±30	615/1334
Beta – 369001	189	Organic sediment	730±30	690/1260
Beta – 368999	209	Organic sediment	830±30	738/1212
Beta – 369000	284	Organic sediment	1070±30	1040/978
Beta – 369002	354	Organic sediment	1580±30	1469/481
Beta – 369003	385	Organic sediment	1810±30	1757/193

Palynology

CONISS analyses suggest that the pollen diagram can be divided into six zones (Figure 3a–c with maximum percentage values in parentheses):

Zone 1 (401–340 cm, 14 samples). Forest elements fluctuate between 10% and 40%, showing peaks at 390, 375, and 350–337 cm (Figure 3a). In the sub-Andean forest, the most abundant taxon is Urticaceae (8%). The Andean forest is dominated by Viburnum (33%), Quercus (14%), Cyatheaceae (10%), Hedyosmum (5%), Melastomataceae (4%), and Podocarpus (3%). The Páramo vegetation is represented mainly by Lycopodiaceae (91%), Poaceae (4–54%), Ericaceae (25%), and Asteraceae (8%). Lycopodiaceae always have values inversely proportional to those of Poaceae, while values of Asteraceae, Pteridophyta, fungal spores, mosses, and forest vegetation increase simultaneously (Figure 3b). Ericaceae is important only in the upper part of the zone. Aquatic vegetation is represented by Apiaceae (Hydrocotyle) (95%), Cyperaceae (97%), and Lachemilla orbiculata (35%). It is noteworthy that when Cyperaceae decrease, Apiaceae (Hydrocotyle) and Lachemilla orbiculata increase (Figure 3c).

Zone 2 (340–280 cm, 14 samples). Urticaceae mainly characterize the sub-Andean forest with an increase in signal from 0% to 15%. The Andean forest taxa are not well represented in this part of the record (Figure 3a). The Hedyosmum and Cyatheaceae values decrease remarkably toward the upper part of the zone. Páramo-related taxa such Poaceae reach 83%, whereas Lycopodiaceae show low values (<5%) and Ericaceae reach 55%. Among the hydrophytes, Lachemilla orbiculata reach 90% at the beginning of the zone and then are replaced by Cyperaceae (Figure 3c). Other local palynomorphs such as fungal spores increase significantly at upper part (1500%), as opposed to the mosses species (Figure 3b).

Zone 3 (280–192 cm, 8 samples). At this interval, regional vegetation of the sub-Andean forest (12%) is represented by Urticaceae (8%) and Acalypha (4%) and the Andean forest by Rosaceae (32%), Rubiaceae (9%), Quercus (7%), Cyatheaceae (11%), Hedyosmum (4%), Viburnum (4%), and Weinmannia (3%). The Páramo is composed of Poaceae (65%), Lycopodiaceae (20%), and Asteraceae (17%) (Figure 3a). Hydrophytic vegetation such as Cyperaceae dominates throughout the record, but there is a noticeable increase in Plantago and Apiaceae (Hydrocotyle) in the lower and middle parts of the zone, respectively (Figure 3c).

Zone 4 (195–103 cm, 16 samples). The sub-Andean forest vegetation remains constant compared with the previous zone, where Urticaceae (10–12%) and Acalypha (2–3%) are dominant (Figure 3). The Andean forest pollen show pulses of Cyatheaceae (14%), Quercus (8%), Hedyosmum (4%), Juglans (7%), Weinmannia (5%), Melastomataceae (4%), and Alnus (2%). Poaceae (80%), a species considered as an indicator of Páramo conditions, is the dominant taxon, while Caryophyllaceae (20%) and Asteraceae (16%) are recorded in the lower proportions (Figure 3a). Other palynomorphs are mainly represented by fungi (>1000%) and Pteridophyta (40%) (Figure 3b). For this zone, Cyperaceae (90%) are the most abundant taxa of the hydrophytic vegetation. Although Apiaceae (Hydrocotyle) (10%), Gunnera magellanica (10%) and Lachemilla orbiculata (25%) exhibit low values, these species increase in percentage in periods when the Páramo vegetation dominates. Lysipomia has two peaks (35%), which are synchronous with the highest values of the sub-Andean and Andean forest (50% and 35%) (Figure 3c).

Zone 5 (100–51 cm, 9 samples). At the bottom of this zone, the forest vegetation reaches 50%, followed by an abrupt decline (10%), and ends with an increase in the top of the zone (25%). Urticaceae (15%) dominate the sub-Andean forest, whereas Cyatheaceae (14%), Hedyosmum (10%), Viburnum (9%), Quercus (7%), and Melastomataceae (1%) dominate the Andean forest. Páramo vegetation, such as Poaceae (25%), decreases significantly, while Lycopodiaceae (80%) and Asteraceae (15%) increase (Figure 3a). Local vegetation elements are indicated by fungi (>1000%), mosses (80%), and Pteridophyta (50%). This last taxon is only abundant at the beginning of the zone (Figure 3b). In contrast, Testaceae (5%) is recovered throughout the whole record. Hydrophytes are represented by Cyperaceae (95%), Apiaceae (Hydrocotyle) (15%), and Gunnera magellanica (5%). These two last taxa predominate at the beginning of the zone (Figure 3c).

Zone 6 (50–1 cm, 11 samples). The forest vegetation shows a progressive increase toward the top of the zone. Urticaceae (11%) and Acalypha (8%) are dominant in the sub-Andean forest and Cyatheaceae (8%), Quercus (7%), Hedyosmum (5%), Melastomataceae (3%), Clusia (4%), Juglans (3%), and Weinmannia (2%) in the Andean forest. Quercus, Clusia, and Weinmannia show a decreasing trend in the last 15 cm, while the Páramo vegetation signal is represented by Poaceae (20–50%) and Asteraceae (2–10%) which progressively increase and Lycopodiaceae (60–10%) which decrease (Figure 3a). Among the local taxa, Pteridophyta and fungal spores reach 20% and 500%, respectively, and mosses virtually disappear at the end of the record (Figure 3b). Cyperaceae (95–5%) dominate the hydrophytic vegetation with a gradual decrease toward the top. On the other hand, Apiaceae (Hydrocotyle) and Gunnera magellanica percentages increase (Figure 3c).

Figure 3.

(a) Pollen diagram of the El Triunfo core showing individual percentages of pollen taxa. From left to right: Lithology, depth (cm), dating (cal. yr BP), relative percentages of the most important pollen taxa arranged in ecological groups, synthesis diagram of general vegetation, and pollen zones. Letters (A, B, C, D, E, F, and G) and dotted lines indicate the stratigraphic location of discrete volcanic layers. (b) Main groups of palynomorphs. From left to right: Lithology, depth (cm), dating (cal. yr BP), non-aquatic or non-regional groups (black), aquatics (marsh and cushion plants), regional vegetation (sub-Andean, Andean, and Páramo biomes), and pollen zones 1–6. Letters (A, B, C, D, E, F, and G) and dotted lines as in Figure 3a and c. Pollen diagram displaying individual percentage records of pollen taxa, which reflect the aquatic vegetation of the El Triunfo core. From left to right: Lithology, depth (cm), dating (cal. yr BP), the most abundant pollen taxa arranged in ecological groups, pollen zones 1–6, and the main pollen diagram. Letters (A, B, C, D, E, F, and G) and patterns as in Figure 2.

Discussion

The lithostratigraphic and palynological data indicate six periods with significant vegetation changes (5–50%) since ~1930 cal. yr BP (~AD 68). They were probably related to regional and local climatic events. Figure 4 shows the comparison of vegetation belts in different localities of Colombian Páramos.

Figure 4.

Paleovegetation records from northwestern Colombia: (a) Llano Grande II (3460 m) and (b) Puente Largo II (3600 m) in Páramo de Frontino (Velásquez, 2005), (c) Lagunares de Santa Isabel (4300 m), and (d) El Triunfo (3800 m) (this work). Lines connect the same periods of time.

Period 1: ~1930–1360 cal. yr BP/AD 128–591 (zone 1). In this zone, Páramo vegetation was dominant, indicating its presence nearby the coring site; however, variations in its abundance suggest temperature fluctuations. High values of Cyperaceae and Plantago cushions at ~1930–1800 cal. yr BP (~AD 128–195) indicate a marsh system with some dry areas. The increase in Apiaceae (Hydrocotyle) and Lachemilla orbiculata between 1800 and 1700 cal. yr BP (AD 195–298) implies increased precipitation, running waters, and thus humid conditions, as is currently the case in the wetland. After this wet pulse, Cyperaceae returned to high proportions suggesting drier conditions (around 1550 cal. yr BP) (AD 335). At the end of the period, the wet conditions returned again (1400 cal. yr BP/AD 550). The increase in forest, cushion plants, Ericaceae shrubs, ferns, and mosses around 1850, 1550 and 1346 cal. yr BP (AD 84,385 and 601) is indicative of short-lived pulses of warm and dry conditions, except in the last pulse when the conditions became wetter. In between these warm pulses, Lycopodiaceae dominated over Poaceae, suggesting very cold conditions.

The forest increase is interpreted as a real climate change and not Superpáramo effect (high percentage of arboreal pollen in elevated areas due to few Páramo vegetation and winds bringing forest pollen) since it was registered simultaneously (~1930–1360 cal. yr BP/AD 128–591) in different regions of Colombia (Figure 4), such as in the Lagunares record, Los Nevados (4300 m a.s.l.) (Salamanca and Noldus, 2003), Páramo de Frontino (3460 m a.s.l.) (Velásquez, 2005; Velásquez and Hooghiemstra, 2013), and La Cocha lake in SW Colombia (2780 m a.s.l.) (González-Carranza et al., 2012). In the Yucatán peninsula, this time marks the beginning of a dry phase corresponding to the Classic Maya period (AD 250–850) (Brenner et al., 2001). In Cariaco, the Titanium record shows a dry tendency similar to that of the Yucatán and the Mg/Ca relation indicates an increase in temperature around 1800 cal. yr BP with a subsequent decline (Haug et al., 2001; Lea et al., 2003). The temperature record is similar to the El Triunfo record, but the moisture signal seems to be in antiphase, changing from dry to wet in our records and from wet to dry in the Caribbean. This climate asymmetry is probably due to variation in position and strength of ITCZ. A close correlation between level changes of Lake Titicaca (Abbott et al., 1997) and our inferred precipitation from El Triunfo reinforces this idea since it is believed that climate out-of-phase between Cariaco basin and Lake Titicaca is due to variations in ITCZ (Black et al., 2007). Thus, it is possible for climate changes occurring in northwestern South America to be contrary to climate changes in northern South America.

Period 2: 1357–1010 cal. yr BP/AD 591–940 (zone 2). Although according to the dominance of Cyperaceae swampy conditions continued, the climate seems to have been wetter than in the previous period, as indicated by the high presence of Lachemilla orbiculata. The maximum values of Poaceae show overall cold conditions except at the end of the period (~1170 cal. yr BP), when a warm and dry phase started as evidenced by the Poaceae decreasing and the Ericaceae shrubs and Asteraceae increasing. Similar climatic conditions were recorded in other places of Colombia (Figure 4) like Páramo de Frontino (Velásquez, 2005), La Cocha (González-Carranza et al., 2012), Reposo 1 and Los Lagos (Rangel-Ch et al., 2005; Velásquez et al., 1999), and Lagunares de Santa Isabel (Salamanca and Noldus, 2003). The cold and wet conditions around 1300 cal. yr BP/AD 648 were not exclusively in northern South America; a phase of maximum ice expanding was registered in Ecuador (Moy et al., 2009). In Cariaco, a temperature decreasing between 1500 and 1200 cal. yr BP/AD 459–723 was also evident (Lea et al., 2003) but with a moisture-decreasing trend (Haug et al., 2001). The conditions in the Perú-Bolivia altiplano (Abbott et al., 1997) were also cold and wet. This simultaneous change in both hemispheres indicates the ITCZ as a likely explanation due to its form and position with regard to the compared sites (Figure 1a).

The end of this period (1200–1010 cal. yr BP) records a gradual change from cold and humid to warm and dry conditions which correspond to the ‘Medieval Warm Period’. Cariaco was similarly warm and dry at ~1060 cal. yr BP and wetter at ~1000 cal. yr BP (Black et al., 2007; Haug et al., 2001). In Yucatán, this phase was very dry with some wet pulses and corresponds to the Maya collapse in the transition between Classic and Postclassic periods (Brenner et al., 2001). Contrary to these records, in the Titicaca region, AD 760–1040 was a wet period (Thompson, 1992). Again a clear asymmetry between Cariaco and Titicaca is evident. Probably in this warm period, the ITCZ was in a northern position as recorded in past warm periods (Schneider et al., 2014). The scope of the west component of ITCZ in the Central region of Colombia could be a good explanation of small observed differences.

Period 3: 1010–701 cal. yr BP/AD 940–1250 (zone 3). The relatively high values of Andean pollen record show that this time interval was a dry and warm period, continuing the temperature increase registered at the end of zone 2. The warming was not a continuous event; instead there were two warm events separated by a cold event around 850 cal. yr BP/~AD 1100, indicated by the increase in Apiaceae (Hydrocotyle) at ~850 cal. yr BP. In the wetlands of El Nevado del Ruiz, Apiaceae (Hydrocotyle) grows nowadays in running water. Warm and dry conditions with some cold and wet pulses continued until 701 cal. yr BP/AD 1250, with a temperature increase of nearly 1.5°C, compared to the last zone. This temperature change is inferred of 15% Andean forest pollen increase, from zone 2 to zone 3, equivalent to an upward movement of Andean forest of 200–300 m with a rate of 0.6°C/100 m (Hooghiemstra, 1984).

Similar climatic conditions, which occur in these places, prevailed across some regions of Colombia (Figure 4) and northwestern South America, such as those registered in Páramo de Frontino in northwestern Colombia (Velásquez and Hooghiemstra, 2013); southern Perú’s Quelccaya ice core, around AD 1000–1490 (Thompson and Mosley-Thompson, 1987, 1989); Titicaca region (Abbott et al., 1997); and western Patagonia (Moy et al., 2009). However, in Cariaco this period was wet with a drier pulse around 850 cal. yr BP (Haug et al., 2001; Lea et al., 2003). This out-of-phase pattern agrees with the postulated climatic conditions in northern South America and Perú–Bolivia altiplano. During warm periods when the ITCZ moves to the most northward position, the Cariaco region receives more precipitation and thus the Altiplano region becomes drier (Haug et al., 2001; Peterson and Haug, 2006). If the ENSO (El Niño) anomaly were stronger during the ‘Medieval Warm Period’, the observed climatic pattern would be the opposite since the ITCZ moves southward during the ENSO phase. This warming had its origin in the Atlantic, and during the ‘Medieval Warm Period’, the ENSO anomaly was probably absent or weaker as postulated by Rein et al. (2004).

Period 4: 701–470 cal. yr BP/AD 1250–1480 (zone 4). The dominance of Páramo vegetation is indicative of cold conditions, especially around 650, 580, 520, and 490 cal. yr BP (AD 1293, 1334, 1430, and 1445), although with short warm events in between. Overall, taking into account the hydrophilic vegetation, the cold periods tended to be wet and the warm ones dry. Inside the wetland, vegetation was dominated by Cyperaceae, but around ~580 and 490 cal. yr BP (~AD 1334 and 1445), Apiaceae (Hydrocotyle), Gunnera magellanica, and Lachemilla orbiculata were important and indicative of wetter events. Similar conditions were recorded in the Páramo de Frontino and Lagunares de Santa Isabel (Salamanca and Noldus, 2003; Velásquez, 2005) (Figure 4). In the lakes La Cocha (González-Carranza et al., 2012) and Titicaca (Abbott et al., 1997; Baker et al., 2003), dry conditions were observed during this period. The clear asymmetry between Cariaco and northwestern South America was due, as stated previously, to variations in the shape, extent, and intensity of the ITCZ on northern South America (Peterson and Haug, 2006; Velásquez, 2005; Velásquez and Hooghiemstra, 2013). This period coincides with the first phase of LIA.

The cold event around 650 cal. yr BP/AD 1293, registered in the Frontino and El Triunfo cores, is possibly related to the Wolf solar minimum, the effect of which has been reported in other parts of the world, especially in the Northern Hemisphere (Stuiver and Quay, 1980). In the Lagunares de Santa Isabel section (IIB), this event could coincide with the cold pulse at the beginning of this zone. The double cold and wet pulse registered between 580 and 470 cal. yr BP/AD 1334–1480 in the Triunfo section is similar to that reported in Llano Grande II and Puente Largo peat cores from Páramo de Frontino (Velásquez, 2005) and La Cocha I (González-Carranza et al., 2012). It could also be equivalent to the Spörer Minimum radiation recorded in the Northern Hemisphere as a double decreasing pulse in solar activity (Stuiver and Quay, 1980).

Period 5: 470–82 cal. yr BP/AD 1480–1867 (zone 5). The wetland had similar conditions to recent time as evidenced by the high values of Cyperaceae and, to a lesser degree, of Apiaceae (Hydrocotyle). The surrounding area was dominated by Páramo vegetation, in particular Lycopodiaceae, except at the beginning of this period (~450 cal. yr BP/~AD 1480) when there was a significant rise in forest pollen (50%) together with other taxa like local ferns, mosses, and fungi. This short event was warm and dry, but the rest of the period was cold and wet. The coldest and wettest event occurred between 230 and 280 cal. yr BP/AD 1716–1590 (72–80 cm) and could represent the Maunder Minimum of solar radiation, characterized by a decrease in sunspots between AD 1645 and 1750 (Maunder, 1904). Similar climate conditions were registered in Páramo de Frontino and Lagunares de Santa Isabel (Figure 4), La Cocha Lake (González-Carranza et al., 2012), ice core records from Perú (Thompson and Mosley-Thompson, 1987, 1989), and Lake Titicaca (Abbott et al., 1997; Baker et al., 2003). In the Cariaco Basin (Haug et al., 2001; Lea et al., 2003), the LIA was cold an dry, and as stated previously, this out-of-phase moisture relation was probably due to the southward migration of the ITCZ.

Period 6: 82 cal. yr BP/AD 1867 to present time (zone 6). The general tendency is the last century is toward warming, as shown by the continuous forest pollen increase recorded in the pollen diagrams (Figure 4). The aquatic vegetation indicates a drying trend as seen in the Páramo de Frontino and Lake La Cocha but contrary to Cariaco basin. It is clear that this asymmetric pattern between northwestern and northern South America registered along late Glacial, early Holocene, and middle Holocene (Velásquez and Hooghiemstra, 2013) continued operating in late Holocene, even when considering very short times scales. If indeed the ENSO anomaly has increased in late Holocene (Yan et al., 2011), the ITCZ strength and location variations seem to be the main causes of the moisture regulation in north South America but with a clear out-of-phase relationship between northwestern and northern regions.

Notwithstanding the warming tendency in the last century, the pollen diagrams show Quercus, Weinmannia, and Clusia decreasing during this period, due to deforestation and the use of these trees as timber material (Álvarez et al., 2007). The large crops of Alnus in the buffer zone, replacing the destroyed forests, is not reflected in the current pollen records, probably due to the long distance of Alnus trees from the wetland in the past, and of course it cannot be considered as a cause of the recent Andean forest pollen increase. On the other hand, it seems that high values of Rumex in the recent pollen records are due to potato crops and extensive livestock surrounding the El Triunfo wetland (Hofstede, 1995; Salamanca, 2003; Verweij et al., 2003).

According to Morales-Benítez (2003), the last 100 years have been characterized by an increase in cultivated vegetation in forest areas due to human settlements during colonization. This human activity in the high mountain includes animal rearing, the cultivation of crops, and deforestation and could also have contributed to evident greenhouse effect recorded in recent years (Hämmerli, 2010).

Conclusions

The pollen and spore record from El Triunfo wetland indicates that Páramo vegetation has prevailed throughout the past 2000 years; however, some oscillations in their relative percentage reflected variations in the temperature and humidity.

The dominance of Cyperaceae in the aquatic vegetation along the record indicates that the basin has been mainly a marsh, but the frequent pulses of cushion plants (dry conditions) and plants growing in flowing water (wet conditions) show possible changes in the wetland physiognomy.

The temperature synchronicity over the past 2000 years in the studied area and different sites of northwestern Colombia, the Caribbean, northern Perú, and Central South America (based on different proxies, many AMS radiocarbon dating, similar age calibration, and decadal resolution) reflects not only the tropical location but also similar responses to regional or global climatic phenomena. The warmest periods were registered at ~1000 cal. yr BP (~AD 1016) (‘Medieval Climatic Optimum’) and in the last century, whereas the coldest ones were around 1200 cal. yr BP (AD 831) and 650–50 cal. yr BP (AD 1368–1759) (LIA). As a rule warm periods were dry and cold periods were wet.

The moisture asynchrony identified among different places of Caribbean region (including Cariaco) and northwestern South America is probably due to minor variations in the shape, size, and latitudinal scope of ITCZ. An intensification of El Niño phenomena could be the cause of warming and dryness during the ‘Medieval Warm period’.

In the last century and particularly in recent decades, intense human activity (farming, agriculture, deforestation, forest fragmentation) has led to an increase in Rumex, the virtual disappearance of Quercus, and a strong decrease in Clusia and Weinmannia.

Vegetation changes are well recorded in El Triunfo wetland for the last two millennia without significant hiatuses. Therefore, this area has great potential for the exploration of Holocene past environments, vegetation changes, climate, and volcanic activity in detail.

Footnotes

Acknowledgements

Thanks to Vicerrectoría de Investigaciones y Posgrados (VIP) of the Universidad de Caldas (Colombia) for financing this Project. The Instituto de Investigaciones en Estratigrafía (IIES, Universidad de Caldas) gave important logistic support. Thanks to Mauricio Reyes, Antonio García, and Daniel Jaramillo for their help in the laboratory work. Valentina Ramírez, Juan David Vallejo, Paula Andrea López, Sergio Rivera, and Cristian Guacaneme were very useful in the fieldwork logistics. Thanks to Néstor Fabio Alzate and David Sanín, professors of the Universidad de Caldas, for their valuable advices. Thanks to all the personnel of the Paleoecology Laboratory of the Nacional University of Colombia (Medellín) for their technical support. Thanks to Nathan Fisher, Juan Pablo Betancourt, and Fernando Henao for their support in the English grammar, maps, and art graphs respectively. Special thanks to Maria I Velez (University of Regina, Canadá), Jaime Escobar (Universidad del Norte, Barranquilla), Orlando Rangel (Universidad Nacional, Bogotá), Catalina González (University of Los Andes, Bogotá), and two anonymous reviewers for their critical reading of the manuscript.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Abbott

Binford

Brenner

et al . (1997) A 3500 ¹⁴C yr high-resolution record of water-level changes in Lake Titicaca, Bolivia/Peru. Quaternary Research 47: 169–180.

Álvarez

Sanín

Alzate

et al . (2007) Plantas de la región centro-sur de Caldas-Colombia (Cuadernos de Investigación). Manizales: Universidad de Caldas, 52 pp.

ASF DAAC (2015) ALOS PALSAR_Radiometric_Terrain_Corrected_low_res; Includes Material © JAXA/METI 2007. Available at: https://www.asf.alaska.edu/ (accessed through ASF DAAC 26 August 2016). DOI: 10.5067/JBYK3J6HFSVF.

Baker

Fritz

Seltzer

et al . (2003) Late Quaternary paleoclimatic history of tropical South America from drilling Lake Titicaca and the Salar de Uyuni. Eos, Transactions American Geophysical Union 84: 46 (Abstract PP21A-07).

Blaauw

Christen

(2011) Flexible paleoclimate age-depth models using an autoregressive gamma process. Bayesian Analysis 6: 457–474.

Black

Abahazi

Thunell

et al . (2007) An 8-century tropical Atlantic SST record from the Cariaco Basin: Baseline variability, twentieth-century warming, and Atlantic hurricane frequency. Paleoceanography 22: PA4204.

Bogotá

Groot

MHM

Hooghiemstra

et al . (2011) Rapid climate change from north Andean Lake Fúquene pollen records driven by obliquity: Implications for a basin-wide biostratigraphic zonation. Quaternary Science Reviews 30: 3321–3337.

Boom

Mora

Cleef

et al . (2001) High altitude C4 grasslands in the northern Andes relicts from glacial conditions? Review of Palaeobotany and Palynology 115: 147–160.

Bosman

Hooghiemstra

Cleef

(1994) Holocene mire development and climatic change from a high Andean Plantago rigida cushion mire. The Holocene 4(3): 233–243.

10.

Brenner

Hodell

Curtis

et al . (2001) Abrupt climate change and pre-Columbian cultural collapse. In: Markgraf

(ed.) Interhemispheric Climate Linkages. San Diego, CA: Academic Press, pp. 7–104.

11.

Bush

Hanselman

Hooghiemstra

(2007) Andean montane forests and climate change. In: Bush

Flenley

Gosling

(eds) Tropical Rainforest Responses to Climatic Change. Berlin, Heidelberg: Springer, pp. 33–54.

12.

Chepstow-Lusty

Frogley

Bauer

et al . (2003) A late-Holocene record of arid events from the Cuzco region, Peru. Journal of Quaternary Science 18: 491–502.

13.

Cuatrecasas

(1958) Aspectos de la vegetación natural de Colombia. Revista de la Academia Colombiana de Ciencias Exactas, Fisicas y Naturales 10(40): 221–268.

14.

DWH DIVER (2015) Web application: Deepwater horizon data integration visualization exploration and reporting application, National Oceanic and Atmospheric Administration. Available at: https://www.asf.alaska.edu/about/how-to-cite-data/ (accessed 26 August 2016).

15.

Flórez

(1992) Los Nevados de Colombia Glaciales y Glaciaciones. Análisis Geográficos, 22. Bogotá: Instituto Geográfico Agustín Codazzi, 95 pp.

16.

Gómez

Berrío

Hooghiemstra

et al . (2007) A Holocene pollen record of vegetation change and human impact from Pantano de Vargas, an intra-Andean basin of Duitama, Colombia. Review of Palaeobotany and Palynology 145: 143–157.

17.

González-Carranza

Hooghiemstra

Vélez

(2012) Major altitudinal shifts in Andean vegetation on the Amazonian flank show temporary loss of biota in the Holocene. The Holocene 22: 1227–1241.

18.

Grabandt

RAJ

(1980) Pollen rain in relation to arboreal vegetation in the Colombian Cordillera Oriental. Review of Palaeobotany and Palynology 29: 65–147.

19.

Grimm

(1987) CONNIS: A Fortran 77 program for stratigraphically constrained cluster analysis by the method of the incremental sum squares. Computers & Geosciences 13: 13–35.

20.

Groot

MHM

Bogotá

Lourens

et al . (2011) Ultra-high resolution pollen record from the northern Andes reveals rapid shifts in montane climates within the last two glacial cycles. Climate of the past 7: 299–316.

21.

Hämmerli

(2010) A review of multiproxy temperature and precipitation reconstructions for South America during the Holocene. MSc Thesis, University of Bern, 61 pp.

22.

Haug

Haugen

Sigman

et al . (2001) Southward migration of the Intertropical Convergence Zone through the Holocene. Science 293: 1304–1308.

23.

Herd

(1982) Glacial and volcanic geology of the Ruiz-Tolima Volcanic Complex Cordillera Central, Colombia. Publicaciones geológicas especiales del Ingeominas 8: 1–48.

24.

Hofstede

RGM

(1995) Effects of burning and grazing on a Colombian Páramo ecosystem. PhD Thesis, University of Amsterdam, 199 pp.

25.

Hooghiemstra

(1984) Vegetational and climatic history of the high plain of Bogotá, Colombia: A continuous record of the last 3.5 million years. Vaduz: J. Cramer, 368 pp.

26.

Hooghiemstra

(1989) Quaternary and upper-Pliocene glaciations and forest development in the Tropical Andes evidence from a long high-resolution pollen record from the sedimentary basin of Bogotá, Colombia. Palaeogeography, Palaeoclimatology, Palaeoecology 72: 11–26.

27.

Hooghiemstra

Wijninga

Cleef

(2006) The paleobotanical record of Colombia: Implications for biogeography and biodiversity. Annals of the Missouri Botanical Garden 93: 297–324.

28.

IPCC (2007) Cambio climático 2007: Informe de síntesis. In: Pachauri

Reisinger

(eds) Contribución de los Grupos de trabajo I, II y III al Cuarto Informe de evaluación del Grupo Intergubernamental de Expertos sobre el Cambio Climático. Ginebra: IPCC, p. 104.

29.

Kuhry

(1988) Palaeobotanical-palaeoecological studies of tropical high Andean peatbog section (Cordillera Oriental, Colombia). Dissertationes Botanicae 116. Vaduz: J. Cramer, 241 pp.

30.

Lea

Pak

Peterson

et al . (2003) Synchroneity of tropical and high-latitude Atlantic temperatures over the last glacial termination. Science 301: 1361–1364.

31.

Marchant

Behling

Berrío

et al . (2001b) Mid-to Late-Holocene pollen-based biome reconstructions for Colombia. Quaternary Science Reviews 20: 1289–1308.

32.

Marchant

Behling

Berrío

et al . (2002) Pollen-based biome reconstructions for Colombia at 3000, 6000, 9000, 12000, 15000 and 18000, 14C yr ago: Late Quaternary tropical vegetations dynamics. Journal of Quaternary Science 17(2): 113–129.

33.

Marchant

Berrío

Behling

et al . (2006) Colombian dry moist forest transitions in the Llanos Orientales – A comparison of model and pollen-based biome reconstructions. Palaeogeography, Palaeoclimatology, Palaeoecology 234(1): 28–44.

34.

Marchant

Berrío

Cleef

et al . (2001a) A reconstruction of Colombian biomes derived from modern pollen data along an altitude gradient. Review of Palaeobotany and Palynology 117: 79–92.

35.

Marchant

Boom

Behling

et al . (2004) Colombian vegetation at the Last Glacial Maximum: A comparison of model- and pollen-based biome reconstruction. Journal of Quaternary Science 19: 721–732.

36.

Maunder

(1904) Note on the distribution of sun-spots in heliographic latitude, 1874–1902. Monthly Notices of the Royal Astronomical Society 64: 747–761.

37.

Melief

ABM

(1989a) Late Quaternary history of vegetation in the Parque Los Nevados and surroundings (Cordillera Central). In: Van der Hammen

Díaz-Piedrahita

Alvarez

(eds) La Cordillera Central Colombiana. Transecto Parque Los Nevados (segunda parte) (Studies on Tropical Andean Ecosystems, vol. 3). Berlin; Stuttgart: J. Cramer, pp. 537–588.

38.

Melief

ABM

(1989b) Relation of pollen rain and vegetation in the Parque Los Nevados transect. In: Van der Hammen

Díaz-Piedrahita

Alvarez

(eds) La Cordillera Central Colombiana. Transecto Parque Los Nevados (segunda parte) (Studies on Tropical Andean Ecosystems, vol. 3). Berlin; Stuttgart: J. Cramer, pp. 517–536.

39.

Mesa

Carvajal

Salazar

et al . (1997) Introducción al clima de Colombia. Bogotá: Universidad Nacional de Colombia, 390 pp.

40.

Mommersteeg

Hooghiemstra

Loutre

et al . (2005) Orbital forcing of climate change in Colombia from Fuquene-7C and Funza- 2A high- resolution pollen records: The last 170 kyr. In: Berger

Ercegovac

Mesinger

(eds) Milutin Milanovitch Anniversary Symposium: Paleoclimate and the Earth Climate System. Belgrade: Serbian Academy of Sciences and Arts, pp. 113–129.

41.

Morales-Benítez

(2003) La Montaña de la dura cerviz. Medellín: Consejo de Medellín. Biblioteca Pública Piloto, 510 pp.

42.

Moy

Moreno

Dunbar

et al . (2009) Climate change in Southern South America during the last two millennia. In: Vimeux

Sylvestre

Khodri

(eds) Past Climate Variability in South America and Surroundings Regions, from the Last Glacial Maximum to the Holocene, vol. 1. Dordrecht: Springer, pp. 353–393.

43.

Munsell (2014) Munsell Rock Color Book. Grand Rapids, MI: Munsell Color.

44.

Murillo

Bless

(1974) Spores of recent Colombian Pteridophyta. I: Trilete spores. Review of Palaeobotany and Palynology 25: 319–365.

45.

Parques Nacionales Naturales de Colombia (2007) Plan de Manejo 2007–2011. Parque Nacional Natural Los Nevados, Dirección Territorial Noroccidente, Medellín, p. 37.

46.

Peterson

Haug

(2006) Variability in the mean latitude of the Atlantic Intertropical Convergence Zone as recorded by riverine input of sediments to the Cariaco Basin (Venezuela). Palaeogeography, Palaeoclimatology, Palaeoecology 234: 97–113.

47.

Poveda

Álvarez

Rueda

(2011) Hydro-climatic variability over the Andes of Colombia associated with ENSO: A review of climatic processes and their impact on one of the Earth’s most important biodiversity hotspots. Climate Dynamics 36: 2233–2249.

48.

Poveda

Jaramillo

Gil

et al . (2001) Seasonality in ENSO-related precipitation, river discharges, soil moisture and vegetation index in Colombia. Water Resources Research 37(8): 2169–2178.

49.

Rangel-Ch

Moyano

Van der Hammen

(2005) Estudio palinológico del Holoceno en la parte alta del macizo Tatamá. In: Van der Hammen

Rangel-Ch

Cleef

(eds) La cordillera Occidental, transecto de Tatamá (Estudios de Ecosistemas Tropandinos-Ecoandes, vol. 6). Berlín; Stuttgaart: J. Cramer (BORNTRAEGER), pp. 757–796.

50.

Rangel-Ch

Idrobo

Cleef

et al . (1995) Segunda lista del material herborizado en el transecto Parque los Nevados. In: Van der Hammen

Dos Santos

(eds) Estudios de ecosistemas tropoandinos: La cordillera Central de Colombia transecto parque los Nevados (Studies on Tropical Andean Ecosystems, vol. 5). Berlin; Stuttgart: J. Cramer, p. 545.

51.

Rein

Lückge

Sirocko

(2004) A major Holocene ENSO anomaly during the Medieval period. Geophysical Research Letters 31: L17211.

52.

Salamanca

(2003) Recovery of the paramo vegetation after the 1985 eruption of the Ruiz Volcano. In: Van der Hammen

Dos Santos

(eds) La Cordillera Central Colombiana. Transecto Parque Los Nevados (Studies on Tropical Andean Ecosystems, vol. 5). Berlin; Stuttgart: J. Cramer, pp. 347–391.

53.

Salamanca

Noldus

(2003) Paleoecological analysis of the lagunares de Santa Isabel section. In: Van der Hammen

Dos Santos

(eds) La Cordillera Central Colombiana. Transecto Parque Los Nevados (Studies on Tropical Andean Ecosystems). Berlin; Stuttgart: J. Cramer, pp. 393–420.

54.

Salomons

(1986) Paleoecology of volcanic soils in the Colombian Central Cordillera (Parque Nacional Natural Los Nevados). Berlin; Stuttgart: J. Cramer, 212 pp.

55.

Schneider

Bischoff

Haug

(2014) Migrations and dynamics of the intertropical convergence zone. Nature 513: 45–53.

56.

Stuiver

Quay

(1980) Changes in atmospheric Carbon-14 attributed to a variable sun. Science 207: 11–19.

57.

Stuiver

Reimer

(2015) Calib radiocarbon calibration, execute version 7.1html. Available at: http://calib.qub.ac.uk/calib/ (accessed 27 February 2015).

58.

Thompson

(1992) Ice core evidence from Peru and China. In: Bradley

Jones

(eds) Climate since A.D. 1500. London; New York: Routledge, pp. 517–547.

59.

Thompson

Mosley-Thompson

(1987) Evidence of abrupt climatic during the last 1,500 years recorded in ice cores from the tropical Quelccaya ice cap, Peru. In: Berger

Labeyrie

(eds) Abrupt Climatic Change – Evidence and Implications. NATO Advanced Science Institutes Series C: Mathematical and Physical Sciences 216. Dordrecht: D. Reidel Publishing Co., pp. 99–110.

60.

Thompson

Mosley-Thompson

(1989) One-half millennia of tropical climate variability as recorded in the stratigraphy of the Quelccaya Ice Cap, Peru. In: Peterson

(ed.) Aspects of Climate Variability in the Pacific and the Western Americas. Geophysical Monograph 55. Washington, DC: American Geophysical Union, 445 pp.

61.

Thouret

Van der Hammen

(1983) La secuencia Holocénica y Tardiglacial en el Parque Los Nevados. In: Van der Hammen

Pérez

Pinto

(eds) La Cordillera Central Colombiana. Transecto Parque Los Nevados (Studies on Tropical Andean Ecosystems, vol. 1). Alemania: J. Cramer, pp. 262–276.

62.

Thouret

Van der Hammen

Juvingé

et al . (1995) Geologia del cuaternario reciente en el macizo volcánico del Ruiz-Tolima (Cordillera Central). In: Van Der Hammen

Dos Santos

(eds) La Cordillera Central Colombiana. Transecto Parque Los Nevados (Studies on Tropical Andean Ecosystems, vol. 4). Berlin; Stuttgart: J. Cramer, pp. 183–239.

63.

Van der Hammen

Hooghiemstra

(1995) The El abra stadial, a younger dryas equivalent in Colombia. Quaternary Science Reviews 14: 841–851.

64.

Van der Hammen

Hooghiemstra

(2003) Interglacial-glacial Fuquene-3 pollen record from Colombia: An Eemian to Holocene climate record. Global and Planetary Change 36: 181–199.

65.

Velásquez

(1999) Atlas palinológico de la flora vascular paramuna de Colombia: Angiospermae. Medellín: Universidad Nacional de Colombia, Colciencias, 173 pp.

66.

Velásquez

(2005) Paleoecología de alta resolución del Holoceno tardío en el Páramo de Frontino Antioquia. Medellín: Universidad Nacional de Colombia sede Medellín, 199 pp.

67.

Velásquez

Hooghiemstra

(2013) Pollen-based 17-kyr forest dynamics and climate change from the Western Cordillera of Colombia; no-analogue associations and temporarily lost biomes. Review of Paleobotany and Palynology 194: 38–49.

68.

Velásquez

Cabrera

Parra

et al . (2004) Aspectos metodológicos a tener en cuenta en estudios paleoecológicos de alta resolución en los páramos andinos. In: Rangel

Aguirre

Andrade

et al . (eds) Memorias Octavo Congreso latinoamericano y segundo colombiano de Botánica. Bogotá: Instituto de Ciencias Naturales, Universidad Nacional de Colombia, pp. 76–95.

69.

Velásquez

Henao

(2007) Diagnóstico de los humedales altoandinos de Caldas. Fundación Pangea, 151 pp. Available at: http://www.corpocaldas.gov.co/publicaciones/640/Diagn%C3%B3stico%20Humedales%20Altonandinos.pdf.

70.

Velásquez

Parra

Sánchez

et al . (1999) Tardiglacial y Holoceno del norte de la cordillera Occidental de Colombia. Medellín: Universidad Nacional de Colombia & COLCIENCIAS, 236 pp.

71.

Verweij

Kok

Budde

(2003) Aspectos de la transformación del páramo por el hombre. In: Van der Hammen

Dos Santos

(eds) La Cordillera Central Colombiana. Transecto Parque Los Nevados (Studies on Tropical Andean Ecosystems, vol. 5). Berlin; Stuttgart: J. Cramer, pp. 329–452.

72.

Wanner

Beer

Bütikofer

et al . (2008) Mid- to Late-Holocene climate change: An overview. Quaternary Science Reviews 27: 1791–1828.

73.

Weng

Hooghiemstra

Duivenvoorden

(2007) Response of pollen diversity to the climate-driven altitudinal shift of vegetation in the Colombian Andes. Philosophical Transactions of the Royal Society of London, Series B: Biological Sciences 362: 253–262.

74.

Yan

Sun

Wang

et al . (2011) A record of the Southern Oscillation Index for the past 2,000 years from precipitation proxies. Nature Geoscience 4: 611–614.