Environmental changes in the highlands of the western Andean Cordillera,southern Peru,during the Holocene

Abstract

In the tropical Andes, impacts of both natural and anthropogenic disturbances have been detected over a period exceeding 4000 years. However, the history of the environment remains unknown in most Andean regions. To infer possible interactions between climate and humans, we analysed the pollen content of an 8.5 m deep peat core extracted from a peat bog located near the Nevado Coropuna volcano on the slope of the Western Cordillera in southern Peru (15°30S, 72°40W, 6380 m). Results showed that taxa of the upper Puna expanded when cooler and moister climatic conditions prevailed during a time period that includes the early and the mid Holocene (9700–5200 cal. yr BP). An increase in shrub pollen frequencies and a decrease in the Poaceae/Asteraceae ratio are attributed to a drier climate during the late Holocene (5200–3000 cal. yr BP). After 3000 cal. yr BP, the vegetation cover resembled that of today. Both archaeological and pollen data attest to the beginning of agriculture from 2200 cal. yr BP. Around 900 cal. yr BP, the vegetation cover suddenly changed, probably because of a colder and drier climate. Past societies continued their agricultural activities despite this abrupt change. Our results emphasize the specific geographical location of Nevado Coropuna astride the Western Cordillera and the western edge of the Altiplano, which is consequently subject to both Pacific and Atlantic influences.

Keywords

climatic change Holocene human impact pollen analysis Puna tropical Andes

Introduction

Long-term climate records are necessary to reveal the influence of climate changes on human societies and of man’s impact on the environment (Brenner et al., 2001; Hodell et al., 1995; Weiss and Bradley, 2001), to increase our knowledge of global climate changes and to calibrate numerical simulations of global circulation (Markgraf et al., 2002; Messerli et al., 1993; Smol and Cumming, 2000; Sterken et al., 2006; Thompson et al., 1998). In the tropical Andes, both climate and humans have impacted the landscape. Several human communities, including the Inca Empire, occupied the tropical Andes for thousands of years while changes in climate and vegetation occurred under the influence of both Pacific and Atlantic circulations. In recent decades, palaeoenvironmental studies have been carried out from southern Ecuador to northern Chile in the Central Andes revealing contrasted situations depending on the latitude (e.g. Abbott et al., 2003). For instance, during the mid Holocene, between 8000 and 5000 cal. yr BP, the region of Lake Titicaca was characterized by a dry episode with the lowest water level recorded in 30 000 years (Abbott et al., 1997a; Baker et al., 2001; Cross et al., 2001; Hanselman et al., 2005; Mourguiart, 1990; Paduano et al., 2003; Seltzer et al., 1998; Ybert, 1988). New radiocarbon dates obtained from recently exposed plant remnants at high elevations (Nevado Quelccaya, eastern Cordillera, southern Peru) also attest to a minimal extension of the glacier at 5200 cal. yr BP (Buffen et al., 2009). In contrast, in northern Chile, packrat middens and geomorphological analyses attest to a more humid episode during the mid Holocene in what is today the Atacama Desert (Betancourt et al., 2000; Quade et al., 2001). Strong latitudinal contrasts may explain these regional differences based on modern settings and observations of the different atmospheric circulation patterns between the Altiplano and the Atacama region characterising a transition area with either an Atlantic or a Pacific influence on the yearly distribution of precipitation (Vuille and Keimig, 2004). Another abrupt and drastic dry episode was recorded at 850 cal. yr BP at different locations in the Central Andes, e.g. Lake Titicaca (Abbott et al., 1997b; Binford et al., 1997) and Nevado Quelccaya (Ortloff and Kolata, 1993; Thompson et al., 1985). This event has been linked with the disappearance of the Tiwanaku civilization (Binford et al., 1997).

However, since their original expansion, economic activities and urban centres have mainly been located on the Pacific coast and in the western Cordillera where the climate is arid, with less than 250 mm precipitation per year. The development of human communities and civilizations that lasted for centuries in these dry areas was possible because man adapted to his environment and water was available from melting glaciers (Chepstow-Lusty et al., 2009). Detailed knowledge on related changes in climate and societies at high elevations is still needed to evaluate the future of some urban centres in a scenario without glaciers and given the predicted changes in the distribution of precipitation (Bradley et al., 2006; Urrutia and Vuille, 2009). To improve the current knowledge on climate and environmental changes in the western Cordillera, we present a new pollen record located in an inhabited highland near the Nevado Coropuna in southwestern Peru. Our interpretation of fossil pollen data is supported by published modern pollen rain results based on 24 samples collected from different types of vegetation (Kuentz et al., 2007). We combined archaeological data with pollen analysis to evaluate the past impact and the adaptation of human activities to arid landscapes.

Modern setting: Climate and vegetation in southern Peru

The Nevado Coropuna (15°30S, 72°40W) is one of the Quaternary volcanoes that form the Peruvian volcanic arc in the Central Andes. It straddles the western Cordillera and the western edge of the Altiplano at an altitude of 6380 m a.s.l. The western Cordillera is characterized by steep temperature and precipitation gradients over a distance of only 120 km between the arid Pacific coastline and the snowline (now c. 5400 m a.s.l.) at latitude 15°S.

Today, variations in insolation between austral winter and summer are high and are responsible for shifts in the Intertropical Convergence Zone (ITCZ). The southern position of the ITCZ allows penetration of humid air masses from the Atlantic favored by the easterly wind circulation termed South American Summer Monsoon SASM (Vuille and Keimig, 2004; Zhou and Lau, 1998) when summer precipitation occurs on the Altiplano and around Nevado Coropuna. About 70–80% of total precipitation falls in summer. On the other hand, a dominance of the westerly wind circulation can favor the installation of dry conditions in our study area. Seasonal and interannual variability is considerable because of an increase/decrease in the surface temperature of the Pacific Ocean during El Niño/La Niña phases of the Southern Oscillation (ENSO). Strong El Niño events such as the one in 1998 can cause dramatic flooding along the northern Peruvian coast (Fontugne et al., 1999), and retreat of mountain glaciers in northern Bolivia (Francou et al., 2003) and in southern Peru (Forget et al., 2008; Thouret et al., 2002). Prolonged La Niña-like conditions are induced by cool ocean surface temperature anomalies in the central and eastern tropical Pacific (Trouet et al., 2009), and can lead to glacier expansion (Arnaud et al., 2001; Francou et al., 2004).

The main vegetation type in the tropical Andes is high elevation grassland also called Puna, which covers an area of c. 85 000 km². Today, Puna grows between 3500 and 5200 m a.s.l., between the steppe and the snowline in the western Cordillera of Peru, and between the Andean rain forest and the snowline in the eastern Cordillera of Peru (Kuentz et al., 2007) (Figure 1A). Five altitudinal vegetation types were identified around Nevado Coropuna and are described in Kuentz et al. (2007) and Kuentz (2009) (Table 1 and Figure 1B). These vegetation types include, from the lowest to the highest in elevation: (1) The Ambrosia steppe (800–2500 m a.s.l.) is mostly composed of Ambrosia shrubs. (2) In the pre-Puna (2500–3500 m a.s.l.), the vegetation cover is denser and includes different shrubs of the Asteraceae family (Ambrosia, Diplostephium, Senecio). (3) The lower Puna (3500–4000 m a.s.l.) is composed of Asteraceae shrubs and herbaceous species of the Fabaceae and Solanaceae families. (4) In the Puna sensu stricto (4000–4800 m a.s.l.) the grassland is mainly composed of grasses (e.g. Calamagrostis, Festuca, Stipa) and shrubs of the Asteraceae family (Baccharis, Chuquiraga, Parastrephia, Senecio) (UNESCO, 1981). Patches of Polylepis forest grow around 4400 m a.s.l. The flat bottoms of glacially shaped valleys are often occupied by peat bogs (Cyperaceae, Juncaceae). (5) In the upper Puna (4800–5200 m a.s.l.), the sparse vegetation is composed of Poaceae and Asteraceae shrubs (Baccharis, Senecio, Wermeria). Cushions of Apiaceae and Caryophyllaceae families are abundant around 5000 m a.s.l.

Table 1.

Vegetation types around Nevado Coropuna with their corresponding pollen indicators and climatic parameters (from Kuentz et al., 2007)

Vegetation types	Vegetation sub-types	Main species present	Representative ollen taxa	Mean annual temperatures	Mean annual rainfall
Upper Puna (4800–5200 m)		Apiaceae Asteraceae Caryophyllaceae Poaceae	Apiaceae (≥30%)	−2–6°C	100–230 mm
Puna sensu strict (4000–4800 m)	Peat bogs	Ranunculaceae Juncaceae Cyperaceae	Cichorioideae (≥1%), Cyperaceae (≥1%)	0–7°C	80–230 mm
Puna sensu strict (4000–4800 m)	Polylepis woodland	Asteraceae Polylepis	Polylepis (≥1%)	3–8°C	80–100 mm
Lower Puna (3500–4000 m)		Asteraceae Cactaceae Fabaceae Solanaceae	Senecio (≥70%)	8–12°C	50–100 mm
Pre-Puna (2500–3500 m)		Ambrosia Diplostephium Senecio, Tarasa	Senecio (≥30%), Malvaceae (≥15%)	10–14°C	40–100 mm
Steppe (800–2500 m)		Ambrosia	Ambrosia (≥80%)	14–24°C	40–100 mm

Figure 1.

The biogeographical and palaeoenvironmental setting of the Nevado Coropuna volcano. (A) Location of Nevado Coropuna in the study area, main vegetation belts in Peru, and palaeoenvironmental record sites quoted in this study. (B) Map of the vegetation types around Nevado Coropuna with the location of the archaeological sites. Distinct symbols are keyed to the archaeological chronological frame. Locations of sampled cores are also shown

Materials and methods

Vegetation types were mapped by remote sensing using a 2002 Landsat ETM image (Figure 1B) (Kuentz et al., 2007). A total of 24 soil-surface samples of all vegetation types were collected. In 1998, an 8.5 m deep sedimentary core (COR300) was drilled with a Russian corer in a peat bog located at 4400 m a.s.l. in a large glacially shaped valley that drains the SSW of Nevado Coropuna (Figure 1B). This core was extruded in a vertical position to maintain stratigraphy, sediment samples were wrapped in the field and shipped unopened in their polycarbonate coring tubes to the University of Montpellier, France. The core (COR300) was analysed with an average sampling resolution of 6 cm. The first 150 cm cover the last millennium and analyses are still in progress. Pollen grains were extracted from the sediments following the standard protocol developed by Fægri and Iversen (1989). The peat is preserved because of slow decomposition under saturated conditions; therefore, significant bioturbation is unlikely. Pollen concentrations were calculated according to the method of Cour (1974). Pollen counts were performed under 600× magnification and morphology was identified by comparison with our reference pollen collection obtained from floristic materials collected in the field and from herbariums (Universidad Nacional San Augustin de Arequipa, Museo de Historia Natural de Lima and Herbario Nacional de Bolivia), as well as by comparison with pollen atlases (Heusser, 1971; Hooghiemstra, 1984). More than 300 pollen grains were counted per sample. The total pollen sum included arboreal and non-arboreal pollen, but excluding Poaceae, Cyperaceae and Juncaceae, which are over-represented with respect to the frequencies of other taxa. Influx of charcoal particles was estimated according to Clark (1982). This is a point counting method to estimate the projected area of charcoal. The pollen diagram was arranged according to results of modern pollen rain analyses. The taxa were organized in accordance with the vegetation strata (trees/shrubs/herbs) except for the peat-bog taxa (local production). These results of the modern pollen rain revealed the presence of pollen taxa at different altitudinal ranges and an altitudinal pollen gradient was deduced: Asteraceae type Ambrosia (1800–2200 m a.s.l.), Malvaceae (2700–3300 m a.s.l.), Asteraceae-type Senecio (3500–4100 m a.s.l.) and Apiaceae (above 4600 m a.s.l.) (Kuentz et al., 2007). Concerning fossil data, an ecological diagram was built to reflect this altitudinal pollen gradient. We extrapolated the precipitation range around Nevado Coropuna based on meteorological data from SENAMHI (Servicio Nacional de Meteorologia e Hidrologia). The Poaceae/Asteraceae (P/A) ratio calculated for our pollen surface samples was plotted and arranged according to a precipitation gradient (Figure 2). Except for one sample (25), the P/A ratio is well correlated with changes in precipitation, in agreement with the humidity index defined by Reese and Liu (2005), and can thus be used to interpret the fossil data. Constrained clustering was performed using Zone software (Juggins, 2005). AMS ¹⁴C datings of seven peat samples was carried out at the Laboratoire de Mesure du Carbone (Gif sur Yvette, France) in 2007. Calibrated ages were obtained with Calib501 software (Stuiver et al., 2005) (Table 2). We applied a Southern Hemisphere correction to all samples (McCormac et al., 2004). Data on human occupation during the Holocene was acquired from previous archaeological and geographical studies (Forget et al., 2008; Ziolkowski, 2005; Ziolkowski and Belán Franco, 2001) and georeferenced in a Geographic Information System.

Figure 2.

(A) Map showing the extrapolation of the precipitation range around Nevado Coropuna based on data from SENAMHI (Servicio Nacional de Meteorologia e Hidrologia) with the location of the pollen surface samples (from Kuentz et al., 2007). (B) Poaceae/Asteraceae ratio of the soil-surface samples

Table 2.

Radiocarbon dates for the COR 300 peat core with calibrated ages using a Southern Hemispheric correction according to CALIB 5.0.1 (Stuiver et al., 2005)

Depth (cm)	Laboratory code	δ¹³C	¹⁴C age yr BP	Cal. yr BP (2σ)	Age ad/bc (2σ)
277–283	SacA 7628	−24.60	1260 ± 30	1120 ± 65	ad 830 ± 65
470–475	SacA 7629	−25.00	2705 ± 30	2790 ± 55	840 ± 55 bc
578–583	SacA 7630	−25.00	4300 ± 30	4835 ± 35	2885 ± 35 bc
681–686	SacA 7631	−26.40	4555 ± 35	5171 ± 134	3221 ± 34 bc
781–786	SacA 7632	−25.30	5220 ± 35	5936 ± 62	3986 ± 62 bc
883–888	SacA 7633	−31.70	5905 ± 40	6644 ± 112	4694 ± 112 bc
991–996	SacA 7634	−25.20	8675 ± 45	9608 ± 90	7658 ± 90 bc

Present human and archaeological data

Present human activities are described as a function of the altitudinal vegetation types found from low to high elevation. The Ambrosia steppe is irrigated in valley floors and used for growing onions, sugar cane, potatoes, and rice. In the pre-Puna, vegetables, maize and fodder crops are the main crops. Pasture is also extensive in this vegetation type. Fuel wood is provided by planting eucalyptus and pines. In the lower Puna, dry agriculture with tubercles (Oxalis, Solanum), Quinoa and Kiwicha predominate. In the Puna sensu stricto, when the ground used for livestock breeding becomes dry and compacted by animals such as lamas and alpacas, the Poaceae Stippa ichu predominates. When livestock graze humid soils, Pycnophyllum molle prevails. In this vegetation type, villages generally depend on glacier meltwater. Polylepis forests are frequently harvested for fuel as they are the only source of wood at this elevation. Medicinal plants have also been identified in this vegetation type (Brack Egg, 1999). In the upper Puna, the impact of human activity is limited to temporary camps and religious monuments.

Numerous archaeological sites located in southern coastal Peru provided evidence for a first occupation by hunter-gatherers between 13 000 and 11 000 cal. yr BP (Sandweiss, 2003; Sandweiss et al., 1998). High elevation areas are thought to have been colonized after c. 11 000 yr BP as evidenced by ancient paths (Aldenderfer, 1999; Baied and Wheeler, 1993; Marchant et al., 2004). The domestication of the camel family in the Andes dates back to 6000 yr BP (Guffroy, 1999; Jakubicka and Woloszyn, 2005). At Nevado Coropuna, two caves occupied by hunter-gatherers have been identified and dated before 5000 cal. yr BP (Figure 1B). No archaeological artefacts have been found for the period between 3000 and 2300 cal. yr BP around Coropuna. Between 2300 and 1000 cal. yr BP, four archaeological sites have been identified in the region of Coropuna (Figure 1B), all located in the pre-Puna vegetation type at 3000 m a.s.l. As many as 30 archaeological sites dated between 1000 and 700 cal. yr BP are located in the pre-Puna, lower Puna and Puna sensu stricto. They attest to the spread and diversification of pastoral and agricultural activities between 2500 and 4800 m a.s.l. at that time. During the Inca Empire (ad 1340–1540), the upper Puna vegetation type (4800 to 5200 m a.s.l.) was probably colonized to raise livestock. The large number (41) of archaeological sites dating from the Inca period attests to the extent of human occupation at that time. Spanish colonization resulted in the redistribution of land and habitats throughout the region.

Results

Chronology and stratigraphy

The core was composed of brown peat with four thin layers of non-organic material (silt and/or clay). The chronology of COR300 core is based on seven AMS radiocarbon dates from bulk. As no age reversal was observed, these dates were all used to reconstruct a timescale. The age model was drawn with Kaleidagraph software from a smooth curve (Figure 3). The oldest age, at the base of the core, was estimated to be 9750 cal. yr BP. Sedimentation rates varied over time. From 9750 to 6600 cal. yr BP, the average rate of sedimentation was 0.5 mm/yr, increased between 6600 and 4700 cal. yr BP (1.7 mm/yr on average), but decreased between 4700 and 1500 cal. yr BP, it decreased (0.7 mm/yr on average). Between 1500 and 800 cal. yr BP, the sedimentation rate was very high (3.7 mm/yr on average).

Figure 3.

Radiocarbon dates used to build the age model and changes in sedimentation rates along the core COR 300

Pollen zones

Cluster analysis was performed on the pollen counts to define pollen zones (Figure 4). However, our results did not enable us to draw any subdivisions in the diagram probably because all taxa have the same statistical weight in the cluster. Consequently, we based our pollen zonation on a comparison with modern pollen rain results and used visible changes in the indicator taxa (mainly Ambrosia, Apiaceae, Poaceae, the Poaceae/Asteraceae ratio and pollen concentration). We were able to characterize four pollen zones (Figure 4). Using light microscopy, pollen analysts are unable to distinguish Acaena from Polylepis. Acaena is currently present in warm humid valleys and can be confused with Polylepis in the eastern Cordillera (Chepstow-Lusty et al., 2005). Because Acaena was not observed during our vegetation surveys and has never been recorded by botanists at the Universidad Nacional San Augustin de Arequipa in the study area, we refer only to Polylepis pollen grains.

Figure 4.

Palynological diagram of core COR 300, taken 8 km SSW of the summit of the Nevado Coropuna volcano (Southern Peru). Pollen frequencies, concentrations and influx, charcoal influx, and a summary ecological diagram are presented along a depth scale. The dotted line indicates the beginning of agriculture around Nevado Coropuna

Zone 1. Early and mid Holocene (996–691 cm, 9700–5200 cal. yr BP, 51 samples)

Zone 1 was characterized by high frequencies of Apiaceae (9.5%), Brassicaceae (3.7%), Poaceae (85%) and Polylepis (5.2%) and relatively low frequencies of Ambrosia (16.6%). Charcoal influx was relatively high (1.2 cm²/cm³ per yr). The rate of sediment accumulation was very low (0.3 mm/yr) in the lower part of the zone, but increased to 1.4 mm/yr at a depth of 840 cm (c. 6300 cal. yr BP). Concentrations were relatively high in the lowest part of this zone and reached maximum at a depth of 930 cm, i.e. c. 7500 cal. yr BP with 12 600 grains/g.

Zone 2. Late Holocene 1 (691–481 cm, 5200–3000 cal. yr BP, 36 samples)

This zone was characterized by very low pollen concentrations (363 grains/g) and a decrease in Apiaceae (6%) and Poaceae (72%) frequencies. Ambrosia (17.3%), Ophryosporus (6%) and Chenopodiaceae/Amaranthaceae (12%) frequencies slightly increased. Polylepis frequencies were low (4.1%) and decreased drastically (2%) after c. 4600 cal. yr BP (at 571 cm in depth). At the same time, Apiaceae and Brassicaceae frequencies also decreased (3% and 2%, respectively). Charcoal particles were rare in this zone (influx of 0.5 cm²/cm³ per yr). Sediment accumulation accelerated from 1.4 mm/yr to 2.2 mm/yr (at 613 depth, i.e. c. 4900 cal. yr BP), before dropping to 0.5 mm/yr at c. 4000 cal. yr BP.

Zone 3. Late Holocene 2 (481–199 cm, 3000–900 cal. yr BP, 43 samples)

This zone contained very high frequencies of Ambrosia (26%). Frequencies of Poaceae (82%) and concentrations values (1030 grains/g) increased in comparison with the previous zone. Ophryosporus (10%) and Chenopodiaceae/Amaranthaceae (15%) frequencies increased slightly whereas frequencies of Apiaceae (5%) and Polylepis (2.2%) remained low. Few charcoal particles were counted in this zone (influx of 0.4 cm²/cm³ per yr). The rate of sediment accumulation was high (2.3 mm/yr on average). A break in pollen spectra was identified at c. 2200 cal. yr BP (at a depth of 433 cm). After this date, frequencies of Ambrosia increased (from 22 to 26%) while frequencies of Apiaceae decreased (from 6 to 4%). Solanaceae were occasionally identified from time to time (0.3%) whereas they were uncommon before c. 2200 cal. yr BP (Figure 5).

Figure 5.

Synthetic palynological diagram from core COR 300 (only the taxa cited in the discussion are shown). Pollen frequencies, concentrations, and charcoal influx are presented along a time scale. P1, P2, P3 and P4 are the pollen zones. The dot indicates the domestication of the camelideae in the Andes. The star indicates the beginning of agriculture around Nevado Coropuna

Zone 4. Recent period (199–157 cm, about 900–800 cal. yr BP, 8 samples)

This zone was characterized by a marked decrease in the frequencies of Poaceae (50%), Ambrosia (18%), Ophryosporus (5%), Chenopodiaceae/Amaranthaceae (4%) and an increase in Apiaceae (10%), Caryophyllaceae (17%) and Brassicaceae (4%) frequencies. Pollen concentrations were very low (500 grains/g). Arboreal pollen taxa Polylepis almost disappeared. Pollen grains of Rosaceae (0.4%) were identified for the first time, and Solanaceae (0.3%) were still present. Malvaceae (1%), Fabaceae (1.9%) and Alnus reached their highest frequencies. The influx of charcoal particles remained low (0.3 cm²/cm³ per yr). This zone showed the highest rate of sedimentation (4.1 mm/yr on average).

Discussion

The early–mid Holocene (9700–5200 cal. yr BP): A moist and cold climate

Between 9700 and 5200 cal. yr BP, Apiaceae cushions and Polylepis forest expanded in Nevado Coropuna, whereas Ambrosia shrubs are rarely recorded (Figure 5). According to pollen rain analysis, this assemblage matches a well-developed upper Puna belt and suggests moister and colder conditions than today (Table 1) (Kuentz et al., 2007). However, in the other Andean palaeoenvironmental records, this time interval is generally divided into two distinct climatic periods. The first period of time between c. 10 000 and 7500 cal. yr BP, represents the early Holocene and the second, between 7500 and 5000 cal. yr BP, characterizes the mid Holocene. During the early Holocene, 120 km northwest of Nevado Coropuna, data based on snail shells and loess attest to moister conditions at c. 9000 cal. yr BP (Unkel et al., 2007). Palynological and sedimentological analyses show that 250 km to the east, the level of Lake Titicaca was relatively high between 10 000 and 8500 cal. yr BP (Baker et al., 2001; Paduano et al., 2003). Geomorphological studies on moraines formed during glacial stages attest to the fact that one glacier in northern Bolivia reached low elevations between 10 000 and 7000 cal. yr BP (Smith et al., 2011). Further south, archeology and lake level analyses show similar moist conditions between 12 900 and 8400 cal. yr BP in the Atacama region (22–24°S) (Grosjean et al., 2001; Messerli et al., 1993; Wolfe et al., 2001). The slightly moister and colder conditions deduced in the Nevado Coropuna record could be linked with prevailing La Niña-like conditions resulting in an increase in precipitation on the Altiplano (Arnaud et al., 2001; Francou et al., 2004). La Niña-like conditions are generally linked with colder/warmer tropical Pacific Ocean surface temperatures in the eastern/western parts of the Ocean than today, associated with weaker and less frequent El Niño events (Shin et al., 2006). Various proxy reconstructions inferred such climatic conditions during the early Holocene (Overpeck and Webb, 2000; Sandweiss et al., 2001; Tudhope et al., 2001).

From c. 8000 cal. yr BP, the climate became progressively drier in the southern Tropics and moister in the northern Tropics (Haug et al., 2001). However, close to Nevado Coropuna and at lower elevations, the level of Lake Pacucha (c. 300 km to the northeast) was lower during this interval, but the vegetation remained abundant, attesting to high seasonal moisture rates (Hillyer et al., 2009; Valencia et al., 2010). Conversely palaeo-proxies from Lake Titicaca records attest to extreme aridity characterized by a sharp decrease in the lake level to as much as 100 m below its present level (e.g. Argollo and Mourguiart, 2000; Baker et al., 2001; Hanselman et al., 2005). Geochemical analyses also revealed an arid period between 8600 and 6400 cal. yr BP at Lake Chungará located at the border between Chile and Bolivia (Moreno et al., 2007). Further south (23°S), the record of Laguna Miscanti also reflects an arid climate during the mid Holocene although frequent changes in moisture rates are attested (Grosjean et al., 2001). A northernmost position of the Intertropical Convergence Zone at millennial-scale is observed either by proxy data or by simulations during the mid Holocene (Braconnot et al., 2007). This situation prevented the summer rainy season in the southern Tropics and produced a dry climate similar to that recorded in the Altiplano. Further south, in the Atacama region, the situation was different. Analyses performed on lake terraces around Lake Aricota (17°S), U/Th analysis of the sediment from the Salar of Atacama (23°S) and fossil rodent middens recorded in the central Atacama (22–24°S) revealed a moister environment during the mid Holocene than at present (Betancourt et al., 2000; Bobst et al., 2001; Latorre et al., 2005; Placzek and Quade, 2001; Placzek et al., 2006). The interpretation of drier mid-Holocene conditions around the Titicaca basin and conversely moister conditions in the Atacama region has been the subject of debate in the scientific community (Betancourt et al., 2000; Grosjean, 2001; Quade et al., 2001). In reality, the different proxies used by these authors (packrat-middens versus lake levels) recorded climate variability at various timescales: consequently the mid-Holocene period was generally dry but included humid intervals at annual and decadal scales (Grosjean et al., 2003). The specific location of Nevado Coropuna in the transition area between the Altiplano and the Pacific slope of the Central Andes could explain why no break in the vegetation cover was observed between the early and the mid Holocene, in contrast to other records in the Central Andes. During the early Holocene, the Titicaca basin, the Atacama region and Nevado Coropuna were correlated, showing similar moister conditions. In contrast, during the mid Holocene, climatic dynamics may have differed between the Altiplano and the western slopes of the Andes. During this period, the Nevado Coropuna record was similar to records of the Atacama region but differed from records of the Altiplano. These conclusions support the hypothesis of a highly variable precipitation pattern over the Central Andes (Vuille and Keimig, 2004; Vuille et al., 2000).

Mid to late Holocene: Installation of drier climatic conditions

Between 5200 and 3000 cal. yr BP, the vegetation around Nevado Coropuna was made up of a majority of shrubs of the Puna sensu stricto. After the early Holocene, the herbaceous vegetation (majority of Poaceae) decreased in favour of the shrub vegetation (Asteraceae mainly). The Poaceae/Asteraceae (P/A) ratio and the pollen concentrations were low between 5200 and 3000 cal. yr BP, attesting to dry climatic conditions (Figures 2 and 6). During this dry interval, the size of the Polylepis forests decreased considerably. For this period, only two caves probably occupied by hunter-gatherers have been identified in the region. Although agriculture started at least as early as 4000 cal yr BP in southern Peru as attested to for instance at Waynuna (Perry et al., 2006), human presence around Coropuna was rare at this time and we consequently excluded any major human impact such as deforestation or terraces constructed for cultivation purposes. At the same latitude as Coropuna but 150 km further east, the Titicaca basin showed arid conditions followed by a slow rise in the lake level after 5000 cal yr BP (Abbott et al., 2003) (Figure 6). On the Nevado Sajama (400 km southeast of the Nevado Coropuna), between 5000 and 3400 cal. yr BP, high dust concentrations in the ice core and low snow accumulation suggest drier conditions (Thompson et al., 1998) (Figure 6). At the end of this interval, around 3000 cal. yr BP, there was an increase in the P/A ratio and in pollen concentrations in the Coropuna pollen record: this suggests the return of moister climatic conditions in the central Andes, in agreement with the ice core record of Nevado Sajama. We interpret these climate changes as being linked with precession cycles and with the southward displacement of ITCZ, which influenced precipitation on the Altiplano (Placzek et al., 2006). Between c. 5000 and 3000 cal. yr BP, the increase of the summer insolation allowed a broader-scale effect of the upper-air easterlies that brought summer precipitation to the Titicaca basin. However we infer that it was only from 3000 cal. yr BP that the intensity of the upper-air easterlies was high enough to allow a weakening of the westerlies that led to a summer precipitation climate regime on Nevados Sajama and Coropuna, as described by Vuille and Keimig (2004) (Figure 6). In addition, the increase in El Niño frequencies observed during this period could also have led to the dry conditions observed at Sajama and Coropuna (Maldonado and Villagrán, 2006; Moy et al., 2002; Riedinger et al., 2002; Rodbell et al., 1999). In conclusion, between 5200 and 3000 cal. yr BP, the human impact alone cannot explain the observed changes in vegetation cover and, here again, the influence of climate prevails.

Figure 6.

Comparison between proxy data for palaeo-precipitations in the Central Andes and results of palynological analysis of core COR 300. (A) Fluctuations in the level of Lake Titicaca, as inferred from δ¹³C measurements (Abbott et al., 2003). (B) Number of particles >0.63 µm/ml and snow accumulation in the Nevado Sajama ice core (Thompson et al., 1998). (C) Poaceae/Asteraceae (P/A) ratio and pollen concentration (Cn) in COR 300 core. (D) January insolation at 15°S (Berger, 1992)

Human impact on vegetation

Between 2200 and 900 cal. yr BP, the taxa indicator of the upper Puna (mainly Apiaceae) decreased while the herbaceous and shrub vegetation (mainly Ambrosia) indicator of the lower Puna increased (Figure 5). This pollen assemblage characterizes similar climatic conditions as today. The expansion of Ambrosia and the regression of Apiaceae could also reflect an increase in temperature and/or drier climatic conditions (Table 1) (Kuentz et al., 2007). One possibility is that these changes were linked with the increase in El Niño frequencies that characterized this time period (Maldonado and Villagrán, 2006; Moy et al., 2002; Riedinger et al., 2002). However, we know that an increase in ENSO frequencies could not have impacted the climate of the Central Andes because neither Lake Titicaca nor the Sajama ice core recorded drier conditions during this period (Figure 6). Around the Nevado Coropuna, climatic conditions were relatively humid with summer precipitation as revealed by the increase in the P/A ratio (Figure 6). A more probable hypothesis is that changes in the vegetation cover were caused by human activities. Indeed, around the Nevado Coropuna, four archaeological sites date back to the Intermediate Period (2300–1000 cal. yr BP) at 3000 m a.s.l., and are linked with the first agricultural activities in the region (Figure 1B) (Kuentz, 2009). The increase in Ambrosia could also reflect the plantation of these shrubs to stabilize agricultural terraces (Chepstow-Lusty et al., 1998). The increase in Solanaceae and Chenopodiaceae/Amaranthaceae, which are native cultivated taxa, also supports this hypothesis (Figure 5). In addition, the fact that Azorella compacta was one of the main sources of fuel for human populations may explain the regression of Apiaceae cushions in the area. Both archaeological and pollen data attest to high human presence and to the beginning of agriculture from 2200 cal. yr BP in the region of Coropuna. Therefore changes in cover vegetation during this period should be attributed to past societies and less to climatic changes.

An abrupt and short environmental change (ad 970–1130)

Between 980 and 820 cal. yr BP (about ad 970–1130), Apiaceae and Caryophyllaceae cushions expanded at Nevado Coropuna. In modern samples, these taxa are associated with the upper Puna altitudinal belt and linked to colder and/or moister climatic conditions (Kuentz et al., 2007). As Poaceae frequencies decreased considerably, moister conditions can be excluded. Moreover, the total pollen concentrations also dropped. East of Nevado Coropuna, the level of Lake Titicaca decreased at 850 cal. yr BP (Abbott et al., 1997b; Binford et al., 1997) and the Quelccaya ice cap also retreated (Ortloff and Kolata, 1993; Thompson et al., 1985). These phenomena are attributed to a short dry climatic episode. The hypothesis of significant volcanic activity that would have disturbed the development of the vegetation is rejected because at this time regional volcanic activity was low around Nevado Coropuna (Thouret et al., 2002). On the other hand, the presence of 30 archaeological sites attributed to the second Intermediate Period (1000–700 cal. yr BP), suggests extensive human occupation during this time interval in different altitudinal belts (pre-Puna, lower Puna and Puna sensu stricto) (Figure 1B). Several palaeo-irrigation canals identified around the Nevado Coropuna may also date back to this period (Forget et al., 2008). An increase in cultivated taxa such as Brassicaceae, Fabaceae, Rosaceae and Malvaceae was also observed in the pollen record (Figure 5). Today fodder plants from these families are cultivated around the volcano, and beans and lupin (Fabaceae) provide the basic food of the population. The expansion of Rosaceae and Malvaceae may also reflect cultivation of fruit trees at lower elevations. A marked increase in human activities was also recorded elsewhere in the Central Andes. For example, between ad 830 and 960, a high dust content observed in the Nevado Quelccaya ice core (Thompson et al., 1988) is attributed to the development of cultivated terraces on the Altiplano (Erickson, 2000). Near Cusco, Chenopodiaceae/Amaranthaceae and maize pollen grains have been identified around ad 920 (Chepstow-Lusty et al., 2009). In conclusion, this short episode is characterized by an abrupt change in plant species assemblages that are not farmable and may be linked with colder and drier conditions. Nevertheless, this climatic change did not prevent past societies from developing and expanding their agricultural activities. Their supremacy and skill in irrigation technologies enabled them to settle in the rather hostile environment which prevailed at this time.

Charcoal record of the core COR300

Because fires were rare in the Amazon basin between 11 000 and 7500 cal. yr BP (Power et al., 2008), the source of the charcoal particles observed in core COR 300 near Nevado Coropuna during this period could be linked with local storms and/or volcanism. On the other hand, between 8000 and 4000 cal. yr BP, fires increased in the Amazon basin and on the eastern slopes of the Andes (Mayle and Power, 2007). Therefore, after 8000 cal. yr BP, the charcoal particles observed in the sediment around Nevado Coropuna may have been carried aloft from their area of origin in the south and southeast.

Despite increasing fires in tropical South America around 3000 cal. yr BP (Power et al., 2008), the frequency of charcoal particles remained low at Nevado Coropuna after the mid Holocene. Our results show that the amount of biomass that could be used as fuel in these arid landscapes was reduced, with rare Polylepis trees and Azorella compacta cushions, thus preventing the occurrence of big fires. However, the presence of small local fires cannot be excluded, either man-induced or natural (e.g. volcanic eruptions in the Andahua field of monogenetic volcanoes 25–30 km east of Nevado Coropuna or thunderstorms).

Conclusion

The ‘Nevado Coropuna’ record is the first pollen record located in the transition area between the western Cordillera and the Altiplano in southern Peru, and which covers the whole Holocene. Our research shows that major environmental changes occurred throughout the Holocene even in the dry vegetation cover at high elevations in the Central Andes. These changes were linked with climatic variability of the Central Andes distinguishing and alternating the influence of Pacific or Atlantic circulations, which is in agreement with modern observations and model predictions for the twenty-first century in the Central Andes (Urrutia and Vuille, 2009; Vuille et al., 2000). The moist conditions that prevailed during the early and mid Holocene (9700–5200 cal. yr BP) were correlated with prolonged La Niña-like conditions and cooler surface temperatures of the Pacific Ocean (Carré et al., 2011), whereas the drier late-Holocene (5200–3000 cal. yr BP) climate was attributed to a weaker influence of the westerlies. Agriculture appears to have begun later in this region than in other parts of the Central Andes but the succession of late-Holocene and historical civilizations strongly impacted the vegetation cover from 2200 cal. yr BP onward. Between 980 and 820 cal. yr BP an abrupt environmental change was probably due to colder and drier conditions although it did not affect agricultural activities. In the twenty-first century, climate simulations (e.g. Intergovernmental Panel on Climate Change (IPCC), 2007) predict an increase in aridity in the southern tropical Andes and/or an increase in rainfall in the northern tropical Andes (Solomon et al., 2009; Urrutia and Vuille, 2009). Many tropical glaciers will almost completely melt in the next 50 years and the continuous increase in the population will lead to more livestock grazing in the highlands (Bradley et al., 2006).The contrasted patterns observed at Coropuna during the Holocene confirm the fragility of these high-elevation environments in the face of a changing climate and reveal the importance of geographical location in evaluating the consequences of probable changes in the lower tropical belt (Seidel et al., 2008). These observations underline the need for more environmental studies in the tropical Andean highlands if appropriate strategies are to be found to enable the populations to survive in a drying landscape, because water resources usually used for irrigation will become scarce in the near future.

Footnotes

Acknowledgements

Radiocarbon dating was performed at the Laboratoire de Mesure du Carbone 14 (LMC14) – UMS 2572 (CEA/DSM – CNRS – IRD – IRSN – Ministère de la culture et de la communication) at Gif sur Yvette, France. This research is part of the IRD (Institut de Recherche pour le Développement) ‘GREAT ICE’ programme and was funded by IFEA (Institut Français des Etudes Andines), IRD and the CERAMAC laboratory (University Clermont-Ferrand 2). AK thanks Denis Didier Rousseau, Rachid Cheddadi and the ‘Paléoenvironnements et Paléoclimats’ team of ISEM, Stephan Beck and the National Herbarium at La Paz (Bolivia) for welcoming her during her PhD. Thanks to an anonymous reviewer who has significantly improved the manuscript.

References

Abbott

Binford

Brenner

Kelts

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

Abbott

Seltzer

Kelts

Southon

(1997a) Holocene paleohydrology of the Tropical Andes from Lake Records. Quaternary Research 47: 70–80.

Abbott

Wolfe

Seltzer

Aravena

Mark

. (2003) Holocene paleohydrology and glacial history of the central Andes using multiproxy lake sediment studies. Palaeogeography, Palaeoclimatology, Palaeoecology 194: 123–138.

Aldenderfer

(1999) The Pleistocene/Holocene transition in Peru and its effects upon human use of the landscape. Quaternary International 54: 11–19.

Argollo

Mourguiart

(2000) Late Quaternary climate history of the Bolivian Altiplano. Quaternary International 72: 37–51.

Arnaud

Muller

Veuille

Ribstein

(2001) El Niño-Southern Oscillation (ENSO) influence on a Sajama volcano glacier (Bolivia) from 1963 to 1998 as seen from Landsat data and aerial photography. Journal of Geophysical Research 106: 17 773–17 784.

Baied

Wheeler

(1993) Evolution of high Andean puna ecosystems: Environment, climate, and culture change over the last 12,000 years in the Central Andes. Mountain Research and Development 13: 145–156.

Baker

Seltzer

Fritz

Dunbar

Grove

Tapia

. (2001) The history of South American Tropical precipitation for the past 25,000 years. Science 291: 640–643.

Betancourt

Latorre

Rech

Quade

Rylander

(2000) A 22,000-year record of monsoonal precipitation from northern Chile’s Atacama Desert. Science 289: 1542–1546.

10.

Binford

Kolata

Brenner

Janusek

Seddon

Abbott

. (1997) Climate variation and the rise and fall of an Andean civilization. Quaternary Research 47: 235–248.

11.

Bobst

Lowenstein

Jordan

Godfrey

Luo

(2001) A 106 ka paleoclimate record from drill core of the Salar de Atacama, northern Chile. Palaeogeography, Palaeoclimatology, Palaeoecology 173: 21–42.

12.

Brack Egg

(1999) Diccionario enciclopedico de plantas utiles del Peru. Cusco: Centro de Estudios Regionales ‘Bartolomé de las Casas’.

13.

Braconnot

Otto-Bliesner

Harrison

Joussaume

Peterchmitt

J-Y

Abe-Ouchi

. (2007) Results of PMIP2 coupled simulations of the mid-Holocene and Last Glacial Maximum – Part 1: Experiments and large-scale features. Climate of the Past 3: 261–277.

14.

Bradley

Vuille

Diaz

Vergara

(2006) Threats to water supplies in the tropical Andes. Science 312: 1755–1756.

15.

Brenner

Hodell

Curtis

Rosenmeier

Binford

Abbott

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

(ed.) Interhemispheric Climate Linkages. San Diego: Academic Press, 87–107.

16.

Buffen

Thompson

Mosley-Thompson

(2009) Recently exposed vegetation reveals Holocene changes in the extent of the Quelccaya Ice Cap, Peru. Quaternary Research 72(2): 157–163.

17.

Carré

Azzoug

Bentaleb

Chase

Fontugne

Jackson

. (2011) Mid-Holocene mean climate in the southeastern Pacific and its influence on South America. Quaternary International doi: 10.1016/j.quaint.2011.02.004.

18.

Chepstow-Lusty

Bennett

Fjeldsa

Kendall

Galiano

Herrera

(1998) Tracing 4000 years of environmental history in the Cuzco area, Peru, from the pollen record. Mountain Research and Development 18: 159–172.

19.

Chepstow-Lusty

Bush

Frogley

Baker

Fritz

Aronson

(2005) Vegetation and climate change on the Bolivian Altiplano between 108,000 and 18,000 yr ago. Quaternary Research 63: 90–98.

20.

Chepstow-Lusty

Frogley

Bauer

Leng

Boessenkool

Carcaillet

. (2009) Putting the rise of the Inca Empire within a climatic and land management context. Climate of the Past 5: 375–388.

21.

Clark

(1982) Point count estimation of charcoal in pollen preparations and thin sections of sediments. Pollen et spores 24(3–4): 523–535.

22.

Cour

(1974) Nouvelles techniques de détection des flux et des retombées polliniques: étude de la sédimentation des pollens et des spores à la surface du sol. Pollen et spores XVI: 103–141.

23.

Cross

Baker

Seltzer

Fritz

Dunbar

(2001) Late Quaternary climate and hydrology of Tropical South America inferred from isotopic and chemical model of Lake Titicaca, Bolivia and Peru. Quaternary Research 56: 1–9.

24.

Erickson

(2000) The Lake Titicaca Basin: Landscape Transformations in the Precolumbian Americas. Lentz

(ed.). Colombia University Press.

25.

Faegri

Iversen

(1989) Textbook of Pollen Analysis. New York: Wiley and Sons.

26.

Fontugne

Usselmann

Lavallée

Julien

Hatté

(1999) El Niño variability in the coastal desert of southern Peru during the mid-Holocene. Quaternary Research 52: 171–179.

27.

Forget

Thouret

Kuentz

Fontugne

(2008) Héritages glaciaires, périglaciaires et évolution récente: le cas du Nevado Coropuna (Andes centrales, sud du Pérou). Géomorphologie: relief, processus, environnement 2: 113–132.

28.

Francou

Vuille

Favier

Caceres

(2004) New evidence for an ENSO impact on low-latitude glaciers: Antizana 15, Andes of Ecuador, 0°28’S. Journal of Geophysical Research 109: 18106.1–18106.17.

29.

Francou

Vuille

Wagnon

Mendoza

Sicart

(2003) Tropical climate change recorded by a glacier in the central Andes during the last decades of the twentieth century: Chacaltaya, Bolivia, 16°S. Journal of Geophysical Research 108: doi 10.1029/2002JD002959.

30.

Grosjean

(2001) Mid-Holocene climate in the south-central Andes: Humid or dry? Reply. Science 292: 2391–2392.

31.

Grosjean

Cartajena

Geyh

Nuñez

(2003) From proxy data to paleoclimate interpretation the mid-Holocene paradox of the Atacama Desert, northern Chile. Palaeogeography, Palaeoclimatology, Palaeoecology 194: 247–258.

32.

Grosjean

Leeuwen

Van der Knaap

Geyh

Ammann

Tanner

. (2001) A 22,000 ¹⁴C year BP sediment and pollen record of climate change from Laguna Miscanti (23°S), northern Chile. Global and Planetary Change 28: 35–51.

33.

Guffroy

(1999) El arte rupestre del Antiguo Peru. Lima: Instituto Frances de Estudios Andinos (IFEA)-Institut de Recherche pour le Développement (IRD).

34.

Hanselman

Gosling

Paduano

Bush

(2005) Contrasting pollen histories of MIS 5e and the Holocene from Lake Titicaca (Bolivia/Peru). Journal of Quaternary Science 20(7–8): 663–670.

35.

Haug

Hughen

Sigman

Peterson

Rohl

(2001) Southward migration of the intertropical convergence zone through the Holocene. Science 293(5533): 1304–1308.

36.

Heusser

(1971) Pollen and Spores of Chile, Modern Types of the Pteridophyta, Gymnospermae, and Angiospermae. Tucson: The University of Arizona press.

37.

Hodell

Curtis

Brenner

(1995) Possible role of climate in the collapse of classic Maya civilization. Nature 375: 391–394.

38.

Hooghiemstra

(1984) Vegetational and Climatic History of the High Plain of Bogotá, Colombia: A Continuous Record of the Last 3.5 Million Years. Vaduz, Liechtenstein: Cramer.

39.

Hillyer

Valencia

Bush

Silman

Steinitz-Kannan

(2009) A 24,700-yr paleolimnological history from the Peruvian Andes. Quaternary Research 71: 71–82.

40.

Intergovernmental Panel on Climate Change (2007) Climate Change 2007: Synthesis Report. Core Writing Team, Pachauri

Reisinger

(eds) Contribution of Working Groups I, II and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Geneva: Intergovernmental Panel on Climate Change.

41.

Jakubicka

Woloszyn

(2005) Pinturas rupestres de Pintasayoc. In: Ziolkowski

Belán Franco

Sobczyk

(eds) Proyecto Arquelogico Condesuyos (vol III). Varsovia: Université de Varsovie, 109–136.

42.

Juggins

(2005) http://www.staff.ncl.ac.uk/staff/stephen.juggins/software.htm.

43.

Kuentz

(2009) Dynamiques actuelle et holocène de la Puna (Andes sèches du Pérou) à partir des observations de terrain, de la cartographie (SIG) et de la palynologie (Région du Nevado Coropuna). PhD. Clermont-Ferrand: Université Blaise Pascal, Universitaires Européennes, 256 pp.

44.

Kuentz

Galán de Mera

Ledru

Thouret

(2007) Phytogeographical data and modern pollen rain of the puna belt in southern Peru (Nevado Coropuna, Western Cordillera). Journal of Biogeography 34: 1762–1776.

45.

Latorre

Betancourt

Rech

Quade

Holmgren

Placzek

. (2005) Late Quaternary history of the Atacama Desert. In: Smith

Hesse

Woldendorp

(eds) 23 Degrees South: Archaeology and Environmental History of the Southern Deserts. National Museum of Australia Press, 73–90.

46.

Maldonado

Villagrán

(2006) Climate variability over the last 9900 cal yr BP from a swamp forest pollen record along the semiarid coast of Chile. Quaternary Research 66(2): 246–258.

47.

Marchant

Hooghiemstra

Islebe

(2004) The rise and fall of Peruvian and Central American civilizations: Interconnections with Holocene climatic change – A necessarily complex model. In: Yasuda

Shinde

(eds) Monsoons and Civilizations. New Delhi: Rolli Books, 351–376.

48.

Markgraf

Baumgartner

Bradbury

Diaz

Dunbar

Luckman

. (2002) Paleoclimate reconstruction along the Pole-Equator-Pole transect of the Americas (PEP 1). Quaternary Science Reviews 19: 125–140.

49.

Mayle

Power

(2007) Impact of a drier early–mid-Holocene climate upon Amazonian forests. Philosophical Transactions of the Royal Society 363: 1829–1838.

50.

McCormac

Hogg

Blackwell

Buck

Higham

TFG

Reimer

(2004) SHCal04 Southern Hemisphere calibration 0 – 1000 cal BP. Radiocarbon 46: 1087–1092.

51.

Messerli

Grosjean

Bonani

Burgi

Geyh

Graf

. (1993) Climate change and natural resource dynamics of the Atacama Desert during the last 18,000 years: A preliminar synthesis. Mountain Research and Development 13(2): 117–127.

52.

Moreno

Giralt

Valero-Garcés

Sáez

Bao

Prego

. (2007) A 14 kyr record of the tropical Andes: The Lago Chungará sequence (18°S, northern Chilean Altiplano). Quaternary International 161: 4–21.

53.

Mourguiart

(1990) Une approche nouvelle du problème posé par les reconstructions des paléoniveaux lacustres: utilisation d’une fonction de transfert basée sur les faunes d’ostracodes. Géodynamique 5: 151–166.

54.

Moy

Setzer

Rodbell

Anderson

(2002) Variability of El Niño/Southern Oscillation activity at millennial timescale during the Holocene epoch. Nature 420: 162–165.

55.

Ortloff

Kolata

(1993) Climate and collapse: Agroecological perspectives on the decline of the Tiwanaku state. Journal of Archaeological Science 20: 195–221.

56.

Overpeck

Webb

(2000) Non-glacial rapid climate events: Past and future. Proceedings of the National Academy of Sciences of the United States of America 97: 1335–1338.

57.

Paduano

Bush

Baker

Fritz

Seltzer

(2003) A vegetation and fire history of Lake Titicaca since the Last Glacial Maximum. Palaeogeography, Palaeoclimatology, Palaeoecology 194: 259–279.

58.

Perry

Sandweiss

Piperno

Rademaker

Malpass

Umire

. (2006) Early maize agriculture and interzonal interaction in southern Peru. Nature 440(2): 76–79.

59.

Placzek

Quade

(2001) Holocene lake-level fluctuations of Lake Aricota, Southern Peru. Quaternary Research 56: 181–190.

60.

Placzek

Quade

Patchett

(2006) Geochronology and stratigraphy of late Pleistocene lake cycles on the southern Bolivian Altiplano: Implications for causes of tropical climate change. Geological Society of America Bulletin 118(5/6): 515–532.

61.

Power

Marlon

Ortiz

Bartlein

Harrison

Mayle

. (2008) Changes in fire regimes since the Last Glacial Maximum: An assessment based on a global synthesis and analysis of charcoal data. Climate Dynamics 30(7–8): 887–907.

62.

Quade

Rech

Betancourt

Latorre

(2001) Mid-Holocene climate in the south-central Andes: Humid or dry? Response. Science 292: 2391.

63.

Reese

Liu

(2005) Inter-annual variability in pollen dispersal and deposition on the tropical Quelccaya ice cap. Professional Geographer 57(2): 185–197.

64.

Riedinger

Steinitz-Kannan

Last

Brenner

(2002) A 6100 ¹⁴C yr record of El Niño activity from the Galapagos Islands. Journal of Paleolimnology 27: 1–7.

65.

Rodbell

Seltzer

Anderson

Abbott

Enfield

Newman

(1999) A 15,000-year record of El Niño driven alluviation in southwestern Ecuador. Science 283: 516–520.

66.

Sandweiss

(2003) Terminal Pleistocene through mid-Holocene archaeological sites as paleoclimatic archives for the Peruvian coast. Palaeogeography, Palaeoclimatology, Palaeoecology 194: 23–40.

67.

Sandweiss

Maasch

Burger

Richardson

Rollins

Clement

(2001) Variation in Holocene El Niño frequencies: Climate records and cultural consequences in ancient Peru. Geological Society of America Bulletin 29(7): 603–606.

68.

Sandweiss

Mclnnis

Burger

Cano

Ojeda

Paredes

. (1998) Quebrada Jaguay: Early South American maritime adaptations. Science 281: 1830–1832.

69.

Seidel

Randel

Reichler

(2008) Widening of the tropical belt in a changing climate. Nature 1: 21–24.

70.

Seltzer

Baker

Cross

Dunbar

Fritz

(1998) High-resolution seismic reflection profiles from Lake Titicaca, Peru-Bolivia: Evidence for Holocene aridity in the tropical Andes. Geology 26(2): 167–170.

71.

Shin

Sardeshmukh

Webb

(2006) Understanding the mid-Holocene climate. American Meteorological Society 19: 2801–2817.

72.

Smith

Lowell

Owen

Caffee

(2011) Late Quaternary glacial chronology on Nevado Illimani, Bolivia, and the implications for paleoclimatic reconstructions across the Andes. Quaternary Research 75: 1–10.

73.

Smol

Cumming

(2000) Tracking long-term changes in climate using algal indicators in lake sediments. Journal of Phycology 36: 986–1011.

74.

Solomon

Plattner

G-K

Knutti

Friedlingstein

(2009) Irreversible climate change due to carbon dioxide emissions. PNAS 106(6): 1704–1709.

75.

Sterken

Sabbe

Chepstow-Lusty

Frogley

Vanhoutte

Verleyen

. (2006) Hydrological and land-use changes in the Cuzco region (Cordillera Oriental, South East Peru) during the last 1200 years: A diatom-based reconstruction. Archiv für Hydrobiologie 165(3): 289–312.

76.

Stuiver

Reimer

(2005) CALIB 5.0. http://calib.qub.ac.uk/calib/.

77.

Thompson

Davis

Mosley-Thompson

Liu

(1988) Pre-Incan agricultural activity recorded in dust layers in two tropical ice cores. Nature 336: 763–765.

78.

Thompson

Davis

Mosley-Thompson

Sowers

Henderson

Zagorodnov

. (1998) A 25,000-year tropical climate history from Bolivian ice cores. Science 282: 1858–1864.

79.

Thompson

Mosley-Thompson

Bolzan

Koci

(1985) A 1500 year record of tropical precipitation recorded in ice cores from the Quelccaya Ice Cap, Peru. Science 229: 971–973.

80.

Thouret

J-C

Juvigné

Marino

Moscol

Legeley-Padovani

Loutsch

. (2002) Late Pleistocene and Holocene tephro-stratigraphy and chronology in Southern Peru. Sociedad Geologica del Peru 93: 45–61.

81.

Trouet

Esper

Graham

Baker

Scourse

Frank

(2009) Persistent positive North Atlantic Oscillation mode dominated the Medieval climate anomaly. Science 324(5923): 78–80.

82.

Tudhope

Chilcott

McCulloch

Cook

Chappell

Ellam

. (2001) Variability in the El Niño-Southern Oscillation through a glacial–interglacial cycle. Science 291: 1511–1517.

83.

UNESCO (1981) Vegetation Map of South America – Explanatory Notes. Paris: Les presses de l’UNESCO.

84.

Unkel

Kadereit

Machtle

Eitel

Kromer

Wagner

. (2007) Dating methods and geomorphic evidence of palaeoenvironmental changes at the eastern margin of the South Peruvian coastal desert (14130°S) before and during the Little Ice Age. Quaternary International 175: 3–28.

85.

Urrutia

Vuille

(2009) Climate change projections for the tropical Andes using a regional climate model: Temperature and precipitation simulations for the end of the 21st century. Journal of Geophysical Research 114: D02108, doi:10.1029/2008JD011021.

86.

Valencia

Urrego

Silman

Bush

(2010) From ice age to modern: A record of landscape change in an Andean cloud forest. Journal of Biogeography 37: 1637–1647.

87.

Vuille

Keimig

(2004) Interannual variability of summertime convective cloudiness and precipitation in the Central Andes derived from ISCCP-B3 data. Journal of Climate 17: 3334–3348.

88.

Vuille

Bradley

Keimig

(2000) Interannual climate variability in the Central Andes and its relation to tropical Pacific and Atlantic forcing. Journal of Geophysical Research 105: 12 447–12 460.

89.

Weiss

Bradley

(2001) What drives societal collapse? Science 291: 609–610.

90.

Wolfe

Aravena

Abbott

Setzer

Gibson

(2001) Reconstruction of paleohydrology and paleohumidity from oxygen isotope records in the Bolivian Andes. Palaeogeography, Palaeoclimatology, Palaeoecology 176: 177–192.

91.

Ybert

(1988) Apports de la palynologie à la connaissance de l’histoire du lac Titicaca (Bolivie-Pérou) au cours du Quaternaire récent. Travaux de la Section Scientifique et Technique, Institute Français de Pondichery XXV: 139–150.

92.

Zhou

Lau

(1998) Does a monsoon climate exist over South America? Journal of Climate 11: 1020–1040.

93.

Ziolkowski

(2005) Apuntes sobre la presencia Inca en la region de los Nevados Coropuna y Solimana. In: Ziolkowski

Belán Franco

Sobczyk

(eds) Proyecto Arquelogico Condesuyos (vo lII). Varsovia: University of Varsovia, 27–63.

94.

Ziolkowski

Belán Franco

(2001) El proyecto arqueologico ‘condesuyos’. Varsovia: University of Varsovia.