Evidence for a Younger Dryas glacial advance in the Andes of northwestern Venezuela
Introduction
The high summits of the eastern cordillera of the Venezuelan Andes lie between 8°30′ and 9°00′ N and 70°30′ and 70°45′ W, at elevations close to 5000 m a.s.l. Along a NE to SW trending drainage divide, ice spilled to the west from Pico Humboldt (Sierra Nevada de Mérida) and Pico Mucuñuque (Sierra Nevada de Santo Domingo—Fig. 1a) during the last (Mérida = Wisconsinan) glaciation (Schubert, 1972, Schubert, 1974, Mahaney et al., 2007a) most significantly through the Mucubají and Coromoto valleys (Fig. 1b and c) which contain glacial geomorphic and sedimentological evidence of Late Pleistocene glacial fluctuations thought to record the Younger Dryas (YD) climatic reversal, the latest event in the Late Glacial (Ralska-Jasiewiczowa et al., 2001, Gibbard, 2004). Both valleys are glacial/fluvial outlets draining across steep gradients (10–35° slopes) into lacustrine basins located between 3000 and 3600 m a.s.l.
The existence of the YD climatic event (11–10 ka; 11.6–12.7 ka cal BP) in tropical South America, as well as its possible timing and/or character, have been subject to considerable discussion and debate, with most evidence derived from either palynological/paleoecological or glacial geomorphological/sedimentological studies in Colombia, Ecuador, Peru, and Bolivia (Salgado-Labouriau and Schubert, 1976, Salgado-Labouriau, 1989, Clapperton, 1990, Hansen, 1995, Osborn et al., 1995, Van der Hammen and Hooghiemstra, 1995, Rodbell and Seltzer, 2000). Heine (1993) criticized much of the available evidence on the basis of crude dating control and/or correlation between sites, and faulty palynological or geomorphological interpretations. While there is growing consensus that one or more climatic reversals did occur during overall deglaciation of the Andes (Osborn et al., 1995), the cumulative evidence suggests earlier cooling events (< 15.5 ka), less well-defined, possibly longer-lived, and with or without associated glacier advances (e.g., Clapperton and McEwen, 1985, Helmens, 1988, Birkeland et al., 1989, Rodbell and Seltzer, 2000). Sites allowing precise dating of glacial advances coeval with the YD as defined for northwestern Europe are extremely limited (Osborn et al., 1995). In summarizing the rather scanty evidence available for Venezuela, Schubert and Clapperton (1990) suggested that glaciers may not have had time to reform during the YD period following their initial Late Pleistocene retreat/disappearance, particularly at lower elevations. Heine (1993) proposed that greater aridity at this time (Curtis et al., 1999, Rodbell and Seltzer, 2000) may have further limited glacier growth despite depressed temperatures.
In the Venezuelan Andes upstream of Laguna Mucubají (3600 m a.s.l.) and Lago Coromoto (3000 m a.s.l.), a distinct series of recessional moraines located between 3200 and 3800 m a.s.l. (Schubert, 1972, Schubert, 1974, Salgado-Labouriau et al., 1977) record several stillstands that occurred following the Last Glacial Maximum (LGM). In the Mucubají Valley, these moraines were previously shown to predate 14.9 ka cal BP based on dated peats exposed in river terraces upstream of the uppermost example (Salgado-Labouriau et al., 1977, Mahaney et al., 2007b). An additional 14C date obtained from the base of a core drilled into bog (over lacustrine) sediments just upvalley of the terrace site (Stansell et al., 2005) provides a revised minimum age for glacier retreat of 15.7 ka cal BP. These ages do not account for additional lacustrine sedimentation below sample depths that may have persisted for several centuries (c.f. Heine, 1993, Dirszowsky et al., 2005). While subsequent glacier fluctuations upvalley may have contributed to variation in glaciofluvial/lacustrine sedimentation and peat accumulation at the Mucubají terrace site (Salgado-Labouriau et al., 1977) as proposed by Stansell et al. (2005), their suggestion that ice readvanced downvalley of the site between 13.8 and 10 ka cal BP is highly improbable given the lack of erosional and/or deformational features in the sediments (Salgado-Labouriau et al., 1977, Mahaney et al., 2007a, Mahaney et al., 2007b).
Based on pollen assemblages within the terrace sediments, Salgado-Labouriau et al. (1977) inferred a sequence of vegetational shifts and climatic variations for the site in which temperatures increased from 14.1 (minimum age) to 13.8 ka cal BP (Mucubají Warm Phase), but then returned to relatively cool conditions (2–3 °C below present) from 13.8 to 13.0 (interpolated) ka cal BP (Mucubají Cold Phase). The latter cold phase was extended to 10.5 ka cal BP (Salgado-Labouriau, 1988), in part based on a much longer pollen record obtained from Páramo de Miranda ∼ 12 km to the NNW (Salgado-Labouriau et al., 1988), and thus may overlap with the YD event. Rull et al. (2005) developed a pollen record for sediments of Laguna Verde Alta near the Páramo de Miranda and concluded that glacial or periglacial conditions persisted until ca. 12.7 ka cal BP with little evidence of glacial advance or climatic reversal before the Holocene. At Los Zerpa, ~ 4 km ENE of the Mucubají Valley, LGM moraines are locally covered (upvalley) by lateral and frontal moraines (La Canoa, La Canoita) evidently developed during a last Pleistocene readvance (Carillo et al., 2006). While not dated directly, these moraines are considered by Carillo et al. (2006) to predate the YD event based, in part, on the Mucubají Valley dates as well as on the relatively minor displacement of the moraines along the Boconó transform fault compared to the LGM moraines.
Within the Mucubají Valley upstream of the previously documented recessional moraine sites and in the Coromoto Valley, new investigations have revealed geomorphologic and stratigraphic information (including surface moraine and outwash fans burying alluvial peat), which appear to indicate a resurgence of ice following the breakup and retreat of glaciers at the end of the Late Glacial and corresponding to the YD. Ice overran older tills, glaciolacustrine, and/or glaciofluvial deposits, leaving buried organic materials dated to between 13.7 and 12.4 ka cal BP together with reworked peat dating to 18.8 ka cal BP. In the Humboldt area, ice overran, buried, and deformed preexisting peat deposits. In the Mucubají Catchment an outwash fan at 3800 m a.s.l., upvalley from a sequence of Late Glacial tills overlies buried alluvial peat dated to between 13.7 and13.3 ka cal BP.
Although several workers [including Schubert (1974), Schubert and Clapperton (1990), Mahaney and Kalm (1996) and Mahaney et al., 2001a, Mahaney et al., 2001b] have discussed the evidence for punctuated glacier recession following the LGM, glacial landforms in the eastern cordillera of Venezuela have not been subjected to detailed mapping or sedimentologic and geochronologic analysis. The areas dealt with in this study, below Pico Humboldt and Pico Mucuñque, witnessed the growth of two of the largest valley glaciers in the entire range during the LGM (Schubert, 1998). The results reported here represent the first terrestrial evidence for resurgence of ice in this area during the YD event complete with bracketing maximum and within-advance radiocarbon ages.
Section snippets
Materials and methods
All sections were dug by hand and cut back to expose fresh sediments. Sediment samples were collected for laboratory analysis and peats and organic clayey silts were collected for radiocarbon dating. Samples were handled with metal implements, air-dried, and wrapped in aluminum foil. Laboratory processing of the samples was carried out with speed: the radiocarbon samples analysed within one month of collection. Radiocarbon samples were washed with distilled water and acid treated to remove
Geomorphic context
The Mucuñuque Catchment shown in Fig. 1b is a nearly linear fault-controlled glacial basin with headwaters on the western slopes of Pico Mucuñuque (4672 m a.s.l.). Trending NW the valley is marked by a number of bedrock bars punctuated about every 100 m by a drop in elevation down to the elevation of Lago Mucubaji at 3600 m a.s.l.). Prominent bogs are found at all levels from 4200 m a.s.l., where the valley bottom is narrowest at ∼ 150 m, to the lower reaches where it widens to ∼ 250 m.
Discussion
The stratigraphies in the two areas indicate different scenarios of ice withdrawal and readvance. In the Humboldt Catchment, punctuated ice withdrawal led to the emplacement of an end moraine in Late Glacial time followed by an ice advance (12.4 ka cal BP) evidently during the YD climatic reversal. In the Mucubají Catchment the Late Glacial appears to have involved two closely related ice advances (marked by two buried tills at site MUM7) during earlier phases of the Late Glacial followed by a
Conclusions
This is the first report of YD-age glacial sediments in the Venezuelan Andes, and the ice activity they record is approximately correlated with similar occurrences in Ecuador (Clapperton and McEwen, 1985, Clapperton, 1990), Patagonia (Hajdas et al., 2003), and possibly Colombia (Van der Hammen and Hooghiemstra, 1995). Most moraines in the South American Cordillera of latest Pleistocene age lack sufficient dating control for correlation or to confirm YD affinity (Osborn et al., 1995). The
Acknowledgements
Funding was provided by Quaternary Surveys, Toronto to WCM. Financial support by Estonian State Target Foundation (project No. 0182530s03) to VK is acknowledged. We thank Norm Catto (Memorial University, St. John's, Newfoundland) and two anonymous referees for their thought-provoking reviews.
References (56)
- et al.
Determination of bacterial manganese oxidation rates in sediments using an in situ dialysis technique, I., Laboratory studies
Geochimica et Cosmochimica Acta
(1983) Composition of some iron–manganese concretions from some New Zealand soils
Geoderma
(1975)- et al.
Lithostratigraphy of the Mérida (Wisconsinan) glaciation and Pedregal interstade, Mérida Andes, northwestern Venezuela
Journal of South American Earth Sciences
(2005) - et al.
Precise radiocarbon dating of Late-Glacial cooling in mid-latitude South America
Quaternary Research
(2003) Late Pleistocene glacial sequence in the area of the high plain of Bogota (eastern cordillera, Colombia)
Palaeogeography, Palaeoclimatology, Palaeoecology
(1988)- et al.
Paleosols and Andean uplift in Venezuela: assessing competing hypotheses of relict tropical soils versus paleohydrogeochemical variations
Journal of South American Earth Sciences
(2002) - et al.
Soil stratigraphy and ecological relationships of a Late Glacial–Holocene fluvial terrace sequence, Sierra Nevada National Park, Northern Venezuelan Andes
Journal of South American Earth Sciences
(2007) - et al.
Potential glacial evidence for the Younger Dryas event in the cordillera of North and South America
Quaternary Science Reviews
(1995) - et al.
Rapid ice margin fluctuations during the Younger Dryas in the Tropical Andes
Quaternary Research
(2000) - et al.
15,000-yr pollen record of vegetation change in the high altitude tropical Andes at Laguna Verde Alta, Venezuela
Quaternary Research
(2005)