* = depth of radiocarbon dates

FIGURE 13 Selected pollen profiles from Laguna de Fuquene in Colombia. (Data from Van Geel and Van der Hammen, 1973.)

the tree line of Andean vegetation (warmer climate). It is unlikely that this site suffered significant moisture deficits in the late Pleistocene.

Sparse paleolimnological data from Laguna de Fuquene and other Colombian records suggest a transition from colder and possibly drier early late glacial climates to warmer and wetter conditions afterwards. A reduction or southerly displacement of northeast trade winds associated with the Intertropical Convergence Zone (ITCZ) seems a logical cause for dry, late glacial environments (Bradbury, 1997). An increase in the water balance could reflect a reestablishment of easterly moisture sources associated with latest glacial and early Holocene increased summer insolation coupled with warm sea surface temperatures (SSTs) and increased fluxes of moisture to the atmosphere.

16.2.13. Laguna Surucucho, Ecuador (2.90S, 78.80W; 3180 m)

Pollen in a 12-m core from Laguna Surucucho on the east flank of the Andes documents abundant Isoetes in probable full and late glacial intervals, although ran dom age reversals make the radiocarbon chronology for this period ambiguous (Fig. 14). Isoetes in presumed full glacial intervals may reflect shallow-water conditions, which are also suggested by sandy, inorganic, clay-rich sediment. However, Isoetes remained abundant during the late glacial period in laminated black gyttja, which probably indicates moderately increased lake productivity and stratification related to warmer summer temperatures. Abundant Polylepis pollen in this interval may imply more effective precipitation. By the Holocene, Isoetes falls to low concentrations, coincident with pollen evidence of an Andean forest with a warmer climate and increased moisture (Colinvaux et al., 1997).

This record, although essentially undated, appears to correlate with other low-latitude, high-elevation records that show high Isoetes values in full or late glacial levels. Inferred shallow-water conditions could suggest modest moisture deficits in large-capacity basins, although the abundance of Isoetes may simply relate to cold, oligotrophic lake conditions and the presence of suitable silty substrates near the lake margin regardless of lake depth.

FIGURE 14 Selected pollen profiles from Laguna Surucucho in Ecuador. (Data from Colinvaux et al., 1997.)

16.2.14. Lago de Junin: West-Central Peru (11.00S, 76.130W; 4100 m)

Silty clay and abundant Isoetes spores characterize the full glacial record of the Junin core (Fig. 15). These indicators are presumed to represent outwash from nearby glaciers entering a cold, shallow, oligotrophic lake (Hansen et al., 1984). Abundant herb (Asteraceae) and Polylepis /Acaena pollen indicate relatively dry, puna-like climates in the vicinity, perhaps at lower elevations (Hansen et al., 1984).

By 15,000 14C B.P., peaty, organic-rich sediments imply greater productivity and decreased influx of glacial silt that lasted to ca. 11,000 14C B.P. Thereafter, banded marl with organic layers suggests more persistent, carbonate-charged groundwater input with biogenic deposition of calcium carbonate from stems and leaves of submerged aquatic macrophytes. Isoetes become rare and pollen (e.g., Moraceae) from east Andean slopes characterize the record throughout the early and middle Holocene (Fig. 15).

The peaty deposits at an extrapolated date of 15,000 14C B.P. suggest low lake levels and possible moisture deficits during the late glacial period, whereas the transition to marl by 11,000 14C B.P. signaled the onset of moister and warmer climates with greater weathering of surrounding Paleozoic carbonate rocks. The appearance of extraregional pollen from the east at that time probably indicates strengthening of the trade winds that serve today as the principal source of moisture in the region. Overall, both the Lago de Junin and the Laguna Surucucho records have similar interpretations.

16.2.15. Lago Titicaca: Bolivian Altiplano (16.00S, 69.00W; 3810 m)

In Lake Titicaca (Huinamarca), ~5 m cores from a water depth of 19 m document lake-level changes reconstructed from sedimentology (Wirrmann, 1987; Wirrmann et al., 1988); Isoetes, pollen, and algae (Ybert, 1992); and diatoms (Servant-Vildary, unpublished data). The interpretation of fossil assemblages is based on modern algal assemblages in surface sediments along two transects located in Lake Titicaca (Ybert, 1992). At present, Botryococcus and Pediastrum are rare (<20%) between 0 and 2 m depth, but increase to 70% between 2 and 4 m and reach 90% between 4 and 200 m. These algae currently associate with the abundant planktonic diatom Cyclotella andina in oligosaline Lake Titicaca (Servant-Vildary, 1992). Percentages of Pedias-trum + Botryococcus in the cores are presumed to indicate similar lake depths in the past and a hydrochemi-cal composition close to that of modern Lake Titicaca. Radiocarbon chronologies for Lake Titicaca require reservoir correction of about —250 years (Abbott et al., 1997).

FIGURE 15 Selected pollen and geochemical profiles from Lago de Junin in west-central Peru. (Data from Hansen et al., 1984.)

Moderate percentages of Botryococcus + Pediastrum in full glacial levels of the core (19,000 14C B.P.) suggest lake elevations similar to those existing now. After 18,200 14C B.P., high percentages of Botryococcus + Pediastrum in a core labeled TD (Fig. 16) indicate freshwater conditions and modern or higher lake levels than those of today at Lake Titicaca. A high stand (Tauca III at 182 cm in core TD1) occurred at 13,200 14C B.P., and an undated lacustrine terrace, located at + 5 m above the present level of Lake Titicaca and attributed to the Tauca phase by Servant and Fontes (1978), suggests that Lake Titicaca could have overflowed to the southern basins (Poopo-Coipasa-Uyuni) at this time. A drastic drop of the lake level was recorded by the disappearance of Botryococcus + Pediastrum before the 9600 14C B.P. date on the correlative algae peak in core TD (Sylvestre et al., 1999). High percentages of fungi after 7000 14C B.P. in core TD1 suggests a hiatus at this Huinamarca locality.

Although a modern 14C reservoir correction is inappropriate for times when the paleolake was lower than it is today, we tentatively correlate this low lake stage

FIGURE 16 Profiles of Botryococcus + Pediastrum from two cores in the Huinamarca basin of Lake Titicaca in Bolivia. (Data from Ybert, 1992.)

with the dry Ticana event. The reappearance of Botry-ococcus + Pediastrum (40%) just after 9600 14C B.P. in a core (TD; Fig. 16) suggests that Lake Titicaca may have been somewhat deeper for a brief period during the early Holocene, which would correspond with increasing lake levels in the Uyuni-Coipasa basin (Coipasa event). However, this increased lake level is not recorded in core TD1 and must be confirmed by further studies in Lake Titicaca.

16.2.16. Salar de Uyuni-Coipasa: Southern Bolivian Altiplano (20.0°S, 67.0°-68.0°W; 3653 m)

The late glacial and early Holocene lake-level changes on the southern Bolivian Altiplano are reconstructed from outcrops and shorelines on the margin of the Uyuni-Coipasa basin (Servant and Fontes, 1978; Servant et al., 1995; Sylvestre et al., 1996; Sylvestre et al., 1999). The elevation of dated lacustrine deposits above the present bottom of the Salar de Uyuni (3653 m) reflects paleolake levels. The ecology of the fossil diatom assemblages documents the limnological character of the paleolakes according to the modern distribution of diatom communities from the Altiplano (Servant-Vildary and Roux, 1990; Sylvestre, 1997) and from other regions when no modern analogs were found (Sylvestre et al., 1996). Fossil diatoms from four selected outcrops on the margin of the basin (Sylvestre et al., 1996) provide the data for this analysis. The lake-level chronology is based on 44 radiocarbon dates on inorganic carbonates. The validity of the radiocarbon ages is verified by comparison with six 230Th/234U ages. This comparison shows excellent agreement between 14C calendar ages and 230Th/234U ages during lake transgressions and high stands. However, during periods of lake regression, apparent 14C ages are greater than U/Th ages, suggesting that radiocarbon ages must be corrected for a reservoir effect.

By ca. 15,500 14C B.P., lacustrine conditions (Tauca Ia) became established on the southern Bolivian Altiplano. At 3657 m (-4 m above the present Salar de Uyu-ni), a thin aragonite crust gave a radiocarbon age of 15,430 ± 80 14C B.P., calibrated to 18,458 cal. B.P. and Th/U dated at 18,860 ± 390 B.P. Fossil diatom assemblages, dominated by benthic Denticula subtilis, indicate shallow, saline water at the lake margin (Fig. 17).

FIGURE 17 Profiles of diatom habitat groups from two outcrops of lacustrine sediment in the Salar de Uyu-ni basin and the reconstructed lake stage elevation profile for the region. Dashed lines represent lake-level changes of less certainty. (Data from Sylvestre et al., 1999; Sylvestre, 1997.)

After 15,500 14C B.P., the level (Tauca Ib) rose rapidly to —27 m above the present salar (Fig. 17). The dominant planktonic diatom, Cyclotella striata, indicated that comparatively deep water covered the 3657-m site. Coeval lacustrine deposits at 3682 m contain the tycho-planktonic diatoms Fragilaria atomus and F. construens subsalina, which indicated water inputs to the Uyuni-Coipasa basin. Between ca. 14,400 and 13,000 14C B.P., the lake level stabilized (Tauca II). Finely laminated sediments at 3685 m contain littoral epiphytic diatoms (Achnanthes brevipes, Cocconeis placentula var. euglypta, Rhopalodia gibberula) and tychoplanktonic diatoms (Staurosirella pinnata). These assemblages indicate shallow water with episodic freshwater input. The lake level is estimated to have been —40 m above the present salar (Fig. 17). The lake level increased abruptly at ca. 13,000 14C B.P. and reached a hydrological maximum with a depth of ca. 100 m above the present bottom of the Salar de Uyuni (Tauca III). Twenty-one radiocarbon ages place this phase between ca. 12,000 and 13,000 14C B.P. An algal bioherm located at 3760 m on the highest shoreline gave a radiocarbon age of 12,930 ± 50, calibrated at 15,330 cal. B.P. and Th/U dated at 15,070 ± 400 B.P. The planktonic diatom Cyclotella striata dominates in all outcrops between 3657 and 3685 m and in the bio-herms at 3760 m. Most outcrop surfaces are eroded, but two radiocarbon ages suggest that the Tauca phase may have ended after ca. 12,000 14C B.P. A sudden drop of the lake level (Ticana event) is recorded at 3657 m by fluvial sands between <12,000 and >9500 14C B.P. These deposits locally contain interbedded clay-silt lenses with mollusk shells and Denticula subtilis that imply shallow residual ponds remained around the main basin, which must have been almost dry. A radiocarbon age (11,980 ± 50 B.P.) on mollusk shells may include a reservoir effect related to old (Tauca stage) groundwater. During the early Holocene period, a slight oscillation of the lake level (Coipasa event) is represented by a calcareous crust widely developed around the basin at 3660 m. This crust supports small algal bioherms dominated by D. subtilis, indicating shallow, saline water. The Coipasa lake level rose at least —7 m above the present salar. Nine radiocarbon ages between 11,400 and 10,500 years on the Coipasa event may be more than —2000 years too old according to comparative radiocarbon and U/Th dates on related samples. These old radiocarbon dates may reflect groundwater discharge of old inorganic carbon from the former lacustrine Tauca transgression (Sylvestre et al., 1999). Consequently, the lacustrine Coipasa event probably occurred between ca. 9500 and 8500 14C B.P. Since the early Holocene, the Salar de Uyuni has remained comparatively low (Sylvestre, 1997).

16.2.17. Laguna Lejía: Atacama Desert, Northern Chile (23.500S, 67.700W; 4325 m)

Diatoms and sediment studies of high-stand lacustrine outcrops on the south side of Laguna Lejía are radiocarbon dated (with variable and often large reservoir corrections) to 11,500-9700 14C B.P. (Geyh et al., 1999). These deposits indicate an initial (11,500 14C B.P.) lake-level rise (2-10 m) from previous lake elevations near modern levels (4325 m) and an abrupt rise to +25 m by <10,800 14C B.P. (Fig. 18). Laguna Lejía then fell to near modern levels shortly afterwards, but rose rapidly thereafter to another high stand (+25 m) during the early Holocene. After 8800 14C B.P., Laguna Lejía progressively fell, reaching modern elevations by ca. 8000 14C B.P. and then desiccating in the mid-Holocene (Grosjean et al., 1995; Geyh et al., 1999).

The highest lake levels, 10-24 m above modern, were previously thought to correlate with the Tauca phase of the Uyuni-Coipasa basins to the north (Messerli et al., 1992), but now appear to postdate the maximum lake-level stages recorded farther north (Geyh et al., 1999). Additional chronological control for the general Atacama lake chronology comes from methodologically reliable (i.e., reservoir effect-free) peat intercalated with diatomite and archaeological charcoal associated with the maximum paleobeach ridges (Geyh et al., 1999). The revised chronology of the Lejía/ Atacama region would indicate an approximate correlation of high lake stands to the Coipasa lake-level rise in the Titicaca/Uyuni area (Fig. 18).

Annual precipitation is estimated to be as much as 2.8 times modern values at the lake-level maximum (Grosjean et al., 1995). Tropical easterly moisture sources expanded to the south and southwest during these times, and increased cloudiness is thought to have maintained the Atacama lakes (Grosjean and Nuñez, 1994).

16.2.18. Salinas del Bebedero: San Luis Province, Argentina (33.330S, 66.750W; 380 m)

Two paleolimnological data sets document past lake-level stages at Salinas del Bebedero. Radiocarbon-dated beach ridges (Fig. 19) indicate high-stand lake elevations of + 25 to + 20 m above the present-day playa surface between 20,100 and 13,300 14C B.P. (González, 1994). Salt-tolerant gastropods associated with the beach ridges reflect dissolution of Mesozoic evaporites in the drainage area of the Río Desaguadero and, as a consequence, the saline composition of the river and the lake basin (Déletang, 1929).

FIGURE 18 (A) Lake stage elevation histogram for Laguna Lejía in Chile. (B) Lake stage elevation curves for the Titicaca-Uyuni region in Bolivia and Laguna Lejía in Chile, and histogram of radiocarbon dates on archaeological material from the Atacama region of Chile. (Data from Geyh et al., 1999.)

By the end of the late glacial period, falling lake levels are recorded by outcrops of diatomaceous silt and clay exposed in the Arroyo Bebedero 5-10 m above the modern playa surface. Freshwater benthic diatoms in deposits dated at between 11,600 and 10,700 14C B.P. and between 10,100 and 9600 14C B.P. alternate with open- (deep-?) water, saline planktonic diatoms that apparently indicate lake-level fluctuations (Fig. 19). Low, but comparatively fresh lake stages may have reflected less saline base flow conditions in Arroyo Bebedero that supported marsh environments surrounding the basin and drained to a shallow saline playa at the basin center. Open-water, planktonic, saline diatoms (Cyclotella choctawatcheeana) characterize the intervening high lake stages (10,700-10,100 and 9600-9100 14C B.P.) and imply abundant surface flow from the saline Río Desaguadero that entered Salinas del Bebedero via Arroyo Bebedero (González and Maidana, 1998). A hiatus and archaeological remains on beach ridges within the basin 9100-8600 14C B.P. suggest a dry period, although the lake rose afterward and remained open and at generally intermediate stages for an unknown period according to the distribution of planktonic saline diatoms (González and Maidana, 1998).

Today, increased flow in the Río Desaguadero responds to precipitation and snowmelt in the Andes, mostly from westerly storms originating in the Pacific. For example, high snowfall during the 1983 El Niño caused major flooding of the Salinas del Bebedero (M.A. González, personal communication). Summer precipitation from Atlantic sources probably does not play an important role in the hydrologic balance of the playa today (González and Maidana, 1998), although it is possible that easterly moisture played an important role in the early Holocene freshwater marsh environments recorded by Pseudostaurosira brevistriata (Fig. 19).

16.2.19. Laguna Cari Laufquen: Río Negro Province, Argentina (41.130S, 69.420W; 800 m)

Radiocarbon-dated lake sediments and lithoid tufa on shorelines above the modern floor of Laguna Cari Laufquen describe a preliminary lake-level curve (Fig. 20) that extends and modifies earlier work on the pa-leolimnology of this basin (Galloway et al., 1988). The largest and most persistent high stage of the lake occurred between 18,400 and ca. 13,000 14C B.P., although the gravels representing the termination of this event

FIGURE 19 Selected diatom profiles from outcrops and elevation of dated beach ridges surrounding Salinas del Bebedero in Argentina. (Data from González, 1994; González and Maidana, 1998.)

are undated (Stine, unpublished). Apparently, the lake did not rise sufficiently at this time to persistently drain. Nevertheless, the 60-m high stand was probably within 10 m of the sill elevation. The calcareous clays

* = radiocarbon age control

FIGURE 20 Lake stage elevation curve from Laguna Cari Laufquen in Argentina. (Data from S. Stine.)

* = radiocarbon age control

FIGURE 20 Lake stage elevation curve from Laguna Cari Laufquen in Argentina. (Data from S. Stine.)

and silts deposited by the lake at this high stand contain rare diatoms and ostracodes and indicate that the late glacial lacustrine system was turbid and saline, but not especially productive. The clear, productive water required to produce stromatolites was apparently confined to protected bays distant from the main sediment sources. Deposits from the second high stand (+35 m), centered on 10,000 14C B.P., suggest limnological conditions similar to those during the late glacial period. The abrupt termination of the 35-m stage has good age control and appears to correlate with the lake stage chronology of the Atacama region. Although clear stratigraphic and geomorphic evidence exists for minor Holocene lake rises, at this time their chronology is unknown.

Lake levels at Laguna Cari Laufquen respond to snowmelt and precipitation on mountain drainages east of the Andes. Presumably, the full glacial (?), late glacial, and early Holocene high stands at Cari Lauf-quen document greater precipitation at this latitude;

this increased precipitation was also expressed by Andean full and late glacial glacier advances recorded near Bariloche (e.g., at Aguado; Markgraf and Bianchi, 1999) and in Lago Mascardi at 15,000-13,000 14C B.P. (Ariztegui et al., 1997). The source of the precipitation is unclear. A westerly source conflicts with pollen evidence at this latitude west of the Andes (Markgraf et al., 1992). Southeasterly moisture related to increased activity of mobile polar highs or to cold air outbreaks from Antarctica and southernmost South America remains a possibility (e.g., Leroux, 1993).

16.2.20. Lago Cardiel: Santa Cruz Province, Argentina (49^S, 71.250W; 276 m)

Lago Cardiel is a large (460 km2), turbid, deep (>70 m), endorheic lake in southern Patagonia occupying a volcano-tectonic depression in Neogene volcanic rocks and underlying Cretaceous sediments. Lake hydro-chemistry is dominated by sodium and sulfate, but the comparatively fresh water (~3 g/L) suggests lake recharge to the groundwater table.

Shoreline dates from lithoid tufa indicate lake depths up to 55 m above present water levels between 9800 and 7700 14C B.P. (Fig. 21). Intermediate to low lake levels existed at 20,000 14C B.P., but transgressive deposits with a minimum age of 31,400 14C B.P. and possibly related strandlines indicate a very high stage preceding the full glacial period. Mid- to late Holocene lake levels higher than modern levels (10-20 m) imply a complex and variable lacustrine history (Stine and Stine, 1990). At present, the lake continues to fall below recorded historical levels.

Was this article helpful?

0 0

Post a comment