Lake Tauca

Lake Tauca is a former lake in the Altiplano of Bolivia. It is also known as Lake Pocoyu for its constituent lakes: Lake Poopó, Salar de Coipasa and Salar de Uyuni. The lake covered large parts of the southern Altiplano between the Eastern Cordillera and the Western Cordillera, covering an estimated 48,000 to 80,000 km2 of the basins of present-day Lake Poopó and the Salars of Uyuni, Coipasa and adjacent basins. Water levels varied, possibly reaching 3800 m in altitude. The lake was saline. The lake received water from Lake Titicaca, but whether this contributed most of Tauca's water or only a small amount is controversial; the quantity was sufficient to influence the local climate and depress the underlying terrain with its weight. Diatoms, plants and animals developed in the lake, sometimes forming reef knolls.

The duration of Lake Tauca's existence is uncertain. Research in 2011 indicated that the rise in lake levels began 18,500 BP, peaking 16,000 and 14,500 years ago. About 14,200 years ago, lake levels dropped before rising again until 11,500 years ago. Some researchers postulate that the last phase of Lake Tauca may have continued until 8,500 BP. The drying of the lake, which may have occurred because of the Bølling-Allerød climate oscillation, left the salt deposits of Salar de Uyuni.

Lake Tauca is one of several ancient lakes which formed in the Altiplano. Other known lakes are Lake Escara, Ouki, Salinas, Minchin, Inca Huasi and Sajsi, in addition to several water-level rises of Lake Titicaca. The identity of these lakes is controversial; Sajsi is often considered part of Lake Tauca, and the lake is frequently divided into an earlier (Ticaña) and a later (Coipasa) phase.

The formation of Lake Tauca depended on a reduction in air temperature over the Altiplano and an increase in precipitation, which may have been caused by shifts in the Intertropical Convergence Zone (ITCZ) and increased easterly winds. It was originally supposed that glacial melting might have filled Lake Tauca, but the quantity of water would not have been sufficient to fill the whole lake. The lake was accompanied by glacial advance, noticeable at Cerro Azanaques and Tunupa. Elsewhere in South America, water levels and glaciers also expanded during the Lake Tauca phase.

Overview
Lake Tauca existed on the Altiplano, a high plateau with an average altitude of 3800 to 4000 m, covering an area of 196000 km2 or 1000 x. The highland is in the Andes, the world's longest mountain chain which was formed during the Tertiary with a primary phase of uplift in the Miocene. Its central area, which contains the Altiplano, is formed by the eastern and western chains: the Eastern and Western Cordillera of Bolivia, which reach an altitude of 6500 m. The Eastern Cordillera creates a rain shadow over the Altiplano. The climate of the Altiplano is usually dry when westerly winds prevail; during the austral summer, heating induces easterly winds which transport humidity from the Amazon. A north-south gradient exists, with mean temperatures and precipitation decreasing from 15 C and 700 mm in the north, to 7 C and 100 mm in the southern Lípez area. Although precipitation decreases from north to south, the evaporation rate throughout the Altiplano exceeds 1500 mm/yr. Most precipitation is recorded between October and April. Occasionally during winter (but also in summer), frontal disturbances result in snowfall. Strong winds and high insolation are other aspects of the Altiplano climate. Much of the water balance in the present-day Altiplano-Atacama area is maintained by groundwater flow. The terrain of the Altiplano consists primarily of sediments deposited by lakes and rivers during the Miocene and Pleistocene. A Paleozoic basement underlies Cretaceous and Tertiary sediments. The Andean Central Volcanic Zone and the Altiplano–Puna volcanic complex are in the Cordillera Occidental.

Lake Tauca was one of many lakes which formed around the world during glacial epochs; others include the Baltic Ice Lake in Europe and Lake Bonneville in North America. Today, the Altiplano contains Lake Titicaca, with a surface area of 8800 km2, and several other lakes and salt pans. The latter include the Salar de Uyuni, at an altitude of 3653 m with an area of 10000 km2, and the Salar de Coipasa, covering 2500 km2 at an altitude of 3656 m. Lake Titicaca and the southern salt flats are two separate water basins, connected by the Rio Desaguadero when Titicaca is high enough. The theory that the Altiplano was formerly covered by lakes was first proposed by J. Minchin in 1882. The formation of such lakes usually, but not always, coincided with lower temperatures. No evidence has been found for lake expansions in the Altiplano region below an altitude of 3500 m.

Geography
Larger than Lake Titicaca, Tauca was over 600 km long and covered the area of the present-day Lake Poopo, Salar de Uyuni and Salar de Coipasa. Lake Tauca was the largest paleolake in the Altiplano in the last 120,000 years at least, and comparable to present-day Lake Michigan. Several different estimates for its surface area exist:

Water depths reached 110 - 120 m. Water levels were about 140 m higher than Salar de Uyuni, or 135 to 142 m. According to research published in 2000, the lake level varied from 3700 to 3760 m. Some disagreement about water levels at various sites may reflect differing isostatic rebound of the land covered by the lake. The original 1978 research on the Tauca phase postulated its shoreline at 3720 m. Of the previous lake cycles in the area, only the Ouki cycle appears to have exceeded that altitude.

A later phase in lake levels (known as the Ticaña phase) was lower, at 3657 m; the drop from Tauca was abrupt. The late phase of Lake Tauca, Coipasa, had a water level of 3660 m or 3700 m and covered an area of about 32000 km2. Transitions between lake cycles occurred in about one thousand years.

Lake Tauca was the largest lake on the Altiplano during the last 100,000-130,000 years. Although the preceding paleolake (Minchin) was probably shallower, there is disagreement about the methods used to ascertain water depth. Some consider Minchin the larger lake; a 1985 paper estimated its size at 63000 km2, compared with Tauca's 43000 km2. Confusion may have resulted from the incorrect attribution of Tauca's shorelines to Lake Minchin; a shoreline at 3760 m formerly attributed to Lake Minchin was dated to the Tauca phase at 13,790 BP. The theory that Tauca is the largest lake follows a deepening trend in the southern Altiplano paleolakes which contrasts with a decreasing trend in the level of Lake Titicaca during the Pleistocene. This pattern probably occurred because the threshold between the two basins progressively eroded, allowing water from Titicaca to flow into the southern Altiplano. The lakes left erosional benches, fan deltas (where the lakes interacted with ice) and lake-sediment deposits, and eroded into moraines. The ridge that separates the Salar de Uyuni and Salar de Coipasa was a peninsula in the lake; San Agustín, San Cristóbal and Colcha formed islands.

The lake and its predecessors (such as Lake Minchin) formed in the area currently occupied by salt flats such as the Salar de Uyuni, Salar de Coipasa, Lake Poopó, Salar de Empexa, Salar de Laguani, and Salar de Carcote—several tens of meters beneath the Tauca water level. The present-day cities of Oruro and Uyuni are located in areas flooded by Lake Tauca. Salar de Ascotán may or may not have been part of Lake Tauca. The submergence of a large part of the Altiplano under Lake Tauca reduced the production of dust there and its supply to Patagonia, but "restocked" the sediments and thus increased dust supply once Lake Tauca dried up. The terrain above 3800 m was affected by glaciation. In the Coipasa basin, a major debris avalanche from the Tata Sabaya volcano rolled over terraces left by Lake Tauca.

Hydrology
At a water level of 3720 m, the total volume of the lake has been estimated to be 1200 km3 to 3810 km3 at a level of 3760 m. Such volumes could have been reached in centuries. The quantity of water was sufficient to depress the underlying bedrock, which rebounded after the lake disappeared; this has resulted in altitude differences of 10 to 20 m. Based on oxygen-18 data in lake carbonates, water temperatures ranged from 2 to 10 C or 7.5 ±. Tauca may have been subject to geothermal heating.

The lake was deep and saline, with salinity increasing from the Tauca to the Coipasa stages. The salt content seems to have consisted of NaCl and Na2SO4. Estimated salt concentrations:

Estimated salt concentrations (based on a lake level of 3720 m, for sodium chloride, lithium and bromine):

Some of this salt penetrated aquifers beneath the lake, which still exist. A significant excess NaCl concentration has been inferred for Lake Tauca, possibly stemming from salt domes whose contents moved from lake to lake. Precipitation of calcium carbonate resulted in lake waters becoming progressively enriched in more soluble salts.

Glacial meltwater may have contributed substantially to Lake Tauca's development. Strontium isotope data indicates that water draining from Lake Titicaca through the Rio Desaguadero may have contributed between 70% and 83% of Lake Tauca's water, an increase of between 8 and 30 times the current outflow of Lake Titicaca via the Desaguadero. A drop in the level of Lake Titicaca about 11,500 BP may have resulted in its outflow drying up, favouring the disappearance of Lake Tauca. According to other research, the increased outflow of Lake Titicaca would have had to be unrealistically large to supply Lake Tauca with water if Titicaca was its principal source. Other estimates assume that one-third of Tauca's water came from Lake Titicaca, no more than 15% for any lake cycle, or the much-lower four per cent (similar to today's five-per cent contribution from Titicaca to Lake Poopó). During the Coipasa cycle, Lake Poopó may have contributed about 13% of the water. About 53% of Lake Tauca's water came from the Eastern Cordillera. About 60,000 years ago, the Desaguadero probably began transporting water from Lake Titicaca to the Uyuni area and the southern paleolakes. Tauca was fed by the Río Grande de Lipez on the south, the Río Lauca on the northwest and the glaciers of the two cordilleras on the east and west. The lake's total drainage basin has been estimated at about 200,000 km2. If lake levels reached an altitude of 3830 m, the lake may have drained into the Pilcomayo River and from there through the Río de la Plata into the Atlantic Ocean. Formerly an outlet may have formed at Salar de Ascotán, into the Pacific Ocean, before it was obstructed by lava flows. A theory proposed by Campbell in 1985 that a former Altiplano-wide lake catastrophically drained into the Rio Beni during the Holocene has not received much support.

Although earlier theories postulated that large lakes formed from glacial meltwater, increased precipitation or decreased evaporation (or both) are today considered necessary for lake formation; a complete glacial melting would have had to occur in less than about a century to produce the required volume. The water volume would be insufficient to explain Lake Tauca's high water levels; however, some smaller lakes in the southern Altiplano probably expanded from glacial meltwater alone. The lake may have contributed to increased precipitation by influencing land breezes. According to strontium isotope data, there may have been little water exchange between Tauca's Uyuni and Coipasa basins. During the Coipasa lake cycle, the Coipasa-Uyuni and Poopó basins had a limited connection. Minor water-level fluctuations occurred during the lake's existence.

Based on a 60000 km2 surface area, the evaporation rate has been estimated at over 70,000,000,000 m3/yr—comparable to the discharges of the Nile or Rhine. Less than half of this evaporation returned to the lake as precipitation; in the central sector of the lake at Tunupa, this would have increased precipitation by 80%, delaying the retreat of glaciers in the area. Groundwater from Lake Tauca may have drained into the Quebrada Puripica, northeast of Laguna Miscanti. Given the height of the sill between the two basins and evidence found at Poopó, water may have drained from the Coipasa-Uyuni basin into Lake Poopó during the Coipasa cycle.

Glacial debris and ice were probably present at the lake, with fan deltas at Tunupa overlapping the Lake Tauca shore. At Tunupa and Cerro Azanaques, glaciers reached their maximum size shortly before the lake level peaked and probably contributed to water levels when their retreat began. Two minor glacial advances, over 12,000 BP and about 11,000 BP, appear to coincide with Lake Tauca.

Lake Tauca left up to 5 m thick sediments in the southern Altiplano, and tufa deposits formed in the lake. The continental environment Pleistocene sediments were formed from lacustrine carbonate deposits. These rocks contain amphibole, clay minerals such as illite, kaolinite and smectite, feldspar, plagioclase, potassium feldspar, pyroxene and quartz. The composition of these rocks resembles that of the Altiplano soils. The sedimentation rate in the Uyuni basin was about 1 mm/year.

Biology
Low concentrations of pollen are found in sediments left by Lake Tauca in the Salar de Uyuni. Lake Minchin sediments contain more pollen (indicating that it may have had a more favourable climate), but the lack of pollen may be the product of a deeper lake. Polylepis may have thrived in favourable salinity and climatic conditions. Increased Polylepis and Acaena pollen is observed towards the end of the Tauca episode.

The lake was deep enough for the development of planktonic diatoms, including the dominant Cyclotella choctawatcheeana. Other diatoms noted in Lake Tauca are the benthic Denticula subtilis, the epiphytic Achnanthes brevipes, Cocconeis placentula and Rhopalodia gibberula, the planktonic Cyclotella striata and the tychoplanktonic Fragilaria atomus, Fragilaria construens and Fragilaria pinnata. Epithemia has also been found.

Sediments at the shoreline contain fossils of gastropods and ostracods; Littoridina and Succineidae snails have been used to date the lake. Other genera included Myriophyllum, Isoetes (indicating the formation of littoral communities) and Pediastrum. Algae grew in the lake and produced reef knolls (bioherms) formed by carbonate rocks. These grew in several phases, and some were initially considered stromatolites. Some dome-shaped bioherms reach a size of 4 m, forming reef-like structures on terraces. They developed around objects jutting from the surface, such as rocks. Tube- and tuft-shaped structures also appear on these domes. Not all such structures formed during the Tauca episode. Similar structures have been found in the Ries crater in Germany, where Cladophorites species were responsible for their construction. Taxa identified at Lake Tauca include Chara species. The water above the tufa deposits was probably less than 20 m deep. In some places (linked to Phormidium encrustatum and Rivularia species), limited stromatolitic development took place.

Research history
Reports of lake deposits on the Altiplano go back to 1861. A John B. Minchin in 1882 reported the existence of encrustations around Lake Poopo and the salars south of Coipasa. He postulated that a lake with a surface area of 120000 km2 left these encrustations and that the nitrate deposits in the Atacama and Tarapaca were likewise formed by water draining for this lake. Some estimates of the size of this lake claimed that it reached from Lake Titicaca as far as 27° South. The name "Lake Minchin" was applied in 1906 by Steinmann, who applied it to the Uyuni basin, while naming the lake covering the Poopo and Coipasa basins "Lake Reck". The name was applied in honour of John B. Minchin. Later it was found that Lake Titicaca was not part of Lake Minchin and the theory was put forward that meltwater from glaciers had formed the lake. A different lake (Lake Ballivian) was also defined which encompassed Lake Titicaca. The lake episodes "Escara" and "Tauca" were first defined in 1978. The relationship between various deposits in the southern Altiplano and these around Lake Titicaca was unclear at the beginning of the research history. Lakes were identified by the lake terraces, sediments, bioherms and drill cores.

Predecessor lakes
Before Lake Tauca, there were Ouki (120,000–98,000 years ago), Salinas (95,000–80,000 years ago), Inca Huasi (about 46,000 years ago), Sajsi (24,000–20,500 years ago) and Coipasa (13,000–11,000 years ago). Inca Huasi and Minchin are sometimes considered the same lake phase, and other researchers have suggested that Lake Minchin is a combination of several phases. The Ouki cycle may be subdivided in the future, and a number of sometimes-contradictory names and dates exist for these paleolakes.

Preceding lake: Escara
Escara was identified in the central Altiplano, it may be the oldest Altiplano lake cycle. Lake levels reached an altitude of 3780 m; perhaps reaching the size of Lake Tauca and Ouki. At the town of Escara, 8 m thick deposits have been left by the lake.

Escara is dated to 191,000 years BP. This date is of a tuff associated with lake deposits, the deposits themselves have not been dated. The L5 sediment and S10 layers in Salar de Uyuni have been linked to Escara. Some tuffs found in Escara lake deposits have been dated to about 1.87 million years ago. During the episode of Lake Escara, Lake Ballivian may have existed in the northern Altiplano as a southward extension of Lake Titicaca; Lake Escara would be thus identical to "lake pre-Minchin" which has left terraces 60 - 70 m above the present-day elevation.

Hypothetical pluvial and lake: Minchin
A humid period 46,000-36,000 years ago has been deemed "Lake Minchin"; it led to the formation of a large body of water on the Altiplano where Lake Tauca would later develop. The layer S4 in Salar de Uyuni drill cores has been linked to Lake Minchin. During this time, a salt lake existed at Laguna Pozuelos, while numerous lakes formed in northwestern Argentina after valleys were dammed by landslides, several lake basins in the Lipez region and many salt flats in the Altiplano filled with lakes, in which bioherms and stromatolites grew, moisture increased in the Brazilian and Bolivian Amazon and sediment accumulated in the Pativilca valley, the Pisco River valley (forming the "Minchin Terrace") and the Lomas de Lachay valleys. Regional glacial advance extending to the southern Altiplano/Puna has been correlated with the Minchin/Inca Huasi stage; the Choqueyapu II glacier advance in the Bolivian Andes, more debatably the Canalaya Phase in the Cordillera Apolobamba and the formation of the N-III moraines at Choquelimpie may coincide with the Minchin pluvial. Sedimentation rates in the main Altiplano lake were much less than during the Tauca pluvial.

The name "Lake Minchin" has been used inconsistently to refer to either the palaeolake at Lake Poopo, a lake existing 45,000 years ago, the highest lake in the Altiplano, or to sediment formations. An alternative theory postulates that Lake Minchin was formed by several lakes, including Ouki and Inca Huasi, and by unreliable radiocarbon dates. Sometimes the term "Minchin" is also applied to the whole hydrological system Titicaca-Rio Desaguadero-Lake Poopo-Salar de Coipasa-Salar de Uyuni, or to the highest ancient lake in the Altiplano (usually known as Lake Tauca). There are also contradictions between lake level records in different parts of the system. This confusion has led to calls to drop the usage of the name "Minchin".

Chronology
The existence of Lake Tauca was preceded by a dry period, with minor lake events recorded in Salar de Uyuni in the Late Pleistocene at 28,200–30,800 and 31,800–33,400 BP. This period was accompanied by the disappearance of ice from Nevado Sajama. A dry period is also noted in Africa and other parts of South America around 18,000 BP, and the retreat of the Amazon rainforest may have produced the lake low-water mark. The era may have been drier than the present. The drying of Lake Minchin left a salt layer about 20 m thick in the Salar de Uyuni, where gullies formed. Some ooid sediments formed before the Lake Tauca phase. Around 28,000 BP, lake levels rose in Lake Huinaymarca (Lake Titicaca's southern basin), preceding Lake Tauca by about two millennia. During this period, lakes in the Uyuni basin were intermittent. Previous lakes in the basin were generally small and shallow.

The radiometric age of Lake Tauca ranges from 72,600 to 7200 BP. The duration of the lake highstands may be overestimated due to radiation scatter. Radiocarbon dates have been obtained on crusts containing calcite, gastropod shells, stromatolites and structures left behind by algae. The Lake Tauca shorelines formed over more than century-long periods.

The first research, by Servant and Fontes in 1978, indicated a lake age between 12,500 and 11,000 BP according to C-14 dating. These were bracketed by dates between 12,360 ± 120 and 10,640 ± 280 BP for the highest deposits at Salar de Coipasa and Salar de Uyuni, and 10,020 ± 160 and 10,380 ± 180 BP for deposits which formed shortly before the lake dried. The reliability of the dates was questioned in 1990, and a later estimate was set at 13,000 to 10,000 BP. In 1990, Rondeau proposed ages of 14,100 to 11,000 BP based on radiocarbon dating and 7,000 to 14,800 BP based on uranium-thorium dating.

In 1993 it was suggested that Lake Tauca had an earlier phase, with water levels reaching 3740 m, and a later phase reaching 3720 m. Research published in 1995 indicated that the lake was shallow for over a millennium before rising to (and stabilizing at) its maximum level. Water levels between 13,900 and 11,500 BP reached 3720 m; 3740 m was reached between 12,475 and 11,540 BP, and 3760 to 3770 m between 12,200 and 11,500 BP.

Research in 1999 indicated an earlier start of the Tauca lake cycle, which was subdivided into three phases and several sub-phases. Around 15,438 ± 80 BP (the Tauca Ia phase), water levels in Salar de Uyuni were 4 m higher than the current salt crust. Lake levels then rose to 27 m above the salt flat, accompanied by freshwater input (Tauca Ib). Around 13,530 ± 50 BP (Tauca II), the lake reached an altitude of 3693 m, not exceeding 3700 m. At this time, strong gully erosion and alluvial fans probably formed in Bolivian valleys. Between 13,000 and 12,000 BP, the lake reached its greatest depth—110 m—of the Tauca III period. Dates of 15,070 BP and 15,330 BP were obtained for the highest shoreline, at 3760 m. After 12,000 BP, water levels decreased abruptly by 100 m. An even-earlier start was proposed by 2001 research, based on sediments in the Uyuni basin, which determined that Lake Tauca began developing 26,100 BP. A 2001 review indicated that most radiometric dates for Lake Tauca cluster between 16,000 and 12,000 BP, with lake levels peaking around 16,000 BP. A drop in oxygen-18 concentration in the Nevado Sajama glaciers has been associated with increased precipitation around 14,300 years ago. A 2005 book estimated the duration of the Lake Tauca phase at between 15,000 and 10,500 BP.

Research in 2006 postulated that the Lake Tauca transgression began 17,850 BP and peaked at altitudes of 3765 to 3790 m between 16,400 and 14,100 years ago. Spillovers into neighbouring basins may have stabilized the lake levels at that point, and the level subsequently dropped over a 300-year period. The following Coipasa phase ended around 11,040 +120/-440 BP, but its chronology is uncertain.

A 2011 lake history study set the beginning of the lake-level rise at 18,500 years ago. Levels rose slowly to 3670 m 17,500 years ago, before accelerating to 3760 m by 16,000 years ago. Contradictions between lake depths determined by shorelines and diatom-fossil analysis led to two lake-level-rise chronologies: one reaching 3700 m 17,000 years ago and the other reaching 3690 m between 17,500 and 15,000 years ago. The lake level would have peaked from 16,000 to 14,500 years ago at 3765 to 3775 m altitude. Shortly before 14,200 BP, the lake level would have begun its drop to 3660 m by 13,800 BP. The Coipasa phase began before 13,300 BP and reached its peak at 3700 m 12,500 years ago. The Coipasa lake's regression was nearly complete around 11,500 years ago. A 2013 reconstruction envisaged a lake level rise between 18,000 - 16,500 years ago, followed by a highstand between 16,500 - 15,500 and a decrease in lake levels between 14,500 - 13,500 years ago.

Lake Tauca is sometimes subdivided into three phases (Lake Tauca proper, Ticaña and Coipasa), with the Tauca phase lasting from 19,100 to 15,600 BP. The Coipasa phase, originally thought to have lasted from 11,400 and 10,400 BP, was corrected to 9,500 to 8,500 BP and later to 12,900 - 11,800 BP; it was preceded by a 400-year long lake level rise and was followed by a 1,600 years long decline. During this phase, lake levels rose to 3660 m altitude or 3700 km2 with a surface area of 28400 km2; the depth of the lake reached 55 m. According to a 1998 publication, Lake Tauca and the Coipasa phase lasted from 15,000 to 8,500 BP. The Coipasa phase has also been identified in Lake Chungará. The Coipasa phase was much less pronounced than the Tauca phase and shorter in duration, and was concentrated on the Coipasa basin, presumably because it receives more water than the Uyuni basin. An earlier lake phase, Sajsi (24,000–20,000 years ago), is sometimes considered part of Lake Tauca with the Tauca and Coipasa cycles. The Sajsi lake phase preceded the Tauca phase by one or two millennia and water levels were about 100 m lower than during the Tauca stage; it coincided with the Last Glacial Maximum.

The Ticaña phase was accompanied by a 100 m drop in water level. The Tauca and Coipasa phases are sometimes considered separate. Lakes Tauca and Minchin have been considered the same lake system and called Lake Pocoyu, after the present-day lakes in the area. "Minchin" is also used by some authors as a name for the system.

The Chita tuff was deposited in Lake Tauca at 3725 m altitude approximately 15,650 years BP, when the lake may have been regressing. Another tuff of uncertain age was deposited above Tauca-age sediments and tufas at the southeastern Salar de Coipasa. Data from Tunupa indicate that lake levels stabilized between 17,000 and 16,000 years ago. A 50 m lake-level drop occurred by 14,500 BP, with the lake drying between then and 13,800 years ago. Rising temperatures and a drop in precipitation were the likely triggers of lake and glacial retreat at the end of Heinrich event 1. In contrast, data from the Uyuni-Coipasa basin indicate that water levels peaked 13,000 years ago. The drying of Lake Tauca during the Ticaña lowstand has been linked to the Bølling–Allerød climate period and increased wildfires on the Altiplano; Lake Titicaca may have dropped beneath its outflow, cutting off the water supply to Lake Tauca. Glacial retreat at the beginning of the Holocene may also have been a contributing factor. As the lake receded, decreased evaporation (and cloud cover) would have enabled sunlight to increase the evaporation rate, further contributing to a decline in lake surface area.

A pattern of lake cycles becoming longer than the preceding one has been noted. Water from the lake may have contributed to increased oxygen-18 at Sajama around 14,300 years ago, possibly triggered by evaporation. As the lake level dropped, Lake Poopó would have been disconnected first; the sill separating it from the rest of Lake Tauca is relatively shallow. Coipasa and Uyuni would have remained connected until later. Water levels in Lake Titicaca's Lake Huinaimarca were low by 14,200 BP. By the Antarctic Cold Reversal, Lake Tauca was dry.

The end of the Tauca phase was followed by dry and cold conditions in the Puna, similar to the Younger Dryas, then by an early-Holocene humid period associated with decreased solar radiation. After 10,000 BP, another drought lasted from 8,500 BP to 3,600 BP, and peaked from 7,200–6,700 BP. The world's largest salt pan was left behind when Lake Tauca dried up, with approximately 10 m of material left at Salar de Uyuni. Lake basins in the Altiplano which had filled during the Tauca phase were separated by lower lake levels. Channels between the lakes testify to their former connections.

Climate
There are few reconstructions of how the climate looked before and after the Lake Tauca highstand. It has been estimated that summer precipitation would have increased by 315 ± and temperature dropped 3 C-change for Lake Tauca to form. According to a 1985 estimate, increased precipitation of 200 mm/yr would be needed; the estimate was subsequently revised to 300 mm/yr. With a 5 to 7 C-change temperature decrease, a 20–75% increase in precipitation would be required to form the lake. Research in 2013 indicated that the climate at the Tunupa volcano (in the centre of Lake Tauca) was about 6 to 7 C-change colder than present, with rainfall estimated at 320 to 600 mm. A 2018 estimate supported by 2020 research envisages a temperature decrease of 2.9 ± and a mean precipitation 130% higher than today, about 900 ±; this precipitation increase was concentrated on the eastern side of the catchment of Lake Tauca while the southernmost watershed was almost as dry as present-day. In a coupled glacier-lake model, temperatures were conditionally estimated at 5.7 ± lower than today. In the southern Altiplano, precipitation exceeded 500 mm during this epoch. In the central Altiplano, precipitation was 1.5 to three times higher than today. In and around the Arid Diagonal, precipitation doubled from 300 mm/year to 600 mm/year. Around the lakes precipitation may have increased nine-fold.

The formation of Lake Tauca coincides with Heinrich event 1 and has been explained with a southward shift of the Bolivian high that increased transport of easterly moisture into the Altiplano and a strengthening of the South American Summer Monsoon due to a decrease in the cross-equatorial transport of heat. Earlier highstands of Altiplano lakes may also correlate to earlier Heinrich events. Increased cloud cover probably increased the effective precipitation by reducing evaporation rates. In contrast, insolation rates do not appear to be linked to lake-level highstands in the Altiplano; the lake expansion occurred when summer insolation was low although recently an insolation maximum between 26,000 and 15,000 years ago has been correlated to the Tauca stage. The humidity above the lake has been estimated at 60%, taking into account the oxygen-18 content of carbonates deposited by the lake.

Coinciding with Lake Tauca, between 17,000 and 11,000 BP glaciers expanded in the Andes between 18° and 24° south latitude. At Lake Titicaca, glacial tongues approached the shore. The equilibrium line altitude of glaciers in the dry Andes decreased by 700 to 1000 m. Such glacial advances may have been preceded by the humid episodes which formed Lake Tauca. Around 13,300 BP, maximum glacier size in southern Bolivia is associated with a highstand of Lake Tauca. Glaciers did not expand everywhere, however, and there is little evidence for glacial expansion at El Tatio, Tocorpuri and parts of the Puna. Glacier expansions at Llano de Chajantor and surroundings may or may not have occurred. Frequent incursions of polar air may have contributed to glacial expansion. At Tunupa, a volcano located in the centre of Lake Tauca, maximum glacial extent lasted until the lake reached its highest level. Glacial shrinkage beginning 14,500 years ago probably occurred at the same time as a drop in lake levels, although dating ambiguity leaves room for debate. The Cerro Azanaques moraines reached their greatest extent from 16,600 to 13,700 BP. The existence of Lake Tauca coincides with the Late Glacial Maximum, when temperatures in the central Altiplano were about 6.5 C-change lower. Part of the glacial advance may have been nurtured by moisture from Lake Tauca, a conclusion supported by oxygen isotope data from the Sajama glaciers and by paleoclimate reconstructions around the former Lake Tauca. The Chacabaya glacial advance may be contemporaneous with Lake Tauca. Just like the Lake Tauca highstand may have coincided with the first Heinrich event, the Younger Dryas may be associated with the Coipasa highstand and the second Central Andean Pluvial Event although the Younger Dryas ended two millennia before the CAPE. The second CAPE was caused either by changes in the South American monsoon or by changes in the atmospheric circulation over the Pacific Ocean, and its end has been attributed to a warming North Atlantic drawing the ITCZ northward. Today, the average temperature at stations at an altitude of 3770 m is 9 C.

Context
The formation and disappearance of Lake Tauca was a major hydrological event that was accompanied by several millennia of wetter climate. Its formation and the later Coipasa lake phase is associated with the Central Andean Pluvial Event (CAPE), which occurred from 18,000–14,000 to 13,800–9,700 BP. During this epoch, major environmental changes occurred in the Atacama as precipitation increased between 18° and 25° degrees south. In some areas, oases formed in the desert and human settlement began. The Central Andean Pluvial Event has been subdivided into two phases, a first one which began either 17,500 or 15,900 years ago and ended 13,800 years ago and a second phase which began 12,700 years ago and ended either 9,700 or 8,500 years ago; they were separated by a short dry period that coincides with the Ticaña lowstand. The second phase of the Central Andean Pluvial Event has been subdivided further into a wetter earlier and a drier later subphase. During the Coipasa lake cycle, precipitation may have focused on the southern Altiplano and been transported there from the Chaco; the main Tauca cycle may have been accompanied by precipitation from the northeast. A glacial advance in the Turbio valley (a feeder of the Elqui River) between 17,000 and 12,000 years ago has been attributed to the Central Andean Pluvial Event. Other indicators point to dry conditions/lack of glacier advances in central Chile and the central Puna during the highstand of Lake Tauca, glaciers had already retreated from their maximum positions by the time it began and the Central Andean Pluvial Event may not have been synchronous between the southern Altiplano and the southern and northern Atacama.

Increased precipitation during the Tauca phase was probably triggered by the southern movement of the ITCZ and the strengthening of the South America monsoon, possibly caused by chilling in the northern hemisphere and North Atlantic, along with higher water temperatures off Northeastern Brazil. Combined with a southern shift of high pressure zones, increased moisture during late glacial times would have flowed from the Amazon. This change, which occurred from 17,400–12,400 years or 18,000–11,000 BP, is recorded in Bolivian Chaco and Brazilian cave records. Some 20th century phases of higher water levels in Lake Titicaca have been correlated with episodes of increased snow cover on Northern Hemisphere continents; this may constitute an analogy to conditions during the Lake Tauca phase. The Tauca phase may have been triggered by the southern shift of tropical atmospheric circulation and a weakening of the Atlantic meridional overturning circulation that decreased northward heat transport. An intensification and southward shift of the South Atlantic Convergence Zone may have contributed to the precipitation increase but not all records agree.

Another theory posits that vegetation changes and lake development would have decreased the albedo of the Altiplano, resulting in warming and moisture advection of moisture towards the Altiplano, but such positive feedback mechanisms were considered questionable in a 1998 study. Persistent La Niña climatic conditions may have contributed to the lake's filling and also to the onset of the first CAPE. Conversely, a global climatic warming and a northward shift of the monsoon occurred around 14,500 years ago, increased occurrence of El Niño and the northward shift of the ITCZ accompanied the Ticaña lowstand. The ideal conditions for the development of paleolakes in the Altiplano do not appear to exist during maximum glaciation or warm interglacial periods.

Related events
During the Tauca phase, Lake Titicaca grew in size; the pampas around Titicaca were left by that lake and the paleolake Minchin. Lake Titicaca rose by about 5 m, reaching a height of 3815 m, and its water became less saline. Another shoreline, at 3825 m altitude, has been linked to a highstand of Lake Titicaca during the Tauca epoch. The highstand, in 13,180 ± 130 BP, is contemporaneous with the Tauca III phase. Titicaca's water level then dropped during the Ticaña phase and probably rose again during the Coipasa. The highstands left terraces at the southern and eastern shores of Lake Titicaca, which were later deformed by tectonic processes.

Lake Titicaca probably overflowed on the south between 26,000 and 15,000 BP, adding water to Lake Tauca. Titicaca's outflow, the Rio Desaguadero, may have been eight times that of today. Lake Titicaca was thought to have had a low water level during the Tauca phase before evidence of deeper water was found. Higher lake levels have been found at the same time in other parts of the Altiplano and areas of the Atacama above 3500 m. This was not the first time Lake Titicaca rose; Pleistocene lake-level rises are known as Mataro, Cabana, Ballivian and Minchin. The overflow from Lake Titicaca into the southern Altiplano was possible for the last 50,000 years; this might explain why there is little evidence of large lakes in the southern Altiplano in the time before 50,000 years ago.

Lakes also formed (or expanded) in the Atacama at that time; highstands in Lejía Lake began rising after 11,480 ± 70 BP, and in Salar Aguas Calientes high-water levels lasted until 8,430 ± 75 BP. Highstands in Laguna Khota occurred around 12,500 and 11,000 BP. The formation of a lake at Salar de Llamara and some Salar de Atacama highstands are associated with Lake Tauca, the Minchin humid period and the Coipasa highstand. Traces of the Tauca humid episode have been found at Salar Pedernales, past 26° south latitude. Between 23,000 and 14,600 a lake formed at Laguna Pozuelos. Lake Tauca's highstand correlates with river terraces in Peru's Pisco River; terraces dated 24,000–16,000 BP in its tributary, the Quebrada Veladera; enlarged drainage systems in the Quebrada Veladera; a humid period at Lake Junin, and new soil formation in the pampas south of the Quinto River in Argentina and in the Ahorcado river valley in Peru. During the second Central Andean Pluvial Event, soils also formed in a wetland of northern Chile.

During the Tauca phase, water levels in Laguna Miscanti were higher than today; shorelines formed from an event in Ch'iyar Quta and Lake Tuyajto; saline lakes formed in the Lipez area, and water levels rose in the Guayatayoc-Salinas Grandes basin, in Laguna de Suches in Peru and lakes at Uturuncu and Lazufre. Some Atacama Altiplano lake levels increased by 30 to 50 m, Lake levels rose in Laguna Mar Chiquita, Laguna La Salada Grande in the Cordillera Oriental (Argentina) and Salina de Bebedero in Argentina.

Downward expansion of vegetation and increased discharge in the rivers draining to the Pacific Ocean has been correlated to the Tauca period. Evidence exists at the Quebrada Mani archeological site for a higher water supply 16,400–13,700 years ago. During the Tauca, greater flow occurred in rivers in the Atacama region as well as a higher groundwater recharge; more precipitation fell in the Rio Salado valley; flooding in the Río Paraguay-Parana basin and the contribution from Andean rivers such as the Rio Salado and Rio Bermejo increased; the excavation of the Lluta River Valley, Quebrada de Purmamarca and the Colca Canyon may have been aided by an increased water supply, river incision changed, river terraces formed in the Lomas de Lachay, erosion occurred along the Pilcomayo, and an increase in Pacific plankton was probably linked to increased runoff (and an increased nutrient supply) from the Andes. groundwater-fed wetlands developed in the Cordillera de la Costa, and valleys and large salt caves formed northwest of the Salar de Atacama.

Glaciers advanced in the Cordillera de Cochabamba. An ice cap formed over the Los Frailes ignimbrite plateau; its demise after the end of the Lake Tauca period may have allowed magma to ascend and form the Nuevo Mundo volcano. Moraine formed at Hualca Hualca and Nevado de Chañi where glaciers expanded; the Choqueyapu II glacier in the Eastern Cordillera advanced; moraines formed from glacial advances in Argentina (including the Sierra de Santa Victoria); basal sliding glaciers formed at Sajama; periglacial phenomena became more significant in northwestern Argentina from increased moisture supply; glaciers and probably also rock glaciers grew at Sillajhuay; snow cover in the Atacama Altiplano increased to about 10% above 4000 m elevation; glacier advanced in the northern Atacama. A glacial advance in central Chile around 15,000 years ago, also associated with increased precipitation and the Lake Tauca period, was probably triggered by tropical circulation changes.

Landslide activity decreased in northwestern Argentina but increased at Aricota, Locumba River, Peru; alluvial fans were active in the Cordillera Oriental of Peru; tufa deposition began in the Cuncaicha cave north of Coropuna; the climate grew wetter over the southern Amazon as evidenced in Brazilian cave deposits; precipitation and forest cover in Pampa del Tamarugal increased with an interruption ("Late Pleistocene Pampa del Tamarugal desiccation event") during the Ticaña lowstand; the vegetation limit in the Atacama desert descended towards the coast; groundwater discharge in the Atacama increased; wetlands developed at Salar de Punta Negra; the "Pica glass" formed in the Atacama as a consequence of increased vegetation and the occurrence of wildfires in this vegetation and plant pathogens such as rust fungi were more diverse than today. Prosopis tamarugo grew at higher altitude thanks to a better water supply; and vegetation coverage increased in the Atacama Altiplano. The well dated record of Lake Tauca has been used to correlate climatic events elsewhere in the region.

Environmental consequences
Paleoindian settlement in South America commenced during the Lake Tauca and Ticaña stages, facilitated by the more favourable environment during the CAPE; the Viscachani culture around Lake Titicaca was contemporaneous with Lake Tauca. The earliest human dispersal in the region around Lake Tauca occurred towards the end of the Ticaña phase, with the Coipasa phase coinciding with the definitive establishment of humans in the region and also their spread through northwestern Argentina, where conditions were favourable. In the Atacama area, Tauca-age paleolakes had provided the environment for first settlers; the end of the paleolake phase coinciding with Lake Tauca was accompanied by the end of the first phase of human settlement, which had occurred during the Central Andean Pluvial Event; now humans left the desert. In the Altiplano, the wet period that was contemporaneous to Lake Tauca allowed the settlement of the region and the Central Andean Pluvial Event did the same in the Pampa del Tamarugal and the southern Atacama valleys. The initial peopling of the Salar de Atacama region was during the Lake Tauca time, but a sharp population drop took place after its drying. Inca towers on the Altiplano have been built with rocks left by Lake Tauca.

Some fossil water reserves in the dry Andes formed during the Tauca phase, the groundwater in the northern Chilean Central Valley, around Peinado in the Puna and part of the groundwater under Pampa del Tamarugal for example date back to the Lake Tauca wet phase. Lake Tauca may have supplied water to the Rio de la Plata region, sustaining life there during dry periods.

The Lake Tauca and preceding cycles left evaporite deposits, with sediment layers left by the lake in the Salar de Uyuni reaching a thickness of 6 m. The salts are continually washed out and re-deposited by ephemeral rainfall, causing the salt surfaces of the Salars to become very flat and smooth. The high aerosol content of the air in the Uyuni region has been attributed to fine sediments left by Lake Tauca. Diatomaceous deposits containing clay or calc were left behind by the lake, and ulexite deposits were formed by sediments in its deltas.

The taxonomic similarity between fish species of the genus Orestias in Lauca National Park and Salar de Carcote has been attributed to these watersheds' being part of Lake Tauca; in general the evolution of these fish was heavily influenced by the various lake cycles including these that preceded the Tauca cycle. The drying of the ancient lakes would have fragmented amphibious habitats, generating separate populations. Lake Tauca and its predecessors may have created a productive environment that was populated by mammals like glyptodonts, Gomphotheriidae, Megatheriidae and Toxodontidae; the Atacama Altiplano had far more life than today during the Tauca cycle, including now-extinct deer and horses. On the other hand, the Altiplano lakes would have separated the animal and plant populations.