Search

The temporal and spatial patterns of late Holocene climate conditions provide valuable information for testing hypothesized mechanisms of recent climate changes. As a proxy for late Holocene climate in the southern tropics, we present a 10Be chronology of moraines deposited by Qori Kalis, an outlet glacier of Quelccaya Ice Cap in Peru. The Qori Kalis moraines are located downflow from the Quelccaya ice cores and provide the first glacial extent record that can be compared directly to annually resolved tropical ice core records. Qori Kalis advanced to its late Holocene maximum extent prior to 520 ± 60 yr before CE 2009, when Quelccaya ice core net accumulation values were at or below their late Holocene average. Subsequent glacial retreat between ~520 and 330 yr before CE 2009 coincides with the highest net accumulation values of the ~1800-yr-long ice core record. Therefore, we suggest that temperature, rather than net accumulation, was the primary driver of these glacial fluctuations. Comparison of the late Holocene fluctuations of Qori Kalis glacier with glaciers in the southern tropical Andes, Patagonian Andes, Switzerland, Alaska and New Zealand suggests globally synchronous, centennial-scale cold events. [ABSTRACT FROM AUTHOR]

We present a hydroclimate synthesis of the southern Great Basin over the last two glacial-interglacial cycles focused on paleolakes in Death Valley (core DV93-1), Searles Valley (core SLAPP-SRLS17), Owens Valley (core OL92), and the Devils Hole cave. There is close agreement between the occurrence of lakes in Death Valley and the height of the water table in the Devils Hole (50 km east of Death Valley) during the last 200 kyr. Death Valley and Devils Hole have adjacent, partly overlapping, drainage areas and most likely did over the last 200 kyr. When the water table in the Devils Hole was above the threshold level of ∼5 m higher than the modern, permanent lakes existed in Death Valley. At water table elevations less than 5 m above the modern, ephemeral lakes, saline pans, and mudflats occurred in Death Valley. The close temporal agreement between inferred paleoenvironments from the sediments in the Death Valley core and the paleowater table elevation in Devils Hole suggests a common forcing and provides insight into climate variability in the southwestern United States over the last 200 kyr. Owens Valley and Searles Valley, which derived inflow waters from the Sierra Nevada via the Owens River, contain paleohydrologic records which match those from Death Valley and the Devils Hole in terms of timing and direction of water availability over the last 200 kyr, indicating a similar paleohydrologic history for the entire southern Great Basin region. Near the end of Marine Oxygen Isotope Stage 6 (MIS 6), 140 ka - 130 ka, Lake Manly in Death Valley became shallow and hypersaline, and ultimately dried up at 127.1 ka ±4.3 ka. The transition from glacial to interglacial vegetation, which involved the loss of Juniperus pollen and an increase in Quercus (oak) pollen, occurred in Death Valley core DV93-1 at 131.3 ka ±4.0 ka. Following the glacial to interglacial pollen shift, a large alkaline lake formed in Death Valley. Similar conditions (freshwater, high productivity, and a mixed, deeply oxygenated water column indicated by biomarkers) existed in Searles Lake between 135.3 +2.7/ -2.9 ka and 130.1+2.7/ -2.6 ka, also following the juniper-oak pollen transition. Sr isotopes in calcite and sulfate minerals (gypsum, glauberite, thenardite), and the rare occurrence of the sodium carbonate mineral northupite with a low 87Sr/86Sr ratio in core DV93-1, together with organic geochemical proxies from Searles core SLAPP-SRLS17, all suggest that at this time, late MIS 6 Lake Manly in Death Valley received alkaline water via spillover from Searles Valley into Death Valley through Panamint Valley. The hydrologic connection between Searles Valley, Panamint Valley, and Death Valley at Termination II (130 ka) is documented here for this system of pluvial lakes for the first time. The Devils Hole water table decreased to +6.5 m at 140.8 ka ±3.2 ka, rose briefly to +8 m at 137.6 ka ±0.5 ka, and then dropped 8 m by 120.36 ka ±0.45 ka, when it reached an elevation similar to the modern. The pluvial lakes in Death Valley and Searles Valley may have coincided with the rise of the Devils Hole water table at ∼137.6 ka ±0.5 ka years ago, although the age models for core DV93-1 and core SLAPP-SLRS17 during the end of MIS 6 carry large uncertainties. • 200,000 year paleohydrologic synthesis of southern Great Basin: Death Valley, Searles Valley, Owens Valley, Devils Hole cave. • New data from Death Valley: U-Th dates (halite) and 87Sr/86Sr ratios (carbonates and sulfates). • Close agreement between Death Valley paleoenvironments and water table elevations in Devils Hole. • Searles Lake overflowed into Death Valley via Panamint Valley near the end of MIS 6. [ABSTRACT FROM AUTHOR]

Abstract: Varve thickness records from glacial Lake Ojibway provide a detailed picture of ice margin dynamics and paleohydrology for the Laurentide Ice Sheet during the early Holocene. The demise of Lake Ojibway occurred when an ice dam of the shrinking Laurentide Ice Sheet failed, perhaps releasing enough freshwater to impede Atlantic meridional overturning circulation and precipitate the 8200 cal BP climatic cooling event. Previously published records and new information from lacustrine sediment cores are combined to reconstruct a varve record over 2100 years long. Between varves numbered 1 to 1500, average rates of ice margin retreat increased from 140 m/y to as high as 400 m/y. Beginning prior to varve year 1800, and lasting to around year 2000, ice readvanced along a front at least 300 km wide during an episode called the Cochrane. The total distance of the advance was at least 75 km, and averaged around 200 m/y. During the advance, a series of ∼50 anomalously thick varves were deposited, the cause of which is uncertain. A widespread unconformity occurred during the end of the Cochrane advance, which was followed at around varve year 2060 by the return of varves, which thicken abruptly. Unlike previous varves, these varves include ice-rafted pellets of Lake Ojibway sediment. These varves may have resulted from rapid melting and calving, rather than a second advance, and may relate to basin re-filling following a lowstand. There are no direct dates on the Lake Ojibway varves, but a sudden increase in varve thickness at year 1528 is a regional stratigraphic marker that may have resulted from glacial sediment and meltwater (including Lake Agassiz overflow), diverted from routes to the Lake Superior basin, into the Lake Ojibway basin. Dates from the Lake Superior basin suggest that the diversion of this water occurred at around 9040 cal BP. Applying this date to varve 1528 constrains the age of the Lake Ojibway varves numbered 1 to 2100 to between 10,570 and 8470 (±200) cal BP. [Copyright &y& Elsevier]

Improving the late Quaternary paleoclimate record through climate interpretations of low-latitude glacier length changes advances our understanding of past climate change events and the mechanisms for past, present, and future climate change. Paleotemperature reconstructions at low-latitude glaciers are uniquely fruitful because they can provide both site-specific information and enhanced understanding of regional-scale variations due to the structure of the tropical atmosphere. We produce Little Ice Age (LIA) and Younger Dryas (YD) paleoclimate reconstructions for the Huancané outlet glacier of the Quelccaya Ice Cap (QIC) and low-latitude southern hemisphere regional sea surface temperatures (SSTs) using a coupled ice-flow and energy balance model. We also model the effects of long-term changes in the summit temperature and precipitiation rate and the effects of interannual climate variability on the Huancané glacier length. We find temperature to be the dominant climate driver of glacier length change. Also, we find that interannual climate variability cannot adequately explain glacier advances inferred from the geomorphic record, necessitating that these features were formed during past colder climates. To constrain our LIA reconstruction, we incorporate the QIC ice core record, finding a LIA air temperature cooling at the ice cap of between ∼0.7 °C and ∼1.1 °C and ∼0.4 °C and regional SSTs cooling of ∼0.6 °C. For the YD paleoclimate reconstructions, we propose two limits on the precipitation rate, since the ice core record does not extend into the Pleistocene: 1) the precipitation rate scales with the Clausius-Clapeyron relationship (upper limit on cooling) and 2) the precipitation rate increases by 40% (lower limit on cooling), which is an increase about twice as great as the regional increases realized in GCM simulations for the period. The first limit requires ∼1.6 °C cooling in ice cap air temperatures and ∼0.9 °C cooling in SSTs, and the second limit requires ∼1.0 °C cooling in ice cap air temperatures and ∼0.5 °C cooling in SSTs. Our temperature reconstructions are in good agreement with the magnitude and trend of GCM simulations that incorporate the forcing mechanisms hypothesized to have caused these climate change events. [ABSTRACT FROM AUTHOR]

Both modern anthropogenic emissions of mercury (Hg), primarily from artisanal and small-scale gold mining (ASGM), and preindustrial anthropogenic emissions from mining are thought to have a large impact on present-day atmospheric Hg deposition. We study the spatial distribution of Hg and its depositional history over the past ∼400 years in sediment cores from lakes located regionally proximal (∼90-150 km) to the largest ASGM in Peru and distal (>400 km) to major preindustrial mining centers. Total Hg concentrations in surface sediments from fourteen takes are typical of remote regions (10-115 ng g-1). Hg fluxes in cores from four lakes demonstrate preindustrial Hg deposition in southeastern Peru was spatially variable and at least an order of magnitude lower than previously reported fluxes in lakes located closer to mining centers. Average modern (A.D. 2000-2011) Hg fluxes in these cores are 3.4-6.9 μg m-2 a-1, compared to average preindustrial (A.D. 1800-1850) fluxes of 0.8-2.5 μg m-2 a-1. Modern Hg fluxes determined from the four lakes are on average 3.3 (±1.5) times greater than their preindustrial fluxes, similar to those determined in other remote lakes around the world. This agreement suggests that Hg emissions from ASGM are likely not significantly deposited in nearby down-wind regions. [ABSTRACT FROM AUTHOR]