The Black Sea is connected to a large drainage area including the European Russian Plain, part of the Alps and southeastern Europe. To study the hydrological changes in this basin over the last 40,000 years, we measured specific terrigenous biomarkers for wetland vegetation in well-dated sediments from the northwestern Black Sea, spanning the last glacial period (lacustrine phase) and the Holocene (marine phase). Low abundances of these biomarkers are observed during the North Atlantic ice melting and cooling events known as Heinrich Events 4 to 2, the Last Glacial Maximum and the Younger Dryas Event. Increased biomarker inputs characterize the mild climate phases known as Dansgaard–Oeschger Interstadials, the Bølling/Allerød and Preboreal Warmings indicating increased erosion due to permafrost degradation, higher primary productivity and/or wetland extension in the drainage basin. The final retreat of the Fennoscandian Ice Sheet from the Russian Plain occurs during the early part of Heinrich Event 1 and is characterized by increased biomarker concentrations in a typical series of four deglacial clay layers. For the last glacial period, the correspondence in timing between the biomarker records and the atmospheric CH4 record from ice cores, suggests an important CH4 contribution due to boreal permafrost thawing and wetland emission.

A 4.9-m-thick lake sequence, formed due to the landslide damming of a stream in the semiarid Garhwal Himalaya, was studied to understand past monsoonal variations in the region. The Optically Stimulated Luminescence (OSL) chronology indicates that the lake existed between ~ 12 and ~ 7 ka ago. Chronologically constrained trends of sand percent, organic phosphorus (OP), apatite inorganic phosphorus (AIP) and parameters of environmental magnetism were measured in the paleolake profile. Measured proxies indicate that the Indian summer monsoon ameliorated in the early Holocene after 12 ka cooling, and it appears that all the proxies from the lake have captured this globally recognized early Holocene warming. Four phases of wet conditions (intensified monsoon) are recognized at ~ 11.5 ka, ~ 11–10.5 ka, ~ 10–9 ka and ~ 8–7 ka with maximum uncertainties of ~ 1000 years. The wet phases are characterized by high magnetic susceptibility, increased OP and reduced AIP. In an attempt to understand the primary forcing of the sharp fluctuations in monsoonal activity in the region, we show that changes in magnetic susceptibility match variations of residual atmospheric δ14C, suggesting a role for solar variability as an explanation of climatic variability.

First-order modeling suggests that a low-water phase in late-glacial Lake Agassiz can be explained through changes in the balance between evaporation, precipitation, and runoff, rather than drainage. The low-water Moorhead Phase is often attributed to drainage through outlets opened by isostatic depression and retreat of the Laurentide ice margin. However, new data indicate that the proposed outlets were ice-covered during the Moorhead Phase. Instead, the lake water levels dropped to the Moorhead Phase before the start of the Younger Dryas chronozone and remained there until 11.3 ka. Thus, drainage seems to be an implausible explanation for Younger Dryas-aged low water levels in Lake Agassiz. An alternative explanation is that evaporation equaled or exceeded water inputs from the adjacent ice margin and the deglaciated parts of the drainage basin. To evaluate whether this hypothesis is plausible, we constructed a simple model that considers the paleo-basin geometry, hydrology, and meltwater production from the adjacent ice margin. Modest hydrologic changes (within the range of present-day variability), coupled with low meltwater production, produce a closed basin. Shifts in the location of the polar jet, driven by increased Arctic albedo, may explain our inferred hydrologic changes.

Land snail shell δ13C value is often used as a paleovegetation proxy assuming that snails ingest all plants in relation to their abundance, and that plants are the only source of carbon. However, carbonate ingestion and variable metabolic rates complicate these relationships. We evaluate if live-collected snails from Lanzarote (Canary Islands) reflect the abundance of C3 and CAM plants. Snails were collected on either CAM or C3 plants for isotope analysis of shell and body, and shell size. Respective shell and body δ13C values of snails collected on CAM plants averaged − 8.5 ± 1.7‰ and − 22.8 ± 1.6‰, whereas specimens from C3 plants averaged − 10.1 ± 0.7‰ and − 24.9 ± 1.1‰. A flux balance model suggests snails experienced comparable metabolic rates. A two-source mass balance equation implies that snails consumed ~ 10% of CAM, which agrees with their abundance in the landscape. Snails collected on CAM plant were smaller than those on C3 plants. Conclusively: 1) snails consume CAM plants when they are available; 2) migration of snails among C3 and CAM plants is a common phenomenon; and 3) C3 plants may be a more energetic food for growth than CAM plants. This study shows that shell δ13C values offer approximate estimates of plants in C3–CAM mixed environments.

Long-term fire histories provide insight into the effects of climate, ecology and humans on fire activity; they can be generated using accumulation rates of charcoal and soot black carbon in lacustrine sediments. This study uses both charcoal and black carbon, and other paleoclimate indicators from Lake Kinneret (Sea of Galilee), Israel, to reconstruct late Holocene variations in biomass burning and aridity. We compare the fire history data with a regional biomass-burning reconstruction from 18 different charcoal records and with pollen, climate, and population data to decipher the relative impacts of regional climate, vegetation changes, and human activity on fire. We show a long-term decline in fire activity over the past 3070 years, from high biomass burning ~ 3070–1750 cal yr BP to significantly lower levels after ~ 1750 cal yr BP. Human modification of the landscape (e.g., forest clearing, agriculture, settlement expansion and early industry) in periods of low to moderate precipitation appears to have been the greatest cause of high biomass burning during the late Holocene in southern Levant, while wetter climate apparently reduced fire activity during periods of both low and high human activity.