Abstract This study comprises the source rock evaluation of 122 Late Middle Eocene–Early Miocene mudstones from the NE margin of the Black Sea. Samples are immature to early mature. The majority of samples have moderate to very good organic richness, poor to moderate source potential and a hydrogen-deficient to gas-prone source rock quality. However, a significant proportion of the samples have good to excellent organic richness and source potential, and an oil- and gas-prone quality derived from amorphous-rich kerogens. These samples would generate significant amounts of oil and associated gas where buried to peak maturity. They come from the lowermost (Rupelian) part of the Maykop Series and the late Bartonian–early Priabonian Kuma Suite or its stratigraphic equivalents. The Rupelian source-rock interval(s) in west Georgia is at least 60 m thick and potentially as much as 200 m thick. It has a source potential index (SPI) of 0.7–2.5 t HC m −2 . The thickness of the Kuma Suite-equivalent source rock interval south of the western Greater Caucasus is unconstrained. Maykop Series source rocks occur in the Black Sea Basin. Prospective Kuma Suite-equivalent samples on both the northern and southern margins of the Black Sea imply that similar sediments may also be present in the basin. Supplementary material: Additional information on the geographical location and age determination of the samples discussed in this paper are available at https://doi.org/10.6084/m9.figshare.c.3841399

Speleothems are valuable archives of climate change because of their extraordinary time resolution, which is unattainable in other terrestrial climate proxies. Analyses of 4796 ultraviolet fluorescent (UVf) layers observed in polished thin sections of a 15-cm-long speleothem collected from Raccoon Mountain Cave near Chattanooga, Tennessee, USA, as well as 200 δ 13 C and δ 18 O measurements and 11 high-precision U/Th dates permit refined interpretations of middle and late Holocene paleoclimate records in the southeastern United States. Speleothem UVf layers average 0.015 mm, identical to the average growth rate determined for the middle and late Holocene portions of the speleothem (ca. 7600–400 yr B.P.) based on the U/Th ages and interval thicknesses. UVf layer counts between paired U/Th ages are also consistent with determined ages and further support their interpretation as annual layers. The middle Holocene is typified by 100–400 yr intervals of higher rainfall characterized by thin UVf layers (0.003–0.010 mm) and more-negative δ 13 C values (−3‰ to −6‰ Peedee belemnite [PDB]), punctuated by shorter periods (5–20 yr, rarely 50–100 yr) of lower rainfall with thicker UVf layers (0.030–0.080 mm) and less-negative δ 13 C values (−1‰ to −3‰ PDB); “extreme drought” events are characterized by both the thickest UVf layers (0.150–0.170 mm) and the least-negative δ 13 C values (+0.05‰ to −1‰ PDB). The late Holocene, in comparison, is characterized by overall wetter conditions and more regular (sinusoidal curve) behavior, suggesting 50–100 yr cycles of higher and lower rainfall, with UVf layers ranging from 0.005 to 0.030 mm/yr. Statistical analyses of UVf layer thicknesses using order-two momentum threshold vector autoregressive models (MTVAR2) quantify the relationship between δ 13 C and δ 18 O, dependent upon the momentum in the climate. This study demonstrates that thickness of annual layers in speleothems can be used to resolve detailed paleorainfall records, provided there is preservation of organic matter sufficient to excite UVf response; however, relationships among changes in rainfall amounts, stable isotope values of speleothem calcite, and thicknesses of UVf annual layers (≈growth rates) are not straightforward.