Massive addition of isotopically-depleted carbon to the ocean and atmosphere caused a carbon isotope excursion (CIE) and global greenhouse warming during the Paleocene-Eocene Thermal Maximum (PETM) circa 56 million years ago. The body of the CIE is followed by a recovery interval that is key to understanding Earth's capacity for carbon uptake, mechanisms of carbon uptake, and biotic responses following an extreme greenhouse warming event. Expanded terrestrial stratigraphic sections in the Bighorn Basin of Wyoming provide exceptionally high-resolution records of the CIE and can be linked directly to the mammalian fossil record.Here, we provide carbon isotope records of unprecedented resolution measured on in-situ pedogenic carbonate nodules in two parallel 8-km-spaced sections of upper Paleocene and lower Eocene fluvial sediments in the northern Bighorn Basin. We find consistent precession-driven sedimentary cycles in the two sections. Cycle thicknesses show significant lateral, and thus vertical, variation, demonstrating that astronomical age models constructed for fluvial successions require detailed sedimentary facies analysis of parallel sites.Plotting the high-resolution carbon isotope records in time using our astronomical age model for the correlated sections indicates a CIE body duration of 101 ± 9 kyr. The CIE shows an initial recovery step of +2.7 ± 1.0 ‰. This step occurs across the single, well-developed paleosol marker bed known as Purple-4, which represents a time interval of up to 15 kyr. The rapidity of recovery at the end of the CIE body is remarkable in light of existing hypotheses for carbon removal from the ocean-atmosphere system. Concurrent mammal finds show that the transition from faunal zone Wa-0 to faunal zone Wa-1 occurred in two steps, with a transitional Wa-R fauna preceding the CIE initial recovery step and Wa-1 fauna following the step.

The uncertainty in flood frequency relations can be decreased by adding reconstructed historic flood events to the data set of measured annual maximum discharges. This study shows that an artificial neural network trained with a 1‐D/2‐D coupled hydraulic model is capable of reconstructing river floods with multiple dike breaches and inundations of the hinterland with high accuracy. The benefit of an artificial neural network is that it reduces computational times. With this network, the maximum discharge of the 1809 flood event of the Rhine River and its 95% confidence interval was reconstructed. The study shows that the trained artificial neural network is capable of reproducing the behavior of the hydraulic model correctly. The maximum discharge during the flood event was predicted with high accuracy even though the underlying input data are, due to the fact that the event occurred more than 200 years ago, uncertain. The confidence interval of the prediction was reduced by 43% compared to earlier predictions that did not use hydraulic models.

No abstract

No abstract

Loess deposits are widespread in the Quaternary, but relatively rare in older geological records. This disparity is commonly linked to the unique climate conditions of the Quaternary, but those cannot fully explain the scarcity of loess in older records. Instead, we propose that the poor preservation of loess also plays an essential role. To test this hypothesis, we assess the preservation potential of loess by quantifying its modern‐day distribution in active sedimentary basins. This analysis shows that on the global scale only 20% of loess occurs in basins of which the majority is in a foreland setting. This could be due to nearby silt‐producing mountains and the effects of rain shadow aridity. The other 80% is ultimately either eroded or reworked and therefore poorly preserved in the long term. This conclusion implies that loess deposits may have been more common in pre‐Quaternary periods, despite being less abundant in the geological record.