A critical reevaluation of palaeoclimate proxy records from loess in the Carpathian Basin

Abstract

In the Carpathian Basin, loess is the most important archive of Quaternary palaeoclimate evolution, but only in the past two decades systematic and high-resolution investigations were conducted. Those studies remarkably improved our knowledge of the regional past environmental change; palaeoclimate inferences based on the magnetic susceptibility and grain-size distribution, as the most commonly used palaeoenvironmental proxies for the Carpathian Basin loess, indicate colder and drier climatic conditions during glacials when compared to interglacials. With an increasing number of studies using novel proxies in loess research, such a traditional understanding of dry and cold glacials and humid and warm interglacials in the Carpathian Basin has been questioned. As an illustrative example, mollusc-based climate reconstructions suggest generally warm and very dry summer conditions with mean July temperatures up to 21 °C for the southern Carpathian Basin during the last glacial. Results based on stable carbon isotopes strongly oppose such high summer temperatures, but studies based on n-alkanes are in general agreement with the mollusc data when it comes to the vegetation reconstruction indicating mostly steppic conditions. However, n-alkanes studies contradict warm and dry glacial conditions as indicated by mollusc-based reconstructions, pointing instead to cold and relatively humid glacials. In addition, there is an ongoing debate whether or not millennial-scale climatic oscillations can be observed in the Carpathian Basin loess, as well as whether this area was an important Northern Hemisphere dust source or rather a sink of far distance dust transport. Consequently, the current state of the art of the palaeoclimate reconstructions from loess in the Carpathian Basin is rather inconsistent.In order to “make sense” of the existing palaeoclimate data from the Carpathian Basin loess, we have reevaluated and reinterpreted the available data. We discuss and propose a coherent interpretation of rock magnetic, grain-size, malacological, stable carbon and nitrogen isotope, n-alkane and bacterial membrane lipid data for the last glacial cycle loess archives from the Carpathian Basin. We show that glacial conditions in the Carpathian Basin led to a notable increasing North-South gradient in temperature and an even stronger expressed decreasing trend in humidity, and that most of the biomarker proxy data conducted in loess for the very dry southern part of the Carpathian Basin show a strong bias towards arid conditions. In particular, palaeotemperature reconstructions seem to be misleading. Glacial conditions were drier and colder than previously proposed (summer temperatures likely under 15 °C during glacials), but notably warmer than in other parts of Western, Central, and Eastern Europe. The vegetation consisted mostly of steppic environments during both, glacials and interglacials. We indicate that the Carpathian Basin has a potential to be a major dust source of the Northern Hemisphere during glacials, although it was at the same time exposed to the deposition of fine far distance travelled dust. Moreover, the main issues in the regional and continental correlation of loess are highlighted, as well as the sensitivity of the Carpathian Basin loess to the millennial-scale climate variability recorded in other Northern Hemisphere records. Finally, we suggest that the onset of loess formation in the investigated area occurred during the Middle Pleistocene Transition, and it was probably related to intensive silt production related to glacier dynamics in the Alpine ice cap.