Paleo±Dust: quantifying uncertainty in paleo-dust deposition across archive types

COSENTINO, NICOLÁS J.; TORRE, GABRIELA; LAMBERT, FABRICE; ALBANI, SAMUEL; DE VLEESCHOUWER, FRANÇOIS; BORY, ALOYS J.-M.

Earth System Science Data

Copernicus Publications

Año: 2024 vol. 16 p. 941 - 959

Mineral dust aerosol concentrations in the atmosphere varied greatly on glacial?interglacialtimescales. The greatest changes in global dust activity occurred in response to changes in orbital parameters (which affect dust emission intensity through glacial activity) and the lifetime of dust in the atmosphere (caused by changes in the global hydrological cycle). Long-term changes in the surface dust deposition rate are registered in geological archives such as loess, peats, lakes, marine sediments, and ice. Data provided by these archives are crucial for guiding simulations of dust and for better understanding the natural global dust cycle. However, the methods employed to derive paleo-dust deposition rates differ markedly between archives and are subject to different sources of uncertainty. Here, we present Paleo±Dust, an updated compilation of bulk and <10 μm paleo-dust deposition rates with quantitative 1-sigma uncertainties that are inter-comparable among archive types. Paleo±Dust incorporates a total of 285 pre-industrial Holocene (pi-HOL) and 209 LastGlacial Maximum (LGM) dust flux constraints from studies published until December 2022, including, for the first time, peat records. We also recalculate previously published dust fluxes to exclude data from the last deglaciation and thus obtain more representative constraints for the last pre-industrial interglacial and glacial end-member climate states. Based on Paleo Dust, the global LGM V pi-HOL ratio of <10 μm dust deposition rates is 3.1±0.7 (1 sigma). We expect Paleo±Dust to be of use for future paleoclimate dust studies and simulations using Earth system models of high to intermediate complexity. Paleo±Dust is publicly accessible at https://doi.org/10.1594/PANGAEA.962969 (Cosentino et al., 2024).