Investigating the Indirect Effects of Dust via Ice-Containing Clouds in the Last Glacial Maximum and Pliocene Paleoclimates

Abstract:

A longstanding issue in paleoclimate research is that proxy data indicates greater polar amplification than that simulated by climate models. This is true for periods both colder and warmer than today, such as the Last Glacial Maximum (LGM) and the Pliocene. Resolving this disparity requires a better understanding of climate processes, particularly in the sensitive polar regions.

Climate records indicate that atmospheric dust load has varied greatly in the past with greater fluxes during the recent glacials compared to the interglacials. Dust fluxes in the Pliocene were reduced compared to the modern, partly due to the northward expansion of vegetation in Africa, which limited the extent of this major dust source.

Mineral dust is an important ice nuclei (IN) in clouds. The presence of dust allows ice-crystals in clouds to form over a wider range of environmental conditions than in its absence. Termed the indirect effect of dust, this process has a considerable impact on the physical and radiative properties of clouds and therefore climate. We investigate the indirect effects of dust on the LGM and Pliocene climates as they represent end member cold / warm climates with a high & low-dust load respectively.

We use CAM5 & CESM, state of the art climate models, and a new empirical parameterization for mixed-phase clouds (DeMott et al., 2015) in which dust is able to act as a cloud IN. Cloud ice-formation in our climate model is based on temperature and IN availability as opposed to the overly simple temperature-only parameterization often used. We modify global dustiness and CO₂concentrations to create idealized LGM and Pliocene simulations.

Our results show that the indirect effects of dust have a strong impact on the climate, particularly in the northern hemisphere. Modifying global dustiness impacts cloud physical and radiative properties, which translate into a large influence on global climate & polar amplification over the Arctic. Global temperatures and polar temperature vary by ~ 5°C and > 10°C respectively between our simulations.

The indirect effect of dust on the climate is an important forcing which has not previously been included in paleoclimate studies. Changes in dust are intrinsically linked to changes in climate, and this is a plausible mechanism by which model-data agreement can be improved for paleoclimates.