Mudball: Surface dust and Snowball Earth deglaciation

Abstract

Recent modeling results have raised doubts about the ability to deglaciate from a global glaciation at atmospheric carbon dioxide levels that are realistic for a Neoproterozoic Snowball Earth. Here we argue that over the lifetime of a Snowball event, ice dynamics should lead to the development of a layer of continental and volcanic dust at the ice surface in the tropics that would significantly lower the tropical surface albedo and encourage deglaciation. This idea leads to the prediction that clay drapes found on top of Neoproterozoic glaciations should be thicker in tropical than extratropical regions. We test this idea by running the FOAM general circulation model (GCM) with an added tropical dust layer of different sizes and albedos and find that the tropical dust layer causes Snowball deglaciation at p CO 2 = 0.01–0.1 bar in a reasonable regime of these parameters. We find similar, though more nuanced, results from a limited number of test cases using National Center for Atmospheric Research's CAM GCM.