An albedo number does its work through the part it leaves behind. A salt crust at 93 percent reflectance passes 7 percent of incoming sunlight into the surface; fresh snow passes 17 percent, and melting bare sea ice passes 33 [15]. Swap snow for salt and the absorbed shortwave falls to roughly two fifths of what it was; swap melting ice for salt and it falls to about a fifth [15]. That is why ten points of albedo adds up to a real swing in an energy budget.
Getting the crust at all requires a surface losing ice to vapour rather than gaining snow. Brine trapped inside cooling sea ice concentrates until salts crystallise, and when the ice above sublimates those crystals stay behind as a thin bright layer [4]. Hence the tropical belt in the model, prescribed as a place where evaporation outpaces snowfall [9]. Bare, sublimating ice is the prerequisite; without it, there is no crust.
What kind of model it is matters for how far the result travels. This is a simplified setup whose purpose is to find the stable states a climate can settle into [8], so its output is an inventory of equilibria and which one the system falls into, not a rate. The case that salt hastens a freeze is an argument about ordering: the earliest salts precipitate some 28 degrees Celsius above the temperature at which seawater would be frozen almost through [16], so the brightening is available while ice is still spreading rather than only after the planet has arrived. How many years that saves is left open.
Whether the crusts were ever actually there is a separate question. The physical anchor is a laboratory crust on very cold sea ice [5], the dynamical anchor is an equilibrium model [8], and the paper appears in Climate of the Past as modelling rather than as a report of a preserved Neoproterozoic salt layer [3]. I would still put this mechanism in the class of things a deep-time sea-ice scheme ought to represent, conditional on the tropical bare-ice belt surviving contact with a model that computes its own circulation instead of assuming it [9]. Bright ice replacing dark ocean has been the standard account of how the glaciations of roughly 700 million years ago ran away [1][14]. If salt behaves the way this model says, that account is the conservative one.
