Observation of isothermal metamorphism of new snow and interpretation as a sintering process

Isothermal snow metamorphism is based on the fundamental process of sintering. However, the relative weights of the physical processes responsible for sintering between ice grains are still debated. The most active are expected to be grain boundary diffusion or sublimation-condensation. We performed four isothermal experiments, starting with fresh snow, at temperatures of -1.6, -8.3, -19.1, and -54 degrees C during nearly 1 year. Monthly, we imaged the snow samples by X-ray microtomography. We monitored the changes in density, specific surface area, structure model index, and several structural parameters based on the distance transform of the ice matrix and pore space in the snow, as the trabecular number and thickness. At -54 degrees C, the metamorphism process was very slow, and during 1 year the specific surface area decreased by only 19%. For the other samples, the results can be interpreted as a two-stage process with a first phase of rapid change in trabecular number and structure model index and a second phase where the trabecular number and structure model index were constant. An increase in density and trabecular thickness together with a decrease of the specific surface area following a power law were observed throughout the experiments. The increase in ice thickness ( coarsening) related linearly to the densification. The continuous densification implies that volume processes like grain boundary diffusion have to occur. The linear relation between densification and coarsening suggests that the same mechanism governs both processes. The logarithmic decrease of the specific surface area is an indication that the coarsening is rate limiting.