A new model of crater degradation on Ceres involving ice sublimation and talus formation

We report on the widespread occurence of talus on cerean crater walls using Dawn framing camera images. This is unexpected for a planetary body with no atmosphere like Ceres since the dominant process in crater degradation was expected to be topographic diffusion associated with impact gardening, like on the Moon or Mercury. To investigate why talus deposits are so common, and whether they could be related to the particular nature of Ceres’ ice-rich crust, we mapped them over the entire surface and studied their morphological characteristics. We classified the features into three different types, indicating their degree of preservation. Our results show that there is no trend between the total surface area of talus (whatever their degree of preservation) and latitude or longitude. However, we found that talus deposits tend to be observed within deep and young impact craters, and that certain craters expose a particularly large surface area of talus deposits: Dantu, Ninsar, Occator, Kupalo and Juling craters. It was also noted that well-preserved talus have a larger surface area than more degraded ones, and have $\approx$20 % more consolidated outcrops per km${}^2$. Age estimates from crater size frequency distributions for the host craters completed these observations, giving a mean age of $20_{-4}^{+4}$ Ma for the craters hosting the more preserved talus deposits, and $280_{-8}^{+8}$ Ma for the craters hosting the more degraded ones. By studying the slope of the talus deposits, we obtained a value of the angle of repose of the cerean regolith: 34.5°$\pm$ 2.8°. We also found that the talus facing the poles are approximately 5°steeper than those facing the equator, have an 18 % larger surface area, and exhibit three times as many outcrops compared to talus on crater walls facing other directions. This suggests that the degradation of outcrops, the source of the material making up the scree, is sensitive to insolation. If we assume that the outcrops are made cohesive by the ice they contain, then we can conclude that sublimation of this ice leads to their fragmentation, at the origin of the talus material. We suggest that outcrop wall retreat due to ice sublimation feeding talus slopes is a predominant process in the earlier phase of crater degradation on Ceres supplanting impact gardening and topographic diffusion for some tens of million years. Once the maximum extent of the talus is reached, the progressively reduced debris production leads to a decreasing talus activity. Impact gardening then takes over and talus deposits are slowly degraded during the next few hundreds million years before fully disappearing. This has an influence on age estimations of the surface of Ceres, as the diameters of the craters grow larger over time due to the ice sublimation, resulting in a slight overestimation of the age of the dated cerean areas. In addition, in the perspective of future missions to Ceres, one of whose objectives would be to precisely quantify the ice content of the crust, the outcrops facing the poles and not yet degraded would be among the most interesting targets to explore.