How can we converge on models of the interior of Enceladus?

Knowledge of the interior structure of Enceladus is important for understanding its origin, evolution, ongoing behavior, and potential for habitability. In particular, the thickness of the ice shell tells us about the energy budget, thermal history, and the nature of its ocean-to-surface pathways. Despite this importance, however, the community has yet to converge on a robust internal structure model and ice shell thickness estimate for Enceladus.

Over the last ten years, researchers have produced an array of interior structure models for Enceladus, with mean ice shell thickness estimates ranging from as little as 14 km to as much as 60 km (Figure 1). These models incorporate a variety of different estimates for the various observational constraints and often adopt distinct sets of modeling assumptions, making it difficult to compare meaningfully between the models or to decide which to adopt in one’s analysis or for mission planning purposes.

Here, we attempt to clarify how interior model results (especially ice shell thickness) depend on each of the various input data estimates (i.e., shape, gravity, and librations) and model assumptions (i.e., related to equilibrium figure theory, isostatic compensation, ice shell dynamics, and other modeling choices). We do this using a framework that allows us to modify each input in isolation, permitting apples-to-apples comparisons and revealing the sensitivity of the outcomes to each input independently.

As an example, we compare the interior structure models of Hemingway & Mittal (2019) and Park et al. (2024)—two similar models but with significantly different preferred ice shell thicknesses (21 vs 30 km). Our analysis allows us to identify exactly which shifts in the input data account for this difference in shell thickness estimates (Figure 2). More generally, we show how sensitive interior models are to shifts in each of the different inputs, revealing which measurements and which modeling assumptions are the most consequential. We argue that our community will not be able to converge on interior models of Enceladus until we can converge on both our estimates of the observational constraints (including how uncertainties are reported) and our choices of modeling assumptions.

Such an effort could have benefits that extent far beyond our understanding of Enceladus. Enceladus is the icy moon for which we so far have the best observational constraints and therefore it makes for an important test case. Gaining a deeper understanding of the current state of diversity among interior models for Enceladus—and hopefully learning to move beyond some of this ambiguity—will thus be crucial as we prepare for new missions to Europa, Ganymede, and eventually the Uranian moons.