Abstract
We extend an existing thermophysical activity model of comet 67P/Churyumov-Gerasimenko to include pressure build-up inside the pebbles making up the nucleus. We test various quantities of H$_{2}$O and CO$_{2}$, in order to simulate the material inside and outside of proposed water enriched bodies (WEBs). We find that WEBs can reproduce the peak water flux observed by Rosetta, but that the addition of a time-resolved heat-flow reduces the water fluxes away from perihelion as compared to the previously assumed equilibrium model. Our modelled WEBs eject dust continuously but with a rate that is much higher than the observed erosion and mass-loss, thus requiring an active area smaller than the total comet surface area or very large quantities of dust fallback. When simulating the CO$_{2}$-rich non-WEB material, we only find the ejection of large chunks under specific conditions (e.g. low diffusivities between the pebbles or intense insolation at southern summer), while we also find CO$_{2}$ outgassing rates that are much greater than observed. This is a general problem in models where CO$_{2}$ drives erosion, alongside difficulties in simultaneously ejecting chunks from deep while eroding the surface layer. We therefore conclude that ejection of chunks by CO$_{2}$ must be a localized phenomenon, occurring separately in space or time from surface erosion and water emission. Simulating the global production rates of gas, dust, and chunks from a comet thus remains challenging, while the activity mechanism is shown to be very sensitive to the material structure (i.e. porosity and diffusivity) at various scales.