Widespread impact-generated porosity in early planetary crusts

Understanding the origin and evolution of planetary crustal porosity is of particular interest because crustal porosity strongly affects thermal, magmatic, and hydrothermal processes early in planetary history and influences potential habitable niches. The Moon offers a window into the evolution of ancient planetary crusts because its early porosity has been primarily modified by subsequent impacts, especially on the lunar farside where volcanism is less extensive. NASA's GRAIL mission showed the lunar crust is less dense and more porous than previously thought: the upper ~4 km of the lunar highland crust averages ~12% porosity, decreasing to ~4% at ~20 km depth. The spatial distribution of porosity suggests impacts are the likely source; however, it remained unclear how substantial porosity is produced at depth and outside crater rims. Previous models explained near-surface porosity (upper kilometers) or porosity within complex craters, but not the deep lying porosity observed by GRAIL. Ejecta deposition can account for shallow porosity, yet ejecta blankets achieve maximum thicknesses of ~5 km near basin rims, insufficient to explain deep porosity. Here the authors posit that in situ tensile fragmentation during impacts generates appreciable porosity deep in the crust and outside crater rim crests. During hypervelocity impacts, a compressive shock is followed by a tensile rarefaction (relief) wave from free surfaces, producing strong tensile stresses and high strain rates that trigger dynamic fragmentation. The resulting fragmented volume should host porosity, and the study examines whether tensile fragmentation can explain GRAIL-observed porosity.

Prior impact simulations including dilatancy demonstrated that shear deformation can create pore space within craters and ejecta blankets, but cannot produce significant porosity at depth outside crater rims. Observations indicate a large fraction of impact-generated porosity occurs outside crater rim crests, implying an additional mechanism beyond ejecta deposition and shear-induced porosity is required to explain deep lunar crust porosity. Earlier work showed gravity strongly affects tensile fragmentation, suggesting a corresponding effect on porosity generation, with lower gravity favoring larger regions of tensile failure and porosity.

The study uses the multimaterial, multirheology iSALE shock physics code to simulate porosity generation during large impacts into lunar, martian, and terrestrial crusts. Simulations incorporate: (i) a dilatancy routine to model shear-induced pore-space generation accompanying shear deformation; (ii) a dynamic fragmentation algorithm accounting for rate-dependent flaw growth and tensile failure due to shock and release; and (iii) a tensile porosity routine that inserts porosity into material in tension to ensure thermodynamic and strength-model self-consistency. Simulations cover impactor diameters from 1 to 1000 km and target surface gravities appropriate for the Moon, Mars, and Earth. Representative cases include impact velocities of 15 km/s (e.g., 10 km and 100 km diameter impactors). Only the excavation stage is simulated, when tensile fragmentation occurs, before gravitational collapse, ejecta landing, and other modification-stage processes alter the initial porosity structure.

Tensile porosity algorithm: In each computational cell, stress is decomposed into isotropic (pressure) and deviatoric parts. Pressure is computed from an equation of state (EOS) as a function of density and specific internal energy; deviatoric stresses follow linear elasticity limited by a shear strength model/envelope that increases with pressure and decreases with strain. The strength model defines a minimum (most negative) pressure the material can support (P_min). The tensile failure model reduces the most tensile principal stress when tensile strength is exceeded. Because EOS and strength models are defined independently, EOS-predicted negative pressures in expanded states (ρ < ρ0) may exceed P_min in magnitude, creating inconsistency if pressure is simply capped at P_min. The implemented approach replaces the cap with porosity insertion: the solid material density is prevented from dropping below that yielding P = P_min, and porosity is inserted to satisfy mass conservation and reach the required bulk density. An iterative procedure guesses the solid density, adjusting porosity so that the EOS and porosity model return P = P_min, ensuring consistency between EOS and strength. This porosity quantifies the pore space created by tensile distension and fragmentation. Material parameters include a basalt EOS for the crust and the ε-α porous compaction framework within iSALE. Outputs include spatial porosity fields at specified times after impact (e.g., 150 s for 10 km impactors; 500 s for 100 km impactors).

• Including tensile porosity produces substantially more and deeper porosity than simulations with only shear-induced dilatancy. Even a 1 km diameter impactor generates significant porosity both near the surface and tens of kilometers deep; compared to shear-only cases, tensile porosity yields 1–100× more porosity at depth and near-surface. In shear-only runs, porosity >0.1% was confined to depths of ~10 km.
• For a 10 km impactor at 15 km/s into intact lunar crust, porosity of ~1% occurs within the upper ~5 km even at ~250 km radial distance. Porosity >0.01% extends to depths of ~10 km at 250 km distance.
• A 100 km diameter impactor (15 km/s) exemplifies basin-forming conditions: beneath the crater, high overburden pressure suppresses tensile porosity, but outside the crater, near-surface porosity is extensive. At 1000 km radial distance, porosity extends to ~50 km depth under lunar gravity. The modeled porosity structure (a few percent down to ~30–50 km) matches GRAIL estimates, implying that 100–1000 km scale lunar basins alone can account for the observed lunar crustal porosity.
• Gravity dependence with 100 km impactors: at 1000 km radial distance, impact-generated porosity extends to ~50 km depth on the Moon, ~25 km on Mars, and ~10 km on Earth. Under Mars gravity, ~1% porosity occurs down to ~3 km depth at ~800 km from impact. At 400 km distance, significant porosity exists to relatively large depths on all three bodies (on the Moon ~1% at ~18 km depth; on Mars and Earth several percent porosity occurs several kilometers deep). At 800 km, porosity is ~1% at ~3 km depth on the Moon and Mars, but on Earth it is <0.1% and limited to the upper ~2–3 km.
• Cumulative effects: Impacts into pre-porous crusts suggest deep porosity production is additive up to a threshold (to be determined). Over lunar bombardment history, multiple impacts could increase bulk crustal porosity.
• Habitability implications: On Hadean Earth, if radial extent scales roughly linearly with impactor size, the cumulative surface area of substantial impact-generated porosity in the upper few kilometers sums to ~1–3× Earth's surface area. Such porosity supports widespread hydrothermal circulation far from crater rims where melts are negligible, offering potential habitats and refugia; impact-induced porosity is a protracted source that could persist for hundreds of Myr.

The results address the longstanding question of how deep and widespread porosity in early planetary crusts—particularly the Moon’s—was generated. Simulations show that tensile fragmentation during the excavation stage of large impacts produces extensive porosity at depth and far outside crater rims, matching the magnitude and depth distribution inferred by GRAIL. This mechanism complements and surpasses shear-induced dilatancy in creating deep porosity, especially beyond crater rims where observations indicate much of the porosity resides. The strong dependence on surface gravity explains variations among the Moon, Mars, and Earth, with lower gravity allowing deeper porosity penetration at a given radial distance. The production of porous, fractured crust at regional scales enhances permeability and fluid circulation, implying far more extensive hydrothermal systems than those confined to crater interiors. On the early Earth and Mars, such environments could have supported subsurface habitability and protected niches for prebiotic chemistry and early life, with cumulative effects over many basin-forming impacts potentially restructuring large fractions of the crust.

Large impacts that form 100–1000 km scale basins can produce the full extent of porosity observed in the lunar crust through in situ tensile fragmentation during the excavation stage. Incorporating a tensile porosity routine in hydrocode simulations yields deep and laterally extensive porosity fields consistent with GRAIL data, and demonstrates that gravity controls porosity depth penetration on different planetary bodies. The widespread, impact-generated porosity likely facilitated pervasive hydrothermal circulation and could have created extensive habitable subsurface environments on early Earth and Mars. Future work should quantify additive porosity thresholds in pre-porous crusts, incorporate modification-stage processes and ejecta blanketing, explore time evolution and healing of porosity, and couple porosity fields with hydrothermal and geochemical transport models.

• Simulations cover only the excavation stage, excluding modification-stage processes such as crater collapse, melt migration, and ejecta emplacement that could alter porosity distribution near craters.
• The effect of ejecta fallback on porosity is neglected, though it would add an additional porous layer in ejecta blankets.
• Thresholds and saturation behavior for cumulative porosity production in pre-porous crusts remain to be determined.
• Results are sensitive to material models (EOS, strength, damage, and porosity compaction parameters) and assume basaltic crustal properties; heterogeneity and anisotropy are not explicitly modeled.
• Melt production within craters can reduce apparent porosity in modeled fields; post-impact thermal and mechanical evolution (healing/closure of pore space) is not simulated.
• Spatial resolution limitations may under-resolve far-field porosity at the largest radial distances.