The Coming Lunar Excavation Reality Check: What NASA's IPEx Will Actually Face
- 3 days ago
- 8 min read
Why Apollo Penetrometer Data Still Matters for IPEx and Every Future Lunar Excavator?
NASA’s Moon Base is advancing as a phased, iterative program focused on the lunar South Pole. The current strategy prioritizes robotic-first infrastructure, long-duration surface operations, and the ability to survive extreme environmental conditions, long periods of darkness, temperature extremes ranging from below –200 °C to above +120 °C, abrasive dust, and rugged terrain. Within this architecture, excavation and site preparation have shifted from supporting activities to foundational requirements. Landing pads, roads, berms, radiation protection, and feedstock for in-situ resource utilization all depend on the reliable movement of large volumes of regolith.
NASA’s Infrastructure Pilot Excavator (IPEx) (Figure 1), also known as the ISRU Pilot Excavator, is the agency’s primary technology pathfinder for this capability. Designed as a lightweight system that uses counter-rotating bucket drums, IPEx is intended to excavate and transport significant quantities of regolith under the low-gravity, high-vacuum, and high-dust conditions of the lunar surface. Its development represents a necessary and well-conceived step toward sustained surface operations.
Yet the mechanical reality that IPEx, and every subsequent excavator, will actually face remains only partially understood. Most current performance models still rely primarily on bulk density, average friction angle, and simple cohesion values. These parameters are useful, but they are incomplete. Apollo Self-Recording Penetrometer data collected more than fifty years ago continue to show that near-surface lunar regolith often exhibits a fabric-controlled resistance that cannot be explained by density and overburden stress alone. This fabric creates a threshold energy requirement that density-only design does not capture.
This article examines that geotechnical gap. It re-examines the Apollo penetrometer evidence, explains why fabric state matters for early robotic excavation, and presents a practical engineering framework (OCR* and LRC) that can improve risk assessment using data already available.
I'd argue: if we size the first generation of lunar excavators without accounting for fabric, what will they actually face when they begin digging?
What IPEx Will Actually Face
Most performance models for lunar excavators still rest on a simplified foundation: bulk density, average friction angle, and a modest cohesion value. These parameters are useful for first-order estimates. They are not sufficient for the first generation of robotic diggers that must operate in undisturbed South Pole regolith.
The most complete set of near-surface penetration data we possess, the Apollo Self-Recording Penetrometer (SRP) measurements from Apollo 15 and 16, later compiled by Mitchell et al. (1974), paints a more demanding picture.
The Figure 2 below shows the most complete compilation of near-surface penetration resistance data remains the envelope published by Mitchell et al. (1974). When properly interpreted, it reveals a critical engineering reality:
When the SRP data are examined carefully (Figure 2), three engineering realities stand out:
Resistance is highly variable at the same depth.
Shaded envelopes from different landing sites and different probe geometries show large overlap. At any given depth between 20 cm and 60 cm, measured resistance can differ by a factor of three or more. Same depth does not mean same resistance.
The material is frequently non-homogeneous with depth.
Sudden increases and decreases in resistance appear in multiple profiles. These are not clean stratigraphic layers. They are expressions of changing mechanical state, fabric development, particle interlocking, and stress history.
A threshold effect is present.
Several tests reached the instrument’s maximum force (110 N on Apollo 15, 215 N on Apollo 16) at relatively shallow depths. On Apollo 15, astronauts encountered a stiff zone at approximately 30-35 cm that could not be scooped and required chipping. This is classic threshold behavior: a developed fabric that must first be broken before the material yields more readily.
These observations have a direct consequence for IPEx and every excavator that follows it.
In the upper tens of centimeters, exactly the zone where early robotic diggers will operate, undisturbed lunar regolith often behaves as a fabric-controlled material. Current overburden stress is extremely low (typically well below 1 kPa at 30 cm). Yet resistance can be high. The most coherent engineering explanation currently available is that long-term impact gardening, thermal cycling, and seismic shaking have created a mechanical fabric that stores “memory.” Breaking that fabric requires a minimum energy input. Once broken, the material often becomes easier to excavate.
Density-only design misses this threshold. A system sized on average bulk density may be under-powered when it first encounters mature fabric, or it may be over-massed (and therefore less efficient) if it is conservatively designed for the highest possible resistance. Neither outcome is acceptable for early Moon Base logistics. This is the geotechnical reality that IPEx will actually face: not a uniform granular soil whose strength grows smoothly with density, but a state-dependent material whose initial resistance is governed by fabric that density parameters alone cannot fully capture.
Practical Engineering Lens: Fabric Awareness, OCR*, and LRC
Recognizing that undisturbed lunar regolith often behaves as a fabric-controlled material is only the first step. The next requirement is a practical engineering language that allows us to interpret existing data, rank sites, and size early excavators with greater realism.
Two complementary tools address this need.
OCR* - A Relative State Index
The OCR* framework treats the mechanical state of lunar regolith as a relative, overconsolidation-like index. It normalizes observed penetration resistance (or equivalent strength indicators) against the extremely low current overburden stress at a given depth.
Because lunar overburden is so small, even moderate absolute resistance produces elevated OCR* values. High OCR* indicates a well-developed fabric that will require higher initial energy to break. Lower OCR* indicates a more disturbed or less structured state where excavation can begin more readily.
OCR* is not a laboratory overconsolidation ratio. It is a screening-level state index calibrated against Apollo ground-interaction observations. Its purpose is relative comparison: to distinguish locations and depth intervals where fabric effects are likely to be significant from those where they are less dominant. Numerical values are illustrative and depend on simplifying assumptions regarding the conversion of penetration resistance into an apparent preconsolidation stress. The method is therefore best used for relative ranking rather than absolute design values.
As with any screening index, OCR* values should be treated as relative indicators and recalibrated as higher-quality in-situ data become available.
LRC - Lunar Regolith Classification for Operations
The Lunar Regolith Classification (LRC) system organizes shallow regolith into indicative mechanical-state domains based on expected operational behavior (Figure 3):
This is a behavioral engineering abstraction, not a geological or stratigraphic profile. Depth ranges and OCR* values are indicative only, intended to support relative screening and early operational planning. Actual transitions are gradual, discontinuous, and highly site-variable. Site-specific in-situ data are required for design.
L1 – Disturbed/low-fabric surface: low initial excavation energy, high dust generation once disturbed.
L2 – Transitional: moderate threshold, variable response.
L3 – Mature fabric: clear threshold energy required; higher long-term stability once prepared.
L4 – High-resistance/strongly structured: significantly elevated excavation energy and tool wear.
These domains are not geological layers. They are engineering abstractions designed for early planning and equipment sizing. Assignment of a location or depth interval to an LRC domain is indicative and should be based on multiple lines of evidence (penetration resistance trends, orbital proxies such as optical maturity and thermal inertia, and historical performance at analogous sites). Domain boundaries will be refined as in-situ data from precursor and early Artemis missions become available.
Why This Matters for IPEx
For IPEx and subsequent excavators, the distinction is operational:
Tool forces, power budgets, and cycle times will be governed first by the energy required to break the existing fabric, not by average bulk density.
A system optimized only for “typical” density may stall or under-perform when it first encounters L3 or L4 material.
Conversely, a system oversized for the worst-case fabric will carry unnecessary mass and consume more resources than necessary in lower-fabric zones.
Combining relative OCR* rankings with LRC domain classification, engineers can begin to screen candidate sites and depth intervals for expected excavation behavior using data that already exist. This does not replace the need for in-situ measurements. It provides a structured way to reduce early risk while those measurements are still being planned and executed.
The alternative, continuing to size the first generation of lunar diggers primarily on density and average strength parameters, leaves the fabric threshold as an unquantified risk. That is the gap the OCR* and LRC frameworks are intended to close at the screening level.
What can Be Done Now?
Several practical steps are already possible using data that currently exist:
Re-interpretation of Apollo SRP profiles
The force-depth curves from Apollo 15 and 16 can be re-examined through a fabric-aware lens rather than density-only models. Relative OCR* trends can be estimated at the locations and depths where high resistance or stiff behavior was recorded.
Relative site screening using orbital proxies
Optical maturity (OMAT), thermal inertia, surface roughness, and crater age can be combined into relative OCR* rankings for Artemis candidate regions (including Mons Mouton Plateau and other South Pole sites). These rankings remain screening-level but allow early identification of zones more likely to exhibit mature fabric.
LRC-informed planning
Candidate sites and depth intervals can be provisionally assigned to LRC domains (L1–L4) using multiple lines of evidence. This supports early decisions on excavation energy budgets, tool design margins, and surface preparation priorities for IPEx-class systems.
Updated excavator performance models
Simple fabric-threshold terms can be added to existing density-based models. Even conservative implementation of a threshold energy requirement improves realism for initial digs in undisturbed regolith.
Test protocol design for IPEx and precursors
Flight and precursor test plans can deliberately include measurements of initial breakthrough resistance and the change in behavior after fabric is disrupted. This turns IPEx into both a construction tool and a geotechnical data source.
These steps do not require new landed missions. They use Apollo ground truth, orbital datasets, and engineering judgment to reduce early risk.
What Still Requires In-Situ Confirmation?
The frameworks remain screening tools. The following measurements are essential before the methods can support design-level decisions:
Direct force-depth and excavation energy profiles at South Pole candidate sites under controlled conditions.
Side-by-side comparison of undisturbed versus deliberately disturbed zones using the same excavator.
Correlation of orbital OCR* proxies with actual ground performance.
Assessment of how fabric evolves under repeated traffic and extreme temperature cycling.
Validation or recalibration of LRC domain boundaries with real robotic excavation data from the lunar surface.
Until these data exist, OCR* and LRC should be treated strictly as relative screening instruments that improve early planning, not as substitutes for site-specific geotechnical investigation.
Final Thoughts
IPEx is an important technology pathfinder. Its success, and the success of every excavator that follows it, will depend on how honestly the lunar construction community confronts the mechanical reality of regolith fabric.
Density alone is not sufficient. Apollo penetrometer data already show that near-surface resistance is highly variable, frequently non-homogeneous with depth, and often governed by a fabric that creates a threshold energy requirement. Ignoring this fabric leaves early robotic diggers exposed to avoidable performance risk.
A fabric-aware, state-based approach using tools such as OCR* and LRC offers a practical way to improve early risk assessment with data that already exist. These frameworks are screening-level instruments. They do not replace the need for in-situ measurements. They do, however, provide a structured language for interpreting Apollo observations, ranking candidate sites, and sizing the first generation of lunar excavators with greater realism.
The window to embed this thinking into tool design, site selection, and operational planning is open now. Waiting for perfect in-situ data before adjusting models is no longer a viable strategy for the systems that will dig the first trenches of the Moon Base.
Reference
Mitchell, J.K., Houston, W.N., Carrier, W.D. III & Costes, N.C. (1974). Apollo Soil Mechanics Experiment S-200 Final Report (NASA CR-134306). University of California, Berkeley.
Roberto Moraes
Lunar Construction Strategist | Author | SpaceGeotech Founder
Comments