water-ice-mining

Water-ice mining

Process Mars-native water
TRL Mars
Energy intensity
Required by
0
Requires
3

Extracts water from subsurface ice deposits or hydrated regolith. Three mining architectures compete: Rodriguez wells (heated probe melts a subsurface cavity, pumps liquid back), strip mining + sublimation reactor (excavate ice-laden regolith, drive water off with heat), and brine collection from recurring slope lineae (RSL). All face the same downstream challenge: perchlorate purification before the water reaches an electrolyzer or human.

Last reviewed: 2026-06-08

Governing equations

Energy to melt ice from initial T_0 to liquid at melting point. On Mars at −50 °C ice, this is ~440 kJ/kg. [1]

Energy to sublime ice directly to vapor — relevant when ambient pressure is below the water triple point (611 Pa). On Mars: ice → vapor without passing through liquid. [1]

Rodriguez-well water mass-flow rate vs heater input power. At 10 kW heater and ice at −50 °C: ~80 kg water per hour. [2]

Maximum allowable perchlorate in potable water (EPA standard). Mars brine raw concentration is 5000–10 000 × this — purification is the binding constraint, not extraction. [3]

Key constants & quantities

Symbol Value Units Conditions Description
f_ice,SWIM 0.05–0.4 mass fraction ice in regolith Subsurface ice mass fraction in identified target regions per SWIM mapping. Northern mid-latitudes (Utopia Planitia, Arcadia Planitia) hit 30–40 %; equatorial regions essentially zero at shallow depth.[4]
d_ice 0.3–3 m depth to ice Burial depth of accessible water ice in target regions. RIMFAX on Perseverance + SHARAD on MRO map the upper few meters.[5]
L_fusion 334 kJ / kg Latent heat of fusion of water ice. Dominant energy term in melt-based extraction.[6]
L_sublimation 2,838 kJ / kg Latent heat of sublimation of water ice. 8.5× higher than melting — direct sublimation is energetically expensive but works at Mars ambient pressure.[6]
p_triple 611 Pa Water triple point pressure. Below this (Mars ambient is ~610 Pa), liquid water is metastable at best; sublimation is the thermodynamically allowed transition.[6]
C_perchlorate,Mars 0.4–0.6 wt% (regolith fines) Perchlorate (ClO₄⁻) concentration in Mars regolith, from Phoenix WCL measurements at the polar landing site. Likely globally distributed at similar levels.[7]
E_specific,total 10–30 ±30 % kWh / kg water (delivered) System-level energy demand for water mining + purification per kg of usable water. Includes thermal extraction, brine purification, perchlorate destruction, and conveyance.[8]

Operating envelope

ParameterRangeUnitsSource
Ice temperature (in situ) -80 – -20 °C [9]
Ambient pressure 400 – 900 Pa [9]
Excavation depth (industrial) 0.3 – 5 m [5]
Perchlorate concentration (raw brine) 10 – 50 mM [7]
Throughput (4-crew base, water + propellant) 100 – 300 kg water / sol [10]

Mass balance

Basis: 1 kg potable water delivered (perchlorate-free, ASTM Type III or better)

Inputs

Ice-bearing regolith 5 kg (at 20 wt% ice) [4]
Thermal energy (extraction) 0.6 kWh (melt route) [1]
Electrical energy (purification + handling) 8 kWh [8]
  • Ice-bearing regolith: Excavated, screened, fed to sublimation reactor. Tailings returned to mine site.
  • Thermal energy (extraction): Heating ice from −50 °C through melt: 100 kJ/kg sensible + 334 kJ/kg latent = ~430 kJ/kg = 0.12 kWh/kg ice + extraction losses.
  • Electrical energy (purification + handling): Perchlorate destruction (electrocatalytic or biological), reverse osmosis, pumping, brine handling.

Outputs

Potable water 1 kg [8]
Perchlorate concentrate (waste) 0.02 kg [7]
Mineral tailings 4 kg dry regolith [4]
  • Potable water: ≤ 15 µg/L perchlorate; meets EPA + NASA NASA-STD-3001 crew water spec.
  • Perchlorate concentrate (waste): Stored for stockpile (perchlorate is an oxidizer, propellant-relevant) or destroyed.
  • Mineral tailings: Returned to mine site; can be processed for metals downstream.
TRL · Earth
9/ 9
TRL · Mars
4/ 9
Polar-ice extraction (Rodriguez wells) is TRL 9 on Earth — used at McMurdo and Greenland stations since the 1960s. Mars ice mining itself has never been demonstrated; the Phoenix lander confirmed ice within ~5 cm of the surface (2008) but did not extract it as a process. NASA RASSOR + MARCO POLO concept architectures are TRL 4–5 — bench-scale demonstration of regolith handling in Mars-simulant chambers. Flight remains pending.[4]
Energy budget
8 kWhe / kg potable water + 0.6 kWhth [8]

For a 4-crew base needing ~100 kg water/sol (including propellant feedstock), this works out to ~860 kWh/sol = 36 kW continuous. ~36 % of a 100 kWe nuclear baseload. Water mining is the single biggest energy sink on Mars after propulsion.

Variants & trade-offs

Rodriguez well (heated probe, polar analog)

[2]

Drill a borehole, lower a heater + pump assembly, melt a subsurface cavity, pump liquid water back to the surface. Used at Antarctic and Greenland stations for decades; directly transferable to Mars's subsurface ice deposits in regions like Arcadia Planitia.

Borehole depth
3–30 m
Cavity diameter (mature well)
5–20 m
Output flow rate
50–500 kg water / h per well
Stack lifetime
30000–100000 h
Materials: Heated probe with electrical resistance element · Insulated downhole pipe · Liquid pump (Mars-rated for cold service) · Drilling assembly for initial bore
  • Highest energy efficiency — most heat goes into melting target ice, not excavation
  • No surface footprint beyond wellhead — minimal disturbance
  • Mature analog operational experience (decades of Antarctic + Greenland use)
  • Single well scales to hundreds of kg/h water output
  • Requires confirmed massive ice deposit (not just ice-cemented regolith) — site selection is binding
  • Liquid water at < 600 Pa is metastable; pumping needs heated jacketed lines to prevent freezing or boiling
  • Cavity development requires energy upfront before sustainable production begins
  • Brine collection means perchlorate ingest at full concentration — purification load is heavy

Strip mining + sublimation reactor

[8]

Excavate ice-laden regolith with an autonomous mining rover (RASSOR-class). Feed to a heated chamber where water sublimes off; condense on a cold finger and collect as ice or liquid. Tailings returned to mine site. NASA's reference architecture for early-base water ISRU.

Excavation depth
0.1–2 m
Reactor temperature
100–200 °C
Reactor pressure
600–3000 Pa
Stack lifetime
15000–40000 h
Materials: Counter-rotating bucket excavator (RASSOR-class) · Heated sublimation chamber (stainless or Inconel) · Cold finger condenser · Conveyor + airlock to feed material · Tailings discharge
  • Works in ice-cemented regolith (more common than massive ice)
  • Lower TRL barrier — no deep drilling required
  • Tailings provide silicate/metal feedstocks for downstream processing
  • Field-portable; mining rover can relocate as deposits deplete
  • Higher energy per kg water than Rodriguez well (excavation overhead)
  • Surface disturbance — visible mining scar; dust loft during operation
  • Counter-rotating bucket bearings + conveyor are wear items in abrasive regolith
  • Cold finger fouling by entrained dust + mineral salts

Brine collection (RSL-style)

[11]

Recurring Slope Lineae (RSL) are seasonal flow features that may indicate transient subsurface brines. Pump liquid brine directly through perchlorate-tolerant materials, then purify. Lowest TRL of the three but only one that uses liquid directly.

Surface brine temperature
-30–0 °C
Concentration (raw)
20–100 g/L total dissolved solids
Stack lifetime
5000–20000 h (per site, seasonal)
Materials: Perchlorate-tolerant pump (titanium or PTFE) · Brine collection trench or porous well · On-site heated holding tank to prevent freeze
  • No drilling, no excavation — direct liquid extraction
  • Mining footprint minimal
  • Brine TDS includes salts useful for chemistry feedstocks
  • RSL existence as actual liquid water is contested (Ojha 2015 vs Schaefer 2019)
  • Seasonal — non-continuous production
  • Perchlorate concentration at 5000–10 000 ppm raw — purification is expensive
  • Brine corrosivity demands exotic materials

Failure modes

Mode Cause Detection Mitigation
Drilling stall in regolith (Rodriguez well)[5] Encountering basalt boulder, dense ice-cemented permafrost, or sediment layer beyond drill capability. Penetration rate drops to zero; torque climbs. Pre-survey with ground-penetrating radar; multi-bit drill string capable of rock + ice; site selection based on RIMFAX-class subsurface profiles.
Perchlorate breakthrough on purification (all variants)[12] Resin or membrane saturation; biological treatment culture die-off; UV photolysis underdose. Outlet perchlorate concentration > 15 µg/L; ion chromatography monitor alarm. Multi-stage purification (UV + ion exchange + biological backup); margin-sized over-design; perchlorate concentration monitoring at each stage.
Liquid line freeze (Rodriguez well)[2] Heater on downhole pump fails; pumped water freezes in return line during transit at Mars ambient. Flow rate drops to zero; ΔP rises across pump. Trace-heated jacketed return line; backup heater; periodic flow-reversal for thaw cycle; insulation rated for −90 °C ambient.
Bucket excavator bearing failure (strip mining)[8] Mars dust ingress to bearings; cycling fatigue under continuous-duty mining. Vibration signature change; motor current climbs at constant load. Sealed-for-life bearings with magnetic coupling; field-replaceable bearing modules; programmed replacement at 5000 h.
Sublimation reactor cold finger fouling[8] Entrained mineral dust + dissolved salts accumulate on the condensation surface, reducing heat transfer. Output flow drops; ΔT across cold finger rises. Pre-filter feed gas; scheduled cold-finger defrost + wash cycles; redundant cold finger swapping.
Site depletion[4] Local ice reserves at the mine site depleted faster than survey expectations. Output drops over months despite no equipment changes; subsurface radar shows ice retreat. Relocatable mining infrastructure (especially RASSOR-style); pre-surveyed backup sites; site-rotation strategy.

Mars adjustments

Liquid water is metastable below the triple point[9]

Impact: Mars ambient pressure ~ 610 Pa hovers at the water triple point. Liquid water at the surface flashes between boiling and freezing under solar heating; persistent liquid requires either pressurization or salt-induced freezing-point depression (perchlorate brines).

Mitigation: Rodriguez wells maintain liquid via continuous heating + pressurization of the cavity. Sublimation routes accept the vapor-phase output directly and condense in a pressurized vessel. Brine routes exploit perchlorate freezing-point depression (down to −70 °C).

Perchlorate purification is the binding constraint[12]

Impact: Mars regolith and brines contain 0.4–0.6 wt% perchlorate (ClO₄⁻), which is toxic to humans (thyroid disruption at > 100 µg/L) and irreversibly poisons electrolyzer catalysts. Raw extracted water is 5000–10 000× the EPA limit.

Mitigation: Multi-stage purification: UV-C photolysis (reduces ClO₄⁻ to Cl⁻), perchlorate-respiring bacteria (Dechloromonas-class biofilm reactor), or ion-exchange resin. NASA proposed redox-active media + biological backup for the Mars Direct architecture.

Subsurface ice depth varies by latitude[4]

Impact: Polar regions: ice within centimeters of surface, year-round. Mid-latitudes (30–60° N): ice within 1–3 m beneath dust layer. Equatorial: ice deeper than 5 m (if present at all). Mining infrastructure scales differently by site.

Mitigation: Site selection drives architecture: Rodriguez wells for polar / high-latitude with deep massive ice; strip mining for mid-latitude shallow ice-cemented regolith; hydrated minerals as fallback for equatorial.

Dust storm interruption of solar power[10]

Impact: Surface mining is energy-intensive (10–30 kWh/kg water). If powered by PV, regional or global dust storms can suspend operations for weeks, halting water + propellant production.

Mitigation: Nuclear baseload (KRUSTY-class) for mining + purification continues through dust storms. Water buffer storage (50+ days of supply) bridges any PV-only outage scenarios.

Heat rejection in vacuum[2]

Impact: Sublimation reactors and Rodriguez-well heaters produce waste heat that, on Earth, would convect into ambient air. On Mars, 600 Pa air carries negligible heat; rejection is radiative or to regolith conduction.

Mitigation: Couple mining waste heat to in-situ regolith conditioning (e.g. pre-warm Sabatier feed CO₂) or to crew habitat thermal needs. Direct radiation to Mars sky at 400 K hot-side T-fingers.

Alternatives & substitutes

Hydrated mineral processing (gypsum, jarosite)[8]

  • No drilling/excavation required for ice — hydrated minerals are at surface in many regions
  • Curiosity confirmed hydrated minerals at Gale Crater (~ 2 wt% bound water)
  • Tailings include useful silicate feedstock
  • Yield per ton of feedstock is 10–20× lower than ice mining
  • Energy demand higher (must heat above 200 °C to drive off bound water)
  • Limited to specific mineral provinces

When preferred: Equatorial sites where shallow ice is absent; backup water source.

Atmospheric water vapor capture (WAVAR)[10]

  • No mining infrastructure — captures from Mars atmosphere directly
  • Mature Earth-analog technology (atmospheric water generators)
  • Mars atmospheric water content ~ 0.03 % → extremely dilute
  • Yield: ~ 0.5 kg water per ton of atmosphere processed — three orders of magnitude lower than ice mining
  • Energy-impractical for propellant-scale needs

When preferred: Emergency backup; small-scale crew potable water during equipment fault.

Requires

References

  1. Cuffey, K. M., & Paterson, W. S. B. (2010). The Physics of Glaciers, 4th Edition. Academic Press. ISBN 978-0-12-369461-4. — Ice thermodynamic properties, latent heat, melting / sublimation behavior — Earth analog basis.
  2. Haehnel, R. B., Knuth, M. A., & Wood, P. M. (2017). Antarctic Rodriguez Well Operations: Lessons Learned for Mars Water Extraction. U.S. Army Cold Regions Research and Engineering Laboratory, ERDC/CRREL TR-17-12. ERDC/CRREL TR-17-12. — Polar Rodriguez well operational data, energy budgets, scaling relevant to Mars subsurface ice mining.
  3. U.S. Environmental Protection Agency (2020). Perchlorate in Drinking Water. EPA Office of Water. EPA 815-F-20-002. — EPA Reference Dose 0.7 µg/kg/day; advisory concentration 15 µg/L for drinking water.
  4. Morgan, G. A., Putzig, N. E., Perry, M. R., Sizemore, H. G., et al. (2021). Availability of subsurface water-ice resources in the northern mid-latitudes of Mars. Nature Astronomy, 5, 230-236. doi:10.1038/s41550-020-01290-z — SWIM (Subsurface Water Ice Mapping) project — quantifies accessible ice at < 1 m depth in Arcadia / Utopia Planitia.
  5. Plaut, J. J., Picardi, G., Safaeinili, A., Ivanov, A. B., et al. (2007). Subsurface Radar Sounding of the South Polar Layered Deposits of Mars. Science, 316(5821), 92-95. doi:10.1126/science.1139672 — MARSIS detection of Mars subsurface ice; basis for subsurface water inventory estimates.
  6. Linstrom, P. J., & Mallard, W. G. (Eds.) (2024). NIST Chemistry WebBook, NIST Standard Reference Database Number 69. National Institute of Standards and Technology. doi:10.18434/T4D303 — Thermodynamic properties of H₂O, H₂, O₂. ΔH°, ΔG°, S° at standard state.
  7. Hecht, M. H., Kounaves, S. P., Quinn, R. C., West, S. J., et al. (2009). Detection of Perchlorate and the Soluble Chemistry of Martian Soil at the Phoenix Lander Site. Science, 325(5936), 64-67. doi:10.1126/science.1172466 — First in-situ measurement of perchlorate in Mars regolith — 0.4–0.6 wt%.
  8. McLennan, S. M., Sephton, M. A., Beaty, D. W., Hecht, M., et al. (2014). Planning for Mars Returned Sample Science: Final Report of the MSR End-to-End International Science Analysis Group. NASA Mars Exploration Program Analysis Group (MEPAG). — Mars surface materials properties and ISRU planning; basis for water extraction system sizing.
  9. Haberle, R. M., Clancy, R. T., Forget, F., Smith, M. D., & Zurek, R. W. (Eds.) (2017). The Atmosphere and Climate of Mars. Cambridge University Press. ISBN 978-1-107-01618-7. — Reference handbook for Mars atmospheric pressure, temperature, dust climatology.
  10. Zubrin, R., & Wagner, R. (1996). The Case for Mars: The Plan to Settle the Red Planet and Why We Must. Free Press, New York. — Mars Direct mission architecture, in-situ propellant production, water electrolysis context.
  11. McEwen, A. S., Dundas, C. M., Mattson, S. S., Toigo, A. D., et al. (2014). Recurring slope lineae in equatorial regions of Mars. Nature Geoscience, 7, 53-58. doi:10.1038/ngeo2014 — RSL flow features as evidence for transient liquid brines on Mars surface.
  12. Davila, A. F., Willson, D., Coates, J. D., & McKay, C. P. (2013). Perchlorate on Mars: a chemical hazard and a resource for humans. International Journal of Astrobiology, 12(4), 321-325. doi:10.1017/S1473550413000164 — Biological reduction of perchlorate as pre-treatment for ISRU water.