Lunar Polar Ice Everywhere, but Only on Surface

Permanently shadowed regions larger than 10 km² were selected (yellow polygons). This study used flat surfaces in PSRs (coral patches) that had adjacent non-PSRs (lime-green patches). Credit: Qiao et al., 2019, Figure 1.

An international team of Chinese, American, and Russian scientists may have conducted the most extensive study of ice within lunar Permanently Shadowed Regions (PSRs) yet performed. By comparing the reflections from flat surfaces in major PSRs to those of adjacent non-PSRs, they determined that the vast majority contain ice. However, the ice appears to be restricted to the uppermost surface.

The study selected PSRs larger than 10 km² in regions at ±80° latitude. The flat surfaces within both PSRs and adjacent non-PSRs were constrained to have slopes less than 5°. The selection included 34 south pole and 41 north pole pairs of regions.



Nearly a decades’ worth of data from the Lunar Orbiter Laser Altimeter (LOLA) was used for determining the reflectance of the target areas. LOLA is one of seven instruments on the NASA Lunar Reconnaissance Orbiter (LRO), which was launched in 2009 to collect data about the Moon. One of LOLA’s primary objectives is to produce a high resolution global topographic model used to aid future landing site selection and surface mobility studies. LOLA’s other primary objective is to study the polar illumination environment and collect data of the Moon’s PSRs.

LOLA is a pulse detection time-of-flight altimeter that measures the precise distance to the lunar surface at five spots simultaneously, each about 25 m apart. This means that LOLA fires five laser beams at the lunar surface at a rate of 28 times a second (28 Hz). If LRO were orbiting the Moon at its nominal ground speed of 1600 m/s (3579 mph), LOLA would capture a measurement every 57 m (LRO’s current elliptical orbit alters these figures somewhat).

Measurements from LOLA combine to show changes in terrain reflectance. The LOLA tracks increase in reflectance as they pass over the flat PSR (green polygon), and decrease reflectance over the adjacent non-PSR (white polygon). Credit: Qiao et al., 2019, Figure 2.

LOLA is a key instrument for searching for water ice in PSRs. LRO is in a polar orbit, meaning that the poles have the most observations of any area on the Moon. LOLA is also well suited for the role of looking for ice because backscatter from the near-infrared laser reflects off surface ice and the laser is fired nearly perpendicular to the surface (zero-phase angle). LOLA can also collect observations in the sunless PSRs, where most other instruments require sunlight.

Previous studies have used LOLA data to explore PSRs, but they only cover single areas or an entire regions at coarse resolution. These studies have found high amounts of laser light reflected within PSRs, but the causes were inconclusive. Potential sources include either surface ice or space weathering. Space weathering is a physical process on airless bodies caused by exposure to space, including interplanetary dust and micrometeorite bombardment, solar and cosmic ray irradiation, and solar wind implantation. The key is determining what is ice and what is weathered material.


Comparing LOLA albedo measurements from PSRs and adjacent non-PSRs. The average PSR was about 5% more reflective than the adjacent non-PSR. The most likely explanation for this is the presence of surface ice. Credit: Qiao et al., 2019, Figure 3.


Using data from LOLA, the team found that the vast majority (71 of 75) of the analyzed PSRs were up to 10% more reflective than their adjacent non-PSRs. On average, the PSRs were about 5% more reflective. Models developed using these observations show consistent water concentrations to the results from the LCROSS ejecta plume observations.

To determine if the differences were due to surface ice or space weathering, the scientists cross-compared the results with observations from many other instruments. The main factors the team identified as potential sources for the reflectance anomalies included solar illumination, surface temperature, surface composition, and boulder abundance. Observations from LRO Diviner, LROC NAC, LRO Mini-RF, and Kaguya MI could not solely explain the differences. The most likely explanation is that the selected PSRs contain ice.


Image of the surface of a permanently shadowed region (PSR). No boulders over the size of 20 m were identified within this PSR, which might interfere with LOLA measurements. The LROC NAC captured this long exposure image using sunlight reflected from surrounding regions to illuminate the shadowed areas (bold white area). Credit: Qiao et al., 2019, Figure 9.

Ice Depth

With ice identified in nearly all PSRs, the next question is how extensive is it? Since LOLA uses a near-infrared laser, it can only measure the uppermost micrometer of surface. Deeper observations require radar echo experiments. However, these experiments do not indicate the presence of thick surface ice deposits. Therefore, the observed ice is either a surface frost layer or a mixture of water ice particles within regolith.

The life cycle of water on the Moon may be highly dynamic. The PSR water ice may start from volcanically erupted water that migrates toward the poles. When it encounters PSRs, it becomes cold trapped and is then subjected to sublimation via impact gardening.


Closeup of Amundsen crater. The blue areas are permanently shadowed regions, with the area enclosed in bold white being a 132 km² flat area. The dashed ellipse represents a 30 x 15 km landing zone. Credit: Qiao et al., 2019, Figure 11.

Going Forward

Even though this study was extensive, there are more items to address. The surface materials in PSRs may be more porous than adjacent non-PSRs, causing higher water ice content in PSRs than surface observations indicate. The extreme temperatures in PSRs may impact how boulders are distributed or cause unknown interactions with the regolith. Additionally, the ice may include other volatiles besides water, altering the amount of laser light reflected.

Ultimately, the best approach for better data is to send a lander or rover directly into a PSR to collect in-situ measurements. The extreme environment of PSRs makes many remote observations difficult, yet even a very short duration surface mission can provide valuable insight.

This study provides excellent insight into how extensive ice is within PSRs, along with the depth of ice to expect. The approach of studying large, flat PSRs is pertinent for near term missions that require large landing ellipses. The two recommended sites include Amundsen crater in the south and Lovelace crater in the north.

Having accessible PSR and non-PSR zones, and unique opportunities for both scientific and resource missions, these craters would make great targets for future missions.

April 27, 2019 - As kindly mentioned by Kevin Cannon on Twitter, the radar data referenced by this paper rules out continuous thick sheets of ice at the surface, like those seen on Mercury. However, this does not rule out a subsurface mixture of diffuse ice infused regolith.