The presence of frozen volatiles (especially H2O ice) has been proposed in the permanently shadowed regions (PSRs) near the poles of the Moon, based on various remote measurements including the visible and near-infrared (VNIR) spectroscopy. Compared with the middle- and low-latitude areas, the VNIR spectral signals in the PSRs are noisy due to poor solar illumination. Coupled with the lunar regolith coverage and mixing effects, the available VNIR spectral characteristics for the identification of H2O ice in the PSRs are limited. Deep learning models, as emerging techniques in lunar exploration, are able to learn spectral features and patterns, and discover complex spectral patterns and nonlinear relationships from large datasets, enabling them applicable on lunar hyperspectral remote sensing data and H2O-ice identification task. Here we present H2O ice identification results by a deep learning-based model named one-dimensional convolutional autoencoder. During the model application, there are intrinsic differences between the remote sensing spectra obtained by the orbital spectrometers and the laboratory spectra acquired by state-of-the-art instruments. To address the challenges of limited training data and the difficulty of matching laboratory and remote sensing spectra, we introduce self-supervised learning method to achieve pixel-level identification and mapping of H2O ice in the lunar south polar region. Our model is applied to the level 2 reflectance data of Moon Mineralogy Mapper. The spectra of the identified H2O ice-bearing pixels were extracted to perform dual validation using spectral angle mapping and peak clustering methods, further confirming the identification of most pixels containing H2O ice. The spectral characteristics of H2O ice in the lunar south polar region related to the crystal structure, grain size, and mixing effect of H2O ice are also discussed. H2O ice in the lunar south polar region tends to exist in the form of smaller particles (similar to 70 mu m in size), while the weak/absent 2-mu m absorption indicate the existence of unusually large particles. Crystalline ice is the main phase responsible for the identified spectra of ice-bearing surface however the possibility of amorphous H2O ice beneath optically sensed depth cannot be ruled out.

The lunar polar regions contain permanently shadowed regions (PSRs) or local topographic depressions that never receive direct sunlight. These environments (< 110 K) have the potential to cold-trap volatile materials in the form of ice, which are essential resources for exploration and industrialization of cislunar space and the Solar System. Orbital observations and those from the LCROSS impactor experiment provide evidence of the existence of water ice and other cold-trapped volatiles in PSRs; however, constraints on volatile abundance and distribution remain ambiguous as individual observations are not always in concord. Here we compile observations indicating the presence of volatiles from ten remotely sensed datasets in 65 PSRs to estimate the locations and mass of water ice deposits. Faustini, Cabeus, de Gerlache, Shoemaker, Haworth, Sverdrup, Slater, and Amundsen are likely the most resource-rich PSRs. Based on co-locations of observations indicative of surface frost and subsurface hydrogen abundance, we find that the craters with the highest potential mass in metric tons (t) of water ice include Cabeus (-11 x 10(6) t), Shoemaker (-5 x 10(6) t), Faustini (-4 x 10(6) t), de Gerlache (-3 x 10(6) t), and Haworth (-3 x 10(6) t). Future prospecting of lunar volatiles and water ice is contingent on filling knowledge gaps in resource potential, notably accurate measurements of grade and depth of volatiles. Our proposed ranking and estimates for resource tonnage are a tool to guide future orbital and landed missions that could accurately determine the resource potential of PSR deposits.