3.1. Cryptomaria classification
Cryptomaria are maria regions on the Moon that are covered by highly reflective materials originating from the ejecta of craters or basins. Based on the source of the ejecta, these cryptomaria can be classified into four types: Copernicus-type, Balmer-type, Proximal basin ejecta-type, and Distal basin ejecta-type1,7,10,20. In this study, the gravity data results are utilized to reveal the origin of mare basalts within the cryptomaria and to classify the cryptomaria accordingly. We have divided the cryptomaria into two main types (Fig. 3): Impact-type and Diffuse-type.
Regions Balmer, Cleomedes, Dewar, Humboldtianum, Langemak, Mendel-Rydberg, Milne, and Smythii (Supplementary Fig. 1) are typical examples of the Impact-type cryptomaria. These regions have experienced significant impact events, ranging from smaller crater formations with diameters of 104.14 km to larger multi-ring basin formations with diameters of 886.77 km. The impact events in these regions have resulted in the thinning of the primitive lunar crust and the uplift of the lunar mantle (Fig. 3a). This geological process has led to the intrusion of mafic magma into the crust21. The presence of mascons, as observed in the gravity anomaly map, indicates the existence of materials within the lunar crust that possess a higher density than the average crustal material. The combination of crater and basin features in these regions suggests that the high-density material is mafic magma. These mafic magma can reach the lunar surface through the dikes, filling the bottom of the craters or basins22. Subsequent high-reflectivity ejecta from external sources have covered the low-reflectivity material, resulting in the characteristic appearance of cryptomaria as we observe them today. Within the regions of Cleomedes, Humboldtianum, Mendel-Rydberg, and Smythii, the mare basalts are not completely covered by ejecta, and there are still exposed continuous maria on the lunar surface. Furthermore, some DHCs can strip away the high-reflectivity material from the surface of the cryptomaria, exposing the underlying basalt to the lunar surface.
Regions Australe, Hercules, Lomonosov-Fleming, Marginis, Mare Frigoris, Taruntius, Van de Graaff, W. Humorum, and W. Procellarum (Supplementary Fig. 2) represent typical examples of the Diffuse-type cryptomaria. These regions show no mascons and are situated either at the edge of the maria or within the maria themselves. As illustrated in Fig. 3b, large craters situated within the maria have penetrated the layer of mare basalt covering the lunar surface. The ejecta surrounding these craters has covered the basalt layer, giving rise to the formation of cryptomaria. Alternatively, the perimeters of the maria are coated with ejecta material from craters, resulting in the formation of cryptomaria along the edges of the maria. The formation of craters also leads to a reduction in local topography, creating favorable conditions for the infilling of the crater floor by diffuse mare basalt. Irregular positive gravity anomalies are observed in these regions. Taking into account the geological background of these areas and existing research findings on gravity anomalies21–24, it is suggested that these gravity anomalies are likely formed due to the presence of mare basalt filling low-lying terrains or magma intrusion into the lunar crust through dikes.
Cryptomaria of both Impact-type and Diffuse-type are found in Regions Schiller-Schickard and South Pole-Aitken. These regions are characterized by the presence of several large impact craters, and cryptomaria can be observed both within and outside of these craters7. In this work, the cryptomaria regions on the Moon have been classified into two types: Impact-type and Diffuse-type, based on the source of the mare basalt. This classification provides valuable insights into the origin of basalt in 19 cryptomaria regions. Moreover, the gravity anomaly features within the lunar crust presented by the gravity model can effectively distinguish between these two types of cryptomaria. These gravity anomalies, combined with the spectral and morphological characteristics of the lunar surface, further support and validate the classification of the cryptomaria regions into these two types.
This new classification model enhances our comprehension of cryptomaria formation and paves the way for future investigations. According to the classification results, it is evident that not all cryptomaria are the outcome of ancient volcanic activity. For instance, the formation time of Diffuse-type cryptomaria is closely with the formation time of the surrounding maria (Supplementary Note 3). Impact-type cryptomaria are more likely to represent the oldest volcanic products on the Moon. This consideration is crucial when studying basalt production during various periods of lunar volcanism.
3.2. Cryptomaria identification
A significant concentration of DHCs involved in the excavation of basaltic materials has been observed in 19 out of the 29 proposed cryptomaria regions worldwide (Table 1), establishing them as prominent cryptomaria regions7. Conversely, the remaining 10 regions, lacking evidence of DHCs, were not classified as cryptomaria. In this work, we investigate the feasibility of determining whether these regions can be classified as cryptomaria through an analysis of gravity anomalies, topographical features, and chemical content of the lunar surface.
In contrast to the presence of DHCs observed in the majority of proposed cryptomaria regions, the five regions depicted in Supplementary Fig. 3 do not exhibit any evidence of DHCs7. However, these regions do feature large-sized craters, thin lunar crusts, and notable mascons in their gravity anomalies, indicating an environment that is more favorable for mare basalt eruptions2. Furthermore, the presence of a mascon at the center of these craters suggests an uplift of the lunar mantle in these regions. These observations are consistent with the characteristics of Impact-type cryptomaria. Based on the results from the morphology and FeO content maps25, typical volcanic formations such as lunar rimas and grabens are present in certain regions (Supplementary Note 5 and Supplementary Fig. 5). Additionally, there is a noticeable increase in FeO content within and around several fresh small craters. These observations provide further evidence supporting the identification of these areas as Impact-type cryptomaria. In contrast, Region Mendeleev is not considered a cryptomare. Although the impact event in Mendeleev caused uplift of the lunar mantle and intrusion of magma into the crust, the magma did not reach the lunar surface, resulting in the absence of surface volcanic features (Supplementary Note 5).
The five Regions Casatus, De Forest, Korolev, Maurolycus, and Zucchius in Supplementary Fig. 4 have all been subjected to impact events, but no mascons are observed in the gravity anomaly data. This indicates that there was no significant uplift of the lunar mantle at the center of these basins and that the intrusion of mafic magma into the lunar crust is more difficult. Additionally, no discernible DHCs are observed on the surface of these regions, and there are no volcanic structures or features indicating an increase in FeO content in the morphology and FeO content maps. These five regions do not exhibit the typical characteristics of Impact-type cryptomaria. Moreover, they are located in isolated areas within the lunar highlands and are not adjacent to any maria, indicating that they do not qualify as Diffuse-type cryptomaria either. These regions are composed of light plains that are not associated with cryptomaria. The formation of these light plains is primarily attributed to the covering effect of ejecta from craters.
Based on the analysis of gravity anomalies and surface features, the initial list of 29 proposed cryptomaria regions has been revised to 23 regions (Table 1). Six regions, Mendeleev, Casatus, De Forest, Korolev, Maurolycus, and Zucchius, do not exhibit characteristics consistent with cryptomaria. While Region Mendeleev cannot be definitively classified as a cryptomaria, it may represent a mafic magma intrusion into the lunar crust. Among the remaining 23 regions, 12 are identified as Impact-type cryptomaria, and 9 are classified as Diffuse-type cryptomaria. Notably, both types of cryptomaria coexist in Regions Schiller-Schickard and South Pole-Aitken.
3.3. Cryptomaria thickness
In this section, our focus is not on estimating the global basalt thickness of the cryptomaria. Instead, we delve into the feasibility of determining the cryptomaria thickness using gravity data and compare the various thickness estimation methods currently available. Gravity anomaly data can provide insights into density anomalies in the lunar crust, which can be attributed to changes in porosity, magma intrusions, or variations in the bulk crustal composition23. However, such data is not readily applicable for directly estimating the thickness of the cryptomaria. Cryptomaria are maria covered by high-albedo ejecta, previous work26 have estimated the thickness of maria using gravity anomaly data. However, these estimations could not accurately account for the thickness of basalt mare at the edges of the maria. Therefore, estimating the thickness of Diffuse-type cryptomaria using gravity anomalies alone may not be feasible. Furthermore, previous studies have revealed a significant disparity between the thickness of cryptomaria estimated using gravity data and that estimated using DHCs3,18 (Supplementary Note 6). The result derived from DHCs are considered more reliable when analyzed in the context of impact simulations and the geological characteristics of the region26–29.
The thickness of the cryptomaria can be estimated from the depth of DHCs3,11,30–32, but this method has limitations. The number of DHCs available for analysis is typically small, and they can only provide localized information about the depth of excavation within the cryptomaria. As a result, the depths obtained from DHCs not represent the average thickness of the cryptomaria. In addition to DHC analysis, flood simulations have been employed to estimate the thickness of cryptomaria7,33. This modeling approach provides an upper limit on the thickness of the cryptomaria and the thickness of mare basalts formed during different time periods. However, it is important to note that these simulations do not account for the influence of exogenous ejecta on basin filling. The process of ejecta deposition from large craters would have occurred concurrently with basalt emplacement34, potentially affecting the filling of the basins. Furthermore, studies utilizing data from the Chang'E-2 microwave radiometer and ground-based radar12,13 have provided insights into the shallow subsurface of the cryptomaria. Compared to spectral data, these techniques have the advantage of detecting deeper information, to infer the boundary between the upper ejecta of the cryptomaria and the lower mare basalts. This information has facilitated estimations of the thickness of the cryptomaria, ranging from a few meters to several hundred meters. In general, the estimation of cryptomaria thickness requires the utilization of multiple methods to enhance accuracy and reliability. Sori et al.17 employed gravity methods to estimate the cryptomaria thickness and subsequently verifying the existence of cryptomaria through the examination of DHCs. By combining various methods, researchers can leverage the complementary strengths of each approach to gain a more comprehensive understanding of the thickness and characteristics of the cryptomaria. We propose that a combination of microwave radiometer data, ground-based radar data, and DHCs features can be employed for estimating the thickness of the cryptomaria. This can be achieved by using the area of DHCs as a calibration reference point and establishing a functional relationship between the excavation depth of DHCs and remote sensing data for inverting the cryptomaria thickness in the region. In cases where the diameter of the DHCs is sufficiently large to penetrate the cryptomaria layer, the thickness of the cryptomaria can be observed from high-resolution images, leading to improved inversion results when used as the calibration point. Additionally, high degree and order gravity field model can provide insights into the lunar subsurface, and future gravity field models may offer enhanced accuracy in revealing the cryptomaria thickness. Utilizing the thickness of the cryptomaria estimated from the DHCs features as a constraint for forward modeling gravity anomalies and comparing this result with existing gravity field models may also present a viable method for improving the accuracy of cryptomaria thickness estimates.
Investigating the thickness of cryptomaria stands as a crucial undertaking in the realm of future lunar volcanism research. The insights garnered from these investigations are indispensable for quantifying the cumulative volume of lunar volcanic products and gaining deeper insights into ancient lunar volcanism.