A total of 242 wrinkle ridges, 137 lobate scarps, and 148 unidentified structures, with a total length of ~10,991 km, were mapped in this study (Fig. 3). The length of individual segments ranges from ~1 km to ~175 km, with a mean length of ~21 km. The mapped wrinkle ridges have a total length of ~7,852 km and range from ~3.7 km to ~175 km. The wrinkle ridge mean length is 32.8 km. Lobate scarps have a total length of ~946.4 km with a minimum length of ~1 km and a maximum length of ~58.5 km. The lobate scarp mean length is ~7 km.

The differences in the appearance of the ridge segments allow distinguishing four different classes. These classes are crisp, moderately degraded, advanced degraded, and heavily degraded. They differ from one another in their erosional state, general structure, surface texture, crosscut relationships, and small graben occurrence. However, transitions between the different degradation classes are gradual. A total of 846 segments of contractional tectonic features were mapped of which 658 segments were classified (Fig. 4). Their appearances and occurrences are described in the following.

A total of 49 segments with a cumulative length of ~451 km and an average length of ~9.2 km were classified as crisp (Fig. 5). Consequently, they represent 5.1% of the total mapped length. Crisp features consist of wrinkle ridges and lobate scarps. All of them occur scattered within Mare Tranquillitatis and are often close to moderately degraded features (Fig. 4). In general, they have a NW – WNW orientation. Crisp features have sharp edges, and steep slopes on a small scale (< 100 m; Fig. 5). They are generally relatively small structures in terms of length and width and have a winding and lobate appearance. They often braid and cross each other along strike. The crisp wrinkle ridges often resemble a lobate scarp morphology, with a simple asymmetrical profile and, in some cases, a seemingly missing broad arch. Often smaller surface-breaking tectonic features occur in their vicinity. Crisp features, generally, crosscut small craters (Fig. 5c; < 50 – 100 m diameter) and wrinkle ridges often appear to be surface breaking when they crosscut craters. Clusters of small (width < 50 m) crisp appearing graben and troughs are present on top of and in the close vicinity of crisp features (Fig. 5). Generally, the graben are located at the hanging wall and are oriented perpendicular and parallel to the latter. Small boulder patches are visible occasionally (Fig. 5b).

About 100 segments were classified as moderately degraded (Fig. 6). They have a total length of ~780 km, which makes up 8.9% of the total mapped length. On average, they have a ~7.8 km length and generally show a NW orientation. Moderately degraded features are comprised of lobate scarps and wrinkle ridges. The structures are similar in size to the crisp segments, but the edges can be more indistinct than crisp features. In general, they have a winding and lobe-like morphology, and they often braid and cross each other. Only a few small craters superimpose the segments. They typically crosscut several craters along their length, which mostly have diameters of larger than 100 m (Fig. 6). Small graben are generally not associated with these structures. They occur throughout Mare Tranquillitatis and can be spatially associated with crisp, advanced, and heavily degraded features.

Advanced degraded features (Fig. 7) are the second most common class and are dominantly comprised of wrinkle ridges. A total of 251 advanced degraded wrinkle ridges with an average length of ~11 km have been mapped, resulting in a total length of ~2,762 km. This class represents 31.5% of the total length. They are generally the most massive wrinkle ridges with respect to width and topography (up to 10s of kilometers wide and hundreds of meters high). Their rounded morphology mostly resembles the traditional wrinkle ridge definition, with a, in some cases km scaled, broad arch and an asymmetric superimposed steep crest (Fig. 7a). The changes in the orientation of the wrinkle ridge asymmetry are either gradual or abrupt. Smaller ridge segments of higher order occur in front or back and on the top of these wrinkle ridges. Structures of higher order can transition to first-order ridges along their strike. On slope maps, advanced degraded wrinkle ridges show slopes up to > 30°. They have a larger number of superimposed craters than the previously described morphological classes. However, the abundance of superimposed craters is often lower than crater abundances in the surrounding mare units. These wrinkle ridges can deform and crosscut craters with diameters of several hundred meters, but most segments do not crosscut any craters. The surfaces often show a crisscross pattern that previous studies described as an “elephant-hide” pattern (Fig. 7b; Gold, 1972; Zharkova et al., 2020; Bondarenko et al., 2022). Extensive boulder fields are associated with some advanced degraded wrinkle ridges (Fig. 7).

The most common class are the heavily degraded features (Fig. 8), which also dominantly consist of wrinkle ridges. 258 segments, with a total length of ~3,140 km and an average length of ~12.1 km, were mapped. As a result, 35.8% of the total length is represented by this class. While their overall structure can be similar to advanced degraded wrinkle ridges, they generally have an indistinctive and diffuse morphology with more rounded edges (Fig. 8), and the classical wrinkle ridge structure is often only visible in topographic data. They have many superimposed craters and generally do not crosscut any craters, but rarely can deform craters with diameters of several hundred meters. There are no associated small graben present. Their surface texture can resemble an elephant-hide structure. In general, advanced and heavily degraded wrinkle ridges are similarly distributed. However, individual wrinkle ridge assemblages are generally represented mainly by one of both classes. In general, heavily degraded wrinkle ridges occur less commonly together with crisp and moderately degraded wrinkle ridges than advanced degraded wrinkle ridges. Both classes represent the largest wrinkle ridge structures in Mare Tranquillitatis in length, width, and height.

The sharp-edged morphology and the relatively small size of the crisp wrinkle ridges and lobate scarps suggest a relatively young formation age in contrast to advanced degraded and heavily degraded features. Crisp features can crosscut craters with diameters of less than 50 - 100 m. Craters of these sizes are estimated to be of Copernican ages (< 800 Ma; Wilhelms et al., 1987), because older craters of this size would have been infilled and degraded since then (Trask, 1971; Basilevsky, 1976; Fassett & Thomson, 2014). Thus, it is possible to establish a Copernican age, i.e., an upper age limit of ~800 Ma for these landforms. Since tectonic activity would result in seismic shaking and thus in enhanced degradation of the small craters, the upper limit is presumably overestimated (Williams et al., 2019). CSFD measurements also support Copernican ages for lobate scarps with similar crisp morphologies (van der Bogert et al., 2018; Clark et al., 2017) and possibly even wrinkle ridges (Valantinas et al., 2018). Accompanying crisp features are small fresh graben and troughs (Fig. 5). The existence of small crisp graben situated near lobate scarps was first documented at the back-limb of the Lee-Lincoln scarp, close to the Apollo 17 landing side (Watters et al., 2010). Since then, more of these structures were found in the vicinity of lobate scarps (French et al., 2015; Watters et al., 2012) and wrinkle ridges (French et al., 2015; Williams et al., 2019). Small graben observed in Mare Tranquillitatis are similar in their dimensions to the graben described in the latter studies. They typically have widths of less than 50 m and, in many cases, of even less than 10 m. Because of their similarity to sizes measured in other studies, depths of ~17 m to ~1 m can be assumed (Watters et al., 2012; Williams et al., 2019). Fill rates of shallow depressions in lunar regolith are estimated to be 5 ± 3 cm/Ma (Arvidson et al., 1975). Therefore, a ~1 m deep graben should be filled entirely with regolith after ~12.5 to ~50 million years, which implies formation ages of less than 50 Ma. Due to their association with lobate scarps, Watters et al. (2012) suggested that these graben form by uplift and flexural bending resulting from the movement at the underlying thrust fault. Thus, these graben can be viewed as possible evidence for tectonic activity of crisp features during the last 50 Ma (French et al., 2015; Watters et al., 2012; Williams et al., 2019). Lu et al. (2019), used ejecta boulders of craters crosscut by small wrinkle ridges in Mare Imbrium to calculate the individual crater ages since boulder abundances decrease with exposure time (Basilevsky et al., 2013; Ghent et al., 2014; Lu et al., 2019). The derived ages support wrinkle ridge formation during the last 10s of Ma (Lu et al., 2019). The morphology of the young wrinkle ridges studied by Lu et al. (2019) are indistinguishable from crisp wrinkle ridges in Mare Tranquillitatis. In summary, different methods indicate the formation of young wrinkle ridges and lobate scarps on the Moon during the last few 10 to 100 Ma. Thus, we propose tectonic activity for crisp wrinkle ridges and lobate scarps in Mare Tranquillitatis at least during the last 50 Ma, which further highlights the global recent wrinkle ridge formation.

Based on our study we cannot conclusively estimate formation ages for moderately degraded features. Crater crosscutting relationships imply younger ages for moderately degraded wrinkle ridges and lobate scarps than advanced degraded features. The main difference between moderately degraded and crisp features, next to a more rounded morphology, is the apparent lack of small graben. However, while small graben can be seen as possible evidence for recent tectonic activity, it is unknown whether they necessarily have to form during recent activity. Therefore, the lack of crisp graben does not necessarily imply an older age. Furthermore, because of the small size and faint appearance of these graben, as well as the missing NAC coverage (incidence angles between 55° and 90°) of some features, a wider distribution of undetected graben is possible. We estimate that moderately degraded wrinkle ridges and lobate scarps have a broad range of formation ages in between crisp and advanced degraded features.

Crisp features occur scattered within Mare Tranquillitatis and do not align with patterns predicted by basin loading and subsidence. Hence, subsidence does not seem to be the major controlling factor of young wrinkle ridge and lobate scarp formation. Additionally, they are not correlated with positive gravitational anomalies within the mare (Fig. 9, 10). However, as previously stated, Mare Tranquillitatis is of irregular shape, which could influence subsidence-induced stress patterns, and the role of the thickness of the elastic lithosphere in wrinkle ridge formation is also a factor (Watters, 2022). Previous studies discussed the prolonged cooling, triggered by the abundance of heat-producing elements, of the Procellarum KREEP Terrane (PKT) to be a factor in recent wrinkle ridge formation (Daket et al., 2016; Lu et al., 2019). However, Mare Tranquillitatis is not associated with the PKT (Wieczorek & Phillips, 2000). Therefore, this model does also not explain the recent formation of wrinkle ridges and lobate scarps in Mare Tranquillitatis. Late-stage global compressional stresses are consistent with both an initially completely molten Moon and an initially hot exterior and magma ocean (Binder & Gunga, 1985; Solomon & Head, 1979; Watters et al., 2019; Williams et al., 2013). The interior cooling of the Moon could result in compressional stresses of ≥ 2, but < 10 MPa (Watters et al., 2015, 2019). For shallow thrust faults to form, an estimated ~2 - 7 MPa are sufficient (Watters et al., 2019; Williams et al., 2013). Small-scale wrinkle ridges were likely formed by shallow thrust faults (Lu et al., 2019; Watters, 2004). The derived depths from Lu et al. (2019) for small wrinkle ridge thrust faults are similar to suggested depths of shallow lobate scarps (~1 km; Williams et al., 2013). Concluding, global compression seems to be a likely candidate as the driving force behind recent wrinkle ridge and lobate scarp formation on the Moon and in Mare Tranquillitatis. Global lobate scarp patterns and the timing of detected moonquakes highlighted the possible influence of tidal forces, such as orbital recession, diurnal tidal stresses, and true polar wander onto the lunar stress field (Matsuyama et al., 2021; Watters et al., 2019). Models of an additional influence of SPA ejecta loading onto the global stress field also showed good fitting results and are discussed as an alternative or addition to the influence of tidal forces (Matsuyama et al., 2021). N to NW orientated faults between ~20°E and ~40°E, and ~0°N to ~20°N are predicted by a combination of recession stresses, diurnal tidal stresses at apogee, and global contraction (Watters et al., 2015, 2019), as well as by a combination of SPA loading, true polar wander, and global contraction (Matsuyama et al., 2021). These predicted trends approximately correspond with the W to NW orientation of crisp ridges and scarps within Mare Tranquillitatis (Fig. 11), suggesting their formation is consistent with those models. However, it should be highlighted that the lithopsheric stressfield is a result from the complex interaction and overlaying of multiple stresses, evolving with time. Additional influences like, e.g., late stage mare basalt cooling (Tian et al. 2021), stresses related to a possible movement of magma in connection with young volcanic activity in Mare Tranquillitatis (Braden et al., 2014; Qiao et al., 2020), or preexisting ancient faults in the basement might have influenced the regional stress field. The patterns of moderately degraded wrinkle ridges allign with both the patterns of advanced and heavily degraded features, as well as with some crisp ridges and scarps (Fig. 4). Hence, moderately degraded wrinkle ridges and scarps could reflect the evolution of the stressfield from dominantly basin-localized to a dominately global stressfield, and they could represent the continued growth of ancient faults.

The large size and strongly degraded morphology of advanced and heavily degraded features suggest an older formation age relative to moderately degraded and crisp features. Advanced and heavily degraded features deform all the mare units defined by Hiesinger et al. (2000), which have ages of ~3.4 to ~3.8 Ga. Consequently, they have an upper formation age limit of at least 3.8 Ga. Since Rupes Cauchy and some large graben are deformed by advanced and heavily degraded wrinkle ridges, some of the wrinkle ridge formation occurred after 3.6 Ga (Lucchitta & Watkins, 1978; Watters et al., 2009). They can deform craters of several hundred meters in diameter but generally deform no craters, which agrees with Nectarian, Eratosthenian, and Imbrian formation ages (Trask, 1971). The deformation of craters with diameter ranges of hundreds of meters implies on the basis of crater degradation (Basilevsky, 1976, Fassett & Thomson, 2014) that at least some wrinkle ridges experienced ongoing activity throughout the Nectarian, Eratosthenian, and possibly even the Copernican. Yue et al. (2017) found a young average age of large wrinkle ridges in Mare Tranquillitatis (~2.4 Ga) relative to other lunar maria (~3.3 Ga). With the focus on Mare Tranquillitatis and the degradation-state approach of our classification, this age discrepancy between wrinkle ridges in Mare Tranquillitatis and other maria cannot be resolved. Relatively younger ages of advanced degraded wrinkle ridges compared to heavily degraded ridges can only be suggested and not conclusively proven. More precise dating methods than our morphological analysis are needed to uncover the early tectonic evolution of the maria basins, however, standard CSFD measurements on wrinkle ridges are challenging, because of steep slopes, small count areas, and the often hummocky and heterogeneous terrain (Frueh et al., 2020). Hence, buffered crater counting might be a more suitable option to obtain formation ages of individual wrinkle ridges.

The occurrence of the advanced and heavily degraded concentric and radial wrinkle ridges in the western mare appears to have been localized by a subsurface feature (Fig. 9, 10; Freed et al., 2001; Schleicher et al., 2019). These concentric and radial wrinkle ridges, as well as several concentric large graben, can be attributed to the Lamont gravity anomaly, which is argued to be a ghost crater (Dvorak & Phillips, 1979; Scott, 1974) or a feature of volcanic origin (Zhang et al., 2018). Next to the Lamont anomaly, western Mare Tranquillitatis is characterized by a positive gravitational anomaly ranging from Mare Nectaris in the south to Mare Serenitatis in the north (Fig. 9, 10). Correlated with this positive anomaly are the thickest basalts in Mare Tranquillitatis (Dvorak & Phillips, 1979; De Hon, 1974, 2017; Konopliv et al., 2001; Zuber et al., 2013). At the surface, this positive anomaly is expressed as an elongated depression. Advanced and heavily degraded wrinkle ridges and large graben within this depression occur parallel to the gravitational anomaly and the topographic depression, which could also imply a subsidence-related origin (Fig. 3; McGovern et al., 2022). Mare Serenitatis most likely influenced the northwestern Mare Tranquillitatis, resulting in a radial wrinkle ridge and parallel graben to Mare Serenitatis (Fig. 3). Wrinkle ridges close to the eastern mare boundary follow a NS trend similar to the general trend of the eastern boundary itself, which is consistent with an origin from basin loading and subsidence. The eastern mare boundary shows no clearly detectable gravitational anomalies that have been associated with deep fractures of the mare basement (Fig. 10; Andrews-Hanna et al., 2018). The fewer number and the less coherent patterns of features in the eastern mare could be a result of the shallower basalts and, therefore, less basin loading induced by subsidence. While basin loading and subsidence influenced the regional stress field and the tectonic patterns in Tranquillitatis, additional global stress fields contributing to wrinkle ridge formation have been proposed. One possible influence on the global stress field are deep transient stresses generated by the South Pole-Aitken (SPA) basin (Schultz & Crawford, 2011), which predicts antipodal failures on the lunar nearside due to extensions deep within the Moon. Schultz and Crawford (2011) suggested reactivated deep-seated faults localized the wrinkle ridges. Wrinkle ridge patterns of the lunar nearside do spatially correlate with the predicted patterns (Schultz & Crawford, 2011; Valantinas & Schultz, 2020), however, it is not clear that SPA-related stresses would not have largely relaxed before the period of wrinkle ridge formation in Tranquillitatis. Another discussed potential model is the fault adjustment correlated with deep-seated intrusions on the lunar nearside. In this case, wrinkle ridges would be the surface expression of these deep-seated intrusions (Andrews-Hanna et al., 2014). GRAIL Bouguer gravity gradient data revealed a possible polygonal pattern of ancient deep intrusion connecting most of the lunar maria and also Mare Tranquillitatis (Andrews-Hanna et al., 2014). The western elongated positive gravitational anomaly of Mare Tranquillitatis is proposed to originate from these deep-seated intrusions. However, following the linear unrestricted growth trends and the similar displacement values of wrinkle ridges associated and not associated with these proposed intrusions, there is little evidence that most of the ridge faults were influenced by buried structures associated with ancient rifts (Watters, 2022). Having said that, we do observe wrinkle ridges correlated with sharp increases in elevation, possible extensive folding, and an elevation offset between both sites of the fold located in this western part of the basin. Possible similar wrinkle ridges have been associated with deeply rooted faults penetrating the base of mare deposits (Byrne et al., 2015; Watters, 2022). The slopes of the trough associated with the positive gravitational anomaly, however, complicates the assessment of the actual extent of folding. Quantifying the deformation in Mare Tranquillitatis will be part of a planned follow-up study. In summary, the compressional stresses that resulted in the formation of advanced and heavily degraded wrinkle ridges in Mare Tranquillitatis originated primarily from load-induced subsidence with other possible sources of regional or global stress, like SPA induced stress and fault adjustment, which changed with time. Individual wrinkle ridges in western Tranquillitatis could be correlated to the deep-rooted faults.

It should be noted that the ancient age of advanced and heavily degraded feature, which we we imply in this study, is the formation age and not necessarily the age of the most recent activity along the fault. Previous studies discussed the possible activity of ancient wrinkle ridges during the last Ma (French et al., 2019; Valantinas & Schultz, 2020). One possible evidence is the abundance of boulders at wrinkle ridge crests (French et al., 2019; Valantinas et al., 2017; Valantinas & Schultz, 2020; Watters et al., 2019). Valantinas and Schultz (2020) suggested that layered mare basalts buckled and regolith drained into small fractures during episodes of uplift, exposing the buckled material below. Basilevsky et al. (2013) and Ghent et al. (2014) found that 50% of rock populations, with fragment diameters larger than 2 m, are destroyed after 40 - 80 Ma and 99% after 150 - 300 Ma. Following the boulder size of wrinkle ridge boulder fields, Valantinas and Schultz (2020) proposed that wrinkle ridges with high boulder abundance were active during the last tens of millions of years. Boulder density increases with increasing slope. This leads to the question whether boulders are simply associated with steep slopes rather than ongoing wrinkle ridge activity since shallow seismic shaking generated by impacts or tectonic activity unrelated to wrinkle ridges could also result in the exposure of boulder fields (French et al., 2019). For our classification, the abundance of boulders was not used to determine the possible erosional state of a wrinkle ridge segment. Crisp and moderately degraded features do usually not appear in DiVINER rock abundance maps and boulders are only visible occasionally in small patches. Thus, no or merely a few boulders have been exposed during their activity. This could be either evidence against boulder fields as a general sign of recent tectonic activity, related to the relation between flow thickness and thrust fault size, or correlated with the physical properties of the basalt flows, which result in different predicted rock abundances (Elder et al., 2022). Boulder-rich wrinkle ridges mapped by Valantinas and Schultz (2020) tend to correlate with advanced degraded wrinkle ridges rather than heavily degraded. However, it should be noted that the boulder fields themselves could influence the morphological classification since they typically appear brighter than the regolith (Fig. 7a), possibly resulting in a greater contrast between the sunlit and shadow side and, therefore, in a seemingly more defined appearance. Five segments from Valantinas and Schultz (2020) can be associated with moderately degraded wrinkle ridges. These are located at the southeastern Lamont ring and a single wrinkle ridge in southwestern Mare Tranquillitatis (Fig. 12). All of these ridges occur together with advanced and heavily degraded wrinkle ridges and are larger in relief than other moderately degraded features. They deform craters with ~100 m in diameter or are accompanied by faint and small graben-like features (Fig. 12b). The size of those moderately degraded wrinkle ridges, their transitional morphology between moderately degraded and advanced degraded features, and their associated patterns with advanced and heavily degraded wrinkle ridges suggest possible ancient wrinkle ridges, which were later modified by more recent activity. Two large wrinkle ridges directly north of the Lamont Anomaly could also contribute to this discussion (Fig. 3, 4, 9). The eastern ridge (7.0°N, 22.7°E) shows a high boulder abundance, while the western ridge (7.0°N, 22.1°E) only exhibits boulder fields in its northern most part (Valantinas & Schultz, 2020). Segments of the eastern ridge are classified as advanced degraded and segments of the western ridge, mainly, as heavily degraded. Additionally, the eastern wrinkle ridge shows one small off-shoot segments, which is classified as moderately degraded (Fig. 4). The off-shoot segment could highlight a continuing growth of the eastern wrinkle ridge, which resulted in the formation of boulder fields. In those cases, boulder fields might be a sign of recent activity of the wrinkle ridges. However, since boulders might also be exposed due to seismic shaking unrelated to wrinkle ridges (French et al., 2019) and boulder fields can also be found on other positive relief features than wrinkle ridges, we are not able to conclude if all boulder rich wrinkle ridges in Mare Tranquillitatis are boulder-enriched due to their recent tectonic activity. In addition, crisp and moderately degraded features have not exposed extensive boulder fields. Thus, we can neither clearly support nor reject boulder fields along wrinkle ridges as a general sign of recent tectonism.

Boulder-rich wrinkle ridges on the lunar nearside are proposed to be part of an active nearside tectonic system (ANTS). A possible origin for this recent activity was assigned to the previously discussed deep transient stresses generated by the South Pole-Aitken (SPA) basin (Schultz & Crawford, 2011; Valantinas & Schultz, 2020) and a continued fault adjustment correlated with deep-seated intrusions (Andrews-Hanna et al., 2014; Valantinas & Schultz, 2020). Recorded deep moonquakes might be evidence of the SPA-induced stresses (Valantinas & Schultz, 2020). However, as previously stated, the influence of these proposed putative mare-filled ancient rifts and intrusions on wrinkle ridge formation has been questioned (Watters, 2022). Also, it remains unknown if the current lunar stress field would allow for ongoing readjustment of those ancient faults. Thus, the question of young activity associated with ancient wrinkle ridges and the implications of boulder fields remains unresolved. The reactivity or prolonged activity of ancient wrinkle ridges, however, seems to be true for at least some individual wrinkle ridges in Mare Tranquillitatis.

In this study, compressional tectonic features were mapped in Mare Tranquillitatis and classified into crisp, moderately degraded, advanced degraded, and heavily degraded, based on their morphology and erosional state. This classification allows to suggest formation ages and possible origins of these features:

Mare Tranquillitatis exhibits compressional tectonic features with a variety of formation ages ranging from ancient to recent. The complex and changing stress field behind wrinkle ridge formation is presumably a result of a combination of different factors, which underlines the need for new studies. Furthermore, our results highlight and strengthen the case for a still tectonically active Moon within and outside of the maria basins. To further uncover the active lunar tectonism, the future installation of a geophysical network on the Moon is highly desirable (Fuqua Haviland et al., 2022).