The lunar surface is dominated by impact cratering, volcanism, and tectonics. One such suggested tectonic feature is the large population of elongate linear to arcuate rilles, which are topographic depressions with steep sides and flat floors. The rilles appear to be broadly circumferential to the large basins on the lunar nearside and extend up to hundreds of kilometers. Previous workers have shown that the rilles ceased formation around 3.6 Ga. These rilles are canonically interpreted to be graben formed by the flexural response of the lunar lithosphere to mare basalt loading within large basins. Additional work has suggested that some of the rilles represent the surface manifestation of volcanic dikes that have stalled in the near-surface, based on the associations of rilles with volcanic features (vents, mounds, pyroclastic materials, sinuous rilles) (Petrycki et al., 2004; LPSC 35). Determining the fraction of the rille population that is of a volcanic origin is important to the understanding of the intrusive magmatic history of the Moon.
In order to distinguish between the two primary rille formation hypotheses (tectonic vs. volcanic), we are generating a global database of lunar rilles in ArcMap building on previous work by Petrycki et al. (2004). Our database includes morphologic and morphometric information designed to differentiate between a tectonic or volcanic origin for each individual rille. For example, we include in our database associations with volcanic features such as vents or mounds, off-shooting sinuous rilles, pyroclastic deposits and the dimensions and variations of these volcanic features along strike. The presence of such volcanic features is considered evidence for a dike origin for the rille, and the absence of such features is more suggestive of a tectonic origin.
In order to estimate the geometry/volume of potential dikes, we also record the rille length, average width, and average floor width. If a given rille is the surface expression of a dike, the associated flexure should be revealed by the rille-normal topographic profile. We record crustal thickness from GRAIL and the average floor depth within the rille ArcMap polygon, which can be used to estimate the depth to the dike-tip. We also record the terminus morphology (flared, rounded, angular); because dikes are disc-shaped and the width of the dike depends on the depth to the dike-tip, a rille formed by a subsurface dike is predicted to exhibit a flared-out terminus, reflecting the deepening of the dike-tip at the terminus. We record the angle of intersection of cross-cutting rilles, which provides information on the orientation of the regional stress field at the time of rille formation. We also record the substrate (mare, highlands) and sinuosity of the rille as a simple measure of whether the rille is linear, arcuate, or more angular/sinuous for comparison and potential correlation with the other datasets described above. Finally, we record embayment relationships to determine stratigraphic age for examination of the temporal history of the rilles. Our initial database reveals numerous examples of linear and arcuate rilles that are interpreted to be the surface manifestation of intrusive dikes.