Floor-fractured craters are a class of lunar craters marked by shallow floors with numerous large fractures on them [1; 2]. There have been two proposed formation mechanisms: 1) viscous relaxation  and 2) magmatic intrusion and sill formation  Recent morphological studies of floor-fractured craters utilizing data from the Lunar Orbiter Laser Altimeter (LOLA) and Lunar Reconnaissance Orbiter Camera-Wide Angle Camera (LROC-WAC) have determined that magmatic intrusion and sill formation is the more favorable formation mechanism for floor-fractured craters . The process associated with deformation by magmatic intrusion and sill formation involves a propagating dike ascending beneath a crater, stalling as a result of density changes in the subdense breccias beneath a crater, then propagating laterally beneath the crater floor, and inflating to form a sill or laccolith . The intrusion of the magma and inflation of the sill results in the shallow crater floor, and the numerous morphologic features seen on the crater floors, including the floor fractures, the floor topographic profile, dark halo craters, and patches of mare material . By introducing a large volume of high density magma beneath the crater floor, the process of magmatic intrusion and sill formation would also have a distinct effect on the observed gravity anomaly (specifically the Bouguer gravity anomaly) of floor-fractured craters [1; 5].
Complex craters on both the Earth  and the Moon  have observed negative Bouguer gravity anomalies, a signal contributed by the underdense breccia region beneath the crater. We thus postulate that floor-fractured craters hosting a subcrater intrusion will display a positive Bouguer anomaly, as the positive gravity contribution from the higher density magma body counteracts, or in some cases, overpowers the negative gravity contribution of the subdense breccias. Using the Goddard Space Flight Center 720 Bouguer Gravity solution of the Gravity Recovery and Interior Laboratory (GRAIL) data, we examine all cataloged  lunar floor-fractured craters with diameter greater than 20 km, and a gravity signature that is not overwhelmed by proximity of the gravity signature of any adjacent lunar basins. Of the eligible population of 88 craters, we determine that 71 have a positive Bouguer gravity anomaly in the crater floor region, 14 of the craters have a net zero Bouguer anomaly which we interpret as a balance between the contributions of the high density magma and the low density breccias. The remaining 3 craters all displayed a negative Bouguer anomaly, coupled with the relatively small degree of floor shallowing, and presence of only sparse floor fractures. Thus, gravity data provided by GRAIL supports floor-fractured crater formation by magmatic intrusion and sill formation. We now use the magnitude of the gravity anomalies in tandem with the topography data to better constrain the thicknesses of both the magmatic intrusion and the breccia region.
References:  P. Schultz (1976) The Moon 15, 241.  Jozwiak et. al. (2012) JGR 117, E11.  Z. Danes (1965) Astrogeologic Studies, Annu. Prog. Rep. A, 81.  Jozwiak et. al. (2013) LPSC XLIV, Abstract #2170.  R. Wichman and P. Schultz (1995) JGR 100. E10.  M. Pilkington and R. Grieve (1992) Rev. of Geophys. 30, 2.  J. Dvorak and R. Phillips (1977) GRL, 4.