Reliability of Assessing Lunar Crustal Density Structures Using the GRAIL Gravity Model

Abstract The spatial resolution of gravity data acquired by the Gravity Recovery and Interior Laboratory (GRAIL) mission is adequate to resolve relatively small-scale density structures larger than approximately tens of kilometers within the lunar crust, which record important information about lunar geological evolution invisible from the surface. Band-limited spherical harmonic (SH) expansion is frequently employed to separate gravity signals induced by small-scale crustal density structures from those generated by large-scale structures. However, this procedure inevitably introduces ringing artifacts, i.e., the Gibbs effect. The potential effects of the Gibbs phenomenon on the identification and interpretation of lunar crustal density structures remain understudied. Here we investigate this effect by modeling gravity signals of density structures in both the lunar crust and mantle, with additional consideration of uncertainties of the GRAIL gravity data. The modeled density structures include shallow dense intrusions and deeper mantle uplifts with diverse geometries. Model results show pronounced effects of the Gibbs phenomenon on the high-pass-filtered gravity anomaly. We find that GRAIL Bouguer gravity models may contain substantial residual signals from large-scale structures due to the Gibbs effect. Therefore, assessment of lunar crustal density structures using high-pass-filtered GRAIL Bouguer gravity anomalies with a specific starting SH degree carries substantial uncertainties. The Gibbs effect can be evaluated in a relatively reliable way by combining full-band gravity anomaly spectra with forward modeling of gravity signals caused by possible density structures. To investigate potential crustal density structures using GRAIL data, we recommend analyzing gravity anomaly spectra in the full range of SH degrees.