重力異常分布図からの伏在カルデラリム抽出法

抄録

Many studies have investigated the relationships between gravity data and subsurface caldera structure beneath the area. Difficulties may arise, however, for these analyses because of different interpretations of the buried rims of the calderas. For buried calderas, ground deformations and erosional processes including landslides have significantly reduced the appearance of the original caldera rims. Generally, the interpretation of caldera rims is based on gravity data together with geological observations, because the preserved rim has a large negative gravity anomaly that reaches minima over the marginal low-density ridges. In this study, we propose an effective method to enhance the delineation of caldera rims from gravity data using watershed delineation in GIS. The proposed method incorporates a correlation analysis to appropriately assume a rock density to correct the gravity anomalies, and a band pass filter to eliminate noises in the data. Using gravity anomalies corrected with 2.60 g/cm³ of assumed rock density, and with bandpass filter that spanned the size-range of calderas, we have successfully extracted rims of the late Cenozoic calderas from NE Honshu, Japan.