1984 年 37 巻 3 号 p. 475-489
A method for evaluating detection capability of earthquakes was developed for an observational network. A function expressing detection probability P(M, x) for an earthquake with magnitude M, located at x was defined by combinating the individual detection probability determined for each station.
By using this formula, the detectable regions of microearthquakes were drawn for the Kanto-Tokai Observational Network of NRCDP, which consisted of 44 seismometer stations as of March 1983. It was confirmed that shallow earthquakes (h≤30km) with magnitude greater than 2.0 could certainly be detected within the network covering the Kanto and Tokai areas. For earthquakes with smaller magnitude, the network has a gap of detectable region around the Boso peninsula.
For those selected regions of relatively high detection capability, frequency distributions of magnitude were investigated, and b-values and other parameters were obtained. It was found that in most cases, the Gutenberg-Richter's relation was satisfied down to fairly small magnitude of 1.1-2.0, with ordinary b-values of about 0.9.