- November 25, 2004 -
|
Unable to cite many earthquake data from the USGS [1], did I investigate into how much percent data the USGS lost. I compared the data of the Central Weather Bureau of Taiwan [2] in the time: 20040101~0803, latitude: 21.8~25 N, longitude: 120~125E, and magnitude: >= 4 with those of the USGS in the same time and area, but all magnitudes, and found that the USGS had lost data of 32.6%. I also did the same work between the data of Bogazici University, Turkey [3] in the time: 20030101~20030630, latitude: 33.9~42.2N, longitude: 24.7~44.4E, and magnitude: >= 4, and those of the USGS, and found a loss ratio of 24.2% for the USGS. Both the Central Weather Bureau of Taiwan and Bogazici University also have a loss ratio over 21% and 31.5% respectively. The details are in the tables [4-5] Like data error [6-7], data loss is a big problem for serious study. For example, my prediction No. 29 to the USGS in Table 5 [8] would be a miss if neither the Northern California Earthquake Data Center [9], nor the Univ. of Nevada [10] had offered the data. The USGS has been playing an important role to offer the data, but the data are not complete, and have large error sometimes. For example, the magnitude of the Sea of Okhotsk earthquake on Nov. 7, 2004 jumped from 5.6 [11] to 6.2 [12]. I hope the USGS and other data producers to peer on those problems for complete and accurate data that are important for earthquake prediction. |
| References |
|
|