中长期地震危险性概率预测中的统计检验方法Ⅱ:N-test和L-test方法

地震 ›› 2013, Vol. 33 ›› Issue (1): 47-55.

中长期地震危险性概率预测中的统计检验方法Ⅱ:N-test和L-test方法

高朝军¹, 蒋长胜², 韩立波², 李艳娥²

1.新疆维吾尔自治区地震局, 乌鲁木齐 830011;
2.中国地震局地球物理研究所, 北京 100081

收稿日期:2012-06-07 修回日期:2012-08-01 出版日期:2013-01-31 发布日期:2020-10-16
作者简介:高朝军(1981-),男,河南南阳人,研究实习员,主要从事数字地震学和地震预测研究。
基金资助:
地震行业科研经费项目(201008007)、国家国际科技合作专项项目(2012DFG20510)和震情跟踪青年课题(2012020113)联合资助

Statistical Test Methods of Intermediate-term and Long-term Probabilistic ForecastingⅡ: N-test and L-test Method

GAO Chao-jun¹, JIANG Chang-sheng², HAN Li-bo², LI Yan-e²

1. Earthquake Administration of Xinjiang Uygur Autonomous Region, Xinjiang 830011, China;
2. Institute of Geophysics, CEA, Beijing 100081, China

Received:2012-06-07 Revised:2012-08-01 Online:2013-01-31 Published:2020-10-16

摘要/Abstract

摘要： 本文介绍了国际“地震可预测性合作研究”CSEP计划中, 用于地震概率预测模型检验的N-test和L-test方法的基本原理、发展动态。以地震活动平滑模型(smoothed seismicity model)在新疆天山中部地区的中长期预测结果为例, 本文尝试使用N-test和L-test统计检验方法对预测结果进行评价。结果表明, 上述方法能够较好地评价预测结果, 并有助于构建更为科学合理的地震预测模型及相关参数设置。

关键词: 地震概率预测模型, 统计检验, N-test方法, L-test方法

Abstract: In this paper, we introduce the principles and developments of the N-test and L-test methods which have been used to evaluate earthquake probabilistic forecasting models as routine statistical test methods in the CSEP (Collaboratory for the Study of Earthquake Predictability) project. Selecting the middle section of Tianshan region as the study area, we perform evaluation of the Smoothed Seismicity model for intermediate-term and long-term retrospective forecasting with N-test and L-test methods. The results show that N-test and L-test methods are efficient to evaluate the earthquake prediction models and can be used to optimize parameter setting in the earthquake forecasting models.

Key words: Earthquake probabilistic forecasting, Statistical test, N-test method, L-test method

中图分类号:

P315.7

高朝军, 蒋长胜, 韩立波, 李艳娥. 中长期地震危险性概率预测中的统计检验方法Ⅱ:N-test和L-test方法[J]. 地震, 2013, 33(1): 47-55.

GAO Chao-jun, JIANG Chang-sheng, HAN Li-bo, LI Yan-e. Statistical Test Methods of Intermediate-term and Long-term Probabilistic ForecastingⅡ: N-test and L-test Method[J]. EARTHQUAKE, 2013, 33(1): 47-55.

参考文献

[1] Schorlemmer D, Zechar J D, Werner M J, et al. First results of the Regional Earthquake Likelihood Models experiment[J]. Pure Appl Geophys, 2010, 167: 859-876.
[2] Schorlemmer D, Gerstenberger M C. RELM Testing Center[J]. Seismol Res Lett, 2007, 78: 30-36.
[3] Werner M J, Zechar J D, Marzocchi W, et al. Retrospective evaluation of the five-year and ten-year CSEP-Italy earthquake forecasts[J]. Ann Geophys, 2010, 53(3): 11-30.
[4] Marzocchi W, Zechar J D. Earthquake forecasting and earthquake prediction: different approaches for obtaining the best model[J]. Seismol Res Lett, 2011, 82(3): 442-448, doi:10.1785/gssrl.82.3.442.
[5] Kagan Y Y, Jackson D D. New seismic gap hypothesis: Five years after[J]. J Geophys Res, 1995, 100(B3): 3934-3959.
[6] Schorlemmer D, Gerstenberger M C, Wiemer S, et al. Earthquake likelihood model testing[J]. Seismol Res Lett, 2007, 78: 17-29.
[7] Zechar J D. Evaluating earthquake predictions and earthquake forecasts: a guide for students and new researchers, Community Online Resource for Statistical Seismicity Analysis[J]. 2010. doi:10.5078/corssa-77337879. Available at http://www.corssa.org.
[8] Frankel A, Mueller C, Barnhard T, et al. National seismic hazard maps[R]. USGS Open-File Report, 1996, 96-532.
[9] Helmstetter A, Kangan Y Y, Jackson D D. Comparison of short-term and long-term earthquake forecast models for Southern California[J]. Bull Seism Soc Amer, 2006, 96: 90-106.
[10] Frankel A. Mapping seismic hazard in the central and eastern United States[J]. Seism Res Lett, 1995, 66: 8-21.
[11] Gutenberg R, Richter C F. Frequency of earthquakes in California[J]. Bull Seism Soc Amer, 1944, 34: 185-188.
[12] Kagan Y Y, Jackson D D. Long-term probabilistic forecasting of earthquakes[J]. J Geophys Res, 1994, 99: 13685-13700.
[13] Kagan Y Y, Jackson D D, Rong Y. A testable five-year forecast of moderate and large earthquakes in Southern California based on smoothed seismicity[J]. Seism Res Lett, 2007, 78: 94-98.
[14] Zechar J D, Jordan T M. Simple smoothed seismicity earthquake forecasts for Italy[J]. Annals of Geophyscics, 2010, 53(3): 99-105.
[15] Nanjo K Z, Tsuruoka H, Hirata H, et al. Overview of the first earthquake forecast testing experiment in Japan[J]. Earth Planers Space, 2011, 63: 159-169.
[16] Gardner J K, Knopoff L. Is t he sequence of eart hquakes in Southern California, with aftershocks removed, Poissonian?[J]. Bull Seism Soc Amer, 1974, 64(5): 1363-1367.
[17] Reasenberg P A. Second-order moment of Central California seismicity, 1962—1968[J]. J Geophys Res, 1985, 90: 5479-5495.
[18] 李志海, 蒋长胜, 黄瑜, 等. 新疆地区地震目录最小完整性震级和台网科学布局研究[J]. 地震学报, 2011, 3(6): 763-775.
[19] Imoto M, Rhoades D A, Fujiwara H, et al. Conventional N-, L-, and R-tests of earthquake forecasting models without simulated catalogs[J]. Earth Planets Space, 2011, 63: 275-287.