基于ETAS模型对青海地区余震预测效能评估的研究

为系统评估青海地区余震短期发生率的预测效能,以及构建适合地震活动特点的余震早期预测策略和预测指标体系,利用国际上当前较为前沿的时间序列ETAS模型和"瘦化算法"对青海地区2009年以来的8个地震的早期余震序列参数进行拟合,并利用N-test检验方法对预测结果进行回溯性的效能评估。研究表明:ETAS模型和"瘦化算法"对青海地区的余震发生率具有很好的预测能力,建议采用3天的预测时间窗,且对序列早期阶段进行应用,或可取得"最佳"的预测效果。

Evaluation of the Forecasting Effectiveness of the Short-term AftershockOccurrence Rate and Forecasting Strategy in Qinghai Region

To evaluate the forecasting effectiveness of the occurrence rate of short-term aftershocks in the Qinghai region, and to construct an early aftershock forecasting strategy and forecast index system that reflects the regional seismic activity, we performed a continuous and sliding fitting of the sequence parameters of eight earthquakes that have occurred in the Qinghai area since 2009. To do so, we used the epidemic-type aftershock sequence (ETAS) model and the ETAS-based thinning algorithm, and used the N-test method to evaluate the forecasting effectiveness. The prediction results show that the ETAS model and ETAS-based thinning algorithm have good forecasting ability regarding the aftershock occurrence rate in the Qinghai region. We found that the "best" forecasting result may be obtained using a 3-day forecasting time window and applying the algorithm in the early stage of the sequence.