中国大陆的特征地震活动及其中一长期预测研究（一）

Studies on Characteristic Earthquake Activities and Their Mid-long-term Forecasting in China Mainland (Ⅰ)

中国大陆不同地点２４个特征地震轮回的震级一时间序列分析表明：在一个完整轮回中，相继特征地震事件ｉ和ｊ之间时段的前１／２极少有破坏性地震发生；而在轮回的后１／２，大多数情况下至少有一次中－强地震发生；其中，中一强地震发生在轮回最后１／３时段的约占５４％。称这种地震为“填隙式中强震”（事件ｆ），其发生可能标志着当地自从上一次特征地震后，应变已重新积累到达距下一次特征地震极限应变不远的背景，故对于发生在轮回末尾的特征事件ｊ来说，具有作为中一长期“前兆”的意义。有可能识别出的首次填隙式中强震（事件ｆ１），其强度通常比同一轮回开始的特征事件ｉ小１—２个震级单位，但明显大于当地背景地震活动强度，并发生在与事件ｉ和ｊ相同的空间位置。文中初步建立起中国大陆完整特征地震活动轮回的震级－时间序列模式（ｉ－ｆ－ｊ模式），它将作为本研究第二部分中进一步发展特征地震中－长期预测方法的基础。

Based on the analysis of the magnitude-time sequences of 24 characteristic earthquake cycles at different localities of China mainland it has shown that during individual complete cycles, few destructive earthquakes occurred within the earliest hi of a cycle duration between sequential characteristic events i and j, but in most cases, at least one rnoderatestrong earthquake occurred within the latest 1/2 of the cycle durations. Of which, at about 54% cases, the moderate-strong earthquakes occurred within the latest 1/3 of the cycle durations. These earthquakes have been termed as ' gap-filling moderate-strong earthquakes' (events f). The occurrence of an event f possibly marks that the rebuilt strain, since the last characteristic earthquake at a seismic source, is already up to the state not far from the ultimate strain in which the next characteristic event would recur at the same place. So, for the characteristic event j occurring at the ends of individual cycles, the events f are of the implication as mid-long-term 'precursor'. It is possible to recognize the first moderate-strong gap-filling earthquakes (event f1), usually it is less 1-2 magnitude units than the characteristic event i occurring at the beginning of the same cycle, but obviously larger than the size of the locally background seismicity, and has the same locality with the characteristic events i and j. A magnitude-time sequence model, called i-f-j model,for complete cycle of characteristic earthquake activity on China mainland has been preliminarily proposed in this paper. This model will be the basis on which the part Ⅱof this study will further develop the methodology for the mid-long-term forecasting of characteristic earthquakes.