In this study, we use version 2.0 of the digital inventory of landslides triggered by the MW 7.8 Kaikōura, New Zealand, earthquake of 14 November 2016 to revaluate some of the controls on their spatial distribution. Version 2.0 contains 29,557 hand-digitised landslides, whilst version 1.0 included only 10,195 mainly larger (> 10,000 m2) landslides. The main results from the re-analysis using version 2.0 of the inventory are the following: (1) the landslide frequency and density results indicate that the power laws fitted to version 1.0 and version 2.0 landslide inventories are similar; (2) the landslide densities decrease with increasing distance from the nearest fault that ruptured to the surface during the earthquake; (3) the larger landslides in the inventory occur on slopes with higher local slope relief; and (4) the results from logistic regression modelling using the version 2.0 inventory show that (a) the geological materials have a larger relative contribution to the fit of the overall model, when compared with the previous model based on the version 1.0 inventory, and (b) the “distance to fault” predictor variable continues to have more statistical power in predicting landslide probability than the modelled peak ground acceleration or peak ground velocity. These results suggest that future research on the Kaikōura earthquake landslide distribution—and the landslide distributions associated with other similar large earthquakes—might be directed towards investigating the interaction between surface fault rupture, fault damage-zone properties, earthquake-induced ground shaking, and the initiation of slope failures.

中文翻译：

---

由 MW7.8 引发的山体滑坡 2016 年 11 月 14 日凯库拉地震：更新

在本研究中，我们使用由 2016 年 11 月 14 日新西兰凯库拉 MW 7.8 地震引发的滑坡数字清单 2.0 版重新评估对其空间分布的一些控制。2.0 版包含 29,557 个手动数字化滑坡，而 1.0 版仅包含 10,195 个主要较大（> 10,000 平方米）的滑坡。使用 2.0 版清单重新分析的主要结果如下： (1) 滑坡频率和密度结果表明拟合 1.0 版和 2.0 版滑坡清单的幂律相似；(2) 滑坡密度随着距地震中最近破裂断层到地表的距离的增加而减小；(3)清单中较大的滑坡发生在局部坡度较高的斜坡上；(4) 使用 2.0 版清单的逻辑回归建模结果表明 (a) 与基于 1.0 版清单的先前模型相比，地质材料对整体模型的拟合具有更大的相对贡献，以及(b) “断层距离”预测变量在预测滑坡概率方面继续比模拟的峰值地面加速度或峰值地面速度具有更大的统计能力。这些结果表明，未来对凯库拉地震滑坡分布——以及与其他类似大地震相关的滑坡分布——的研究可能会转向研究地表断层破裂、断层损伤带特性、地震引起的地面震动和边坡失稳的开始。