Riverine floods are often caused by prolonged or heavy rainfall. These heavy or extreme rainfall events are mainly driven by various climate mechanisms, which have often been ignored in flood risk assessments. This study presents a climate mechanism-based flood frequency analysis approach, accommodating a direct linkage between risk metrics for decision making and the dominant causal climate mechanisms driving riverine flooding. The approach adopts several techniques including Markov chain analysis, K-nearest neighbor (KNN) resampling, and Z-score-based jittering method to mimic historical climate data with the heterogeneous consideration of the climate mechanisms. After that, the effects of climate change are incorporated through the manipulation of the transition matrix and the utilization of a quantile mapping approach. Future flood frequencies are quantified using the generated climate sequence and a rainfall-runoff model. Our proposed approach is described as a case study for the Nam River Basin, South Korea using two climate mechanisms, the Tropical Cyclone (TC) and non-TCs. Our results indicate that both mechanisms could potentially contribute to future flood extremes, especially when subjected to climate change. However, TCs have been proven to be the primary driving climate mechanism that shows more significant and immediate effects on the probable extremes in the study area. The approach presented in this study can provide new insights into future flood management.

河流洪水通常是由长时间或强降雨引起的。这些强降雨或极端降雨事件主要是由各种气候机制驱动的，而这些因素在洪水风险评估中往往被忽视。本研究提出了一种基于气候机制的洪水频率分析方法，适应决策风险指标与驱动河流洪水的主要因果气候机制之间的直接联系。该方法采用马尔可夫链分析、K-最近邻（KNN）重采样和基于Z-score的抖动方法等多种技术来模拟历史气候数据，同时考虑气候机制的异质性。之后，通过转换矩阵的操作和分位数映射方法的使用，将气候变化的影响纳入其中。使用生成的气候序列和降雨径流模型量化未来的洪水频率。我们提出的方法被描述为韩国南河流域的案例研究，使用两种气候机制，热带气旋 (TC) 和非 TC。我们的结果表明，这两种机制都可能导致未来的极端洪水，尤其是在受到气候变化的影响时。然而，TCs 已被证明是主要驱动气候机制，对研究区可能出现的极端事件显示出更显着和直接的影响。本研究中提出的方法可以为未来的洪水管理提供新的见解。韩国使用两种气候机制，热带气旋 (TC) 和非 TC。我们的结果表明，这两种机制都可能导致未来的极端洪水，尤其是在受到气候变化的影响时。然而，TCs 已被证明是主要的驱动气候机制，对研究区可能出现的极端事件显示出更显着和直接的影响。本研究中提出的方法可以为未来的洪水管理提供新的见解。韩国使用两种气候机制，热带气旋 (TC) 和非 TC。我们的结果表明，这两种机制都可能导致未来的极端洪水，尤其是在受到气候变化的影响时。然而，TCs 已被证明是主要的驱动气候机制，对研究区可能出现的极端事件显示出更显着和直接的影响。本研究中提出的方法可以为未来的洪水管理提供新的见解。