Global losses caused by floods are gradually increasing under the influence of climate change and human activities. The Yangtze River (YR) economic belt has continued to experience frequent flood disasters over the years; therefore, clarifying the significant flood risk factors is highly relevant for the development of the region. In this study, we investigated the flood risk factors in the Yangtze River Basin (YRB) during 2002–2018 based on meteorological data, reconstructed terrestrial water storage (TWS) data from Gravity Recovery Climate Experiment (GRACE) (GRACE-TWS), and Landsat data. The major conclusions were as follows: (1) the principal components (i.e., the first, third, and fourth principal components (PC1, PC3, and PC4)) of Standardized Precipitation Evapotranspiration Index (SPEI) at a 12-month scale (SPEI12) were more significantly (P < 0.05) correlated with teleconnection indices (i.e., Atlantic Multi-decadal Oscillation (AMO), Pacific Decadal Oscillation (PDO), Southern Oscillation Indices (SOI), and Multivariate El Niño-Southern Oscillation (ENSO) Index (MEI)), than the SPEI at 6-month scale (SPEI6) in the YRB during 2002–2018; (2) the changes in water bodies reflected by Landsat data (i.e., Landsat 5, Landsat 7, and Landsat 8 OLI) for the YRB confirmed the wet periods (i.e., 2002–2005, 2010, 2012–2013, and 2015–2018) monitored by SPEI (i.e., SPEI6 and SPEI12); and (3) while the Pearson correlation coefficients indicated a significant linear relationship between the major hydrological factors (e.g., precipitation, runoff, GRACE-TWS, flood potential index (FPI), and soil moisture (SM)) in the YRB, the precipitation detected by the GRACE-TWS and SM with one-month lag phase, had the maximum correlation coefficients of 0.83 and 0.85, respectively. Significant relationships (r² = 0.48 and r² = 0.92, P < 0.05) were found between variable infiltration capacity (VIC)-runoff and predicted runoffs based on these factors (i.e., precipitation, GRACE-TWS, flood potential index (FPI), and SM) and the random forest regression. This study provides significant hydrological information and contributes to approaches aimed at supporting climate resilience and investments in the YRB.

在气候变化和人类活动的影响下，洪水造成的全球损失正在逐渐增加。多年来，长江经济带洪涝灾害频发；因此，澄清重要的洪水风险因素与该地区的发展高度相关。在本研究中，我们基于气象数据、重力恢复气候实验 (GRACE) (GRACE-TWS) 重建的陆地蓄水 (TWS) 数据，调查了 2002-2018 年长江流域 (YRB) 的洪水风险因素，以及陆地卫星数据。主要结论如下：（1）12个月尺度（SPEI12）标准化降水蒸散指数（SPEI）的主成分（即第一、第三和第四主成分（PC1、PC3和PC4）） ) 更显着 (磷 < 0.05) 与遥相关指数（即大西洋多年代际振荡 (AMO)、太平洋年代际振荡 (PDO)、南方涛动指数 (SOI) 和多元厄尔尼诺-南方涛动 (ENSO) 指数 (MEI)）相关，而不是2002-2018 年 YRB 6 个月规模的 SPEI（SPEI6）；(2) Landsat 数据（即 Landsat 5、Landsat 7 和 Landsat 8 OLI）反映的 YRB 水体变化证实了湿润期（即 2002-2005、2010、2012-2013 和 2015-2018） ) 由 SPEI（即 SPEI6 和 SPEI12）监控；(3) 虽然 Pearson 相关系数表明长江流域主要水文因素（如降水、径流、GRACE-TWS、洪水潜力指数（FPI）和土壤水分（SM））之间存在显着的线性关系，但由 GRACE-TWS 和 SM 检测到的滞后期为 1 个月，的最大相关系数分别为 0.83 和 0.85。显着关系 (r²  = 0.48 和 r ²  = 0.92，P  < 0.05) 在可变渗透能力 (VIC)-径流和基于这些因素（即降水、GRACE-TWS、洪水潜力指数 (FPI) 和 SM）的预测径流之间发现和随机森林回归。这项研究提供了重要的水文信息，并有助于制定旨在支持 YRB 气候复原力和投资的方法。