Sufficient historical flood inventory data (FID) are crucial for accurately predicting flood susceptibility using supervised machine learning models. However, historical FID are insufficient in many regions. Remote sensing provides a promising opportunity to expand the FID. However, whether the FID expanded by remote sensing can improve the accuracy of flood susceptibility modeling needs further study. In this study, a framework was proposed for improving the accuracy of flood susceptibility prediction (FSP) by combining machine learning models and the expanded FID using Sentinel-1A radar images. Five widely used machine learning models were employed to verify the accuracy of the proposed method by taking Wuhan City as a case study, including the random forest (RF), gradient boosting decision tree (GBDT), k-nearest neighbor (KNN), support vector machine (SVM), and artificial neural network (ANN) models. Sentinel-1A images from time points before, during, and after flood events were used to expand the FID for training the machine learning models. The results showed that the performance of the machine learning models for predicting flood susceptibility was improved greatly by considering the expanded FID, being improved by approximately 1.14–19.74% based on the area under the receiver operating characteristic curve (AUC). Among the used machine learning models, taking into account all the statistical indicators, the ANN showed the best performance, while the SVM showed the best generalization performance in Wuhan City. According to the results of the ANN model, approximately 19% of the area in Wuhan City, mainly distributed near rivers and lakes, is at a high flood susceptibility level. This study provides an essential reference for flood susceptibility analyses in regions with limited flood sampling data.

River-lake water systems (RLS) are important carriers for matter transformation and energy transmission. Influenced by accelerated social and economic development, the structural, functional, and environmental states of RLS have been seriously damaged. It is an important problem for human beings to coordinate the contradiction between socio-economic development and the protection of RLS. In order to quantitatively study the harmonious relationship between socio-economic development and the state of RLS, the harmony theory method was used to analyze the degree of harmonious development between socio-economy and RLS in this study taking Xiangyang City as an example, and formulating corresponding harmonious optimization schemes. The results indicate that: (1) the state of RLS had a relatively small change during 2009–2014, and its spatial distribution shows a decreasing trend with the Han River as the central axis decreases on both of its sides; (2) before 2011, the driving force of socio-economic development in Xiang yang City mainly originated in the peripheral regions such as Laohekou City, Zaoyang City, and Gucheng County, but after 2011, it migrated rapidly towards Downtown, and reached the maximum in 2014; (3) when the influence of regional socio-economic development on RLS is small, socio-economic development is the main factor driving the change of the overall harmonious development degree of socio-economy and RLS. However when the influence is big, it is combined, driven by socio-economic development and the state of RLS; (4) the main factors affecting the overall harmonious degree of socio-economy and RLS in Xiangyang City include: river length, standard ratio of water quality, water consumption per capita, reservoir regulation capability, farmland irrigation water consumption per Mu (Mu is an area unit in China, 1 Mu approximately equals to 666.67 m²), and sewage treatment rate. This study can provide a reference for the future analysis of the harmonious development between socio-economy and RLS, as well as in formulating corresponding harmonious optimization schemes in China and other countries in a similar situation.

This Special Issue is the second edition following the publication of the first Issue, “Yellow River Basin Management under Pressure: Present State, Restoration and Protection”, in 2021

Green infrastructure (GI) has been implemented globally to mitigate the negative effects of urbanization. GI also regulates the urban runoff process and reduces non-point source pollution by intercepting initial runoff pollution and stormwater storage. In this paper, the impacts on GI were quantified and analyzed, considering eight designed storms with a 24 h duration and eight others with a 2 h duration with the combination of two characteristics (return period and peak time). The runoff process and reduction effect of pollutants were simulated for GI combinations (green roofs, vegetative swale, bio-retention units, and permeable pavement) using the Storm Water Management Model, taking the Dongshan campus of Shanxi University as an example case study. The results show that the GI combination can reduce runoff, suspended solids (SS), and chemical oxygen demand (COD). For short- and long-duration rainstorms, the average reduction rates of runoff, SS, and COD were 39.7%, 38.8%, and 39.6%, and 36.5%, 31.7%, and 32%, respectively, indicating its better effectiveness for short-duration storms. The GI’s effect was more sensitive during the short-duration storms owing to the greater absolute value of the 2 h elastic coefficients versus that of the 24 h, and the best reduction effect was observed with a rainfall peak coefficient of 0.1. These results provide a scientific reference for GI planning and implementation under a changing climate in the future.