Due to the complex interaction of urban and mountainous floods, assessing flood susceptibility in mountainous urban areas presents a challenging task in environmental research and risk analysis. Data-driven machine learning methods can evaluate flood susceptibility in mountainous urban areas lacking essential hydrological data, utilizing remote sensing data and limited historical inundation records. In this study, two ensemble learning algorithms, Random Forest (RF) and XGBoost, were adopted to assess the flood susceptibility of Kunming, a typical mountainous urban area prone to severe flood disasters. A flood inventory was created using flood observations from 2018 to 2022. The spatial database included 10 explanatory factors, encompassing climatic, geomorphic, and anthropogenic factors. Artificial Neural Network (ANN) and Support Vector Machine (SVM) were selected for model comparison. To minimize the influence of expert opinions on model training, this study employed a strategy of uniformly random sampling in historically non-flooded areas for negative sample selection. The results demonstrated that (1) ensemble learning algorithms offer higher accuracy than other machine learning methods, with RF achieving the highest accuracy, evidenced by an area under the curve (AUC) of 0.87, followed by XGBoost at 0.84, surpassing both ANN (0.83) and SVM (0.82); (2) the interpretability of ensemble learning highlighted the differences in the potential distribution of the training data’s positive and negative samples. Feature importance in ensemble learning can be utilized to minimize human bias in the collection of flooded-site samples, more targeted flood susceptibility maps of the study area’s road network were obtained; and (3) ensemble learning algorithms exhibited greater stability and robustness in datasets with varied negative samples, as evidenced by their performance in F1-Score, Kappa, and AUC metrics. This paper further substantiates the superiority of ensemble learning in flood susceptibility assessment tasks from the perspectives of accuracy, interpretability, and robustness, enhances the understanding of the impact of negative samples on such assessments, and optimizes the specific process for urban flood susceptibility assessment using data-driven methods.

The issues with downscaling the outputs of a global climate model (GCM) to a regional scale that are appropriate to hydrological impact studies are investigated using the random forest (RF) model, which has been shown to be superior for large dataset analysis and variable importance evaluation. The RF is proposed for downscaling daily mean temperature in the Pearl River basin in southern China. Four downscaling models were developed and validated by using the observed temperature series from 61 national stations and large-scale predictor variables derived from the National Center for Environmental Prediction–National Center for Atmospheric Research reanalysis dataset. The proposed RF downscaling model was compared to multiple linear regression, artificial neural network, and support vector machine models. Principal component analysis (PCA) and partial correlation analysis (PAR) were used in the predictor selection for the other models for a comprehensive study. It was shown that the model efficiency of the RF model was higher than that of the other models according to five selected criteria. By evaluating the predictor importance, the RF could choose the best predictor combination without using PCA and PAR. The results indicate that the RF is a feasible tool for the statistical downscaling of temperature.

Flash flood, one of the most devastating weather-related hazards in the world, has become more and more frequent in past decades. For the purpose of flood mitigation, it is necessary to understand the distribution of flash flood risk. In this study, artificial intelligence (Least squares support vector machine: LSSVM) and classical canonical method (Logistic regression: LR) are used to assess the flash flood risk in the Yunnan Province based on historical flash flood records and 13 meteorological, topographical, hydrological and anthropological factors. Results indicate that: (1) the LSSVM with Radial basis function (RBF) Kernel works the best (Accuracy = 0.79) and the LR is the worst (Accuracy = 0.75) in testing; (2) flash flood risk distribution identified by the LSSVM in Yunnan province is near normal distribution; (3) the high-risk areas are mainly concentrated in the central and southeastern regions, where with a large curve number; and (4) the impact factors contributing the flash flood risk map from higher to low are: Curve number > Digital elevation > Slope > River density > Flash Flood preventions > Topographic Wetness Index > annual maximum 24 h precipitation > annual maximum 3 h precipitation.