Machine learning has been used to mine the massive data collected by automatic environmental monitoring systems and predict the changes in the environmental factors in lakes. However, further study is needed to assess the feasibility of the development of a universal machine-learning-based turbidity model for a large shallow lake with considerable spatial heterogeneity in environmental factors. In this study, we collected and examined sediment and water quality data from Lake Taihu, China. Three monitoring stations were established in three lake zones to obtain continuous time series data of the water quality and meteorological variables. We used these data to develop three turbidity models based on long short-term memory (LSTM). The three zones differed in terms of environmental factors related to turbidity: in West Taihu, the Lake Center, and the mouth of Gonghu Bay, the critical shear stress of bed sediments was 0.029, 0.055, and 0.032 N m⁻², and the chlorophyll-a concentration was 23.27, 14.62, 30.80 μg L⁻¹, respectively. The LSTM-based turbidity model developed for any zone could predict the turbidity in the other two zones. For the model developed for West Taihu, its performance to predict the turbidity in the local zone (i.e., West Taihu) was inferior to that for the other zones; the reverse applied to the models developed for the Lake Center and Gonghu Bay. This can be attributed to the complex hydrodynamics in West Taihu, which weakens the learning of LSTM from the time series data. This study explores the feasibility of the development of a universal LSTM-based turbidity model for Lake Taihu and promotes the application of machine learning algorithms to large shallow lakes. Full article

Inland river runoff variations in arid regions play a decisive role in maintaining regional ecological stability. Observation data of inland river runoff in arid regions have short time series and imperfect attributes due to limitations in the terrain environment and other factors. These shortages not only restrict the accurate simulation of inland river runoff in arid regions significantly, but also influence scientific evaluation and management of the water resources of a basin in arid regions. In recent years, research and applications of machine learning and in-depth learning technologies in the hydrological field have been developing gradually around the world. However, the simulation accuracy is low, and it often has over-fitting phenomenon in previous studies due to influences of complicated characteristics such as “unsteady runoff”. Fortunately, the circulation layer of Long-Short Term Memory (LSTM) can explore time series information of runoffs deeply to avoid long-term dependence problems. In this study, the LSTM algorithm was introduced and improved based on the in-depth learning theory of artificial intelligence and relevant meteorological factors that were monitored by coupling runoffs. The runoff data of the Yarkant River was chosen for training and test of the LSTM model. The results demonstrated that Mean Absolute Error (MAE) and Root Mean Square error (RMSE) of the LSTM model were 3.633 and 7.337, respectively. This indicates that the prediction effect and accuracy of the LSTM model were significantly better than those of the convolution neural network (CNN), Decision Tree Regressor (DTR) and Random Forest (RF). Comparison of accuracy of different models made the research reliable. Hence, time series data was converted into a problem of supervised learning through LSTM in the present study. The improved LSTM model solved prediction difficulties in runoff data to some extent and it applied to hydrological simulation in arid regions under several climate scenarios. It not only decreased runoff prediction uncertainty brought by heterogeneity of climate models and increased inland river runoff prediction accuracy in arid regions, but also provided references to basin water resource management in arid regions. In particular, the LSTM model provides an effective solution to runoff simulation in regions with limited data. Full article

Regression-based frameworks for streamflow regionalization are built around catchment attributes that traditionally originate from catchment hydrology, flood frequency analysis and their interplay. In this work, we deviated from this traditional path by formulating and extensively investigating the first regression-based streamflow regionalization frameworks that largely emerge from general-purpose time series features for data science and, more precisely, from a large variety of such features. We focused on 28 features that included (partial) autocorrelation, entropy, temporal variation, seasonality, trend, lumpiness, stability, nonlinearity, linearity, spikiness, curvature and others. We estimated these features for daily temperature, precipitation and streamflow time series from 511 catchments and then merged them within regionalization contexts with traditional topographic, land cover, soil and geologic attributes. Precipitation and temperature features (e.g., the spectral entropy, seasonality strength and lag-1 autocorrelation of the precipitation time series, and the stability and trend strength of the temperature time series) were found to be useful predictors of many streamflow features. The same applies to traditional attributes such as the catchment mean elevation. Relationships between predictor and dependent variables were also revealed, while the spectral entropy, the seasonality strength and several autocorrelation features of the streamflow time series were found to be more regionalizable than others. Full article

Predictive uncertainty in hydrological modelling is quantified by using post-processing or Bayesian-based methods. The former methods are not straightforward and the latter ones are not distribution-free (i.e., assumptions on the probability distribution of the hydrological model’s output are necessary). To alleviate possible limitations related to these specific attributes, in this work we propose the calibration of the hydrological model by using the quantile loss function. By following this methodological approach, one can directly simulate pre-specified quantiles of the predictive distribution of streamflow. As a proof of concept, we apply our method in the frameworks of three hydrological models to 511 river basins in the contiguous US. We illustrate the predictive quantiles and show how an honest assessment of the predictive performance of the hydrological models can be made by using proper scoring rules. We believe that our method can help towards advancing the field of hydrological uncertainty. Full article

Accurate and reliable predictors selection and model construction are the key to medium and long-term runoff forecast. In this study, 130 climate indexes are utilized as the primary forecast factors. Partial Mutual Information (PMI), Recursive Feature Elimination (RFE) and Classification and Regression Tree (CART) are respectively employed as the typical algorithms of Filter, Wrapper and Embedded based on Feature Selection (FS) to obtain three final forecast schemes. Random Forest (RF) and Extreme Gradient Boosting (XGB) are respectively constructed as the representative models of Bagging and Boosting based on Ensemble Learning (EL) to realize the forecast of the three types of forecast lead time which contains monthly, seasonal and annual runoff sequences of the Three Gorges Reservoir in the Yangtze River Basin. This study aims to summarize and compare the applicability and accuracy of different FS methods and EL models in medium and long-term runoff forecast. The results show the following: (1) RFE method shows the best forecast performance in all different models and different forecast lead time. (2) RF and XGB models are suitable for medium and long-term runoff forecast but XGB presents the better forecast skills both in calibration and validation. (3) With the increase of the runoff magnitudes, the accuracy and reliability of forecast are improved. However, it is still difficult to establish accurate and reliable forecasts only large-scale climate indexes used. We conclude that the theoretical framework based on Machine Learning could be useful to water managers who focus on medium and long-term runoff forecast. Full article