River ecological flow early warning forecasting using baseflow separation and machine learning in the Jiaojiang River Basin, Southeast China.

Ecological flow early warning is crucial for the rational management of watershed water resources. However, determining of accurate ecological flow threshold and choosing the appropriate forecasting model are challenging tasks. In this study, we initially developed a baseflow separation and Tennant method-based technique for calculating ecological river flow. Then an ecological flow early warning model was created using the machine learning technique based on distributed gradient enhancement framework (LightGBM). Finally, we utilized the framework of Shapley Additive Planning (SHAP) to explain how various hydrometeorological factors affect the variations in ecological flow conditions. The Jiaojiang River basin in southeast China is selected as the study area, and the hydrological stations in upstream of Baizhiao (BZA) and Shaduan (SD) are chosen for key analysis. The results of these applications show that the monthly baseflow frequency of the river ecological flow conditions of the two stations in the dry season is 20 % (7.49 m3/s) and 30 % (4.79 m3/s), respectively. The ecological flow level early warning forecasting accuracy is close to 90 % in the BZA and SD stations during dry and wet seasons. The variations of ecological flow are most affected by evaporation and base flow index. The results of this study can serve as a strong basis for the effective allocation and utilization of locally available water resources.

Modeling and Mapping High Water Table for a Coastal Region in Florida using Lidar DEM Data.

Predicting and mapping high water table elevation in coastal landscapes is critical for both science application projects like inundation risk analysis and engineering projects like pond design and maintenance. Previous studies of water table mapping focused on the application of geostatistical methods, which cannot predict values beyond an observation spatial domain or generate an ideal pattern for regions with sparse measurements. In this study, we evaluated the multiple linear regression (MLR) and support vector machine (SVM) techniques for high water table prediction and mapping using fine spatial resolution lidar-derived Digital Elevation Model (DEM) data, and designed an application protocol of these two techniques for high water table mapping in a coastal landscape where groundwater, tide, and surface water are related. Testing results showed that SVM largely improved the high water table prediction with a mean absolute error (MAE) of 1.22 feet and root mean square error (RMSE) of 2.22 feet compared to the application of the ordinary Kriging method which could not generate a reasonable water table. MLR was also promising with a MAE of around 2 feet and RMSE of around 3 feet. The study suggests that both MLR and SVM are valuable alternatives to estimate high water table elevation in Florida. Fine resolution lidar DEMs are beneficial for high water table prediction and mapping.

Assessment of weather-based influent scenarios for a WWTP: Application of a pattern recognition technique.

This study proposes an integrated approach by combining a pattern recognition technique and a process simulation model, to assess the impact of various climatic conditions on influent characteristics of the largest Italian wastewater treatment plant (WWTP) at Castiglione Torinese. Eight years (viz. 2009-2016) of historical influent data namely influent flow rate (Qin), chemical oxygen demand (COD), ammonium (N-NH4) and total suspended solids (TSS), in addition to two climatic attributes, average temperature and daily mean precipitation rates (PI) from the plant catchment area, are evaluated in this study. Following the outlier removal and missing-data imputation, five influent climate-based scenarios are identified by K-means clustering approach. Statistical characteristics of clustered observations are further investigated. Finally, to demonstrate that the proposed approach could improve the process control and efficiency, a process simulation model was developed and calibrated. Steady-state simulations were conducted, and the performance of the plant was studied under five influent scenarios. Further, an optimization scenario-based method was conducted to improve the energy consumption of the plant while meeting effluent requirements. The results indicate that with the adaptation of suitable aeration strategies for each of the influent scenarios, 10-40% energy saving can be achieved while meeting effluent requirements.