Exploring a multi-objective optimization operation model of water projects for boosting synergies and water quality improvement in big river systems.

Due to excessive nutrient enrichment and rapidly increasing water demand, the occurrence of riverine environment deterioration events such as algal blooms in rivers of China has become more frequent and severe since the 1990s, which has imposed harmful consequences on riverine ecosystems. However, tackling river algal blooms as an important issue of restoring riverine environment is very challenging because the complex interaction mechanisms between the causes are impacted by multiple factors. The contributions of our study consist of: (1) optimizing joint operation of water projects for boosting synergies of water quality and quantity, and hydroelectricity; and (2) preventing algal bloom from perspectives of hydrological and water-quality conditions by regulating water releases of water projects. This study proposed a multi-objective optimization methodology grounded on the Non-dominated Sorting Genetic Algorithm to simultaneously minimize the excess values of algal bloom indicators (water quality, O1), minimize the used reservoir capacity for water supply (water quantity, O2), and maximize the hydropower generation (hydroelectricity, O3). The proposed methodology was applied to several catastrophic algal bloom events that took place between 2017 and 2021 and thirteen water projects in the Hanjiang River of China. The results indicated that the proposed methodology largely stimulated the synergistic benefits of the three objectives by reaching a 36.7% reduction in total nitrogen and phosphorus concentrations, a 33.1% improvement in the remaining reservoir capacity, and a 41.0% improvement in hydropower output, as compared with those of the standard operation policy (SOP). In addition, the optimal water release schemes of water projects would increase the minimum streamflow velocity of downstream algal bloom control stations by 8.6%-9.4%. This study provides a new perspective on water project operation in the environmental improvement in big river systems while boosting multi-objectives synergies to support environmentalists and decision-makers with scientific guidance on sustainable water resources management.

Exploring a multi-output temporal convolutional network driven encoder-decoder framework for ammonia nitrogen forecasting.

Artificial neural networks exhibit significant advantages in terms of learning capability and generalizability, and have been increasingly applied in water quality prediction. Through learning a compressed representation of the input data, the Encoder-Decoder (ED) structure not only could remove noise and redundancies, but also could efficiently capture the complex nonlinear relationships of meteorological and water quality factors. The novelty of this study lies in proposing a multi-output Temporal Convolutional Network based ED model (TCN-ED) to make ammonia nitrogen forecasts for the first time. The contribution of our study is indebted to systematically assessing the significance of combining the ED structure with advanced neural networks for making accurate and reliable water quality forecasts. The water quality gauge station located at Haihong village of an island in Shanghai City of China constituted the case study. The model input contained one hourly water quality factor and hourly meteorological factors of 32 observed stations, where each factor was traced back to the previous 24 h and each meteorological factor of 32 gauge stations was aggregated into one areal average factor. A total of 13,128 hourly water quality and meteorological data were divided into two datasets corresponding to model training and testing stages. The Long Short-Term Memory based ED (LSTM-ED), LSTM and TCN models were constructed for comparison purposes. The results demonstrated that the developed TCN-ED model can succeed in mimicking the complex dependence between ammonia nitrogen and water quality and meteorological factors, and provide more accurate ammonia nitrogen forecasts (1- up to 6-h-ahead) than the LSTM-ED, LSTM and TCN models. The TCN-ED model, in general, achieved higher accuracy, stability and reliability compared with the other models. Consequently, the improvement can facilitate river water quality forecasting and early warning, as well as benefit water pollution prevention in the interest of river environmental restoration and sustainability.

Exploring a similarity search-based data-driven framework for multi-step-ahead flood forecasting.

Due to a small proportion of observations, reliable and accurate flood forecasts for large floods present a fundamental challenge to artificial neural network models, especially when the forecast horizons exceed the flood concentration time of a river basin. This study proposed for the first time a Similarity search-based data-driven framework, and takes the advanced Temporal Convolutional Network based Encoder-Decoder model (S-TCNED) as an example for multi-step-ahead flood forecasting. A total of 5232 hourly hydrological data were divided into two datasets for model training and testing. The input sequence of the model included hourly flood flows of a hydrological station and rainfall data (traced back to the previous 32 h) of 15 gauge stations, and the output sequence stepped into 1- up to 16-hour-ahead flood forecasts. A conventional TCNED model was also built for comparison purposes. The results demonstrated that both TCNED and S-TCNED could make suitable multi-step-ahead flood forecasts, while the proposed S-TCNED model not only could effectively mimic the long-term rainfall-runoff relationship but also could provide more reliable and accurate forecasts of large floods than the TCNED model even in extreme weather conditions. There is a significant positive correlation between the mean sample label density improvement and the mean Nash-Sutcliffe Efficiency (NSE) improvement of the S-TCNED over the TCNED at the long forecast horizons (13 h up to 16 h). Based on the analysis of the sample label density, it is found that the similarity search largely improves the model performance by enabling the S-TCNED model to learn the development process of similar historical floods in a targeted manner. We conclude that the proposed S-TCNED model that converts and associates the previous rainfall-runoff sequence with the forecasting runoff sequence under a similar scenario can enhance the reliability and accuracy of flood forecasts while extending the length of forecast horizons.

Deep learning-based neural networks for day-ahead power load probability density forecasting.

Energy efficiency is crucial to greenhouse gas (GHG) emission pathways reported by the Intergovernmental Panel on Climate Change. Electrical overload frequently occurs and causes unwanted outages in distribution networks, which reduces energy utilization efficiency and raises environmental risks endangering public safety. Electrical load, however, has a dynamically fluctuating behavior with notoriously nonlinear hourly, daily, and seasonal patterns. Accurate and reliable load forecasting plays an important role in scheduling power generation processes and preventing electrical systems from overloading; nevertheless, such forecasting is fundamentally challenging, especially under highly variable power load and climate conditions. This study proposed a deep learning-based monotone composite quantile regression neural network (D-MCQRNN) model to extract the multiple non-crossing and nonlinear quantile functions while conquering the drawbacks of error propagation and accumulation encountered in multi-step-ahead probability density forecasting. The constructed models were assessed by an hourly power load series collected at the electric grid center of Henan Province in China in two recent years, along with the corresponding meteorological data collected at 16 monitoring stations. The results demonstrated that the proposed D-MCQRNN model could significantly alleviate the time-lag and biased-prediction phenomena and noticeably improve the accuracy and reliability of multi-step-ahead probability density forecasts on power load. Consequently, the proposed model can significantly reduce the risk and impact of overload faults and effectively promote energy utilization efficiency, thereby mitigating GHG emissions and moving toward cleaner energy production.

Multivariate framework for the assessment of key forcing to Lake Malawi level variations in non-stationary frequency analysis.

Lake Malawi in south eastern Africa is a very important freshwater system for the socio-economic development of the riparian countries and communities. The lake has however experienced considerable recession in the levels in recent years. Consequently, frequency analyses of the lake levels premised on time-invariance (or stationarity) in the parameters of the underlying probability distribution functions (pdfs) can no longer be assumed. In this study, the role of hydroclimate forcing factors (rainfall, lake evaporation, and inflowing discharge) and low frequency climate variability indicators (e.g., El Nino Southern Oscillation-ENSO and the Indian Ocean Dipole Mode-IODM) on lake level variations is investigated using a monthly mean lake level dataset from 1899 to 2017. Non-stationarity in the lake levels was tested and confirmed using the Mann-Kendall trend test (α = 0.05 level) for the first moment and the F test for the second moment (α = 0.05 level). Change points in the series were identified using the Mann-Whitney-Pettit test. The study also compared stationary and non-stationary lake level frequency during 1961 to 2004, the common period where data were available for all the forcing factors considered. Annual maximum series (AMS) and peak over threshold (POT) analysis were conducted by fitting various candidate extreme value distributions (EVD) and parameter fitting methods. The Akaike information criteria (AIC), Bayesian information criteria (BIC), deviance information criteria (DIC), and likelihood ratios (RL) served as model evaluation criteria. Under stationary conditions, the AMS when fitted to the generalized extreme value (GEV) distribution with maximum likelihood estimation (MLE) was found to be superior to POT analysis. For the non-stationary models, open water evaporation as a covariate of the lake levels with the GEV and MLE was found to have the most influence on the lake level variations as compared with rainfall, discharge, and the low frequency climatic forcing. The results are very critical in flood zoning especially with various planned infrastructural developments around the lakeshore.

Exploring multidecadal changes in climate and reservoir storage for assessing nonstationarity in flood peaks and risks worldwide by an integrated frequency analysis approach.

The changing climate and reservoir storage have a far-reaching influence on the nonstationarity in flood peaks worldwide, but the quantification of the relative contribution of each covariate (i.e., climate and reservoir storage) is fundamentally challenging especially under the time-varying mechanisms in statistical properties. This study proposed an integrated flood frequency analysis for assessing the impacts of changing climate and reservoir storage on the nonstationarity in flood peaks and flood risks worldwide. The 32 major river catchments covering more than 60% of hydro-meteorological observation stations and 70% of reservoir storage worldwide constituted the case study. The proposed three-faceted approach was explored systematically through: modeling the nonstationarity in global flood peaks, identifying the contribution of changing climate and reservoir storage to the nonstationarity of flood peaks, and quantifying the change in flood risks under the nonstationary condition. The findings pointed out that global flood trends varied from increasing +19.3%/decade to decreasing -31.6%/decade. Taking the stationary flood frequency analysis as the benchmark, the comparative results revealed that the flood risk in 5 rivers under the nonstationary condition in response to warming climate significantly increased (1% â†’ 5%) over the historical period whereas the flood risk in 7 rivers in response to increasing reservoir storage largely reduced (1% â†’ 0.5%). Despite the spatiotemporal heterogeneity of observations, the changes in flood peaks evaluated here were explicitly in lined with the changing climate and reservoir storage, supporting the demand for considering the nonstationarity of flood peaks and risks in social infrastructure planning and designing as well as water management.

Explore a Multivariate Bayesian Uncertainty Processor driven by artificial neural networks for probabilistic PM2.5 forecasting.

Quantifying predictive uncertainty inherent in the nonlinear multivariate dependence structure of multi-step-ahead PM2.5 forecasts is challenging. This study integrates a Multivariate Bayesian Uncertainty Processor (MBUP) and an artificial neural network (ANN) to make accurate probabilistic PM2.5 forecasts. The contributions of the proposed approach are two-fold. First, the MBUP can capture the nonlinear multivariate dependence structure between observed and forecasted data. Second, the MBUP can alleviate predictive uncertainty encountered in PM2.5 forecast models that are configured by ANNs. The reliability of the proposed approach was assessed by a case study on air quality in Taipei City of Taiwan. We consider forecasts of PM2.5 concentrations as a function of meteorological and air quality factors based on long-term (2010-2018) hourly observational datasets. Firstly, the Back Propagation Neural Network (BPNN) and the Adaptive Neural Fuzzy Inference System (ANFIS) were investigated to produce deterministic forecasts. Results revealed that the ANFIS model could learn different air pollutant emission mechanisms (i.e. primary, secondary and natural processes) from the clustering-based fuzzy inference system and produce more accurate deterministic forecasts than the BPNN. The ANFIS model then provided inputs (i.e. point estimates) to probabilistic forecast models. Next, two post-processing techniques (MBUP and the Univariate Bayesian Uncertainty Processor (UBUP)) were separately employed to produce probabilistic forecasts. The Bayesian Uncertainty Processors (BUPs) can model the dependence structure (i.e. posterior density function) between observed and forecasted data using a prior density function and a likelihood density function. Here in BUPs, the Monte Carlo simulation was introduced to create a probabilistic predictive interval of PM2.5 concentrations. The results demonstrated that the MBUP not only outperformed the UBUP but also suitably characterized the complex nonlinear multivariate dependence structure between observations and forecasts. Consequently, the proposed approach could reduce predictive uncertainty while significantly improving model reliability and PM2.5 forecast accuracy for future horizons.

Multi-output support vector machine for regional multi-step-ahead PM2.5 forecasting.

Air quality deteriorates fast under urbanization in recent decades. Reliable and precise regional multi-step-ahead PM2.5 forecasts are crucial and beneficial for mitigating health risks. This work explores a novel framework (MM-SVM) that combines the Multi-output Support Vector Machine (M-SVM) and the Multi-Task Learning (MTL) algorithm for effectively increasing the accuracy of regional multi-step-ahead forecasts through tackling error accumulation and propagation that is commonly encountered in regional forecasting. The Single-output SVM (S-SVM) is implemented as a benchmark. Taipei City of Taiwan is our study area, where three types of air quality monitoring stations are selected to represent areas imposed with high traffic influences, high human activities and commercial trading influences, and less human interventions close to nature situation, respectively. We consider forecasts of PM2.5 concentrations as a function of meteorological and air quality factors based on long-term (2010-2016) observational datasets. Firstly, the Kendall tau coefficient is conducted to extract key spatiotemporal factors from regional meteorological and air quality inputs. Secondly, the M-SVM model is trained by the MTL to capture non-linear relationships and share correlation information across related tasks. Lastly, the MM-SVM model is validated using hourly time series of PM2.5 concentrations as well as meteorological and air quality datasets. Regarding the applicability of regional multi-step-ahead forecasts, the results demonstrate that the MM-SVM model is much more promising than the S-SVM model because only one forecast model (MM-SVM) is required, instead of constructing a site-specific S-SVM model for each station. Moreover, the forecasts of the MM-SVM are found better consistent with observations than those of any single S-SVM in both training and testing stages. Consequently, the results clearly demonstrate that the MM-SVM model could be recommended as a novel integrative technique for improving the spatiotemporal stability and accuracy of regional multi-step-ahead PM2.5 forecasts.

Exploring synergistic benefits of Water-Food-Energy Nexus through multi-objective reservoir optimization schemes.

This study proposed a holistic three-fold scheme that synergistically optimizes the benefits of the Water-Food-Energy (WFE) Nexus by integrating the short/long-term joint operation of a multi-objective reservoir with irrigation ponds in response to urbanization. The three-fold scheme was implemented step by step: (1) optimizing short-term (daily scale) reservoir operation for maximizing hydropower output and final reservoir storage during typhoon seasons; (2) simulating long-term (ten-day scale) water shortage rates in consideration of the availability of irrigation ponds for both agricultural and public sectors during non-typhoon seasons; and (3) promoting the synergistic benefits of the WFE Nexus in a year-round perspective by integrating the short-term optimization and long-term simulation of reservoir operations. The pivotal Shihmen Reservoir and 745 irrigation ponds located in Taoyuan City of Taiwan together with the surrounding urban areas formed the study case. The results indicated that the optimal short-term reservoir operation obtained from the non-dominated sorting genetic algorithm II (NSGA-II) could largely increase hydropower output but just slightly affected water supply. The simulation results of the reservoir coupled with irrigation ponds indicated that such joint operation could significantly reduce agricultural and public water shortage rates by 22.2% and 23.7% in average, respectively, as compared to those of reservoir operation excluding irrigation ponds. The results of year-round short/long-term joint operation showed that water shortage rates could be reduced by 10% at most, the food production rate could be increased by up to 47%, and the hydropower benefit could increase up to 9.33 million USD per year, respectively, in a wet year. Consequently, the proposed methodology could be a viable approach to promoting the synergistic benefits of the WFE Nexus, and the results provided unique insights for stakeholders and policymakers to pursue sustainable urban development plans.