Water, energy, and food are essential resources for humanity. The growing shortages of these resources and serious deterioration of river environments are having a big impact on the sustainable development of the economy and society in China. Water, energy, and food support human life and yet coexist in different ways, and therefore it is critical to find a way for all three key elements to be secured in order to support high standards of sustainable development in China. We used the criteria of stability, coordination, and sustainability of symbiotic systems to select 33 indexes that were then used to establish an index system. The weight of index was determined by using the entropy weight method combined with Analytic Hierarchy Process. The fuzzy comprehensive evaluation method was used to calculate the collaborative security index, which was the basis of our evaluation of the collaborative water–energy–food security of China in time and space. The results show that North China and Northwest China are at high water–food–energy security risk, while East, Central, and South China are at moderate risk. With the exception of Southwest China and South China, risk in most parts of the country has risen over the past decades, while it has fallen in Shandong, Henan, Sichuan, and Yunnan provinces.

In addition to agriculture, the water–land nexus (WLN) also profoundly affects the sustainable development of industry and residents’ lives. However, little research has been designed to assess the water–land nexus from the perspective of industry development and people’s quality of life. In the current paper, Wi, a regional industrial water–land nexus matching index, and Wd, a matching index of the domestic water–land nexus, were proposed for evaluating the water–land nexus from the industry development and quality of life perspectives separately in the current paper. Furthermore, climate change and human activities have significant impacts on the water–land nexus. The WLNs were assessed spatially and temporally for the first time based on these two indexes in 128 municipalities in the Huang-Huai-Hai River Basin of China from 1951 to 2017 to analyze the impacts of the changing environment on them. The impact of changing environment was explored based on changes of some climate factors and land use. The value of Wi are higher in the eastern and southern cities than the western and northern cities, while Zhenjiang city in Jiangsu Province has the highest Wi. For Wd, there are two low Wd zones across the basin, while the minimum values occurred in Linxia Hui Autonomous Region (Wd = 35.34 mm). Wi and Wd in most cities in the basin showed a significant downward trend, and some cities in the southwest of the basin have the fastest-decreasing of Wd. Wt and Wa were also calculated to assess the total and agricultural water–land nexus separately based on existing research. The Wt for the Huang-Huai-Hai River Basin gradually increases from northwest to southeast, and its spatial distribution characteristics are similar to precipitation in the river basin. In addition, the government should simultaneously implement water transfer plans to reduce the agricultural water pressure in Ningxia and Gansu provinces. Dynamic driving factors of change of the four assessment indexes (Wt, Wa, Wi, Wd) are briefly analyzed in the end of the paper.

Short-term water demand forecasting is crucial for constructing intelligent water supply system. There are plenty of useful models built to address this issue. However, there are still many challenging problems, including that the accuracies of the models are not high enough, the complexity of the models makes them hard for wide use in reality and the capabilities of models to catch peaks still have much room for improvement. In order to solve these problems, we proposed an ensemble deep learning model named STL-Ada-LSTM for daily water demand forecast by combining STL method with AdaBoost-LSTM model. After data preprocessing, the smoothed series is decomposed by STL to gain three input series. Then, several LSTM models are integrated by the AdaBoost algorithm to construct the ensemble deep learning model for water demand forecast. At last, the superiority of the proposed model is demonstrated by comparing with other state-of-art models. The proposed method is applied for water demand forecast using daily datasets from two representative water plants in Yiwu, East China. All models are assessed by mean absolute scaled error (MAE), mean absolute percentage error (MAPE), mean square error (MSE), root mean square error (RMSE), coefficient of determination (R2) and Akaike information criterion (AIC). The results show that the proposed model not only enhances the accuracy of the forecast, but also improves the stability and conciseness, which make it a practical daily water demand forecast model.