Forecasting environmental water availability of lakes using temporal fusion transformer: case studies of China's two largest freshwater lakes.

Preserving lacustrine ecosystems is vital for sustainable watershed development, and forecasting the environmental water availability of lakes would support policymakers in developing sound management strategies. This study proposed a methodology that merges the lake water level prediction and environmental water availability evaluation. The temporal fusion transformer (TFT) model forecasted the lake water levels for the next 7 days by inputting the streamflow and lake water level data for the past 30 days. The environmental water availability was assessed by comparing the forecasted lake water levels with the environmental water requirements, resulting in adequate, regular, scarce, and severely scarce environmental water availability. The methodology was tested in two case studies: Poyang Lake and Dongting Lake, the two largest freshwater lakes in the Yangtze River Basin, China. The TFT model performed well in forecasting the lake water levels, as shown by the high coefficient of determination and finite root mean square error. The coefficients of determination exceeded 0.98 during the model training, validation, and test for both Poyang Lake and Dongting Lake, and the root mean square errors ranged from 0.06 to 0.46 m. The accurate prediction of lake water level promoted the precise forecasting of the environmental water availability with the high Kappa coefficient exceeding 0.90. Results indicated the rationality and effectiveness of integrating the lake water level prediction and environmental water availability evaluation. Future research includes the applicability of the TFT model to other lakes worldwide to test the proposed approach and investigate strategies to cope with environmental water scarcity.

Evolution of landscape pattern and the association with ecosystem services in the Ili-Balkhash Basin.

The Ili-Balkhash Basin in Central Asia is an arid endorheic basin shared by China and Kazakhstan. Population growth, socio-economic development, ecological conservation measures, and climate change have spurred land use and land cover changes and ecosystem services variations. This study used the long-term dataset from 1992 to 2018 to detect the landscape pattern evolution and its association with ecosystem services. The landscape pattern was quantified using landscape metrics, and the GeoDetector model quantified the driving factors of landscape pattern evolution. The ecosystem service value was assessed using the benefit transfer method. The time series trend was identified by the linear regression analysis combined with the Mann-Kendall algorithm, and Pearson's correlation coefficient was used to confirm the correlation. The temporal dynamics of the landscape pattern indicated the significant conversion of bare land to grassland. Cropland and urban land expanded significantly at the expense of forestland, grassland, and bare land. Various landscape elements tended to be more uniformly distributed across the basin with more regular shape and higher aggregation. The ecosystem service value increased significantly, and its correlation with the landscape pattern varied according to land use and land cover (LULC) types. The weakened shape complexity, the strengthened aggregation degree, and the more uniform distribution of different LULC types helped elevate total ecosystem service value. The results advanced the understanding of landscape pattern evolution and provided the scientific reference for land management regarding ecosystem services. Given the watershed ecosystem's integrity, transboundary cooperation between China and Kazakhstan was suggested to reinforce watershed sustainability through integrated watershed land resource planning and the joint adaptive strategies to climate change.

Assessing environmental water requirement for groundwater-dependent vegetation in arid inland basins by combining the copula joint distribution function and the dual objective optimization: An application to the Turpan Basin, China.

Preserving groundwater-dependent terrestrial ecosystems through environmental water allocation is critical for sustainable development in arid inland basins. Assessing the environmental water requirement is challenging due to the complex relationship between vegetation growth and groundwater depth. This study proposed a new assessment method by combining the copula joint distribution function and the dual objective optimization. The copula joint distribution function was used to describe the relationship between vegetation and groundwater depth instead of the traditional regression analysis. Given an ecological protection target, the conditional probability of achieving the target was estimated using the copula joint distribution. The groundwater depth interval with relatively high probability was suitable for vegetation growth and correspondingly conducive for ecological protection. In addition to ecological protection, the socio-economic water requirement was incorporated into the environmental water assessment, resulting in a dual optimization problem that could be resolved by the ideal point method. The optimization analysis revealed a groundwater depth with a high probability of successful ecological protection and low groundwater evapotranspiration to balance vegetation and human demands for groundwater. The proposed method of environmental water assessment by combing copula joint distribution function and dual objective optimization was applied in the Turpan Basin, an arid inland basin in Northwest China. The environmental groundwater depth ranged between 6 and 20 m, and the optimized interval was 7-8 m. The optimal environmental groundwater depth resulted in a probability of 0.46 to achieve the ecological protection target and annual evapotranspiration of 983 mm. The proposed method was practical and reliable and could be an effective tool for assessing the environmental water requirement of groundwater-dependent vegetation in arid inland basins.

A framework to assess the impact of ecological water conveyance on groundwater-dependent terrestrial ecosystems in arid inland river basins.

Overexploitation of water resources at middle reaches has threatened downstream oasis in arid inland river basins. To achieve sustainable development, ecological water conveyance is one effective measure to reallocate water resources between socio-economic and natural systems. A comprehensive impact assessment of ecological water conveyance on groundwater-dependent ecosystems that are common in downstream inland river basins is needed. The present study integrated the technologies of remote sensing analysis, trend detection, and numerical simulation into a technical framework that identifies the spatial response and temporal dynamics of vegetation to groundwater changes induced by ecological water conveyance. The Normalized Difference Vegetation Index (NDVI) was used as a proxy to estimate vegetation. The structure of the framework we used is clear and reasonable. We used remote-sensing data and ground truth information, and a decision tree that incorporates an iterative self-organizing data analysis technique. The decision tree classifies the land cover into affected and unaffected areas, and the results indicate the spatial range of impact. The Mann-Kendall algorithm and Sen's slope detect the tendency in NDVI series that indicates the temporal response of vegetation. The Verhulst logistic function, combined with environment carrying capacity function, constitutes a simplified vegetation dynamic model that can be used to predict potential impact through scenario analysis. An application in the Shiyang River basin in Northwest China evaluated the performance and usefulness of the framework; the accuracy of the results suggested that the framework is effective and practical. Additional case studies are required to assess the reliability and applicability of the framework and identify the factors that affect assessment results beyond our case study in the Shiyang River basin.