Attribution of hydrological droughts in large river-connected lakes: Insights from an explainable machine learning model.

Large lakes play an important role in water resource supply, regional climate regulation, and ecosystem support, but they face threats from frequent extreme drought events, necessitating an understanding of the mechanisms behind these events. In this study, we developed an explainable machine learning (ML) model that combines the Bayesian optimized (BO) long short-term memory (LSTM) model and the integrated gradients (IG) interpretation method to simulate and explain lake water level variations. In addition, the hydrological drought trends and extreme drought events in Poyang Lake from 1960 to 2022 were identified using the standardized water level index (SWI) and run theory.

Nonstationary frequency analysis and uncertainty quantification for extreme low lake levels in a large river-lake-catchment system.

Extreme hydrological events have become increasingly frequent on a global scale. The middle Yangtze River also faces a substantial challenge in dealing with extreme flooding and drought. However, the long-term characteristics of the extreme hydrological regime have not yet been adequately recognized.

Multiple latent variables but functionally dependent output mappings underlying the recognition of own- and other-race faces for Chinese individuals: Evidence from state-trace analysis.

To explore the number of latent variables underlying recognition of own- and other-race faces for Chinese observers, we conducted a study-recognition task where orientation, stimuli type, and duration were manipulated in the study phase and applied state trace analysis as a statistic method. Results showed that each state trace plot on each pair of stimuli types matched a single monotonic curve when stimuli type was set to state factor, but separate curves between face and non-face showed up when the state factor was orientation. The results implied that at least two latent variables affected recognition performance in the inversion paradigm.

[Zonation on non-point source pollution risk in the lower reaches of Zijiang River based on the "source-sink" theory].

Non-point source pollution risk assessment and zonation research are of great significance for the eco-environmental protection and optimization of land use structure. We identified the "source" and "sink" landscape using the "source-sink" landscape pattern theory based on the two phases of land use data in the lower reaches of Zijiang River in 2010 and 2018. We comprehensively considered the non-point source pollution occurrence and migration factors, and used location-weighted landscape contrast index (LCI) and non-point source pollution load index (NPPRI) to analyze non-point source pollution risk spatio-temporal characteristics in the study area.