Runoff modeling is one of the key challenges in the field of hydrology. Various approaches exist, ranging from physically based over conceptual to fully data driven models. In this paper, we propose a data driven approach using the state-of-the-art Long-Short-Term-Memory (LSTM) network. The proposed model was applied in the Poyang Lake Basin (PYLB) and its performance was compared with an Artificial Neural Network (ANN) and the Soil & Water Assessment Tool (SWAT). We first tested the impacts of the number of previous time step (window size) in simulation accuracy. Results showed that a window in improper large size will dramatically deteriorate the model performance. In terms of PYLB, a window size of 15 days might be appropriate for both accuracy and computational efficiency. We then trained the model with 2 different input datasets, namely, dataset with precipitation only and dataset with all available meteorological variables. Results demonstrate that although LSTM with precipitation data as the only input can achieve desirable results (where the NSE ranged from 0.60 to 0.92 for the test period), the performance can be improved simply by feeding the model with more meteorological variables (where NSE ranged from 0.74 to 0.94 for the test period). Moreover, the comparison results with the ANN and the SWAT showed that the ANN can get comparable performance with the SWAT in most cases whereas the performance of LSTM is much better. The results of this study underline the potential of the LSTM for runoff modeling especially for areas where detailed topographical data are not available.

Littoral wetlands are globally important for sustainable development; however, they have recently been identified as critical hotspots of nitrous oxide (N₂O) emissions. N₂O flux from subtropical littoral wetlands remains unclear, especially under the current global warming environment. In the littoral zone of Lake Poyang, a simulated warming experiment was conducted to investigate N₂O flux. Open-top chambers were used to raise temperature, and the static chamber-gas chromatograph method was used to measure N₂O flux. Results showed that the littoral zone of Lake Poyang was an N₂O source, with an average flux rate of 8.9 μg N₂O m⁻² h⁻¹. Warming significantly increased N₂O emission (13.8 μg N₂O m⁻² h⁻¹ under warming treatment) by 54% compared to the control treatment. N₂O flux in the spring growing season was also significantly higher than that of the autumn growing season. In addition, temperature was not significantly related to N₂O flux, while soil moisture only explained about 7% of N₂O variation. These results imply that N₂O emission experiences positive feedback effect on the ongoing warming of the climate, and abiotic factors (e.g., soil temperature and soil moisture) were not main controls on N₂O variation in this littoral wetland.

The complex water regimes and fragile ecological systems in Dongting Lake and Poyang Lake, located in the middle reach of the Yangtze River, have been significantly affected by regional climate change and anthropogenic activities. The hydrological data from the outlets of Dongting Lake (Chenglingji station) during 1955–2016 and Poyang Lake (Hukou station) during 1953–2014 were divided into two periods: the pre-impact period and the post-impact period. Four statistical tests were used to identify the change years: 1979 at Chenglingji and 2003 at Hukou. The indicators of hydrologic alteration and range of variability approach were used to assess alterations in water level regimes. Results show that the severely altered indicators were January water level at both lake outlets, and 1-, 3-, 7- and 30-day minimum water level at Chenglingji, with the degree of hydrological alteration being larger than 85%. The overall degrees of hydrological alteration at Chenglingji and Hukou were 52.6% and 38.2%, respectively, indicating that water level regimes experienced moderate alteration and low alteration or that ecosystems were at moderate risk and low risk, respectively. Changes in water level regimes were jointly affected by climate change and anthropogenic activities. Water level regimes at Dongting Lake outlet were mainly affected by increased rainfall and dam regulation. Decreased rainfall, dam regulation, and sediment erosion and deposition were the main impact factors of water level regimes at Poyang Lake outlet. These changes in water level regimes have greatly influenced both aquatic and terrestrial ecosystems, especially for fish and vegetation communities. This study is beneficial for water resource management and ecosystems protection under regional changes.