2026-10-13T16:37:19Z https://www.scidb.cn/oai

10.57760/sciencedb.44361 2026-07-31T16:02:56Z

2026-07-31 Daily model-reconstructed water-temperature profiles for Lake Qiandao, China, 1960–2025 doi:10.57760/sciencedb.44361 en This dataset contains 24,106 daily, model-reconstructed water-temperature profiles for Daba station in Lake Qiandao (Xin'anjiang Reservoir), eastern China, covering 1960-01-01 to 2025-12-30. The dataset is organized into 66 annual CSV files. Each daily profile contains temperatures at 300 model-resolved depths from 0.05 to 58.50 m below the water surface, with a median vertical spacing of approximately 0.20 m. Temperature values are reported in kelvin and can be converted to degrees Celsius by subtracting 273.15.The profiles were reconstructed using the multi-parameter scientific knowledge-guided neural network (MP-KgNN) developed for Lake Qiandao. The model is constrained by the one-dimensional heat-diffusion equation, was pre-trained with WRF-Lake simulations, and was fine-tuned against high-frequency water-temperature profiles collected at Daba station. The dataset provides the primary temperature-profile input used to examine long-term changes and event-scale vertical structure of compound surface-bottom heatwaves in Lake Qiandao. These data are model estimates rather than direct observations. No values were extrapolated below the deepest model-resolved layer at 58.50 m. water temperature; Lake Qiandao; heatwaves Miao Liu Zhigang Cao Shi Kun PUBLIC https://creativecommons.org/licenses/by/4.0/ dataset Science Data Bank