The paper presents a 1-km annual Prolonged Artificial Nighttime-light Dataset of China (PANDA-China) from 1984 to 2020, generated using a Night-Time Light convolutional LSTM network. This dataset addresses the limitations of existing long-term nighttime light (NTL) data, which are either too short in duration or lack temporal consistency. The proposed method, NTLSTM, combines a spatiotemporal attention module and a convolutional LSTM unit to model the dynamic changes in NTL data. The dataset is validated through various assessment criteria, including RMSE, R², and linear slope, showing high accuracy and temporal consistency. PANDA-China is compared with other NTL datasets and socioeconomic variables (built-up areas, GDP, population), demonstrating superior performance in terms of temporal dynamics and spatial consistency. The dataset provides valuable insights into the spatiotemporal urbanization process in China, particularly before 1992 and after 2013, and offers opportunities for further research on economic and energy-related topics.The paper presents a 1-km annual Prolonged Artificial Nighttime-light Dataset of China (PANDA-China) from 1984 to 2020, generated using a Night-Time Light convolutional LSTM network. This dataset addresses the limitations of existing long-term nighttime light (NTL) data, which are either too short in duration or lack temporal consistency. The proposed method, NTLSTM, combines a spatiotemporal attention module and a convolutional LSTM unit to model the dynamic changes in NTL data. The dataset is validated through various assessment criteria, including RMSE, R², and linear slope, showing high accuracy and temporal consistency. PANDA-China is compared with other NTL datasets and socioeconomic variables (built-up areas, GDP, population), demonstrating superior performance in terms of temporal dynamics and spatial consistency. The dataset provides valuable insights into the spatiotemporal urbanization process in China, particularly before 1992 and after 2013, and offers opportunities for further research on economic and energy-related topics.