Boosik Kang

Conclusions and Future Work This study demonstrated the effectiveness of LSTM-based surrogate modeling—specifically the ConvLSTM—for predicting flood inflows at the Namgang multipurpose dam. Compared with the traditional HEC-HMS physical model, ConvLSTM showed significantly higher computational efficiency and prediction accuracy, particularly in peak flow magnitude and timing, while drastically reducing runtime (under 0.05 s per event) and memory use.

ANA anti-nuclear antibodies APC antigen-presenting cells IRF interferon regulatory factor 1 Introduction Evapotranspiration (ET), a critical component of the global and local water cycles, is strongly influenced by climatic factors such as precipitation, atmospheric temperature, and solar radiation. It plays a pivotal role in hydrological processes and water resource management, while also being a key indicator of climate change.

This is an early access version, the complete PDF, HTML, and XML versions will be available soon. Open AccessArticle by Jiyoung Sung SciProfiles Scilit Preprints.org Google Scholar and Boosik Kang SciProfiles Scilit Preprints.org Google Scholar * Department of Civil & Environmental Engineering, Dankook University, Gyeonggi-do 16890, Republic of Korea * Author to whom correspondence should be addressed.