Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Prediction of Qinghai Lake's level under future climate change: A hybrid modeling approach based on hydrological model and deep learning method
Study region: The Qinghai Lake Basin on the Northeast Tibetan Plateau. Study focus: Qinghai Lake has experienced significant water level fluctuations with potential ecological and climatic implications. This study proposes an innovative hybrid modeling framework combining deep learning and hydrological models to enhance the precision of meteorological data acquisition and the accuracy of runoff and lake level simulations and forecasts. Additionally, we apply SHAP explanation and water balance theory to explain how meteorological and hydrological features contribute to the water level fluctuations. New hydrological insights for the region: This study proposes an effective and convenient bias correction method for GCM and hybrid physical-neural models to accurately simulate historical runoff and lake level changes in the basin during 1980–2014. Based on the optimal GCM and future runoff, the lake level is predicted to rise by approximately 0.82–1.67 m between 2015 and 2050, primarily driven by substantial increases in runoff. Further, we conduct a comparative water balance analysis for two periods: declining period (1980–2004) and ascending period (2005–2050), and further divide these periods into freezing and thawing sub-periods. Comparing the ascending and decline period, we observe that runoff remains stable during the freezing period, while the annual runoff during the thawing period increased by 1.19–1.39 × 108 m3. These results suggest the impact and ongoing influence of permafrost degradation on runoff and thus water level.
Prediction of Qinghai Lake's level under future climate change: A hybrid modeling approach based on hydrological model and deep learning method
Study region: The Qinghai Lake Basin on the Northeast Tibetan Plateau. Study focus: Qinghai Lake has experienced significant water level fluctuations with potential ecological and climatic implications. This study proposes an innovative hybrid modeling framework combining deep learning and hydrological models to enhance the precision of meteorological data acquisition and the accuracy of runoff and lake level simulations and forecasts. Additionally, we apply SHAP explanation and water balance theory to explain how meteorological and hydrological features contribute to the water level fluctuations. New hydrological insights for the region: This study proposes an effective and convenient bias correction method for GCM and hybrid physical-neural models to accurately simulate historical runoff and lake level changes in the basin during 1980–2014. Based on the optimal GCM and future runoff, the lake level is predicted to rise by approximately 0.82–1.67 m between 2015 and 2050, primarily driven by substantial increases in runoff. Further, we conduct a comparative water balance analysis for two periods: declining period (1980–2004) and ascending period (2005–2050), and further divide these periods into freezing and thawing sub-periods. Comparing the ascending and decline period, we observe that runoff remains stable during the freezing period, while the annual runoff during the thawing period increased by 1.19–1.39 × 108 m3. These results suggest the impact and ongoing influence of permafrost degradation on runoff and thus water level.
Prediction of Qinghai Lake's level under future climate change: A hybrid modeling approach based on hydrological model and deep learning method
Kaixun Liu (Autor:in) / Na Li (Autor:in) / Sihai Liang (Autor:in)
2025
Aufsatz (Zeitschrift)
Elektronische Ressource
Unbekannt
Metadata by DOAJ is licensed under CC BY-SA 1.0
Analysis and Prediction of Poyang Lake’s Navigable Conditions under a New Hydrological Regime
| DOAJ | 2023
THE HISTORY IN HARRIETT LAKE'S CLOSET
| British Library Online Contents | 2009
Salt lake's 250-mile sewer program
| Engineering Index Backfile | 1956
Carbon Balance of Grasslands on the Qinghai-Tibet Plateau under Future Climate Change: A Review
| DOAJ | 2020