Application of the CNOP‐P Ensemble Prediction (CNOP‐PEP) Method in Evapotranspiration Forecasting Over the Tibetan Plateau to Model Parameter Uncertainties

Abstract

Evapotranspiration (ET) is recognized as an integral component of the regional and global energy and water cycles and is associated with the exchange of mass and energy between atmospheric and land systems (Jung et al., 2010;Trenberth et al., 2009). ET significantly impacts regional and global climate changes (Milly & Dunne, 2016;Scheff & Frierson, 2014). Hence, it is critical for evaluating ET variations to understand land-atmosphere interactions and global climate change. As the highest plateau in the world, the Tibetan Plateau (TP) observably regulates the hydrological cycle at the local and global scales, including the ET cycle, and further affects global and regional climate changes (Kang et al., 2010;Ma et al., 2017;Yao, 2019). Therefore, obtaining accurate simulations and predictions of ET over the TP can facilitate the addressing of the energy and water cycles on the plateau and analyses of regional and global changes (Li et al., 2012).

Verification of Long‐Term Ensemble Evapotranspiration Hindcast Using a Conditional Nonlinear Optimal Parameter Perturbation Ensemble Prediction Method on the Tibetan Plateau

“…In the study of Sun et al (2023), just one parameter perturbation was generated by the CNOP-P method to carry out ensemble forecast. In order to assess the differences between the above methods, we implemented ensemble forecasting with the CNOP-PEP method proposed by Sun et al (2023). Numerical results show that the ensemble forecasting skill is highest if all the parameter perturbation ensemble members are generated by the CNOP-PEP method.…”

“…From the above description, the CNOP-PEP method is different from that in the study of Sun et al (2023). Just one ensemble member is generated using the CNOP-P method in the study of Sun et al (2023). Their results showed that the CNOP-P-type parameters perturbations helped to improve the ensemble prediction skills of ET.…”

“…Finally, the q ensemble members corresponding to the m-dimensional parameter vector are generated by the CNOP-P method under a framework called the CNOP-P ensemble prediction (CNOP-PEP) method. From the above description, the CNOP-PEP method is different from that in the study of Sun et al (2023). Just one ensemble member is generated using the CNOP-P method in the study of Sun et al (2023).…”

“…The conditional nonlinear optimal perturbation related to parameter (CNOP‐P, Mu et al., 2010) method is a nonlinear optimization method that can calculate the fastest‐growing parameter perturbation member, which could the maximal uncertainties in numerical simulation and prediction. Although the CNOP‐P method has been applied to predictability studies of the atmosphere, ocean and land surface, there are few studies on its application to ensemble forecasting of land processes (Sun et al., 2023). The work of Sun et al (2023) suggested that the model perturbation of rapid growth properties could improve ensemble forecasting skill of ET.…”

“…Although the CNOP‐P method has been applied to predictability studies of the atmosphere, ocean and land surface, there are few studies on its application to ensemble forecasting of land processes (Sun et al., 2023). The work of Sun et al (2023) suggested that the model perturbation of rapid growth properties could improve ensemble forecasting skill of ET. Unfortunately, only one member was generated using the CNOP‐P method.…”