Combinatorial Optimization of Physics Parameterization Schemes for Typhoon Simulation Based on a Simple Genetic Algorithm(SGA)OACSTPCD
Each physical process in a numerical weather prediction(NWP)system may have many different parameterization schemes.Early studies have shown that the performance of different physical parameterization schemes varies with the weather situation to be simulated.Thus,it is necessary to select a suitable combination of physical parameterization schemes according to the variation of weather systems.However,it is rather difficult to identify an optimal combination among millions of possible parameterization scheme combinations.This study applied a simple genetic algorithm(SGA)to optimizing the combination of parameterization schemes in NWP models for typhoon forecasting.The feasibility of SGA was verified with the simulation of Typhoon Mujigae(2015)by using the Weather Research and Forecasting(WRF)model and Typhoon Higos(2020)by using the Coupled Ocean–Atmosphere–Wave–Sediment Transport(COAWST)modeling system.The results show that SGA can efficiently obtain the optimal combination of schemes.For Typhoon Mujigae(2015),the optimal combination can be found from the 1,304,576 possible combinations by running only 488 trials.Similar results can be obtained for Typhoon Higos(2020).Compared to the default combination proposed by the COAWST model system,the optimal combination scheme significantly improves the simulation of typhoon track and intensity.This study provides a feasible way to search for the optimal combinations of physical parameterization schemes in WRF and COAWST for more accurate typhoon simulation.This can help provide references for future development of NWP models,and for analyzing the coordination and adaptability of different physical process parameterization schemes under specific weather backgrounds.
Zebin LU;Jianjun XU;Zhiqiang CHEN;Jinyi YANG;Jeremy Cheuk-Hin LEUNG;Daosheng XU;Banglin ZHANG;
China Meteorological Administration–Guangdong Ocean University Joint Laboratory for Marine Meteorology,South China Sea Institute of Marine Meteorology,Guangdong Ocean University,Zhanjiang 524088 College of Ocean and Meteorology,Guangdong Ocean University,Zhanjiang 524088 Guangzhou Institute of Tropical and Marine Meteorology/Guangdong Provincial Key Laboratory of Regional Numerical Weather Prediction,China Meteorological Administration,Guangzhou 510640China Meteorological Administration–Guangdong Ocean University Joint Laboratory for Marine Meteorology,South China Sea Institute of Marine Meteorology,Guangdong Ocean University,Zhanjiang 524088 Shenzhen Institute of Guangdong Ocean University,Shenzhen 518120China Meteorological Administration–Guangdong Ocean University Joint Laboratory for Marine Meteorology,South China Sea Institute of Marine Meteorology,Guangdong Ocean University,Zhanjiang 524088 College of Ocean and Meteorology,Guangdong Ocean University,Zhanjiang 524088Guangzhou Institute of Tropical and Marine Meteorology/Guangdong Provincial Key Laboratory of Regional Numerical Weather Prediction,China Meteorological Administration,Guangzhou 510640Guangzhou Institute of Tropical and Marine Meteorology/Guangdong Provincial Key Laboratory of Regional Numerical Weather Prediction,China Meteorological Administration,Guangzhou 510640 College of Atmospheric Science,Lanzhou University,Lanzhou 730000
计算机与自动化
simple genetic algorithm(SGA)combinatorial optimizationtyphoon forecastnumerical weather prediction(NWP)
《Journal of Meteorological Research》 2024 (001)
P.10-26 / 17
Supported by the National Natural Science Foundation of China(42130605);Shenzhen Science and Technology Program(JCYJ20210324131810029);Guangdong Province Introduction of Innovative Research and Development Team Project China(2019ZT08G669)。
评论