Optimization And Application Of Algorithm Of The Terrain-following Hybrid Coordinate Based On Pressure

In recent years,with the continuous improvement of computer technology,the NWP(numerical weather prediction)based on high performance computer has developed rapidly.The vertical coordinate,which is the basis of vertical discretisation of NWP model,has an important influence on the performance of NWP model.Because of its unique advantages,the hybrid ?-p coordinate has been widely used in NWP model of many NWP centers around the world.At present,however,there is no universally-accepted algorithm for designing hybrid coordinate.Therefore,it is necessary to explore and study the design of hybrid coordinate.At present,there are two kinds of hybrid coordinate designing algorithms,i.e.Benard algorithm and Eckermann algorithm.In this paper,on the basis of the theoretical analysis of the hybrid coordinate and the implementation of the above two algorithms with Fortran language,we first proposed a method for getting the value of parameter a in the Benard2008 algorithm,and then we proposed a new composite designing algorithm by effectively combining the former two algorithms.Finally,based on the above three algorithms,we designed four kinds of hybrid coordinates.Based on the global spectral model YHGSM T511,a total of four sets of comparative experiments are carried out.The results show that:(1)compared with ? coordinate,the hybrid coordinate can significantly improve the prediction accuracy of the upper section of the model levels,with the amplitude increasing when the altitude increases;(2)the four kinds of hybrid coordinates can all accurately predict the abrupt adjustment of the stratospheric circulation situation and the distribution of the global temperature field;(3)keeping the total number of layers unchanged,a local increase of layer distribution near the tropopause can improve the prediction accuracy of upper section of the model,but this is at the expense of reducing the accuracy of the lower section,and the effect on the accuracy of different predictive variables at the tropopause is different;(4)compared with Benard algorithm,except for individual local areas,the composite algorithm can improve the accuracy of model prediction on the whole.