CONSTRUCTION OF REGIONAL WEIGHTED MEAN TEMPERATURE MODEL BASED ON OPTIMIZATION BP NEURAL NETWORK

We developed a new global atmospheric weighted mean temperature (T_m) model considering time-varying lapse rate. Firstly, a global multidimensional T_m lapse rate model (NGGTm-H model) was developed using the sliding window algorithm. Secondly, the daily variation characteristics of T_m and its relationships with geographical situation were investigated. Finally, a hybrid-grid global T_m model considering time-varying lapse rate (NGGTm model) was developed.

In this study, we have developed a model (RF-PWV) to characterize precipitable water vapor (PWV) variation with altitude in the study area. RF-PWV can significantly reduce errors in vertical correction, enhance PWV fusion product accuracy, and provide insights into PWV vertical distribution, thereby contributing to climate research.

The existing zenith tropospheric delay (ZTD) models have limitations such as using a single fitting function, neglecting daily cycle variations, and relying on only one resolution grid data point for modeling. This model considers the daily cycle variation and latitude factor of ZTD, using the sliding window algorithm based on ERA5 atmospheric reanalysis data. The ZTD data from 545 radiosonde stations and MERRA-2 atmospheric reanalysis data are used to validate the accuracy of the GGZTD-P model.