Multi-source Data Fusion And Application Research Base On LAPS/STMAS

With our observing systems becoming maturing and observational data increasingly rich, how to effectively use these observations for public weather information services, drought monitoring, diagnostics and driving numerical models of weather disasters, has gradually become a hot research topic of meteorology. Using the LAPS (Local Analysis and Prediction System) and STMAS (Space-Time Multiscale Analysis System) developed by ESRL of NOAA (The National Oceanic and Atmospheric Administration), this study has researched integration analysis techniques of sounding, ground stations and satellites, radar and other unconventional observational data.In order to confirm the result of the ground temperature integration by LAPS and STMAS systems in different spatial range and different spatial resolution, in this study, using the LAPS and STMAS respectively, we have integretated automatic weather station observations and GFS (Global Forecast System) background information on the grid of1km,5km,10km,20km resolutions, in Beijing, North China, and the whole nation, then compared the integrated result of difference techniques of LAPS and STMAS from the point of view of the algorithm, and tested integretaed results by automatic weather station data independently. The results show that:the STMAS could resolve the fine characteristics of the observations, the interetaed results is much closer to the actual observation, however, LAPS is more smooth and loss of observational information. LAPS and STMAS both have good performance in the eastern area with dense observations. In data-sparse areas, STMAS has larger correction for background temperature than LAPS. The root mean square error of STMAS is much smaller than the LAPS in long time series statistics. The LAPS temperature is lower than STMAS affected by the background field.In order to study the high spatial and temporal resolution analysis field how to affect land surface processes, using1km×1km surface analysis field of STMAS output and the ground data of GFS forecast products we have respectively driven NoahMP,and simulated land surface process of the Guangxi for a period of one year, and compared soil moisture and soil temperature in different forcing data. The results showed that:high resolution forcing data could improve the simulation effect of soil temperature and revise the systematic bias, and has more delicate variation characteristics. In terms of soil moisture, it is more sensitive to precipitation, and the surface temperature, pressure, humidity and wind could also have effect on soil moisture by influencing the soil evaporation. The changing of the surface temperature, pressure, humidity and wind can help to correct estimation for the amount of soil moisture evaporation.Using LAPS (STMAS) to comprehensively analysis sounding, radar, satellite data and NCEP/GFS background field, we have choosen one case (7.21heavy rain in Beijing,2012) to compare the simulated ability of severe weather system of each module product. Base on the above analysis results and the NCEP/GFS forecasts field to initialize the WRF to study how the non-adiabatic LAPS (STMAS) initializing numerical model has effect on precipitation forecast. The results show that:through integrating radar radial wind data, the STMAS has a greater change in the center of the location and intensity of convergence in low-level divergence field. Convergence and divergence of wind field is more obvious, which is more related to the position and intensity of precipitation observed. The cloud analysis module could correct the cloud cover, cloud height, which can provide a more realistic cloudiness initial field for model. However, the correction for cloud height is discontinuous; STMAS non-adiabatic initialization for WRF simulation of precipitation have an great improvement in a few hours. Precipitation and its location is close to observation in the fist hour. WRF initializing by STMAS can simulate extinction and reproducibility process of thunderstorms in cloud system, which was reflected from the transformation and movement of rainband center.