The Experiment Of Construct For High Resolution Climatic Field Of Liaoning Area Using WRF Model With Automatic Weather Station Data Assimilation

Firstly, according to the characteristics of automatic station data, an automatic station data quality control module has been designed. Many quality control techniques have been integrated into the quality control module, such as the format logic check, the climate extremes record check, the time consistency check and the space consistency check. The AWS data quality control module has been applied into AWS data in 2012 of Liaoning Province. Then by use of numerical model WRF and its observation nudging, Liaoning Province 4km resolution temperatures and wind fields in January and July were established.3 different experiments were conducted:without assimilation, assimilate sparse field data and assimilate automatic station observation data. Three different experiments’simulation capabilities of temporal and spatial variability in monthly average temperature, monthly average wind speed, daily temperature and hourly temperature were analyzed. The results show that:(1) Quality control module can effectively identify problematic data, provides quality assurance for the application of assimilate automatic weather station data.(2) WRF model combined with assimilation automatic stations data can create high quality 4km resolution temperature field in Liaoning Province in January and July. With assimilation automatic stations data, the deviation of daily average temperature of January compare to observation is lower than±0.5℃, the absolute deviation of daily temperature is generally less than 0.6℃,the correlation coefficient is above 0.95; the absolute deviation of hourly temperature is generally less than 1℃ and correlation coefficient is greater than 0.92. The deviation of daily average temperature of July compare to observation is lower than±0.5℃ in most area of Liaoning Province, the absolute deviation of hourly temperature is generally less than 1℃ and correlation coefficient is greater than 0.91. In both January and July, the result of assimilate automatic station observation data is better than the result of without assimilation and result of assimilation sparse field data. In complex terrain, the improvement of assimilate automatic station observation data is dramatically, the absolute deviation of hourly and daily temperature can be reduce by 0.5℃.(3) WRF model combined with assimilation automatic stations data can create high quality 4km resolution wind field in Liaoning Province in January and July. With assimilation automatic stations data, the deviation of wind speed of January compare to observation is lower than±1m/s in most area of Liaoning Province. Only a few areas such as mountain in western part of Liaoning Province and western part of Liaoning Province has the deviation of wind speed greater than 1.5m/s. The deviation of wind speed of July compare to observation is lower than ±1m/s in most area of Liaoning Province. Only a few areas such as northern part of Liaoning Province has the deviation of wind speed greater than 1.5m/s. Among all three experiments, the result of assimilate automatic station observation data is better than the result of without assimilation and result of assimilation sparse field data. By compare the results of different result, the result of mountain area is poor, the correlation coefficient is only 0.55, the mean square error is over 2.6m/s; Simulation capabilities of plains and hills is much better, the correlation coefficient is around 0.8, the mean square error is less than 1.8m/s.