Improvement Research Of Stable Component Extraction Based On Historical Data And Its Application On Numerical Model

Under the background of global warming, the extreme weather and climate events occurred frequently. The current short-term weather forecasts can not meet the social demand,thus it is necessary to do some research about the 10-30 days extended forecast, which can reduce or eliminate major disastrous weather and climate anomaly to people’s production and life caused by the negative impact.Based on the numerical model, in present paper, the state variables are divided into two parts, incloulding the stable components which can be pridicted and the chaotic component which can’t be pridicted, and the former will be studied. Besides, based on the historical data,a further study of the method extending is proposed to extract the stable components which can display the information of 10-30 days. The stable components are used to explain the characteristics of atmospheric motion system, which avoidding the traditional research that the stable components are depending on the performance of the numerical model. Specific conclusions are as follows:1) Improving research of extended forecast period by using the historical data from stable component method, the stable component method is based on the extraction further consider the ocean before the external forcing for climate background selection. New method can effectively improve the proportion of abnormal stable components in the case study, and to research for circulation system in different climate mean state abnormal characteristics analysis, which is more prominent role of abnormal stable components in in the event of an exception.2) Through continuous hot weather on August 2013 in South China and the transition from the early May 2014 to the end of the cold weather of high temperature weather process in North China that two cases 10-30 day extension analysis of evolution characteristics of stable components, found the contributions in case of abnormal difference stable components corresponding to substrate contributions and climate state base is large, and the corresponding basal climatic stable components is little difference, which shows that abnormal stable components can supplement the information beyond the description of climate theinformation given in the characterization of persistent abnormal process, it will be more conducive to the research on anomaly circulation system is different from the average state of the climate, and climate just reflects the steady state component of the large scale circulation system in the high latitude area adjustment, further shows the improved method effect.3) by using the improved extraction method with stable component, based on different ENSO phase of the typical cases in, analysis of the extracted from the extension of the steady component of variance contribution rate(accounted for the proportion of total variance) in the seasonal scale changes over time. It is proved that the stable components of atmosphere of the overall trend has a very good memory. Stable component can be well characterized in 10-30 days scale weather systems of subject information and this part of the information in the following time period has a very good continuity, characterization of the atmospheric motion system of general trend.4) Therefore, based on the stable components of atmosphere of the overall trend has a very good memory, using 10-30 days extended period of stable components, since 2001,Eurasia mid latitude region of extreme low temperature events occur frequently, but SST background of different reason. In the La Nina SST anomalies in the state of the overall climate background makes the two Asia, mid latitude ridge one trough circulation situation of establishment and maintenance of guide cold air from the tank area south, and corresponds to the latitude of the upper troposphere to the Westerlies weakened and in the surface layer of Siberia high pressure in the presence of, resulting in a continued cold weather in winter. The El Nino type cold winter is also is in the mid latitudes of the northern hemisphere two ridge one trough circulation situation, but the anomalous circulation situation, just superimposed on climate background significantly prolonged relative departure of abnormal state that El Nino SST background is not caused by the direct cause of the cold winter.5) In view of the stable component of the atmospheric system, it is particularly studied,and the main characteristics of the stable component of the system can be improved. Theory based on extracted from historical data in 10-30 days stable components, and its application in the numerical model of the extended period prediction, and original mode of simulation results are compared, the for the level of model prediction ability to improve diagnosis, found in analog mode and 10 days after, improvement of prediction skill regardless of in the SST conditions under which are better than the original scheme, improved scheme of ACC in the extension of the period of the remained at about 0.2, and the original scheme of ACC is almost zero. This explains the improved test in extending the scale of the period to improve the forecast model has a certain degree of that with the extension of time prediction andmodel for the initial places to provide information has gradually dissipated, and the improved test exceptionally stable component of the characterization of extension information starting to have an effect, can more effectively improve mode in the following period predication skill.