The Correction Of Umkehr Observations Above Kunming And The Comparison With The Data Of ECC And Brewer

In this paper, we corrected the observations of ozone profiles above Kunming area by improving the Umkehr algorithm of Dobson spectrophotometer, and proved the improvments by comparison with the data from electrochemical ozonesonde carried by ballons, then studied the accuracy of Dobson and Brewer spectrphotometer by compared the sounding data with the two kinds of ground observation data, and preliminarily researched the characteristics of ozone vertical distribution above the region.The improved algorithm to Kunming, to a certain extent, reduced the errors due to the multiple scattering, aerosol scattering and absorption and refraction, even the original system error caused by the algorithm itself. Comparative analysis for the ozone vertical profile before and after the improvement algorithm for inversion showed that after the correction of ozone vertical profile there was significant differences in the lower-middle area. Considering the theory of reverse and ozone sounding data the data after the correction is better than that of the original data; Comparied with the ozone sounding data and the two ground-based observations, the differences were significant in the troposphere and the lower-middle stratosphere, especially for the Brewer spectrophotometer observations, it may because of the automatically running of Brewer spectrophotometer influenced by the weather or other factors; we researched the characteristics of ozone vertical distribution above Kunming area with the results of Dobson spectrophotometer observations, and found the characteristics nearly the same in four seasons, shown as:ozone from the surface to the tropopause near slowly decreases with the increase of height, and it rose sharply in the lower stratosphere, and between 22 and 27 km after reaching the maximum then rapidly diminishing. In addition, the analysis of the distribution of the ozone profile changes day by day showed in the winter and spring the changes significantly stronger than the summer and fall. The interannual and interdecadal variability is still not clear at present due to the data length is not enough, we still need a long time to get enough data.