Reconstruction of Data Gaps in Total-Ozone Records with a New Wavelet Technique

This study introduces a new technique to fill and reconstruct daily observational of Total Ozone records containing void data for some days based on the wavelet theory as a linear time-frequency transformation, which has been considered in various fields of science, especially in the earth and space physics and observational data processing related to the Earth and space sciences. The initial corrupted records consist of six years of daily total Ozone measured by Dobson Photo-Spectrometer Instrument of Institute of Geophysics, University of Tehran. To verify the filled gaps resulted from this technique, the outputs of the proposed method are compared with the Total Ozone Mapping Spectrometer (TOMS) for the year ۲۰۰۵ and Ozone Monitoring Instrument (OMI) for the years ۲۰۰۶ – ۲۰۱۰ satellite data (hereafter used as TOMS/OMI data). The proposed technique consists of three steps: (۱) quality control and denoising; (۲) data-reconstruction based on Daubechies parent function (DB۱); and (۳) the combination of approximation and complementary coefficients using the Inverse Discrete Wavelet Transform (IDWT). Results show that this method was able to successfully reconstruct the missing data for gaps lasting no greater than ۱۸ days. For gaps beyond this ۱۸-day limit, however, this method was unable to reconstruct the voided data. As most instruments, including Dobson and Brewer Spectrometer, are working based on the optical interaction of stratospheric Ozone and sunshine, gaps in the Total-Ozone for more than ۱۸ days should happen in atmospheric systems with longevity over ۱۸ days in which overcast clouds persist longer than the ۱۸-day limit. The proposed method could be applied with high efficiency.

Keywords:

Total-Ozone , Discrete Wavelet Transform , Missing data , interpolation , Signal reconstruction