Journal of Earth System Science

, Volume 123, Issue 8, pp 1749–1758

A modified sliding spectral method and its application to COSMIC radio occultation data


DOI: 10.1007/s12040-014-0507-z

Cite this article as:
Xu, XS., Guo, P. & Hong, ZJ. J Earth Syst Sci (2014) 123: 1749. doi:10.1007/s12040-014-0507-z


In the moist lower troposphere, a limitation of the sliding spectral (SS) method is the restriction of the resolution of bending angle profiles because of the atmospheric multipath effect and noise. A modified sliding spectral (MSS) method is proposed in this paper to improve the inversion resolution of SS method in the moist lower troposphere. Simulation results show that the noise in the signal may cause inversion error in the classical SS method. The MSS method can decrease the influence of the noise to some extent. The SS and MSS methods were used to process COSMIC (Constellation Observing System for Meteorology, Ionosphere, and Climate) atmPhs profiles from DOY (day of year) 71–DOY 100 in 2007. The retrieved refractivity profiles were compared with those from the corresponding ECMWF (European Centre for Medium-Range Weather Forecasts) analysis. The results show that the SS method contains systematic positive biases in the 3–10 km height range and systematic negative biases below 3 km. The MSS method, in comparison to SS method, has decreased the maximum positive bias in the range of 3–10 km height from 0.37% to 0.23% in the northern hemisphere, from 1.3% to 0.25% in the tropics, and from 0.60% to 0.35% in the southern hemisphere. The biases of the MSS method are comparable to those announced for the COSMIC atmPrf profile; the latter is inverted by full spectrum inversion (FSI) method.


COSMIC occultation multipath inversion MSS method 

Copyright information

© Indian Academy of Sciences 2014

Authors and Affiliations

  1. 1.Ningbo Institute of TechnologyZhejiang UniversityZhejiangChina
  2. 2.Shanghai Astronomical Observatory, NAOCASShanghaiPR China
  3. 3.School of Mathematics & Information ScienceWenzhou UniversityZhejiangChina

Personalised recommendations