Estimating the observation errors of FY-3C radio occultation dataset using the three-cornered hat method

Abstract

This study uses the three-cornered hat (3CH) method to estimate the observation error variances (ErrVars) of FY-3C RO refractivity, temperature, and specific humidity for the first time. The FY-3C RO data was compared to the three reference datasets including radiosonde observations and NCEP and ERA-Interim reanalyses. The ErrVars of FY-3C RO data are estimated at 18 globally distributed radiosonde stations by using the three reference datasets and are compared to corresponding gridded ErrVars estimated using only the two model datasets as references. The two types of estimates show good correlations at different heights, while the gridded estimates are generally the smaller ones, which may be attributed to the neglection of error correlations among the datasets when applying the 3CH method. Due to the lack of radiosonde data in oceanic and polar regions, the global distributions of FY-3C RO observation errors are presented based on the estimated 5° × 5° gridded ErrVars. The global distribution of the FY-3C RO fractional error standard deviations (ErrSDs) demonstrates that the observation error varies greatly at different pressure levels and latitudes. Specifically, the refractivity ErrSDs at 200 hPa and 50 hPa are significantly higher around 30°N and 30°S than in other areas. The specific humidity ErrSDs generally increase as pressure levels decrease. In addition, statistics show that the fractional ErrSDs of refractivity are generally the lowest between 45° N–75° N and 45° S–75° S at all pressure levels, and land-sea differences exist in the fractional ErrSDs for all three types of RO data.

An Appraisal of the Progress in Utilizing Radiosondes and Satellites for Monitoring Upper Air Temperature Profiles

“…The latter issues have led to the regional reduction in RS reports and However, the exact number of active RS launching sites that exist globally to date is still not clear as the number fluctuates due to a couple of factors. For instance, most of the RS stations are sparsely distributed and are densely concentrated in Northern Hemisphere land areas (see Figure 3), leaving large regions of the world's oceans and the Southern Hemisphere essentially unmonitored [26,74]. Figure 3 shows that regions such as Africa, South America, the Southern Ocean, and the Antarctic are particularly underrepresented in terms of radiosonde observations.…”