17:15 〜 18:45
[AAS10-P10] Representation of JRA-3Q quasi-biennial oscillation in the zonal wind and ozone
キーワード：長期再解析、JRA-3Q、QBO、オゾンQBO
This study evaluates the representation and consistency of the Japanese Reanalysis for Three Quarters of a Century (JRA-3Q) quasi-biennial oscillation in zonal wind and ozone (zonal wind QBO and ozone QBO). The JRA-3Q is the latest third generation of global atmospheric reanalysis spanning late 1940s onwards, and it is based on the Japan Meteorological Agency (JMA)’s operational system using an atmospheric model with a reduced horizontal resolution of TL479 and 100 vertical layers up to 0.01 hPa with 6-hourly 4D-Var data assimilation as of December 2018.
The monthly mean zonal wind and ozone mixing ratio are analysed from global reanalysis datasets of JRA-3Q, ERA5, MERRA-2, and JRA-55, -55C, -25. A combined radiosonde dataset at the equatorial belt compiled by the Free University of Berlin (FUB) is used for the evaluation, along with additional Singapore sonde data derived from a National Aeronautics and Space Administration (NASA) site. The ozone measurements produced by the SPARC Data Initiative from 1978 to 2010 and Aura MLS satellite ozone dataset from 2005 to 2022 are used in the form of monthly zonal mean.
In the satellite era all reanalysis QBOs are fairly in good agreement with the observations, with only slight differences that are seen in MERRA2 that has stronger westerlies in the upper stratosphere. In the pre-satellite era, 10 hPa zonal wind for JRA-55 and ERA5 shows fairly in good agreement with observations while JRA-3Q QBO is fairly degraded before 1980s, even compared with that in JRA-55. It is noted that representing modelled QBO in JRA-3Q is more sophisticated than the one in JRA-55 because JRA-3Q uses a scheme of non-orographic gravity wave drag parameterization whereas JRA-55 simply applied Rayleigh friction for layers above 50 hPa. Degraded JRA-3Q QBO is probably due to the fact that the background error covariances used for JRA-3Q are optimized for the current enhanced observing system with a shorter horizontal correlation length than the ones used for JRA-55, which might be why errors in background fields were not sufficiently corrected where observations were sparse.
The annual oscillation of zonal winds usually has a peak at 2 hPa, but there is no annual oscillation for JRA-55 in the pre-satellite era. Because JRA-55 adopted rather strong Rayleigh friction at upper levels, such an unrealistic effect on the zonal wind and temperature fields would result in the annual oscillation to disappear in the pre-satellite era.
Ozone is used as a boundary condition in the JRA-3Q reanalysis system and plays an important role in determining temperature and wind distribution. Ozone data used or produced in long-term reanalysis by the JMA or other organizations express a realistic total ozone distribution by assimilating ozone observations, and the vertical distribution is also relatively well represented in the stratosphere. The power of the observed ozone QBO (SPARC-DI + AURA-MLS) is clearly represented with two major power regions: the lower QBO driven by dynamics and the upper QBO mainly driven by chemistry, and there is a minimum power of the ozone QBO that is clearly seen at around 15 hPa. However, the separation of the upper and lower parts of the ozone QBO in the reanalysis is not clear. These results indicate that the observational data for data assimilation those is not homogeneous throughout and that the data assimilation system used in the reanalysis is inconsistently between dynamics and chemistry.
The monthly mean zonal wind and ozone mixing ratio are analysed from global reanalysis datasets of JRA-3Q, ERA5, MERRA-2, and JRA-55, -55C, -25. A combined radiosonde dataset at the equatorial belt compiled by the Free University of Berlin (FUB) is used for the evaluation, along with additional Singapore sonde data derived from a National Aeronautics and Space Administration (NASA) site. The ozone measurements produced by the SPARC Data Initiative from 1978 to 2010 and Aura MLS satellite ozone dataset from 2005 to 2022 are used in the form of monthly zonal mean.
In the satellite era all reanalysis QBOs are fairly in good agreement with the observations, with only slight differences that are seen in MERRA2 that has stronger westerlies in the upper stratosphere. In the pre-satellite era, 10 hPa zonal wind for JRA-55 and ERA5 shows fairly in good agreement with observations while JRA-3Q QBO is fairly degraded before 1980s, even compared with that in JRA-55. It is noted that representing modelled QBO in JRA-3Q is more sophisticated than the one in JRA-55 because JRA-3Q uses a scheme of non-orographic gravity wave drag parameterization whereas JRA-55 simply applied Rayleigh friction for layers above 50 hPa. Degraded JRA-3Q QBO is probably due to the fact that the background error covariances used for JRA-3Q are optimized for the current enhanced observing system with a shorter horizontal correlation length than the ones used for JRA-55, which might be why errors in background fields were not sufficiently corrected where observations were sparse.
The annual oscillation of zonal winds usually has a peak at 2 hPa, but there is no annual oscillation for JRA-55 in the pre-satellite era. Because JRA-55 adopted rather strong Rayleigh friction at upper levels, such an unrealistic effect on the zonal wind and temperature fields would result in the annual oscillation to disappear in the pre-satellite era.
Ozone is used as a boundary condition in the JRA-3Q reanalysis system and plays an important role in determining temperature and wind distribution. Ozone data used or produced in long-term reanalysis by the JMA or other organizations express a realistic total ozone distribution by assimilating ozone observations, and the vertical distribution is also relatively well represented in the stratosphere. The power of the observed ozone QBO (SPARC-DI + AURA-MLS) is clearly represented with two major power regions: the lower QBO driven by dynamics and the upper QBO mainly driven by chemistry, and there is a minimum power of the ozone QBO that is clearly seen at around 15 hPa. However, the separation of the upper and lower parts of the ozone QBO in the reanalysis is not clear. These results indicate that the observational data for data assimilation those is not homogeneous throughout and that the data assimilation system used in the reanalysis is inconsistently between dynamics and chemistry.