5:15 PM - 7:15 PM
[ACG47-P03] Intercomparison of dynamic energy cycles between JRA-55 and JRA-3Q
Keywords:Wave-mean flow interaction, Meridional circulation, Energy cycle
In this study, the dynamic energy balances of reanalyses are estimated using the three-box model in the mass-weighted isentropic zonal mean framework (Uno and Iwasaki, 2006). In this box model, atmospheric energy is classified into three forms: zonal available potential energy (Az), zonal mean kinetic energy (Kz), and wave energy (W). By computing residuals between energy conversion and generation, the dynamic energy balances for Az (Error-Z) and W (Error-W) were evaluated as indices of the precision of reanalysis datasets.
Reanalysis datasets are crucial for better understanding atmospheric dynamics and the historical climate changes. It is useful to clearly evaluate the precision of reanalysis. Kosaka et al. (2024) reported that forecast models used for the Japanese Reanalysis for Three Quarters of a Century (JRA-3Q) improved temperature and heating profiles, as well as radiation budgets, compared to the previous Japanese 55-year Reanalysis (JRA-55; Kobayashi et al. 2015). JRA-3Q was validated with other datasets or observations. Meanwhile, evaluating reanalysis datasets from various perspectives remains essential. Therefore, in this study, JRA-3Q was assessed from the standpoint of dynamic energy balance consistency.
We evaluated the JRA55 and JRA3Q using the two indices for the period from December 1990 to November 2020, focusing in particular on December–February (DJF) and June–August (JJA). The mean values of the two indices show that dynamic energy balances are improved in JRA-3Q compared to those in JRA-55. In DJF (JJA), Error-Z decreased from 3.2% (22%) in JRA-55 to 2.8% (20%) in JRA-3Q. Similarly, Error-W decreased from -9.4% (19%) in JRA-55 to -6.1% (-9.9%) in JRA-3Q. These results imply that the systematic part of increment from data assimilation in JRA-3Q could be smaller, resulting from the improvement of physics parameterization packages of the global NWP model. The errors in the dynamic energy balance are much larger in JJA than in DJF, which is caused by strong diurnal variability of global mean diabatic generation of Az and W in JJA.
Reanalysis datasets are crucial for better understanding atmospheric dynamics and the historical climate changes. It is useful to clearly evaluate the precision of reanalysis. Kosaka et al. (2024) reported that forecast models used for the Japanese Reanalysis for Three Quarters of a Century (JRA-3Q) improved temperature and heating profiles, as well as radiation budgets, compared to the previous Japanese 55-year Reanalysis (JRA-55; Kobayashi et al. 2015). JRA-3Q was validated with other datasets or observations. Meanwhile, evaluating reanalysis datasets from various perspectives remains essential. Therefore, in this study, JRA-3Q was assessed from the standpoint of dynamic energy balance consistency.
We evaluated the JRA55 and JRA3Q using the two indices for the period from December 1990 to November 2020, focusing in particular on December–February (DJF) and June–August (JJA). The mean values of the two indices show that dynamic energy balances are improved in JRA-3Q compared to those in JRA-55. In DJF (JJA), Error-Z decreased from 3.2% (22%) in JRA-55 to 2.8% (20%) in JRA-3Q. Similarly, Error-W decreased from -9.4% (19%) in JRA-55 to -6.1% (-9.9%) in JRA-3Q. These results imply that the systematic part of increment from data assimilation in JRA-3Q could be smaller, resulting from the improvement of physics parameterization packages of the global NWP model. The errors in the dynamic energy balance are much larger in JJA than in DJF, which is caused by strong diurnal variability of global mean diabatic generation of Az and W in JJA.