DRP-4DVar-based Coupled Data Assimilation for Decadal Climate Prediction

A coupled data assimilation (CDA) system using the DRP-4DVar approach (Wang et al, 2010, 2018) was developed based on the Grid-point Version 2 of Flexible Global Ocean–Atmosphere–Land System (FGOALS-g2), one of the CMIP5 models (Li et al, 2013), and the nudging-based CDA system of this model (Wang et al, 2013). Thirty ensemble members required by the DRP-4DVar system are chosen from the outputs of the Pre-Industrial control experiment of FGOALS-g2. The time window of assimilation is one month. A preliminary effort was made to conduct a 62-year CDA cycle for initializing the ocean component of FGOALS-g2, which assimilated the monthly mean analyses of sea temperature and salinity in the deeps up to 1000 m from the dataset ds285.3 (Ishii, 2005, 2006) during 1945–2006. The initialization improved the climate variabilities at annual to decadal scales not only in the ocean but also in the atmosphere, such as El Nio-Southern Oscillation (ENSO), Pacific Decadal Oscillation (PDO), North Pacific index (NPI), Pacific-North American teleconnection (PNA), Atlantic Multidecadal Oscillation (AMO), and summer precipitation over the Tibetan Plateau, comparing with the uninitialized simulation (i.e., the history run) by the same model.

To investigate the impact of the initialization on the prediction skill of FGOALS-g2, thirty six sets of 10-year-long hindcasts were conducted with the analysis as initial conditions, starting from 1961 and then every year until 1996 with ten ensemble members generated by a Lagged Average Forecast method (Hoffman et al. 1983). Higher decadal prediction skills on these climate variabilities were basically achieved. In particular, significant improvements in both precipitation and surface air temperature in summer over the TP in terms of the spatial distribution and annual variability were sighted, comparing with the results from the uninitialized simulation.

References

Hoffman R N and Kalnay E. Tellus A 1983; 35: 100–18.

Ishii M, Shouji A and Sugimoto S et al. Int J Climatology 2005; 25: 865–79.

Ishii M, Kimoto M and Sakamoto K et al. J Oceanography 2006; 62: 155–70.

Li L J, Lin P F and Yu Y Q et al. Adv Atmos Sci 2013; 30: 543–60.

Wang B, Liu J J and Wang S D et al. Adv Atmos Sci 2010; 27: 715–27.

Wang B, Liu M M and Yu Y Q et al. Adv Atmos Sci 2013; 30: 674–83.

Wang B, Liu J J and Liu L et al. PLoS One 2018; 13: e0191088.