Prediction - CMA-CPSv3
CMA-CPSv3
The China Meteorological Administration Climate Prediction System version 3 (CMA-CPSv3) is an integrated ensemble prediction system for sub-seasonal, seasonal, and interannual climate prediction. It is built on the high-resolution, fully coupled climate system model BCC-CSM2-HR, which incorporates interactions among the atmosphere, ocean, land, and sea ice. The system comprises three main components:
1. The high-resolution version of the Beijing Climate Center Climate System Model (BCC-CSM2-HR)
2. Coupled Data Assimilation System
3. Ensemble Forecast System
The high-resolution version of the Beijing Climate Center Climate System Model (BCC-CSM2-HR)
- Atmosphere component: BCC-AGCM3.0 with a horizontal resolution of T266 (~45 km), 56 vertical layers, and a model top at 0.01 hPa.
- Land component: BCC-AVIM2 with a horizontal resolution of T266 and 10 vertical layers.
- Ocean component: MOM5 with a horizontal resolution of 0.25° and 50 vertical layers.
- Sea Ice component: GFDL’s SIS5 with a horizontal resolution of 0.25°.
Coupled Data Assimilation (CDA) System
Developed on the BCC-CSM2-HR, CDA integrates:
- An Ensemble Optimal Interpolation (EnOI) scheme for ocean data assimilation.
- An Optimal Interpolation (OI) scheme for sea ice data assimilation.
- An atmospheric Nudging scheme.
CDA is a weakly coupled assimilation system which harmonizes multi-source observational data (ocean, sea ice, land, and atmosphere) to provide coordinated initial conditions for climate predictions, ensuring reliable analysis of climate system components for CPSv3.
Ensemble Forecast System
The CMA-CPSv3 Ensemble Forecast Syste consists of two Subsystems:
- 1.Sub-seasonal to Seasonal (S2S) Prediction System:
o Includes 4 ensemble members: 1 control member and 3 perturbed members using the Stochastic Perturbed Parameterization Tendency (SPPT) scheme.
o Operates every day, providing 60-day predictions.
- 2.Seasonal to Interannual Prediction System:
o Includes 21 ensemble members: 1 control member and 20 perturbed members combining Lagged Average Forecast (LAF) and SPPT schemes (including the 4 S2S members).
o Operates every month(initialized at the beginning of each month), providing predictions for the next 7 months (extended to 13 months for predictions initialized in March, October, and December).
Products:CPsv3 provides daily, pentad, ten-day, monthly, and seasonal ensemble mean and probabilistic forecasts figures over global, Asia, China, including atmospheric, ocean, land, sea ice variations.
Operational Status:
- CMA-CPSv3 S2S contributed to the Sub-seasonal to Seasonal Prediction Project (S2S) Phase 2 in November 2019, with data publicly available.
- The system was operationalized run in January 2015.