The Korea Meteorological Administration (KMA) is advancing its prediction system to produce better climate forecasts in collaboration with academic societies and the APEC Climate Center  (APCC) as the research partner of KMA. 

The operationalization of techniques should be decided upon based on the consistent verification criteria. Therefore, the standard verification framework is needed to assess the techniques developed objectively and comprehensively and ensure fairness. To address this, APCC is constructing the Climate forecast model Evaluation & Management system by APCC (CrEMA), which consists of two parts, namely skill assessment of various climate variables and diagnostics for major climate variability. The former was developed in 2022 and presents the prediction skill of deterministic and probabilistic forecasts, while the latter was developed in 2023 and diagnoses the reproducibility of most variability in the model environment based on El Nino-Southern Oscillation (ENSO), Madden-Julian Oscillation (MJO), and the East Asian summer monsoon (EASM).

The CrEMA diagnostic metrics were designed concerning the CLIVAR ENSO metrics and CLIVAR MJOWG diagnostics, which evaluate the specific climate variability of coupled models such as CMIP and are reorganized into the factors suitable for climate prediction diagnosis. The CrEMA diagnostic metrics are categorized into performance, process, and teleconnection, which aims to provide comprehensive evaluation information for each climate mode.

The results are expressed in a scorecard format, indicating relative skill differences compared with the reference model, and present the level of the current operation model and effectiveness of the developed technology.  The goal of the CrEMA is to identify improvements and weaknesses in the model briefly and suggest ways to improve the model based on process-based metrics.