Assessment of the FIO-CPS v2.0 in predicting 2-meter air temperature over China

Abstract

The climate prediction system is an essential tool for predicting climatological state and variability. Systematic evaluation of the output is critical for assessing the prediction performance and making improvement. In this study, we evaluate the prediction capability of the First Institute of Oceanography-Climate Prediction System version 2.0 (FIO-CPS v2.0), a short-term climate prediction system, on the 2-meter air temperature over China using five criteria, namely prediction score (PS), prediction consistency (PC), correlation coefficient (CC), root mean square error (RMSE), and distance between indices of simulation and observation (DISO). The results showed that FIO-CPS v2.0 has higher accuracy in summer, and its performance varies with different lead times depending on the evaluation criteria used. Higher overall prediction skill was mostly found in the northeastern region during July and September, and the southeastern coastal region during June–September. Our findings provide insights into the prediction ability of the FIO-CPS v2.0 on air temperature and may help to facilitate its development.