How skillful was the projected temperature over China during 2002-2018?

Area of Science:

• Climate Science
• Atmospheric Science
• Environmental Science

Background:

• Climate change necessitates accurate regional projections for adaptation and mitigation.
• The skill of past climate model projections over China is largely unknown.
• Assessing model performance is crucial for reliable future climate predictions.

Purpose of the Study:

• To retrospectively evaluate the skill of climate models from IPCC Assessment Reports (TAR, AR4, AR5) for near-term temperature projections in China.
• To analyze model performance in projecting temperature climatology and trends from 2002-2018.
• To investigate the impact of weighting methods and emission scenarios on projection uncertainty.

Main Methods:

• Retrospective analysis of climate model outputs from IPCC TAR, AR4, and AR5.
• Comparison of model-simulated near-surface (2m) air temperature changes against observational data in China (2002-2018).
• Evaluation of model skill for both temperature climatology and trends, employing arithmetic and reliability ensemble averaging.

Main Results:

• Climate models demonstrated skill in projecting China's temperature climatology and trends several years ahead.
• Models generally exhibited cold biases, with TAR models overestimating and AR4 models underestimating the national average trend.
• Weighting methods reduced projection uncertainty, while emission scenarios showed minor differences for climatology but significant differences for trends.

Conclusions:

• IPCC climate models are skillful for near-term temperature projections in China, particularly for climatology.
• Model biases and uncertainties in trend projections require careful consideration.
• Ensemble weighting and emission scenarios influence the reliability and range of future climate predictions.