Probabilistic Simulations and Projections of Global Temperature Based on Greenhouse Gases

ABSTRACT Climate simulations and projections play a critical role in the research of climate change. Their estimated values and ranges are often calculated by the multi‐model ensemble. Therefore, in order to enrich research methods and reduce uncertainty, this paper simulates and projects global temperature anomalies based on a non‐stationary probability model, using greenhouse gases as covariates and calculates the time profile of the exceedance probabilities for each temperature under various emission scenarios. The results show that the simulated interval includes the majority of the actual temperatures, and the projected future temperatures are close to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change. The very likely (likely) ranges of uncertainty are 0.38°C (0.20°C), which reduce the uncertainty. The probability of global temperature anomalies exceeding 3°C is very small, and the 1.5°C global warming level is unlikely (very low scenario), about as likely as not (low scenarios) and very likely (intermediate scenarios) to be exceeded. Furthermore, the time profiles can accurately represent the probability of exceeding a certain temperature for each year.