Statistical properties of internal climate variabilities for studying climate changes and potential emergent constraints

Climate variabilities consist of variations internal to the earth system and climate changes caused by external forcing such as the increase of greenhouse gases. Because the internal variabilities are correlated spatially and temporally, it is necessary to account for the statistical properties of internal variabilities when estimating the effects due to external forcing. We study the internal climate variabilities by examining outputs from multiple climate models that participated in the Coupled Model Intercomparison Projects. We characterize the spatial and temporal covariances using auto-correlations and statistics across different ensemble runs for temperature, top-of-atmosphere fluxes, precipitation, and circulation. We study how to perform statistical tests to assess the significance of climate change signals for forecasting experiments. We compare internel variabilities of multiple models and observational data to examine the consistencies and diversities. We also explore connections between internal variabilities and responses to external forcing, which may lead to emergent constraints for future projections.