Seminar: Na Li

Global atmospheric growth rate (AGR) shows large year-to-year variations. Quantifying and understanding the patterns of this variations and their drivers is crucial to better understand global carbon cycle dynamics and better predict the future climate. Primarily contributed by land-atmosphere carbon flux exchange, AGR variations is largely affected by internal climate variability (such as large-scale atmosphere circulation modes, e. g., ENSO), as well as externally forced climate changes (such as human induced climate change). Detecting and separating the internal vs external forced signals in carbon variations is the key to understand the driving mechanisms of interannual variations in AGR.

This PhD study is to disentangle the internally vs. externally driven component in the global AGR interannual variations. We separate the study into 3 topics: 1) investigate the relationship between global AGR variations and internal climate variability (here refer to large atmospheric circulation variations). We use Ridge Regression and sea level pressure (as the main proxy of large atmospheric circulation variations) to predict the global AGR, and to depict the dominant spatial domains that are driving the global AGR variations. 2) evaluate if a suggested pattern in global carbon cycle is anthropogenic driven, or just internal climate variability. We employ several climate model large ensembles to depict the distribution of internal climate variability, and check if a suggested pattern in global AGR can be randomly generated by internal climate variability only, and 3) investigate the time for a forced signal to emerge from internal variability in global and regional carbon cycle variations. Through this study, we hope to gain more understanding of the driving mechanisms of global AGR variations and more accurately detect and attribute the human induced signals from internal climate variability in global carbon cycle.