Ecosystem carbon transit versus turnover times in response to climate
warming and rising atmospheric CO<sub>2</sub> concentration
Abstract. Ecosystem carbon (C) transit time is a critical diagnostic parameter to characterize land C sequestration. This parameter has different variants in literatures, including a commonly used turnover time. However, neither of them has been carefully examined under transient C dynamics in response to climate change. In this study, we estimated both C turnover time as defined by the conventional stock-over-flux (i.e., Olson method) and mean C transit time as defined by the mean age of C mass leaving the system (i.e., Rasmussen method). We incorporated them into Community Atmosphere-Biosphere-Land Exchange model (CABLE) to estimate C turnover time and transit time, respectively, in response to climate warming and rising atmospheric [CO2]. Modeling analysis showed that both C turnover time and transit time increased with climate warming but decreased with rising atmospheric [CO2]. The increase of C turnover time with warming was estimated to be 2.4 years with Olson method whereas the transit time increased by 11.8 years with Rasmussen method. The decrease with rising atmospheric [CO2] was estimated to be 3.8 years with Olson method and 5.5 years with Rasmussen method. Our analysis based on Rasmussen method showed that 65 % of the increase in global mean C transit time with climate warming results from the depletion of fast-turnover C pool. The remaining 35 % increase results from accompanied changes in compartment C age structures. Similarly, the decrease in mean C transit time with rising atmospheric [CO2] results approximately equally from replenishment of C into fast-turnover C pool and subsequent decrease in compartment C age structure. Greatly different from the Rasmussen method, the Olsen method, which does not account for changes in either C age structure or composition of respired C, underestimated impacts of either warming or rising atmospheric [CO2] on C diagnostic time and potentially lead to biases in estimating land C sequestration.