Frameworks for Analyzing the Terrestrial Carbon Flux in Response to Climate Variability and Extreme Events
The terrestrial biosphere can switch between carbon source/sink by releasing/absorbing carbon to/from the atmosphere depending on the climate variability or extreme events. A lack of knowledge exists as to what extent the terrestrial biosphere responds to increasing temperature or extreme events (e.g., heatwaves, droughts), particularly at high temporal resolutions, due to missing dense in-situ spatio-temporal carbon flux observations. To understand, assess, and quantify the terrestrial biosphere carbon flux exchange in response to increasing temperatures and extreme events (i.e., heatwave), I have (1) measured continuous high frequency soil respiration (Rs) through a unique experimental setup and (2) used global daily carbon flux from satellite observations and model simulations. Using continuous high frequency in-situ measurements, I present a novel probabilistic framework in which I provide new insights in detecting changes in the Rs distribution in response to shifts in hydroclimate drivers. Furthermore, I employ the continuous high frequency Rs data and ten additional data sites of Rs spanning the contiguous United States (CONUS), to characterize the relationship between Rs and heatwaves. Applying the probabilistic framework mentioned earlier, I conclude that during heatwaves Rs rates increase significantly, on average, by ~27% relative to that of non-heatwave conditions. This indicates that the terrestrial feedback to the carbon cycle may be largely underestimated without capturing these high frequency extreme events (i.e., multi-day heatwaves). Using the available remotely sensed global daily carbon flux data, I quantify the response of the net ecosystem exchange (NEE) to warming, both at global and regional scales. I display that 1.5 oC of warming above the long-term average temperature (Tlgtrm) increases the likelihood of the terrestrial biosphere acting as a carbon sink by 16.3%. Shifting to 2.0 oC above Tlgtrm does not substantially elevate the probability of carbon assimilation above that of 1.5 oC of warming, probably leading to higher anthropogenic carbon emissions in the atmosphere. Moreover, I show that the length of the carbon uptake period increases for study regions below 50o N by ~53 days, on average, which is almost twice as much as that for sites above 50o N (~27 days).