Detecting Disturbance and its Impact on Ecosystem Carbon Balance from Global to Regional Scales

Abstract:

One of the most vital ecosystem services currently provided by the terrestrial biosphere is the removal of approximately one quarter of the anthropogenic CO₂ emitted to the atmosphere. However, as patterns of temperature and precipitation change so is the frequency and intensity of ecosystem disturbance. Despite evidence that ecosystem disturbance regimes have shifted leading to widespread forest mortality, the net effect of disturbance on the carbon (C) balance of forest ecosystems remains uncertain. We will use satellite and atmospheric observations to deconvolve net carbon exchange (NEE) into its component fluxes of gross primary productivity and total respiration (e.g. NEE= GPP – R) at global to regional scales. At the global scale we find that NEE has increased over the last 50 years and appears to have accelerated as a result of diminished R over the last 15 years. However the variance in global NEE has also increased perhaps due to inter-annual variability in R, especially within semi-arid ecosystems. These global trends are not necessarily consistent with regional patterns in the net carbon balance, especially across the western US. Atmospheric mass balance suggests that ecosystems of North America have shifted from a net C sink to a net C source. While prolonged drought across the Western US has likely caused this shift in continental scale NEE, attributing this shift in the net C balance to any one mechanism of disturbance (e.g. drought, insect infestation, and fire) or their interactions is challenging. Lastly, we will evaluate existing observing networks, such as NOAA/ESRL and Ameriflux, and how they can be combined with nascent networks, such as NEON, EarthNetworks, and OCO-2, to identify regional disturbance processes that may be causing increasing variance in the global C cycle.