Climate Variability Impact on Regional Carbon Fluxes over Temperate and Boreal North America

A comprehensive analysis is conducted to quantitatively investigate the relationship between climate variability and regional net carbon exchanges over North America. The net carbon fluxes are derived from multiple improved time-dependent inversion models based on an extended time period (1980-2008). The potential effects of climate variability on variations in net carbon fluxes in six sub-regions of North America are characterized by performing correlation analysis with global surface air temperature and precipitation fields via lagged time series correlation and spatial correlation. A further analysis is performed to mechanically quantify these relationships through comparison with observationally-based data-products such as the Palmer Drought Severity Index (PDSI). We set out to test the hypothesis that the evolution of the carbon fluxes in space and time is determined by the large-scale El Niño-Southern Oscillation (ENSO) teleconnection patterns. Our analysis largely confirms this connection, in that the sub-continental North American carbon fluxes are associated with large-scale temperature and precipitation anomalies that are characteristic of ENSO. Correlations between regionally-integrated fluxes are found to be significant with ENSO-like climate patterns at lags of zero to a few months. The correlations are shown to be consistent with a propagation pattern from the equatorial Pacific to the extra-tropics. Importantly, the relationship between carbon flux variability over North America and large-scale patterns of variability in the physical state of the atmosphere are considered for its mechanistic implications.