While CarbonTracker applies bottom-up estimates of fossil fuel and
wildfire emissions directly, it attempts to estimate optimal fluxes
from the ocean and land biosphere. This optimization starts with a
first guess of the flux component and its uncertainty (left panel).
The one standard-deviation level of uncertainty for the land flux is
shown with a thin green vertical line, and the uncertainty for oceanic
exchange is shown with a thin blue vertical line. The total
uncertainty, extending to the end of the vertical bar, incorporates
within-model and across-model uncertainties. This is due to the use
of the 8 prior flux models and two transport estimates used in CT2013; within-model error is the
average posterior uncertainty from these 16 simulations, and
across-model error is the additional uncertainty from the scatter of
the 16 inversions. As shown in the legend, the within-model error,
which usually dominates the total error, is indicated by the small
horizontal cross bar in each vertical error bar. Fossil fuel
uncertainty is shown by the darker tan bar inset into the main fossil
bar. The two fossil fuel emissions estimates used in CT2013 are at
the ends of the darker inset bar.
fluxes are subsequently evaluated against atmospheric CO2 observations and modified to be consistent with
those measurements. The optimized results (right panel) have
different mean fluxes and reduced uncertainty ranges, indicating the
influence of information brought by the observations. The uncertainty
on the net flux (black vertical line) is the RMS sum of land biosphere
and oceanic flux uncertainties.
Note that fossil fuel emissions can occur over regions characterized
as ocean or non-optimized regions such as ice, polar deserts, and
inland seas. This is partly due to real emissions from international
shipping, and partly due to emissions occurring in coastal land
regions that are assigned to the ocean in our coarse 1° x 1°
division scheme. Similarly, land regions may have a small
contribution from air-sea gas exchange, and vice versa.
Cumulative flux anomalies (bottom panel). Cumulative anomalies of the natural flux for each year, resolved by month. Flux anomalies are defined as the optimized flux for a given year and month minus the climatological mean flux for that same month. These anomalies are accumulated from January to the current month of each year in order to show the impact of unusual seasons on the total annual flux.
Natural flux (top panel). First guess (red) and final (black) natural fluxes from the specified region, resolved by week. "Natural flux" is the sum of land biosphere, fire, and air-sea CO2 emissions to the atmosphere. It does not include fossil fuel emissions. Units are PgC yr-1. Faint background lines are the raw, weekly fluxes integrated over the specified region. The dark black and red lines in the foreground have been smoothed using a low pass filter. The filter used is an eight-week Hanning window, which has a similar frequency response to a four-week boxcar filter, but superior performance in attenuating high frequencies.
Fossil fuel flux (bottom panel). As in the top panel, but for fossil fuel emissions. Note that fossil fuel emissions are not optimized in CarbonTracker.