Numerical models of the carbon cycle are becoming increasingly sophisticated. One result of this is that these models now require fossil-fuel carbon-dioxide emissions data with sub-annual (e.g., seasonal) time resolution. They also require finer spatial resolution than national averages (i.e., than one point per nation). Finer spatial resolution is especially needed for countries as large in area as the United States of America (U.S.A.). Here we present a summary of monthly data for the entire nation, and annual data for each state in the U.S.A.

Current predictions of future CO₂ sink strength vary widely as a result of different model representations of the carbon cycle. A sound characterization of these prediction uncertainties is crucial for the design of economically efficient carbon management strategies. We use a mechanistically sound and statistically tractable model of the global carbon cycle to (1) assimilate historical observations of atmospheric CO₂ concentrations and oceanic CO₂ fluxes, (ii) derive probabilistic predictions of future CO₂ concentrations and fluxes, and (iii) compare the utility of terrestrial and oceanic observations to constrain predictive uncertainties. We found that terrestrial and oceanic flux observations have nearly equal ability to constrain these uncertainties, if terrestrial observations include both net primary productivity (NPP) and respiration. Model predictions are dependent on the choice of historical land use emissions dataset. The probability density function (PDFs) of model parameter estimates are not normally distributed, and neglecting autocorrelation in the CO₂ concentration signal during model calibration causes overconfident results.

In order to facilitate future decision-making regarding regional carbon fluxes, it is essential to better quantify uncertainty in inverse carbon flux models. At Colorado State University, research is being performed in order to better quantify sources and sinks and associated uncertainties on a mesoscale level, through a coupled atmospheric (RAMS and PCTM) and terrestrial carbon flux (SiB3) model (Denning, 2003).  Carbon-dioxide flux and mixing ratio data were collected from a ring of towers (WLEF tall tower and nearby smaller towers) in northern Wisconsin over the summer of 2004.  The fully coupled terrestrial-atmospheric model, SiB/RAMS, will be forced with 2004 reanalysis data to predict fine scale weather in the vicinity of these towers for the summer of 2004. Relevant portions of this simulated weather, including wind fields and pertinent turbulence components, are extracted and used to create backward-in-time Lagrangian Particle Dispersion Modeled (LPDM) influence functions.  Pseudo spatial carbon-dioxide mixing ratio and flux data created by SiB/Rams is then used as input to several different estimation routines in order to try and predict pseudo tower data at different heights.  Different temporal and spatial aggregation lengths are considered as means of data reduction. Particular attention will be paid to Ensemble Kalman Filter (EnKF) techniques as well as geo-statistical methods as a means of estimation.

Measurements of the partial pressure of CO₂ in surface seawater (pCO₂^w) have been made frequently and extensively in the western North Pacific (3-35°N, 132-142°E) since 1990. Based on the time series analysis of pCO₂^w data, we obtained a “climatological view” of seasonal variation in pCO₂^w in the western North Pacific. We have examined the relationship between pCO₂^w and sea surface temperature (SST). The pCO₂^w–SST relationship varies spatially and temporally. The pCO₂^w showed an average growth rate of 1.6 µatm yr^-1 (nearly equal to that of the air, pCO₂^a) with large variability (±8.9µatm yr^-1). In 1998, larger growth rates of pCO₂^w occurred in the subtropical gyre and the western equatorial Pacific, which was probably associated with the 1997/98 El Niño phenomena. To know processes affecting long-term variations in pCO₂^w, we have examined seasonal variation in growth rate of pCO₂^w. The linear growth rate of pCO₂^w during the winter season ranged from 1.3±0.2 to 2.1±0.2µatm yr^-1 with an average of 1.7±0.2µatm yr^-1. During spring/summer seasons, the average growth rate of pCO₂^w was larger than 2µatm yr^-1 north of 27°N, and within the range from 0 to 1µatm yr^-1 in the North Equatorial Current. These increases were mostly caused by the oceanic uptake of anthropogenic CO₂, and to some extent, other processes controlling the pCO₂^w change: thermodynamic effect, lateral transport and vertical mixing, and biological activity.

Measurements of CO₂ concentration are carried out on a weekly basis since 1992 on the island of Lampedusa (35.5°N, 12.6°E), in the Mediterranean. Measurements are based at the Station for Climate Observations, which rests on a rocky plateau (45 m asl) on the North-Eastern coast of the island, and are made with a NDIR analyzer. World Meteorological Organization (WMO) reference standards are used for calibrations.  Continuous measurements were started in 1998; they were interrupted in early 2003, and activated again in 2005. The continuous observations show evidence of a small daily cycle (amplitude < ±1 ppm) only during the months of June, July, and August. Mean annual cycles derived from weekly flask measurements show a dependency on the wind origin: the annual cycle and the annual CO₂ mean are smaller for winds originating from the Southern sectors, than for winds from Northern sectors. The continuous measurements were combined with daily backward airmass trajectories to identify the dependency of the CO₂ amount on the airmass origin. Trajectories provided by National Oceanic and Atmospheric Administration / Air Resources Laboratory (Hysplit) are used. During winter, low CO₂ is generally connected to Southern/South-Eastern airmasses. In summer airmasses from North often display lower CO₂ content, due to the influence of the European sink.

CO₂ emissions from fossil-fuel combustion can be estimated at the state or monthly level even when full data on fuel combustion are not available. Our hypothesis is that a representative proxy can accurately estimate the pattern of CO₂ emissions if a sufficient fraction of the total can be represented, even if the dataset used does not cover all energy consumption sectors. Our approach employs monthly sales data for each state from the U.S. Department of Energy’s Energy Information Administration (EIA). This is used to estimate the relative proportions of solid, liquid and gaseous fossil fuels for each state for each month.