Abstract

We are interested in the spatial distribution of fossil‐fuel‐related emissions of CO₂ for both geochemical and geopolitical reasons, but it is important to understand the uncertainty that exists in spatially explicit emissions estimates. Working from one of the widely used gridded data sets of CO₂ emissions, we examine the elements of uncertainty, focusing on gridded data for the United States at the scale of 1° latitude by 1° longitude. Uncertainty is introduced in the magnitude of total United States emissions, the magnitude and location of large point sources, the magnitude and distribution of non‐point sources, and from the use of proxy data to characterize emissions. For the United States, we develop estimates of the contribution of each component of uncertainty. At 1° resolution, in most grid cells, the largest contribution to uncertainty comes from how well the distribution of the proxy (in this case population density) represents the distribution of emissions. In other grid cells, the magnitude and location of large point sources make the major contribution to uncertainty. Uncertainty in population density can be important where a large gradient in population density occurs near a grid cell boundary. Uncertainty is strongly scale‐dependent with uncertainty increasing as grid size decreases. Uncertainty for our data set with 1° grid cells for the United States is typically on the order of ±150%, but this is perhaps not excessive in a data set where emissions per grid cell vary over 8 orders of magnitude.

Key Points

CO₂ emissions data at 1° resolution carry large uncertainty The dominant source of uncertainty is in using population density as a proxy for CO₂ emissions Uncertainty in the location and magnitude of large point sources is a major source of uncertainty