Publication Date:
2019
Description:
Abstract
We present an observationally‐constrained United States black carbon emission inventory with explicit representation of activity and technology between 1960 and 2000. We compare measured coefficient of haze data in California and New Jersey between 1965 and 2000 with predicted concentration trends and attribute discrepancies between observations and predicted concentrations among several sources based on seasonal and weekly patterns in observations. Emission factors for sources with distinct fuel trends are then estimated by comparing fuel and concentration trends and further substantiated by in‐depth examination of emission measurements. We recommend (1) increasing emission factors for pre‐regulation vehicles by 80%‐250%; (2) increasing emission factors for residential heating stoves and boilers by 70% to 200% for 1980s and before; (3) explicitly representing naturally aspired off‐road engines for 1980s and before; (4) explicitly representing certified wood stoves after 1985. We also evaluate other possible sources for discrepancy between model and measurement, including bias in modeled meteorology, sub‐grid spatial heterogeneity of concentrations, and inconsistencies in reported fuel consumption. The updated United States emissions are higher than the a priori estimate by 80% between 1960 and 1980, totaling 690 Gg/yr in 1960 and 620 Gg/yr in 1970 (excluding open burning). The revised inventory shows a strongly decreasing trend that was present in the observations but missing in the a priori inventory.
Print ISSN:
2169-897X
Electronic ISSN:
2169-8996
Topics:
Geosciences
,
Physics
Permalink