Spatially Distributed Fossil Fuel CO2 Emissions in Two U.S. Cities Using Activity Data: Applicability for Global Cities and High-resolution Atmospheric Inversion Modeling

Abstract:

Urban fossil fuel CO₂ (FFCO₂) emissions play a significant role in the global C cycle and climate change. To better understand and monitor urban FFCO₂ emissions, we need timely estimates at fine spatial resolution. However, currently available global estimates have coarse resolution of 10km or more except for some US cities which have finer FFCO₂ estimates at ~250m (Hestia Project; Gurney et al. 2012). We construct an urban sectoral emission model for the U.S. based on multiple cities and spatially disaggregate each sector to arrive at finely resolved emissions data products. We then calibrate our results with other datasets to confirm whether this approach can be applicable in any global urban domain.

We acquire 2012 annual emissions estimates from EPA’s national emissions inventory for the Los Angeles megacity and Indianapolis and apply our U.S. urban sectoral emission model to derive sectoral estimates. We then spatially distribute these sectoral emissions based on activity and other proxy data. We combine remote sensing and open source data such as national land cover data, population density, impervious surface, and road maps to develop intensity metrics of energy use within each sector. These intensity metrics are then used to spatially allocate emissions within each sector. We incorporate global powerplant emissions data to complete our emissions datasets. We validate our urban FFCO2 emissions datasets, both at sectoral and city scales, against Hestia results for two cities and, in case of Indianapolis, compare to results from inverse modeling of atmospheric CO2 concentrations. This study will guide the next phase of research by developing the methodology to determine the spatial variation of FFCO₂ emissions in select cities around the world.