Errors and uncertainties in a gridded carbon dioxide emissions inventory

Thomas Lauvaux,Zbigniew Nahorski,Gregg Marland,Vitaliy Kinakh,Olha Danylo,Matthias Jonas,Petro Topylko,Shamil Maksyutov,Mariia Halushchak,Rostyslav Bun,Joanna Horabik-Pyzel,Myroslava Lesiv,Tomohiro Oda

doi:10.1007/s11027-019-09877-2

Abstract

Emission inventories (EIs) are the fundamental tool to monitor compliance with greenhouse gas (GHG) emissions and emission reduction commitments. Inventory accounting guidelines provide the best practices to help EI compilers across different countries and regions make comparable, national emission estimates regardless of differences in data availability. However, there are a variety of sources of error and uncertainty that originate beyond what the inventory guidelines can define. Spatially explicit EIs, which are a key product for atmospheric modeling applications, are often developed for research purposes and there are no specific guidelines to achieve spatial emission estimates. The errors and uncertainties associated with the spatial estimates are unique to the approaches employed and are often difficult to assess. This study compares the global, high-resolution (1 km), fossil fuel, carbon dioxide (CO2), gridded EI Open-source Data Inventory for Anthropogenic CO2 (ODIAC) with the multi-resolution, spatially explicit bottom-up EI geoinformation technologies, spatio-temporal approaches, and full carbon account for improving the accuracy of GHG inventories (GESAPU) over the domain of Poland. By taking full advantage of the data granularity that bottom-up EI offers, this study characterized the potential biases in spatial disaggregation by emission sector (point and non-point emissions) across different scales (national, subnational/regional, and urban policy-relevant scales) and identified the root causes. While two EIs are in agreement in total and sectoral emissions (2.2% for the total emissions), the emission spatial patterns showed large differences (10~100% relative differences at 1 km) especially at the urban-rural transitioning areas (90–100%). We however found that the agreement of emissions over urban areas is surprisingly good compared with the estimates previously reported for US cities. This paper also discusses the use of spatially explicit EIs for climate mitigation applications beyond the common use in atmospheric modeling. We conclude with a discussion of current and future challenges of EIs in support of successful implementation of GHG emission monitoring and mitigation activity under the Paris Climate Agreement from the United Nations Framework Convention on Climate Change (UNFCCC) 21st Conference of the Parties (COP21). We highlight the importance of capacity building for EI development and coordinated research efforts of EI, atmospheric observations, and modeling to overcome the challenges.

Highlights

Emission inventories (EIs) are the fundamental tool to quantify the amount of man-made emissions, such as those of greenhouse gases (GHGs) and other air pollutants, and to keep track of their changes over time
This study evaluates the Open-source Data Inventory for Anthropogenic CO2 (ODIAC) high-resolution emission fields by comparing them with a locally developed, fine-grained EI, the geoinformation technologies, spatiotemporal approaches, and full carbon account for improving the accuracy of GHG inventories (GESAPU, Bun et al 2018; Charkovska et al 2019)
We evaluated the global high-resolution, gridded EI ODIAC using the multiresolution EI GESAPU over the domain of Poland

Summary

Introduction

Emission inventories (EIs) are the fundamental tool to quantify the amount of man-made emissions, such as those of greenhouse gases (GHGs) and other air pollutants, and to keep track of their changes over time. For GHGs, nationally reported EIs are generally compiled following the guidelines prepared by the Intergovernmental Panel on Climate Change (IPCC) (e.g., IPCC 2006). Emissions are reported by countries in order to monitor international compliance of GHG reductions (e.g., under the Kyoto Protocol or Paris Agreement). The IPCC Guidelines provide “best practice” to compile EIs in a consistent manner, regardless of the data availability in different countries. The uncertainties associated with national estimates for fossil fuel carbon dioxide (CO2) emissions (FFCO2) are often relatively small, especially for developed countries (e.g., ± 4% for the USA). As previously discussed in Liberman et al (2007), White et al (2011), and Ometto et al (2015), studying the variety of sources of errors and uncertainties is crucial in order to make EIs more robust and accurate for providing science-based guidance to global climate mitigation

Results

Discussion

Conclusion