Near-real-time methodology for assessing global carbon emissions

Abstract

Climate change is a major challenge for the sustainable development of mankind. Carbon emissions from human activities are the main driving force of global climate change, and the quantification of carbon emissions is the basis for coping with global changes and achieving carbon neutrality. Developing more spatially and temporally fine-grained carbon emission data to achieve more precise, accurate, and timely carbon emission monitoring is at the current forefront of the field and a major national demand. Here, a carbon emission quantitative method for near-real-time global carbon emissions is proposed, based on multisource activity data such as statistics, satellite remote sensing, and observation. By parameterizing the extent of daily human activity, it can achieve a near-real-time quantitative estimation of global and regional carbon emissions according to the methodology of the IPCC 2006 guidelines, resolving preexisting challenges, including time lag of yearly emission inventories and how to spatialize the national inventories in high temporal resolution. This paves the way for more accurate, reliable, and verifiable carbon monitoring. Specifically, near-real-time estimates can reveal daily, weekly, monthly, and seasonal changes in global carbon emissions. Results show that emissions are highly related to human activity (e.g., the emissions from Monday to Friday are at a high level but return to a relatively low level during weekends). In addition, winter emissions are higher than those of summer, reflecting the greater demand for heating in the winter for populations in the northern hemisphere and cooling demands in summer. This phenomenon can indicate the variations of seasonal changes in each country, where temperatures at different latitudes reflect heating and cooling demands. Sectoral emissions demonstrate the seasonality of power, including that used by residential sectors. During the COVID-19 pandemic, emissions dropped unprecedentedly, with the emissions of the power sector decreasing rapidly. During the pandemic, domestic aviation emissions were similar to ground transport emissions, while international aviation emissions remained low due to the restrictions imposed on entering and leaving countries. Spatialized daily emissions reveal discrepancies of fine-grained sectoral emissions with a spatial resolution of 0.1°×0.1°. Global daily average emissions show that emissions are concentrated within eastern America, western Europe, southeastern China, etc., with the emerging hotspots being the megacities in each region. Sectoral emissions vary because the sources of emissions of each sector are diverse. Uncertainties are crucial for evaluating the performance of spatialization and fine-grained temporal discretization of activity data. This methodology considers per-sector uncertainties according to the IPCC 2006 guidelines. The uncertainties for power, industry, ground transport, residential, aviation, and international shipping sectors are 14%, 36%, 9.3%, 40%, 10.2%, and 13.0%, respectively. For spatialization, the uncertainties come from national emission data and the EDGAR and GID datasets. The baseline emissions, point-source emissions and scale, non-point-source distribution, and proxy data contribute to the uncertainties. In the future, additional high spatiotemporal resolution data will be used in extra cross-validation and corrections to achieve more precise carbon monitoring.