June 3, 2020
Summary of New Project Publication for Non-Specialists
Publication: Archana Dayalu, J. William Munger, Yuxuan Wang, Steven C. Wofsy, Yu Zhao, Thomas Nehrkorn, Chris P. Nielsen, Michael B. McElroy, and Rachel Chang. 2020. “Evaluating China's anthropogenic CO2 emissions inventories: a northern China case study using continuous surface observations from 2005 to 2009.” Atmospheric Chemistry and Physics.
Archana Dayalu is a staff scientist at Atmospheric and Environmental Research. She conducted the study as part of her Ph.D. dissertation research in the Harvard Department of Earth and Planetary Sciences, advised by J. William Munger and Steven C. Wofsy and as a student affiliate of the Harvard-China Project. In addition to Dayalu, Munger, and Wofsy, co-authors Yuxuan Wang, Yu Zhao, Chris P. Nielsen, and Michael B. McElroy are also affiliates of the Harvard-China Project. The CO2 observational data were collected at the Miyun atmospheric observatory operated jointly by the Harvard-China Project and Tsinghua University.
This HCP Research Brief for non-specialists was written on behalf of the team by Archana Dayalu.
Summary of the Research
China has pledged to reduce carbon dioxide (CO2) emissions per unit GDP by 60–65% relative to 2005 levels, and to peak carbon emissions overall by at least 2030. However, there is significant disagreement among estimates (or “inventories”) of Chinese anthropogenic CO2 emissions, which makes it difficult for China to track progress toward its goals and evaluate the efficacy of regional control measures.
So-called “bottom-up” CO2 emissions inventories combine “emission factors,” estimates of emissions by type of combustion or other processes that are either calculated or derived from measurement of resulting waste gases, with estimates of the extent of those emitting activities to approximate total emissions at a much larger scale. Due to the uncertainties associated with assumptions in these calculations, it is understandable that there can be differences in predictive skill between inventories. As a result, developing one “true” emissions inventory is not possible.
However, observations of ambient atmospheric concentrations – the result of emissions – can be used “top-down” as an independent truth against which emissions estimates can be evaluated. To this end, our study sought to answer the following:
1. Can we use existing atmospheric observations to determine which anthropogenic CO2 emissions inventories for China are the most accurate?
Our study used observations from a surface site in northern China managed by collaborators at Tsinghua University. The observations were collected hourly from 2005 through 2009, spanning some key events in China’s atmospheric history: 2005 is the benchmark for many of China’s CO2 targets; the Beijing Olympics and many associated emissions reduction measures happened in 2008; and 2008 through 2009 saw the global financial crisis which, as is typical for economic downturns, is linked to additional emissions impacts.
We examined three different inventories spanning a range of typical inventory construction approaches for China. One inventory was China-specific, relying on subnational statistics and domestic field data; the remaining two were spatial subsets of global inventories that relied more on global averages and nationally reported statistics. We found the China-specific CO2 inventory – estimating 30% higher emissions in the northern China region relative to the global inventory subsets – agreed well with the observations and significantly better than the two global inventories. Our results, exploiting a robust time series of continuous observations, support the rates and geographic distribution of northern China’s CO2 as suggested by the China-specific inventory.
2. Can this methodology help with better monitoring and evaluation of China’s CO2 emissions going forward?
Our results support the increased use and development of China-specific inventories that rely on subnational and domestic field data in tracking China's progress as a whole towards reducing emissions. It is likely that inventories relying largely on global average data fail to capture important features in emissions sources specific to China. With our northern China case study, we have also demonstrated how long-term observations even at a single site can reveal key differences among inventories. Importantly, measuring and verifying CO2 emissions across all of China – both for other administrative regions and nationally – requires a denser observational network. In combination with existing satellite missions, such a network will have a profound impact on China’s ability to track its progress as a potential world leader in emissions reductions.
3. What do China’s “CO2 per unit GDP” targets mean for real climate progress?
We find that carbon intensity in the northern China region decreased by 47% from 2005 to 2009, which appears promising for climate goals. But this can be misleading. We show there was a corresponding 18% increase in the region’s absolute emissions over the same time period, affirming a critical point that carbon intensity targets in emerging economies can be highly insufficient for making real climate progress. For instance, a reduction in in CO2/GDP can happen through increased efficiency of the production process – but if more production is happening, net emissions can still increase. And while CO2/GDP can be useful as a guideline (increased efficiency is a good thing!) climate change is ultimately driven by net emissions and not relative reductions. CO2/GDP should not be used as a stand-alone metric for assessing climate progress.
So-called “bottom-up” CO2 emissions inventories combine “emission factors,” estimates of emissions by type of combustion or other processes that are either calculated or derived from measurement of resulting waste gases, with estimates of the extent of those emitting activities to approximate total emissions at a much larger scale. Due to the uncertainties associated with assumptions in these calculations, it is understandable that there can be differences in predictive skill between inventories. As a result, developing one “true” emissions inventory is not possible.
However, observations of ambient atmospheric concentrations – the result of emissions – can be used “top-down” as an independent truth against which emissions estimates can be evaluated. To this end, our study sought to answer the following:
1. Can we use existing atmospheric observations to determine which anthropogenic CO2 emissions inventories for China are the most accurate?
Our study used observations from a surface site in northern China managed by collaborators at Tsinghua University. The observations were collected hourly from 2005 through 2009, spanning some key events in China’s atmospheric history: 2005 is the benchmark for many of China’s CO2 targets; the Beijing Olympics and many associated emissions reduction measures happened in 2008; and 2008 through 2009 saw the global financial crisis which, as is typical for economic downturns, is linked to additional emissions impacts.
We examined three different inventories spanning a range of typical inventory construction approaches for China. One inventory was China-specific, relying on subnational statistics and domestic field data; the remaining two were spatial subsets of global inventories that relied more on global averages and nationally reported statistics. We found the China-specific CO2 inventory – estimating 30% higher emissions in the northern China region relative to the global inventory subsets – agreed well with the observations and significantly better than the two global inventories. Our results, exploiting a robust time series of continuous observations, support the rates and geographic distribution of northern China’s CO2 as suggested by the China-specific inventory.
2. Can this methodology help with better monitoring and evaluation of China’s CO2 emissions going forward?
Our results support the increased use and development of China-specific inventories that rely on subnational and domestic field data in tracking China's progress as a whole towards reducing emissions. It is likely that inventories relying largely on global average data fail to capture important features in emissions sources specific to China. With our northern China case study, we have also demonstrated how long-term observations even at a single site can reveal key differences among inventories. Importantly, measuring and verifying CO2 emissions across all of China – both for other administrative regions and nationally – requires a denser observational network. In combination with existing satellite missions, such a network will have a profound impact on China’s ability to track its progress as a potential world leader in emissions reductions.
3. What do China’s “CO2 per unit GDP” targets mean for real climate progress?
We find that carbon intensity in the northern China region decreased by 47% from 2005 to 2009, which appears promising for climate goals. But this can be misleading. We show there was a corresponding 18% increase in the region’s absolute emissions over the same time period, affirming a critical point that carbon intensity targets in emerging economies can be highly insufficient for making real climate progress. For instance, a reduction in in CO2/GDP can happen through increased efficiency of the production process – but if more production is happening, net emissions can still increase. And while CO2/GDP can be useful as a guideline (increased efficiency is a good thing!) climate change is ultimately driven by net emissions and not relative reductions. CO2/GDP should not be used as a stand-alone metric for assessing climate progress.
Have a question for the researchers? Email us at harvardchinaproject@harvard.edu.
The study was funded in part by the Harvard-China Project’s award from the Harvard Global Institute on the theme of "China 2030/2050: Energy and Environmental Challenges for the Future.”