Model analysis of atmospheric observations reveals methane leakage in North China

1. Background

Natural gas is a relatively clean burning fossil fuel, that causes less air pollution than coal and is widely used in the world. Recent studies have shown that the natural gas leaks from production, supply chain, and end-use facilities are a large source of atmospheric methane (CH₄), and the leaking budget is underestimated in many places by bottom-up inventories. CH₄ is the second most important greenhouse gas (GHG) contributing to global warming after carbon dioxide (CO₂), with a relatively shorter lifetime, making the reduction of CH₄ emission a suitable target for implementing rapid and achievable mitigation strategies of the Paris Agreement.

Over the last decade, natural gas has become the fastest-growing fossil energy source in China due to the coal-to-gas government initiative that has been implemented to reduce air pollution and CO₂ emissions. Natural gas consumption has increased dramatically from 108.5 billion standard cubic meters (bcm) (4% of primary energy consumption) in 2010 to a record level of 280 bcm (7.6% of primary energy consumption) in 2018. In addition, according to China’s energy plan, the share of primary energy from gas will keep increasing and is likely to reach 15% by 2030, while coal and oil consumption will decline. From 2010 to 2018, the length of gas supply pipelines in urban areas of China increased approximately three-fold from 298 to 842 thousand kilometers. However, CH₄ leakage from those pipelines has not been actively reported, and there is limited publicly available data on upstream emissions and local distribution of natural gas emissions in China.

2. Research Outline and Results

In this study, we used nine years (2010–2018) of CH₄ observations by the Greenhouse gases Observing SATellite “IBUKI” (GOSAT) and surface station data from the World Data Centre for Greenhouse Gases (WDCGG) to estimate CH₄ emissions in different regions of China. GOSAT observes the column-averaged dry-air mole fractions of CH₄ in the atmosphere, and the surface stations monitor CH₄ concentrations near surface. The observation data were used for simulations by the high-resolution inverse model NTFVAR (NIES-TM-FLEXPART-variational) to infer the surface flux of CH₄ emissions. Inverse modelling optimizes prior flux estimates, which are constrained so that an acceptable agreement between the simulated and observed atmospheric concentrations is achieved.

Figure 1 shows the model-estimated CH₄ fluxes in four regions of China. The four regions, North China (NE), South China (SE), North-west China (NW), and the Qinghai-Tibetan Plateau (TP), vary with respect climate, geographical features, types of agriculture, major economic activities, and CH₄ emission sources. The model-estimated average CH₄ emissions from the four subregions over the period 2010–2018 are 30.0±1.0 (average ± standard deviation) Tg CH₄ yr^-1 from the SE region, 23.3±2.7 Tg CH₄ yr^-1 from the NE region, 2.9±0.2 Tg CH₄ yr^-1 from the NW region, and 1.7±0.1 Tg CH₄ yr^-1 from the TP region. The trends in CH₄ emissions have varied in the different regions of China over the last nine years, with significant increase trends detected in the NE region and the whole China.

We focused our analysis on the NE region where natural gas production and consumption have increased dramatically and are likely one of the main contributors to the increase estimated in regional total CH₄ emissions. The CH₄ emissions from natural gas, including leakage from fuel extraction, processing, transport, and the end-use stage, were estimated using an approach that combined data for the province-level emissions inventory and published inverse model studies. The model-estimated total CH₄ emissions and the estimated natural gas emissions both increased significantly during 2010–2018 (Figure 2). The total amount of natural gas emissions due to leakages constitutes a significant waste of energy and value. For example, in 2018, natural gas consumption in the NE region was 101.5 bcm and the estimated total natural gas emissions were 3.2%–5.3% of regional consumption.

Figure 3 shows the changes in estimated CH₄ emissions from natural gas and the model-estimated total CH₄ emissions for 2010-2018 compared to previous years in the NE region. The year-over-year change in the model-estimated total CH₄ emission closely follows the changes in CH₄ emissions from natural gas. In January 2016, record cold wave hit the region causing a sudden increase in natural gas use, and natural gas suppliers recorded an increase in natural gas loss (i.e., the difference between the amount of gas purchased and the amount of gas sold). Simultaneously, the atmospheric observations also captured the emission changes, as reflected in our inverse estimates (Figure 3). The analysis shows a strong correlation between trends in natural gas use and the increase in the atmospheric CH₄ concentration over the NE region, which is indicative the ability of GOSAT to monitor variations in regional anthropogenic sources.

3. Future Perspectives

The findings of our study highlight that the increase in natural gas use threatens China’s carbon reduction efforts. The increase in CH₄ leaks from natural gas production and the supply chain will adversely affect the interests of diverse stakeholders, despite the introduction of carbon reduction measures. Given that the large natural gas distribution pipelines span more than 900 thousand kilometers in China, natural gas leaks constitute a significant waste of energy and value. The year-over-year changes in regional emissions and trends were detected by satellite and surface observations in this study. In the future, additional observations using high-resolution satellites will help to more accurately quantify emissions and provide scientific directions for emission reduction measures. There is also a need to further detect and locate such leaks using advanced mobile platforms in order to effectively mitigate CH₄ emissions in China and bring about economic, environmental, and health benefits.

4. Data Availability

GOSAT data used in this study are available from the GOSAT Data Archive Service https://data2.gosat.nies.go.jp/index_en.html

In-situ methane observation data are archived on the WDCGG Global Network: https://gaw.kishou.go.jp/

Emissions Database for Global Atmospheric Research (EDGAR) emission inventories are available for download at

https://edgar.jrc.ec.europa.eu/

Global Fire Assimilation System (GFAS) fire emissions Database are from https://www.ecmwf.int/en/forecasts/dataset/global-fire-assimilation-system

Wetland emission by Vegetation Integrative SImulator for Trace gases (VISIT) model are available at

https://www.nies.go.jp/doi/10.17595/20210521.001-e.html

The NIES airborne and Japan-Russia Siberian Tall Tower Inland Observation network (JR STATION) data are available at

https://db.cger.nies.go.jp/ged/en/index.html

The Japanese 55-year Reanalysis (JRA-55) data from the Japanese Meteorological Agency (JMA) are available at

https://search.diasjp.net/en/dataset/JRA55

5. Supplementary Information

○ Greenhouse gases Observing SATellite “IBUKI” (GOSAT)

The Greenhouse Gases Observing Satellite "IBUKI" (GOSAT) is the world’s first spacecraft to monitor the concentrations of the two major GHGs CO₂ and CH₄ from space. NIES has promoted the GOSAT series projects for GHG observation from space, together with the Ministry of the Environment, Japan (MOE) and the Japan Aerospace Exploration Agency (JAXA). GOSAT (IBUKI) is the first satellite in the series and has been observing column-averaged concentrations of CO₂ and CH₄ for more than 13 years since its launch in 2009. The second satellite, GOSAT-2 (IBUKI-2) was launched in 2018 and started observing carbon monoxide in addition to CO₂ and CH₄. Furthermore, the third satellite, Global Observing SATellite for Greenhouse gases and Water cycle (GOSAT-GW) is under development and due for launch in Japanese fiscal year 2023.

○ Lifetime of methane in the atmosphere

Methane is the second most important well-mixed GHG contributing to human-induced climate change after CO₂. The lifetime of CH₄ in the atmosphere refers to the time that CH₄ stays in the air after being emitted from a variety of sources. CH₄ is removed from the atmosphere mostly by chemical reactions. The atmospheric lifetime of CH₄ is 10 ± 2 years, which is relatively shorter than that of CO₂ (approximately 5 to 200 years) (IPCC, 2013).

○ Methane emission sources

Methane is emitted from a variety of anthropogenic and natural sources. Approximately 60% of all CH₄ emissions come from anthropogenic sources, such as agricultural activities, waste treatment, oil and natural gas systems, coal mining, stationary and mobile combustion, and certain industrial processes. Natural emissions include wetlands, freshwater bodies such as lakes and rivers, and geological sources such as terrestrial and marine seeps and volcanoes. Other smaller sources include ruminant wild animals, termites, hydrates and permafrost.

○ Underestimation of methane emissions from oil and gas using bottom-up inventories

Methane can leak into the atmosphere from upstream/downstream natural gas operations (i.e., extraction and gathering, processing, transmission and storage, and distribution) and end-use combustion. Atmospheric measurement studies have shown that a large amount of CH₄ emissions from oil and gas production are unaccounted for in bottom-up inventories. Using high-resolution satellite observations, Zhang et al. (2020) estimated a leakage equivalent to 3.7% (~60% higher than the national average leakage rate) of all the gas extracted from the largest oil-producing basin in the United States. Chan et al. (2020) reported eight-year estimates of CH₄ emissions from oil and gas operations in western Canada and found that they were nearly twice that from inventories. Weller et al. (2020) used an advanced mobile leak detection (AMLD) platform combined with GIS information of utility pipelines to estimate CH₄ leakage from pipelines of local distribution systems in the United States. They found that the leakage from those pipelines was approximately five times greater than that reported in inventories compiled based on self-reported utility leakage data.

○ High-resolution inverse model NIES-TM-FLEXPART-variational (NTFVAR)

Inverse modeling is an important and essential method for estimating GHGs emissions. The model uses atmospheric observation data as a controller in atmospheric models to optimize bottom-up emission inventories (prior fluxes).

The NIES-TM-FLEXPART-variational (NTFVAR) global inverse model was developed by Dr.Shamil Maksyutov’s group at NIES. NTFVAR is combined with a joint Eulerian three-dimensional transport model, the National Institute for Environmental Studies Transport Model (NIES-TM) v08.1i, and a Lagrangian model, the FLEXPART model v.8.0. The transport model is driven by JRA-55 meteorological data from JMA. The prior fluxes include gridded anthropogenic emissions from the EDGAR database, such as energy, agriculture, waste and other sectors; wetland emissions estimated by the Wetland emission by the VISIT model; biomass burning emissions estimated by GFAS; and climatological emissions from oceanic, geological, and termite sources. The inverse modeling problem is formulated and solved to find the optimal value of corrections to prior fluxes minimizing mismatches between observations and modelled concentrations. Variational optimization is applied to obtain flux corrections to vary prior uncertainty fields at a resolution of 0.1° × 0.1° with bi-weekly time steps. A variational inversion scheme is combined with the high-resolution variant of the transport model and its adjoint described by Maksyutov et al. (2021).

References:

Chan, E. et al. Eight-Year Estimates of Methane Emissions from Oil and Gas Operations in Western Canada Are Nearly Twice Those Reported in Inventories. Environmental Science & Technology 54, 14899-14909, doi:10.1021/acs.est.0c04117 (2020).

IPCC 2013: Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. [Stocker, T. F. Q. et al.]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA.

Maksyutov, S. et al. Technical note: A high-resolution inverse modelling technique for estimating surface CO₂ fluxes based on the NIES-TM - FLEXPART coupled transport model and its adjoint. Atmospheric Chemistry Physics 21, 1245–1266 doi:10.5194/acp-21-1245-2021(2021).

Weller, Z., Hamburg, S. & von Fischer, J. A National Estimate of Methane Leakage from Pipeline Mains in Natural Gas Local Distribution Systems. Environmental Science & Technology 54, 8958-8967, doi:10.1021/acs.est.0c00437 (2020).

Zhang, Y. et al. Quantifying methane emissions from the largest oil-producing basin in the United States from space. Science Advances 6, doi:10.1126/sciadv.aaz5120 (2020).