This is the wiki for the Global Carbon Project's new emerging activity 'Towards a routine update of the CH4 global budget'.

Members are welcome to edit the pages, add new information and upload files.

This site is under constant change and improvement.

Click here to go to the Meetings page.

Click here to go to the Resources page (resources = presentations, data files, plots, etc.).

GCP Methane - Towards a routine update of the CH4 global budget

Our understanding of the global CH4 budget is limited, and our ability to quantify its underlying drivers and its evolution is poor (Dlugokencky et al., 2003, 2009). Whereas the atmospheric concentration of CH4 has increased nearly continuously since the pre-industrial revolution, its rate of growth has varied drastically. Large growth rates were observed during the late 1970s and early 1980s. Average growth rate was 12.3±3.4 ppbv from 1983-1989, while during the 1990's a sharp decrease in growth rate was observed (average of 6.1±4.3 ppbv). The atmospheric CH4 abundance was constant from 1999 through 2002, implying a steady state in the global atmospheric CH4 budget. Near-zero growth characterized the early 2000s, and from 2007-2009 growth rates increased again, leading to an average growth rate of 2.5±3.8 ppbv. During that time of strongly varying growth rates, decadal budgets of the main source categories (wetland, fire, fossil fuel and agriculture/waste emissions) remained relatively constant. The figure below illustrates this:

Fig 1: Atmospheric concentration, growth rate variability and trends for the last three decades (1983-2009), plus bar charts showing decadal budgets of the natural (wetlands, fires) and anthropogenic (fossil fuel, agriculture + waste) sources for same period. Kirschke et al. (in prep.)

CH4 atmospheric data are provided by several atmospheric networks (e.g., NOAA, CSIRO, LSCE, AGAGE), and several emission inventories, such as EDGAR, GEIA, GFED, provide bottom-up estimates of CH4 emissions from various source categories. Isotopic data of δ13C of CH4 can be used to extract the different processes contributing to the global CH4 budget.

This project aims to:

  • Establish a common timeline for the annual update of atmospheric CH4 observational data, and establish links to data providers/PI’s.
  • Work with emission inventories for regular updates of CH4 inventories.
  • Work with research groups and projects (e.g., Wetchimp) operating several wetland and rice emission models, in order to establish links between these bottom-up estimates and top-down approaches (inverse modeling community).

Elements of the budget

The CH4 budget will be based on a combination of observations, model results, and their analysis/synthesis.


  • Number of livestock, and area of rice cultivation, every year. It may be difficult to have annual updates. FAO mostly likely to be the key data source. Contact person: V. Gitz; Data gathered from FAO, emission data also available from EDGAR
  • Fossil fuel methane emissions, leaks, etc. Very complex set of sources, many components. Contact person: Gregg Marland (alternative: JRC/EDGAR); no update from CDIAC planned, emission data available from EDGAR
  • Fire emissions every year: global dataset from Guido van der Werf. Publically available and Guido can advance quickly the most recent year to us. Contact person: Guido van der Werf (for historical budget: GEIA data base); Data gathered from GFED


  • Wetlands biogeochemistry models. We need an ensemble of models; three models would be appropriate. Run every year. LPJ-WHyMe, contact person: Rita Wania, Renato Spahni Data gathered from Renato, ORCHIDEE, contact person: Bruno Ringeval Data gathered from Bruno, LPJ wetland model, contact person: Elke Hodson Data gathered from Elke, WETCHIMP, contact person: Joe Melton has been contacted, workshop will be held in October, Stefanie might participate - possibility of up to 10 global wetland model estimates.
  • Flooded area. Every year we need to re-calculate the flooded area globally depending on the amount of precipitation. There is at least one top-down model that would be able to provide an estimate. Contact person: Catherine Prigent; has been contacted by: Philippe Bousquet, no update beyond 2006 Note that precipitation data from CRU will be updated every year from 2011.
  • Explore satellite data to further support the estimate of global flooded area. One or two groups, a few years only, every 10 years. A composite of micro-radar, and other remote sensing products. Contact person: Anthony Bloom, Paul Palmer; have been contacted by Pep, use of GRACE data not very promising
  • Atmospheric Inversions, Philippe Bousquet is already working on it. A number of TransCom members working on CH4 inversions. We need a commitment from principle investigators of gas datalabs to release the data as soon as possible. Otherwise, CH4 data is usually available one year later than CO2 data for a given year. Several modeling groups have been identified and contacted; Model results gathered from Philippe Bousquet, Sander Houweling, Lori Bruhwiler, will possibly get results from Paul Palmer Data providers: NOAA, CSIRO and LSCE networks need to be contacted for clear agreement on data provision, e.g. data published in May each year
  • OH sink – long-term trends, not a component of the year-to-year variability. Who is doing it? Contact person: Sophie Szopa, Ivar Isaksen have been contacted by: Stefanie Kirschke and Philippe Bousquet

Timelines and Products

  • First draft of the budget including a draft paper by November 2011, in time to be considered for the upcoming IPCC report. An additional 6 months to complete and submit major paper/s.
  • Total of two or three high-profile papers. One focusing on the historical evolution of the methane budget from pre-industrial (as exists for CO2), and a second paper focusing on the most recent period with higher precision measurements. Focus on three or four major excursions in the growth rate and try to explain.

–> The current manuscript will focus on past and current changes in the CH4 budget, ca. for the last three decades.


  • Stefanie Kirschke leads the group with support from Philippe Ciais, Philippe Bousquet and GCP SSC.
  • Pep Canadell and others help with contacts and getting commitment from modelers and data providers.


Three activities as outlined below have been identified as steps towards a routine update of the CH4 global budget. We want to start with a CH4 budget for the last three decades, focusing on decadal variatons in the growth rate, regional/latitudinal budgets, and interannual variability (IAV). Can we compare regional budgets to UNFCCC-reported national/regional GHG emissions? Later on we will also try to shed light on the long-term budget since 1850 to help us better understand and explain the continuous growth rate on a longer time scale. Read more below the figures.

Fig 2: Regional methane budgets from top-down (light bars) and bottom-up (dark bars) approaches. Regional budgets were calculated for 13 regions (9 Transcom regions, plus separate regions for India, China, and SE Asia). Green: wetland, red: fire, brown: fossil fuel, blue: agriculture and waste. Kirschke et al. (in prep.)

Fig 3: Interannual variability (IAV) of all fluxes (from 11 inversions, Bousquet et al., 2011) as a function of latitude. Range of all inversions is shown as gray shaded area; individual inversions are shown as colored lines, reference inversion: black line. The last panel shows in addition the IAV of the OH loss. Kirschke et al. (in prep.)

Activity 1

The goal of this activity is to collect bottom up flux data and atmospheric data to close the global methane budget for different time periods.

1. Historical period (1850-1980) The evolution of the mix of different sources of CH4 will be compiled from the literature/references. The evolution of OH radicals is estimated from models (e.g., Prinn et al., 1995; Chen and Prinn 2005; Chen and Prinn, 2006; Krol et al., 1998; Krol et al., 2008). Atmospheric CH4 observations are derived from smoothed ice core data. Given the time smoothing in ice core CH4 observations, a similar time smoothing may need to be applied to the mix of sources and the OH sink. How can uncertainties be estimated (expert judgment? use of different estimations?)?

2. Modern period

2.1 The high growth rate period: 1980-1990 2.2 The low growth rate period: 1991-2006

Open questions: Is the CH4 cycle in equilibrium (i.e. sources in balance with sinks) on the long term, or do we merely see opposing trends in different sources (e.g. as suggested in Bousquet et al., 2006)? What is the role of the recent acceleration in the emissions of fossil CO2 on the CH4 growth rate? What is the role of recurrent drought on wetland and fire emissions?

2.3 The recent anomaly: 2007-ongoing

Open questions: What is the global magnitude? What are the links to climate anomalies? Which regions are important, which processes? For example, Dlugokencky et al. (2009) – tropical and northern (arctic) wetlands, higher than average temperature and precipitation.

  • Compile historical and modern budget global
  • Global total flux
  • Global error
  • Spatio-temporal distribution

Table 1 - Data for contemporary budget, 1980-1990: table1_data_contemp_budget_1980-1990.pdf

Table 2 - Data for contemporary budget, 1990-2006: table2_data_contemp_budget_1990-2006.pdf

Table 3 - Data for contemporary budget, > 2006: table3_data_contemp_budget_gt_2006.pdf

Activity 2

The goal of this activity is to run global inverse models using different inverse systems from various modeling groups. Within this activity annual updates of global CH4 inversion fluxes will be produced, i.e. global top-down estimates of CH4 emissions will be provided on a regular basis (annual, bi-annual?).

Open questions: At what scales and time resolution? How will atmospheric data and a priori information be gathered? Which models will be used?

  • Compile annual update of CH4 inversion fluxes
  • Inversion results global/regional/latitudinal bands
  • Global error estimates
  • Spatio-temporal distribution/flux maps

List of modeling groups (not exhaustive):

Please note: Bold indicates groups that have already provided their model results A * marks groups that have been contacted but have not yet provided results

  • JRC – Peter Bergamaschi (TM5, 4D-VAR inversion)*
  • LSCE – Philippe Bousquet (LMDZ, 4D-VAR & analytical inversions)
  • NOAA/ESRL – Lori Bruwhiler (Carbontraker CH4)
  • SRON – Sander Houweling (TM5)
  • KNMI – Michiel van Wheele (TM5, 4D-VAR)*
  • MIT – Chen / Rigby / Prinn (2D Box model, Kalman filter)*
  • UNIVERSITY EDINBURGH - Paul Palmer/Annemarie Fraser*
  • HARVARD – James Wang / Jennifer Logan (forward modeling) - not participating
  • + Regional inverse modeling : Met Office (A. Manning), ECN (A. Vermeulen),…
  • + probably other atmospheric chemistry groups

Activity 3 - to be discussed

The goal of this activity is to provide a product of estimated CH4 fields/an optimized atmosphere which could possibly be provided for forward-modeling efforts. Model data could be extracted for station locations and these extractions be provided to the experimental groups for feedback and analysis.

  • Compile optimized atmosphere fields and station location data
  • Assess uncertainties
  • Generate maps
start.txt · Last modified: 2011/09/15 13:22 by skirschk