This page describes the Natural Environment Research Council project, MOYA.

The Methane Observations and Yearly Assessments or MOYA project is a consortium of UK scientists funded by the Natural Environment Research Council. The collective aim is to close the global methane budget by undertaking new observations, model simulations and utilising existing data.

The key aims of the project are to:

  • achieve a radical improvement in the measurement of methane and its isotopologues in the atmosphere, to understand what changes are happening at global and regional scales.

  • understand why these changes are occuring through targeted field campaigns and the interpretation of atmospheric measurements using advanced modelling methods.

  • provide predictive power on how methane sources and sinks may change in the future.

Our measurement strategy focuses on integrating data from multiple platforms (fixed surface monitoring sites, ships, aircraft and satellites), making use of new measurement technology, particularly with regards to methane isotopologues, and establishing new long-term monitoring stations in data-poor (but key) regions of the globe. Dedicated field campaigns will close gaps in our understanding of source and sink magnitudes for specific major but poorly quantified processes (e.g. tropical wetlands), and how they respond to change. Finally, these advances will be brought together through innovative modelling methods at global and regional scales, to reapportion sources, attempt to close the methane budget, and identify remaining uncertainties. A framework for collaborations across groups is outlined in the figure below. This project, which is very ambitious but carefully targeted and costed on long experience, aims to achieve a major step forward in our understanding. It is also designed to dovetail closely with the work of partner groups worldwide. The project will build a coherent greenhouse gas observation network partnership, Equianos, able to make a substantial contribution to global measurements, to support the ongoing task of tracking the global methane budget in the future. Given the historic importance of methane in changes in global climate, we need urgently to improve our understanding of methane to predict more accurately its future role in climate change.

Project flowchart outlining interactions within the MOYA project.

Our purpose is to tackle the major unknowns in the methane budget. The fundamental goal is to achieve a more accurate understanding of the global methane budget, using improved observations and observation networks, targeted process studies and state-of-the art modelling of methane in the global atmosphere. We aim:

  1. To make Observations of methane that will support spatial and temporal quantification of the atmospheric burden;

  2. To use Process Studies of methane source and sink processes by targeted field campaigns, to estimate fluxes to the air from land surfaces, especially in tropical wetlands, and to improve source characterisation through isotopic tracer measurement in the atmosphere;

  3. To integrate the observational data through numerical Modelling, improving the representation of processes in models, and thus interpreting past and present trends, including the causes of the dramatic recent growth and 'biogenic' isotopic shift.

To ensure that NERC research investment will bring the largest gains with respect to these objectives, the project will:
  • Characterise methane in the Atlantic sector through an improved in situ network. Currently, coverage is especially limited in the Atlantic. Our network will integrate with the US NOAA work and enhance it. The UK is ideally placed to lead this effort as it has sovereignty over all remote S. Atlantic islands, and a long-term interest in the Antarctic. The N-S axis fits closely to NERC capabilities, with its presence from Spitsbergen to Antarctica.

  • Improve observations of methane in the tropics. Team members have close institutional relationships at key areas in the tropics, including India, Bolivia, Amazonia and South East Asia. We will use these links to augment and create new tropical methane measurement capabilities.

  • Develop new isotopic measurements at remote sites to discriminate between sources and sinks. In particular, D/H measurements will fill a major gap in source understanding, and will help constrain the OH sink. For this, inexpensive, high-precision, high-sample-number measurements are needed. A central methane isotope facility will be developed to do this.

  • Carry out targeted field campaign studies in key areas to determine source fluxes, seasonality behaviour, and particularly to determine source isotopic signatures.

  • Use aircraft and satellite remote-sensing techniques to quantify sources and sinks, assessing seasonal, vertical and latitudinal methane distributions, and measuring reactive species in the tropical troposphere. The observations will improve our knowledge of the OH sink, a major unknown in the global budget. Satellite data will improve spatial coverage.

  • Using atmospheric chemical transport and inverse models, we will integrate the data streams from the observations to determine methane sources,by seasonality and fluxes, and to quantify the major sinks.

WP 1 Equianos Observation Network

Objective: to extend the measurement base, as specified in the Announcement.

WP 1.1 Atlantic continious measurement

All Atlantic sites will operate in situ CH4 and CO2 cavity ring-down instruments for high-frequency measurement. Ascension Island measurement, currently by RHUL Picarro 1301, will be upgraded to a 2300-series instrument. 6-gas calibration will be maintained and the air inlet and autosampler and dryer upgraded. E. Falkland Island (Sapper Hill) RHUL measurement will be upgraded by BAS to a 2300-series instrument and to achieve better automation. RRS JC Ross 2300-series instrument (RHUL) will be serviced by BAS. All these installations will terminate in early 2017 unless supported in this project. Halley Bay (BAS) and Cape Verde (Exeter/MPIJena) will be continued. Intercomparison will ensure compatibility with NOAA and French networks.

WP 1.2 Flask and bag network, for δ13CH4 and δDCH4

Isotopic support for global modelling radically to enhance the global methane isotopologue dataset. Arctic sampling will be daily flasks from Zeppelin (Spitsbergen) in collaboration with NILU Norway, and weekly flasks from Alert Canada, and Pallas Finland; Flask will also come from Ascension and Cape Point, S. Africa. Tedlar bags will be sampled regularly from Kjolnes Norway, Pallas Finland, Barra Scotland, Azores, Chacaltaya Bolivia, Kajjansi Uganda, Darjeeling India, and Hong Kong. Cape Verde flasks will be measured by Max Planck Inst, Jena.

WP 1.3 New sampling programme (UEA)

Shipboard sampling aboard the Cap San Lorenzo; in-situ for CH4, automated flask sampling every 10° latitude for δ13CH4 and δDCH4. Repeated Atlantic crossings (8 weeks return trip; continued for years 2-4) with in-situ data ‘binned’ into 9 virtual stations, every 10° latitude from 55°N (Germany) to 35°S (Argentina).

WP 1.4 Equianos Observation network support structure

Integrate the network and ensure compatibility with US NOAA and other international WMO Global Atmosphere Watch observations. a) Calibration facilities (UEA) will include supply of target gases and secondary standards. b) Intercomparison flasks (UEA) will be circulated to US NOAA, NOAA-INSTAAR, LSCE Paris, and NIWA NZ to maintain rigorous intercomparability. c) Data archiving (NCAS, UEA) to maintain and deliver data to international databases.

WP 2 Aircraft Observation and Satellite Retrievals and Validation

Objective: to perform aircraft observations and develop satellite products.

WP 2.1 FAAM aircraft sampling

Sampling will be conducted at six-monthly intervals across two years with flights along an Atlantic meridian spanning 100° of latitude: UK - Azores – Cape Verde - ITCZ – Ascension Island. Highprecision, 1 Hz in situ geo-located measurements of CH4, N2O, CO2, O3, CO, VOCs and NMHCs will be recorded along with samples of δ13CH4 and δDCH4. Together, this dataset will allow us to calculate the local oxidising capacity of air (OH) and apportion methane enhancements in air masses targeted and sampled from the aircraft. ARIES will provide rapid 3D mapping of trace gases including CH4 (Allen et al., 2014, 2015). Measurements will inform regional process studies of continental outflows of methane from Africa and S. America, providing snapshot case study datasets for spatial scaling and assessment. FAAM datasets will support TCCON validation at Ascension and validation of new and existing satellite instruments in WP 2.3.

WP 2.2 BAS aircraft

Continuous measurements of CH4, as well as air sampling for δ13CH4 and δDCH4, will be made on the BAS MASIN aircraft during two of its annual northbound transect flights (65°S to 50°N) from Antarctica to the UK. S. American wetlands will be overflown (at ~10,000 ft) during the tropical rainy season. The spatial heterogeneity of methane will be established as will regional background isotopic signatures. Specific methane source regions will be identified using back-trajectory modelling techniques, and isotopic signatures determined, providing regional context to parallel contemporaneous flux studies of WP4.1. The aircraft will cross the Andes near Chacaltaya, thus providing broad spatial context to those long-term measurements. Data will help constrain regional inversion studies (3.3.6).

WP 2.3 Development of satellite products

Development of GOSAT, Sentinel 5 Precursor (SP5), NIR, SWIR datasets through improved ground-based validation. To ensure the consistency between GOSAT and S5P and to rigorously characterize errors in the retrieved CH4, we will use observations from the TCCON network with a focus on the Tropics (e.g. Ascension or Manaus) and boreal regions. Two dedicated campaigns using a portable Bruker EM27SUN spectrometer (from University of Leicester) will be carried at Uganda (in conjunction with the field deployment of WP 4.2.1) to give validation opportunities in tropical Africa. Further validation opportunities in Namibia, Brazil and India are expected to become available from the EM27SUN -COCCON network (Frank Hase, KIT).

WP 3 Isotope services

Objective: to perform and develop isotope measurements.

WP 3.1 δ13CH4 measurements

Measurement of Tedlar bag and flask samples collected by partners at Observation sites and in Process Study campaigns. Samples will be analysed for δ13CH4 at RHUL. Currently RHUL provides measurements on large numbers of flask and bag samples from across Europe as Trans-National Activities in the EU-InGOS project (until end-2015). High-precision analyses are performed by GC-CF-IRMS (Fisher et al. 2006).

WP 3.2 Development of CF-IRMS D/H measurement system

Development of the δDCH4 instrument is state-of-the-art and therefore inherently higher risk. Successful instrument development will permit large gains in model source apportionment and uncertainty reduction. WP 3.1.1 Develop high-precision, high-sample number δDCH4 consortium facility. In Europe, δDCH4 in ambient air is only currently measured at Utrecht and Max Planck Inst. Jena. The new IsoFlow system built by Isoprime Ltd. (Manchester, UK) promises a major step ahead in fast, small sample δDCH4 analysis. RHUL will test the system for high precision δDCH4 measurement and develop an automated sample inlet for the analysis of bag samples. WP 3.1.2 Measurement of samples from regular collection sites and field campaigns. Process study campaign samples (WP4) will give δDCH4 methane source signatures determination by 17 Keeling plot methods. Then δDCH4 measurement at observing stations will be applied to regional and global source apportionment in regional trajectory studies (between source areas and remote measurement sites). With large numbers of δ13CH4 and δDCH4 measurements generated in this project, plus ongoing US NOAA-INSTAAR and NIWA data sets, a genuine 3D global appreciation of the methane budget can be achieved.

WP 4 Field studies in key tropical regions

Objective: to determine fluxes, including seasonality, temperature response, and C3/C4 impact on isotopic signatures and soil uptakes.

WP 4.1 South American wetland flux studies

WP 4.1.1 Field Campaigns (OU, Rio University): 4 x 2-3 week campaigns at two seasonally flooded forest locations in the central Amazon, in the areas integrated by two of the regular flight locations. Timing will include high water, low water and intermediate stages to allow for flux measurements and δ13CH4 and δDCH4 characterisation of emissions under varying hydrological regimes. In addition, 2 x 3-week campaigns will be conducted in the peat swamps near Iquitos (OU, Aberdeen) and in the Pantanal during low and high water. WP 4.1.2 Regional integration of emissions from principal wetland types (Leeds, São Paulo). Biweekly CH4 measurement flights near Iquitos and in the Pantanal to supplement other projects' flights in 4 locations for regional integration of emissions from major S. American wetland types.

WP 4.2 African wetland flux studies

Activities will focus on wetlands in Uganda and Botswana, in 4 different Papyrus wetlands that are broadly representative of major wetlands in these countries, including: i) deep peat wetland (SW Uganda); ii) floodplain wetlands (Lake Kyoga or Lake Victoria, Uganda); iii) seasonally-flooded deltaic wetlands (Okavango Delta, Botswana); and iv) perennially-flooded deltaic wetlands (Okavango Delta, Botswana). WP 4.2.1. Across all 4 wetlands, we will use a unified multi-scale sampling approach at the field scale (eddy covariance for 1 year); plot scale (chambers, monthly or seasonal campaigns); isotope sampling from chambers and air (monthly Keeling plot sampling for δ13CH4 and δDCH4); and ecosystem process modelling (ECOSSE). Campaign-based chamber measurements will also be used to complement the eddy covariance measurements and further constrain estimates of dry season CH4 uptake. In Uganda, we will deploy a Picarro CRDS isotope analyzer to determine realtime δ13CCH4 variations coupled to concomitant eddy covariance fluxes on a campaign basis. WP 4.2.2 Road campaigns (RHUL) using bag/flask collection in Botswana (2 weeks), Zambia and Uganda. Keeling plots will be made to identify δ13CH4 and δDCH4 signatures, targeting key sources (wetlands, fires, cows) for which isotopic signatures are highly uncertain.

WP 4.3 Asian natural and anthropogenic emissions

WP 4.3.1 India (UBris) Bag/flask measured for δ13CH4 and δDCH4 bulk isotopic signatures of emissions from anthropogenic sources (diurnal studies) ( ruminant emissions, rice) WP 4.3.2 Malaysia (UCam, UEA, U Malaya). Use of Bachok Observatory (NE Malaysia). Regional studies using bag/flask δ13CH4 and δDCH4 for bulk isotopic signatures of emissions. WP 4.3.3 Kuwait (U Kuwait, RHUL) Bag/flask δ13CH4 and δDCH4 measured for bulk isotopic signatures from anthropogenic sources (oil and gas industry). WP 4.3.4 Hong Kong/China (HK Univ, RHUL) Bag/flask measured for δ13CH4 and δDCH4 bulk isotopic signatures (anthropogenic emissions from Chinese coal industry, rice).

WP 5 Methane source and sink estimation using observations and models

Objective: Use models constrained by observations to derive CH4 source and sink changes.

WP 5.1 Quantification of atmospheric methane sinks

a) Prediction of OH concentration and variability b) Evaluation of OH fields in forward and inverse models c) Quantification of tropospheric Cl sink in global methane models. Simulations of the magnitude and variability of the major methane sink, OH, using the “full chemistry” models UKCA, GEOS-Chem and TOMCAT will be carried out. Fields derived from these models will be tested and constrained (off-line) in MOZART, GEOS-Chem and TOMCAT inverse frameworks using observations of CH4, δ13CH4, δDCH4 and CH3CCl3. Sensitivity studies of the impact of OH chemistry on these species will use off-line chemistry in the UKCA. Independent estimates of OH variability will be obtained using a 2D model and AGAGE measurements of CH3CCl3 (Rigby et al., 2013), improved in this project by adding an interactive ocean to the 2D model. Potential systematic errors in the CH3CCl3 emissions model on the outcome of OH inversions will be investigated using a hierarchical Bayesian inverse framework. Tropospheric Cl distribution will be quantified using the TOMCAT 3D CTM, which has a detailed description of tropospheric chlorine chemistry. The impact of halogen chemistry on OH will be investigated. TOMCAT fields of Cl will be provided for forward and inverse model simulations of methane and its isotopologues to identify the role of uncertainties in Cl oxidation on derived methane fluxes.

WP 5.1 Development of methane isotopologue models

a) Re-evaluation of global isotopic source ratios; b) Forward model simulations of CH4, δ13CH4 and δDCH4; c) Sensitivity experiments to quantify the influence on observation network of variability from different sources and regions; d) Quantification of the impact of tropospheric Cl on the isotopic fractionation of atmospheric methane. A thorough re-evaluation of δ13CH4 and δDCH4 source ratios will be carried out to improve representation in models, including spatial and temporal gradients (from WP 5.3 and existing datasets). Reference simulations with the UKCA model (Telford et al., 2010) will be carried out, with CH4 sources partitioned according to previously used δ13CH4 and δDCH4 source ratios (Rigby et al., 2012). Methane will be simulated both online within the model’s chemistry scheme and offline using regional and source-type tagged methane tracers and prescribed OH and Cl distributions from WP 5.1. Model simulations will be compared with observation data from this project and partners (NOAA, MPI). δ13CH4 and δDCH4 source ratios, including new findings from WP 4, will be used to develop geographically and seasonally varying isotopic source ratio maps. New flux inventories will be run through the model to quantify improvements in the spatial and temporal gradients in 13C/12C and D/H predicted by the model. Flux fields will be used as an input to the global inverse models. Using tagged tracers, sensitivity assessments will be performed of distributions of CH4, δ13CH4 and δDCH4 in UKCA to changes in fluxes from each of the major sources, including wetlands, fossil fuel and biomass burning. A variety of wetland emission scenarios will be considered, including JULES and those that participated in the WETCHIMP model comparison (Melton et al., 2013).

WP 5.3 New estimates of wetland methane flux

WP 5.3.1 JULES simulations of wetland flux (CEH, Met Office) Objectives: a) Targeted development of the JULES wetland methane scheme; b) Simulation of JULES wetland methane emission datasets for use in the atmospheric models. Improved parameterisation of the Q10 temperature response, controlling the magnitude of the methane flux, will be made using the new measurement data from the tropics derived in WP4 and existing datasets from the Alaskan Arctic (Oechel, OU). The need for explicit representation of release of methane from forested wetlands will be investigated. Addition of wetland methane isotope functionality to JULES using data from the process-based studies (WP4) and information on soils and vegetation (e.g., 13C/12C and D/H ratios of the substrate carbon) will provide estimates of emission fluxes of δ13CH4 (i.e δ13CCH4 isotopic signature) and CH3D (as δDCH4). Modelled and observation-based wetland extents will be compared and the sensitivity of emissions to wetland extent uncertainties quantified. Emission estimates will be iteratively refined by comparing to top-down atmospheric modelling studies. WP 5.3.2 Inter-comparison of top-down and bottom-up wetland methane fluxes. (UBris, UEdin, ULeeds, UCam, CEH, Met Office) Objective: Observational constraints on wetland fluxes using multiple top-down methods and models at multiple scales. Forward model runs and sensitivity test will be performed by the UKCA model. Global inverse model estimates, with uncertainties, from MOZART, GEOS-Chem and TOMCAT, will be performed using hierarchical Bayesian MCMC, 4Dvar and EnKF inverse methods (described in Section 3.3.2 – 3.3.5); Regional model NAME simulations will be carried out for wetland areas close to monitoring sites (3.2.6). GRACE/GOSAT/S5P remotely sensed data will be used to constrain wetland flux estimates (Bloom et al., 2012 and Section 3.2.7). The work will deliver a set of estimates, using multiple techniques, of the magnitude and variability of wetland methane flux over a decadal timescale. Results will improve the JULES model, improve our understanding of the drivers of seasonal and inter-annual wetland flux variability, and provide a platform for better predictions of wetland flux.

WP 5.4 New estimates of anthropogenic and biomass burning flux

a) Provide a suite of top-down anthropogenic and biomass burning flux estimates using observations of CH4 and its isotopologues; b) develop space-based biomass burning tracer simulations. Top-down estimates at global and regional scales will be used to evaluate bottom-up estimates of anthropogenic and biomass burning emissions. The range associated with structural model errors will be estimated. Top-down estimates will benefit from multi-species observational constraints, including δ13CH4 and δDCH4, and space-based products. Satellite observations of land surface properties related to fire (data from S5P and GOSAT) will be used to determine observed variability of CH4 due to combustion using concurrent CO and/or HCHO measurements from 2016 onwards. Emission ratios of CH4:CO2 (GOSAT) and CH4:CO:HCHO (S5P) will be assessed using the method of Ross et al., 2013. IASI and GOSAT data will be incorporated in the GEOS-Chem EnKF inversion model framework to improve CH4 biomass burning flux estimates inferred from the space-borne sensors.

WP 6 Integration

WP 6.1 Integration across work packages

Objectives: a) Coordinate appropriate use of datasets, including uncertainties, for atmospheric modellers; b) Create NAME back-trajectory maps of all new observations to aid with data interpretation; c) Develop global and regional model products to aid data synthesis; d) Evaluate biases in remotely-sensed products for GOSAT and S5P using assimilated in situ and remote sensing data. Specifically, NAME model simulations for all new datasets will determine the origin of air masses for each observation point and partition data by source or meteorological conditions to develop estimates of “bulk” source isotopic signatures (WP 3). Global model simulations will be used to help with the interpretation of the observed latitudinal and vertical gradients within WP1 to reconcile remotely sensed “column” and in situ measurement data. In an effort to solve potential biases in the remote sensing data both the regional and global model inversions will incorporate in situ data with an additional benefit of providing an inter-comparison of both datasets.

WP 6.2 Integration with wider modelling community

Objectives: a) Provide full access to the datasets and all necessary supporting information for the appropriate use of the modelling community worldwide; b) Serve as a link between the UK and European Carbon Measurement Community (with ICOS as an external EU partner) to integrate the newly obtained datasets with those provided by our European partners; d) Create a task force involving the key project partners to oversee the production of a series of synthesis publications aimed at interpreting and summarising the new observational findings from WP1 to 4 to make them accessible to the wider scientific community over the course of the project.

Moya @ Leeds Website 2017