NERC Highlight MOYA (Methane Observations and Yearly Assessment)

Methane (CH₄) is a potent greenhouse gas (GHG) with 25 times the global warming potential of carbon dioxide (CO₂) and is the second-most abundant GHG after CO₂.

After almost a decade of stagnation, atmospheric methane has been rising steadily since 2007 but the causes for this renewed increase are still largely unknown. The increase in global concentrations has been accompanied by a shift in isotopic signatures, which suggests that biogenic sources (rather than anthropogenic ones) are driving the rise in atmospheric methane.

Emissions from tropical wetlands and agriculture are believed to be the major drivers behind the renewed increase in atmospheric methane but there are currently insufficient measurements in the tropics to assess the magnitude of these emissions and to understand how these might respond to changes in environmental and meteorological conditions.

There is hence a timely need to identify the biogenic sources of methane and their controls, study the processes underpinning them and quantify their emissions in order to better predict their future impact on climate and identify possible mitigation options.

MOYA (Methane Observations and Yearly Assessment) is a NERC-funded Highlight project aimed at improving the measurement and modelling approaches of methane in the atmosphere and better constraining the global budget.

The four-year project, which began in 2016, brings together a consortium of 14 UK universities and research institutes and sets out an ambitious measurement programme at multiple spatial scales (from process studies at the plant scale to earth observation) across a broad latitudinal range spanning the Artic to the Antarctic.

The GHG team at CEH's Edinburgh site, led by Prof Ute Skiba in collaboration with scientists from the Okavango Research Institute (University of Botswana, Maun), is responsible for the quantification of methane emissions and the study of their environmental controls in two contrasting wetlands in the Okavango Delta, Botswana.

The Okavango, a UNESCO World Heritage site located in north-western Botswana, is one the world's largest inland deltas; it does not have an outlet into the sea and its waters flow into the Kalahari desert. The Delta is fed by the Okavango River, which receives its waters from the Angolan Highlands, but due to the low topographical gradient, peak flood occurs during the dry season. The wetlands of the Okavango delta are in pristine condition and can be separated into three categories: permanently flooded, seasonally flooded (3-6 months per year) and occasionally flooded (typically once per decade). 

Because methane is produced under anaerobic conditions and emissions are linked to soil moisture, our measurements efforts are concentrated in the permanently- and the seasonally flooded wetlands. 

1. Permanent wetland (Guma Lagoon)

We established an eddy-covariance (EC) measurement site on the western shore of Guma Lagoon to capture methane emissions from an extensive papyrus bed, which stretches several hundred meters from the shore towards the open water.  

Eddy-covariance is a powerful technique which allows for in-situ measurements of spatially-integrated fluxes of mass, energy and momentum over large areas. Fluxes are derived from the covariance of the vertical wind speed and the concentration of the compound of interest (in the case of gas fluxes) measured at high frequency (>= 10 Hz) and averaged over a longer period (typically 30 minutes). 

The eddy-covariance instrumentation consists of a Campbell Scientific IRGASON (open-path Infrared Gas Analyser for CO₂/H₂O integrated within the head space of a CSAT3 ultrasonic anemometer) and a Licor 7700 open-path CH₄ analyser mounted atop a walk-up tower giving us a measurement height of 5.5 meters above ground. In addition, we collect meterological variables such as PAR (Photosynthetically Active Radiation; Skye Instruments Quantum sensor), total radiation (Skye Instruments pyranometer), relative humidity, ambient pressure and temperature (Vaisala WXT520 weather station).

2. Seasonal wetland (Chief's Island)

An eddy-covariance system, identical to the one at Guma Lagoon, was installed at Nxaraga, on the south-western edge of Chief's Island, overlooking a seasonal floodplain. The vegetation of the flooplain is dominated by species such Cyperus articulatus L., Schoenoplectus corymbosus (Roth ex Roem. & Schult) and grasses such as Panicum repens L and Oryza longistaminata. In contrast, the sparse inland vegetation consists mainly of grasses and small shrubs and the soil is sandy and salt-custed.

In addition to the flux measurements by eddy-covariance, the exchange of CH₄ and CO₂ betwen soil/vegetation and air is also measured by closed, clear chambers connected to an ultra-portable gas analyser (Los Gatos Inc.). This allows for the study of emission dynamics at smaller spatial scales compared to EC and a closer investigation of the effects of the changing hydro-meteorological conditions on the measured fluxes. The locations of the closed chambers are chosen so as to measure along a soil moisture gradient in order to evaluate the seasonal variations in emissions in response to the gradual drying out of the floodplain. 

The transect measurements by clear dynamic chambers are carried out monthly by our collaborators at ORI (led by Dr Mangaliso Gondwe) in order to capture the seasonality of the emissions. ORI scientists also collect soil and water samples which are analysed for dissolved inorganic and organic carbon (DIC & DOC), total phosphate, total nitrogen, soil organic content, pH, electrical conductivity and texture.

Measurements are ongoing but initial observations revealed contrasting spatial patterns of methane emissions at the two sites as well as large differences in emissions intensities.

• At Nxaraga, the seasonal floodplain, we observed an  emissions gradient, with near-zero or even negative (i.e. methane destruction) fluxes in the areas of sandy soil and larger emissions in the floodplain. The gradual increase in methane emissions away from the island is consistent with the spatial and seasonal distributions of soil moisture within the floodplain, with larger emissions correlating with wetter soils. These results, obtained by eddy-covariance, are corroborated by the smaller scale chamber measurements which suggest that soil moisture is probably the dominant control of methane emissions at this site. As the soil dries out methane production (methanogenesis)  will decline and/or an increase in methanotrophic activity will  enhance methane oxidation.   
• The picture is very different at Guma lagoon, where we observed spatial hotspots characterised by very large emissions (> 5-fold the largest emissions measured at Chief's Island) within the floating papyrus bed and the open water areas of the lagoon. Further investigation is necessary to interpret these emission patterns and understand their environmental controls which probably include plant-mediated transport, diffusion at the water-air interface, bubbling and convective flow within the water column.