Spatial-Temporal Resolution of Dissolved Ch4 Concentrations and Fluxes from Different Freshwater Types in an Agricultural Irrigation Watershed in Eastern China

Inland freshwater ecosystems are of increasing concerns on global methane (CH4) budget in the atmosphere. Agricultural irrigation watershed is a potential CH4 emission hotspot owing to the anthropogenic carbon and nutrient loading. However, large-scale spatial variations of CH4 concentrations and fluxes in agricultural catchments remain poorly understood, constraining an accurate regional estimate of CH4 budgets. Here, we examined the spatial-temporal resolution of dissolved CH4 concentrations and fluxes from different freshwater types (ditch, reservoir and river) in an agricultural irrigation watershed from Hongze catchment, under intensive agriculture and rural activities in Eastern China. Annual mean CH4 concentration and flux were 0.12 μmol L-1 and 0.58 mg m-2 d-1, respectively. Total CH4 flux from this agricultural irrigation watershed was estimated to be 0.002 Gg CH4 yr-1. Diffusive CH4 flux varied in samples taken from different freshwater types, the annual mean CH4 flux for ditch, reservoir and river was 0.31±0.06, 0.71±0.13 and 0.72±0.25mg m-2 d-1, respectively. Midstream (reservoir) and downstream (river) waters are hotspots, in which the CH4 production rates were two times greater than in non-hotspot locations. The dissolved CH4 concentrations and fluxes in three water types showed similar temporal variations, with the highest levels in summer and the lowest levels in winter. Water velocity and wind speed are important in explaining variable of dissolved CH4 concentration and flux. The responses of CH4 fluxes to water dissolved oxygen (DO), nitrate nitrogen (NO3--N) and sediment dissolved organic carbon (DOC) concentrations were significantly stronger in reservoir and river than in ditch, indicating that intensive human activities may affect the CH4 fluxes from different freshwater types. These results highlighted a need for more field measurements with wider spatial coverage and finer frequency, which would further improve the reliability of flux estimates for assessing the contribution of agricultural watersheds to the regional and global CH4 budgets.