Temporal dynamics and environmental controls of carbon dioxide and methane fluxes measured by the eddy covariance method over a boreal river

Boreal rivers and streams are significant sources of carbon dioxide (CO₂) and methane (CH₄) for the atmosphere. Yet the controls and the magnitude of these emissions remain highly uncertain, as current estimates are mostly based on indirect and discrete flux measurements. In this study, we present and analyse the longest CO₂ and the first ever CH₄ flux dataset measured by the eddy covariance (EC) technique over a river. The field campaign (Kitinen Experiment, KITEX) was carried out during June-October 2018 over the river Kitinen, a large regulated river with a mean annual discharge of 103 m³ s^-1 located in northern Finland. The EC system was installed on a floating platform, where the river was 180 m wide and with a maximum depth of 7 m. The river was on average a source of CO₂ and CH₄ for the atmosphere. The mean CO₂ flux was 0.36 ± 0.31 μmol m^-2 s^-1, and the highest monthly flux occurred in July. The mean CH₄ flux was 3.8 ± 4.1 nmol m^-2 s^-1, and it was also highest in July. During midday hours in June, the river acted occasionally as a net CO₂ sink. In June-August, the nocturnal CO₂ flux was higher than the daytime flux. The CH₄ flux did not show any statistically significant diurnal variation. Results from a multiple regression analysis show that the patterns of daily and weekly mean fluxes of CO₂ are largely explained by partial pressure of CO₂ in water (pCO_2w), photosynthetically active radiation (PAR), water flow velocity and wind speed. Water surface temperature and wind speed were found to be the main drivers of CH₄ fluxes.