Abstract Rivers are a dynamic source of greenhouse gases (GHGs), yet the temporal variability and controlling mechanisms of their CO 2 :CH 4 :N 2 O ratios remain poorly constrained. We monitored the three GHGs in the German river Elbe over 5 years at two sites to identify seasonal controls as well as travel time related and site‐specific mechanisms driving GHG concentration ratios and fluxes. CO 2 concentrations ranged from 2.8 to 125 μmol L −1 and showed a clear seasonal pattern with minimum values below saturation in summer, mainly correlating with indicators of planktonic photosynthesis, light availability, and chlorophyll concentration. CH 4 concentrations ranged from 0.0006 to 0.85 μmol L −1 and showed an opposite seasonal dynamic with maximum values in summer, correlating with temperature and particulate organic carbon. N 2 O concentrations were between 0.007 and 0.07 μmol L −1 , mostly near saturation and mainly determined by temperature dependent solubility. We did not observe large differences between the two study sites except for elevated CH 4 concentrations at the downstream site during summer. While CO 2 was regulated by the metabolic balance of the water column, CH 4 was more locally controlled, probably by hydrodynamic conditions affecting particle sedimentation. The total GHG‐potential of the three gases in terms of CO 2 equivalents was dominated by CO 2 and its seasonal cycle. Higher CH 4 emissions during summer were compensated by CO 2 uptake. Data‐driven models based on machine‐learning methods revealed that in the Elbe River, it is probably possible to predict GHG concentrations based on seasonal indicators without the need for water quality parameters.
Koschorreck et al. (Fri,) studied this question.