Relationship between ecosystem respiration and aeration constant in open channel dissolved oxygen analysis

Abstract:

Using the open channel diel method of Odum (1956) and the night-time regression method (Hornberger and Kelly, 1985), we analysed a time series of dissolved oxygen (DO) in two slow flowing streams for a two month period in summer 2014 and obtained values for ecosystem respiration and the aeration constant for each day in the period. We then used the standard dissolved oxygen lumped model to generate a DO time series behaviour for one of those rivers selecting respiration and aeration parameters by randomly sampling from the values obtained from the data. Two synthetic time series were created, one where respiration and aeration were independent of temperature and a second where respiration and aeration were affected by temperature according to the modified Arrhenius relationship. With these two synthetic time series, we again recovered the respiration and aeration input parameters using the night- time regression method and compared those recovered parameters with the input parameters. Because the simulations were conducted with parameters that were known, the values recovered using the night-time regression method (i.e post-simulation) could be compared with parameters driving the simulation (i.e. pre-simulation input values). For values based on data, we found a strong correlation between the aeration constant and respiration for both rivers. For the synthetic time series, no such correlation was found, either with the temperature independent or temperature dependent time series. The night-time regression method also recovered perfectly the input parameters, so the correlation was not brought about as a result of implementing the method itself. We are currently investigating the cause of the correlation.