Entwicklung und Test einer robusten und zuverlässigen Feldmethode zur in-situ Messung gelöster Gase in Fließgewässern
Zusammenfassung der Projektergebnisse
The reliable estimation of dissolved-gas concentrations in streams is paramount for the assessment of ecological functioning, among others in the context of the quantification of metabolic rates. This relies on an accurate and precise estimation of oxygen reaeration rates, which can be achieved through artificial gas-tracer tests. In these, a gas tracer is continuously injected into the stream, and its concentration is recorded at several measurement stations downstream of the injection. From the gas loss observed along the reach, gas exchange rates of the tracer gas and reaeration rates of oxygen can be estimated. These need to be precise, as all uncertainties of the reaeration rate coefficients carry over to the estimation of the rates of respiration and primary production. A primary source of uncertainty in the assessment of gas tracer tests is the collection of gas samples, which are analyzed ex-situ. Gas loss may occur during sampling, storage, transport and analysis of the samples, leading to high uncertainties in calculated reaeration rate coefficients. To overcome these problems associated with the ex-situ analysis of gas samples, we built two battery-powered, field-deployable gas-equilibrium membrane-inlet mass spectrometers (GE-MIMS). These had been designed for quasi-continuous measurements of dissolved atmospheric gases in groundwater, and we adapted them for the application to gas-tracer tests in streams. This required extensive testing in the laboratory, including the selection of adequate gas-tracer compounds and the development of a calibration technique for propane, which cannot be calibrated using ambient air (contrary to noble gases), since it is not present in the atmosphere. We conducted gas-tracer tests in a number of smaller streams near Tübingen, where we successfully injected and measured krypton and propane as tracer compounds. We additionally recorded argon concentrations to verify the equilibrium between the stream water and the atmosphere. Furthermore, normalization of propane signals to those of argon allowed us to obtain relative concentrations of propane between measurement stations. From the recorded concentrations of propane and krypton, we were able to determine reaeration rate coefficients of oxygen with much lower uncertainty than through standard headspace sampling techniques, validating that in-situ methods lead to more reliable results than ex-situ methods.
