Whole-stream metabolism: strategies for measuring and modeling diel trends of dissolved oxygen

Abstract

Stream metabolism is used to characterize the allochthonous and autochthonous basis of stream foodweb production. The metabolic rates of respiration and gross primary production often are estimated from changes in dissolved O2 concentration in the stream over time. An upstream–downstream O2 accounting method (2-station) is used commonly to estimate metabolic rates in a defined length of stream channel. Various approaches to measuring and analyzing diel O2 trends have been used, but a detailed comparison of different approaches (e.g., required reach length, method of measuring aeration rate [k], and use of temperature-corrected metabolic rates) is needed. We measured O2 upstream and downstream of various reaches in Kings Creek, Kansas. We found that 20 m was the approximate minimum reach length required to detect a significant change in O2, a result that matched the prediction of a calculation method to determine minimum reach length. We assessed the ability of models based on 2-station diel O2 data and k measurements in various streams around Manhattan, Kansas, to predict k accurately, and we tested the importance of accounting for temperature effects on metabolic rates. We measured gas exchange directly with an inert gas and used a tracer dye to account for dilution and to measure velocity and discharge. Modeled k was significantly correlated with measured k (Kendall's τ, p < 0.001; regression adjusted R² = 0.70), but 19 published equations for estimating k generally provided poor estimates of measured k (only 6 of 19 equations were significantly correlated). Temperature correction of metabolic rates allowed us to account for increases in nighttime O2, and temperature-corrected metabolic rates fit the data somewhat better than uncorrected estimates. Use of temperature-correction estimates could facilitate cross-site comparisons of metabolism.