Dissolved oxygen isotope modelling refines metabolic state estimates of stream ecosystems with different land use background

• Published in: Scientific reports Vol. 12; no. 1; p. 10204
• Main Authors: Piatka, David R., Venkiteswaran, Jason J., Uniyal, Bhumika, Kaule, Robin, Gilfedder, Benjamin, Barth, Johannes A. C.
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 17.06.2022; Nature Publishing Group; Nature Portfolio
• Subjects: 704/172; 704/47; Dissolved oxygen; Ecological monitoring; Ecosystem services; Ecosystems; Environmental changes; Gas exchange; Groundwater; Humanities and Social Sciences; Isotopes; Land use; Metabolic rate; Metabolism; multidisciplinary; Natural streams; Nutrient availability; Nutrient concentrations; Nutrient sources; Oxygen isotopes; Photosynthesis; Rivers; Sampling; Science; Science (multidisciplinary); Stable isotopes; Urban agriculture
• ISSN: 2045-2322; 2045-2322
• DOI: 10.1038/s41598-022-13219-9

Dissolved oxygen (DO) is crucial for aerobic life in streams and rivers and mostly depends on photosynthesis (P), ecosystem respiration (R) and atmospheric gas exchange (G). However, climate and land use changes progressively disrupt metabolic balances in natural streams as sensitive reflectors of their catchments. Comprehensive methods for mapping fundamental ecosystem services become increasingly important in a rapidly changing environment. In this work we tested DO and its stable isotope ( 18 O/ 16 O) ratios as novel tools for the status of stream ecosystems. For this purpose, six diel sampling campaigns were performed at three low-order and mid-latitude European streams with different land use patterns. Modelling of diel DO and its stable isotopes combined with land use analyses showed lowest P rates at forested sites, with a minimum of 17.9 mg m −2  h −1 . Due to high R rates between 230 and 341 mg m −2  h −1 five out of six study sites showed a general heterotrophic state with P:R:G ratios between 0.1:1.1:1 and 1:1.9:1. Only one site with agricultural and urban influences showed a high P rate of 417 mg m −2  h −1 with a P:R:G ratio of 1.9:1.5:1. Between all sites gross G rates varied between 148 and 298 mg m −2  h −1 . In general, metabolic rates depend on the distance of sampling locations to river sources, light availability, nutrient concentrations and possible exchanges with groundwater. The presented modelling approach introduces a new and powerful tool to study effects of land use on stream health. Such approaches should be integrated into future ecological monitoring.