Rivers are important contributors to global greenhouse gas (GHG) exchange with the atmosphere. However, much less is known about biogeochemical dynamics in rivers when they dry, particularly in isolated pools created by drying. Our objective was to examine the effects of water temperature and allochthonous organic matter (OM) quantity on carbon dioxide (CO2) and methane (CH4) fluxes in isolated pools. We used an automated analyzer to measure CO2 and CH4 from 36 mesocosms filled with sediments and water from a non-perennial river, with temperature (20, 25, 30 °C) and Alnus glutinosa leaf litter (2g, 5g, 10 g) manipulations in triplicate. We found positive individual effects of water temperature and OM quantity on CO2 fluxes, and a synergistic effect of water temperature and OM on CH4 fluxes during the late stages of the incubation. Given the increase in water temperature and OM inputs in rivers associated with climate change, our results indicate an associated increase in CO2, and a disproportionate increase in CH4 fluxes to the atmosphere, potentially contributing towards a positive climate feedback loop.

Stream metabolism is a key biogeochemical process in river networks, synthesising the balance between gross primary production (GPP) and ecosystem respiration (ER). Globally, more rivers and streams are drying due to climate change and water abstraction for human uses and this can alter the organic carbon residence time, leading to decoupled ER and terrestrial organic matter supply. While the consequences of drying on CO2 emissions have been recently quantified, its effects on stream metabolism are still poorly studied. We addressed the short- and long-term effects of drying on stream metabolism by monitoring oxygen dynamics at 20 reaches across a drying river network, including perennial (PR) and non-perennial reaches (NPR) for one year. We also calculated several climatic, land use variables and characterized and local abiotic conditions, and biofilm and sediment communities at five sampling dates. ER was significantly higher in NPR than in PR reaches demonstrating in-situ the effects of drying on stream metabolism. When analyzing the long-term drivers of ER and GPP, we found a direct positive effect of drying on ER and a negative effect on GPP. Drying also altered microbial community composition, with algal communities from NPRs being different from those in PRs. In the short-term, the amount of C emitted during rewetting events was positively related to the duration of precedent non-flow period. Our results show that drying had an important effect on stream metabolism both in the short and long term, and supports the need of including NPRs in global estimates of stream metabolism.