Idiosyncratic phenology of greenhouse gas emissions in a Mediterranean reservoir

Extreme hydrological and thermal regimes characterize the Mediterranean biome and can significantly impact the phenology of greenhouse gas (GHG) emissions in reservoirs. Our study examined the seasonal changes in GHG emissions of a shallow, eutrophic, hardwater reservoir in Spain. We observed distinctive seasonal patterns for each gas. CH4 emissions substantially increased during stratification, influenced predominantly by the rise of water temperature and gross primary production and the drop in reservoir mean depth. N2O emissions mirrored CH4's seasonal trend, significantly correlating to water temperature, wind speed, and net primary production. Conversely, CO2 emissions decreased during stratification and displayed a quadratic, rather than a linear relationship with water temperature -an unexpected deviation from CH4 and N2O emission patterns- likely associated with calcite formation coupled to photosynthesis. This investigation highlights the need to integrate these idiosyncratic patterns into GHG emissions models, enhancing the prediction of global GHG emissions in the global change era.

Methods
This study was conducted at the eutrophic Cubillas reservoir in southern Spain, from March 2021 to July 2022. It focused on weekly monitoring of CO2, CH4, and N2O emissions, capturing both diffusive and ebullitive fluxes. Measurements of these greenhouse gases were taken at the reservoir's surface using a Cavity Ring-Down Spectrometer (PICARRO G2508) connected to a floating chamber, with 4 to 6 readings recorded daily during daylight hours. In addition to greenhouse gas monitoring, the study also involved assessing environmental and biological factors that influence the seasonal patterns of these gases. This included measuring water temperature, oxygen concentration, depth, and wind speed. Furthermore, nitrate levels, Gross Primary Production (GPP), respiration (Res), and Net Ecosystem Production (NEP) were also systematically measured and analyzed. Finally, a multiple mixed linear model approach was employed to identify the primary drivers of greenhouse gas (GHG) emissions.