Umeå universitets logga

Ongoing climate change is projected to extend the warmer and therefore the biologically productive season, reducing ice cover, ice thickness, and quality, potentially influencing biodiversity, and productivity of aquatic ecosystems. Changed influence of dissolved organic matter is one factor that can contribute to those effects. Winter ecology is little studied, and the advancement of knowledge would benefit from controlled experiments on the mesocosm scale. To investigate the capability of mesocosm experimental infrastructures for winter ecological research, a 5-months long experiment during the sub-arctic winter in 2021/2022 was conducted in Umeå, Sweden. Simultaneously, the performance of an outdoor and indoor mesocosm facility with ice-forming capability at the same site was compared. Boundary conditions for hydrographic, chemical, and biological variables were determined.The facilities were operated successfully over winter and treatments caused similar effects in both systems, despite some differences presented below. Salinity and temperature were similar between the facilities throughout the experiment. Ice was markedly thicker on the sea compared to in the indoor facility. Further the ice inside the outdoor mesocosms, was significantly thicker than on the surrounding natural sea. Light irradiance indoors correlated with the outdoor facility, but light irradiance indoors could not reach the outside values in the lightest months of the experiment (after mid-March). Both dissolved organic carbon and dissolved nitrogen was higher in the outdoor facility, possibly caused by a pump effect increasing organic carbon and nitrogen concentrations. Most other nutrient levels remained similar. Chlorophyll-a was comparable between the facilities, while plankton respiration was twice the rate outdoors compared to indoors. Two substances were used to simulate browning, HuminFeed® (a commercially available leonardite) and soil extract, causing similar treatment effects in both facilities for 75% of measured variables. HuminFeed caused a marked increase in CDOM (coloured dissolved organic matter) and nitrite during spring. Treatment with soil extract resulted in slightly higher phosphorus concentrations.The indoor mesocosm facility was thus comparable to the outdoor facility regarding experimental effects, despite facility differences observed. The organic matter sources HuminFeed and soil extract differ in some experimental effects that need to be considered. These results should provide basic knowledge for improving experimental design in future winter mesocosm studies.