Climate change will shift mean environmental conditions and also increase the frequency and intensity of extreme events, exerting additional stress on ecosystems. While field observations on extremes are emerging, experimental evidence of their biological consequences is rare. Here, we introduce a mesocosm system that was developed to study the effects of environmental variability of multiple drivers (temperature, salinity, pH, light) on single species and communities at various temporal scales (diurnal - seasonal): the Kiel Indoor Benthocosms (KIBs). Both, real- time offsets from field measurements or various dynamic regimes of environmental scenarios, can be implemented, including sinusoidal curve functions at any chosen amplitude or frequency, stochastic regimes matching in situ dynamics of previous years and modelled extreme events. With temperature as the driver in focus, we highlight the strengths and discuss limitations of the system. In addition, we examined the effects of different sinusoidal temperature fluctuation frequencies on mytilid mussel performance. High-frequency fluctuations around a warming mean (+2°C warming, ±2°C fluctuations, wavelength=1.5 days) increased mussel growth as did a constant warming of 2°C. Fluctuations at a lower frequency (+2 and ±2°C, wavelength=4.5 days), however, reduced the mussels' growth. This shows that environmental fluctuations, and importantly their associated characteristics (such as frequency), can mediate the strength of global change impacts on a key marine species. The here presented mesocosm system can help to overcome a major short-coming of marine experimental ecology and will provide more robust data for the prediction of shifts in ecosystem structure and services in a changing and fluctuating world.
