2.2. Experimental set up
The experiment was set up in six outdoor tanks (1.2 m wide, 2.4 m long and 0.4 m deep water) at Ruakura experimental facility, Hamilton, New Zealand. Detailed physical and operational description of the tanks is given in Burnett et al. (2007). The experiment was a 3×2 combination of temperature and light availability (+5 ◦C of ambient, −5 ◦C of ambient and ambient) and two light availabilities (25% and 50% of incoming light corresponding to 150–250 and 300–500 mol m−2 s−1, respectively). Light was manipulated using shading clothes. Temperature was maintained by drawing water past heating or cooling elements by an electric pump and recirculated into the tank at 1200 L h−1. In the low temperature treatment mean temperature was 20.6 ◦C (±1.8), in the intermediate 25.5 ◦C (±1.8) and in the high temperature treatment 30.0 ◦C (±1.7). The temperatures are hereafter called 20 ◦C, 25 ◦C and 30 ◦C according to mean temperatures during the experiment, and the light availability referred to as 25% and 50% light. These temperatures were chosen because they represent general differences between mean summer temperature in many temperate (20–25 ◦C) and subtropical freshwater systems (25–30 ◦C), e.g. in northern versus southern Europe and North Island versus South Island in New Zealand. In order to maintain same chemical conditions and thereby lower the risk of any unknown factors due to the study design of only one experimental unit, a small amount of water was continuously replaced in the tanks from a common header tank for all three tanks. The exchange rate of water for each tank was about 18 L h−1 (300 mL min−1) in each tank and was so low compared to the whole volume (2304 L) that it did not influence the temperature regime.
The water in each tank was then renewed approximately every five days. Furthermore, the system is set up to avoid any other factors such that, e.g. wind and animals cannot interfere with the experiment.