Experimental set-up


Water column model with a thermal gradient

The development and construction of two tanks (one with a thermocline and CIL and one without a temperature gradient to serve as control) began in 2002 at the Maurice Lamontagne Institute (MLI).


Water Column Experimental Model Thermally Insulated Glass Panel

The column is a vertical structure of about 1.6 m in height with an internal horizontal cross section of 20.3 x 30.4 cm and is made of clear 1.9 cm thick Plexiglas®.

The column is made of two parts: a moulded section, bent at high temperature, that forms the sides and the back, and a flat section glued to the moulded section to make the front. A thermally insulated glass panel is affixed to the front to avoid surface water condensation and to facilitate the observation of organisms inside the column (B).

An 11.5 cm diameter opening at the base of the column allows access to the inside for cleaning the bottom and, eventually, will be used for the attachment of a mechanism allowing the introduction of organisms at the base of the column.

Heat Exchanger

Heat Exchanger

One of the key elements of the system that creates a stable thermocline and CIL was the conception, the construction and the installation of a heat exchanger in the central section of the column. The heat exchanger is made of a 9.5 cm diameter stainless steel tube. To cool the water at that level, a non toxic anti-freeze fluid cooled at -1.2°C is continuously circulated in the exchanger using a thermostatically controlled circulating water bath.

A standard 400 Watt laboratory hot plate, set at about 92°C and placed a few centimetres from the bottom in a space located under the column heats the water at the base and creates the 'deep warm layer' below the CIL.


Lighting system

Most planktonic organisms, including young stages of crustacean decapods such as the Northern shrimp, react strongly to the visible light (between about 450 and 600 nm). That phenomenon is called phototaxis. This reaction seriously complicates behavioural studies of organisms facing other environmental factors such as a thermal gradient. To allow for the observation of the larval stages without influencing their behaviour, a red light source (maximum emission at 665 nm, power of 6.3 µW cm-2 at the source) is placed behind the back of the column.

Light Emission Spectrum

Emission spectrum of the column's back wall light source. It was shown that crustacean decapod larvae are not sensitive to wavelength higher than 600 nm (Cronin and Jinks, 2001; Johnson et al. 2002).


At intervals of about 18 cm along the entire height of the water column, 8 digital temperature sensors (1-Wire Dallas Semiconductors DS18B20) are positioned at the centre of the horizontal cross section. The 8 sensors form a parallel network and are linked by telephone-type connectors to the reading/display/recording device (accuracy 0.1°C in the range -50 to 150 °C) on a personal computer. A user interface (ZOC for Windows™ interface, BMT Micro INC, Wilmington, NC. USA) controls temperature monitoring and recording.


During the spring of 2004 experiments, the system was located at an ambient temperature of 6°C in a controlled atmosphere unit (CAU) at the Maurice Lamontagne Institute (MLI). Preparation of the thermal profile begins at the end of the day preceding an experiment by the cleaning, filling (sea water filtered at 30 µm) and aerating of the water in the column. The morning of an experiment, work begins with circulating the anti-freeze fluid cooled at -5.5°C in the heat exchanger until the water temperature at that level reaches 0°C or less. The temperature of the fluid in the exchanger is then set at -1.2°C.

The temperature in the CAU determines the surface layer water temperature (about 6°C), the heat exchanger creates a thermocline and a CIL where the minimum water temperature is maintained at about 0.5°C, and the heat source under the bottom of the column produces a deep warm layer where temperature varies between 3 and 4°C.

Once established, water temperature profiles at each level remain very stable for the duration of an experiment.


Temperature Recorded in the Column


Temperatures recorded by each of the 8 sensors located at intervals of about 18 cm along the height of the column. The graph shows the high stability of the temperatures recorded at each level during an experiment. Larger fluctuations are recorded near the bottom due to the hotplate intermittent operation.

The simulated vertical temperature profile also displays all the characteristics of a natural profile as observed in the spring in the Northern Gulf of St. Lawrence.


Comparison of the Mean Water Temperature Profile

Comparison of the mean water temperature profiles. Left: mean and standard deviation for all simulations done in the laboratoratory in the spring of 2004 (number of observations = 20). Right: temperature profile recorded at a station in the Northern Gulf of St. Lawrence in the spring 2002. Proportionally, the surface layer is more important in the laboratory experimental column however the three-layered typical profile observed at sea is well represented.