oronto’s new Ripley’s Aquarium of Canada is the largest facility of its kind in the country. Built on three-levels, this 12,500 m2 attraction is home to more than 15,000 aquatic animals, held in 50 tanks, containing nearly six million litres of water.
HGC Engineering was retained by Ripley Entertainment to provide comprehensive acoustical consulting for the Aquarium complex, which opened in October of 2013. The facility is enviably situated at the base of Toronto’s famed CN Tower, in the city’s downtown core. It is also situated directly to the south of Toronto’s central train station and its major rail corridor.
During the project’s planning and design phase in 2010, Ripley’s animal husbandry department, (which strictly adheres to the standards for animal welfare regulated by the Canadian Association of Zoos and Aquariums) expressed some concerns that ground-borne perceptible vibrations and both ground-borne and airborne sound transmission from nearby rail traffic could pose a potential risk to the well being of the 450 different aquatic species to be housed in the aquaculture tanks; in particular acoustically sensitive sharks and “bony fish”.
As a precaution, Ripley asked HGC Engineering to undertake a study that included investigating and identifying the threshold sound or vibration levels that might adversely affect these aquatic species in terms of their health, behaviour and reproductive capabilities.
In 2010, HGC Engineering visited the yet undeveloped site and conducted measurements of the ground-borne vibration and outdoor sound levels. Vibrations from various trains were measured, including passenger and freight trains that passed through the rail corridor. The vibration produced by the trains were found to be intermittent throughout the day and night. When vibrations from train pass-bys were at a maximum, they were dominated by rumble in the 63 Hz octave band.
TRAIN TRAFFIC IMPACT ON THE AQUARIUM AND ITS SEA-LIFE
Ripley had informed us that the aquarium tanks would house sharks and various other species of bony fishes. The hearing range of bony fishes varies greatly with species. Ambient noise levels in the ocean generally peak at around 100 dB1 at lower frequencies, which is near the frequency at which the vibration levels from the trains would be produced. 
Fishes are subdivided into two categories, hearing generalists and hearing specialists. Hearing specialists have a lower hearing threshold than hearing generalists, making them more susceptible to noise in their environment. Hearing generalists generally detect sound at levels no higher than 1 to 1.5 kHz, whereas specialists are generally able to detect sounds above 1.5 kHz. 
Studies determining the effects of long term exposure to noise have been performed on hearing generalists and specialists. During these studies, hearing generalists were exposed to moderate sound levels (170-180 dB), and little effect was found on the generalists. Hearing specialists were exposed to similar sound levels (170 dB) and suffered hearing loss. Hearing loss and recovery is dependent on duration of exposure; in test cases there was a 5 dB temporary threshold shift after only ten minutes of exposure . Several other studies have recommended that for long term exposure to noise, noise levels should be kept below 150 dB in order to eliminate hearing damage. 
Most published studies looking at the effects of noise and vibration on sharks used lemon sharks as the basis for their analysis, (as is the case with this report). Noise and vibration can be detected by sharks in two ways; with their ears and through the use of their lateral lines.
Auditory Responses of Sharks
Sharks are classified as hearing generalists, as they can generally hear noise up to only 1000 Hz. This is a general range which is applied to rays, sharks, and skates and is likely to be different depending on the specific species of shark (either increasing or decreasing the hearing range).
Studies performed by A. Peter Klimley and Arthur A. Myrberg, Jr. (1978) determined the acoustic stimuli underlying a withdrawal response from a sound source by adult lemon sharks. During this testing, broadband ambient noise levels were measured at 105 dB. The lemon sharks were then exposed to several different noise sources. If the shark retreated from the source upon being exposed to the noise, it was counted as a withdrawal response. The sharks exhibited withdrawals 33% of the time at noise levels which exceeded the ambient noise by 30 dB (ie 135 dB). The lemon sharks exhibited withdrawals 66% of the time at noise levels which exceeded ambient sound levels by 33 dB (ie 138 dB). 
During testing conducted by Banner and Hyatt (1973), the viability of eggs in one particular species of fish was reduced when exposed to noise of 105-120 dB at a frequency of 40-1000 Hz. These sound levels were maintained for several consecutive days . No studies were found on the effects on reproduction due to intermittent sounds.
Low Frequency Rumble and Lateral Lines of Sharks
Lateral lines are fluid filled tubes lined with tiny hairs, which are stimulated due to changes in fluid particle motion. The lateral line is used to determine current direction and the location of objects present in the water.
As a result, Ripley’s animal husbandry department was initially cautious and apprehensive about whether or not the lateral lines of the sharks could be affected by the low frequency rumble (63Hz) of the passing trains. However, while the lateral lines of sharks are sensitive to signals from below 1 Hz to several hundred Hz, the lateral line is most sensitive to stimuli which occur within several body lengths of the animal , and thus are more attracted to gross motion in the water as opposed to acoustical effects. We therefore did not anticipate that induced levels of rumble within the water of the tank would significantly affect the lateral line sensory system of the sharks.
ACOUSTICAL ASSESSMENT AND RECOMMENDATIONS
Noise levels in the water within aquarium tanks are generated by the movement of the boundary. As tanks are supported from the ground, ground-borne vibration will cause corresponding motion of the water at the bottom of the tank.
Resulting sound pressure levels in the water depend on various factors, such as the size of the tank and the depth of the water, but for the purposes of our study reverberation effects were discounted, and acoustic waves in the tanks were assumed to propagate as free plane waves at the frequencies of interest.
Noise and Vibration Impact on Sea-life Health and Well-being
Given these assumptions and the measurement of ground-borne vibration performed at the approximate location of the planned facility, noise levels within the tanks during train pass-bys were estimated to be approximately 150 dB at 31.5Hz. These noise levels are unlikely to affect hearing in fish species which are both hearing generalists, or hearing specialists; exposure to sound levels above 170 dB are clearly correlated to hearing damage in hearing specialist species.
Noise and Vibration Impact on Sea-life Behaviour
However studies involving lemon sharks showed that they are more sensitive to sound, responding to levels that are well below 170 dB. When lemon sharks were exposed to levels of 135-138 dB they withdrew from the noise source. These levels are considerably less than those which might result in hearing damage, but suggest that if the sharks were exposed to these noise levels, their behaviour would be affected at the very least.
Noise and Vibration Impact on Sea-life Reproduction
It can be assumed that the fish will likely interact with each other in the aquarium as they would in their natural environment and that reproduction is likely. While testing has shown that exposing fish eggs to sound levels between 105 and 120 dB reduces egg viability, this is for a constantly maintained noise level over several consecutive days. Our study revealed that the train noise would be intermittent; An average of 1.6 trains per hour pass by the location throughout the day lasting approximately 1 minute in duration. One of these trains is a freight train lasting approximately 8 minutes as opposed to the average 1 minute duration of passenger trains. Three trains per hour pass by the location throughout the night, lasting approximately 1 minute in duration. Therefore the noise from the pass-by will be maintained for one minute intervals throughout the day, with a single eight minute interval for freight trains. Behavioural reaction of the fish to such intermittent stimuli is not known, as they may or may not become conditioned to such sounds. Sharks do not produce eggs like other fish, they give birth to live offspring and as such the sound levels should not affect the viability of the offspring.
Based on all of the above information and findings, HGC Engineering proposed that low frequency noise intrusions from train pass-bys be limited to approximately 135 dB or lower. This would help ensure that behavioural reactions of the fish and sharks are minimized and that long-term hearing damage does not occur in any of the species. As a result of our recommendations, Ripley’s chose to be vigilant and err on the side of caution by incorporating vibration isolation features below the tanks that we had designed to help achieve the desired 135 dB or lower target.
Note: All sound levels referenced in this report represent the sound level in water, re 1 micropascal.
1. G. Vella, I Rushforth, E Mason, A Hough, R England, P Styles, T Holt, P Thorne, “Assessment of the effects of noise and vibration from offshore wind farms on marine wildlife”, ETSU W/13/00566/REP/A DTI/Pub URN 01/1341, © Crown copyright 2001
2. Arthur N. Popper, “Effects of Mid- and High-Frequency Sonars on Fish”, Contract N66604-07M-6056 Naval Undersea Warfare Center Division Newport, Rhode Island, February 21, 2008
3. Mardi C. Hastings and Arthur N. Popper, “Effects of Sound on Fish”, California Department of Transportation Contract No. 43A0139, Task Order 1, January 28, 2005
4. A. Peter Klimley and Arthur A. Myrberg, Jr., “Acoustic Stimuli Underlying Withdrawal from a Sound Source by Adult Lemon Sharks”, Bulletin of Marine Science, 29(4): 447-458, 1979