Language selection

Government of Canada / Gouvernement du Canada

Search

Integrated Hatchery Facility for the Optimization of Sustainable Land-Based Arctic Charr Production

Final Report
Millbrook First Nations
AIMAP 2012-M06

1.0 Executive Summary

Millbrook First Nation has been involved in the aquaculture sector since 2003, when it opened its land based Arctic charr facility in Truro, NS. Since that time numerous production and sales issues have been addressed to improve the sustainability of the operation. This project focuses on the design and operation of a new hatchery facility that would streamline fry production and facilitate maximum growth and production of Arctic charr.

The goal of this project was to design and construct a stand-alone fish hatchery based on innovative water treatment technology. 

The objective of this project was to improve production systems, operational efficiencies and health, and provide quality management. The project was divided into four (4) phases. The hatchery design and development of performance indicators (phase 1) was completed in May of 2012. Performance worksheets were developed to measure required inputs (power, labour, and feed) and expected outputs (healthy fish). Building construction and installations (phase 2) began in December 2012 and was completed in June of 2013. Operating and monitoring (phase 3) began in May of 2013. First stocking of eggs into the hatchery was on May 9, 2013. During the sac fry development period the systems' two chillers failed and temperatures were elevated for a period of 5 days. The performance results of these fish would not be representative so a second batch of eggs was stocked on August 12, 2013. The assessment and reporting (phase 4) was initiated in February 2013 with the baseline monitoring of 60,000 eggs reared in the old hatchery system (cohort 1). These performance results were compared with the data collected from eggs (13,000) stocked and reared in the new hatchery system from Aug to Nov 2013 (cohort 2).

In analyzing fish performance data between the two systems the results showed overall mortality, up to 45 days post feeding, was reduced by 3.8% and growth in the same time period increased 14%. Food conversion ratios (FCR) were similar for both cohorts with cohort 1 and 2 having FCR's of 0.73 and 0.68 respectively.

Operational efficiencies were also gained with the new hatchery using an average of 1.39 kw-h per 1000 fish versus 2.59 kW-h per 1000 fish in the hatchery in the main farm building. This is an increase in energy savings of 46% with the production of fry in the new hatchery. In addition, labour required to pick eggs / mortalities, clean, and feed the fish was reduced from 2.61 minutes/day per 1000 fish in the old hatchery (cohort 1) to 1.72 minutes per 1000 fish in the new hatchery. This is a reduction in labour of 34%. Cost savings related to reduced power and labour inputs translate into approximately 24% in the first 90 days of production in the new hatchery.

As the hatchery completes its first few cohorts of fish further savings in labour will be achieved by making minor changes to the way in which tasks are completed. The goal now will be to make the new hatchery as efficient as possible and to ensure that the fish performance gains are translated into production gains required to achieve the maximum output from the grow-out operation. There are also plans at looking at how to maximize the capacity of the hatchery by producing more charr fry for others. It has been noted that there is very little disease free egg/fry/fingerling stock available to the Arctic charr industry. Thus, based on our results from year one it will be possible to market the output from the hatchery to other growers.

2.0 Introduction and Background

Millbrook First Nation has been involved in the land based culture of Arctic charr since 2003. As a result of the Marshall legal decision to allow First Nations access to the commercial fishery, Millbrook began a process to acquire assets and training to become involved in the commercial fishery. As part of its fishery development strategy it commissioned a study to evaluate the potential of a land based aquaculture facility on the Millbrook Reserve. In 2003 it opened the first commercial land based Arctic charr facility of its size in Atlantic Canada. The operating company was a registered company in Nova Scotia, was wholly owned by the Millbrook First Nation and operated under the name of Nova Scotia Arctic Charr. The original design capacity of the farm was for 125 mt/yr and was designed to grow fish from 100 gram fingerlings to 1,500 g. There was no hatchery or processing facility planned at that time.

During its initial operation it became apparent that the water treatment system and equipment was incapable of operating the facility at its original design capacity of 125 mt/yr. The operation of the facility also faced marketing and sales challenges which combined with production issues limited the revenues of the farm. After making strides with sales and marketing, the Band then made the decision to upgrade the water treatment system. Prior to implementing the upgrades the Band signed a joint venture partnership with Sustainable Fish Farming Canada Limited (SFFC) and formed Blue Two Limited Partnership which is 50% owned by the Band and 50% owned by SFFC. All of the assets remain wholly owned by the Band. SFFC would be involved in the daily running of the facility and marketing of the products.

The upgrading of the water treatment system involved shutting the system down for several months. The farm was officially closed and emptied to allow for the significant retrofit of the new water treatment system. This fallow period also allowed for a complete disinfection of the farm and eradication of disease vectors. Previously, the farm had been designed to bring in charr fingerlings at 100g and on-grow them to 1,500g (in a 12 month period). This strategy exposed the farm to outside diseases from the farms which were supplying the fingerlings. Even when fish are tested for the known diseases there is always a risk of bringing in a disease. In a system of high recirculation, such as the Millbrook facility, once a disease organism is introduced it is extremely difficult to get rid of it. 

Thus, when the facility reopened in 2010, decisions were made that the farm would bring in eggs and hatch them at the facility to mitigate against the introduction of outside disease vectors. Imported eggs can also bring in disease, but are more easily disinfected at the time of introduction. The first eggs were hatched at the aquaculture facilities of the Agricultural College in Truro, NS which was known to be disease free. This proved to be sub-optimal as the College did not have the staff or facilities to continue with this service. Thus, subsequent batches would have to be hatched at the Millbrook facility.

Four ‘combi-tank' hatching tanks were loaned to Millbrook from the Agricultural College and were set up in available spaces between two culture tanks and existing equipment at the existing Millbrook grow-out farm. The hatching tanks are plumbed into one of the existing grow-out tanks which is used as a chilled reservoir and supplies water to the hatching tanks. This system is undesirable as there is no useable space for the hatchery tanks, no dedicated cooling/cold water system, no independent light control, no additional biosecurity control, no biosecurity separation between hatchery and grow-out farm, no ability to on grow to proper stocking sizes. Once the eggs are off chilled water they are put on flow through well water (8oC) which is make up water that the farm can no longer access as it is channelled to waste after passing through the fry tanks. Upon reaching 1 gram in weight the fry are transferred from the combi-tanks to a 37m3 grow-out tank for the start of on-growing. Due to the dedication of two 37m3 tanks to fry production the farm output capacity has also been reduced.

The operation is involved in the DFO Fish Health Protection Program and as such is regularly tested (4 times/yr for first two years and then twice per year) for a predetermined list of reportable salmonid diseases. The rationale for internalizing egg hatching and fry production has been supported as the farm has been operating for over two years now with no disease. With the improvement in water quality from the new treatment system production and growth are both improved. The design output of the grow-out farm is now 200 mt/yr. However, the dedication of tanks to egg hatching and fry production has reduced the farm potential output to approximately 180 mt/yr. Thus, there are clear economic and operational benefits to be achieved from the establishment of a new integrated hatchery/nursery facility.

3.0 Sustainable Fish Production

Sustainability has become a critical decision factor in the marketplace in North America when buying seafood. Millbrook First Nation recognized this when it established the facility over ten years ago and the importance of this type of fish production has continued to increase to the present day. By investing in a new water treatment facility and applying these principles to the design of the new hatchery, Millbrook has continued to ensure that the production of Arctic charr can be achieved with no impact on the surrounding natural environment and supply high quality fish at a competitive price on a consistent basis. 

Millbrook's commitment to sustainable operations is demonstrated through its low operational impact to the environment by employing state-of-the-art water processing systems and technology that allows the operation to have no impact to natural water resources and habitats. With arrangements using outside nursery farms to culture eggs to fry size to supply the operation, the company was not able to exercise control over water treatment and discharge practices of its suppliers. An additional benefit of the project is the application of Blue Two's sustainable water processing technology to a hatchery and nursery operation that will follow industry best practices for recirculation, treatment, and conservation.

During the time of the installation of the new water treatment for the grow-out facility, a new processing facility was also constructed. The integration of hatchery, grow-out and processing allows the highest degree of control over the product through production and post-harvest handling which ensures that the product is of the highest quality when it reaches the marketplace. This approach has allowed Millbrook and its partner Sustainable Fish Farming Canada to garner a highly respected reputation in the marketplace for its fish products which are marketed under the name of Sustainable Blue.

Feedback from our customers has confirmed that because this Arctic charr is grown in a sustainable manner it is differentiated from its competitors and elevates it to a status that can command a premium price. Maintaining the sustainability factor in our operations is paramount and thus it was important that the new hatchery also be designed with sustainability in mind.

Closed containment fish farms have been gaining attention lately as sea based farming comes under social pressure related to disease and pollution. It is important that addition of an integrated hatchery not only be sustainable in its design but also improve the financial position of the farming operation and thus be economically sustainable as well. A key critique of commercial land based farming operations is that they are not economically sustainable. The technology employed at the Millbrook grow-out and new hatchery facilities allow for commercially viable sustainable production of Arctic charr.

4.0 Project Overview

The execution of this project was divided into four phases, namely;

The project will improve the immediate operating performance of the Millbrook production facility, through the development of an integrated hatchery facility, and will provide immediate commercial benefits from the following desired outcomes:

The facility was completed in June 2013 and has been operating for a period of five months at the time of this report. To date it is demonstrating that it will be an excellent addition to the farming operations at Millbrook and is providing the benefits outlined above.

Challenges to date include the failure of both of the new chillers at a critical time during start up. It was also necessary to dilute the flow of ozone to the contact vessels in order to have enough control over the dosing amount. At full strength ozone supply the on/off control system of the contact vessels is too course due to the relatively light organic loads on the system. With the addition of an air/ozone mixing system we can now increase/decrease the ozone supply concentration and thus respond to the varying organic load situations in the hatchery without the use of variable drives or other control measures. This innovation will also be incorporated into the design of similar hatcheries being developed. It was found that once the system is receiving feed on a regular basis then full strength ozone can be used in the contact vessel.

5.0 Results

5.1 Phase I – Development of Performance Indicators and Hatchery Facility Design (May – June 2012)

Objective: To finalize the design of the hatchery, and develop criteria for evaluation of the impacts of hatchery integration into the farm.

5.1.1 Performance Indicators

The performance indicators were designed to show how the facility performed with respect to required inputs i.e. power, labour, and feed and the expected outputs e.g. healthy fry. Thus, several worksheets were designed to record the labour required to pick eggs/morts, clean tanks/trays, feed fry, weigh fish and other labour requirements. Power was estimated by the chillers, pumps and lighting used in the production of the fry. Fish performance was monitored with respect to mortalities, deformities, growth and feed conversion.

These performance indicators were developed in conjunction with the production staff of the facility and were to be used with the first cohort of charr to be raised in the new hatchery and compared to those raised in the existing facility.

The delineation of performance indicators was completed in May 2012 and the initial monitoring of eggs stocks was undertaken from Feb to Apr 2013.

5.1.2 Hatchery - Overall Design

The design process for any aquaculture facility starts with the output capacity which in this case is dictated by the capacity of the grow-out facility. Once the capacity is determined then the size and number of tanks can be determined based on required stocking densities. This will then allow for the determination of feed requirements which determines oxygen requirements, flow rates, and sizing of water treatment equipment.

Hatchery capacity minimums would be the requirement for the existing fish farm grow-out facility. With a current capacity of 200 mt/yr and an average weight of 1.2kg/fish, the farm would harvest 167,000 fish. Assuming a 5% mortality it would require 175,000 fry to be stocked. It will take approximately 6 months from hatch to 5 grams in the hatchery so the hatchery should have a minimum of 87,500 fry per batch. Based on the fact that Millbrook may wish to assist other Bands with the development of similar fish farms and/or develop more farms of its own, a capacity of 200,000 fry per batch was chosen. The hatchery was divided into two distinct systems based on a need for two separate temperature regimes. Arctic charr eggs require water at 6oC upon initial hatching and yolk sac absorption. After first feeding, the temperature can be increased gradually up to 10-12oC. System 1 was to permit egg hatching, yolk sac absorption and first feeding and on-growing up to 1 gram. System 2 was a nursery system which would allow on-growing from 1 - 5 grams at a temperature of up to 12oC.

Stocking densities should be kept low but optimal to encourage feeding behaviour but allow for good water circulation and waste removal. New fry will be stocked at densities that will require supplemental oxygen.

Water Supply/Drainage:  The water supply to the hatchery is from two sources. The primary source is an existing well which was drilled in 2001 but never used. This well was fitted with a new pump and plumbed to the hatchery building (approximately 200 m away). The second supply is from the existing well and make up water supplying the main grow-out facility. If the primary well pump is out of service then the secondary supply can easily be put into use. New water requirements for the hatchery will be very low and only for incidental losses and evaporation.

The waste lines from the floor sump and the backwash lines of the sand filters are plumbed to the settling tank adjacent to the facility where they are pumped to a secondary settling tank and then into the municipal sewage distribution system and carried to the Colchester Municipal tertiary treatment facility.

Tanks:  System 1 - In order to minimize space and fish handling, system 1 is comprised of eight (8) fibreglass combi-tanks. Combi-tanks are aptly named as they combine hatching trays, yolk sac absorption tanks, first feeding tanks, and early rearing tanks in one nested tank system. Each combi-tank has a diameter of 1.5m and a depth of 0.7 m (1.2m3). Each tank in the combi-tank system is capable of hatching and rearing up to 25,000 eggs.

System 2 - Tanks for system 2 are four (4) circular fibreglass tanks 2.7 m diameter x 1.2 m depth (7.1 m3) c/w a double drain for enhanced flushing of solids.

Water Supply to Tanks:  The tanks are supplied via a pump capable of achieving a turnover time i.e. complete volume replacement, in 30 minutes.

System 1 - Water is pumped to a ring main to provide equal flows at all discharge points regardless of the position in the distribution line. The flow to each tank is regulated by a flow meter (10 to 70 LPM) and passes through a 1 m x .150 m diameter degassing column before entering the tank. Each combi-tank is fitted with its own degassing column to ensure that no elevated gas pressures are present as these can have detrimental effects on eggs and small fish.

System 2 - Water is pumped to a ring main to provide equal flows at all discharge points regardless of the position in the distribution line. Water flow is regulated by a ball valve. Degassing columns are not required on individual nursery tanks.

Water from the tanks is designed to flow via gravity back to the collecting sump (dirty sump). Under normal conditions these sumps would be buried into the ground to accommodate draining and flow rates. However, to facilitate a simpler and less costly construction the building floor levels were split such that the plant room floor was 0.6 m lower than the tank room. This allowed for slab on grade construction but could accommodate the drainage flow rates. The only drawback is that to completely drain the tanks a portion of the water must be drained to waste.

As the tank head levels were kept to a minimum the drainage lines were sized accordingly to allow for the appropriate flow rates. System 1 drainage line was 150mm diameter and System 2 drainage line was 200 mm.

Collection Sumps and Water Treatment: Each system has its own pair of water collection sumps namely a ‘dirty sump' and a ‘clean sump'. Each sump is approximately 9 m3 (18 m3 per system) and was fabricated on-site from 12 mm polypropylene sheet. Polypropylene is a desired plastic over fibreglass and polyethylene as it is more inert and will not leach harmful substances over time.

Water flowing back from the tanks carrying waste is emptied into the ‘dirty sump'.

Solids Removal:  Water from the dirty sump is pumped to an ozone contact vessel where it is subjected to ozone gas which serves to breakdown the solids and to give the particles a small electrical charge. Ozone also serves to breakdown dissolved organics in the water. The ozonated water flows back into the dirt sump. Water from the dirty sump is also pumped to a reflux sand filter which removes particulate materials via adsorption and physical capture. 

Sand filtration is an excellent means to filter water but can become plugged overtime and thus requires more water to back wash and loses efficiency over time. The reflux sand filter incorporates a series of high pressure water jets that serve to assist cleaning during the backwash cycle so that the sand is cleaned to its original condition and does not lose efficiency nor require increased amounts of back wash water over time. Water flowing out of the sand filter is discharged to the top of the biofilter where in it flows down through the biofilter and is deposited in the ‘clean sump'. The design and use of reflux sand filters is a bespoke product and is part of the innovation used in this hatchery. Furthermore in operating the ozone contact vessels is was determined that when there was a small organic load on the system the redox level (ozone dosing) was too course and could not be controlled within the desired specifications. To remedy this a the ozone supply line was fitted with a mixing valve to allow the introduction of compressed air thereby diluting the concentration of the ozone gas mixture used in the contact vessels.

Temperature Control:   Water from the clean sump is pumped to a dedicated chiller (7.5HP) located outside adjacent to the building. After passing through the chiller the water is discharged back to the top of the biofilter and ultimately ends up in the clean sump. An identical system is used for System 2 but it has a larger chiller (10 HP).

Biofiltration and Carbon Dioxide Removal:  These steps are combined in the same unit. Directly above the clean sumps are two trickling biofilters (one each for System 1 and System 2). 

The biofilters are vessels filled with plastic ‘pall rings' (high surface area plastic media with a high void ratio). Water flowing to the top of the biofilter flows over a distribution plate, which creates a standing head of water. It then flows though small holes in the plate and passes over the plastic media. The media serve as home to nitrifying bacteria which are required to convert ammonia into nitrate. After passing through the media the water is drained via a pipe to the clean sump directly below. It is important to note that the pipe draining the biofilter has its discharging end submerged in the clean sump. By having both the entrance and exits of the filter covered with water creates a pressurized air space in the filter. Trickling filters are efficient to use as they require no moving media and thus no energy to do so, are highly oxygenated at all times, and are self-cleaning to maintain a healthy biofilm on the media.

Each of the biofilters is fitted with a series of 50 mm pipes which deliver compressed air (supplied by an air blower) into a space created by a false bottom in the biofilter vessel. This compressed air is then forced to pass up through the media to the distribution plate which has a matrix of 50mm pipes which pass through the standing head of water on the distribution plate. These pipes are joined together and ultimately expel the air to the outside. This rising air through the media is in contact with the downward flowing water and extracts the carbon dioxide from the water and expels it outside via the pipework.

The biofilters for each system are of the same design but of different volumes due to the different amounts of wastes (nitrogen) generated by each system. The biofilter volumes are 3.9 m3 and 11.1 m3 for Systems 1 and 2 respectively.

Oxygen Supply: The oxygen supply for the facility is sourced from a branch line coming from the main oxygen generation system supplying servicing the grow-out facility.

System 1 oxygen is delivered via oxygen diffusers and is regulated via flow meters. As the oxygen requirements are fairly low in the combi-tanks the oxygen diffusers are regulated manually.

System 2 oxygen is delivered via an oxygen saturation vessel with two distribution pumps and is manually regulated by a ball valve and monitored via a flow meter. The oxygen saturation vessel is of a similar design as that of the ozone contact vessels. Water enters the top of the vessel and flows over plastic pall rings creating a high surface area. Oxygen enters from the lower part of the vessel and maintains an oxygen rich environment at a predetermined level. As the water flows though the oxygen it absorbs it and becomes supersaturated. This oxygen rich water fills a predetermined portion of the bottom of the vessel and is then distributed to the tanks via a ring main. 

Each tank in System 2 is also fitted with oxygen diffusers which serve primarily as emergency back up to the delivery of oxygen.

Monitoring and Control:  Key to the efficient operation, management, and emergency response of the hatchery is monitoring and control. The system is fitted with various probes, a Programming Logic Control (PLC) and a Human Machine Interface (HMI). 

The HMI Temperature for both systems is monitored and trended on the HMI screen. If the temperature falls outside of user defined limits then an alarm is triggered and cell phones notified via an auto dialer. 

Water flow to each system is also monitored and if flow stops for any reason then an alarm is triggered. 

The use of ozone to treat the water is highly beneficial but if levels exceed safe limits then it can become harmful to the fish. An indirect measure of the level of ozone is the reduction –oxidation potential or ‘redox' potential. This measures the potential for ozone to react with organic matter in the water. The cleaner the water is the higher the redox level. In natural systems a redox level of 300 to 350 is normal for clean water. Thus each system's redox level is measured. If it exceeds a user defined level then the ozone system is shut down and isolated. Once the redox level begins to fall again the ozone system is activated. The redox level is trended on the HMI to observe the performance of the system. If a ‘high' level is reached then an alarm is triggered.

Oxygen levels are monitored and trended on the HMI for the tanks of System 2. The PLC controls the start/stop of the two delivery pumps to the oxygen saturation vessel. If the oxygen level falls below a set point one of the pump is started. Under normal circumstances this will be sufficient to maintain the oxygen level in the tanks. However, if the system falls below a set point then the second supply pump will be started. Once the oxygen level reaches the desired level then the second pump will be stopped. The second pump also serves as a back-up in the case of a pump failure.

If the oxygen levels in any of the tanks in System 2 falls to a user defined low level an alarm will be triggered and supplemental oxygen via the oxygen diffusers will be activated. In the event of a power failure the oxygen diffusers will come on automatically without the requirement for power. 

Back Up: The critical element in the case of a power failure is the loss of oxygen. To this end a compressed oxygen cylinder is plumbed into the oxygen distribution line. In the event of a power loss a valve is automatically opened to allow the flow of oxygen from this cylinder. If this automated valve fails then there is a manual bypass valve which can be opened. All tanks would then receive oxygen via oxygen diffusers.

Over an extended power outage other water quality parameters would start to become an issue, thus a small generator is available to run the sand filtration and main circulation pumps for each system. This will control carbon dioxide and ammonia levels until main power can be restored.

The PLC will be fitted with a UPS power supply which will allow for alarm activation and emergency notification to be made to the appropriate persons.

5.2 Phase II – Construction and Installation (December 2012 to May 2013)

Objective:  To construct a dedicated hatchery in a stand-alone insulated building to the adjacent farm with its own water supply, and recirculation and disinfection systems.

While AIMAP funding was approved in May 2012, delays in securing additional funding resulted in the construction phase officially starting in Dec 2012 (7 months later than planned) and was completed in May, 2013. The construction phase entailed site preparation, slab pouring, building structure erection, insulation, wiring and lighting, ordering and fabrication of equipment, fabrication of sumps and installation of equipment, and commissioning of systems. Due to winter construction there were a few delays due to weather/cold.

The building has a footprint of 8 m x 30 m and a ceiling height of 3.6 m in the tank room and 4.2 m in the plant room. The building sits on a 6” slab on grade with an 8ft frost wall at the point where the floors differ in elevation. Both floors are fitted with a trench drain which drains to a floor sump located in the plant room. The building is a standard 2” x 6” stud wall construction with R24 insulation and metal sheeting on the outside walls and roof. The inside walls and ceiling are covered with a 10 mm hi-core ‘coroplast' plastic sheet. The dividing wall between the tank room and plant room is also insulated for sound attenuation.

Prior to installation of the floor buried pipe work was installed including delivery pipe from the supply wells and drainage pipe work from the floor drain sump and sand filter backwash lines.

5.2.1 Site Preparation and Building Construction

The site preparation was undertaken by Warren Dill Contracting and included removing the existing hoop house greenhouse, removal of overburden/gravel bed to the appropriate depth and grading. The slab design included building a frost wall 6ft deep across the width of the building, at the point between the plant room and tank room. A gravel base was then laid and compacted, and then a rebar matrix laid in place prior to pouring the floor slab. Prior to pouring the slab it had snowed and a day and a half was required to remove the snow so that there would be a clean area for pouring the concrete floor. It was also necessary to wait for a break in the weather so that it was not raining/snowing and the temperature was above freezing to pour the slab. 

During this same period the underground water supply and drainage lines were installed. This procedure went as scheduled.

The building framing went up as scheduled but the application of the metal siding and roofing was delayed several days due to wind and freezing temperatures. Once the building was enclosed it was insulated. It was possible to bring in some temporary heat and light at this time which aided the internal work. The internal walls and ceiling of the hatchery were lined with a 10 mm white coroplast plastic sheet. Upon installation of the sheets it was determined that the sheets were not square. This required two additional days to square the sheets.

Completed in mid-February, the building was then ready for equipment fabrication and installation.

5.2.2 Equipment and Systems Installation

The supply/drainage pipework and fibreglass tanks were installed in the plant room while the main pieces of equipment were installed in the tank room. 

Fabrication of the sumps and biofilter tanks at the rear of the plant room required the construction of a wooden support structure to frame the lower sump tanks and to support the elevated biofilter vessels. The sumps and biofilter vessels were lined with ½” polypropylene plastic sheets and welded together to form a water tight vessel. The support structure also included a set of stairs and an elevated walkway so that the biofilter distribution plate could be serviced.

The two main reflux sand filters were plumbed in place and required fabrication and installation of the internal pipework necessary to achieve ‘reflux' during the backwash/cleaning cycle. The reflux system requires a 3” ring main above the sand bed which delivers high pressure water flow into the sand via ½” pipes (approx. every 2” along the ring main). During the backwash cycle this high pressure water flow system is activated and the sand if thoroughly cleaned, without the need for additional backwash water.

Routing of all of the pipework proved to be a challenge as the design was modified slightly to allow the pipework to/from the pumps to be supported by the wooden support structure of the sumps/biofilters. This reduced the requirement for additional overhead pipe support but meant that each pipeline path had to be designed so that it did not cross with another.

The two system chillers were installed but were delivered late due to a manufacturing error at the factory. Shortly after commissioning both chillers failed and required extensive servicing.

Following the installation of all the water/oxygen pipework and associated flow meters the control system wiring and monitors/sensors were installed. This included all oxygen sensors, redox sensors, temperature sensors, sump level switches, system flow switches and associated PLC equipment, HMI module, and telephone/alarm systems.

5.2.3 Commissioning

Commissioning involved the running and testing of all equipment and sensors. All tanks/sumps/vessels were tested for water tightness. A leak was detected in the biofilter vessel of system 2 which required emptying the vessel of all pall rings, locating the leak and re-welding. As system 1 was required first it was commissioned prior to system 2. System 1 started operating (laying down of eggs) while system 2 was still being commissioned. 

During the commissioning of chiller no. 1 (system 1) all went well and the chiller worked for several days without issue maintaining a constant temperature of 6°C. However, shortly after the eggs hatched the chiller stopped working and temperatures began to quickly elevate. The backup to this was to use system no. 2 chiller which had been tested but not run for any length of time. When system no. 2 chiller was brought on line it ran for a short time and then also stopped working. Service technicians indicated that it would be several days to repair the two chillers. A back up chiller was brought in from another farm which brought the temperature back down. During the period (4 days) without chilling the system was run on flow through in order to keep the temperature as low as possible. Temperatures rose to approximately 10°C which increased the rate of yolk sac absorption. Following the lowering of the temperature the fry did not come onto feed well and mortalities were elevated. It was decided that this batch would not be ‘representative' for comparison purposes and a new batch of eggs was brought in.

During commissioning a pipe in the Reflux Sand Filter of system no. 2 broke and sand was exiting during the backwash cycle. This required by-passing the sand filter, emptying it of sand, locating the broken pipe and reinstalling a new pipe. This repair took 6 days.

5.3 Phase III - Operating and monitoring (May to November, 2013)

Objective: To operate the new hatchery and monitor the operational parameters outlined in phase I as well as the performance of the stock in the new hatchery.

In the past eggs from Icy Waters (Yukon, NWT) have been used as the primary strain of charr grown at the facility. The majority of charr in the North American market place is a strain of charr from Iceland. Due to a more uniform uptake of pigment in the Icelandic strain a decision was made to switch to this strain of charr. Thus, the last cohort of fish hatched in the existing facility (Feb 1, 2013) and the first cohort of fish to be hatched in the new facility (May 9, 2013) are Icelandic Arctic charr.

Phase III was initiated with the introduction of the first cohort of eggs (Feb 2013). Approximately 18,000 Icelandic Arctic charr eggs were introduced into two combi-tanks in the new facility on May 9, 2013. The performance of this cohort of fish was to be compared with the Icelandic eggs which were imported and hatched in the existing facility on Feb 1, 2013. Due to a chiller failure during the sac-fry development period, the fry were not able to develop at the appropriate rate and it was decided that they should be terminated and a second cohort was brought in on Aug 12, 2013.

Water quality parameters including temperature, pH, carbon dioxide (CO2) and Oxidative-Reduction Potential (ORP) were recorded daily. In addition, mortalities and feed quantities were also recorded daily. Weights were recorded at a month after first feeding, bi-weekly for the next month and then weekly up to the point of transfer to grow-out tanks.

The amount of time required (i.e., labour) was recorded for the tasks of egg picking and mort removal, tank cleaning, feeding, and fish weighing. The equipment which required electricity i.e. pumps, chillers, lights were monitored for their usage (run time) and calculations performed to determine the amount of electricity used.

5.4 Phase IV: Assessment and Reporting (May to November 2013)

Objective: To assess and compare the monitored parameters of the operation of the new hatchery (first three months) with that of the old hatchery. 

The assessment of the performance of the new hatchery with respect to its impact on efficiency, fish growth and health, and production costs was determined following the collection and analysis of the operational data on the fish currently in the hatchery. 

Initial stocking of the hatchery was in May and thus monitoring would be completed in July 2013. Due to a chiller failure and elevated temperatures during the sac fry stage of development, the initial stocking cohort was terminated and a second cohort of eggs were brought in on August 12, 2013. The first cohort included 60,000 eggs which is a standard number hatched twice a year. The second cohort was only 13,000 eggs as the production cycle at the farm was not appropriate for hatching more eggs. Monitoring of this cohort was completed in at the middle of November.

5.4.1 Fish Performance

A summary of the performance data collected on the Icelandic fish imported on Feb 1, 2013 (Cohort 1) and those of August 12, 2013 (Cohort 2) are presented as follows;

Egg and Fry Mortality:
Overall mortality was reduced in cohort 2 (new hatchery) with the reduction occurring in the egg mortality rate. Fry mortality was actually higher in cohort 2. As illustrated in the above graph there were three short periods where the mortality of fry became elevated relative to the periods before. There were no apparent causes identified in any water quality parameter being measured. Investigations are on-going as to the role of water quality changes from the new well. Often the water quality from new wells can change suddenly as new water infiltrates the surrounding soils/rocks. Regular water testing will be conducted on the well over the next year to ensure that changes are not occurring. We would expect that fry mortalities will also be reduced in subsequent batches and thus further reducing overall mortality rates.

Feeding, Weight Gain, and Food Conversion:
Feeding, weight gain, and food conversion was similar for the two cohorts. Within any weight sampling there is a variance as only a subset of the fish is weighed. Weight gains are also quite variable at the early stages of the fish's life. Generally once fish come ‘on' feed they are very aggressive and convert very efficiently with food conversions less than 1.0

Fish were fed to satiation at the point of first feeding. Once all fish were on feed they were fed hourly 12 hrs/day. Fish were fed a 0.5 mm crumble diet prepared by Corey Feeds (New Brunswick). This regime was similar for both cohorts with the exception that cohort 2 was fed by automatic feeders from day 30 to day 45 from first feeding. Cohort 1 was fed by hand. Both cohorts were feeding at approximately 3% of their body weight per day by day 30 and 2.5% of their body weight by day 45 from first feeding.

Weight  - Fish were weighed 30 and 45 days after first feeding. Cohort 1 had an average weight of 0.351 g (as of April 30, 2013) 45 days after first feeding (73 days post hatch). Cohort 2 had an average weight of 0.390 g (as of November 16, 2013) 45 days after first feeding (70 days post hatch).

The weights of the fry after 30 and 45 days on feed were similar with cohort 2 slightly ahead after 45 days. Of interest is that cohort 2 was slightly behind after only 30 days on feed. This can be a reflection of variance in samples tested and/or a true increase in growth after 30 days by cohort 2. As the fish continue to grow it will become more evident. 

Both cohorts had excellent food conversion ratio's (FCRs) i.e the ratio of feed used to achieve the recorded weight gain. Cohort 1 FCR was 0.73 and cohort 2 FCR was 0.68. Cohort 2 was fed more with the use of automatic feeders which would allow for the feed to be consumed over a longer period during the day. This may have contributed to the better FCR.

5.4.2 Water Quality Performance:

Water quality in the new hatchery was similar in all respects to that of the flow through hatchery system used for cohort 1. The variance in temperature readings of cohort 2 are slightly higher than that recorded for cohort 1. This is due to the variation in the recording sensor and can be reduced with a change to the sampling/averaging time in the software.

Egg hatching temperature was slightly lower but egg hatching timing was similar for both batches. This is probably a result of the degree days prior to receiving the eggs. Temperature after hatch up to first feeding was also slightly lower and fish in cohort 2 took a 3 to 5 days longer before first feeding.

The pH in cohort 1 was that of the well water (avg. ph=6.9) and would not be expected to change over the period. The pH of cohort 2 was controlled in the hatchery and was adjusted with a buffering solution of calcium carbonate to maintain a slightly higher pH (avg. ph=7.6) in order to stimulate growth in the biofilter.

Carbon dioxide readings were slightly higher in the new hatchery (cohort 2) due to a residual accumulation of carbon dioxide and the limitations of the carbon dioxide removal system. The actual maximum level reached of 4 ppm is still well within the target of less than 10 ppm. 

ORP readings were also initially higher in the new hatchery (cohort 2) as the redox of the new well was higher than that used for cohort 1. As the recirculation system matured the redox naturally dropped due to the introduction of organics. After a few weeks of adjusting the redox control system it was stabilized. This illustrates that the recirculation system of the new hatchery is capable of achieving water quality results similar to those of flow through system on well water.

5.4.3 System Efficiency

Labour:
The overall labour required to maintain the eggs/fry during the first 90 days was lower in the new hatchery by approximately 34% (.015 hrs/day/1000 fish saved eg. 81 hrs for 60,000 fish over a 90 day period). Egg picking and removal of mortalities was reduced primarily due to a lower number of overall mortalities in the new hatchery, but can also be attributed to better lighting and ergonomics of the egg trays and combi-tanks, making it easier to see and remove the dead eggs/fish. Cleaning was easier as well due to the ability to better control water flows and concentrate waste in the tanks. Modifications were also made to the combi-tank fry trays to allow for an even easier cleaning of the tanks. Presently the internal screens surrounding a short standpipe and scrubbed by a brush and waste is directed to the outflow of the tank. The new modification will use a double standpipe such that the internal standpipe can be pulled to allow the tank to self-clean over a short interval. The actual time spent will probably remain unchanged but it will require less of an effort by the operator and be less intrusive for the fish. Feeding and weighing of the fish is relatively the same as each tank uses a combination of hand feeding and automatic feeding. Taking fish weights on a regular interval is the same process as was used previously.

Energy Usage:
The energy usage in the new hatchery was approximately 46% lower than that of the old hatchery system in the main farm during the first 90 days. The efficiency gains are due to a more efficient chiller in the new hatchery but also to the fact that the introduction of new water (requiring chilling) is significantly reduced in the new hatchery over that used in the hatching/rearing of cohort 1.

Production Costs:
In assessing the core production costs of fry production the following cost items were reviewed – eggs, labour, energy, and feed. Egg costs were the same for both cohorts so they were not included in the overall cost comparison. However, there were differences between the cohorts of the other three cost items.

The overall production cost for feed, energy, and labour for cohort 1 was $1.52 per 1000 fish (first 90 days). The similar production cost for cohort 2 was $1.15 per 1000 fish. This is a cost reduction of approximately 24%. The continuation of cost savings will be analyzed up to the target weight of 5 grams over the coming months.

6.0 Summary and Next Steps

The benefits of the project include:

The results of the assessment of the operational parameters show that the system has gained efficiencies in labour and power inputs for the production of fry less than 1 gram. In addition, the recirculation system has been able to demonstrate that it can maintain water quality parameters that are similar to those of a flow through system on well water.

Although it is early still, the design of the hatchery is achieving its desired goals. In the existing hatchery it was difficult to maintain a steady temperature due to the size of the chiller and the fact that the hatchery tanks were in a large room with a varying temperature. To date the temperature in the new hatchery has been stable with a variance of less than 1°C. Some of the components of the hatchery design namely the polypropylene sumps are a new design and are already being incorporated into designs for a new salmon smolt hatchery being constructed in Centre Burlington, Nova Scotia. In addition, the hatchery design is being included as part of an overall integrated land based salmon farm for the Batchewana First Nation in Sault Ste. Marie, Ontario. 

Morality rates are slightly improved and it is anticipated that they will be further gains in the coming cohorts. Growth and food conversion rates were also equal to or slightly improved in the hatchery. As the data pool is very small the overall intended gains in achieving a 5 gram fry for stocking into the grow-out facility will not be known for another 3 months. 

Given the current operational and fish performance data that has been analyzed, it is anticipated that the integration of the new hatchery will provide strong and healthy fry of a minimum weight of 5 grams to the grow-out facility, and allow the facility to optimize production to achieve 200 mt/yr.

7.0 Communications

In order to further aquaculture enhancements in the region it is important to communicate the findings of this type of aquaculture development to the public and to industry.

Site Visits:
Millbrook has already had several visitors to the new facility including: Alfie MacLeod (MLA Cape Breton West), local primary and elementary schools, Patrick Swim (local lobster/fish businessman), and Murray Schalin (President, Coldwater Fisheries Ltd.). Representatives of the Bolivian First Nation Band also toured the new hatchery in (insert month and year) with an interest in learning more about land based aquaculture. In Nov 2013, industry and government representatives from Quebec visited with an interest in recirculation design in general, but specifically in the use of ozone in recirculation freshwater systems. In addition, a First Nation Band from Newfoundland and a group of fishermen from northern New Brunswick are scheduled to tour the new hatchery in (Feb 2014), both with a view to establishing land based aquaculture. 

Articles/Publications:
It is the intent that once the Phase IV is finalized (November 2013) the results will be published in regional and national First Nation newsletters (Mi'kmaq Maliseet News, Windspeaker) In addition an article will be published in Hatchery international magazine and a pdf file of the report will be available for electronic distribution upon request from the public.

Presentations:
Following the first year of operation, the impact of the hatchery and the overall vertically integrated design will be presented at the annual AANS Fish Farmers conference held in January each year in Halifax, NS. Presently, a feasibility study is being undertaken for another First Nation Band in Ontario who are interested in constructing a vertically land based aquaculture operation. The technology used in the design of this hatchery will be part of this project.

Date modified: