Principal Investigator(s): Hunt Howell, Win Watson, Michael Chambers

I. Accomplishments

A. Scheduled Tasks
1. Verify the results of the first halibut grow-out study, and learn more about the behavior of halibut in submerged, offshore cages.
2. Finish harvesting the 2003 cod, and stock and maintain the next group of cod.
3. Evaluate the physiology and behavior of cod and halibut in offshore net pens using video and acoustic telemetry systems, and correlate the behavior of fish to natural perturbations in environmental conditions by continuously monitoring the conditions inside the cage while collecting our video and tracking data.
4. Use behavioral and physiological data to improve the production of fish.
5. Collect data that will be used in economic analyses.
6. Disseminate results of the project.

B. Progress on Tasks
1. Halibut transfer and culture.
In May of 2001, 2000 juvenile halibut (30g mean weight) were purchased from R&R Development Ltd. in Digby, Nova Scotia, and transferred to one of our 600 m³ Sea Station cages. Monthly samples of the fish allowed us to track growth and survival. Growth was steady throughout the 3-year experiment, and fish were harvested at a mean weight of 2.9kg in May 2004, after 31 months in the cage. Specific growth rate (SGR) for the entire period was 0.4g/d. Food conversion efficiency was 1.0. Survival, from stocking to harvest, was 68%. Full details of the halibut work are provided in the 2004 Annual Report, as well as in Howell and Chambers (2005). A second round of halibut culture was scheduled to start in early summer 2006. The goals were to verify the results of the first study, and to learn more about the behavior of halibut in submerged, offshore cages. We chose to delay this work because we had only a single, small feed buoy at the site during 2006, and this was being used to feed the cod. Our plan is to purchase and stock the halibut in the summer of 2007. Fish will be purchased from Scotia Halibut, a Canadian commercial halibut producer. Thirty-five hundred fish, with an average weight of 1 kg, and vaccinated for Vibrio, will be transported the Port Authority in Portsmouth, NH in a smolt truck. From there, the 5m³ live haul tanks will be lifted with a crane and placed onto the deck of a contracted fishing vessel. Once the halibut are relocated offshore, they will be fed diets specially formulated for halibut from the 20 ton feed buoy (3% body weight/day). Divers will sample the fish monthly to obtain growth data. We anticipate that the fish will reach market size (4-5kg) by the autumn of 2010. These fish will be used in some of the proposed research (outlined below). Upon harvesting, all fish will be enumerated and weighed to enable precise quantification of survival, growth and feed conversion during culture in the net pen. The harvested fish will be transported to the Portsmouth Fishermen's Cooperative in Portsmouth, NH, where they will be processed and sent to market. All proceeds will be returned to the OOA project.

2: Cod.
A.) Harvest: In January 2006, we harvested the last of the cod stocked into the cage in 2003. To facilitate this harvest during the rough winter weather, the cage was moved to a semi- protected area near the Isles of Shoals. Here the remaining cod were harvested live and transferred back to 35m³ holding pens under the pier near the Coastal Marine Lab. Cod were held until the summer of 2006 before they were sold live to Pacific Trade at a 1.7 kg mean weight. Some cod were also provided to Great Bay Aquaculture to add to their broodstock.

B.) Stocking: Great Bay Aquaculture (GBA) produced the cod presently in one of the cages in 2006. On April 28th and 29th 2006, a total of 50,000 cod (31.5g mean weight) were transferred offshore. The fish were starved for 24 hours before being pumped from nursery tanks in the hatchery and counted into 5m³ smolts tanks on a flatbed trailer. The cod were transported to Rye Harbor and sluiced from the truck into tanks on the project vessel. During the 1-hour transfer offshore, the fish were provided with flowing seawater and oxygen. At the site, fish were sluiced from the boat into a 200m³ nursery net inside the 3000m³ Sea Station cage. Mortality in the two weeks following transfer, which we associated with transfer stress, was low (6%), as was the mortality of the fish during their approximately 5.5 months in the nursery net (3%). Some of the fish exhibited abrasions on their pelvic, pectoral and caudal fins which we assume was caused by abrasion against the twine of the net. Over-crowding was probably the root cause. We did have a single incident of acute mortality (3%) that was related to partial cleaning of the nursery net. Cod were released from the nursery net into the main net on October 17, 2006 at a mean weight of 150g, and there has been virtually no mortality since their move into the larger net.

C.) Maintenance and Sampling: The 43,000 remaining cod will be held in the submerged cage (15-25 meters), monitored, fed and maintained until they are harvested in late 2008. Monthly samples of fish will we used to calculate mean weight, length, gonsosomatic index (GSI), and hepatosomatic index (HSI). Fish will be fed twice per day, and the amount will be adjusted based upon size and temperature. Fish were fed a Burris™ diet (semi-sinking pellet, 50% protein, 14% lipid) until November 2006, and diet was then changed to a Skretting™ diet due to density problems with the original feed. Food conversion ratio to date has been good (0.98), as has the specific growth rate (3.2%/d). Growth in weight has been steady (Figure 1), and existing biomass in the cage is estimated at 12,900kg.

3. Evaluate the physiology and behavior of cod and halibut in offshore net pens.
We had three major goals in 2006:
1. Finish testing our new instrumentation buoy and deploying it at the study site.
2. Collecting activity data with the new buoy, starting as soon as new cod were placed in the net pen.
3. Obtaining respirometry data in the laboratory so that fish swimming data from the field could be converted into metabolic rate values.

We accomplished all three of these objectives. Details are provided below.

A. Instrumentation Buoy
Most of 2005 was spent building a new instrumentation buoy devoted solely to collecting data about the behavior of cod in the net pen, as well as oceanographic data from within the pen (currents, light, tides, temperatures, etc.). This effort was successful and the buoy was ready to test and modify as necessary in the beginning of 2006. The new buoy was fabricated from a surplus Coast Guard navigational buoy (Figure 2). One of its two waterproof tubes was used for batteries and the other for our instrumentation. We also added a “switch box” to the buoy to make it easier to connect all the input cables for our underwater cameras and other instrumentation. Finally, we also added two solar panels and a wind generator. A diagram of the new instrumentation buoy is shown in Figure 2, along with pictures of the actual buoy being prepared for deployment. other waterproof tube (A) in the buoy was used to hold the instrumentation (computer, HTI receiver for tracking, DVR for video). A box (C) was mounted on the top of the buoy to serve as a type of switchbox for all the cables leading from the buoy to the cage, located ~50-100 ft. under the surface. This new buoy housed: 1) the HTI telemetry system, capable of fine-scale ultrasonic tracking of tagged fish within the cage; 2) A four channel digital video recorder (DVR), installed to analyze the movements of the population via 3 underwater cameras (the extra channel monitored video taken from a camera on the buoy to examine activity and weather on the surface) and; 3) An additional environmental monitoring system was added to examine current, temperature, salinity and depth within the cage. This buoy has been very successful, exceeding our expectations. We were able to continuously record video and telemetry data for between 4 and 7 days, using one set of batteries and the additional power provided by the wind generator and solar panels. Then we would download the data from the computer, swap one hard drive in the DVR for a fresh one, and replace the batteries with a fully charged set. There were no leaks and the buoy weathered several challenging storms in the late fall without a problem.

B. Tracking animals in the net pen
During the spring of 2006 the buoy was tested on a nursery nets located under the pier at the Coastal Marine Lab (CML) in New Castle. In early July, when a new group of cod was placed in the net pen, the instrumentation buoy was deployed at the site and the cables were installed on the SS3000 cod cage. During the past summer the buoy has been successfully collecting vast amounts of data from the telemetry system as well as the digital video recorder (DVR). Currently we have an extensive data set of fish movements from July to October. To date we have tracked a total 48 juvenile cod for a total of 1,718 hours, yielding an estimated ~45 billion data points. Additionally, the DVR has captured large-scale movements of the population for hundreds of hours. We are currently analyzing those data and we plan to submit a paper for publication in the spring, and present our results at the World Aquaculture Conference in February.

We are currently collecting three types of data from the fish cage. First, we are tracking ~6 cod at a time using the HTI telemetry system. The four hydrophones, receiver and software allows us to detect the position of a fish with an implanted transmitter every 2 seconds, in 3 dimensions. Second, the three cameras inside the cage allow us to monitor the overall position of the population of fish in the cage, as well as their approaches to the vicinity of the feeding tube. Finally, the CTD inside the cage will allow us to correlate fish behavior with various environmental parameters.

One of the major questions being considered concerns the feeding behavior of cod in these net pens. We are trying to determine if they express any feeding anticipatory activity (FAA), and if specific fish feed each time they are presented with food. Both telemetry and video data are being used to address these questions. In both cases, part of the problem we are facing is the total amount of data available. This year we addressed this issue for our video data using a novel approach, summarized below.

We have attempted to count the number of fish approaching the feed tube in order to quantify feeding activity. However, we found this to be extremely difficult. Therefore, we decided to take another approach and develop a system to automate the process. We purchased a video system (BTV) that is activated by movement. It detects movement by changes in pixel color or shade of gray, from one frame to the next. Thus, if no movement occurs, nothing changes. When lots of movement takes place, many pixels change. We have found that this system provides a fairly reliable index of feeding activity, and it is much less time consuming than counting fish. Figure 3 shows a one-day sample of data taken from a camera mounted over the feed tube.

HTI monitoring activities were conducted through November 2006. The buoy was then recovered from the grid in December and towed back to a mooring near the Port Authority where it will remain for the winter.

C. Respirometry
During the summer and fall a sample of cod was taken from the cage and transferred to a holding tank at CML. These fish were used in a set of respirometry experiments to investigate the relationship between swimming speed and metabolic rate. Each experiment takes approximately 8 hours, so that we can obtain oxygen consumption data from each fish over a range of swimming speeds, while still leaving time for the fish to recover between each trial. We have run 12 fish to date and the relationship between swimming speed and metabolism obtained is consistent with previously published data. During the next few months our goal is to combine these lab based measurements with data concerning swimming speeds in the field, to yield hourly and daily estimates of cod metabolic rate in the offshore net pen. Eventually, we plan to use this information to modify our aquaculture practices to they are more efficient.

II. Tasks and activities for next reporting period
1. Continue to monitor the growth of the cod.
The 43,000 remaining cod will be held in the submerged cage (15-25 meters), monitored, fed and maintained until they are harvested in late 2008. Monthly samples of fish will we used to calculate mean weight, length, gonsosomatic index (GSI), and hepatosomatic index (HSI). Fish will be fed twice per day, and the amount will be adjusted based upon size and temperature.

2. Cod harvest 2007.
Live harvesting will commence once the new Coastal Marine Lab pier has been built in the summer of 2007. Markets in Boston desire live cod at a size ranging from 500-1000g. The pier is an important staging area that is used to transition fish from offshore cages to holding pens where they can be counted and weighed prior to their transfer to market. Harvest techniques will include the isolation and decompression of 500-2000kg of fish at one time. The fish will be pumped into 5m³ live haul tanks on board the R/V Meriel B and transferred to 35m³ holding pens under the pier. During the transfer, the live haul tanks will be supplied with flowing seawater and oxygen. Cod will be held here until a refrigerated, live haul truck relocates them to a warehouse in Boston where they are finally distributed to restaurants in the area.

3. Continue to study cod behavior and physiology.
During the early winter we plan to continue data analysis to prepare for publication of our findings and the presentation two talks at the World Aquaculture Society meeting in San Antonio in February. We also plan to continue our research this spring and redeploy the system to further investigate the current cod stock as well as gaining preliminary information on halibut. With the baseline information we have on cod, we can begin manipulation experiments to further investigate optimum feeding regimes and feeding anticipatory activity. Also, because much of our data in 2006 were obtained from young fish in a small net pen, it will be very interesting to compare their activity with larger fish in a bigger cage. We also hope to begin experiments designed to explore the influence of sounds on cod behavior. Finally, the one technical change we hope to make is to add a high-speed wireless connection so that we can observe video and download data remotely. We can get cell phone coverage near the pens and cell phone based Internet services should be appropriate for this task.

4. Stock halibut and monitor their growth
Fish will be purchased from Scotia Halibut, a Canadian commercial halibut producer. Thirty-five hundred fish, with an average weight of 1 kg, and vaccinated for Vibrio, will be transported the Port Authority in Portsmouth, NH in a smolt truck. From there, the 5m³ live haul tanks will be lifted with a crane and placed onto the deck of a contracted fishing vessel. Once the halibut are relocated offshore, they will be fed diets specially formulated for halibut from the 20 ton feed buoy (3% body weight/day). Divers will sample the fish monthly to obtain growth data. We anticipate that the fish will reach market size (4-5kg) by the autumn of 2010.

5. Multibeam sonar study.
The UNH aquaculture site and the live harvest cage will be used to support a project entitled “Development of multi-beam sonar as a fisheries tool for stock assessment and essential fish habitat identification of groundfish in the Western Gulf of Maine”. The purpose of the proposed research, being funded by the Northeast Consortium, is to prove the concept of using multi-beam sonar to characterize the spatial distribution, behavior, and relative abundance of cod aggregations. The research will be cooperative effort among fisheries scientists from the University of New Hampshire (UNH), acoustic scientists from the UNH Center for Coastal and Ocean Mapping (CCOM) and the National Marine Fisheries Service, commercial fishermen, and the UNH Open Ocean Aquaculture (OOA) project. We intend to answer 3 fundamentally important questions:

1. Can Atlantic cod be detected using multi-beam sonar?
2. How does acoustic information (back scatter) from multi-beam sonar relate to known densities (number/unit volume) and sizes of cod?
3. How do results from multi-beam sonar compare to the more accepted technique of split-beam scientific echo sounders?

Although the details of experimental design are beyond the scope of this report, a basic description follows:
Adult cod (70-80 cm TL) will be collected by a commercial trawler. Fish will be individually weighed, measured, and placed into the live harvest cage at the OOA research site. Fish will be stocked in the empty live harvest cage at three consecutive stocking densities (0.25, 1, and 2 fish per m³) starting with the highest density. The bottom of the cage will be submerged to 35 m and acclimated for approximately 24 hours before acoustic measurements are collected. The cage will be lowered again to a bottom depth of 55 m, acclimated, and surveyed. After acoustic measurements are collected for the high density at both depths, the cage will be raised following a decompression schedule, and then fish will be caught with nets and then released back to the ocean. At the surface, divers will collect and count fish that died from gear damage, transport, and handling stress. Densities will be adjusted for mortality after each two-depth trial. Acoustic data would be collected for individual targets and schools for a low-, mid- and high density at mid- and bottom depths similar to those expected to be encountered in the field during night and day sampling. All acoustic surveying will be done by the R/V Coastal Surveyor, built specifically for coastal multi-beam hydrography and operated by Center for Coastal and Ocean Mapping. If time and resources permit, an additional density and depth will be tested.

6. International collaboration.
We have agreed to participate in a collaborative, multi-year research program with SINTEF Fisheries and Aquaculture AS (Norway) and the Norwegian University of Science and Technology. The research program, entitled CREATE (Centre for Research-based Innovation in Aquaculture Technology) is multifaceted, but is focused on technological research for aquaculture. Topics will include offshore cage design and testing, and interactions between wild fish and farm fish, including fish escapement. In 2007, we will begin to collect data on the maturation and spawning of cod in our net pen, and the potential for escapement of gametes and embryos should the cod spawn. In particular, we will collect monthly samples of gonad weights (both sexes), collect ovarian and testicular tissue for histological analyses of oogenesis and spermatogenesis, and check for spermiation and ovulation. In subsequent years we intend to investigate sperm viability and quality (mobility), investigate egg size and egg quality, and fertilize the eggs of captive females with the sperm of captive males to investigate percent fertilization, larval size, larval quality (survival to various stages, ability to feed, etc.).

III. Important Results or Findings:
1. Previous results with halibut and cod suggest that both species are excellent candidates for cold water, marine aquaculture. Both seem resistant to disease, and both shown very good growth performance.

2. Cod stocked in 2006 have grown reasonably well. During the first 7 months (May to December), the cod increased from 31 to about 230g mean weights. The stocking methods employed were good, and resulted in almost no transfer mortality. Abrasions on the fins indicate that over-crowding occurred in the nursery net as the fish increased in size. We also learned that attempting to clean the net when high stocking densities exist can result in some mortality so is not advisable.

3. Examination of the telemetry data obtained so far has been very illuminating. While we are still processing the information, and no firm conclusions can be put forth until more data are available, it is possible to offer several observations:

1. Fish do not appear to utilize the entire cage, at least within the time frame of 24 hours. Instead, they appear to spend most of their time around the outside edge, near the 25 m diameter rim. Typically, they do not swim around the cage in a circle, but confine their activity to one area. At some times of day they are very localized in one region of the pen. This contrasted diver observations of continuous swimming around the rim of the cage.
2. The cod tend to be most active during the daylight hours.
3. The fish appear to anticipate feeding and become more active in the hours prior to feeding. When food is administered they often rise toward the feeding tube to feed. However, they do not appear to rise to the feeding tube each time food is released during the 1-hour feeding period. Furthermore, some fish don’t appear to feed on some days.
4. When tracking multiple fish, they do not appear to school, or move together. Rather, each fish appears to move independently of the others even though they may follow the same overall pattern.
5. Using lights to delay sexual maturation may have costs and gains. Each time the lights come on and off it leads to a doubling of metabolic rate, which may reduce the gain in growth due to delayed maturation.
6. The video data suggests that there maybe some feeding anticipatory activity, but further manipulations must be made to confirm this. In particular, we need to alternate feeding fish from one day to the next.

All of the instrumentation worked as planned, proving the feasibility of studying cod in offshore pens. With the addition of a high-speed Internet connection, the system could be easy to maintain and provide continuous information to managers on shore.

IV. Difficulties Encountered
1. Some of the difficulties encountered in 2006 can be traced back to the loss of the 1-ton feed buoy, and its associated equipment, in late 2004. The loss affected the research in three ways: First, we lost the ability to feed the cod regularly, which undoubtedly led to some loss in growth performance in the cod we finished harvesting in January 2006. Although we are using a small (1/4 ton) feed buoy to feed the present group of cod, its capacity is limited, so frequent hand feeding has been required. When we lost the 1-ton feed buoy in 2004 we also lost the underwater lights designed to prevent precocious sexual maturation of the fish. The ensuing early maturation in the cod harvested in late 2005 and early 2006 probably led to a loss in growth performance, and may well have contributed to the suboptimal flesh quality of the harvested cod. It is hoped that the new 20-ton buoy being constructed, which will house our underwater lights, will be completed before early maturation becomes an issue with the current group of small cod. Finally, we lost the telemetry system with the loss of the 1-ton buoy, and much of 2005 and 2006 were spent putting together, and testing, a new system.

2. The noticeable fin erosion of fish in the nursery net, presumably due to over-crowding, was a problem. It suggests that: a) we should stock fewer fish in the future; and b) the nursery net has a suboptimal shape (small volumes in the corners).

3. There was insufficient power generation capability in the modified telemetry buoy. While the wind and solar generators were able to power the telemetry equipment for days at a time, there was a need to install freshly charged batteries on a frequent basis during periods of peak data collection. It is anticipated that the installation of the 20-ton feed buoy will eliminate this need.

V. Anticipated Success in Meeting Project Objectives
We anticipate meeting all of the project objectives. To date, we have grown halibut and cod to market size in our offshore cages. We have monitored the growth performance and survival of both species at monthly intervals, we have been collecting data that will be useful in economic analyses, and we have been disseminating the results through scientific publications and presentations. We have also made good progress with the cod telemetry work. We are collecting the type of data we had hoped to obtain and we are now fully capable of analyzing and visualizing those data so that we can use it to test our various hypotheses concerning the behavior and physiology of cod in offshore net pens.

VI. Reports, Manuscripts and Presentations
Presentations:
Chambers, M.D, W.H. Howell, R. Langan, , B. Celikkol and R. Barnaby. Recent Developments at the University of New Hampshire Open Ocean Aquaculture Project. Aquaculture Canada, November, 2006. Halifax

Howell, W.H. Open Ocean Aquaculture at UNH. UNH Emeriti and Retired Faculty Association. Nov. 15, 2006.

VII. Expenditures
Expenditures were in the range anticipated for the work accomplished to date.