Cancer productus Predator Management Technique for Intertidal Shellfish Aquaculture Phase 2

Below Sea Level Oysters Farms
AIMAP-2009-P40

Table of contents

• Executive Summary
• Conclusion
• Outline
• Phase 1
• Description of Work
• Scheduling
• Design
• Fabrication
• Installation
• First Observations
• Trial Design
• Location
• Layout
• Setup
• Results
• Discussion
• Visual and Physical Disruption
• Conclusion
• Acknowledgements
• Appendix 1
• Conclusion

Executive Summary

Intertidal shellfish culture is under intense predation from the Cancer productus. This project has installed two sizes and designs of a subsurface barrier to prevent the movement of C. productus onto the oyster beds (Crassostreo gigas). The treated area was contained within a barrier boundary, and an adjacent non-treated area was used as a control in the trials. In the trials, both the treated area and the control area had 300 oysters laid out in a similar method, daily counts determined the amount of predation in each area. The barrier boundary that is parallel to the Low Water Mark (LWM) had “T" intersections of the same material that emptied towards deep-water, and allowed the crabs to return to the ocean. The objectives for the project were:

• There has to be non-lethal control of the predator;
  
• It has to avoid the riparian rights issue;
  
• It needs to be cost effective for the oyster farmer; and
  
• There must be minimal maintenance requirements for the system.

The main objective of the study was to determine if the Crab Barrier was effective in reducing the crab predation upon C. gigas. We had expected that several crabs could remain in the barrier system and either leave, succumb to high temperature, or be preyed upon by birds. Now we are certain that the crabs ex it through the barrier system.

Conclusion

The study as conducted did provide strong support for the objectives of the research.

• Documented a high degree of crab predation on intertidal oyster culture;
  
• The control, 12" and 10" barrier had a predation rate of 5.9211%, 1.0702% and 2.5526% oysters/day respectively;
  
• Both designs for the barrier worked as desired both with benefits and drawbacks;
  
• The barriers only had minimal siltation;
  
• The larger barriers reduced predation on intertidal shellfish;
  
• This results in increased production and reduced seed costs;
  
• This allows a larger portion of the tenure useable for grow out;
  
• This results in increased production per area; and
  
• This improves the market value of the land for the farmer

Outline

Intertidal shellfish culture is under intense predation from the C. productus. This is an innovative example of sustainable non-lethal predator management. This project has installed two sizes of a subsurface barrier utilizing two different designs to prevent the movement of C. productus onto the oyster beds. The treated area will be contained within a crab barrier boundary, and an adjacent non-treated area will be used as a control in the trials. In the trials, both the treated area and the control area will have 300 oysters laid out in a similar method. The crab barrier boundary that is parallel to the LWM will have "T" intersections of the same material that will empty towards deep-water. A crab will come up on to the shore with a flood tide, if it encounters the barrier boundary, it will fall into the barrier where it is free to move around.  When the tide ebbs, the crabs will leave the intertidal area utilizing the Tintersections. C. productus is the most damaging predator for intertidal oyster culture on tenures in the Gorge Harbour, in other areas of Cortes Island, and also in Baynes Sound.

The evidence of this predation is most obvious through visual confirmation (caught in the act) and the amount of shell debris left on the beach that has the telltale damage signature of C. gigas. Most of this knowledge is by word of mouth among oyster farmers; Cec Robinson, John Shook, Pat McDonnell, Fred Picard, Ian Winter, Helen Radosevicon Cortes Island, Gordy Mclellan, Clark Munro, Keith Reid, and others. These farmers have quoted estimates of 10% to 100% of stock mortality due to C. gigas.

Various techniques have been used to impede crab predation. Crab fences have been erected, usually a short 12-24" high net of PVC coated galvanized wire. Oysters are often grown in vexar (polyethylene) pouches, which are completely enclosed. The pouches can be grown either directly on the ground or used in a rack and bag technique. Some farmers have grown their oysters in an intertidal tray system that is secured to the beach. Others have grown their oysters high up the intertidal area, where they survive, but often don't have good meat quality. Some of these techniques would require written consent from an upland owner before this technique would be approved on a shellfish management plan. This requires an amenable upland owner, but still does not promise any long term security.  An upland owner is not obliged to provide this security to the oyster farmer, and is allowed this power through legislation called The Riparian Rights Act.

One technique that Below Sea Level Oyster Co (BSL) experimented with in August 2005 was the Crab Bucket. BSL had a one year scientific permit for this technique.

The bucket trap is a five gallon plastic pail sunk on the beach so that the top is level with the surrounding beach. The bottom must be cut out or at least perforated to allow water to rise and fall with the tide. The bucket trap was effective in trapping C. productus, often 6-10 crabs per bucket per day. But the buckets required regular and daily maintenance to remove the crabs and relocate them to deep-water.

It was critically evident that we needed some ability to control the crab predation but it had to meet several criteria:

• There has to be non-lethal control of the predator to satisfy DFO, the oyster farmer, and the general public;
  
• It has to avoid the riparian rights issue so that the solution would be transferable to other suitable locations;
  
• It needs to be cost effective for the oyster farmer to make the upfront investment of a system; and
  
• There must be minimal maintenance requirements for the system.

The bucket traps were discontinued due to the high maintenance of emptying the buckets everyday, but the effectiveness of the simple technique led the way to what is termed the Crab Barrier.

Phase 1

Phase 1of this project was completed in August of 2009. In phase 1four different sizes were evaluated based upon their ability to reduce the predation of C. productus on C. gigas in a cost effective and non­ lethal manner. Of the four sizes tested only the largest (12"depth) was shown to have a significant effect upon the predation rate. Unfortunately, the design of the crab barrier was not optimized and showed significant decrease in width overtime. This effect was so significant that the average width of the 12" barrier was actually 8".

Project Objectives

• Cost effective
  
• Non-lethal
  
• Reduce predation
  
• Increase structural rigidity of crab barrier

Description of Work

Scheduling

Fabrication of the barrier occurred during March 2010. The installation of the barrier occurred on June 3 and June 4, 2011.  The subsequent oyster trials occurred over four low tide cycles from June 11, 2011 to August 3, 2011.

Design

Through discussions with other oyster farmers and lessons learned in phase 1, two designs of the barrier were developed. In essence the crab barrier is a 'U' shaped trench liner. It was decided to go with a flat, solid base with a pattern cut out of the bottom which would provide the access for water to enter and leave the barrier. There was a two inch wide flange that ran along the topside of the barrier that provided rigidity to the unit.  There is a two inch wide flange at the end of each eight foot sect ion so the units could be attached end to end with stainless steel screws. The addition of a lip which extended out from the bottom of the barrier would provide a surface for the substrate to rest on making it more resistant to the effects of waves and tide.  Half the sections of both sizes of the barrier then had buttress reinforcements every two feet (Design A). The other half had solid metal bars bolted across the top every two feet (Design B).

Fabrication

Fabrication of the main barrier channels and the exit channels were done in Campbell River and transported to the site on Cortes Island. The crab barrier was fabricated out of fibreglass as it was most easily formed to fit the shape and barrier dimensions for this phase of the study.

Installation

The project required an application for a scientific license; this was to provide authority to approve any incidental capture, movement or mortality of C. productus during this experiment. We had several discussions with DFO about the project and the installation procedures. In April 2010 a CEAA screening was requested and then terminated in April 2011. Upon approval BSL was provided with guide lines similar to those set out in phase 1along with the letter of approval.

Our beach is a typical shelf beach at the five foot tide level. From the Hugh Water Mark (HWM) there is bedrock and very large rock down to the level shelf, then there is a fairly steep drop off to the LWM. The shelf section of beach is really a collection of boulders that provide the fabric of the beach, with a relatively thin layer of shell, gravel and fines covering the boulders.

The actual area where the barrier was to be installed was measured out and surveyors tape was tied onto rocks and used as a guide for where the machine would dig. There were numerous small boulders (9"-24" in diameter) that lay just under the surface that are disturbed in the excavation.  A small excavator was used to dig the furrow that our barrier would sit in. The excavator had a 24" wide bucket. Our excavator operator tried to be sensitive to the environment that he was working in. Often one rock would be removed at a time, with the thumb holding the rock against the bucket and the machine rotating to lay the rock down towards the LWM, with other similar rocks. The fines and gravel were deposited upslope to be used as backfill later. The excavator did most of the work digging the main channel and the exit channels every 16 feet. The field crew removed small rocks and other obstructions to fine tune the fit. The main barrier was laid in the furrow; it was then leveled to the existing beach substrate and held in place while it was back filled underneath by hand. Once one section was done, the next section was lined up and screwed to the previous piece.

The critical part was to be able to keep the barrier at the height of the natural beach. This was very tricky due to the fact that the excavator had to dump the material it removed right next to the furrow which obscured the level of the beach. This was solved in two ways, firstly by laying down rebar perpendicular and beside the furrow before the excavator moved the material. The other method of keeping the barrier level with the beach was looking at the striations in the sand on the side of the furrow for the top of the level.  The main channels were pre cut to accept the exit channels every 16feet. As required the exit channels were put into place. The exit channel sections were fabricated with the slope taken into account and a flange provided to screw together with the main channel.

The amount of excess material left over after the installation was very minimal. Over all the amount of rock that was taken out was equal to the volume taken up by the barrier. Therefore there was very little excess material to be spread out over the adjacent area, and no material brought in from off site.

Signage was installed around barrier to warn beach walkers of its presence as a hazard. There is very little foot traffic along the beach due to access and not being sandy.

First Observations

The installations were very successful in many regards.  One of the major successes was the ability to make the top of the barrier level with the rest of the beach. The appearance of the beach adjacent to the barrier returned to almost its pre-installation state within two tide days.

Trial Design

Location

Both sizes and designs of the barrier were installed on the north side of the Gorge adjacent to the mouth. The beach is at the five feet tide level. It is a level intertidal shelf beach with a steep drop off towards the LWM.  The beach is made up of a series of large rocks with pea gravel, shell fragments and sand filling in the spaces between the rocks. Both treated areas and the control area are contained within a similar area adjacent to one another on this beach.

Layout

The treated areas are directly adjacent to each other with a 10" barrier separating. The control area is beside the 12" barrier. The main portion of the barrier runs parallel to the LWM with exit channels attached every sixteen feet pointing towards the LWM. Side channels run up the side perpendicular to the LWM from the main barrier up to the bedrock near the HWM.  This effectively seals off the entire treated area, even between the two treated areas. The area of each treated section is approximately 4096 ft 2 •

Setup

The treatment area and the control area each had 300 oysters planted in their plot. The test plots were 10' wide by 5' long. The treatment plot was about 10' away from the barrier and the control plot was a similar layout on the same shelf.

The oysters were laid out in five rows of 60 oysters parallel to the LWM. The oysters were from our deep water site (license #105454) in the Gorge Harbor and were a small grade, 3-4" in length. All of the oysters used in the trial were harvested from tray culture and put in vexar pouches on the beach a week previous to the start of all of the trials.  The oysters used had very soft shells and had some fouling on them when put onto the beach for the trials.

The test plots were set up on the first day of a low tide series that had a low tide below the five feet tide height. The plot was counted every day during the low tide cycle. After the count everyday all dead oysters were removed and then replaced with oysters from pouches that were harvested at the same time.  The trial was completed when the tide no longer allowed access to the beach. The next trial would begin when the beach was next exposed. At the beginning of each trial all of the oysters in the plot were removed and replaced with fresh oysters from the pouches. The number of days in a trial did vary as the tide cycles varied.

Results

Barrier Redesign
The 10" design A barrier was found to be on average 9.7" wide whereas the design B barner was found to be on average 10" wide.  The 12" design A barrier was found to be on average 11.7" wide whereas the design B barrier was found to be 12" wide.

Crab mortality
Over a three month period two crabs were found dead in the barrier.  This occurred on June 24, 2011 and August 2, 2011.

Oyster Counts
Four trials were completed over a 53 day period. The trials were executed from June 11-21; June 26-July 5; July 9-19; and July 25-August 3, 2011. The average percent daily mortality for the 10", 12" and control areas was found to be 2.55%, 1.07% and 5.92% respectively.

T-tests were performed on the data.  A T-test is utilized to compare two sets of data and determine whether there is any chance that they are the same. Once a T-test is performed a number is generated called the P value which is the probability that the two samples are different. A P value of less than 0.05 is generally accepted as a definitive result in favor of there actually being a difference.  From the data collected on daily mortality, two-sample T-tests assuming unequal variances were conducted comparing the 10" barrier and the control, the 12" barrier and the control and the 10" barrier and the 12" barrier. The P values for the comparisons are 5.88E-14, 2.42E-20 and 9.91E-09 respectively.

Discussion

Barrier Redesigns
As evidenced in the results both designs of barrier retained the desired width.  The conclusion is not so simple because each of the designs has its own strengths as well as drawbacks. Design A may in the future decrease it's width to such a point as to become non-functioning because it is not as sturdy as design B. Another problem with it is that it is more substantially more expensive to fabricate because of the extra time and materials involved with making the buttresses. The major benefit to this design is that it is easy to clean if too much silt builds up in the barrier.  A shovel can be run along the bottom scooping out most of the obstruction. Design B will never, barring a complete disintegration of the barrier, decrease in width because of the aluminum bars across the top.  Another benefit of this design is that it was cheaper to manufacture be cause the bars can be bolted on at any point before they are installed and the material and time are much cheaper. The only problem with this design is the complete inability to clean it efficiently. Because the bars across the top inhibit any shovel being able to scoop out the detritus it will take a large amount of labor to clean any significant section of it. The ultimate and deciding factor is the cost of fabrication of design A ultimately will far outweigh the cost of cleaning design B because siltation occurs no matter where it is installed.

Crab Mortality
Only two crabs were found dead in the barrier. The barrier was checked every day for crabs during the trials and every couple of days after the trials were finished.  It is impossible to tell whether the crabs perished in the barrier or on the beach and were washed into the barrier on the ebb tide.

Oyster Counts
Phase 1 determined that the 12" barrier had an effect upon the mortality of the oysters unfortunately the 12" barrier wasn't actually 12" across which is why phase 2 was undertaken. As has been previously stated the barriers did hold up to their manufactured dimensions. This part of the project was designed to determine whether there was an optimum size and as well as make sure that the results we obtained in phase 1 were in fact accurate.

The 10" barrier had an effect upon the mortality of oysters. With a P value of 5.88E-14, which is considerably less than .05, there was a statistical difference between the area enclosed by the 10" barrier and the 12" barrier.  There is a similar story for the 12" barrier as well.  The area enclosed by the 12" barrier as compared to the control area gave a P value of 2. 42E-20. Here there is a statistically significant difference between the two areas. The more important value that was obtained was in the comparison between the mortality rates in the 10" barrier and the 12" barrier.  A P value of 9.91E-09 which is also considerably less than 0.05 was obtained in this comparison which means that there was a statistically significant difference between the two barriers. In this case the 12" barrier was more efficient than the 10" barrier at keeping the C. productus at bay. There were only two sizes of barrier tested in this project so it is difficult to predict the optimum size of barrier.  If we had tested a 14" barrier and it had shown a statistically significant difference than the 12" barrier the n that may be the optimum size. That being said the 12" barrier had an average daily mortality rate of 1.07% with a standard deviation of 1% which would be very hard to beat given the natural variation present in all of these trials.  That means that without testing a 14" barrier we can say that the 12" barrier is the optimal size.

Visual and Physical Disruption

Visual Aesthetics
The greatest problem with the aesthetics of phase 1 was the thumb tacks that held the barrier in place. They stuck up above the beach 4" and were spaced every 2'. There were approximately 900 of them used in phase 1. Through the removal of these we have effectively reduced the visual impact of the barrier by a great deal because there is nothing above the level of the beach in phase 2. Note that the barrier is not readily visible from the fore shore, barely 3 months after its installation. As time progresses and the barrier plays host to more marine organisms it will become less and less noticeable.

Habitat Disruption
The use of an excavator on the beach disrupted some habitat. The driver of the excavator attempted to disrupt as little of the beach as possible.  With a 24" wide bucket the excavator dug a furrow approximately 36" wide and 16" deep. Substrate was moved out of the furrow in such a fashion as to put like by like. This means that large rocks were put closer to the LWM where there were other similar rocks and fine material towards the HWM. Because the vast majority of the rocks were completely buried this practice made the end result look as close to the pre -installation beach as possible. The movement of the rocks also had the added side effect that there was neither excess material leftover, nor any brought in.  Within a couple of tides there was barely any visual evidence of the installation process.

Biodiversity  Comparison
One of the concerns that the barrier brought up was the fact that it would change the species cross section in that specific area aver a long term period. The barriers used in phase 1 of the project were installed October 2008 and so would qualify as being long term installations. A survey of all the organisms in the barrier was undertaken and then compared to a similar adjacent area. Using the Simpson biodiversity index, the treated area had a number of 0.480066 and the untreated area had a number of 0.438957.  These numbers are sufficiently similar to say that amount of biodiversity did not change when the barrier was installed. The species makeup did change. There was a shift from surface dwelling species such as Fucus gardneri, Mytilus edulis and Crassostrea gigas to species that require shelter such as Pogurus hirsutiusculus, Hemigrapsus oregonensis and Oligocottus maculosus.

Conclusion

Barrier Redesign
The two designs have both advantages and disadvantages making them able to better serve specific areas.  Design A is better suited to an area of low flow where more cleaning needs to take place.  Design B is better suited to high flow areas where less detritus needs to be removed from the barrier.

Crab Mortality
The level of crab mortality due to the barrier was sufficiently low as to satisfy the  non-lethal requirement put forward by the DFO.

Oyster Counts
Both the 10" and 12" barrier were proven effective at decreasing the amount of predation by C. productus. The 12" barrier was found to be most efficient at reducing predation.

Cost Effective
The savings was calculated to between $21.21/day and $345/day in an area of 4096 ft2.  With a cost of manufacturing cost of $15/ft and an installation cost of $15/ft for the barrier a total cost for that area would be $1920.  This would mean that the owner would recoup the barrier investment in between 5.5days and 90.5 days. The variability in this number is due to the fact that C. productus feeding behavior was not studied and is very difficult to predict.

Acknowledgements

Thanks to the following people:

•           Matt Swaile for authoring the paper and being the field technician
•           Cam Creighton for assistance during the installation and trials
•           Brett Melvin, Brett Tracy and Martin Nikleva, installation field crew
•           Robbie Driediger for his use of the excavator
•           De bra Hughes for help with the CEAA screening
•           Shelly Jepps, DFO, for facilitating the approval of the project
•           Sean Irvine, DFO, for assisting in the development of the AIMAP application and subsequent contribution agreement.
•           DFO, AIMAP for funding the project

Appendix 1

Comparison of biodiversity in the untreated area against treated area three years post installation.

Objective
To determine whether there w as a change in the amount of biodiversity a s well as a shift in t he species present on the barrier.

Introduction
One of the concerns expressed with this project was the fact that over a long period of time the ecology of the area disturbed would change detrimentally. A change in the ecology was always predicted but whether this change is detrimental would be the important factor. A detriment to the ecology would be defined as a decrease in the amount of biodiversity in the area as well as an undesirable change in the species cross section.

Method
Quadrat sampling using a random number generator was undertaken. The size of the quadrat was a square of 50cm on each side. A number was assigned to the barrier every 50cm. Five samples were taken of both the treated and untreated sites in which every surface organism was counted. Sampling of the untreated section was done by drawing a line parallel to the LWM at the same height and adjacent to the barrier that was the same length as the barrier.  Every surface organism was counted and marked down in the quadrat.

Result
The Simpson biodiversity index was used to quantify the biodiversity of the areas. The untreated section generated a number of 0.438957 whereas the treated section was 0.480066. The difference between the two areas is 0.041109.

Discussion
The Simpson biodiversity index is the most commonly used biodiversity index used in modern ecology surveys. It takes into account both the number of species present as well as the evenness among species.  A difference of 0.041109 is negligible. A difference that small means that the biodiversity did not change when the barrier was introduced.

A different species cross section between the two areas was noted.  In the treated area we detected the presence of species that require more shelter in order to thrive while in the untreated section those species were not present.  In the treated section more species of crabs were present such as Elassochirus tenuimanus and Pagurus hirsutiusculus. In the untreated section these species of crabs were not present but species that do not require shelter were present such as Mytilus edulis, Fucus gardneriand Crassostrea gigas.  This is easily explained because of the nature of the barrier itself. The barrier provides shelter for organisms both in the bottom as well as on the sides. The shelter is from desiccation as well as fast moving water. The barrier provides shade to the organisms in the bottom of the barrier which means that they can stay moist for longer than on the untreated areas. Because the beach is flat the water can move very quickly both onto and off of the beach which if not sheltered may displace some of the organisms.

Conclusion

A detriment to the ecology of the beach was defined as a decrease in the biodiversity or an undesirable change in the species cross section. With a difference of 0.041109 between the treated and untreated areas with regards to the Simpson biodiversity index the level of biodiversity did not change between the two areas. The change in cross section of the species present was not determined to be detrimental or unusual.