Investigation Into The Cause Of The Clam Decline: 2013

Downeast Institute Field Studies – Freeport, Maine

Jump to:  Intro     2014 Results     2015 Results     2016 Results     2017 Results

Video from the Portland Press Herald.

Summary

In 2012, members of the Maine Clammers Association shared their alarm with local officials about the dramatic increase in invasive green crab populations in Casco Bay and the corresponding decrease in soft-shell (steamer) clams. Immediate action was needed to safeguard this economically and culturally important food source.

In 2013, the Town of Freeport took the initiative to address the problem, approving a total of $170,000 to enable its municipal shellfish program to gather information about green crabs and declining clam stocks. A portion of those funds was used to conduct two field studies – green crab trapping and green crab fencing. Downeast Institute (DEI) was one of the scientific organizations contracted to help design field studies and analyze the data, and worked collaboratively with clammers to install the studies. The experiments examined if green crab trapping, netting, and/or fencing could reduce green crab predation on young-of-the-year clams (i.e., spat, “recruits,”). DEI also conducted a third study to examine the interactive effects of clam size and types of predator netting on the survival and growth of clams.  

The fencing experiment was compromised due to lack of routine maintenance and was repeated in 2014. Trapping yielded 13,065 pounds of crabs, with an average size of 59.5 mm, 72% male and 28% female. Approximately 1% of females measured were bearing eggs. The third predator exclusion experiment showed that less than 2% of clams survived in unprotected experimental units, whereas the average rate of clam survival was as high as 55% when protected from predators.

In order to ensure as much confidence in our results as possible, experiments were designed to be statistically valid.  The details, experimental design, and results can be found below.

Experiments & Results

#1: Green Crab Fencing

Green Crab Exclusionary Fencing

During the 1950s, when Maine experienced a period of warming seawater temperatures and a corresponding rise in green crab populations, state biologists installed fences in coves (Bremen, Jonesport) in an attempt to protect clams from green crab predation (Smith et al., 1955). However, no reliable data about the effectiveness of the green crab exclusionary fencing exists from the 1950s.  During 2013, DEI designed and deployed a study to determine if fences installed in the intertidal could successfully deter crabs, thus increasing clam survival behind the fencing.

Freeport clammers volunteered their time through a program called “conservation hours”, in which clammers are required to volunteer 12 hours per year in order to renew their town clamming license.

Little River and Recompense Coves were chosen for deployment of two different green crab fencing experiments designed to help clammers understand the effectiveness of each method in excluding predators. For the first method, clammers installed a 2,100 ft long fence across the mouth of Recompense Cove in the hopes that completely closing off the cove would deter green crabs, resulting in an enhancement of wild seed clams that could be seen, measured, and documented in the fall.  See 1st picture below for a map of the Recompense fence.

At Little River a more complex experiment was deployed to determine the success of various predator exclusion methods (fencing vs. netting vs. controls) in enhancing clam recruits (see second image below to view the schematic of the 2013 Little River site). See page 14 of Final Report for more details about the study design.

Fencing was deployed at Recompense and Little River Coves on 26-27 July 2013, with assistance from Freeport clammers. A survey was conducted at both locations to measure the initial baseline densities of clams (see Final Report). The fences were pre-fabricated by clammers in 8-, 10-, and 12-ft lengths and taken by airboat to both flats (see photo album below for pictures of the process of building and installing the fences).

It was planned that the experiments would be sampled in November, due to the fact that soft-shell clams spawn in the spring or early summer, swim and develop in the water column as planktonic larvae 3-4 weeks, then settle to the mudflats. By October or November, the juvenile clams (spat) that have survived typically have 2-8 mm shells and are easily seen.  Spat settlement density can be measured by taking uniform bottom samples that are washed through fine- mesh screens.

Upon sampling both fenced sites on November 16 and 17, it was unfortunately discovered that a lack of routine maintenance resulted in a loss of the fence’s structural integrity. This meant that the effects on fencing to deter green crabs could not be measured properly. To remedy this, DEI was able to secure funding to repeat the study in 2014.

DEI did complete final sampling at all three tidal heights inside the fence at Recompense (26 benthic cores per tidal height), as well as sampling under the netted and unnetted plots. The results showed that the mean number of clams in the fully-netted plots was nearly an order of magnitude greater (1.7 individuals per core, or 92.3 individuals per square meter) than the mean of the other five treatments. 

It was also found that the density of 0-year class (2013 clam ‘recruits’) individuals was nearly 10x greater in the netted plots compared to the plots without nets or in “fenced” structures.

Images (click to view slideshow)

  • Approximate position of the Recompense Cove fence that was erected July 27 and 28, 2013 and completed August 6 when the last section was installed on Indian Island Shore.
  • Schematic of the 2013 Little River fencing experiment.
  • Mesh netting was stapled to a section of green crab fencing.
  • Aluminum flashing was added to the top of each section of fencing. The flashing is too slippery to provide a foothold for crabs to climb and keeps them out of the fenced areas.
  • Sections of fencing prior to deploying at Little River flat.
  • Sections of fencing at Little River, Freeport, Maine.
  • Thanks to Capt. Weston Watts, a volunteer clammer from Harpswell, we were able to use an airboat to take sections of fencing to the field sites. This saved us from transporting them by hand across the mud.
  • 30-ft x 30-ft fenced plot at Little River. Fences are 18-inches tall, and the mesh netting has 1/2-inch apertures. It took about three hours to deploy and construct six fences.
  • Flexible, plastic netting (1/6=inch aperture; 14-ft x 22-ft) was also used to deter green crabs from the mudflats. Five styrofoam floats under the netting lifted it during high tide. (Otherwise, soft sediments tend to build up on the top of the net, depleting oxygen that clams need in the mud beneath.) In November, samples were taken in both the netted and fenced plots to determine the number of wild spat (2-10 mm soft-shell clams) that had settled (recruited) to these sites.
  • Closeup of a netted plot at Little River flat in Freeport Maine, showing large numbers of mud snails, Ilyanassa obsoleta, crawling on top of the net.
  • Sections of fencing being deployed at the mouth of Recompense Flat (28 July 2013).
  • Due to the threat of entrapping endangered Atlantic sturgeon, Acipenser oxyrinchus oxyrinchus, and/or short-nosed sturgeon, Acipenser brevirostrum, the Army Corps of Engineers required that every 150 feet of fencing have a 2-ft wide gap, as well as one 10 ft gap. Although crab traps eventually were placed at or near these openings, the gaps prevented us from being able to close the cove completely.
  • Sections of fencing being stapled together to ensure a continuous line.
  • Unfortunately, by mid-September, routine maintenance of fencing at Little River and Recompense ceased, resulting in erosion at the fence/sediment interface, as well as compromised structural integrity of the above-ground portion of the fence.
  • Erosion on inside of fenced plot at Little River Flat (12 October 2013).
  • Gaps between the fencing and the surface of the flat allowed green crabs to enter the plots.
  • A large female green crab found within a fenced plot at Little River Flat (12 October 2013).
  • On 16 November, 2013, the fenced plots, netted plots, and control plots were sampled. Unfortunately, most of the fenced plots had collapsed or were missing sections, allowing green crabs and other predators unfettered access to the “fenced” areas.
  • Most of the fences had at least one collapsed section or side by November 2013. Regardless, samples were taken from inside these “fenced” plots and the contents of each were washed through a 1 mm sieve to determine the density and size of wild clam spat.
  • The netting is peeled back to allow samples to be taken.
  • Here, the netting was lost within days of the 16 November 2013 sampling. Nonetheless, samples were taken from this “ghost net” and results compared to the plots that remained netted for the entire field trial.
  • At Recompense Flat, samples were taken via airboat on 17 November.
  • A view of the fence at Recompense, showing the Army Corps of Engineers required sturgeon gap and green crab trap. While some sections of fencing at Recompence Flat were intact, many sections had eroded so that crabs and other predators could have burrowed, swum, or otherwise managed to counteract the fencing.
  • Sections of fencing missing from Recompence Flat. These are large enough for any clam predator to access the flats behind the fencing.
  • More sections of collapsed fencing at Recompence Flat (17 November 2013).
  • Samples were taken at three tidal heights between the fencing and the shore. Few soft-shell clam spat were found in any of the 72 samples taken on 17 November 2013.
  • Mean number of soft-shell clam juveniles that occurred in benthic cores in Experiment I & II at Little River flat, Freeport, Maine on 16 November 2013. The black, horizontal reference line through five of the treatments is the overall mean for those treatments (0.175 individuals per core, or 9.5 individuals m-2). The mean clam number in the fully netted plots was nearly an order of magnitude greater (1.7 individuals per core, or 92.3 individuals m-2) than the mean of the other five treatments.
#2: Green Crab Trapping

Green Crab Trapping

The second study the clammers were interested in trying was green-crab trapping. This was conducted to quantify populations of green crabs at different geographic locations, as well as intertidal and subtidal sites.

The study was initiated during the last week of May 2013 and continued through early November. Fifteen clammers participated in trapping, with almost half (7) fishing their traps regularly during this period. Information collected from the trapping studies included the effect of “soak time” (length of time the traps were fished) on catch mass, whether there were differences in crab abundance between traps fished in the intertidal vs. subtidal, and how sex ratios and size frequencies varied with times and locations.

Over the course of the experiment (162 days), 300 hauls of 1-10 traps per haul were completed by clammers. This resulted in a harvest of 13,065 pounds (about 6 metric tons). Of those, 11,715 crabs were measured and the sex of each was determined. The average size (carapace width) of the crabs was 59.5 mm (2.35 in), and the overall sex ratio was 72% male to 28% female. Approximately 1% of females measured were ovigerous (holding eggs), and the last day that an egg-bearing female was trapped was August 22.

Interactive Effects of Clam Size and Type of Predator Deterrence

The third study was designed to determine the interactive effects of clam size and predator netting. A field experiment was deployed at both Little River and Recompense flats on August 18, 2013.

The three types of predator deterrents a) open pot – the rim of netting around the periphery ensures that clams remain enclosed in the six-inch diameter area, but does nothing to deter predators; b) a pot that is covered in two types of netting – a piece of 6.4-mm extruded (hard) plastic netting closest to the pot, held in place with a piece of 4.2-mm flexible netting; c) a pot that is covered with a piece of 4.2-mm flexible netting.

Experimental units (6-in. diameter plant pots) were placed in the flats. Mud was added to the experimental unit (EU) and twelve clams of different sizes were placed inside each plant pot. The clams were one of three sizes: 8.2 mm, 14.2 mm, and 19.4 mm.  Three types of predator deterrents were used: a rim of netting around the pot, a pot covered in two types of netting, and a pot covered in one kind of netting (see photo).

Every combination of the 3 clam sizes and 3 predator exclusion methods was used to form nine treatments, either unprotected or protected with one of two types of netting. Five replicates of each treatment were deployed at each site. 

After 90 days at Little River, an average of only 1.7 ± 1.9% clams survived in unprotected pots, whereas the average rate of survival was 55.6 ± 11.3% when clams were protected from predators. This demonstrates how important predator exclusion is to clam survival.

The results from this particular study were published in the journal Maine Pollution Bulletin in 2015.

To review the results of all three experiments, see the Final Report to the Freeport Town Council that was completed in early January 2014.

Images (click to view slideshow)

  • Locations where clammers hauled green crab traps in 2013
  • Green crab trap in the intertidal at Pettengill Flat in the Upper Harraseeket River (27 June 2013)
  • Crushed, commercial-size soft-shell clams were used as bait for the entire study. Clams were used as bait because it was thought that other forms of bait may attract atypically high numbers of green crabs to the traps. For the experiment we wanted to create as-close-to-natural conditions as possible, so used clams as they already inhabited the areas where the traps were fished.
  • Traps were fished in groups of five at ten places in the Upper Harraseeket River, and the contents from all five traps were pooled into a single tote that was weighed.
  • Chad Coffin (left) and Clint Goodenow (right) removing green crabs from an Acer green crab trap in the Harraseeket River.
  • Here, a tote of green crabs representing the total amount from five traps is weighed. This information was used to determine if catch-per-unit-effort (CPUE) changed through time, and whether there were any differences in CPUE from one place to another.
  • Over the season, carapace width and gender of green crabs in a 10% (by weight) subsample from each tote (catch from 5 traps) were recorded.
  • Female green crab with a just-released egg mass (27 June 2013).
  • Clammers also discovered that the edge of some upper intertidal areas colonized by salt marsh grass, Spartina alterniflora, provides refuge for many green crabs. The crabs burrow into the roots and rhizome structures of the plants and carve out elaborate subterranean burrows and galleries.
  • Few large males (typical sex ratio is 8 female: 1 male).
  • As green crabs continue to create burrows that kill the underground roots and rhizomes of the marsh grass and seawater enters these burrows each day, large portions of the bank eventually begin to erode. Over time the banks slough onto the flats. Erosion of this type is a major issue for adjacent landowners and municipal planners.
  • Piece of salt marsh bank on a mudflat near Flying Point, Freeport, Maine (28 June 2013).
#3: Effects of Clam Size and Netting on Survival

Interactive Effects of Clam Size and Predator Protection (Netting) on Clam Survival

The third study was designed to determine the interactive effects of clam size and predator netting. A field experiment was deployed at both Little River and Recompense flats on August 18, 2013.

Experimental units (6 in. diameter plant pots) were placed in the flats. Mud was added to the experimental unit (EU) and twelve clams of different sizes were placed inside each plant pot. The clams were one of three sizes: 8.2mm, 14.2mm, and 19.4 mm.  Three types of predator deterrents were used: a rim of netting around the pot, a pot covered in two types of netting, and a pot covered in one kind of netting (see photo).

Every combination of the 3 clam sizes and 3 predator exclusion methods was used to form nine treatments, either unprotected or protected with one of two types of netting. Five replicates of each treatment were deployed at each site. 

After 90 days at Little River, an average of only 1.7 ± 1.9% clams survived in unprotected pots, whereas the average rate of survival was 55.6 ± 11.3%  when clams were protected from predators. This demonstrates how important predator exclusion is to clam survival. The results from this particular study were published in the journal Marine Pollution Bulletin in 2015.

Images (click to view slideshow)

  • The three types of predator deterrents a) open pot - the rim of netting around the periphery ensures that clams remain enclosed in the six-inch diameter area, but does nothing to deter predators; b) a pot that is covered in two types of netting - a piece of 6.4 mm extruded (hard) plastic netting is closest to the pot, held in place with a piece of 4.2 mm flexible netting; c) a pot that is covered with a piece of 4.2 mm flexible netting
  • Chad Coffin, President of Maine Clammers Associations, inspects a matrix of 45 pots at Little River Flat on October 13, 2013.
  • These clams had about 20 mm shells when they were planted in protected pots at Little River flat. The visible hatchery mark makes it easy to measure the new growth that occurred between August 18 and November 16, 2013.re the "Large" clams (initially about 20 mm in shell length) that were planted in protected pots at Little River flat. The experiment ran from 18 August to 16 November 2013. The "hatchery mark" is clearly visible at a size of around 20 mm, and all new growth can easily be measured.

Summary of Final Results:

The studies conducted in 2013 showed that green crabs had become the major predator of soft-shell clams in the Harraseeket River and adjacent intertidal areas extending to the east of Wolfe’s Neck. Clams as large as 20 mm (0.79 in) in shell length (SL) that were not protected by netting, fencing, or some other deterrent, were quickly consumed. However, it was possible to deter green crabs’ predatory activities in small, routinely-maintained areas, enhancing both wild and cultured soft-shell clam juveniles. As the third experiment showed, as well as the Little River fencing and netting experiment, netting (i.e., plastic, flexible with 1/6- inch (4.2-mm) aperture) can enhance wild clam survival. 

Trapping during summer and fall can provide information about green crab abundance and population dynamics that can be used to make more efficient decisions about future attempts to capture crabs. 

Based on these results, the following next steps were recommended to Freeport town officials: 

  • Continue to trap green crabs in discrete areas of the Harraseeket River to determine if trends (e.g., Catch Per Unit Effort (CPUE), frequency distributions of male and female crabs; sex ratios) observed in 2013 continue in the future.
  • Re-examine the effects of fenced plots vs. control plots on the abundance patterns of 0-year class clam recruits. Fencing would need to be properly maintained and plots positioned to take advantage of tides and currents.
  • Re-examine the use of netted plots to enhance wild recruits and protect cultured seed to determine whether these are cost-effective measures that communities or individuals could use to enhance existing stocks of clams.

Final Report:

To review the results of the green crab trapping, fencing, and small-scale predation/clam size experiment, see the Final Report to the Freeport Town Council that was completed in early January 2014.

References

Smith, O.R., Baptist, J.P., Chin, E. 1955. Experimental farming of soft-shell clam, Mya arenaria, in Massachusetts, 1949-1953. Comm. Fish. Rev., Volume 17, No. 6, 1-16.
, , , , , , , , , , , , ,
Recommended Posts

Start typing and press Enter to search