Northampton Shellfish Mortality Response

Generating Scientifically Defensible Estimates of Mortality in Shellfish Farm Grow-Out

These methods of estimating mortality are provided by researchers at the Virginia Institute of Marine Science (VIMS) as guidance for rigorously assessing mortality across a farm. These are methods that would be used in our own research to meet our scientific standards. If you have crop insurance, please refer to your insurance provider for any requirements.

After a sudden mortality event on a shellfish farm, a grower may need accurate and defensible estimates of losses. As any grower knows, the natural (background) mortality rate of shellfish species varies by size as does the value. That means that estimates of mortality must be done by size class. The USDA Farm Service Agency (FSA) has established size classes for both oysters and clams and it is advisable to determine estimates of mortality for each of the size classes on any given farm. Here guidance is provided for oysters given recent reports, but similar advice for clams can be made available.

Oyster Categories on Farms in inches (with smaller sizes of 1-4 mm and 5-12 mm also listed):

• ½" to 1"
• 1" to 1 ¼"
• 1 ½" to 3"
• 3" +

When?
Assessment has no strict timeline, but generally assessment should be done within 2-3 weeks after the event. While additional losses may occur over time, assessments done closer in time to the event are more clearly linked to the event.

Units?
For off-bottom gear, decide what your units are (cages, bags, etc.). Then decide what area you're sampling; this could be the whole farm or different areas on the farm. Do this if you have any reason to believe that mortality rates were different across different areas.

How Many?
Within the area(s) that you’re assessing, determine how many units (bags, cages, etc.) there are for each size class. There is no magic percentage to sample but it is critical that the selection of units is random. You can find random number generators online or through an app. Document how the units were selected. This will avoid any bias in the result.

For estimation, it is recommended to sample at least 20 units or 2% of the total number for each size class for each section that you’re assessing. If the percent mortality is very different among units (e.g., some are at 25% and some are at 80%), increase the number of units, up to 5-10% of the total number of units that are being sampled to get a better idea of the average.

What Are You Counting?
For any closed unit (bag, cage), note the size and area (lease number and specific location on the lease if applicable) that you’re sampling and document with a picture. Empty the bag out on a surface and get a total count of live and freshly dead oysters. Recent dead will typically have the two shells still attached (‘boxes’). Make note of the number of other dead that don’t appear to be freshly dead - for example, those with inner shell surfaces already fouled by biofilm or colonizing invertebrates--as these won’t be counted in your estimate of mortality from the event. Record the totals for the unit where percent mortality for the unit will be calculated using the formula below. Repeat for each unit for that size and area. After 20 units are done, decide if additional units should be assessed based on how much the estimates differ.

Percent Mortality for any Unit = (# of Fresh Dead / (# of Fresh Dead + Live))*100

Repeat this process for each size class that you have in each area. Determine the percent mortality in each size class for each area by averaging the estimates for each bag.

*If the total number of live and fresh dead oysters in a unit (bag, cage, etc.) are too many to count in a reasonable amount of time, you will need to sub-sample. One way of doing this is to divide the total contents into approximately equal piles with a divider (without regard to how the live and dead oysters are distributed) and again randomly choose one of the piles and use the method above to estimate the percent mortality in that pile as an estimate of the percent mortality in the unit.

What else will you need?
Provide as much evidence as you can of what counts were in the bags immediately before the event. This may be from the last time that you stocked them.

Example
A grower has two areas and she thinks there might be one area that was hit harder. She calls them A and B. In Area A, she has seed oysters that are ¼” to ½” and ½” to 1” (two size classes), where the smallest seed are in 10 bags and the larger seed are in 150 bags. In Area B, she has 1” and up to market size oysters, where the 1” to 1 ¼” are in 40 bags, the 1 ¼” to 3” oysters are in 5,000 baskets and the oysters over 3” are in 4,600 bags.

She starts with Area A. Since the smallest seed (¼” to ½” in 10 bags) are in less than the recommended 20 bags, she counts the live and freshly dead oysters in all 10 bags. In the ½” to 1” seed oysters (in 150 bags), she calculates that 2% would be 30 bags (which is over 20) so she randomly chooses 20 bags from Area A. When she’s done, she notices that some bags had 10-20% mortality, while most had 60-80%. She decides she wants to sample more bags to get a better estimate of the average and decides to sample 25 more bags (45 total for this size from Area A, or 3% since she saw so much variation).

In Area B, she samples 20 of her 40 bags of 1” to 1 ¼” oysters and gets consistent numbers so moves on to the next size class. For the oysters that are 1 ¼” to 3” (held in 5,000 bags), she starts with 20 bags and again gets consistent mortality across those, so she moves on to the next size class. For the largest oysters (3”+, held in 4,600 bags), she starts with 20 bags again, but gets a lot of variation. She calculates that 2% would be 92 bags so she opts to do live/dead counts in another 26 bags to get to 46 bags total (or 1% of the bags). She gets a lot of variation still so doubles the number to 92 bags so that she has randomly sampled 2% of the bags.

In total, she counted live/dead oysters in 10 + 45 + 20 + 20 + 92, or 187 bags. She photographed each bag and kept careful notes.

She calculates percent mortality in each bag and then gets the average for each size class for each area. She documents what totals she had before the event for each size class and area. Total losses can be calculated by multiplying the average percent mortality by the pre-event inventory for each size class and area.

At this point, she can discard dead oysters but documents this with video to show that qualitatively the bags that weren’t sampled were similar to what was sampled. If you have crop insurance, be sure to first check with your adjuster before discarding dead oysters.

For any questions about the sampling protocol, please reach out to Dr. Bill Walton - walton@vims.edu - (804) 684-7238

Dear shellfish and fishery producers,

Thank you again to everyone who joined us last night. The level of loss many of you have experienced from the recent harmful algae bloom is deeply concerning, and we want to acknowledge the emotional and financial strain this event has placed on your operations. Your willingness to share candidly is essential as we work together toward coordinated solutions.

Based on our discussion, we want to provide a clear update on next steps and points of action as well as key summary of discussions:

IMMEDIATE ACTION POINTS:

• VIMS is finalizing a unified sampling protocol for loss assessments. This will include guidance on timing, sample handling, and documentation. We expect to share this protocol as soon as possible.
• Virginia Cooperative Extension leads agricultural damage assessment for the Commonwealth of Virginia. To support this effort, please report your estimates of losses to my office  (by email, in person, or over the phone). We must provide the best loss estimate that we can, as soon as we can.  This figure will likely change over time as this often occurs during chronic situations. We will then continue to revise and report figures as information evolves.   It does not need to be exact right now.  These reports are kept anonymous by VCE. Individual grower names or identifying details are not shared. Instead, all information is aggregated at the county level and provided to Northampton County leadership and the local FSA office.

County‑level summaries are also shared with the VCE Associate Director, who compiles statewide estimates and communicates them to the Chair of State Emergency Support Function (ESF) 11 within VDEM, the Commissioner of Agriculture, the Office of the Secretary of Agriculture and Forestry, the Office of the Secretary of Natural and Historic Resources, and the State FSA Executive Director. These combined estimates help determine whether local or state‑level response or relief actions are warranted.

Because VCE reports are anonymous, producers also need to report directly to the local FSA office. FSA must collect individual producer information in its data management system to support any potential federal relief programs. VCE participates on the FSA State Emergency Board to ensure consistency in damage assessment data and reporting throughout the process.

I understand that this process can feel frustrating, but the need to report both anonymously through VCE and individually through FSA reflects the reality of how state and federal systems differ in their requirements for documenting agricultural losses. Accurate reporting—both anonymous reporting to VCE and individual reporting to FSA—is essential to document the true extent of agricultural losses and to support any potential local, state, or federal assistance efforts.

ADDITIONAL SUMMARY OF KEY POINTS FROM LAST NIGHT’ S MEETING

• We discussed the need for stronger inclusion of aquaculture in state and federal disaster frameworks, along with clearer protocols for how HAB-related losses should be evaluated.

• Many of you highlighted the lack of viable insurance options for aquaculture and wild harvest operations. This gap will be elevated in our communications with state and federal partners.

• There is a strong need for better coordination among universities for HAB detection, sampling, monitoring, and alerting—particularly with an emphasis on pragmatic, actionable tools for producers.

• Similarly, improved coordination among agencies and academic institutions for reporting algal blooms is essential to ensure timely and accurate information.

• We also identified the need for broader public education to clarify that harmful algae blooms do not make aquaculture products unsafe for consumption. This will be an important part of our outreach moving forward.

We will continue to keep you updated as protocols are finalized and additional resources become available through email. We are also creating a centralized webpage where producers can access the most current information as it becomes available (under development). The link will be provided as soon as possible. This will serve as the primary hub for updates on sampling protocols, reporting processes, and HAB monitoring. Please continue reporting initial losses, and reach out if you have questions or need assistance. My complete contact information is located below. 

Thank you again for your time, your openness, and your commitment to working together through this challenging event.

Hélène B. Doughty, Associate Extension Agent (Agriculture & Natural Resources)