Using aquaculture to help mitigate impacts of harmful algal blooms on crustacean fisheries: accelerating depuration to produce safe and marketable seafood products

Award Period
Award Amount
Agency Name
UC San Diego
Award Number
PI First Name
PI Last Name
MSI People
Area/s of Research
Ecology and Evolution
Marine Conservation, Policy and Education
Natural Marine Resources

To assess whether depuration may provide a cost-effective means for producing a safe, live, whole product we will conduct a series of laboratory experiments evaluating the potential acceleration of depuration through four parameters: 1) feeding; 2) temperature; 3) emersion; and 4) cumulative parameters. Our null hypothesis for each experiment is that the parameter will not accelerate depuration. We will first test if feeding increases the depuration rate using three treatments; no food, a frozen seafood mix that includes mussels and clams that we and others often use when holding crabs and lobster, and a commercially available pelleted diet used in crab aquaculture (Turbo® 49.7% crude protein, 6.7% lipid, CP Feeds Thailand; Truong et al., 2008).

The no food treatment will serve as an experimental control and as a baseline assessment for depuration rates of each species. The feeding experiment will be followed by a second experiment evaluating whether depuration can be accelerated by water temperature. For this experiment, test animals will be exposed to three water temperatures appropriate for the test species; red rock and Dungeness crab, 9°C, 14°C, 19°C; lobster, 12°C, 17°C, 22°C. Third, we will determine the effect of emersion on excretion rate by exposing test animals daily to one of three emersion periods (0, 30 and 60 minutes) followed by re-immersion. Our fourth experiment will entail combining a subset of the most promising treatments for each species to identify optimal (fastest and most cost-effective) depuration methods.

We will conduct these experiments using methods similar to those used by Lund et al. (1997).

Prior to each experiment, test animals will be dosed with DA by feeding them razor clam (Siliqua patula) meat with quantified, high levels of DA. Razor clams containing high DA levels are typically available along the west coast because they retain DA for very long periods of time.

Samples will be collected throughout the project period and frozen until used in the various experiments. Intentional dosing of crabs and lobsters will ensure the animals have been exposed to DA in a consistent manner, as opposed to collecting crabs and lobster from the wild during and after DA-producing blooms and assuming they contain similar DA concentrations.

For the factorial experiments, individual test animals will be assigned randomly to a treatment with a 6 replicates per treatment – the same number of individual sampled by the California Department of Public Health (CDPH) when evaluating DA levels in crabs and lobsters to inform seafood health advisories and fishery closures (and the same sample size used by Lund et al. (1997)). That said, if variation in DA levels among animals fed DA-contaminated clams in the laboratory necessitates increased replicate numbers, we will compensate by reducing the number of treatment levels and experiments (e.g. only two temperature levels). In addition to the treatments we will have two additional sets of animals that will be used to identify DA levels of animals when brought into the laboratory (pre-holding) and DA levels after being fed DAcontaining clams (post-fed). Test animals will be held in separate tanks with filtered, sterilized (UV-treated) seawater, as is typical for depuration systems (Lee et al. 2008, 2010), and fasted for 3 days prior to initiating laboratory-induced DA accumulation to ensure that their digestive systems are completely evacuated (Curtis and McGaw 2010).

DA will be analyzed among a group of replicate animals at specified intervals: pre-fed (1 sample), post-fed (1 sample), and at days 5 and 10 of the experiment (1 sample from each of 3 treatment groups on each day). We have chosen day 10 as our cut-off sampling period because literature shows it to be a reasonable time to elicit significant DA reduction (Lund et al., 1997) but short enough to be plausible as a commercial fishery mitigation method. The additional few days will enable us to identify factors that look promising and that may be further accelerated when used in combination with other promising conditions. Because our primary interest is in developing methods to improve food safety, we will focus on quantifying DA load in the hepatopancreas (viscera); for the optimized final depuration methods we will measure DA in the meat as well. At the appropriate sampling time, and following ice bath immersion, crabs and lobsters will be sacrificed and their viscera and body/tail (meat) removed.

We will use CDPH protocols and associated methods (Quilliam et al. 1989, Quilliam 2003) to determine DA levels in samples. Analyses will be done on raw rather than cooked organisms/tissue, a deviation from CDPH protocols. (The use of cooked organisms is based on the assumption that most people boil crab and lobster before eating it, where in fact there is a range of preparation methods.) Analysis of raw tissue will 1) reduce loss of DA to the cooking process (Hatfield et al. 1995; Costa et al. 2003) that could mask differences in depuration efficacy reduce the effort and time required for processing, 2) reduce the effort and time required for processing, 3) provide baseline information that can be used to estimate risks associated with multiple cooking methods (e.g., grilling, steaming, stir-frying), and 4) provide information on potential risk associated with consumption of raw product.

Once the tissues have been processed and DA extractions completed, presence and levels of DA will be determined using similar CDPH HPLC-UV (DAD) and mass spectrometry analytical methods (Dhoot et al. 1993), methods we have been using in our laboratory and that have been validated through comparisons with CDPH DA-analyzed shellfish samples. DA calibration curves will be created by running known dilutions of a DA standard and plotting against peak area determined from the UPLC UV detection at 242 nm. Peak area calculation will be done using peak integration with Waters MassLynx 4.2 software (Waters Corporation 2018). Samples will be analyzed using the same method as the DA standard.

Our depuration experiments will be conducted for each species. This is necessary because depuration of DA, and biotoxins generally, varies among species (see reviews by Shumway 1995; Trainer et al. 2012). Based on our field data of DA-producing HABs in the Santa Barbara Channel, and data from CDPH and others, the California spiny lobster may be impacted for less time than Dungeness crab, although both have experienced prolonged, high DA levels.

Depuration rates may explain, at least in part, differences among species. The levels of DA in the viscera and, where appropriate, the body/tail will be compared using an analysis of variance and post hoc univariate F-tests. Data will be transformed prior to analysis as needed to satisfy assumptions of normality and homoscedasticity (Zar 1999).

Research results will be shared through publication of a summary report, a scientific manuscript, and materials for fishing communities. The outreach materials will be distributed to interested individuals and groups, such as CDFW, CDPH, OEHHA, the Legislature’s Joint Committee on Fisheries and Aquaculture, the state’s Dungeness Crab Task Force, the California Ocean Protection Council, and aquaculture and fishing communities.