Article: Ferriss, B. E., Reum, J. C., McDonald, P. S., Farrell, D. M., & Harvey, C. J. (2016). Evaluating trophic and non-trophic effects of shellfish aquaculture in a coastal estuarine foodweb. ICES Journal of Marine Science: Journal du Conseil, 73(2), 429-440. DOI: 10.1093/icesjms/fsv173
Today is the last day in our “All About Aquaculture” theme week; by now, you have read a lot about the benefits, troubles, and unique species associated with “farming” ocean critters. But how does cultivating a species in a designated area affect the food web?
The ecosystem model Ecopath with Ecosim (EwE) is a tool used to look at specific tropic effects in foodwebs, such as their role of bivalves as prey. But what about non-trophic effects such as competition for space? EwE can look at mediating effects, the underlying processes that cause the observed effect between organisms in an ecosystem. For example, how does the concentration of a tasty prey source in one area (such as aquacultured shellfish) alter that ecosystem’s predator structure?
In Washington State, geoduck (pronounced gooey-duck) aquaculture (Figure 1) is an over 10 million dollar industry! It can take 5 to 7 years for a single geoduck to get from its larval size to market size, so a significant effort is taken to reduce predation. In the early stages, geoduck “seed” is protected from predation by placing them in net-covered PVC pipes (Figure 2). This change in the benthic habitat type is likely to change predator-prey interactions and ultimately have effects on this ecosystem by changing which predators will be present.
The EwE model is an open source and free software that allows users to create ecosystem models for specific areas or to adapt pre-existing models to address specific ecological questions. A Puget Sound-specific EwE model was adapted to predict the effects expanding geoduck aquaculture could have on the ecosystem.
Essentially, the Ecopath part of EWE balances gains and losses in biomass (the total weigh present of an organism) by looking at parameters such as diet, mortality, production (photosynthesis), and consumption (predation). Ecosim allows the modelers to predict temporal-based changes to a foodweb by including factors such as harvest, immigration, and emigration.
Mediation effects are complex since they can have positive and negative effects on ecosystem dynamics. For example, increasing eelgrass biomass (a habitat type) could positively affect juvenile salmon biomass by allowing more prey and negatively affect juvenile salmon by increasing refuge space for predators of salmon. For geoducks, the mediation effects are mostly related to the PVC anti-predator structures (Figure 2).
The effects geoduck aquaculture has on the Puget Sound ecosystem were analyzed in two approaches. First, the carrying capacity of geoducks was estimated (just how big a single organism’s population can get before it will negatively affect other organisms). In a marine system, this is often related to phytoplankton production. Secondly, the trophic and non-trophic effects of an increasing geoduck industry were estimated on other functional groups (predators, prey, and other organisms).
What did they find?
Purely looking at carrying capacity (no mediating effects), the current geoduck aquaculture industry could increase by 120% without affecting the Puget Sound ecosystem. The current day geoduck standing stock was estimated to be at only ~0.1% of the maximum carrying capacity.
That is neat, but that estimate is only considering geoducks and not the interconnected effects this increase would have on the entire ecosystem. So next up was to look at the trophic interactions, without any mediating effects. It turns out, that 120% increase also would have little effects, since there was plenty of phytoplankton to go around. Additionally, there was only a predicted 2% increase in the two biggest geoduck predators, sea stars and the Dungeness crab, and a <1% increase in the other major critters in the Puget Sound biomass.
Here is where this picture changes: when mediating effects (effects not directly tied to trophic interactions) are taken into consideration, huge changes are simulated. That 120% geoduck increase was associated with increases in other Puget Sound organisms. For example, Great Blue Herons, a top predator, were predicted to decrease by 23%, while the Pacific cod increased by 7% (Figure 3).
The PVC anti-predator guards for geoducks were modeled to have a large influence on the type of predator that thrived in Puget Sound, the effects of which propagated up the food web. The PVC caused an increase in demersal fish (bottom-dwellers), which altered the prey type available for the even bigger predators. In other words, this change in fish-type caused bird populations to suffer due to a lack of their preferential food source, observed as a decrease in small crustaceans.
Including mediation effects, such as changes in predator refuges, can greatly improve ecosystem modeling. Aquaculture is almost certainly going to increase in Puget Sound as demand for geoducks (and other organisms) increase. Better understanding which marine organisms and functional groups (demersal vs flatfish) could be affected is important so we can better monitor these vulnerable organisms to inform better management decisions, such as restoring a certain habitat type that may encourage declining organisms to recover.