Essays on Science and SocietyEcology and Environment

The ecology of an olfactory trap

See allHide authors and affiliations

Science  23 Nov 2018:
Vol. 362, Issue 6417, pp. 904
DOI: 10.1126/science.aav6873

Clean up after yourself. We are taught this adage from a young age, and yet these words often seem to be discarded before adulthood, especially in regard to the global commons (1). Our largest global commons is the ocean. It is also the sink for much of our waste.

Plastics are a ubiquitous form of marine litter; these synthetic materials have been found in oceans worldwide as well as in sea ice, seafloor sediment, and distressingly, in seafood (2, 3). To date, nearly 1000 marine species, ranging from invertebrates to whales, have been found to ingest plastic debris.

Since the late 1960s, researchers have focused on documenting not only which species are eating plastic but in what quantities, in what regions, and if/how ingested plastic affects the consumers and associated food webs. Despite burgeoning scientific interest on the interactions between plastic waste and marine wildlife, far less research has focused on the behavioral mechanisms underlying the maladaptive decision to ingest plastic. Most of what has been published is largely speculative and based solely on the visual cues of plastic debris.

Think of the common explanation: “Sea turtles eat plastic bags because they look like jellyfish.” My doctoral research found that in addition to having the appearance of food, plastic also releases a biochemical signature that may trick organisms into confusing our trash for their treasure (see the figure).

The story begins with the odoriferous algal-derived molecule dimethyl sulfide (DMS). When phytoplankton are consumed by zooplankton, DMS is released en masse—a result of dimethylsulfoniopropionate (DMSP) catabolism—from the depredated phytoplankton (4). Marine predators—including certain species of fish, seabirds, and sea turtles—track relative concentrations of DMS over the open ocean in order to find productive regions to forage [reviewed in (5)]. I first used a 50-year seabird diet database, coupled with decades of data on seabirds' olfactory preferences, to discover that DMS-tracking is an ecologically predictable trait.

I determined that procellariiform seabirds—a group that includes burrow-nesting petrels and shearwaters—were likely to follow an odor trail of DMS, not to find the phytoplankton producing it but rather to locate nutrient-rich zooplankton (such as krill) (6). Zooplankton consumption of phytoplankton is often responsible for local elevations of DMS (4). In addition, these recruited seabirds provide limiting nutrients such as iron via their waste to the phytoplankton (5, 6). This work provided evidence of a chemically mediated tritrophic mutualism in the marine environment, but the connection to marine debris was yet to be explored.

Procellariiform seabirds possess an acute olfactory sense and are also severely affected by the ingestion of plastic (7). The notion that these birds consume plastic simply because of its appearance is incongruous with decades of research demonstrating that procellariiformes are highly reliant on odor cues for foraging (810). In addition, some species never ingest plastic, whereas others seem unable to avoid it. This unexplained interspecific heterogeneity is puzzling.

To confront this problem scientifically, I created a database by amassing every published record of plastic ingestion in all procellariiform species. This database ultimately included 73 studies, with plastic ingestion data from 20,922 individuals of 65 species. For approximately half of these species, their behavioral attraction, or lack thereof, to DMS was known. When I analyzed how often each species ingested plastic debris in relation to their olfactory attraction to DMS, the result was startling: Those species that use DMS to forage were almost six times more likely to consume plastic than those that are not attracted to DMS (11). Finding a result this pronounced on a dataset this robust is unusual in ecology, so I knew I needed to dive deeper.

The next question to consider was, what does ocean-seasoned plastic smell like to a foraging seabird? To answer this, I tethered three types of preproduction plastic—polypropylene and high- and low-density polyethylene, which together make up two-thirds of consumer plastics—to buoys at two different locations off the coast of California during the summers of 2013 and 2014.

Evidence for an olfactory trap. (A) Simplified diagram of the tritrophic interaction between phytoplankton and top predators (such as Procellariiform seabirds), mediated by DMS. [from (5).] (B) Plastic ingestion and DMS responsiveness among procellariform species. Horizontal bars represent plastic ingestion prevalence by species ordered from high to low in each group. The red bars depict DMS-responsive species (n = 13 ), and the blue bars depict non–DMS-responsive species (n = 12 ) [Modified from (11)]. (C to G) School-wide behavioral responses of northern anchovy (E. mordax) to treatments of (C) control (dyed seawater), (D) food, (E) clean plastic odor, (F) biofouled plastic odor, and (G) food odor, taken from videos 30 and 60 s after treatment introduction. [Modified from (12)]


After the plastic was left at sea for a month, I brought these samples to the Robert Mondavi Institute for Wine and Food and Sciences at the University of California, Davis. I used a gas-chromatography technique that is typically used on wine and food products and found that every sample of ocean-weathered (i.e. biofouled) plastic emitted DMS at concentrations six orders of magnitude above the detection threshold for procellariiforms and three orders of magnitude above background DMS in the environment (11). Furthermore, there was no DMS signature on clean plastics (11).

The distinct DMS signature associated with marine plastic debris originates from the algal biofilm that rapidly coats plastic at sea. The implication is that marine plastics may falsely amplify an olfactory signal that certain species associate with foraging opportunities.

This compelling evidence required further support, in the form of a behavioral assay that would test how the chemical signature of marine plastic affected the foraging behavior of a marine organism. In the summer of 2015, I tested this question on Northern Anchovy (Engraulis mordax), an ecologically critical forage fish species and known plastic consumer (2). Anchovy responded to odors of food and plastic debris similarly, increasing clustering behavior and reducing school-wide rheotaxis in a manner that is characteristic of olfactory search (12). These behaviors are consistent with foraging in this species.

To understand a problem as pervasive and perplexing as plastic debris ingestion requires careful consideration of how the organisms we study experience their world. Taken together, these results suggest that plastic debris may be more confusing and appetizing to marine organisms than previously thought possible. Moreover, the results of my work underscore the urgency of mitigating plastic pollution at a global scale.



Matthew Savoca

Matthew Savoca received his undergraduate degree from Cornell University and his Ph.D. in Ecology from the University of California, Davis. He is currently a postdoctoral researcher at the Hopkins Marine Station of Stanford University studying the foraging behavior of baleen whales and their possible interactions with plastic debris.


Navigate This Article