Estimating the size distribution of plastics ingested by animals

Abstract

The ingestion of plastics appears to be widespread throughout the animal kingdom with risks to individuals, ecosystems and human health. Despite growing information on the location, abundance and size distribution of plastics in the environment, it cannot be assumed that any given animal will ingest all sizes of plastic encountered. Here, we use published data to develop an allometric relationship between plastic consumption and animal size to estimate the size distribution of plastics feasibly ingested by animals. Based on more than 2000 gut content analyses from animals ranging over three orders of magnitude in size (lengths 9 mm to 10 m), body length alone accounts for 42% of the variance in the length of plastic an animal may ingest and indicates a size ratio of roughly 20:1 between animal body length and the largest plastic the animal may ingest. We expect this work to improve global assessments of plastic pollution risk by introducing a quantifiable link between animals and the plastics they can ingest.

Fig. 1. Allometric relationship between animal size and ingestible plastic size.

a Allometric size relationship (log10-log10 linear regression; R² = 0.42, F_1,63 = 46.06, p = 4.7e⁻⁰⁹), including 99% (light grey) and 95% (dark grey) CIs, between animal body length (mm) and the longest piece of ingested plastic (mm) found during gut surveys (longest axis of largest piece of plastic found). Animal images are for illustration only and are not to scale. Each data point (n = 65) corresponds to the largest piece of plastic found within an animal taxon. b Distribution of field studies that provided data for the allometric relationship. Size of data points in a and b correspond to the number of individual animal specimens surveyed. Similar taxa from separate studies are plotted separately.

Fig. 2. Animal depth ranges in the water column exceed plastic pollution models.

Water column depth range of species included in meta-analysis compared to depth range of established global plastic models. Cózar et al., Eriksen et al. and van Sebille et al. are well cited distribution models of buoyant plastics floating at the surface of the Earth’s oceans (674, 946 and 255 citations, respectively (Google Scholar 2019)). We assigned a crude water depth value of 25 m to each of these models, to account for the oceanic surface mixed layer, according to Kukulka et al..

Fig. 3. Detection limits scale with animal size.

Studies of larger animals tended not to specify the use of magnifying equipment. Weak relationship (log10-log10 linear regression; R² = 0.10, F_1,61 = 7.58, p = 0.008), including 99% (light grey) and 95% (dark grey) CIs, between animals and the smallest piece of ingested plastic found during gut surveys (plastics measured along their longest axes; n = 63).

Fig. 4. Global risk of plastic entering zooplankton communities.

a Using the animal-plastic size relationship to estimate the density of ingestible plastics (0.33–1.00 mm in length) divided by zooplankton density. b Current ‘state of the art’: as for a, with ingestible plastic densities substituted with total plastic densities. Legend coarsely estimates the level of plastic ingestion risk for zooplankton.