We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Hold me Back! Does vegetation density shape the polymer-specific fate of MP in stream mesocosms
Summary
Underwater plants in streams act like nature's filter, trapping different types of microplastic pollution (like tire dust and plastic bottle fragments) rather than letting it flow downstream unchecked. This matters because it means healthy, plant-filled waterways may help keep more plastic particles out of the water we eventually drink or the fish we eat—though how much plastic gets trapped depends on both the plant density and the type of plastic involved.
We investigated the fate of three polymer types—polystyrene (PS), polyethylene terephthalate (PET), and tire wear particles (TWP)—within replicated 45 m stream mesocosms containing three vegetation densities of Elodea nuttallii (0%, 25%, 100%, n = 3). Following a controlled pulse release, MPs concentrations were quantified in the water column, sediment, and macrophyte biomass. We used log-scaled particle concentrations to monitor the retention of MPs in sediment and macrophyte biomass (number of MP g-1) through different macrophyte densities and the longitudinal transport over the length of the artificial channels. Percentage sedimentation for each MP and treatment was calculated based on the total number of particles found during density separation. The time-weighted average concentration (TWAC, L-1) was calculated by trapezoidal integration to evaluate the MP concentration over all sampling time points at a given sampling site. To determine the role of submerged vegetation in mediating MP fate, bootstrapped linear mixed-effects models were used to assess possible interactions between vegetation densities, polymer types, and stream distances and to account for the dependence of measurements within each stream. Additionally, as complementary, empirical indicators of MP transport, water- and sediment-distance-decay relationships were quantified using log-linear regression models.