WRS Doctoral Defense

Microplastic Pollution in Agricultural Watersheds

Microplastics (< 5 mm) are an ongoing global environmental pollution problem first documented as early as the 1960s. Microplastics are traditionally thought of as inextricably linked to marine and urban environments. Recently agriculture has become recognized as a potential source of microplastics with the use of plastic mulch, polymer coated seed, pesticides, and fertilizers. Additionally, the majority of microplastic research is focused on quantifying snapshot concentrations in different environmental settings. Little information exists on quantifying loads and fluxes of microplastic pieces entering and leaving compartments over a range of temporal and spatial scales. This dissertation investigates microplastics in a freshwater agricultural setting and provides timely evidence to help answer these key knowledge gaps. The first study analyzed microplastic concentration and microplastic mass in the surface water of 10 small agriculturally dominated subwatersheds of the Cannon River in southeastern Minnesota, over one hydrologic year. Statistically significant differences were observed in the microplastic concentration across sampling locations. The mass of microplastics was significantly different across sampling locations and dates. Polynomial regression and general additive models revealed significant positive correlations between the mean mass of microplastics and mean watershed slope, drainage class percent, hydric soil percent, and water table depth. Significant negative correlations were found between mean microplastic mass, tile drain percent, and hydric soil percent. The mean microplastic concentration was positively correlated related to watershed area. Subsequent positive correlations were revealed through a post-storm analysis between microplastic concentration and microplastic length, surface water residence time, water table depth, and available water storage. The mean microplastic concentration was negatively correlated with developed land percent. The post-storm microplastic mass was positively correlated to soil erosion risk percent. Surprisingly, there were no relationships observed between the mean microplastic concentration and mass, or the post-storm microplastic concentration and mass when compared against the mean crop cover percent of each subwatershed. This study highlights diverse source pathways over time and the importance of collecting time-series water samples with spatial distribution to achieve the data resolution needed to detect trends related to watershed processes. In addition, measuring or converting values to microplastic mass and concentration may reveal hidden environmental relationships and facilitate toxicological risk assessment. The second study monitored microplastic loads in wet and dry atmospheric deposition, as well as stream outflow, for approximately 3 years in a predominantly agricultural watershed, Belle Creek, in southeastern Minnesota. Significant differences in microplastic load were observed, indicating substantially more microplastic is leaving the watershed via stream outflow compared to that entering the watershed from the atmosphere, by an order of magnitude. Hydrologic analysis and baseflow separation of the stream outflow load of microplastics revealed significantly greater loads of microplastics under baseflow conditions compared to stormflow conditions. Cumulative loads in wet deposition and dry deposition were relatively equal over the entire sampling period, demonstrating that microplastic contributions were similar from precipitation and dust deposition, though there was seasonal variation. Evaluation of wet deposition across seasons revealed significantly greater concentrations of microplastics in snow versus rain. The large difference between microplastics leaving the watershed in streamflow and microplastics entering the watershed from the atmosphere indicates substantial in situ sources of microplastics, perhaps a surprising result for a completely rural watershed.  The third study analyzed microplastics in the snow cover, soil, and groundwater of the Belle Creek watershed. Microplastic concentrations were greatest in soil compared to snow and groundwater, suggesting the soil of this watershed may be accumulating microplastics. The microplastic storage potential of the snow and soil of this system was estimated at 4 cm and 5 cm depth, respectively, and scaled to the area of the watershed. The magnitude of estimated storage of microplastics in snow and soil raises concern for the influence that heavy precipitation combined with snowmelt may have on sediment enrichment and transport of microplastic pollution with infiltration, percolation, and runoff into groundwater and surface water, particularly in the early spring seasons. The mean size and mass of microplastic fragments in soil were smaller and lighter in mass compared to snow, and microplastic particles were smaller in soils compared to both snow and groundwater, indicating the longer residence time of the soil system in this watershed allows more time for microplastics to be subject to degradation processes and thus break down into smaller, lighter pieces. Chemical speciation of sampled plastic from these studies is a necessary next step to determine the primary source(s) of plastic pollution.