Microplastic Contamination of a Karst Spring During Flooding

Karst aquifers are important drinking water sources, but high connectivity between the surface and subsurface makes them susceptible to anthropogenic contamination. Agricultural runoff and wastewater from leaking infrastructure can contaminate karst aquifers with biological (fecal bacteria) and chemical (nutrients, pharmaceuticals, toxic metals) pollutants, but they may also be significant sources of microplastics. Previous work has noted the presence of microplastics in karst springs, but little is known about the sourcing and transport of these contaminants in karst aquifers, particularly under a range of flow conditions. Thus, this study explored whether microplastics varied under flood conditions in a karst spring and if they occurred with other chemical tracers indicative of agricultural (N, P) and wastewater (B, optical brighteners) inputs. Water samples were collected every 2-4 h from a karst spring-fed stream during a flood event in April 2019 using an autosampler located ~ 5 m from the spring orifice. We also used the autosampler to collect blanks (i.e., deionized water) to determine the extent to which our field and lab methods contaminated the samples. Our blanks showed that, on average, our methodology contributed microplastics at 11 ± 3 counts/L to the samples. During the flood, we observed microplastics in every sample ( n = 29; 37 ± 24 counts/L), with fibers comprising 98% of the microplastics transported during the event. Microplastic concentrations immediately following rainfall were 61 counts/L, but rapidly diluted to 9 counts/L before any large changes in spring discharge. They increased to 107 counts/L by the end of the flood event. Microplastics were not correlated with suspended solids, agricultural indicators, or wastewater tracers (R 2 < 0.16; p > 0.05). These somewhat unexpected results may be due to microplastics being sourced from a road that was ~ 1 m from our autosampler intake. Indeed, microplastics may have been initially flushed from the road to the spring-fed stream before the spring itself responded. Grab samples will be collected directly from the spring orifice to determine if the observed microplastics were being sourced from the road or the karst aquifer. Nevertheless, our study indicates that microplastics can be transported in highly variable quantities during flood events.