Influence of Land Use, Nutrients, and Geography on Microbial Communities and Fecal Indicator Abundance at Lake Michigan Beaches

Abstract

Microbial communities within beach sand play a key role in nutrient cycling and are important to the nearshore ecosystem function. Escherichia coli and enterococci, two common indicators of fecal pollution, have been shown to persist in the beach sand, but little is known about how microbial community assemblages are related to these fecal indicator bacteria (FIB) reservoirs. We examined eight beaches across a geographic gradient and range of land use types and characterized the indigenous community structure in the water and the backshore, berm, and submerged sands. FIB were found at similar levels in sand at beaches adjacent to urban, forested, and agricultural land and in both the berm and backshore. However, there were striking differences in the berm and backshore microbial communities, even within the same beach, reflecting the very different environmental conditions in these beach zones in which FIB can survive. In contrast, the microbial communities in a particular beach zone were similar among beaches, including at beaches on opposite shores of Lake Michigan. The differences in the microbial communities that did exist within a beach zone correlated to nutrient levels, which varied among geographic locations. Total organic carbon and total phosphorus were higher in Wisconsin beach sand than in beach sand from Michigan. Within predominate genera, fine-scale sequence differences could be found that distinguished the populations from the two states, suggesting a biogeographic effect. This work demonstrates that microbial communities are reflective of environmental conditions at freshwater beaches and are able to provide useful information regarding long-term anthropogenic stress.

FIG 1

Map of sampling sites and surrounding land use. Wisconsin sites: Point Beach State Park (PB), Kohler-Andrae State Park (KA), Atwater Park Beach (ATW), and Bradford Beach (BB). Michigan sites: Pere Marquette Park Beach (PM), PJ Hoffmaster State Park (PJ), North Beach Park (NB), and Grand Haven City Beach (GH). (Map created using ArcMap 10.2.)

FIG 2

E. coli (shown in light gray) and enterococcus (shown in dark gray) densities in water (A), berm sand (B), and backshore sand (C) samples collected throughout the summer of 2013. Both Wisconsin and Michigan beaches are ordered from north to south.

FIG 3

Relative abundances of dominant taxa in backshore sand, berm sand, submerged sand, and water samples. Bars represent the averaged relative abundances of taxa for three samples from each beach zone.

FIG 4

Nonmetric multidimensional scaling (NMDS) plot based on the Bray-Curtis distances of microbial communities from backshore sand, berm sand, submerged sand, and water. The relative abundances of all taxa were used for comparisons. The dashed colored lines surrounding each sample type represent covariance ellipsoids.

FIG 5

Total phosphorus, total organic carbon, and total nitrogen concentrations in backshore, berm, and submerged sands. The average concentrations from three samples taken at the same location at different time points are shown, and error bars represent the standard errors of measurements among all samples for a particular site.

FIG 6

Oligotypes generated from the genus Terrimonas in all berm samples. (A) NMDS based on Bray-Curtis distance for Terrimonas oligotypes. (B) Oligotype relative abundance. Each color represents a unique oligotype. (C) Oligotype dendrogram, produced using a Bray-Curtis distance and Ward's method of linkage.

FIG 7

Oligotypes generated from the genus Ferruginibacter in all berm samples. (A) NMDS based on Bray-Curtis distance for Ferruginibacter oligotypes. (B) Oligotype relative abundances. Each color represents a unique oligotype. (C) Oligotype dendrogram, produced using Bray-Curtis distance and Ward's method of linkage.