Search

Accurately identifying the species present in an ecosystem is vital to lake managers and successful bioassessment programs. This is particularly important when monitoring cyanobacteria, as numerous taxa produce toxins and can have major negative impacts on aquatic ecosystems. Increasingly, DNA-based techniques such as metabarcoding are being used for measuring aquatic biodiversity, as they could accelerate processing time, decrease costs and reduce some of the biases associated with traditional light microscopy. Despite the continuing use of traditional microscopy and the growing use of DNA metabarcoding to identify cyanobacteria assemblages, methodological comparisons between the two approaches have rarely been reported from a wide suite of lake types. Here, we compare planktonic cyanobacteria assemblages generated by inverted light microscopy and DNA metabarcoding from a 379-lake dataset spanning a longitudinal and trophic gradient. We found moderate levels of congruence between methods at the broadest taxonomic levels (i.e., Order, RV=0.40, p0.0001). This comparison revealed distinct cyanobacteria communities from lakes of different trophic states, with Microcystis, Aphanizomenon and Dolichospermum dominating with both methods in eutrophic and hypereutrophic sites. This finding supports the use of either method when monitoring eutrophication in lake surface waters. The biggest difference between the two methods was the detection of picocyanobacteria, which are typically underestimated by light microscopy. This reveals that the communities generated by each method currently are complementary as opposed to identical and promotes a combined-method strategy when monitoring a range of trophic systems. For example, microscopy can provide measures of cyanobacteria biomass, which are critical data in managing lakes. Going forward, we believe that molecular genetic methods will be increasingly adopted as reference databases are routinely updated with more representative sequences and will improve as cyanobacteria taxonomy is resolved with the increase in available genetic information.

Eukaryotic pathogens including fungi and enteroparasites infect humans, animals and plants. As integrators of landscape catchment, lakes can reflect and record biological and geochemical events or anthropogenic changes and provide useful knowledge to formulate public health, food security and water policies to manage and prevent diseases. In this context, potentially pathogenic fungi and parasites were sampled using 18S rRNA gene amplicon sequencing in 382 lakes displaying a broad range of sizes and human impact on the watershed in 10 ecozones across Canada. Based on pathogen classifications from the ePATHogen database published by the Public Health Agency of Canada, we identified 23 health-relevant genera for human and animal hosts, including Cryptococcus and Cryptosporidium. Our study investigated the potential of remote sensing and geomatics to predict microbial contamination in a tele-epidemiological approach. We used boosted regression tree modeling to evaluate the probability of occurrence of the most common genera found in our dataset based on 10 satellite-derivable, geomatics and field survey variables which could be potential sources or transport mechanisms through the watershed or survival factors in the water. We found that southern ecozones that possess the highest agricultural and pasture activities tend to contain lakes with the largest number of potential pathogens including several fungi associated with plant diseases. Bio-optical factors, such as colored dissolved organic matter, were highly related to the occurrence of the genera, potentially by protecting against damage from ultraviolet light. Our results demonstrate the capability of tele-epidemiology to provide useful information to develop government policies for recreational and drinking water regulations as well as for food security.