Search

© 2023 The Authors. Freshwater Biology published by John Wiley & Sons Ltd. This is is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes. 1. Cyanobacterial harmful algal blooms (cyanoHABs) are assumed to be increasing in abundance and toxicity, but comprehensive analysis of change through time is limited, in part, because some key taxa (e.g., Microcystis) leave ambiguous evi- dence of historical abundance and toxicity. Sedimentary DNA (sedDNA) can allow the reconstruction of the cyanobacteria community as well as the frequency of genes specific to cyanotoxin production, enabling us to determine which taxa are present and their potential for toxin-production. 2. Using a combination of droplet digital polymerase chain reaction (ddPCR) and high-throughput sequencing (HTS), we quantified the abundance of cyanobacte- rial genes of known function and changes in cyanobacteria taxa from sedDNA over the last century in nine lakes along a gradient of lake size, depth and trophic state in Minnesota, U.S.A. Using ddPCR, we quantified genes associated with mi- crocystin toxin-producing potential (mcyE), total cyanobacteria (CYA, 16S rRNA) and the genus Microcystis (MICR, 16S rRNA). Using HTS on a subset of lakes, we investigated how the abundance of this toxin-producing gene covaried with the cyanobacteria community composition. We also compared ddPCR and HTS data to fossil pigments, a well-established palaeolimnological method used to track changes in primary producers over time. 3. Our results showed a significant correlation between MICR and the quantity of mcyE gene and cyanobacterial taxa with known toxin- production potential. The abundance of both genes likewise increased concomitantly through time. 4. Community analyses of HTS data showed significant change in cyanobacte- rial communities commencing c. 1950 when major land-use change in this re- gion led to increased lake productivity, and c. 1990 when Dolichospermum and Microcystis genera increased in abundance, and the subtropical exotic cyanobac- teria Raphidiopsis raciborskii and Sphaerospermopsis aphanizomenoides became abundant. Cyanobacteria pigment data reflected these changes only in deeper lakes, suggesting issues related to benthic production or biomarker preservation in shallower systems. This research was funded by a grant from the Gold Bay Foundation (https://proje cts.propu blica.org/nonpr ofits/organ izati ons/47400 2921) to A.J.H. and by the Minnesota Environment and Natural Resources Trust Fund as recommended by the Legislative-Citizen Commission on Minnesota Resources to A.J.H. P.R.L. was funded by a Canada Research Chair and Canada Foundation for Innovati Faculty yes

Summary In agricultural catchments, streams are intimately connected with croplands and are exposed to agrochemicals such as fertilisers and herbicides. In North American streams, the effects of nutrient enrichment on riparian and aquatic plant community structure are not well established and the effects of herbicides are largely unknown. We identified the riparian and aquatic plant community at 12 paired stream/river sites (24 in total, Strahler stream order 3–6) located across a large North American agricultural catchment. Field sites ranged in surrounding agricultural land use (2.7–100% annual crops) and in-stream June concentrations of reactive phosphate (6–65 μg L−1 as phosphorus) and nitrate (3–3981 μg L−1 as nitrogen). Atrazine, a herbicide commonly used on corn crops in North America, was widespread throughout the catchment with time-weighted-average concentrations ranging from 4 to 412 ng L−1 over a 56-day period. Atrazine concentrations were >100 ng L−1 at over half the field sites. In total, 285 riparian and aquatic plant taxa were identified. Effects of agrochemicals on descriptors of plant communities were assessed by comparing paired sites surrounded by low or high levels of agriculture and examining changes across the catchment. Agrochemicals had no effect on overall species richness. However, the percentage of non-native species increased across the catchment along a gradient of increasing nitrate. A decline in submerged macrophytes was associated with enrichment of both nitrate and reactive phosphate, whereas no direct effects of atrazine were observed. A floristic quality assessment was used to determine further effects of agrochemicals on riparian and aquatic plant communities. A ranking system, based on native status and tolerance to disturbance, was used to assign coefficient of conservation values to each species. Floristic quality was subsequently calculated as a function of species richness and the average coefficient of conservation value at each site. Overall floristic quality decreased as nitrate increased. Species positively associated with nitrate were typically fast-growing, weedy, riparian species, whereas species negatively associated with nitrate were characteristic of less disturbed areas and represented a range of growth forms. We conclude that nutrient enrichment, especially of nitrate, has multiple negative effects on riparian and aquatic plant communities that override effects of the herbicide atrazine. Our results highlight the need to reduce inputs of agrochemicals to streams and to regulate nitrogen in addition to phosphorus.