Mink fur farming was once a widespread agricultural activity in southwestern Nova Scotia. Some freshwaters near mink fur farm operations now show severe water quality issues. Notably, the watershed of Nowlans Lake (southwestern Nova Scotia, Canada) once contained six mink farms as well as a fish meal feed processing plant. It is now one of the most productive lakes in Atlantic Canada, with exceedingly high measured Total Phosphorus concentrations.

Here, we provide data from a paleolimnological investigation on the long-term environmental changes that occurred in Nowlans Lake, and how these changes have impacted cladoceran communities. First, we identify abundances of cladoceran taxa throughout the sediment core, corresponding to assemblages from the ~1900s to present day (determined by 210Pb dating analyses – see Table 1 of cited manuscript), and provide raw counts. We detected shifts in the dominant pelagic cladoceran taxa beginning in the early 1900s, with decreases in small-bodied bosminids, while taxa such as Chydorus brevilabris and Daphnia pulex spp. increased. We identified assemblage “zones” using a constrained hierarchal clustering analysis, denoting three distinct assemblage groups (Script 03). We then provide empirical measures of the body sizes of bosminids in each sediment interval, as body sizes of cladocerans are good proxies of shifts in predation. We identified bottom-up ecological factors as the likely drivers of these assemblage shifts, as body sizes were found to be consistent through time with a Mann Kendall Monotonic Trend Analysis (Script 02). Finally, we reconstructed and provide data on trends in sedimentary chl-a concentrations using visible reflectance spectroscopy (VRS). After conducting a PCA on cladoceran assemblages and extracting PC1 scores, we observed a strong relationship by VRS chl-a and PC1 scores, suggesting similar timing between increases in lake productivity and major cladoceran changes (Script 01).

All data presented in this folder originate from sediment cores collected from the deepest basin of Nowlans Lake using a gravity corer in mid-May 2018. Cores were sectioned at 0.5-cm intervals using a vertical extruder. Chronology of the sediment core was established through standard ²¹⁰Pb dating techniques (See cited Manuscript Methodology, and Table 1 for reported dates). Trends in sedimentary chl-a concentrations were inferred by visible reflectance spectroscopy (VRS) using a FOSS NIRSystems model 6500 rapid content analyzer. The concentration of VRS chl-a and its major degradation products was calculated from the resulting reflectance spectra between 650–700 nm using a linear regression equation.

Twenty-eight sediment samples were processed for Cladocera subfossil remains. Approximately 1–2g of wet sediment per interval was deflocculated in 5% KOH solution and heated. After digestion, the sample was sieved on a 38-µm mesh and rinsed with deionized water. Materials were put into labelled glass scintillation vials with 1% safranin staining solution and 95% ethanol. Next, sample were pipetted onto glass coverslips, and contents were evaporated. Glass coverslips were permanently mounted onto slides using Entellan®. Sedimentary remains were identified to genus or species level using bright-field microscopy at 200× or 400× magnification. Coverslips were scanned in entirety for all identifiable remains. Identifications were performed using subfossil taxonomic guides for eastern North America (See cited Manuscript Methodology). Non-fragmented remains of bosminid carapaces were measured using Zeiss Zen imaging software.