You are here:
Characterizing temporal variation in diatom rbcL sequences in streams to inform nutrient indicator development using DNA metabarcoding
Citation:
Smucker, N., E. Pilgrim, C. Nietch, J. Darling, H. Wu, M. Molina, AND B. Johnson. Characterizing temporal variation in diatom rbcL sequences in streams to inform nutrient indicator development using DNA metabarcoding. IDS 2021 – Online International Diatom Symposium, Virtual, JAPAN, August 23 - 25, 2021.
Impact/Purpose:
Nutrient pollution is a widespread problem for freshwater ecosystems throughout the United States and around the world. Its negative effects on biota and ecosystems increase as human populations grow and as watersheds become increasingly altered to produce food and to meet housing and socioeconomic demands. As a result, interest in developing diatom indicators of nutrients continues to grow because their relationships with nutrient concentrations can help inform the assessment, conservation, and management of stream ecosystems and watersheds. However, temporal variability could affect their applications in monitoring programs. Here, we used metabarcoding of diatoms (rbcL) to characterize how temporal variability potentially affects their use as total phosphorus (TP) and total nitrogen (TN) indicators.
Description:
Diatom relationships with nutrient concentrations can be used to develop indicators that help inform the assessment, conservation, and management of stream ecosystems and their watersheds. With molecular-based approaches emerging as an effective way to characterize diatom-environment relationships, we were interested in examining how temporal variability could affect interpretations of monitoring data and assessment outcomes based on DNA metabarcoding results. We used DNA metabarcoding of diatoms (rbcL) and their relationships with total phosphorus (TP) and total nitrogen (TN) concentrations in streams to examine their potential use as nutrient indicators. From July to October 2016, we sampled 25 sites in a watershed (1293 square kilometers) 12-14 times (typically weekly) to examine temporal variation in nutrients and diatoms. Nonmetric multidimensional scaling showed that diatom assemblages were strongly associated with a TP gradient related to percent agriculture in upstream watersheds. Indicator species analysis identified taxa that decreased or increased with TP or TN and we used relative abundances of gene copy numbers for these low P, high P, low N, and high N taxa as metrics. Temporal variability in low P and low N diatom relative abundances was much greater in streams with lower mean nutrient concentrations, despite nutrients becoming more variable as concentrations and watershed percent agriculture increased. This might suggest that (a) in low nutrient streams, relative abundances of low nutrient taxa are highly susceptible to other taxa responding to small, transient increases in nutrient availability over time, and (b) in high nutrient streams, the established dominance by high nutrient taxa likely leads to consistently low relative abundances of low nutrient taxa despite the greater TP and TN variability over time. Regressions showed that diatom assemblages were more strongly associated (1) with TP than with TN and (2) with mean TP and TN concentrations than with same week concentrations at each site. Despite nutrient and diatom variability over time, relationships remained strong for most weeks except during, and immediately after, two high flow events that temporarily weakened diatom-nutrient relationships. Correlations of TP concentrations with low P diatoms and high P diatoms became evident again within one and three weeks, respectively. In general, these results indicate that diatom indicators developed using DNA metabarcoding in this watershed integrated seasonal nutrient concentrations and performed consistently well throughout a summer monitoring period despite temporal variability and a brief breakdown in correlations following disturbance. Our results further highlight the potential for using diatom DNA metabarcoding to develop indicators that can be used in monitoring programs.
