NETFRESH: Networking the response of freshwater ecosystems to environmental change - metabarcoding data

Biodiversity in freshwaters is declining at a rate far greater than marine or terrestrial biomes, yet we lack a mechanistic understanding. A key to elucidating the mechanisms driving these losses is the study of ecological interactions (e.g., predation, competition, parasitism, mutualism), which translate environmental change into alterations in the structure and function of biological communities. Our knowledge of ecological networks in freshwaters, however, remains limited due to technological constraints associated with identifying and quantifying ecological interactions in underwater environments.Here, we propose an ambitious project developing our ability to quantify ecological interactions in running freshwater environments with the aim of understanding ecosystem responses to environmental change. We will combine established (gut content analysis, stable isotopes, production estimates) and novel techniques (metabarcoding, video footage combined with machine learning/artificial intelligence, macronutrients) to explore ecological interactions in real time. Using these methods, we will construct the first multilayer ecological networks (networks including multiple interaction types, e.g., competition, predation, facilitation) in running waters, and test the ability of the framework to detect changes in ecosystems using micro- and meso-cosms. For the latter, we focus on low flows as a pivotal component of environmental change in freshwater ecosystems - with existing mechanistic understanding surrounding the individual-level responses of different species (i.e., increased metabolic demand) and clear hypotheses surrounding the potential responses of ecological interactions (i.e., reduced habitat size leads to increased resource competition).The project will support the generation of: (i) novel methods for identifying and quantifying ecological interactions in freshwaters; and (ii) fundamental knowledge on ecological interactions in running waters. Furthermore, findings from this work will also help to inform national monitoring strategies to better detect the response of ecosystems to environmental change and methods could be developed further to potentially provide an early warning system for species loss and ecosystem collapse in freshwater ecosystems. These data were generated via DNA metabarcoding using general invertebrate PCR primers and nanopore sequencing of invertebrates collected from a mesocosm experiment. The mesocosm experiment followed a Before-After Control-Impact (BACI) design (Stewart-Oaten & Bence, 2001), and included a 6-month colonisation period (January to July 2023). Discharge treatments were generated by adjusting intake valves on the individual mesocosms, measuring the depth of the water and using pre-existing stage-discharge curves to calculate reductions in flow. Hess samplers (165cm2, 500um mesh gauge) were used to collect macroinvertebrates at each sampling point. There were three experimental treatments included within each block (100% [control], 50%, and 10% of the baseflow discharge). Samples were collected before the treatments, and after 30 days of exposure to the treatments. Taxa were identified down to family level for abundance measurements.