Data from: Multiple stressors in river networks: Local and downstream effects on freshwater macroinvertebrates
Data files
Oct 09, 2025 version files 100.04 KB
-
DISCONNECTED_alldata.xlsx
94.70 KB
-
README.md
5.35 KB
Abstract
River networks are complex ecosystems characterized by a continuous exchange of material and energy through longitudinal gradients. These ecosystems are threatened by various human-induced stressors, which frequently co-occur and may interact in complex ways, potentially triggering cascading effects in the river network. Aiming at assessing the single and combined effects of flow intermittency and light pollution on macroinvertebrate communities, we performed a multiple stressor experiment in 18 flow-through mesocosms. Each mesocosm was designed to mimic a simplified river network, with two upstream tributaries merging downstream, to assess both local and cascading effects. The experiment was performed in Summer 2021 for seven weeks, applying the stressors either separately or co-occurring in the upstream sections, following a randomized block design. Flow intermittency was simulated as the ponded phase of the drying process, whereas light pollution was applied with LED strips (set at 10 lux). Drifting macroinvertebrates were sampled weekly during the treatment phase, and benthic macroinvertebrates were sampled at the end of the treatment phase. Both stressors individually applied reduced benthos richness and abundance, whereas drift decreased with flow intermittency and increased with light pollution. When co-occurring upstream, stressors showed dominant effects of flow intermittency on the benthos and interactive effects on the drift. The effects of the single stressors and their interactions cascaded along the river network, with stronger downstream effects when stressors co-occurred upstream. These findings showed that the spatial distribution of multiple stressors along the river network can affect their resultant downstream effects, highlighting the importance of framing multiple stressors research in a spatial context. Considering the pressing needs of the growing human population, our results represent a step forward in anticipating cumulative stressors' effects, informing efficient conservation strategies for protecting freshwater ecosystems.
https://doi.org/10.5061/dryad.rfj6q57k5
Description of the data and file structure
The experiment was performed to assess the local and downstream effects of multiple stressors (flow intermittency and light pollution) on macroinvertebrate communities in a river network. The study took place in a system of 18 flow-through mesocosms, with each mesocosm unit designed to mimic a simplified river network. The experiment ran for seven weeks in summer 2021, consisting of 26 days of colonization followed by 23 days of treatment. The Excel spreadsheet includes four sheets (drift, benthos, background_data, and README) that are described below.
Files and variables
File: DISCONNECTED_alldata.xlsx
Description: Data from the Disconnected project
Variables
- drift (sheet 1): data for drifting macroinvertebrates
- ID (column A): unique identifier of each drift sample collected during the experiment
- HT (column B): header tank identifier (HT1-HT6)
- flume (column C): mesocosm unit (i.e. flume) number identifier (F1-F18)
- side (column D): identifier of the side (left or right for the two upstream tributaries)
- treatment_up (column E): identifier of the treatment applied for each drift sample. This column has four levels, being: CT (control treatment in upstream sections), L (light treatment in upstream sections), F (flow treatment in upstream sections), L+F (light and flow treatment in upstream sections)
- treatment_down (column F): identifier of the treatment applied at flume level. This column is used in the script when considering the total downstream export. This column has three levels, being CT (control flume), CS (flume with co-occurring stressors), and SS (flume with separated stressors)
- date (column G): sampling day (in date format) for each drift sample
- time (column H): sampling day in factor format (from T1 to T4)
- columns I-AT: drifting macroinvertebrate taxa with abundances for each sample
- benthos (sheet 2): data for benthic macroinvertebrates.
- ID (column A): unique identifier of each benthos sample collected during the experiment
- HT (column B): header tank identifier (HT1-HT6)
- flume (column C): mesocosm unit (i.e. flume) number identifier (F1-F18)
- section (column D): section of each flume (being up for upstream tributaries and down for the downstream part)
- side (column E): identifier of the side (left or right for the two upstream tributaries). Samples collected in downstream sections are marked with "__"
- treatment (column F): identifier of the treatment applied for each sample. This column has sevel levels being: CT_up (control treatment in upstream sections), L (light treatment in upstream sections), F (flow treatment in upstream sections), L+F (light and flow treatment in upstream sections), CT_down (control treatment in downstream sections), CS (flume with co-occurring stressors), and SS (flume with separated stressors)
- columns G-AT: benthic macroinvertebrate taxa with abundances for each sample.
- background_data (sheet 3): physical (conductivity, dissolved oxygen, pH, and water temperature, with multiparametric probe WTW Multi 3630 IDS Set G) and hydraulic (flow velocity and water depth, with an OTT MF pro electromagnetic current meter) parameters monitored weekly throughout the experiment (during both colonisation and treatment phases).
- ID (column A): unique identifier of each measurement
- flume (column B): mesocosm unit (i.e. flume) number identifier (F1-F18)
- section (column C): section of each flume (being up for upstream tributaries and down for the downstream part)
- side (column D): identifier of the side (left or right for the two upstream tributaries). Samples collected in downstream sections are marked with "__"
- date_probe(column E): measuring day (in date format) for physical parameters
- date_hydro (column F): measuring day (in date format) for hydraulic parameters
- time (column G): measuring day in factor format (from T-3 to T4)
- phase (column H): phase of the experiment (being colonisation or treatment phase)
- treatmet (column I): identifier of the treatment applied. This column has sevel levels being: CT_up (control treatment in upstream sections), L (light treatment in upstream sections), F (flow treatment in upstream sections), L+F (light and flow treatment in upstream sections), CT_down (control treatment in downstream sections), CS (flume with co-occurring stressors), and SS (flume with separated stressors)
- O2_perc (column J): dissolved oxygen (percentage)
- O2_conc (column K): dissolved oxygen (concentration)
- temp (column L): water temperature
- pH (column M): water pH
- cond (column N): water conductivity
- vel (column O): flow velocity
- depth (column P): water depth
- README (sheet 4): a summary of the physical and hydraulic parameters, reporting the meaning of the abbreviations in sheet 3 and the units.
Code/software
R code used for this study can be found at https://github.com/GSBurg/Disconnected. A list of the R package used is reported at the beginning of the script.
Data were collected during a field experiment in Summer 2021, using the Disconnected system, a system of 18 flow-through mesocosms, at the Eußerthal Ecosystem Research Station (EERES, RPTU Kaiserslautern-Landau), in the Palatinate Forest (Rheinland-Pfalz, Germany). The Disconnected system is a mesocosm system designed to mimic a simplified river network, with two upstream tributaries merging together downstream. The experiment lasted seven weeks, with three weeks of colonisation and four weeks of treatment. Drift data were collected weekly during the treatment phase of the experiment. Benthos data were collected at the end of the experiment.
Drift samples were collected by placing drift nets at the outlet of each upstream section. Benthos samples were collected from sampling areas of 150 cm2 with two replicates in each section (six samples per flume).
All macroinvertebrate samples were processed, counting and identifying organisms to a mixed taxonomic level (family or genus).
