Impacts of environmental stressors on food webs are often difficult to predict because trophic levels can respond in divergent ways, and biotic interactions may dampen or amplify responses. Here we studied food-web level impacts of urban wastewater pollution, a widespread source of degradation that can alter stream food webs via top-down and bottom-up processes. We selected ten pairs of sampling locations (i.e., twenty sites total) in ten small-to-medium tributaries of the Ebro River (NE Iberian Peninsula; see "Related article" for locations). These streams were located in low-mountain habitats (365 m – 950 m a.s.l.) under the influence of Mediterranean climate and flow regimes (mean daily discharge 0.05 ± 0.09 m³·s^-1). Hydrology in the region is characterized by dry summer periods with reduced flows, and rainfall episodes in spring and late fall. We surveyed the sites in early spring (April 15 - 23, 2016) to avoid spring floods; the last scouring flood preceding our sampling took place 5 months before sampling (November 2nd, 2015). Each sampling location was upstream or downstream of an urban wastewater effluent. Sewage impacts represented the main stressor influencing downstream community composition. The present dataset contains  (i) aquatic invertebrate and vertebrate community composition, and (ii) Stable Isotope Analysis data (13C/12C and 15N/14N) for each taxa and reach.

Characterization of faunal community composition: We sampled macroinvertebrate and vertebrate communities at each site. In sites dominated by cobbles (n=5), we collected five macroinvertebrate samples using a Surber sampler (30 x 30 cm, mesh size 250 µm), covering the different microhabitats. In streams where sand dominated (n=5), we obtained the five samples using a core sampler (Ø = 25 cm). Samples were preserved in 4% formalin, and invertebrates were sorted, counted, and identified to the lowest taxonomic level possible in the laboratory (usually genus). We measured body length of the first twenty-five individuals of each taxon, from which we derived biomass (mg dry weight) using published length-mass relationships (e.g., Burgherr and Meyer, 1997). We sampled aquatic vertebrates (fish, amphibians and reptiles) at each site via 3-pass depletion of a 100 m reach delimited by blocking nets, using a portable generator and an Electracatch WFC4 with pulsed direct current (DC) output at a frequency of 50 Hz and 1.5 amps. All vertebrates were identified at the species level, measured (total length), and weighed. Three individuals by site and size class were euthanized and frozen in the field (-20 ºC, for SIA). Species of conservation concern (e.g., the Mediterranean pond turtle, Mauremys leprosa) were measured and released immediately.

Stable Isotope Analyses (SIA): We collected samples for SIA at each reach, targeting the diversity of trophic levels and feeding strategies. We collected and sorted macroinvertebrates in the field to the lowest level possible, and immediately froze them. In the laboratory, invertebrates were thawed and identified mostly to species level (except for Diptera, where we used a combination of genus and subfamily; and Oligochaeta, family level). Up to three independent samples (each 0.4 mg of dry weight) were prepared for every taxon to be processed for SIA. These samples included one or several individuals from the same taxa, size, and reach. Invertebrate gut contents and calcified structures such as shells were removed prior to processing. We obtained a muscle tissue sample from the flanks of each vertebrate. We dried samples at 60ºC until they reached constant weight (usually for 24 hours), and we later ground them using a mortar and pestle. Samples were analyzed for stable isotope ratios (13C/12C and 15N/14N) on a Flash 1112 elemental analyzer connected to a Delta C isotopic ratio mass spectrometer with Conflo III interface (Thermo Scientific, Inc.). Analytical precision from multiple runs was 0.1‰ for δ13C, and 0.2‰ for δ15N. Stable isotope results were expressed as the difference between sample ratios and standards (PeeDee Belemnite limestone for δ13C and atmospheric nitrogen for δ15N). We used these stable isotope data to examine the mechanisms of CSS-slope change, and potential niche compression (see related article).

Complementary data:  

(1) Concentrations of pharmaceutical products can be found in: Mandaric, L., J. R. Mor, S. Sabater and M. Petrovic. 2018. Impact of urban chemical pollution on water quality in small, rural and effluent-dominated Mediterranean streams and rivers. Science of the Total Environment 613–614: 763–772. doi: 10.1016/j.scitotenv.2017.09.128

(2) Gross primary production (GPP) values for each site can be found in: Pereda, O., D. von Schiller, G. García-Baquero, J. R. Mor, V. Acuña, S. Sabater, and A. Elosegi. 2021. Combined effects of urban pollution and hydrological stress on ecosystem functions of Mediterranean streams. Science of The Total Environment, 753, 141971.

Mor_et_al_Ecology_community_composition: Vertebrate and invertebrates abundance (ind m^-2; "Abd_m2") and mean species body size per site (mg; "CWMBS"). Up (upstream) and Do (downstream) denote pairs of sites located upstream-downstream of wastewater effluents. 

Mor_et_al_Ecology_SIA_fauna: Stable isotope analyses (SIA) data by taxon and site. Up (upstream) and Do (downstream) denote pairs of sites located upstream-downstream of wastewater effluents. “Species” category denotes lowest taxonomic level possible (usually species or genus).