Projected effects of Climate‐change‐induced flow alterations on stream macroinvertebrate abundances

Abstract

Global change has the potential to affect river flow conditions which are fundamental determinants of physical habitats. Predictions of the effects of flow alterations on aquatic biota have mostly been assessed based on species ecological traits (e.g., current preferences), which are difficult to link to quantitative discharge data. Alternatively, we used empirically derived predictive relationships for species’ response to flow to assess the effect of flow alterations due to climate change in two contrasting central European river catchments. Predictive relationships were set up for 294 individual species based on (1) abundance data from 223 sampling sites in the Kinzig lower‐mountainous catchment and 67 sites in the Treene lowland catchment, and (2) flow conditions at these sites described by five flow metrics quantifying the duration, frequency, magnitude, timing and rate of flow events using present‐day gauging data. Species’ abundances were predicted for three periods: (1) baseline (1998–2017), (2) horizon 2050 (2046–2065) and (3) horizon 2090 (2080–2099) based on these empirical relationships and using high‐resolution modeled discharge data for the present and future climate conditions. We compared the differences in predicted abundances among periods for individual species at each site, where the percent change served as a proxy to assess the potential species responses to flow alterations. Climate change was predicted to most strongly affect the low‐flow conditions, leading to decreased abundances of species up to −42%. Finally combining the response of all species over all metrics indicated increasing overall species assemblage responses in 98% of the studied river reaches in both projected horizons and were significantly larger in the lower‐mountainous Kinzig compared to the lowland Treene catchment. Such quantitative analyses of freshwater taxa responses to flow alterations provide valuable tools for predicting potential climate‐change impacts on species abundances and can be applied to any stressor, species, or region.

Evaluating ecosystem effects of climate change on tropical island streams using high spatial and temporal resolution sampling regimes

“…As predicted, we found a consistent decrease in invertebrate biomass with projected declines in flow rates and increases in flow variability, suggesting that climate change will affect the standing stocks of these important stream consumers. Previous studies have shown that flow alterations triggered by climate change can decrease invertebrate abundance by predictive models (Kakouei et al, ) and long‐term monitoring of two Costa Rican streams (Gutiérrez‐Fonseca, Ramírez, & Pringle, ). In our study, we found that both abundance and biomass of invertebrates declined up to 20× and that caddisflies and midges were the dominant invertebrate taxa that drove these changes in biomass and abundance (Table S4).…”

When is a hydrological model sufficiently calibrated to depict flow preferences of riverine species?

“…This is necessary because hydrological models perform weakly in predicting IHAs if they are not included in the optimization process (Kiesel et al, ; Pool et al, ; Vigiak et al, ). The IHAs were selected based on Kakouei et al (), who investigated the most important and not cross‐correlated IHAs for the occurring species in the Treene and Kinzig. Kakouei et al () selected the most important variables for the taxa of each catchment by applying a boosted regression tree analysis.…”

“…The IHAs were selected based on Kakouei et al (), who investigated the most important and not cross‐correlated IHAs for the occurring species in the Treene and Kinzig. Kakouei et al () selected the most important variables for the taxa of each catchment by applying a boosted regression tree analysis. A stepwise process was then applied to exclude an IHA variable with the lower relative influence on the taxa, once cross‐correlation between two variables reached the sensitivity threshold of 0.7.…”

“…In particular, macroinvertebrates have evolved distinct adaptations to flow conditions and hence are affected by their changes (Domisch et al, ). These dependencies on flow have been used frequently to assess the occurrence (Pyne & Poff, ) and diversity (Poff & Zimmerman, ) of riverine species, such as fish (O'Keeffe et al, ), benthic invertebrates (Armanini, Horrigan, Monk, Peters, & Baird, ), or phytoplankton (Qu, Wu, Guse, & Fohrer, ), using microcosm experiments (Ceola et al, ), statistical models (Kakouei et al, ), or process‐based models (Mondy & Schuwirth, ). In the absence of direct measurement data or for scenario assessments, modelled streamflow is often used as a data basis for such analysis.…”

The role of floods and droughts on riverine ecosystems under a changing climate

“…In a more complex approach, but typically applied at finer spatial scales, climate change forcing is propagated through a modelling cascade consisting of a hydrological model loosely coupled with a habitat suitability or species distribution model (Jaeger, Olden, & Pelland, ; Kakouei et al, ; Kuemmerlen et al, ; Morid, Delavar, Eagderi, & Kumar, ; Muñoz‐Mas, Lopez‐Nicolas, Martinez‐Capel, & Pulido‐Velazquez, ; Mustonen et al, ; Viganò et al, ; Woznicki, Nejadhashemi, Tang, & Wang, ). For example, Jaeger et al () predicted a higher frequency of zero‐flow days in an intermittent stream in Arizona, United States, which would inevitably lead to increased channel fragmentation and a reduced network‐wide hydrological connectivity during spawning of native fish.…”