Authors of a recently published research article discuss the ecological consequences of large dam removal through their study of instream habitat and invertebrate communities in a temperate, forested catchment in northern Spain.
River restoration by dam removal
During the last decades, increased awareness on the consequences of a degraded environment has led to new legislation, such as the EU Water Framework Directive or the Nature Restoration Law, as well as an exponential increase in the number of river restoration projects. Given the global prevalence of flow regulation and river fragmentation by dams and their strong impact on freshwater ecosystem health, the removal of these infrastructures has become a key restoration action in many river networks.
Nonetheless, knowledge on the ecological effects of dam removal is still scarce, thus impeding a proper design and implementation of dam removal projects. This emerging practice, though potentially beneficial for river ecosystems in the long term, can be a source of impacts in the short term. For instance, the decommissioning of a reservoir can mobilize the accumulated sediments which can negatively affect the biodiversity and functioning of rivers.
Understanding these impacts and the subsequent recovery requires collecting, interpreting, and understanding data. This is precisely what we have done in the study of the Enobieta reservoir.
Enobieta Dam decommissioning as a case study
The Enobieta Dam was designed with a water storage capacity of 2.5 hm3, but geotechnic issues forced the municipality to reduce its capacity to 1.6 hm3 when finished in 1947. The reservoir supplied water to San Sebastian for some decades, but metal concentrations (especially Fe and Mn) were often over legal thresholds for drinking water, and the town faced water shortages because of the small capacity of the modified dam. Therefore, the Añarbe Dam (79-m tall and 43.8 hm3 reservoir capacity) was built further downstream in the catchment in 1976.
Afterwards, the Enobieta Dam lost its strategic value, fell progressively into disuse, and had little or no maintenance, to the point of becoming a safety issue. Indeed, for decades the dam gate and pipes remained closed, and the reservoir was not actively managed, thus meaning that it was permanently full of water. In 2014, to restore the hydrological connectivity in the basin, the municipality removed seven weirs that remained as legacies from past activities (e.g., ironworks). Then, in 2016, managers decided to decommission the last artificial obstacle in the entire headwaters: the Enobieta Dam.
To allow the stabilization of the emerging sediment by the colonizing vegetation and minimize the volume of sediment exported, the reservoir was slowly emptied during 2018 using some old siphons and water serving pipes that mainly released surface water. In December 2018, the bottom gate was repaired and opened, thus starting a period of sediment release with high turbidity episodes. When the reservoir was empty, an older 3.5 m-tall weir emerged 200 m upstream from the large dam.
The local managers demolished this weir in October 2019, and during the very rainy month of November 2019 the Enobieta Stream carved a new channel across the sediment retained by the weir, thus producing a last period of high turbidity. From this moment on, we considered the drawdown process to be finished, thus giving start to what we call the after period.
Key findings and implications
During the before period, the hypoxic conditions in the hypolimnion during stratification resulted in high concentrations of Fe and Mn, although these concentrations diminished with distance downstream. Thus, as expected, benthic macroinvertebrate density, taxa richness and diversity were lower below the dam before decommissioning. During the drawdown period we observed peaks of Fe and Mn, but water quality swiftly recovered during the after period, and nowadays water physicochemical characteristics are similar in control and impact reaches.
Surprisingly, the drawdown of the reservoir did not affect macroinvertebrate communities. Indeed, instead of being affected by the mobilization of the sediments stored in the reservoir, they improved, and thus, the differences between control and impact reaches during the drawdown period decreased. The communities fully recovered and the differences disappeared by the end of the study.
Overall, this study evidences a positive effect of reservoir drawdown on stream structure. Yet, the impact and the recovery process of the decommissioning highly depend on site-specific conditions such as the catchment conservation status or the decommissioning procedure. The success of this project is partly attributed to the excellent conservation status of the Artikutza Valley and the slow drawdown carried out in the Enobieta Reservoir.
Read the full article “Slow drawdown, fast recovery: Stream macroinvertebrate communities improve quickly after large dam decommissioning” in Journal of Applied Ecology.