We tested whether the recovery of riparian vegetation along rapids that have been restored after channelisation for timber floating can be predicted based on floristic and geomorphic characteristics of surrounding landscape units. Our study was located along tributary stream networks, naturally fragmented in rapids, slow-flowing reaches, and lakes (i.e. process domains), in the Vindel River catchment in northern Sweden.

We tested whether landscape characteristics, specifically to what extent the geomorphology (affecting local abiotic conditions), species richness, and species composition (representing the species pool for recolonisation), as well as the proximity to various upstream process domains (determining the dispersal potential), can predict post-restoration recovery of riparian vegetation.

Our results indicate that post-restoration recovery of riparian vegetation richness or composition is not strongly related to landscape-scale species pools in these streams. The restored rapids were most similar to upstream rapids, geomorphically and floristically, including plant traits. Species richness of adjacent landscape units (upstream process domains or lateral upland zone) did not correlate with that of restored rapids, and proximity of upstream rapids or other process domains was only weakly influential, thus diminishing support for the hypothesis that hydrochory or other means of propagule dispersal plays a strong role in riparian vegetation community organisation after restoration in this fragmented stream network.

We conclude that, in these naturally fragmented stream systems with three discrete process domains (rapids, slow-flowing reaches and lakes), hydrochory is probably not the main predictor for short-term riparian vegetation recovery. Therefore, other factors than landscape context can serve in prioritising restoration and, in these systems, local factors are likely to outweigh landscape connectivity in the recovery of riparian vegetation.

The configuration of channels in stream networks is vital for their connectivity, biodiversity, and metacommunity dynamics. We compared the capacity of three process domains-lakes, slow-flowing reaches, and rapids-to disperse and retain plant propagules by releasing small wooden cubes as propagule mimics during the spring flood and recording their final locations. We also measured the geomorphic characteristics (planform, longitudinal profile, cross-sectional morphology, and wood) of each process domain. The three process domains all differed in morphology and hydraulics, and those characteristics were important in shaping the transport capacity of mimics. On average, lakes retained more mimics than slow-flowing reaches but did not differ from the retainment of rapids. Living macrophytes were the most efficient element trapping mimics. In rapids and slow-flowing reaches, most trapped mimics remained floating, whereas in lakes, most mimics ended up on the banks. The decay curves of retention varied substantially among and within process domains. The results suggest that managers who rely on natural recovery of restored sites by means of plant immigration may benefit from understanding landscape patterns when deciding upon the location of restoration measures in stream networks.

Stream ecological theory predicts that the use of allochthonous resources declines with increasing channel width, while at the same time primary production and autochthonous carbon use by consumers increase. Although these expectations have found support in several studies, it is not well known how terrestrial runoff and/or inputs of primary production from lakes alter these longitudinal patterns. To investigate this, we analyzed the diet of filter-feeding black fly and caddisfly larvae from 23 boreal streams, encompassing gradients in drainage area, land cover and land use, and distance to nearest upstream lake outlet. In five of these streams, we also sampled repeatedly during autumn to test if allochthony of filter feeders increases over time as new litter inputs are processed. Across sites, filter-feeder autochthony was 21.1-75.1%, did not differ between black fly and caddisfly larvae, was not positively related to drainage area, and did not decrease with distance from lakes. Instead, lake and wetland cover promoted filter-feeder autochthony independently of stream size, whereas catchment-scale forest cover and forestry reduced autochthony. Further, we found no seasonal increase in allochthony, indicating low assimilation of particles derived from autumn litter fall. Hence, catchment properties, rather than local conditions, can influence levels of autochthony in boreal streams.

Many stream restoration projects aim to increase geomorphic complexity, assuming that this increases habitat heterogeneity and, thus, biodiversity. However, empirical data supporting these linkages remain scant. Previous assessments of stream restoration suffer from incomplete quantification of habitat complexity, or a narrow focus on only one organism group and/or one restoration measure, limiting learning. Based on a comprehensive quantification of geomorphic complexity in 20 stream reaches in northern Sweden, ranging from streams channelized for timber floating to restored and reference reaches, we investigated responses of macroinvertebrates, diatoms, and macrophytes to multiple geomorphic metrics. Sediment size heterogeneity, which was generally improved in restored sites, favored macroinvertebrate and diatom diversity and macroinvertebrate abundance. In contrast, macrophyte diversity responded to increased variation along the longitudinal stream profile (e.g., step-pools), which was not consistently improved by the restoration. Our analyses highlight the value of learning across multiple restoration projects, both in identifying which aspects of restoration have succeeded, and pinpointing other measures that might be targeted during adaptive management or future restoration. Given our results, a combination of restoration measures targeting not only sediment size heterogeneity, but also features such as step-pools and instream wood, is most likely to benefit benthic biota in streams.

Ecosystem engineers substantially alter physical flow characteristics and shape a river's form and function. Because the recurrence interval of geomorphic processes and disturbances in rivers commonly match the temporal scale of plants' life cycles or alterations by animals, the resulting feedbacks are an important component of rivers. In this review, we focus on biota that directly or indirectly induce a physical change in rivers and cause positive feedbacks on the functioning of that organism. We provide an overview of how various ecosystem engineers affect rivers at different temporal and spatial scales and plot them on a conceptual gradient of river types. Various plants engineer the river environment through stabilizing sediment and reducing flow velocities, including macrophytes, woody plants, and algal mats and biofilms. Among animals that engineer, beaver that build dams cause substantial changes to river dynamics. In addition, benthic macroinvertebrates and mussels can stabilize sediment and reduce velocities, and aquatic and riparian grazers modulate the effect of plants. Humans are also considered river ecosystem engineers. Most of the ecosystem engineers reported in literature occur in rivers with low to intermediate relative stability, intermediate channel widths, and small to intermediate grain sizes. Ecosystem engineers that create positive biogeomorphic feedbacks are important to take into account when managing river systems, as many common invasive species are successful due to their engineering capabilities. River restoration can use ecosystem engineers to spur holistic recovery. Future research points towards examining ecosystem engineers on longer spatial and temporal scales and understanding the co-evolution of organisms and landforms through engineering. 

1. Habitat restoration is increasingly undertaken in degraded streams and rivers to help improve biodiversity and ecosystem functioning. Follow-up assessments focused on outcomes for biodiversity have often found scant evidence for recovery, raising concerns about the efficacy of habitat restoration for improving ecological integrity. However, responses of other ecological variables, such as ecosystem process rates and the functional trait composition of biological assemblages, have been little evaluated.

2. We assessed how the restoration of habitat heterogeneity affected multiple functional parameters in 20 boreal stream reaches encompassing both more and less extensively restored sites, as well as channelised and natural reference sites. We further assessed relationships between our functional parameters and a fluvial geomorphic measure of habitat heterogeneity.

3. Leaf decomposition was positively related to habitat heterogeneity. This was associated with shifts in the functional composition of detritivore assemblages, with the most obligate litter consumers more prominent in reaches showing higher habitat heterogeneity. The deposition of fine particulate organic matter was consistently higher in restored than channelised sites, and was positively related to the heterogeneity gradient. Algal biomass accrual per unit area did not vary either with restoration or the heterogeneity gradient.

4. Synthesis and applications. Our findings demonstrate that restoration of river habitat heterogeneity can enhance retention and decomposition of organic matter, key ecosystem properties underpinning ecosystem functioning and service delivery. Significantly, enhanced litter decomposition was linked with a change in the functional composition rather than diversity of detritivore assemblages. Future evaluation of the success of habitat restorations should incorporate quantification of ecosystem processes and the functional traits of biota, in addition to measures of fluvial geomorphology and more traditional biotic metrics, to facilitate a more comprehensive and mechanistic assessment of ecological responses.

Although ecological restoration generally succeeds in increasing physical heterogeneity, many projects fail to enhance biota. Researchers have suggested several possible explanations, including insufficient restoration intensity, or time-lags in ecological responses that prevent detection of significant changes in short-term monitoring programs. This study aims to evaluate whether benthic macroinvertebrate communities responded to an expanded set of stream restoration measures within a study period of one to five years after completion of the restoration project. We studied 10 forest streams in northern Sweden that were channelized in the past for timber floating. Managers subjected six of these streams to habitat restoration, on each of these we selected two reaches, located in close proximity but differing in restoration intensity. In basic restored reaches, the restoration managers broke up the channelized banks and returned cobbles and small boulders to the main channel. In enhanced restoration reaches, they added additional large wood and boulders to reaches previously subjected to basic restoration, and rehabilitated gravel beds. The remaining four streams were not restored, and thus represent the baseline impacted (channelized) condition. We surveyed stream benthic assemblages before the enhanced restoration (year 2010) and three times afterward between 2011 and 2015. Five years after restoration, macroinvertebrate assemblages at the enhanced restored reaches were more differentiated from channelized conditions than those at basic-restored reaches. This reflected increased relative abundances of the insect orders Ephemeroptera and Trichoptera and the bivalve molluscs Sphaeriidae and decreased relative abundances of Chironomidae (Diptera). Analysis of functional traits provided further insights on the mechanistic explanations driving the recovery, e.g., indicating that the augmented channel retention capacity at enhanced restored reaches favored taxa adapted to slow flow conditions and more effectively retained passive aquatic dispersers. The increased restoration intensity in enhanced restored reaches has resulted in shifts in the composition of benthic macroinvertebrate assemblages, including increases in more sensitive taxa. These shifts became fully apparent five years after the enhanced restoration. Our results emphasize the value of longer-term monitoring to assess ecological responses following restoration, and of undertaking additional restoration as a valuable management option for previously restored sites that failed to achieve biotic recovery.

Restoration of channelized streams by returning coarse sediment from stream edges to the wetted channel has become a common practice in Sweden. Yet, restoration activities do not always result in the return of desired biota. This study evaluated a restoration project in the Vindel River in northern Sweden in which practitioners further increased channel complexity of previously restored stream reaches by placing very large boulders (> 1 m), trees (> 8 m), and salmonid spawning gravel from adjacent upland areas into the channels. One reach restored with basic methods and another with enhanced methods were selected in each of ten different tributaries to the main channel. Geomorphic and hydraulic complexity was enhanced but the chemical composition of riparian soils and the communities of riparian plants and fish did not exhibit any clear responses to the enhanced restoration measures during the first 5 years compared to reaches restored with basic restoration methods. The variation in the collected data was among streams instead of between types of restored reaches. We conclude that restoration is a disturbance in itself, that immigration potential varies across landscapes, and that biotic recovery processes in boreal river systems are slow. We suggest that enhanced restoration has to apply a catchment-scale approach accounting for connectivity and availability of source populations, and that low-intensity monitoring has to be performed over several decades to evaluate restoration outcomes.

We review the predicted changes in extreme events following climate change in flowing waters in arctic and subarctic regions. These regions are characterised by tundra or taiga ecosystems in either erosional or depositional glacial landforms or presently glacierised areas of the Northern Hemisphere. The ecological and geomorphic effects of extreme meteorological and hydrological events, such as episodes of strongly increased precipitation, temperatures and flows, can be exacerbated by altered base conditions. For example, winter temperature variations between frost and thaw will become more frequent at many places because mean temperature during the winter is closer to 0 °C, potentially leading to changes in the production of ice and intensified disturbance of riparian and aquatic habitats during extreme floods. Additionally, thawing of permafrost and glaciers can lead to increased bank erosion because of thaw slump and glacial outburst floods. We discuss the abiotic and biotic effects of these and other extreme events, including heavy precipitation, floods, drought and extreme air or water temperatures, and summarise our findings in a model that aims to stimulate further research in this field.