Norway launched a national action plan on wetland restoration in 2016. So far, 90% of the restoration effort has been on peatland restoration, with about 140 mires restored so far. There are three main restoration goals stated in the action plan: 1) limit greenhouse gas (GHG) emission, 2) climate adaptation and 3) improved ecological condition. Quantifying the outcome of the restoration actions is necessary to evaluate whether the goals of the action plan are met. A vegetation monitoring protocol was suggested before restoration started and has been implemented at five restoration sites. As the peatland restoration effort in Norway is increasing, it is timely to evaluate if the data currently collected can measure peatland restoration outcome. We evaluate the monitoring protocol based on statistical analyses of the data collected at two sites, describe how indicator species can be identified using generalized composition data used as the basis for classifying habitats in Norway (EcoSyst framework), and suggest the way forward for peatland restoration monitoring in Norway. Data collected according to the monitoring protocol can document changes in species composition at restoration sites, but has limitations when the ecological complexity at the sites increases and reference sites are unavailable. We argue that adjusting the monitoring protocol will: 1) facilitate alignment with existing peatland research; 2) connect better with monitoring programs where data is collected applying EcoSyst framework principles; and 3) enable upscaling to cover the wide variation emerging in peatland restoration.

Like large carnivores, hunters both kill and scare ungulates, and thus might indirectly affect plant performance through trophic cascades. In this study, we hypothesized that intensive hunting and enduring fear of humans have caused moose and other forest ungulates to partly avoid areas near human infrastructure (perceived hunting risk), with positive cascading effects on recruitment of trees. Using data from the Norwegian forest inventory, we found decreasing browsing pressure and increasing tree recruitment in areas close to roads and houses, where ungulates are more likely to encounter humans. However, although browsing and recruitment were negatively related, reduced browsing was only responsible for a small proportion of the higher tree recruitment near human infrastructure. We suggest that the apparently weak cascading effect occurs because the recorded browsing pressure only partly reflects the long‐term browsing intensity close to humans. Accordingly, tree recruitment was also related to the density of small trees 5–10 years earlier, which was higher close to human infrastructure. Hence, if small tree density is a product of the browsing pressure in the past, the cascading effect is probably stronger than our estimates suggest. Reduced browsing near roads and houses is most in line with risk avoidance driven by fear of humans (behaviorally mediated), and not because of excessive hunting and local reduction in ungulate density (density mediated).

Aims: The area influenced by road construction is large, and measures to re‐establish vegetation in disturbed areas are routinely carried out to reduce impacts on biodiversity. However, goals of mitigation measures are often unclear, and the effects on biodiversity of mitigation measures is rarely monitored. We assessed the effects of different revegetation treatments (natural revegetation, seeding, planting) on vegetation development along highways, and on wildlife crossings of different age.Location: Highways in southeast Norway.Methods: We collected data on vascular plant species, vegetation cover and height, soil grain size and organic matter content, and compared the species composition, richness, and diversity of the restored sites with reference plots in the adjacent target vegetation (mature forest).Results: Our results show a significantly higher richness and diversity in restored plots compared to reference plots, and an increased similarity of species composition over time. Species composition was most similar to reference plots in naturally revegetated plots and seeding seemed to reduce both species and functional trait composition similarity.Conclusions: It is unrealistic that the defined target vegetation will develop on restored sites. Defining a realistic and achievable target vegetation for each road construction project in relation to land use, adjacent vegetation type and successional stage, as e.g., forest edge instead of forest, would be useful. While this may require more effort for management it will translate to higher mitigation success.

Alpine areas worldwide are under heavy land‐use pressure and degradation. Active restoration treatments can contribute to accelerating recovery of degraded areas. However, monitoring data are needed to understand the contribution of restoration treatments to long‐term management and to predict time to recovery (TR). In this study, we used monitoring data on removed roads in an alpine area in Norway to investigate TR of three vegetation‐based indicators. Four restoration treatments were tested: 1) removal of added gravel down to original terrain surface, and stirring of topsoil; 2) adding fertilizer to the stirred topsoil; 3) adding seeds to the fertilized topsoil; and 4) no removal of added gravel, but stirring of top layer (gravel and soil). The restoration of roads took place in 2002, and monitoring of permanent plots was carried out in 2004, 2009, 2014, and 2019. Reference plots in intact vegetation next to removed roads were monitored in 2014 and 2019. We used species composition and species richness of vascular plants as well as total vegetation cover as indicators of restoration outcome and investigated predicted TR for these indicators under different restoration treatments. Species composition changed significantly with time since restoration in all treatments, approaching that of the reference vegetation. The recovery of species composition was slowest in fertilized and seeded plots, where estimated TR was 2–3 times longer (> 45 years) than in the other treatments (< 20 years). Species richness of vascular plants was restored quickly (< 5 years) within all restoration treatments, whereas recovery of vegetation cover varied more (20–30 years). Our study confirms that vegetation recovery in alpine environments is a long‐term process, but that adding seeds and nutrients is unnecessary for, and even inhibits, the recovery of narrow, disturbed sites such as former roads.

In this article, we present the case for Generalized Linear Latent Variable Models (GLLVMs) as a go-to choice of statistical method for any community ecologist wanting to tackle a range of present-day ecological research questions. GLLVMs bring tools and capabilities from classic (mixed-effects) regression models to multivariate community analysis, providing a number of novel ways to tailor models specifically to one’s study questions and data properties not available when using non-model-based multivariate methods. In order to facilitate further adoption of these methods by community ecologists, we provide 1) a practitioner-focused and practical overview of the advantages the GLLVM framework brings to the table when addressing different core ecological questions, 2) a number of concrete suggestions for how GLLVMs best can be incorporated into the analytical workflow of community ecologists, and 3) two illustrative worked examples of this workflow in action on real-world data.