Update manuscript.Rmd

manuscript.Rmd

@@ -44,7 +44,7 @@ Restoration Ecology, TUM School of Life Sciences, Technical University of Munich
 1.  Sowing is a well-established restoration technique to overcome dispersal limitation. Seed mixtures adapted to certain environmental conditions, like substrate or microclimate, are most effective to achieve functional communities. This is especially important if the restored vegetation has to protect critical infrastructure like roadsides and dikes. Here, an improved seed--substrate combination will secure slope stability, make restorations more effective, and generate species-rich grasslands.
 2.  A factorial field experiment addressed this topic on a dike at River Danube in SE Germany in 2018--2021. Within 288 plots, we tested three sand admixtures, two substrate depths, two seed densities and two seed mixture types (mesic hay meadow, semi-dry calcareous grassland) in north and south exposition, and measured the recovery completeness by calculating the successional distance to reference sites, the persistence of sown species, and the Favourable Conservation Status (FCS) of target species.
 3.  Overall, the sown vegetation developed in the desired direction, but a recovery debt remained after four years, and some plots still showed similarities to negative references from ruderal sites. In north exposition, hay meadow-seed mixtures developed closer to the respective reference communities than dry-grassland mixtures.
-4.  In south exposition, the sown communities developed poorly which might be due to a severe drought during establishment. This initial negative effect remained over the entire observation period.
+4.  In south exposition, the sown communities established poorly which might be due to a severe drought during establishment. This initial negative effect remained over the entire observation period.
 5.  Sand admixture had a slightly positive effect on target variables, while the tested substrate depths, seed densities and mixture types had no effects on species persistence or FCS.
 6.  Synthesis and applications: Site-adapted seed mixtures make restoration more effective, while applying several seed--substrate combinations might foster beta diversity. Furthermore, additional management efforts are recommended, as they might be necessary to reduce the recovery debt, as well as re-sowing after unfavourable conditions.
 
@@ -78,7 +78,7 @@ Species composition
 
 Grasslands can support an exceedingly high biodiversity and they provide several ecosystem services [@Dengler.2014; @Bardgett.2021]. However, they are globally endangered [@Bardgett.2021], and in Europe, calcareous grasslands and hay meadows are red-listed habitats [Category 3, 'vulnerable', @Janssen.2016]. Restoration is seen as a key factor to sustain biodiversity and ecosystem services [@COP.2014; @UnitedNations.2019b], and sowing is a well-established approach to establish species-rich grasslands [@Kiehl.2010]. Sowing high-diversity mixtures of regional provenance [@bucharova2019] produced by specialized companies is a promising way to scale up restoration efforts [@Freitag.2021], and to overcome dispersal filters [@myers2009; @orrock2023]. However, there are still open questions about adjusting seed mixtures to specific site conditions and future climate conditions [@Torok.2021].
 
-Restoration ecology can increase the predictability of restoration approaches [@Mouquet.2015] by using rigorous, repeatable, and transparent experiments based on advanced theory, which will finally strengthen evidence-based restoration [@Cooke.2018; @Wainwright.2018]. Local site conditions and the restoration method are key predictors for vegetation development after sowing [@Brudvig.2017b]. The main assembly processes are habitat and biotic filtering which can be manipulated by the choice of seed--substrate combinations [@Torok.2017]. This means a close adaptation of the substrate to the niche of the target species or of the seed mixtures to the characteristics of the chosen substrate. Suitable substrates reduce habitat filtering of the seeded species, while specific seed mixture minimises competitive exclusion of desired species and simultaneously prohibit invasive species [@Funk.2008]. Modifying seed mixtures to match the site conditions could be based on functional plant traits [@Funk.2008; @Laughlin.2014; @Balazs.2020], although this is not easy to implement [@merchant2022; @Bauer.2022]. This challenge is particularly interesting for artificial substrates that can be easily modified and that are used in urban areas [@Bauer.2022], in quarries [@chenot-lescure2022], or on dikes [@Liebrand.1996].
+Restoration ecology can increase the predictability of restoration approaches [@Mouquet.2015] by using rigorous, repeatable, and transparent experiments based on advanced theory, which will finally strengthen evidence-based restoration [@Cooke.2018; @Wainwright.2018]. Local site conditions and the restoration method are key predictors for vegetation development after sowing [@Brudvig.2017b]. The main assembly processes are habitat and biotic filtering which can be manipulated by the choice of seed--substrate combinations [@Torok.2017]. This means a close adaptation of the substrate to the niche of the target species or of the seed mixtures to the characteristics of the chosen substrate. Suitable substrates reduce habitat filtering of the seeded species, while specific seed mixture minimises competitive exclusion of desired species and simultaneously prohibit invasive species [@Funk.2008]. Modifying seed mixtures to match the site conditions could be based on functional plant traits [@Funk.2008; @Laughlin.2014; @Balazs.2020], although this is not easy to implement [@merchant2022; @Bauer.2022]. This challenge is particularly interesting for artificial substrates that can be modified and are often used in urban areas [@Bauer.2022], quarries [@chenot-lescure2022], or dikes [@Liebrand.1996].
 
 Dikes are promising sites for the restoration of species-rich grasslands because they can increase habitat area and connectivity of semi-natural grasslands and therefore significantly contribute to biodiversity conservation in agricultural landscapes [@Batori.2020]. Several ecosystem functions can be reconciled by dike grasslands like erosion control [@husicka2003; @berendse2015] and biodiversity [@teixeira2022], which can be fostered by an adapted seed--substrate combination. Experiments with such seed-substrate combinations on dikes benefit from contrasting microclimates of the different expositions of the steep slopes (\>1:3) [@suggitt2010].
 
@@ -102,7 +102,7 @@ The success of restoration, i.e., the difference from desired conditions, is eva
 
 Specific combinations of seed mixtures and substrates ('seed--substrate combinations') were tested on an existing dike covered by grassland at the Danube River in SE Germany (Figure \@ref(fig:map); 314 m a.s.l.; WGS84: lat/lon, 48.83895/12.88412). The climate of the region is temperate-suboceanic with a mean annual temperature of 8.4 °C and an annual precipitation of 984 mm [@dwd2021]. During the study, three exceptionally dry years (2018--2020) occurred [Appendix A1, @hari2020], as well as three minor floods, which, though, did not reach the plots (Appendix A1). The substrates consisted of calcareous sand (0--4 mm grain size) and agricultural soil obtained from a nearby dike construction site near the village of Steinkirchen. A big roller mixed both components and an excavator filled the substrates in the dug plots.
 
-The target vegetation types were typical grassland types for Central Europe: lowland mesic hay meadows and semi-dry calcareous grassland [EUNIS codes: R22, R1A, @Chytry.2020; Arrhenatherion elatioris and Cirsio-Brachypodion pinnati according to the EuroVegChecklist: CM01A, DA01B, @mucina2016]. The species pool for seed mixtures of hay meadows and dry grasslands consisted of 55 and 58 species, respectively. The seeds were supplied by a commercial producer of autochthonous seeds [Co. Krimmer, Pulling, source area 16, @prasse2010]. From these species pools, 20 species were selected for each plot in a stratified randomised manner (Appendix A2). The aim of these random and unique subsamples was to test types of seed mixtures and not a certain species composition. Each mixture contained seven grasses (60wt% of total seed mixture), three legumes (5%) and ten further non-legume forbs (35%) (Table \@ref(tab:seedmix)). The hay-meadow mixtures had higher community-weighted means (CWM) for specific leaf area (SLA), lower for seed mass, and higher for canopy height than the dry-grassland mixtures (Appendix A3). The south-exposed plots were sown in mid-April 2018 and the north exposition 14 days later. In October 2018, *Bromus hordeaceus* was sown as a nursery grass to provide safe sites under drought conditions. In late-April 2018 due to the drought, the south exposition was protected by a geotextile consisting of straw chaff (350 g m^-2^) which was removed after two weeks due to unsatisfactory effects on seedling emergence. The management started with a cut at 20 cm height without hay removal in August 2018, followed by standard deep cuts with hay removal in July 2019 and 2020. The surrounding area of the plots was mown thrice a year and the first time before flowering in May.
+The target vegetation types were typical grassland types for Central Europe: lowland mesic hay meadows and semi-dry calcareous grassland [EUNIS codes: R22, R1A, @Chytry.2020; Arrhenatherion elatioris and Cirsio-Brachypodion pinnati according to the EuroVegChecklist: CM01A, DA01B, @mucina2016]. The species pool for seed mixtures of hay meadows and dry grasslands consisted of 55 and 58 species, respectively. The seeds were supplied by a commercial producer of autochthonous seeds [Co. Krimmer, Pulling, source area 16, @prasse2010]. From these species pools, 20 species were selected for each plot in a stratified randomised manner (Appendix A2). The aim of these random and unique subsamples was to test types of seed mixtures and not a certain species composition. Each mixture contained seven grasses (60wt% of total seed mixture), three legumes (5%) and ten further non-legume forbs (35%) (Table \@ref(tab:seedmix)). The hay-meadow mixtures had higher community-weighted means (CWM) for specific leaf area (SLA), lower for seed mass, and higher for canopy height than the dry-grassland mixtures (Appendix A3). The south-exposed plots were sown in mid-April 2018 and the north exposition 14 days later. In October 2018, *Bromus hordeaceus* was sown as a nursery grass to provide safe sites under drought conditions. In late-April 2018 due to the drought, the south exposition was protected by a geotextile consisting of straw chaff (350 g m^-2^) which was removed after two weeks due to unsatisfactory effects on seedling emergence. The management started with a cut at 20 cm height without hay removal in August 2018, followed by standard deep cuts with hay removal in July 2019 and 2020. The surrounding area of the plots was mown thrice a year, and the first time before flowering in May.
 
 We used 288 plots of the size 2.0 m × 3.0 m, vertically oriented, halfway up the dike slopes (1:2), distributed over the north and south exposition, and arranged in six blocks (=replicates). The experiment used a split-plot design combined with a randomised complete block design (Figure \@ref(fig:map)). The split plot was created by the two expositions of the dike, where all 24 treatment combinations were tested, i.e., sand admixtures (0, 25, and 50%), soil depths (15 vs. 30 cm), the two seed mixture types, and two seed densities (4 vs. 8 g m^-2^). @Kiehl.2010 recommend 1--5 g m^-2^ for grassland restoration and @KleberLerchbaumer.2017 recommend an increased density of 5--8 g m^-2^ for slopes.
 
@@ -172,15 +172,15 @@ Our aim was to identify perfect seed--substrate combinations regarding restorati
 
 Our results suggest that adapted seed mixtures can increase restoration effectiveness by sowing hay meadows in the north but not necessarily in south exposition of dikes. Furthermore, the reduction of the nutrient load through sand admixture was positive, albeit with small effect size. The question remains if sand admixture is the most efficient restoration measure to promote diversity on dikes. Increasing seed density on dike slopes does not appear to be necessary which contradicts common recommendations [@KleberLerchbaumer.2017], and soil depths of 30 cm are not adverse compared to 15 cm thick substrates.
 
-There were no perfect seed--substrate combinations and thus we conclude that a variation of seed mixture types and different substrates along restoration sections would promote biodiversity more than a single uniform solution [@bauer2023; @Holl.2022]. Negative effects of drought in the sowing season might require re-sowing. Restoration projects should account for the increasing frequency of droughts [@naumann2018] by re-sowing or by combining seeding with hay transfer [@török2012] to improve the microclimate during establishment [@eckstein2005]. We expect a minor effect of succession in the next ten years, which requires further interventions to close the recovery debt. Management adaptation modifies the biotic filter and is a crucial factor in addition to the restoration approach and the site characteristics for restoration success [@Grman.2013; @tölgyesi2022]. For example, the introduction of sheep grazing on the experimental plots, which already exists in the surroundings, will modify the disturbance regime and improve dispersal. Overall, our results support the finding that restored dike grasslands can promote biodiversity in agricultural landscapes [@Batori.2020]. However, the recovery debt highlights the fact that restored grasslands cannot substitute old-growth grasslands [@nerlekar2020].
+There were no perfect seed--substrate combinations and thus we conclude that a variation of seed mixture types and different substrates along restoration sections would promote biodiversity more than a single solution [@bauer2023; @Holl.2022]. Negative effects of drought in the sowing season might require re-sowing. Restoration projects should account for the increasing frequency of droughts [@naumann2018] by re-sowing or by combining seeding with hay transfer [@török2012] to improve the microclimate during establishment [@eckstein2005]. We expect a minor effect of succession in the next ten years, which requires further interventions to close the recovery debt. Management adaptation modifies the biotic filter and is a crucial factor in addition to the restoration approach and the site characteristics for restoration success [@Grman.2013; @tölgyesi2022]. For example, the introduction of sheep grazing on the experimental plots, which already exists in the surroundings, will modify the disturbance regime and improve dispersal. Overall, our results support the finding that restored dike grasslands can promote biodiversity in agricultural landscapes [@Batori.2020]. However, the recovery debt highlights the fact that restored grasslands cannot substitute old-growth grasslands [@nerlekar2020].
 
 \clearpage
 
 \newpage
 
 # Acknowledgements {.unnumbered}
 
-We would like to thank our project partners Dr. Markus Fischer, Frank Schuster, and Christoph Schwahn (WIGES GmbH) as well as Stefan Radlmair and the late Klaus Rachl (Government of Lower Bavaria) for numerous discussions on restoration and management of dike grasslands. Fieldwork was supported by Clemens Berger and Uwe Kleber-Lerchbaumer (Wasserwirtschaftsamt Deggendorf). We thank Holger Paetsch, Simon Reith, Anna Ritter, Jakob Strak, Leonardo H. Teixeira, and Linda Weggler for assisting with the field surveys or soil analyses in 2018--2020. The German Federal Environmental Foundation (DBU) supported MB with a doctoral scholarship. Thank you also to three anonymous referees for their valuable suggestions improving the manuscript.
+We would like to thank our project partners Dr. Markus Fischer, Frank Schuster, and Christoph Schwahn (WIGES GmbH) as well as Stefan Radlmair and the late Klaus Rachl (Government of Lower Bavaria) for numerous discussions on restoration and management of dike grasslands. Fieldwork was supported by Clemens Berger and Uwe Kleber-Lerchbaumer (Wasserwirtschaftsamt Deggendorf). We thank Holger Paetsch, Simon Reith, Anna Ritter, Jakob Strak, Dr. Leonardo H. Teixeira, and Linda Weggler for assisting with the field surveys or soil analyses in 2018--2020. The German Federal Environmental Foundation (DBU) supported MB with a doctoral scholarship. Thank you also to three anonymous referees for their valuable suggestions improving the manuscript.
 
 # Author contribution {.unnumbered}