Restore/create peatland vegetation (multiple interventions)

Supporting evidence from individual studies

1. A replicated, controlled, before-and-after study in 2007–2010 in a degraded blanket bog in England, UK (Anderson et al. 2011) reported that areas restored using multiple interventions developed a different plant community to unrestored areas, and found that they had greater vegetation cover. All areas were initially bare peat. Three years after intervention, the overall plant community composition differed between restored and unrestored areas (data reported as a graphical analysis; difference not tested for statistical significance). Restored areas had developed greater cover than an unrestored area of total vegetation (60–88% vs 15%), mosses (13–25% vs 1%) and heather Calluna vulgaris (2–25% vs 1%). Heather cover was particularly high in plots covered with heather brash. Note that most of the vegetation cover in restored areas was the nurse grass (33–47% cover). In winter 2007/2008, four areas (bare gully sides) were restored by sowing grass seed as a nurse crop (41 kg/ha, mix of six species), fertilization (nitrogen-phosphorous-potassium, 250 kg/ha), liming (1 t/ha) and gully blocking (with stone or heather bales). Two areas were also covered with heather brash (including heather seeds) and one covered in geojute matting. One additional area was not restored (received no intervention). Before monitoring began, sheep were excluded from the entire bog. Vegetation cover was estimated before (summer 2007) and after (summer 2010) restoration, in thirty 2 x 2 m quadrats/area.

   Study and other actions tested

   Referenced paper

   Anderson P., Worrall P., Ross S., Hammond G. & Keen A. (2011) United Utilities Sustainable Catchment Management Programme Volume 3: The Restoration of Highly Degraded Blanket Bog. Penny Anderson Associates report. Penny Anderson Associates Report.

2. A study in 1999–2007 in a historically mined bog in the Czech Republic (Horn 2012) reported that following multiple restoration interventions, bare peat was coloinized by vegetation including Sphagnum moss, cottongrasses Eriophorum spp. and beaked sedge Carex rostrata. These results were not tested for statistical significance. Of the bare peat present one year after restoration began, 30% was covered by vegetation seven years later. Over this time, Sphagnum cover increased from <2% to 8%. Cover also increased of common cottongrass Eriophorum angustifolium, sheathed cottongrass Eriophorum vaginatum and beaked sedge (data reported as maps). Between 1999 and 2004, a historically mined bog (with some remnant vegetation in drainage ditches) was subjected to multiple restoration interventions: rewetting by blocking drainage ditches, excavating shallow 10 x 10 m basins, planting cottongrasses and beaked sedge, sowing Sphagnum moss, mulching (both sown and unsown areas) with sedge cuttings, removing trees and stabilizing the peat surface with tree trunks. In 2000 and 2007, maps were made of where each plant species was dominant (or co-dominant).

   Study and other actions tested

   Referenced paper

   Horn P. (2012) Ecology of peat bogs in Šumava. PhD Thesis. University of South Bohemia.

3. A study in 2003–2008 in a historically mined raised bog in France (Pôle-Relais Tourbières 2013) reported that following restoration by multiple interventions, vegetation cover increased. No statistical tests were carried out. When intervention began, total vegetation cover was 14%. Five years later, it had increased to 51%. Restoration of 0.2 ha of drained, nutrient-enriched bog began in 2003. The top 20 cm of peat were removed, leaving an uneven surface. The stripped peat was used to build embankments and block a drainage ditch, rewetting the area. It was then planted (with sheathed cottongrass Eriophorum vaginatum, common cottongrass Eriophorum angustifolium and Sphagnum mosses) and mulched with straw. Annually between 2003 and 2008, vegetation cover was estimated in six 1 m² quadrats across the restored area.

   Study and other actions tested

   Referenced paper

   Pôle-Relais Tourbières (2013) Restauration des milieux tourbeux: Essais de techniques de restauration des casiers d’exploitation de la tourbières “sur les Seignes” à Frambouhans. French Peatland Coordination Centre report.

4. A replicated before-and-after study in 2008–2011 in a tree-colonized peatland in Scotland, UK (Kirkland 2014) reported that plots restored using multiple interventions developed herb and bryophyte cover. After 2–3 years, restored plots had 42% cover of rush Juncus spp., 21% cover of bryophytes, 10% cover of devil’s bit scabious Succisa pratensis and 20% cover of other vegetation (including heather Calluna vulgaris, grasses, sedges and other herbs). Cover of devil’s bit scabious did not significantly differ between grazed and ungrazed plots (data not reported). In October 2008, eight 16 m² plots were restored by cutting and removing all conifer trees and sowing seeds of devil’s bit scabious. In four plots, conifer brash was burned after tree removal. Four plots were fenced to exclude deer. In August 2011, vegetation cover was visually estimated in five random 2 x 2 m quadrats/plot.

   Study and other actions tested

   Referenced paper

   Kirkland P. (2014) Experimental marsh fritillary habitat restoration project Avongovie, Islay. NatureScot report, no. 544.

5. A replicated, controlled, site comparison study in 2012–2014 in a historically mined bog in Estonia (Karofeld et al. 2016) found that restoration by multiple interventions increased cover of bryophytes and vascular plants, and created a plant community more like the natural donor bog. After 1–2 years, restored plots had greater cover than an unrestored plot of all bryophytes combined (52–65% vs <1%), Sphagnum mosses (50–54% vs <1%) and vascular plants (17–23% vs 11%). Sheathed cottongrass Eriophorum vaginatum and sedge Carex sp. were present in at least one restored plot (cover <1%), but not in the unrestored plot. After two years, the overall plant community in restored plots was 40–67% similar to the unmined donor bog, compared to 21–29% similarity between the unrestored plot and donor bog. In spring 2012, three plots of almost-bare peat were restored by reprofiling (top 20 cm of peat pushed into ridges around the plot), rewetting (blocking a drainage ditch), adding plant fragments (mostly Sphagnum mosses) from the surface of a nearby bog and mulching with straw. One adjacent plot received no intervention. In June and September 2013 and 2014, vegetation cover was estimated in ten 50 x 50 cm quadrats in each plot and the donor bog.

   Study and other actions tested

   Referenced paper

   Karofeld E., Müür M. & Vellak K. (2016) Factors affecting re-vegetation dynamics of experimentally restored extracted peatland in Estonia. Environmental Science and Pollution Research, 23, 13706-13717.

6. A controlled study in 1997–2014 in a degraded fen in France (Pôle-Relais Tourbières 2016) reported that following restoration by multiple interventions, plant species richness and cover of fen-characteristic plants increased. No statistical tests were carried out. Fifteen years after intervention, there were 20 plant species in the fen (vs 12 when intervention began). Cover of fen-characteristic plants, including sedges Carex spp., was 50% (vs 2% when intervention began and vs 12% in an unrestored area of the fen). Restoration of a drained, abandoned, overgrown fen began in 1997–1999. The fen was rewetted by remeandering an adjacent river. Willow shrubs were cut and removed. Existing herbaceous vegetation and surface peat were shredded. Each summer between 1999 and 2004, the fen was grazed by horses. From 2004, a four year cycle of grazing-rest-mowing-rest was implemented in a mosaic across the fen. Between 1999 and 2014, cover of plant community types was estimated in the managed area (in four quadrats along a transect) and adjacent unmanaged fen (details not reported).

   Study and other actions tested

   Referenced paper

   Pôle-Relais Tourbières (2016) La gestion par fauche des milieux humides de la Réser ve naturelle nationale du lac de Remoray. French Peatland Coordination Centre report.

7. A site comparison study in 2008–2014 in a historically mined bog in Quebec, Canada (Rochefort et al. 2016) reported that an area restored using multiple interventions developed a different plant community to, with less vegetation cover than, nearby natural fens. These results were not tested for statistical significance. Note that the aim of this study was to create a fen, as the post-mining peat chemistry was more like a fen than a bog. Five years after intervention, the overall plant community composition of the restored area was different from three nearby natural fens (data reported as a graphical analysis). In the restored area, Sphagnum moss cover was only 3% (vs 15–25% in natural fens) and other moss cover only 3% (vs 12–55%). Vascular plant cover was only 26% (vs 59–86%), although the dominant species were similar in the restored fen and its donor fen. In winter 2009/2010, part of a historically mined bog (abandoned for nine years) was restored by rewetting (blocking drainage ditches with peat), excavating peat basins (removing surface peat and building embankments), sowing vegetation fragments from a moss-dominated donor fen, and mulching with straw. Vegetation cover was estimated in 2008 (donor fen; in 16 quadrats along a transect) or 2014 (restored area: in five 25 m² plots).

   Study and other actions tested

   Referenced paper

   Rochefort L., LeBlanc M.-C., Bérubé V., Hugron S., Boudreau S. & Pouliot R. (2016) Reintroduction of fen plant communities on a degraded minerotrophic peatland. Canadian Journal of Botany, 94, 1041-1051.

8. A replicated, paired, controlled, before-and-after, site comparison study in five historically mined peatlands in Canada (Strack et al. 2016) found that restoration by multiple interventions increased cover of mosses, grass-like plants and vascular plants, but not shrubs. Restored and unrestored plots were initially bare peat. After 1–15 years, restored sites had significantly greater cover than unrestored sites of mosses (38 vs 3%), grass-like plants (22 vs 5%) and total vascular plants (33 vs 11%), but there was no significant difference in shrub cover (9 vs 3%). Relative to natural, undisturbed sites, restored sites had lower cover of mosses (38 vs 77%), shrubs (9 vs 27%) and total vascular plants (33 vs 44%), but higher cover of grass-like plants (22 vs 3%). Five degraded peatlands were restored (dates unclear) using a mixture of techniques. All received fresh vegetation fragments from the surface of natural peatlands and were mulched with straw. Some sites were levelled, rewetted and/or fertilized. Summer vegetation cover was estimated in restored sites after 1–15 years. For each restored site, vegetation cover of a natural peatland was estimated either before restoration or in separate untreated areas after restoration.

   Study and other actions tested

   Referenced paper

   Strack M., Cagampan J., Hassanpour Fard G., Keith A.M., Nugent K., Rankin T., Robinson C. & Strachan I.B. (2016) Controls on plot-scale growing season CO₂ and CH₄ fluxes in restored peatlands: do they differ from unrestored and natural sites? Mires and Peat, 17, Article-5.