Refine
Document Type
- Article (2)
Language
- English (2)
Has Fulltext
- yes (2)
Is part of the Bibliography
- no (2)
Keywords
- target species (2) (remove)
In Central Europe stands of the Koelerion glaucae vegetation complex are threatened and mostly highly fragmented. Knowledge about the impact of abiotic and biotic measures to restore this vegetation complex is crucial. Therefore, an inland sand dune complex (around 2 ha) was created in 2009 as a study model in the Upper Rhine Valley (Germany), which included sites with different substrate conditions as well as grazing impact and minimized inoculation with plant material.
The restoration area is divided into two halves with different substrate conditions (sites 1 and 2), on which inoculation with raked plant material and grazing by donkeys was studied on systematically arranged plots with randomised treatment distribution (32 plots). Additionally the whole area was monitored by a grid-plot approach to show the floristic background (43 plots). Minimized inoculation was conducted with rare Koelerion glaucae plant material in small plots covering around 5–7% of the restoration sites. During the four-year study, vegetation development was recorded and examined in relation to the donor site and an older restoration site. Soil seed bank and seed rain in the newly deposited restoration sites were also investigated, as well as the endozoochorous seed-dispersal by donkeys. Target species ratios (TSR) were calculated to estimate the restoration success. We used mixed linear models and detrended correspondence analysis for data evaluation.
Substrate conditions had an impact on the number of target species and on phanerogam and cryptogam cover. Inoculation enhanced both number and, without grazing, cover of target species since the first year. On not-inoculated plots and on grid-plots, target-species numbers increased gradually. Grazing by donkeys did not affect target-species numbers, but had a decreasing effect on target-species cover. Grazing reduced bryophyte cover, especially on inoculated plots. DCA revealed development of the experimental plots towards the donor site, as has occurred on the older restoration site. Soil seed bank and seed rain were characterized by ruderal species, and did not show similarities to the donor site. Endozoochory revealed some target species to be effectively dispersed by donkeys.
Minimized inoculation is suitable to overcome seed limitation and build up starter populations of target species for the colonization of larger restoration sites. However, within four years species composition of the donor site was not achieved. Grazing by donkeys had mainly structural effects for the studied time period.
Inland sand vegetation, in our case steppic sandy grassland on base-rich soils, is highly endangered in Europe and therefore in the focus of restoration ecology. While there are studies which deal with short-term restoration success, results for an extended time are rare. We were able to analyse the success of a three-step restoration measure for 10 years.
The experiment was established on an exarable field in the Upper Rhine valley, Hesse, Germany. The three-step restoration approach comprised 1) abiotic restoration by deep-sand deposition, 2) inoculation with raked/mown plant material from two different donor sites with well-developed Koelerion glaucae/Allio-Stipetum vegetation and 3) low-intensity grazing by donkeys. The vegetation of the restoration and donor sites (also serving as reference sites to assess restoration success) was sampled on six permanent plots, respectively. Data analyses included ordination, classification and target-species ratios (TSR: relation of target species to all species).
Detrended correspondence analysis revealed a continuous succession of the restoration plots towards the corresponding reference plots: open soil decreased, ruderal species declined and target species increased. While speed of succession decreased, there was still a further improvement in the tenth year. The qualitative TSR (number of target species) reached a plateau after the sixth year with values only slightly lower than at the reference sites. The quantitative TSR (cover of target species) showed a steady improvement and even excelled one reference site. Koelerion glaucae species were present with constancy 17–67% since the 3rd year, with 33–100% since the 7th year. It does not completely resemble either reference site due to a mixture of propagules of both donor sites (e.g. by wind and donkeys) and input from the surroundings. Already in the first year, three Red-list species established themselves, since the 8th year 23 Red-list/near-threatened species have been present. Some ruderal species colonised the restoration site and occurred permanently.
Additionally, we studied the establishment of the highly threatened species Bassia laniflora after inoculation for 6–12 years on three further plots adjacent to the other ones. One of these plots was located on a former sandy field without abiotic restoration; two other plots represented typical Koelerion glaucae vegetation on a newer deep-sand deposition. Bassia laniflora established itself well on all plots. We conclude that restoration of steppic sandy grassland, including highly threatened species, was not only permanently, but increasingly successful over a time span of 10 years. Management by grazing, however, will remain essential to suppress ruderalisation.