Semi-natural grasslands are among the most endangered vegetation types in Europe,
threatened by inadequate management, habitat fragmentation and excessive nitrogen
deposition. As a consequence, many of them have received a protection status as nature
reserves. Yet this protection applies almost exclusively to large contiguous areas whereas
small grasslands often do not receive the right protection status they deserve. This is especially
the case for the many historical unimproved grasslands which have now been reduced to road
verges and unimproved grassland margins and which often still contain relict populations of
valuable plant species. Given their small size and increased isolation, such populations have
become relict populations with expected extinctions as a result. However, these grassland
relicts are important for preserving high species diversity and ecosystem dynamics at the
landscape-scale, for instance as corridors and stepping stones or as microhabitats.
Vegetations at the fringes of the pastures, mainly under the barbed wire
(“prikkeldraadvegetaties”), are a specific example of such grassland relicts.
As a result of agricultural intensification, wooded edges were replaced by barbed wire as
enclosures. Although intensified grazing threatened the unique grassland vegetation through
trampling and defaecation, cattle acted as the desired nature managers along the fringes. By
grazing and trampling the border of the fringes, they create a high-quality microhabitat under
the barbed wire. Moreover, the microrelief, consisting of mounds and lynchets, allows the
small-scale biodiversity under the wire to increase.
Even though the concept of barbed wire vegetation is well established in botanical societies,
little is known about these vegetations in terms of their species richness and the drivers of their
plant species composition. Nevertheless, they have been shown to harbour higher plant
richness than adjacent pastures, which could make them a species resource pool for future
conservation projects. In this study, we want to investigate barbed wire fringe vegetations by
conducting a descriptive analysis (RQ1 + RQ2) and determining the environmental conditions
for high-quality barbed wire vegetations (RQ3 + RQ4), followed by establishing guidelines for
conservation and management of the fringes (RQ5).
In June 2021, we inventoried 40 fringe vegetations in the Hageland-Zuiderkempen region
(Belgium). We mapped the vegetation cover of plants in a 1 m * 2 m plot. We identified the
environmental parameters: ‘presence of mounds’, ‘presence of lynchet’, ‘presence of an
adjacent historical permanent grassland’, ‘grazing regime’ and ‘adjacent land use’. In addition,
we collected soil samples to analyse pH, organic matter, moisture, and Olsen-phosphorus. We
calculated land use intensity (LUI) of the fringe combining the LUI of the immediate
surroundings, the vegetation of the pasture and the grazing regime. Our dataset was
supplemented with Ellenberg-values for light, nitrogen, and mowing-tolerance as well as the
functional group and conservation status of each species (habitat specialists).
(RQ1 + RQ2) To describe the ecology of fringes, we divided sites into groups based on the
distinctive and indicator species of each fringe. We determined the associated plant community
and alliance of each group and tested in which of the abovementioned environmental
parameters they differed. Using multivariate analysis of variance, we determined which
variables explained the variation in the vegetation dataset. (RQ3) To evaluate which
environmental variables had a significant correlation with the habitat specialist richness and
Shannon index, we performed a full linear regression on the environmental variables followed
by model selection using the Akaike information criterion (AIC) to find the most parsimonious
model. To determine whether the biodiversity indices were highest at an intermediate land use,
we built a quadratic model using total LUI and its quadratic term as explanatory variables.
(RQ4) To uncover the specific mechanisms by which fringe dynamics improve the habitat
quality, we determined (1) the nitrogen/phosphorus levels at an intermediate LUI, (2)
differences in Olsen-phosphorus and soil moisture due to the presence of a lynchet, (3) the
effect of grazing on Olsen-phosphorus and habitat specialist richness and (4) the influence of
LUI on the cover of functional groups.
Fringe plant species benefit from the specific conditions under the barbed wire promoted by
intermediate grazing. They can escape from the negative effects of trampling and defaecation.
These plants can be classified into two groups dependent on the history (relict-species) and
abiotic conditions (grazing-avoiding species) of the vegetation. In our study, we first
investigated which plant communities the groups belonged to. We found four different plant
alliances (M1 – 4), three of which could be classified as the Molinio – Arrhenatherata
community, while fringes of group M2 belonged to the Koelerio – Corynepherata community.
Group M2 was considered as the vegetation group with the highest quality, having the highest
habitat specialist species richness, nitrogen-poorest conditions and intermediate LUI. The
fringes of M2 were located on sandy soils, typical of the Koelerio – Corynepherata community,
explaining lower nitrogen levels and pH values. Overall, variation in the vegetation could be
explained by all the environmental characteristics except the presence of an adjacent historical
permanent grassland. Soil pH was the most significant explanatory variable, but we attribute
the differences in soil type to geographical location.
We were able to establish that a fringe with an intermediate LUI of 4 (on a scale of 0 to 6), dry
soil and the presence of a lynchet was optimal. At an intermediate LUI, plant species richness
was highest and most nitrogen-avoiding species were present. An intermediate LUI is the
perfect balance between vegetation encroachment and overgrazing. The presence of a
lynchet could cause a run-off of phosphorus, although Olsen-P-levels were still too high for all
fringes except one. High phosphorus levels may lead to future habitat degradation and
extinction of vulnerable species, although some species may still persist for an unexpectedly
long time. Soil pH was significantly negatively correlated with species richness and Shannon
diversity, but again, geographical location was likely a confounding factor. Furthermore, we
expected that the soil maturity and the presence of ant mounds of Lasius flavus were important
as well, although we could not demonstrate their effect. High-quality barbed wire fringes can
harbor a rich biodiversity of remarkable plants. A total 37 habitat specialist were found in our
study, but 24 of which have disappeared over the past 30 years. The main reason for their
decline is believed to be the negative effects of small populations and excessive nitrogen and
phosphorus deposition.
Barbed wire fringe vegetations can substantially contribute to the landscape-scale biodiversity
by both increasing the available habitat and their connection between habitats, and as ex-situ
back-up of remnant rare plant populations. Although neglected in most conservation projects,
these vegetations are easy and inexpensive to conserve. Only an enclosure, which encloses
the livestock and allows their grazing underneath, is required. Possible management goals are
1) grassland restoration of the adjacent barbed wire fringe vegetation to allow grassland relict
species of the fringe to expand their territory and 2) preservation of their seeds for future
reintroductions. However, seed preservation is only possible if the plant populations are still
viable. An increase of fringes is not only beneficial for the biodiversity. Indeed, fringes produce
healthier food for the livestock, resulting in higher quality of dairy products and meat. This, in
turn, leads to healthier consumers. Conserving and establishing new barbed wire fringe
vegetations can therefore be important both for the welfare of animals and their consumers,
and for biodiversity in the agricultural landscape.