Aspects of Applied Biology 81, 2007
Delivering Arable Biodiversity
Improved field margins for a higher biodiversity in agricultural
By K JACOT
, L EGGENSCHWILER
, X JUNGE
, H LUKA
and A BOSSHARD
Agroscope Reckenholz- Tänikon Research Station ART, 8046 Zurich, Switzerland
Research Institute of Organic Agriculture (FiBL), 5070 Frick, Switzerland
Institute of Environmental Sciences, University of Zurich, 8057 Zurich, Switzerland
Species-rich seed mixtures were developed and tested to establish
improved field margins (IFM) which increase floral and faunal diversity,
effects of IFM on bio indicators were studied and the public’s valuation of
IFM was evaluated. The use of species rich seed mixtures with regional
ecotypes was able to re-introduce rare plant species that formerly have been
typical in crop-dominated landscapes in Switzerland. Already one year after
sowing, IFM attracted many times more butterflies and grasshoppers than
the conventional margins, and more than most other ecological
compensation area types in the area. Compared with sown wildflower strips
and conventional field margins, IFM had an intermediate position for
ground beetles and spiders. Overall, farmers and non-farmers valued the
establishment of IFM positively, mainly because they appreciated the
increased biodiversity in the cultivated landscape.
Key words: ecological compensation, target plant species, spiders,
ground beetles, grasshoppers, butterflies, acceptance
In order to counteract the decline of biodiversity and promote more sustainable farming
practices in Switzerland, the development and promotion of Ecological Compensation Areas
(ECA) became important (Bundesamt für Landwirtschaft, 1998). Until now, there have been
ECA such as extensively managed meadows and sown fallows. However, there is a lack of a
permanent ECA in arable land. Field boundaries play an outstanding role for biodiversity in
agricultural, crop dominated landscapes, as examples from eastern and southern Europe show
(Knop & Reif, 1982). In contrast, field boundaries along crop fields in Switzerland are narrow,
normally mown or mulched three to five times a year, often affected by herbicide, and
accordingly harbour a low biodiversity (Theato, 2001).
To optimize the ecological compensation program of Switzerland, we aimed to establish the
new ECA-Type "Improved Field Margins" (IFM). Species-rich seed mixtures were tested, the
effects of IFM on the abundance of selected bioindicators were studied and the farmer and non-
farmer’s perception and valuation of IFM was evaluated. The seed mixtures were designed to
establish a long-lasting, floristically diverse and flower-rich vegetation, which provides multiple
benefits to the typical fauna of crop-dominated landscapes, such as shelter, food, or suitable
microclimate (Phillippi, 1971; Klotz & Kock, 1986; Knop & Reif , 1982). Also the IFM should
improve the aesthetical value of the landscape.
Materials and Methods
Study area and target system
The on-farm experiment is located in the Midland zone in Switzerland in nine regions
(Klettgau, canton Schaffhausen; Zurich; Wauwil, canton Lucerne; Aesch, canton Baselland;
Nyon, canton Waadt; Grosses Moos, canton Freiburg; Rheintal, canton St. Gallen, Litzibuch and
Fricktal, canton Aargau). A temperate climate dominates in the Midland with a mean annual
temperature of about 10°C and annual precipitation average of 600-1050 mm. Soils in the
Midland are dominated by luvisols with rendzina and gley soil in places.
In 2001 and 2003, a total of 70 IFM of 5 m x 120 m were established on farms. Each strip
consisted of 3 block-wise replicates containing 5 m x 10 m plots. The field margins were sown
with different project-developed seed mixtures containing up to 38 annual and perennial grass
and wildflower species for moist, fresh and dry conditions (Bosshard et al., in prep). The seed
mixture composition was based on data from existing field margins (Theato, 2001) and
additionally on plant sociological literature (Phillippi, 1971; Klotz & Kock, 1986; Knop & Reif,
1982). Except for some grass species, only seeds from Swiss regional ecotype quality, available
on the market, were used. Half of each strip was mown lengthwise and alternately each year in
the second half of August. The air-dried biomass was removed.
We aimed to develop field margins with at least 20 target plant species per 20 m
. An IFM-
target plant species fulfilled at least three of the following four requirements: (1) (at least
formerly) regularly distributed within the Swiss Midlands, (2) typical for extensively used field
and wood margins, (3) rarely found in intensively cultivated field margins or meadows, (4)
particular aesthetic or valuable for fauna. None of the IFM-target plant species was an
agronomical problematic species.
Vegetation was recorded in the plots of all IFM on 20 m
each year in June (2002 to 2005)
using an adapted Braun-Blanquet (1964) methodology: Each species with more than one
specimen and > 0.1% cover per plot was recorded and the cover estimated within 10 classes.
To assess arthropod diversity (Bosshard et al., in prep.), in one of the project regions (Basel in
northern Switzerland) butterflies (macrolepidoptera) and grasshoppers (excl. tetrigidae) were
determined and counted five times between the end of May and the end of August in each of
two years (2003 and 2005, i.e. 2 and 4 years after establishment of the IFM) along a transect of
5m x 120 m according the methodology of Pollard & Yates (1993). Seven IFM were compared
with ten sown short-lived wildflower strips, eight extensively managed, sown hay meadows
with high plant biodiversity and nine with low plant biodiversity (according to the Swiss
Meadow Biodiversity Scheme), five biodiversity hotspots (authochthone, diverse, extensively
used meadows), and ten conventional field margins (narrow, normally mown or mulched three
to five times a year) as controls. The conventional field margin was the only habitat type that
was not included in an ecological compensation scheme.
Surveys of epigeal arthropods (ground beetles and spiders) were conducted in two of the nine
regions of the study area: in the Klettgau area in canton Schaffhausen, and in Litzibuch in
canton Aargau. In each of the two areas, two IFM were examined, as well as two sown
wildflower strips and two conventional field margins, as reference elements. The selected
bioindicators, ground beetles and spiders, were surveyed by means of pitfall traps during five
weeks between April and July in 2002 and 2004 (Duelli et al., 1990).
To study the acceptance of IFM by farmers and non-farmers, 13 IFM in five cantons in the
Swiss Midlands (cantons of Aargau, Baselland, Lucerne, Schaffhausen and Zurich) were chosen
(Junge et al., in prep). These IFM were all in close vicinity to settlement areas and thus
frequently visited by walkers or cyclists.
In summer 2003 and 2004, the IFM were presented on 29 days to 39 farmers and to 233
randomly selected passers-by (all non-farmers). The farmers were living in the surrounding
municipalities of the IFM and were invited by mail.
With the help of a written questionnaire, all study participants were asked to look at the IFM, to
rate the attractiveness of each IFM by using a six-step rating scale (1 = dislike it very much, 6 =
like it very much), and to explain their decision. Out of the 39 farmers who participated in the
survey, 8 were involved in the present project and had already established one or several IFM.
Plant species richness
Plant species richness (10 to 27 target species per 20 m
) on all site types significantly differed
between the seed mixtures and in most years after establishment independently of site or region
(Tukey HSD, p<0.05, n=21-216). No difference in number of plant species (22 to 26 target
species per 20 m
) in different seed mixtures sown in 2003 were detected in the second year
after establishment. From the third year onwards, the IFM sown in 2001 became more species
rich and the species richness increased from the first to the second year after the establishment
in 2003 (p<0.05). For the first series of the experiment (sown in 2001), the threshold of 20 target
species per 20 m
was only surpassed by the mixture for dry conditions four years after
establishment. All mixtures sown in 2003 surpassed the threshold in the second year after
establishment. More than 90% of all recorded species were target plant species.
Grasshoppers and butterflies
In the 49 investigated habitat plots, 47 butterfly species and 15 grasshopper species were
detected. The analysed habitat types differed significantly both for butterflies and grasshoppers,
and both concerning species number and frequency of individuals (ANOVA). The IFM did not
differ significantly from the biodiversity hotspots of the region, but they contained significantly
more species and individuals than the wildflower strips, and 4 (grasshoppers) to 40 times more
species and individuals than the conventional field margins. With 3.0 non-ubiquists and red list-
species on average, IFM were less attractive to butterflies compared with hotspots (4.8 non-
ubiquists and red list-species), but even more attractive than the latter for grasshoppers (5.4 non-
ubiquists in IFM compared with 4.4 in the hotspots). Extensive meadows were positioned
between the hotspots, IFM and the wildflower strips.
Two years after establishment of the IFM, the same species number was present as four years
later, while the frequency increased slightly.
At the 12 sites, a total of 21,000 ground beetles in 93 species and 11,000 spiders in 100 species
were counted (Fig. 1). IFM had an intermediate position compared with the sown wildflower
strips and the conventional field margins.
Fig. 1. Total activity density and total number of species of ground beetles and spiders in two
different regions. IFM: improved field margins (n=8), WS: wildflower strips (n=8), CFM:
conventional field margins (n=8). 02: 2002, 04: 2004.
Table 1. Pair comparisons of the middle activity density and the diversity of species of the
ground beetles and spiders in the improved field margins (IFM) compared with wildflower
strips (WS) and conventional field margins (CFM). Arranged according to group of animals and
investigation year 2002 and 2004. IFM, WS or CFM denotes the habitat with the higher value. *
Significant differences with p<0.05.
Number of species
Ground beetles Spiders
2002 Activity density
2004 Activity density
2002 Number of species
2004 Number of species
2002 Activity density
2004 Activity density
2002 Number of species
2004 Number of species
Group of animals Year Parameter
Comparing the activity density and the diversity of species of the ground beetles and spiders in
the improved field margins (IFM) with wildflower strips (WS) and conventional field margins
(CFM), the wildflower strips exhibited higher activity and diversity in 48% (22 cases) of all 48
cases (Table 1). The IFM had higher activity in 15 cases (31%), 10% more than the
conventional field margins (21%).
Acceptance by farmers and non-farmers
Both farmers and non-farmers liked the improved field margins that were presented to them (on
average 5.1 and 5.2 respectively on the 6-step rating scale).
All study participants were asked in open questions to justify their evaluation of an IFM. The
most frequent reasons among non-farmers for evaluating an IFM positively were that it looked
natural and diverse. Farmers evaluated an IFM positively because of its species richness, flowers
and colours. For more details see Junge et al., in prep.
Plant species richness
The on farm field experiment proved that improved field margins can successfully be
established from a floristic point of view. The data show that the seed mixtures with 60 to 80%
forbs proportion in the seed mixture lead to higher species richness in the plot. The average of
plant species richness for the first run of the experiment is lower than expected. Thus, it is
suitable to raise the forbs component to optimise the success of the seed mixtures. It is known
from a study by Theato (2001) who explored 112 conventional field margins in all parts of the
Swiss Midlands and adjacent regions that conventional field margins in Switzerland contain an
average of 15 species (3 m
) and 26 species (whole margin) respectively. However, the
uppermost number of these field margins did not contain target plant species. With appropriate
seed mixtures, rare plant species (target plant species) that formerly have been typical in crop-
dominated landscapes in Switzerland can be re-introduced. The success of the method is to an
extent independent of site (soil) conditions, and the established vegetation is stable, with the
tendency to increase diversity during the first years after establishment. Several perennial sown
species only germinated and established after a few years so that they may have contributed to
the increase of plant richness over time.
Grasshoppers and butterflies
The IFM provide an attractive habitat for both indicator taxa macrolepidoptera and saltatoria. In
a region with a relatively good ecological infrastructure like the Basel region, the two arthropod
taxa were able to colonise restored IFM in one or two years, reaching both species and
individual numbers similar to those in the regional biodiversity hotspots. Similar observations
were made by Bosshard & Kuster (2001) with the restoration of species and flower rich
meadows. The relatively good flower offer, the botanical composition which is comparable with
that one of extensively used traditional meadows and embankments, and the permanent
vegetation structure of the IFM seemed to be particularly important for the two animal groups
(Sänger, 1977; Erhardt, 1985; Feber et al., 1996).
The impact of semi-natural landscape elements, such as wildflower strips and field margins, on
the species diversity is well-documented (Nentwig, 2000; Kopp, 1998). During the growing
season, wildflower strips represent a habitat for several fallow species as well as for eurytopic
grassland specialists in older strips (Luka et al., 2001; Herzog & Walter, 2005; Pozzi et al.,
2005). On the other hand, in conventional field margins, various stenotopic grassland specialists
find favourable living conditions (Schmidt et al., 2005; Pfiffner & Luka, 1996). Herzog &
Walter (2005) have shown that wildflower strips strongly stimulate the development of ground
beetle and spider populations, which has also been confirmed in this study. However, these
strips alone are not able to promote the stenotopic grassland specialists. Our data show that IFM
provide living conditions for both fallow species and grassland specialists.
IFM provide a valuable complement to sown wildflower strips and meadows. They contribute
to increasing and preserving the species diversity of ground beetles and of epigeal spiders in
Acceptance by farmers and non-farmers
It was shown that both farmers and non-farmers responded positively to IFM. They especially
appreciated the IFM’s diversity. Recently, other studies have shown that the more species-rich a
plant community was, the more it appealed to people (Junge et al., 2004; Lindemann-Matthies et
al., 2003). Study participants also appreciated the IFM’s naturalness. Previous research has
shown that naturalness is a powerful factor for landscape preference (e.g. Kaplan et al., 1972;
Ulrich, 1986). In contrast to the common perception that species-rich semi-natural vegetation
such as wildflower strips in set-aside land may appear untidy and scrubby (Nassauer, 1995;
Hands & Brown, 2002), the results of the present study indicate that non-farmers in Switzerland
would like species-rich and structurally diverse IFM near their places of living. Similarly, the
farmers’ acceptance of IFM was very high, although it might have been a positive selection of
farmers, considering that only interested farmers might have followed the invitation to
participate in the survey.
Because of the comprehensive approach and the positive results, the Swiss federal
administration will introduce IFM as a new type of ecological compensation area for agriculture
in 2008. A good acceptance of IFM is expected by the farmers and may have a negative impact
on the abundance of large-area fallows. These fallows have an essential value e.g. for birds and
hares (Herzog & Walter, 2005). Therefore, it is important to establish both IFM and fallows for
a high diversity on landscape level.
Bosshard A, Kuster D. 2001. Bedeutung neu angelegter Extensivwiesen für Tagfalter und
Heuschrecken. Agrarforschung 8 (7): 252-257.
Braun-Blanquet J. 1964. Pflanzensoziologie: Grundzüge der Vegetationskunde. Verlag
Springer. Wien: 865 pp.
Bundesamt für Landwirtschaft. 1998. Verordnung über die Direktzahlungen an die
Landwirtschaft (Direktzahlungsversordung, DZV). 42 pp.
Duelli M, Studer M, Obrist M, Wirz P. 1990. Minimalprogramme für die Erhebung und
Aufbereitung zooökologischer Daten als Fachbeiträge zu Planungen am Beispiel
ausgewählter Arthropodengruppen. Schriftenreihe für Landespflege und Naturschutz
Erhardt A. 1985. Diurnal Lepidoptera: Sensitive Indicators of cultivated and abandoned
Grassland. Journal of Applied Ecology 22: 849-861.
Feber R E, Smith H, MacDonald D W. 1996. The effects on butterfly abundance of the
management of uncropped edges of arable fields. Journal of Applied Ecology 33: 1191-
Hands D E, Brown R D. 2002. Enhancing visual preference of ecological rehabilitation sites.
Landscape and urban Planning 8:57-70.
Herzog F, Walter T. 2005. Evaluation der Ökomassnahmen, Bereich Biodiversität.
Schriftenreihe der FAL 56:105-114.
Junge X, Lindemann-Matthies P, Matthies D. 2004. People’s perception and appreciation of
plant diversity along waysides. Verhandlungen der Gesellschaft für Ökologie 34: 420.
Kaplan S, Kaplan R, Wendt J. 1972. Rated preference and complexity for natural and urban
visual material. Perception and Psychophysics 12:354-356.
Klotz S, Kock U V. 1986. Comparative geobotanical studies in the Bashkir ASSR. 4. Meadow
and border communities. Vergleichende geobotanische Untersuchungen in der
Baschkirischen ASSR. 4. Wiesen- und Saumgesellschaften. Feddes-Repert. Berlin, E. Ger. :
Knop C, Reif A. 1982. Die Vegetation auf Feldrainen Nordost- und Ostbayerns – natürliche
und anthropogene Einflüsse, Schutzwürdigkeit. Berichte der ANL 6:254-279.
Kopp A M. 1998. Ackerrandstreifen als Lebensraum für Laufkäfer und deren Einfluss auf
Getreideblattläuse. Agrarökologie 28:1-159.
Lindemann-Matthies P, Junge X, Matthies D. 2003. An experimental study on people’s
perception and appreciation of plant diversity. Verhandlungen der Gesellschaft für Ökologie
Luka H, Lutz M, Blick T, Pfiffner L. 2001. Einfluss von eingesäten Wildblumenstreifen auf
die epigäischen Laufkäfer und Spinnen (Carabidae & Araneae) in der intensiv genutzten
Agrarlandschaft "Grosses Moos". Schweizerische Peckiana 1:45-60.
Nassauer J I. 1995. “Messy ecosystems, orderly frames.” Landscape Journal 14: 161-170.
Nentwig W. 2000. Streifenförmige ökologische Ausgleichsflächen in der Kulturlandschaft:
Ackerkrautstreifen, Buntbrache, Feldränder. Agrarökologie:293 pp.
Pfiffner L, Luka H. 1996. Förderung der Artenvielfalt von Laufkäfern und Spinnen im
Agrarland. Ökologie & Landbau 99:49.
Phillippi G. 1971. Sand community, steppe swards and border associations of Schwetzinger
Hardt (Rhine Plain, northern Baden) with special consideration to the nature reserve at
Sandhausen. Landesstelle Naturschutz Landschaftspflege Veröffentlichungen 39:67-130.
Pollard E, Yates T J. 1993. Monitoring Butterflies for Ecology and Conservation. Chapman &
Pozzi S, Jeanneret P, Luka H, Wermeille E. 2005. Evaluation des mesures de compensation
écologique de la région de Nuvilly/Combremont. Revue Suisse Agriculture. 37 (6): 1-12.
Sänger J. 1977. Über die Beziehungen zwischen Heuschrecken und der Raumstruktur
ihrer Habitate. Zoologisches Jahrbuch für Systematik 104: 433-488.
Schmidt M H, Roschewitz I, Thies C, Tscharntke T. 2005. Differential effects of landscape
and management on diversity and density of ground-dwelling farmland spiders. Journal of
Applied Ecology 42:281-287.
Theato C. 2001. Säume als ökologische Ausgleichsflächen. Diplomarbeit, ETH Zürich.
Ulrich R S. 1986. Human responses to vegetation and landscapes. Landscape and Urban