Page 1
Contents lists available at
ScienceDirect
Agriculture, Ecosystems and Environment
journal homepage:
www.elsevier.com/locate/agee
Use of organic inputs by arable farmers in six agro-ecological zones across
Europe: Drivers and barriers
R. Hijbeek
a
,
⁎
, A.A. Pronk
b
, M.K. van Ittersum
a
, A. Verhagen
b
, G. Ruysschaert
c
, J. Bijttebier
c
,
L. Zavattaro
d
, L. Bechini
e
, N. Schlatter
f
, H.F.M. ten Berge
b
a
Plant Production Systems, Wageningen Plant Research, Wageningen University and Research, the Netherlands
b
Agrosystems Research, Wageningen University and Research, the Netherlands
c
Research Institute for Agriculture, Fisheries and Food (ILVO), Belgium
d
Department of Agricultural Forest and Food Sciences, Università degli Studi di Torino, Italy
e
Department of Agricultural and Environmental Sciences, Università degli Studi di Milano, Italy
f
Department for Soil Health and Plant Nutrition, Austrian Agency for Health and Food Safety, Austria
A R T I C L E I N F O
Keywords:
Drivers
Barriers
Manure
Straw
Compost
Europe
A B S T R A C T
Soil organic matter (SOM) in agricultural soils builds up
via
–
among others - the use of organic inputs such as
straw, compost, farmyard manure or the cultivation of green manures or cover crops. SOM has bene
fi
ts for long-
term soil fertility and can provide ecosystem services. Farmer behaviour is however known to be motivated by a
larger number of factors. Using the theory of planned behaviour, we aimed to disentangle these factors. We
addressed the following research question: What are currently the main drivers and barriers for arable farmers in
Europe to use organic inputs?
Our study focuses on six agro-ecological zones in four European countries (Austria, Flanders [Belgium], Italy
and the Netherlands) and four practices (straw incorporation, green manure or cover crops, compost and
farmyard manure). In a
fi
rst step, relevant factors were identi
fi
ed for each practice with farmers using 5 to ten
semi-structured interviews per agro-ecological zone. In a second step, the relevance of these factors was
quanti
fi
ed and they were classi
fi
ed as either drivers or barriers in a large scale farm survey with 1263 farmers.
In the semi-structured interviews, 110 factors that in
fl
uenced farmer decisions to use an organic input were
identi
fi
ed. In the larger farm survey, 60% of the factors included were evaluated as drivers, while 40% were
evaluated as barriers for the use of organic inputs. Major drivers to use organic inputs were related to the
perceived e
ff
ects on soil quality (such as improved soil structure or reduced erosion) and the positive in
fl
uence
from social referents (such as fellow farmers or agricultural advisors). Major barriers to use organic inputs were
fi
nancial (increased costs or foregone income) and perceived e
ff
ects on crop protection (such as increased weeds,
pests and diseases, or increased pesticide use).
Our study shows that motivating farmers to use organic inputs requires speci
fi
c guidance on how to adapt
cultivation practices to reduce weeds, pests and diseases for speci
fi
c soil types, weather conditions, and crops. In
addition, more research is needed on the long-term
fi
nancial consequences of using organic inputs.
1. Introduction
Using organic inputs, such as straw, green manures, compost or
animal manure, contributes to the soil organic matter (SOM) content of
a soil (
Panagos et al., 2015
). SOM is generally considered an important
indicator for soil fertility (
Christensen and Johnston, 1997
;
Reeves,
1997
) as it is known to promote soil aggregation, stabilize soil struc-
ture, increase nutrient availability and improve water holding capacity
(
Johnston et al., 2009
;
Watts and Dexter, 1997
). In addition, SOM
content can have environmental bene
fi
ts such as supporting soil bio-
diversity (
Chang et al., 2007
), sequestering carbon (
Freibauer et al.,
2004
) or immobilizing toxic organic pollutants (
Bollag et al., 1992
).
Recently, a decline in SOM has been identi
fi
ed as a threat to soil
quality on a European scale (
Toth et al., 2008
;
Stolte et al., 2015
).
Precise magnitudes of decline are unknown because monitoring of SOM
dynamics is currently rarely done (
ten Berge et al., 2017
) or gives
https://doi.org/10.1016/j.agee.2019.01.008
Received 6 August 2018; Received in revised form 17 January 2019; Accepted 18 January 2019
Abbreviations:
AEZ, agro-ecological zone; FYM, farmyard manure; SOM, soil organic matter; CT, conservation tillage; GM, green manure; SI, Supporting Information
⁎
Corresponding author.
E-mail address:
renske.hijbeek@wur.nl
(R. Hijbeek).
Agriculture, Ecosystems and Environment 275 (2019) 42–53
0167-8809/ © 2019 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/BY-NC-ND/4.0/).
Page 2
contrasting results. For Great Britain, studies suggest either a decline or
no changes in SOM content over the past decades (
Bellamy et al., 2005
;
Chapman et al., 2013
;
Reynolds et al., 2013
). For Austria and the
Netherlands, positive changes in SOM content have been reported
(
Dersch, 2015
;
Reijneveld et al., 2009
). In contrast, for Germany, Italy,
Belgium, Finland and Denmark, a decline in SOM content of arable soils
over the last decades has been reported (
Capriel, 2013
;
Fantappiè et al.,
2011
;
Goidts and van Wesemael, 2007
;
Heikkinen et al., 2013
;
Taghizadeh-Toosi et al., 2014
).
Currently, maintenance of SOM content is included in the cross-
compliance standards of Good Agricultural and Environmental
Conditions proposed by the EU Council (GAEC;
EC, 2013
). Most
member states have only agreed to a ban on burning of stubble, but a
small number of countries has added additional measures including the
use of solid manures or other solid organic inputs such as compost
(
Frelih-Larsen et al., 2016
).
Frelih-Larsen et al. (2016)
recommend a
strengthening of GAEC by adding a requirement to incorporate crop
residues, either directly or following mulching by composting or use as
animal bedding. This would be in line with the 4/1000 initiative to
increase SOM as a climate change mitigation measure (
UNFCCC, 2015
),
which would also require increasing the amount of organic inputs to
agricultural soils. Before enforcing such a measure, it is important to
fi
rst understand the current drivers and barriers experienced by Eur-
opean farmers with regards their use of organic inputs.
Van den Putte et al. (2010)
argue that farmers in Europe often have
few incentives to use soil conservation measures because the productive
capacity of their farms is often not a
ff
ected by soil degradation in the
short term.
Glenk et al. (2017)
made an economic farm level analysis of
SOM management in two areas of Scotland and Spain and found both
negative and positive
fi
nancial consequences, depending on input type
and region. They did however note that farmer behaviour may be
motivated by factors which are not directly economic such as perceived
workability of the soil, or soil health for future generations. Simple cost-
bene
fi
t models might hence not capture the complexity of farmer be-
haviour and attitudes (
Lynne et al., 1988
).
Burton (2004)
argues that
understanding farmer willingness to adopt certain management prac-
tices improves when using the theory of planned behaviour (
Ajzen,
1991
) as a conceptual framework.
The theory of planned behaviour has recently been successfully
applied in an agricultural context (see for example
Werner et al., 2017
;
Bijttebier et al., 2018
). In this study, the theory of planned behaviour is
used to identify drivers and barriers for the use of organic inputs by
farmers. The aim of this research is: 1) to understand which outcomes of
using organic inputs are considered positive or negative by European
farmers; 2) to investigate whether the subjective norm linked to using
organic inputs is positive or negative across Europe; and 3) to identify
which factors currently stimulate or prevent the utilization of organic
inputs on European farms. We made the assumption that speci
fi
c dri-
vers and barriers might be context-speci
fi
c, depending among other
things on soil type, climate, farm type and the socio-economical con-
text. Drivers and barriers were therefore assessed separately in six re-
gions with di
ff
erent soil types and climates, hereafter called agro-eco-
logical zones (AEZ).
2. Materials and methods
2.1. Study area
This study formed part of the European project CATCH-C in which
AEZs in Europe were identi
fi
ed based on three environmental factors:
climate, slope and soil texture. Each AEZ has more or less homogeneous
conditions for each factor (
Hijbeek et al., 2013
). Six AEZs were studied
in four countries (Austria, Belgium [Flanders], Italy and the Nether-
lands). Selection of AEZs in each country was mainly based on the area
of the AEZ and the economic importance of agriculture in the AEZ. Six
AEZs were selected (
Fig. 1
and
Table 1
): Austria (AT1), Belgium (BE1),
Italy (IT1, IT2) and the Netherlands (NL1, NL2). Within each AEZ, we
focussed the data collection on arable farmers, because SOM contents in
arable farms are often lower than in grasslands and therefore farmers
are more likely to bene
fi
t from using organic inputs.
In Austria, AT1 has a relatively dry climate (mean aridity index of
0.51), level or gentle slopes and a medium soil texture (meaning > 15%
sand and 18
–
35% clay or 15
–
65% sand and < 18% clay). In Belgium,
BE1 has a relatively wet climate (mean aridity index 0.73), a medium
fi
ne soil texture (meaning < 15% sand, < 35% clay) and level to gently
sloping lands.
In Italy, IT1 and IT2 both have a very dry climate. IT1 is mainly
present in the Po plain and has more level land. IT2 is mostly located in
the hills of centre and southern Italy and has steeper slopes. To dif-
ferentiate this di
ff
erence, in
fi
gures and tables IT1 is labelled IT1-level
and IT2 is labelled IT2-sloping. To keep the text
fl
owing, this adjective
is not included in the text.
In the Netherlands, NL1 and NL2 both have a wet climate and level
land. NL1 however has medium or medium
fi
ne soil textures and NL2
has only coarse soil textures (meaning > 65% sand and < 18% clay).
To di
ff
erentiate this variety in soil texture between the two AEZs in the
Netherlands, NL1 was labelled NL1-clay and NL2 is labelled NL2-sand
in
fi
gures and tables.
2.2. Study approach
To identify drivers and barriers, we used the theory of planned
behaviour. According to the theory of planned behaviour (
Ajzen, 1991
),
people base their behaviour on three main constructs: 1) their attitude,
2) their subjective (social) norms and 3) the degree of perceived be-
havioural control. In this framework, an attitude refers to the degree to
which a person expects a certain outcome, together with the desirability
of that outcome. An outcome is an expected result or impact of the
practice, for example increased weed pressure or improved soil fertility.
Subjective norm refers to the in
fl
uence from social referents to perform
a behaviour. Referents are a social in
fl
uence, such as a fellow farmer or
an agricultural advisor. Perceived behaviour control refers to perceived
support or hindrance from control factors. A control factor is a speci
fi
c,
often local, condition that governs the impact of the practice, or facil-
itates or hampers its adoption. These three constructs together lead to
an intention, which might lead to a certain behaviour (
Ajzen, 1991
).
Farmers were questioned about four types of inputs: incorporation
of small grain cereal straw or grain maize straw; cultivation of green
manures or cover crops; application of compost; and application of
farmyard manure (FYM). Not all practices were included in the ques-
tionnaire of each AEZ. Incorporation of straw was included in
fi
ve AEZs.
Cultivation of green manures or cover crops was included in six AEZs.
Application of compost was included in two AEZs. Application of FYM
was included in one AEZ (
Table 2
).
Our study approach consisted of two steps: 1) a qualitative step to
identify relevant outcomes, referents and control factors for each
practice; 2) a quantitative step to quantify each relevant outcome, re-
ferent or control factor. For the
fi
rst step, semi-structured interviews
were conducted with a smaller number of farmers, whilst for the second
step, a much larger farm survey was conducted. In this manner, we used
a bottom-up approach by using farmer knowledge on their local
farming conditions in each region.
2.3. Step 1: semi-structured interviews
In the
fi
rst step, 5
–
10 farmers were interviewed in each AEZ in order
to identify outcomes, referents and control factors for each practice.
Farmers selected for the semi-structured interviews were located in the
speci
fi
c AEZ and had the desired slope and soil type (
Fig. 1
and
Table 1
). The objective of these semi-structured interviews was to use
farmer knowledge to identify relevant factors for each practice, based
on the theory of planned behaviour. Farmers were asked
fi
ve questions
R. Hijbeek et al.
Agriculture, Ecosystems and Environment 275 (2019) 42–53
43
Page 3
to identify outcomes of each practice, four questions to identify re-
ferents for each practice and three questions to identify control factors
for each practice (see for a list the Supplementary Information [SI]
Table 1
).
The semi-structured interviews in the six AEZs resulted in 298
identi
fi
ed outcomes, referents and control factors. Each of these was
related to a practice and an AEZ. As could be expected, a substantial
number of these items were similar, or almost similar, between prac-
tices and AEZs. Therefore, mentioned outcomes, referents and control
factors were clustered in 110 classes (40 outcome classes, 13 referent
classes and 57 control factor classes). Following this, they were given
shorthand labels (such as soil structure, or contract workers). The 110
classes were clustered into nine main categories (
Table 3
). The nine
main categories were: soil type & climate, soil quality, crop protection,
land use, technical,
fi
nancial, environmental impact, legal and social.
As such, the categories were based on insights provided by farmers.
2.4. Step 2: farm survey
The outcomes, referents and control factors for the use of organic
inputs identi
fi
ed in the semi-structured interviews were quanti
fi
ed in a
large farm survey. For each outcome, referent and control factor, two
questions were asked, based on the theory of planned behaviour. In this
manner, the attitude, subjective norm and perceived behavioural con-
trol on the use of organic inputs could be calculated. In addition,
farmers were also asked to report on their current usage of each organic
input. Farmers were asked whether they cultivate green manures or
cover crops, incorporate straw and/or use FYM or compost, on at least
one of their
fi
elds.
The sampling for the questionnaires depended on the availability of
a valid sampling frame of arable farmers within each country (
i.e.
contact details of farmers). The most ideal sampling frame to obtain a
Fig. 1.
Map of study area indicating the six agro-ecological zones in four European countries (the Netherlands, Belgium, Austria and Italy).
Table 1
Characteristics of each agro-ecological zone. Climate zones follow
Metzger et al. (2005)
. Aridity index is calculated by dividing annual precipitation by potential
evapotranspiration, using spatial climatic data for the period 1975-2009(
Janssen et al., 2009
). A lower aridity index indicates a drier climate. Soil texture classes
follow classes used in the European soil database (
EC and ESBN, 2004
).
Climate zone
Average aridity index of the climate zone
Slope
Soil texture
AT1
Pannonian
0.51 (dry)
level (0°) or gentle (2-3°)
Medium
BE1
Atlantic Central
0.73 (wet)
nearly level (1°)
Medium
fi
ne
IT1-level
Mediterranean North
0.38 (very dry)
level (0°) or gentle (2-3°)
Coarse to medium
fi
ne
IT2-sloping
Mediterranean North
0.38 (very dry)
gentle to moderate (2-7°)
Medium or medium
fi
ne
NL1-clay
Atlantic North and Central
0.76 (wet)
level (0°)
Medium or medium
fi
ne
NL2-sand
Atlantic Central
0.76 (wet)
level (0°)
Coarse
Table 2
Overview of agro-ecological zones and practices included in survey.
Incorporation of
straw
Cultivation of green
manures/catch crops
Compost
FYM
AT1
X
BE1
X
X
X
X
IT1 (level)
X
X
IT2 (sloping)
X
X
NL1 (clay)
X
X
NL2 (sand)
X
X
X
R. Hijbeek et al.
Agriculture, Ecosystems and Environment 275 (2019) 42–53
44
Page 4
Table
3
Classi
fi
cation
of
identi
fi
ed
outcomes,
referents
and
control
factors
into
nine
categories.
Soil
type
&
climate
Soil
quality
Crop
protection
Land
use
Technical
Financial
Environmental
impact
Legal
Social
Control
factors
Outcomes
Outcomes
Outcomes
Outcomes
Outcomes
Outcomes
Referents
Referents
Clay
soil
Long
term
quality
Attracting
insects
Crop
rotation
Heterogeneous
spread
of
input
Costs
Landscape
view
Government
Advisors
Weather
Reduced
erosion
Diseases
Extensi
fi
cation
Machinery
Costs
cultivation
Nitrogen
leaching
Control
factors
Agricultural
education
Wet
climate
Soil
fauna
Fungal
diseases
Control
factors
Polluted
compost
Crop
yields
Soil
carbon
sequestration
Administrative
demands
Contract
workers
Soil
fertility
Herbicide
use
Compatibility
GM
Residue
incorporation
Fertiliser
use
Control
factors
Burning
not
allowed
Family
Soil
health
Nematodes
Crop
rotation
Sowing
cover
crops
Fuel
use
Nitrogen
leaching
Legal
nutrient
limits
Fellow
farmers
Soil
nutrients
Pesticide
use
Cultivation
potatoes
Timing
application
Income
Legal
sowing
restrictions
Government
Soil
organic
matter
Weeds
Cultivation
silage
corn
Control
factors
Labour/time
Subsidy
Land
owners
Soil
structure
Weeds,
pests,
diseases
Cultivation
winter
wheat
#
fi
eld
operations
Protein
content
Livestock
farmers
Soil
water
Sod
seeding
Adjacent
agricultural
land
Risk
low
biomass
GM
Magazines
Soil
workability
Compatibility
CT
Control
factors
Study
club
Straw
decomposition
Knowledge
Availability
of
FYM
Suppliers
Uncertainty
N
release
Legislation
on
e
ff
ective
N
Availability
of
straw
Control
factors
Machinery
Av.
of
compost
Contract
workers
Night
harvests
Costs
Cooperation
livestock
Ploughing
straw
Costs
chopping
straw
Land
exchange
Residue
incorporation
Costs
fertilisers
Protein
content
contracts
Seeding
Fertiliser
use
Timing
application
(of
input)
Fodder
need
Variation
in
quality
High
biomass
straw
Income
Labour/time
Labour
for
incorporating
Labour
for
sowing
Land
availability
Compatibility
livestock
Loss
of
income
Market
demands
(Availability
of)
other
organic
inputs
Seed
costs
Storage
facilities
Straw
as
crop
cover
Straw
prices
Subsidy
R. Hijbeek et al.
Agriculture, Ecosystems and Environment 275 (2019) 42–53
45
Page 5
completely random sample is a nation-wide database of farmer ad-
dresses. Such a database was available for Belgium (Flanders) and the
Netherlands. In these countries farmers were invited by letter to par-
ticipate in the farm survey. In Austria and Italy researchers depended
on farmer associations, farmer extension services or other contacts to
distribute the questionnaires. Questionnaires were
fi
lled online (Austria
and the Netherlands) or as a paper questionnaire (Belgium and Italy).
Response rates varied from 12 to 59% (
Table 4
).
To reduce the length of the questionnaires and increase the response
rate, questionnaires in Belgium and the Netherlands were split into two
or three parts allocated to di
ff
erent groups of farmers. As such, each
group of farmers received a questionnaire with a common introduction
section combined with in-depth questions for only one or a few speci
fi
c
types of organic input(s). In addition, to reduce errors in farmers
’
an-
swers, respondents were not obliged to
fi
ll out every question. In the
case of partly completed questionnaires, only the
fi
lled parts were used
in the analysis. For these two reasons, numbers of farmers di
ff
ered
between analyses. Questionnaires were returned during the summer
and autumn of 2013. Filled questionnaires were checked for irregula-
rities (typing mistakes or extreme numbers) and if any were found these
answers were removed from analysis.
Targeting farmers with similar conditions for slope and soil texture
within one AEZ proved to be di
ffi
cult as these characteristics di
ff
er on
small geographical scales. Overall, reported soil textures and slopes of
the respondents in each AEZ showed correspondence with the targeted
AEZs (
Table 1
) but there was also considerable variation within each
AEZ (
Table 5
).
In Austria, respondents to the farm survey had a very diverse range
of slopes and soil textures. In Belgium, farmers had mainly loamy soils
(91%) and mostly level to gently sloping lands. In Italy, IT1 farmers had
mainly level land and in IT2 farmers had mainly gentle to steep sloping
land. In the Netherlands, farmers in NL1 had mainly clay soils and fa-
mers in NL2 had mainly sandy soils.
2.4.1. Quanti
fi
cation of attitude, subjective norm and perceived behavioural
control
In the semi-structured interviews, relevant outcomes, referents and
control factors were identi
fi
ed for each combination of organic input
and AEZ. In the larger farm survey (step 2), these were quanti
fi
ed on a
scale of 1 to 10 by asking pairs of questions. For each identi
fi
ed out-
come, two types of questions were asked in the farm survey: 1) To
which degree a farmer expects a certain outcome (
i.e.
, result, e
ff
ect)
from the given practice (
‘
belief strength
’
) and 2) how is the outcome
valued on a scale from
‘
bad
’
to
‘
good
’
(
‘
outcome evaluation
’
), both on a
Likert scale from 1 to 5. The Likert scale was proposed by Rensis Likert
(Likert 1932) as a means to measure the level of agreement or dis-
agreement with given statements in questionnaires and has been com-
monly used since then. For example: a possible outcome from the ap-
plication of compost is improved soil structure. In the farm survey,
farmers were
fi
rst asked to which degree they expected compost to
improve soil structure (1 = very little, 5 = very much) and second how
they evaluated an improvement in soil structure (1 = very bad, 5 =
very good).
For each identi
fi
ed referent, two types of questions were asked in
the farm survey: 1) how motivated are they to comply with the re-
ferents
’
view (
‘
motivation to comply
’
) and 2) to which degree the re-
ferent is positive or negative towards a practice (
‘
normative belief
’
),
both on a Likert scale from 1 to 5.
For each identi
fi
ed control factor, two types of questions were asked
in the farm survey: 1) to which degree that control factor is valid for the
farm (
‘
control strength
’
) and 2) to which degree the control factor
makes the practice attractive or di
ffi
cult (
‘
control power
’
), both on a
Likert scale from 1 to 5.
Values for outcome evaluations, normative belief and control power
were lowered by three points to give a negative to positive scaling (-2 to
+2). Next, values for each pair of questions were multiplied to obtain a
score for attitude, subjective norm and perceived behavioural control
(scores ranging from -10 to +10), see equations
1
to 3.
=
−
attitude
belief strength
outcome valuation
*(
3)
i
i
i
(1)
=
−
subjective norm
motivation to comply
normative belief
*(
3)
k
k
k
(2)
=
−
perceived behavioural control
control strength
control power
*(
3)
m
m
m
(3)
2.4.2. Classi
fi
cation as drivers or barriers
For each attitude, subjective norm and perceived behavioural con-
trol, the mean value across all respondents in an AEZ was calculated,
together with the 95% con
fi
dence interval. The 95% con
fi
dence in-
terval was calculated as
1.96 *
sd
N
in which sd means standard deviation
and N is the number of farmers who answered the question. Values
were considered signi
fi
cantly di
ff
erent from zero when the 95% con-
fi
dence interval did not include 0. Signi
fi
cant positive values were
Table 4
Numbers of farmers included in the semi-structured interviews (step 1) and the
larger farm survey (step 2).
Step 1
Step 2
Semi-structured
interviews
Questionnaires sent
Response
Response rate
AT1
8
open online survey
34
NA
BE1
7
1600
430
27%
IT1-level
8
211
124
59%
IT2-sloping
9
165
98
59%
NL1-clay
10
2700
336
12%
NL2-sand
5
2000
241
12%
Table 5
: Characteristics (slopes and soil texture) of the agro-ecological zones and their farms in the survey, averaged per agro-ecological zone. N indicates number of farmers
on which the analysis is based; if a range is speci
fi
ed, this means that the number of farmers varied by practice.
AT1 (N = 34)
BE1 (N = 371-395)
IT1-level (N = 102-114)
IT2-sloping (N = 82-85)
NL1-clay (N = 331-333)
NL2-sand (N = 217-219)
Slope (mean percentage farm area)
Level (0°)
18%
39%
80%
8%
100%
100%
Nearly level (1°)
22%
31%
10%
8%
0%
0%
Gently sloping (2-3°)
32%
22%
5%
31%
0%
0%
Sloping (4-7°)
16%
7%
4%
39%
0%
0%
Steep (> 8°)
12%
2%
1%
14%
0%
0%
Soil texture (mean percentage farm area)
Sand
26%
8%
12%
1%
1%
93%
Loam
40%
91%
72%
62%
7%
1%
Clay
32%
1%
16%
38%
91%
1%
Peat
0%
0%
0%
0%
0%
4%
R. Hijbeek et al.
Agriculture, Ecosystems and Environment 275 (2019) 42–53
46
Page 6
classi
fi
ed as drivers. Signi
fi
cant negative values were classi
fi
ed as bar-
riers.
As described above, attitude, subjective norm and perceived beha-
vioural control were based on two questions. The
fi
rst question de-
termines the
‘
strength
’
of a factor (from 1 to 5), while the second
question illustrates how a farmer
‘
evaluates
’
a factor (from negative to
positive). Whilst an outcome or a control factor can be evaluated very
negatively, if the strength is very low (
e.g.
below 3), than the chance of
this occurring is still considered very small. We therefore added a
second dimension to the classi
fi
cation of drivers and barriers, denoting
all drivers and barriers which had an average
‘
strength
’
score below 3 as
‘
potential
’
. These latter factors were perceived to have a positive or
negative impact, but farmers considered the likelihood of this occurring
less than average (
Fig. 2
).
Each factor belonged to one of the 110 classes and nine main ca-
tegories (
Table 3
). If two identi
fi
ed factors for a practice and AEZ fell
into the same class (
e.g.
soil fertility) and they were evaluated both
positive (or both negative), the average was taken. When two identi
fi
ed
factors for a practice and AEZ fell into the same sub category but the
signs di
ff
ered (the one being negative and the other positive), we
wanted to keep this information visible in the results. Therefore, in
these cases they were both kept separate in the analysis (meaning the
speci
fi
c class can act both positively or negatively for applying given
practice).
3. Results
3.1. Current use of organic inputs
In all AEZs, farmers used organic inputs, but at varying quantities
(
Table 6
). When cultivating cereals (including grain maize), most
farmers incorporated straw, intermittently or consistently over the
years, except for NL1. In BE1, fewer farmers incorporated straw (26%),
probably due to the high use of FYM (straw is exported to livestock farm
and imported as FYM).
A large di
ff
erence in the cultivation of green manures or cover crops
is observed between the six AEZs (
Table 6
). While in AT1, BE1, NL1 and
NL2 most farmers (> 80%) cultivated green manures or cover crops on
at least one of their
fi
elds, only a minority (10%) cultivated green
manures or cover crops in IT1 and IT2.
A higher percentage of farmers in IT1 and IT2 (42 and 55%) used
compost compared to farmers in AT1 and BE1 (17% and 8%), and with
NL1 and NL2 in between (35% and 29%). The use of FYM varied
widely, with the highest percentage of farmers using FYM in BE1 and
IT2 (71% and 67%) and the lowest in NL2 (16%). In the Netherlands,
farmers on sandy soils are often located near pig farms, which create a
large availability of slurry, possibly explaining the low use of FYM.
3.2. Drivers and barriers for the incorporation of straw
Incorporation of straw was included in the farm survey of
fi
ve AEZs
(BE1, IT1, IT2, NL1 and NL2). The total number of outcomes, referents
and control factors identi
fi
ed was 96 (summed over the
fi
ve AEZ
’
s, or
on average 19.2 per AEZ). Out of these, 48 were evaluated positively as
drivers, of which 6 potential drivers. 35 factors were evaluated sig-
ni
fi
cantly negatively as barriers, of which 16 potential (
Fig. 3
). Of the
drivers, around 40% were related to soil quality (N = 19). Of the bar-
riers, almost half were
fi
nancial (N = 17), yet there were also some
fi
nancial drivers (N = 6). All outcomes and control factors related to
crop protection were evaluated as barriers (4 out of 4).
In all AEZs, e
ff
ects on soil structure and SOM content were identi-
fi
ed as outcomes and evaluated as drivers for straw incorporation
(
Fig. 3
).
In almost all AEZs - besides BE1
–
impacts on weeds, pests and
diseases in general or speci
fi
cally fungal diseases were evaluated as a
barrier for straw incorporation. In addition, costs of straw incorporation
and loss of income from selling were also evaluated as barriers for straw
incorporation in all AEZs. In most AEZs, the subjective norms of re-
ferents were positive for straw incorporation. Only in BE1, a negative
subjective norm exists from fellow farmers.
In BE1 (
Fig. 3
a), legal nutrient limits were seen as a driver as well as
a barrier. Like the Netherlands, the region of Flanders in Belgium where
the study was conducted, is a Nitrate Vulnerable Zone, which gives
stricter limitations on nutrient inputs for farmer. Straw is however not
counted within the legal nutrient limits, therefore it was seen by some
farmers as an easy manner to increase SOM content. At the same time,
other farmers prefer to add N to straw for decomposition, in which case
the legal nutrient limits are a barrier. In addition, it is a legal obligation
for farmers in Flanders (Belgium) to measure nitrate residues in autumn
to prove that their
fi
elds will not cause excessive nitrate leaching. Some
farmers also indicated that an advantage of incorporating straw is that
it immobilises N and thus lowers the nitrate residue in autumn. The
apparent contradictory results regarding straw incorporation and
manure legislation due to the Nitrate Directive illustrates the com-
plexity of the decision making process. As Belgian farmers often apply
FYM
–
which also has a positive e
ff
ect on SOM, this makes the in-
corporation of straw less attractive or necessary to maintain soil
quality.
In Italy (both IT1 and IT2;
Fig. 3
bc), legal prohibition of burning
straw was considered a driver for straw incorporation. In IT1 speci
fi
-
cally, e
ff
ects of straw incorporation on product quality (protein con-
tent) were considered a driver, while in IT2 availability of adequate
machinery for chopping and incorporating residue was considered a
driver.
In the Netherlands (NL1 and NL2;
Fig. 3
de), reduced labour or time
Fig. 2.
Graphical illustration of the division on potential barriers, barriers,
potential drivers and drivers, based on average scores on underlying questions.
* combined score refers to the mean value for attitude, subjective norm or
perceived behavioural control. ** strength refers to the mean value for the
‘
strength
’
questions (outcome strength, motivation to comply and control
strength).
Table 6
Use of organic inputs in each agro-ecological zone. Bold numbers indicate
combinations of AEZ and input type for which further questions on outcomes,
referents and control factors are included in the survey. For straw incorpora-
tion, percentages are calculated as share of the farmers who cultivate the spe-
ci
fi
ed crop.
AT1
BE1
IT1-
level
IT2-
sloping
NL1-
clay
NL2-
sand
Straw incorporation (% farmers incorporating straw sometimes to always)
a
small grain cereal
straw
71%
26%
–
–
68%
60%
maize straw
88%
94%
–
–
38%
65%
cereals, maize or
sun
fl
ower
–
–
93%
69%
–
–
Green
manure or cover crops (% farms cultivating GM or CC)
GM or CC
93%
87%
10%
10%
84%
83%
Application FYM or compost (% farmers applying compost or FYM on at least some part of
the land)
Compost
17%
8%
42%
55%
35%
29%
FYM
46%
71%
57%
67%
39%
16%
a
Only farmers are included when speci
fi
c crop (maize, cereal) was present in
crop rotation of farm. In Italy, sun
fl
ower was also included in the question on
straw incorporation.
R. Hijbeek et al.
Agriculture, Ecosystems and Environment 275 (2019) 42–53
47
Page 7
requirements for straw incorporation (compared to removing it from
the land) and positive e
ff
ects on workability of the soil were considered
important drivers. Fertiliser use was a strong barrier, because farmers
believe additional fertiliser is needed for the decomposition of straw.
3.3. Drivers and barriers for the cultivation of green manures or cover crops
Cultivation of green manures or cover crops was evaluated in the
farm survey of all six AEZs. The total number of outcomes, referents and
control factors identi
fi
ed for the cultivation of green manures or cover
crops was 144 (summed over the six AEZs, or on average 24 per AEZ).
Out of these, 81 were evaluated positively as drivers and 44 were
evaluated negatively as barriers. Of the drivers, 35% were related to
soil quality (N = 28). One third of the barriers was
fi
nancial (N = 14),
yet there were also some
fi
nancial drivers (N = 10). All outcomes and
control factors for green manures related to environmental impact were
evaluated as drivers (6 out of 6).
As for straw incorporation, e
ff
ects on soil structure and SOM content
were identi
fi
ed and evaluated as drivers in each AEZ for the cultivation
of green manures or cover crops (
Fig. 4
). The e
ff
ect of green manures or
cover crops on soil erosion was also identi
fi
ed and evaluated as a driver
in all AEZs except IT1 (this factor was not included in survey because it
was not mentioned in the interviews). In the Netherlands (NL1 and
NL2) subjective norms of all referents were positive, unlike for example
Italy (IT1 and IT2) where especially fellow farmers and family were
perceived to have a negative view on the cultivation of green manures
or cover crops.
In four AEZs (AT1, BE1, NL1 and NL2) weeds, pests and diseases (as
a general term) or more speci
fi
cally increases in weeds, nematodes,
herbicide use or pesticide use (as speci
fi
c terms) were evaluated ne-
gatively as barriers for the cultivation of green manures or cover crops.
In BE1 however, perceived negative e
ff
ects on weeds and herbicide use
were only potential (having a lower strength). In IT1 and IT2 issues
with crop protection were not mentioned in the semi-structured inter-
views with the farmers and therefore not included in the farm survey. In
all AEZs, costs and e
ff
ects on income were barriers for the cultivation of
Fig. 3.
Drivers and barriers for incorporation
of straw per agro-ecological zone.
■
= mean
attitude score on outcome;
●
= mean sub-
jective norm of referent
▲
= mean perceived
behavioural control on control factor. Lines
indicate 95% con
fi
dence interval. * indicates
an
underlying
strength
score < 3.
N = numbers of farmers included in analysis.
Numbers in [] link to farmers
’
descriptions in
SI
Table 2
. GM = green manure. Colours in-
dicate main categories. (For interpretation of
the references to colour in this
fi
gure legend,
the reader is referred to the web version of this
article).
R. Hijbeek et al.
Agriculture, Ecosystems and Environment 275 (2019) 42–53
48
Page 8
green manures or cover crops.
In AT1 (
Fig. 4
a), e
ff
ects of cultivating green manures or cover crops
on the water holding capacity of the soil was evaluated as a driver as
well as a barrier, consistent with previous experimental
fi
ndings
(
Bodner et al., 2011
;
Bodner, 2013
). Austrian farmers can receive a
subsidy for the cultivation of green manures or cover crops if they
follow a training, which was evaluated as a driver.
In Flanders (BE1;
Fig. 4
b), farmers could also receive a subsidy for
cultivating green manures or cover crops at the time the survey was
conducted. To be eligible for subsidy, farmers in Flanders have to sow
the green manure before a certain date (before the 1
st
of September or
the 15
th
of October, depending on their location). As the time period
between harvest of some crops and legal sowing date is perceived as too
short, this preconditions was considered a barrier. Crops harvested late
in autumn were alsoconsidered a barrier for sowing cover crops. One of
the most important perceived drivers for farmers for sowing cover crops
in Flanders was that they result in uptake of remaining mineral N in the
soil pro
fi
le and thus prevent N leaching.
In Italy (IT1 and IT2;
Fig. 4
cd), costs were a major barrier for the
cultivation of green manures or cover crops. Additional costs include
production factors such as seeds, but also mechanical operations such as
tillage or sowing. Interestingly, farmers perceived a positive e
ff
ect of
Fig. 4.
Drivers and barriers for cultivation of
green manures or cover crops per agro-ecolo-
gical zone.
■
= mean attitude score on out-
come;
●
= mean subjective norm of referent
▲
= mean perceived behavioural control on
control factor. Lines indicate 95% con
fi
dence
intervals. * indicates an underlying strength
score < 3. N = numbers of farmers included in
analysis. Numbers in [ ] link to farmers
’
de-
scriptions in SI
Table 3
. GM = green manure.
CT = conservation tillage. For legend of col-
ours see
Fig. 6
or 8 (For interpretation of the
references to colour in this
fi
gure legend, the
reader is referred to the web version of this
article).
R. Hijbeek et al.
Agriculture, Ecosystems and Environment 275 (2019) 42–53
49
Page 9
green manure or cover crops on fertiliser use in IT2, with savings of
fertiliser in the following cash crop. In IT1, having clay soils was seen as
a barrier for the cultivation of green manure or cover crops. On clay
soils, seedbed preparation for green manures might be more compli-
cated due to excessive soil water content.
In the Netherlands (NL1 and NL2;
Fig. 4
ef), improved workability of
the soil was evaluated as an important driver. This relates to perceived
reductions in fuel costs and perceived reductions in the loss of soil
during harvest. In the Netherlands (NL1 and NL2), requirements in
labour and time to cultivate green manures or cover crops was con-
sidered a barrier. This relates speci
fi
cally to the small time frame
available at the end of summer after harvesting the main crop for the
sowing of the green manure.
3.4. Drivers and barriers for the application of compost or FYM
Application of compost was included as a practice in two AEZs (BE1
and NL2-sand), while the use of FYM was included in one AEZ (BE1). As
both types of input are often imported on arable farms, we present the
results of these two input types together. The total number of outcomes,
referents and control factors identi
fi
ed for the application of compost or
FYM was 58 (summed over the six AEZs, or on average 19.3 per AEZ).
Out of these, 20 were evaluated positively as drivers and 28 were
evaluated negatively as barriers. Of the drivers, more than half were
related to soil quality (N = 13). Almost half of the barriers were
fi
-
nancial (N = 13) and a quarter of the barriers were technical (N = 7).
There were no
fi
nancial drivers for the use of compost or FYM, nor was
any driver related to crop protection or environmental impact.
For compost, the e
ff
ect on SOM content was evaluated positively by
farmers as a driver in both AEZs (
Fig. 5
ab). In BE1, e
ff
ects of compost
on various aspects related to soil quality were also evaluated as drivers
(
e.g.
soil life, soil health, soil fertility and soil structure). In the Neth-
erlands compost may be applied throughout the year - in contrast to
animal manures - making it a more attractive option compared to other
organic inputs. Therefore, the legally allowed timing of application was
also evaluated as a driver for compost in NL2. Availability of alternative
organic inputs such as animal manure, pollution of compost and re-
quired labour and time for spreading were considered barriers for using
compost in both BE1 and NL2. In addition a range of other issues were
evaluated as barriers in BE1, such as availability, costs, little knowledge
on the composition, variation in quality, lack of experience and a risk
for high nitrate residue in autumn (see also above), increases in weeds,
pests and diseases, and legal nutrient restrictions. A more thorough
discussion of the data for compost can be found in
Viaene et al. (2016)
.
For the application of FYM, a similar range of positive e
ff
ects were
found on soil quality as from compost (
e.g.
soil life, soil organic matter,
reduced erosion, soil fertility and soil structure). An important barrier
here is the need for appropriate storage facilities for FYM in Flanders
during winter. Many farmers stated that they do not have adequate
storage place and that if they must have such a place it makes the use of
farmyard manure unattractive. Another perceived barrier was the al-
ternative availability of slurry, which has lower costs.
3.5. Drivers and barriers aggregated per category
Of all the 298 identi
fi
ed outcomes, referents and control factors in
the semi-structured interviews, 149 were evaluated positively as dri-
vers, of which 21 had a mean strength score below 3 and were therefore
Fig. 5.
Drivers and barriers for application of
compost or FYM per agro-ecological zone.
■
= mean attitude score on outcome;
●
= mean
subjective norm of referent
▲
= mean per-
ceived behavioural control on control factor.
Lines indicate 95% con
fi
dence intervals. * in-
dicates an underlying strength score < 3.
N = numbers of farmers included in analysis.
Numbers in [ ] link to farmers
’
descriptions in
SI
Tables 4 and 5
. FYM = farmyard manure.
N = nitrogen. Colours indicate main cate-
gories.
R. Hijbeek et al.
Agriculture, Ecosystems and Environment 275 (2019) 42–53
50
Page 10
considered as
‘
potential
’
drivers (
Fig. 6
). Out of all factors, 107 were
evaluated negatively as a barrier, of which 46 were considered as
‘
po-
tential
’
barriers.
Aggregating drivers across AEZs, a positive impact on soil quality
stands out as the most important driver for using organic inputs. For all
types of organic input, soil quality consistently has the highest count
and percentage of cases in which it is evaluated as a driver (except
environmental impact of green manures or cover crops,
Fig. 6
b).
For each type of organic input,
fi
nancial consequences are most
often mentioned as a barrier, but
‘
negative impacts on crop protection
’
has the highest percentage of cases in which it is evaluated as a barrier
compared to the total sum of barriers and drivers in that category. Both
fi
nancial consequences and crop protection are therefore important
barriers for the types of organic input considered in this study.
4. Discussion
4.1. Identi
fi
ed drivers and barriers
In this study, a large number of outcomes, referents and control
factors (298) were identi
fi
ed as relevant to farmers in their use of
organic inputs.
For all types of organic inputs, important drivers were
‘
perceived
e
ff
ects on soil quality
’
and a positive in
fl
uence from social referents.
Farmers in our survey showed a great interest in maintaining long-term
soil fertility and SOM contents. This seems in contradiction to the as-
sumption that farmers in Europe have too few incentives to use soil
conservation measures because productive capacities are not a
ff
ected
by soil degradation in the short-term (
Van den Putte et al., 2010
). Ef-
fects on SOM content and soil fertility were consistently amongst the
top drivers for each type of organic input (
Fig. 3, 4 and 5
). We con-
fi
rmed the notion that farmer behaviour is motivated by a wide array of
factors which are not all directly economic (
Glenk et al., 2017
) as
fi
-
nancial aspects were only a selected part of the total listed drivers and
barriers. Identi
fi
ed drivers more often had a higher mean strength (> 3)
than identi
fi
ed barriers, indicating a larger likelihood of occurrence. If
mentioned,
fi
nancial consequences of using organic inputs were more
often evaluated as negative than positive.
In our farm survey, farmers consistently evaluated the e
ff
ects of
organic inputs on crop protection negatively. Cereal straw may spread
an important fungal disease causing grain contamination (Fusarium
head blight, FHB) to subsequent cereals in crop rotations (
Maiorano
et al., 2008
), although some measures to relieve this pressure are also
known. The spread of FHB can be reduced when crop residues are in-
corporated into the soil (
Blandino et al., 2012
). Organic inputs have
been observed to reduce soil-borne diseases, but only if they are rich in
nitrogen (
Bailey and Lazarovits, 2003
). In a recent literature review on
the role of compost on soil-borne plant diseases,
Mehta et al. (2014)
found a large variation in e
ff
ects, but mainly positive e
ff
ects on disease
suppression when composts were enriched with bene
fi
cial soil mi-
crobes. As indicated by
Bailey and Lazarovits (2003)
, the e
ff
ect on
disease suppression in soils might manifest itself only after long-term
continuous application of organic inputs. If farmers experiencenegative
e
ff
ects on crop protection in the
fi
rst year of using a speci
fi
c type of
organic input, they might however not be willing
–
without convincing
evidence - to wait and see if this changes in the future.
4.2. Variation in drivers and barriers and the use of organic inputs between
AEZs
When considering the use of organic inputs in each AEZ (
Table 6
),
the most striking is the di
ff
erence in cultivation of green manures. In
both IT1 and IT2, only 10% of farmers cultivated green manures or
cover crops, whilst in the other AEZs this percentage was above 83%. In
IT1 and IT2 speci
fi
cally, fellow farmers and family were perceived as
having a negative view on the cultivation of green manures whilst
suppliers and advisors were somewhat neutral. (
Fig. 3
b,c). Therefore,
the community surrounding farmers does not promote the adoption of
green manures. More generally, from personal experience it is known
that, because green manure is still a relatively new practice for Italian
farmers, communication and learning from peers are very important
factors in increasing its adoption. From the experience of farmers who
are already successful with green manures, other farmers could learn
what are the best species or mixes, seed rate, seeding date, seeding
technique, and termination technique in a given soil and climate.
Currently the adoption of green manures is, however, low, so farmers
have little exposure to these practices.
4.3. Straw incorporation and N requirements
For the incorporation of straw, legal nutrient limits were mentioned
to be relevant in BE1, NL1 and NL2. Flanders (Belgium) and the
Netherlands are designated Nitrate Vulnerable Zones (
EEC, 1991
). In
these cases, farmers are restricted by legal nutrient limits when using
mineral fertilisers and/or organic inputs. In BE1, legal nutrient limits
were perceived both a driver and a barrier for the incorporation of
straw. This corresponds with the observation that fertiliser use is both a
Fig. 6.
Counts and percentages of drivers, potential drivers, barriers and po-
tential barriers per practice and category.
R. Hijbeek et al.
Agriculture, Ecosystems and Environment 275 (2019) 42–53
51
Page 11
driver and a barrier in IT1. In certain cases, incorporation of straw can
supply N to crops, whilst in others, it can induce temporary N im-
mobilization and needs N for decomposition (
Hijbeek et al., 2018a
;
Silgram and Chambers, 2002
).
Moreover, straw returns phosphorus (P) and potassium (K) to soil
(which explains the driver
‘
expected reduced fertiliser use
’
). It seems
that both in BE1 and IT1, farmers value the positive e
ff
ects of straw
incorporation in the long-term (build-up of SOM content; return of P
and K), but have a negative view on the short-term needs of mineral N
fertiliser for the decomposition of straw. In Italy, farmers preferred
burning (despite the loss of organic matter and risk of
fi
re), selling or
even giving away the residues. Burning is now forbidden, which then
acts as a driver for incorporating straw in both IT1 and IT2.
4.4. Limitations of the study
This study has used a sociological methodology to improve our
understanding of agronomic practices. This resulted in a number of
interesting insights, but also involved a number of limitations. Below
we discuss the most relevant limitations.
First, slope and soil texture varied greatly within each agro-ecolo-
gical zone. Second, means of the combined scores of attitude, subjective
norm and perceived behavioural control were assessed. These concepts
are based on underlying farmer beliefs. To account for this, we divided
drivers and barriers into
‘
potential
’
drivers and barriers when their
underlying
‘
strength
’
was below 3 on average. When zooming in on one
speci
fi
c practice in one region, one can also look in more details for the
variation in the underlying beliefs, such as was for example done by
Bechini et al. (2015)
for straw incorporation in Italy and by
Viaene
et al. (2016)
for compost in Flanders.
Thirdly, drivers and barriers were analysed separately for the use of
each type of organic input. Farmers, however, indicated that the use of
one type of organic input might depend on the use of another. For
example, if they use FYM, they might not incorporate straw and if a
farmer already uses slurry this might be a reason not to use FYM. Both
the Netherlands and Belgium have a large supply of animal manure due
to the presence of a relatively large livestock sector (
Oenema and
Berentsen, 2004
;
Viaene et al., 2016
). Sometimes, animal slurry is even
o
ff
ered for free to arable farmers. In these cases, this cheap alternative
organic input is a barrier for the use of for example compost (BE1 and
NL2;
Fig. 5
ab). If all the amounts of organic inputs used by a farmers
are known, one can also express the total sum in an aggregated measure
(for example as total C or e
ff
ective C, see
Hijbeek et al., 2018b
). This
information was, however, not available for all our regions. In addition,
aggregating all organic inputs may ignore robust drivers and barriers
for a speci
fi
c practice, which was the focus of this study.
5. Conclusion
Farmers across all six AEZs considered improvements in soil quality
(
e.g.
soil structure, soil fertility, soil health) as drivers and in most cases
there was a positive in
fl
uence from social referents. Many perceived
barriers related to crop protection (
e.g.
e
ff
ects on the incidence of
weeds, pests and diseases or e
ff
ects on herbicide or pesticide require-
ments) and
fi
nancial consequences (higher costs; income forgone).
Other barriers were more context-speci
fi
c such as manure legislation or
a large and free availability of slurry.
Our
fi
ndings show that farmers perceive a trade-o
ff
between posi-
tive e
ff
ects on soil quality but negative ones on soil born parasites or
plant diseases while using organic inputs. Farmers, therefore, require
speci
fi
c guidance on how best to reduce the pressure of weeds, pests
and diseases with site- or farming-speci
fi
c advice. These are relevant
insights if agricultural and environmental policies aim to include the
use of organic inputs to maintain or increase SOM contents.
Currently from a farmers
’
perspective,
fi
nancial consequences of
using organic inputs cannot be assumed to be neutral in the short-term.
This highlights the need for more research into the long-term
fi
nancial
consequences of using organic inputs. In the meantime, when pro-
moting societal bene
fi
ts of using organic inputs beyond the farm-gate
(
e.g.
biodiversity conservation or soil carbon sequestration),
fi
nancial
consequences should be accounted for.
Financial support
This publication was
fi
nancially supported by the Dutch Ministry of
Economic A
ff
airs (KB14 and PPS Sustainable Soil Management TKI-AF-
12054A) and by the European Commission under the CATCH-C project
(Grant Agreement N° 289782) within the 7th Framework Programme
for Research, Technological Development and Demonstration. Its con-
tent does not represent the o
ffi
cial position of the EC and is entirely the
responsibility of the authors.
Acknowledgements
Our gratitude goes to all European farmers who spared some of their
time to
fi
ll in the questionnaire. Setting up a large scale farm survey
would not have been possible without numerous colleagues across
di
ff
erent institutes. In Austria, main contacts were the provincial and
regional o
ffi
cials of the chamber of agriculture in Lower Austria. In
Belgium, data was collected with the help of Erwin Wauters (ILVO). In
Italy, we received support from universities of Ancona, Bari, Milano,
Napoli Udine, Pisa, Teramo, Tuscia, Ancona and Perugia; provincial
and regional o
ffi
cials; farmers unions (Coldiretti, Confagricoltura,
Confederazione Italiana Agricoltori); CadirLab (Quargnento, AL);
Consorzio Agrario di Siena; advisors of the Italian Farm Accountancy
Data Network; Chiara Costamagna; students of the University of Torino
and University of Milano. In the Netherlands, Rob Smidt helped with
extracting farmers
’
addresses from the database and Bert Rijk supported
the survey logistics. We thank all colleagues for their e
ff
orts and sup-
port. In addition, we thank two anonymous reviewers for their con-
structive comments and Laura Webb for reading a draft version of the
manuscript.
Appendix A. Supplementary data
Supplementary material related to this article can be found, in the
online version, at doi:
https://doi.org/10.1016/j.agee.2019.01.008
.
References
Ajzen, I., 1991. The theory of planned behavior. Organ. Behav. Hum. Decis. Process. 50,
179
–
211
.
Bailey, K.L., Lazarovits, G., 2003. Suppressing soil-borne diseases with residue manage-
ment and organic amendments. Soil Tillage Res. 72, 169
–
180
.
Bechini, L., Costamagna, C., Zavattaro, L., Grignani, C., Bijttebier, J., Ruysschaert, G.,
2015. Barriers and drivers towards the incorporation of crop residue in the soil.
Analysis of Italian farmers
’
opinion with the theory of planned behaviour. Ital. J.
Agron. 10, 178
–
184
.
Bellamy, P.H., Loveland, P.J., Bradley, R.I., Lark, R.M., Kirk, G.J.D., 2005. Carbon losses
from all soils across England and Wales 1978
–
2003. Nature 437, 245
.
Bijttebier, J., Ruysschaert, G., Hijbeek, R., Werner, M., Pronk, A.A., Zavattaro, L., Bechini,
L., Grignani, C., ten Berge, H., Marchand, F., Wauters, E., 2018. Adoption of non-
inversion tillage across Europe: Use of a behavioural approach in understanding
decision making of farmers. Land Use Policy 78, 460
–
471
.
Blandino, M., Haidukowski, M., Pascale, M., Plizzari, L., Scudellari, D., Reyneri, A., 2012.
Integrated strategies for the control of Fusarium head blight and deoxynivalenol
contamination in winter wheat. Field Crops Res. 133, 139
–
149
.
Bodner, G., 2013. Hydrologische E
ff
ekte des Zwischenfruchtanbaus im Trockengebiet.
Innovation Universität für Bodenkultur Wien, Austria
.
Bodner, G., Summerer, H., Ecker, F., Rosner, J., 2011. Zwischenfruchtbau ist auch im
Trockengebiet machbar Ländlicher Raum.Online-Fachzeitschrift des
Bundesministeriums für Land- und Forstwirtschaft. Umwelt und Wasserwirtschaft
.
Bollag, J.-M., Myers, C.J., Minard, R.D., 1992. Biological and chemical interactions of
pesticides with soil organic matter. Sci. Total Environ. 123
–
124, 205
–
217
.
Burton, R.J., 2004. Reconceptualising the
‘
behavioural approach
’
in agricultural studies: a
socio-psychological perspective. J. Rural Stud. 20, 359
–
371
.
Capriel, P., 2013. Trends in organic carbon and nitrogen contents in agricultural soils in
Bavaria (south Germany) between 1986 and 2007. Eur. J. Soil Sci. 64, 445
–
454
.
R. Hijbeek et al.
Agriculture, Ecosystems and Environment 275 (2019) 42–53
52
Page 12
Chang, E.-H., Chung, R.-S., Tsai, Y.-H., 2007. E
ff
ect of di
ff
erent application rates of or-
ganic fertilizer on soil enzyme activity and microbial population. Soil Sci. Plant Nutr.
53, 132
–
140
.
Chapman, S.J., Bell, J.S., Campbell, C.D., Hudson, G., Lilly, A., Nolan, A.J., Robertson,
A.H.J., Potts, J.M., Towers, W., 2013. Comparison of soil carbon stocks in Scottish
soils between 1978 and 2009. Eur. J. Soil Sci. 64, 455
–
465
.
Christensen, B.T., Johnston, A.E., 1997. Chapter 18 Soil organic matter and soil
quality
—
Lessons learned from long-term experiments at Askov and Rothamsted. Dev.
Soil Sci. 25, 399
–
430
.
Dersch, G., 2015. Die Humusgehalte der heimischen Ackerböden haben sich positiv en-
twickelt. BauernZeitung p. 6
.
EC, 2013. Regulation (EU) No 1306/2013 of the European parliament and of the council
of 17 D ecember 2013 on the
fi
nancing, management and monitoring of the common
agricultural policy. Annex II. O
ff
. J. Eur. Union
.
EC and ESBN, 2004. The European Soil Database Distribution Version 2.0, CD-ROM, EUR
19945 EN. European Commission and the European Soil Bureau Network
.
EEC, 1991. Council Directive 91/676/ EEC of 12 December 1991 concerning the pro-
tection of waters against pollution caused by nitrates from agricultural sources. O
ff
. J.
l 375, 12
.
Fantappiè, M., L
’
Abate, G., Costantini, E.A.C., 2011. The in
fl
uence of climate change on
the soil organic carbon content in Italy from 1961 to 2008. Geomorphology 135,
343
–
352
.
Freibauer, A., Rounsevell, M.D., Smith, P., Verhagen, J., 2004. Carbon sequestration in
the agricultural soils of Europe. Geoderma 122, 1
–
23
.
Frelih-Larsen, A., Bowyer, C., Albrecht, S., Keenleyside, C., Kemper, M., Nanni, S.,
Naumann, S., Mottershead, D., Landgrebe, R., Andersen, E., Ban
fi
, P., Bell, S.,
Brémere, I., Cools, J., Herbert, S., Iles, A., Kampa, E., Kettunen, M., Lukacova, Z.,
Moreira, G., Kiresiewa, Z., Rouillard, J., Okx, J., Pantzar, M., Paquel, K., Pederson, R.,
Peepson, A., Pelsy, F., Petrovic, D., Psaila, E.,
Š
arapatka, B., Sobocka, J., Stan, A.-C.,
Tarpey, J., Vidaurre, R., 2016. Updated Inventory and Assessment of Soil Protection
Policy Instruments in EU Member States. Ecologic Institute, Berlin
.
Glenk, K., Shrestha, S., Topp, C.F., Sánchez, B., Iglesias, A., Dibari, C., Merante, P., 2017.
A farm level approach to explore farm gross margin e
ff
ects of soil organic carbon
management. Agric. Syst. 151, 33
–
46
.
Goidts, E., van Wesemael, B., 2007. Regional assessment of soil organic carbon changes
under agriculture in Southern Belgium (1955
–
2005). Geoderma 141, 341
–
354
.
Heikkinen, J., Ketoja, E., Nuutinen, V., Regina, K., 2013. Declining trend of carbon in
Finnish cropland soils in 1974
–
2009. Glob. Change Biol. 19, 1456
–
1469
.
Hijbeek, R., Wolf, J., van Ittersum, M.K., 2013. A Typology of Farming Systems, Related
Soil Management and Soil Degradation in Eight European Countries. D2.242, CATCH-
C Report. Wageningen University and Research Centre, Wageningen, pp. 226
.
Hijbeek, R., ten Berge, H.F.M., Whitmore, A.P., Barkusky, D., Schröder, J.J., van Ittersum,
M.K., 2018a. Nitrogen fertiliser replacement values for organic amendments appear
to increase with N application rates. Nutr. Cycl. Agroecosyst. 110, 105
–
115
.
Hijbeek, R., Pronk, A., van Ittersum, M., ten Berge, H., Bijttebier, J., Verhagen, A., 2018b.
What drives farmers to increase soil organic matter? Insights from the Netherlands.
Soil Use Manag. 34, 85
–
100
.
Janssen, S., Andersen, E., Athanasiadis, I.N., van Ittersum, M.K., 2009. A database for
integrated assessment of European agricultural systems. Environ. Sci. Policy 12,
573
–
587
.
Johnston, A.E., Poulton, P.R., Coleman, K., Donald, L.S., 2009. Chapter 1 soil organic
matter: its importance in sustainable agriculture and carbon dioxide
fl
uxes. Advances
in Agronomy. Academic Press, pp. 1
–
57
.
Lynne, G.D., Shonkwiler, J.S., Rola, L.R., 1988. Attitudes and farmer conservation
behavior. Am. J. Agric. Econ. 70, 12
–
19
.
Maiorano, A., Blandino, M., Reyneri, A., Vanara, F., 2008. E
ff
ects of maize residues on the
Fusarium spp. infection and deoxynivalenol (DON) contamination of wheat grain.
Crop. Prot. 27, 182
–
188
.
Mehta, C.M., Palni, U., Franke-Whittle, I.H., Sharma, A.K., 2014. Compost: its role, me-
chanism and impact on reducing soil-borne plant diseases. Waste Manag. 34,
607
–
622
.
Metzger, M.J., Bunce, R.G.H., Jongman, R.H.G., Mücher, C.A., Watkins, J.W., 2005. A
climatic strati
fi
cation of the environment of Europe. Glob. Ecol. Biogeogr. 14,
549
–
563
.
Oenema, O., Berentsen, P., 2004. Manure Policy and MINAS: Regulating Nitrogen and
Phosphorus Surpluses in Agriculture of the Netherlands. OECD
.
Panagos, P., Borrelli, P., Meusburger, K., Alewell, C., Lugato, E., Montanarella, L., 2015.
Estimating the soil erosion cover-management factor at the European scale. Land Use
Policy 48, 38
–
50
.
Reeves, D.W., 1997. The role of soil organic matter in maintaining soil quality in con-
tinuous cropping systems. Soil Tillage Res. 43, 131
–
167
.
Reijneveld, A., van Wensem, J., Oenema, O., 2009. Soil organic carbon contents of
agricultural land in the Netherlands between 1984 and 2004. Geoderma 152,
231
–
238
.
Reynolds, B., Chamberlain, P.M., Poskitt, J., Woods, C., Scott, W.A., Rowe, E.C.,
Robinson, D.A., Frogbrook, Z.L., Keith, A.M., Henrys, P.A., Black, H.I.J., Emmett,
B.A., 2013. Countryside Survey: National
“
Soil Change
”
1978
–
2007 for Topsoils in
Great Britain
—
Acidity, Carbon, and Total Nitrogen Status. Vadose Zone J. 12
.
Silgram, M., Chambers, B.J., 2002. E
ff
ects of long-term straw management and fertilizer
nitrogen additions on soil nitrogen supply and crop yields at two sites in eastern
England. J. Agric. Sci. 139, 115
–
127
.
Stolte, J., Tesfai, M., Øygarden, L., Kværnø, S., Keizer, J., Verheijen, F., Panagos, P.,
Ballabio, C., Hessel, R., 2015. Soil Threats in Europe. Status, Methods, Drivers and
E
ff
ects on Ecosystem Services. Deliverable 2.1 of the RECARE Project. EUR 27607
EN, JRC Technical Reports. JRC, Luxembourg, pp. 207
.
Taghizadeh-Toosi, A., Olesen, J.E., Kristensen, K., Elsgaard, L., Østergaard, H.S.,
Lægdsmand, M., Greve, M.H., Christensen, B.T., 2014. Changes in carbon stocks of
Danish agricultural mineral soils between 1986 and 2009. Eur. J. Soil Sci. 65,
730
–
740
.
ten Berge, H.F.M., Schröder, J.J., Olesen, J.E., Giráldez Cervera, J.V., 2017. Preserving
Agricultural Soils in the EU, Research for AGRI Committee European Parliament.
Policy Department for Structural and Cohesion Policies, Brussels
.
Toth, G., Montanarella, L., Rusco, E., 2008. Threats to soil quality in Europe. EUR 23438
EN - 2008. JRC Scienti
fi
c and Technical Reports. JRC, Luxembourg
.
UNFCCC, 2015. Join the 4/1000 Initiative. Soils for Food Security and Climate, Lima-
Paris Action Agenda
.
Van den Putte, A., Govers, G., Diels, J., Gillijns, K., Demuzere, M., 2010. Assessing the
e
ff
ect of soil tillage on crop growth: a meta-regression analysis on European crop
yields under conservation agriculture. Eur. J. Agron. 33, 231
–
241
.
Viaene, J., Van Lancker, J., Vandecasteele, B., Willekens, K., Bijttebier, J., Ruysschaert,
G., De Neve, S., Reubens, B., 2016. Opportunities and barriers to on-farm composting
and compost application: a case study from northwestern Europe. Waste Manag. 48,
181
–
192
.
Watts, C.W., Dexter, A.R., 1997. The in
fl
uence of organic matter in reducing the desta-
bilization of soil by simulated tillage. Soil Tillage Res. 42, 253
–
275
.
Werner, M., Wauters, E., Bijttebier, J., Steinmann, H.-H., Ruysschaert, G., Knierim, A.,
2017. Farm level implementation of soil conservation measures: farmers
’
beliefs and
intentions. Renew. Agric. Food Syst. 1
–
14
.
R. Hijbeek et al.
Agriculture, Ecosystems and Environment 275 (2019) 42–53
53
Document Outline