Content uploaded by Guido Haas
Author content
All content in this area was uploaded by Guido Haas on Jan 19, 2022
Content may be subject to copyright.
Agriculture, Ecosystems and Environment 83 (2001) 43–53
Comparing intensive, extensified and organic grassland farming in
southern Germany by process life cycle assessment
Guido Haas∗, Frank Wetterich, Ulrich Köpke
Institute of Organic Agriculture, University of Bonn, Katzenburgweg 3, D-53115 Bonn, Germany
Received 8 October 1999; received in revised form 11 February 2000; accepted 24 March 2000
Abstract
To reduce the environmental burden of agriculture, suitable methods to comprehend and assess the impact on natural
resources are needed. One of the methods considered is the life cycle assessment (LCA) method, which was used to assess the
environmental impacts of 18 grassland farms in three different farming intensities — intensive, extensified, and organic — in
the Allgäu region in southern Germany. Extensified and organic compared with intensive farms could reduce negative effects
in the abiotic impact categories of energy use, global warming potential (GWP) and ground water mainly by renouncing
mineral nitrogen fertilizer. Energy consumption of intensive farms was 19.1GJha−1and 2.7 GJt−1milk, of extensified and
organic farms 8.7 and 5.9GJha−1along with 1.3 and 1.2 GJ t−1milk, respectively. Global warming potential was 9.4, 7.0
and 6.3 CO2-equivalentsha−1and 1.3, 1.0 and 1.3 CO2-equivalentst−1milk for the intensive, extensified and organic farms,
respectively. Acidification calculated in SO2-equivalents was high, but the extensified (119kg SO2ha−1) and the organic
farms (107 kg SO2ha−1) emit a lower amount compared with the intensive farms (136 kg SO2ha−1). Eutrophication potential
computed in PO4-equivalents was higher for intensive (54.2 kg PO4ha−1) compared with extensified (31.2kg PO4ha−1) and
organic farms (13.5kg PO4ha−1). Farmgate balances for N (80.1, 31.4 and 31.1kg ha−1) and P (5.3, 4.5 and −2.3kgha−1)
for intensive, extensified and organic farms, respectively, indicate the different impacts on ground and surface water quality.
Analysingthe impact categoriesbiodiversity,landscape image andanimal husbandry,organicfarms had clear advantagesin the
indicatorsnumberof grassland species, grazingcattle,layout of farmstead andherdmanagement, but indices in thesecategories
showed a wide range and are partly independent of the farming system. © 2001 Elsevier Science B.V. All rights reserved.
Keywords: Life cycle assessment; Agriculture; Intensive; Extensive; Organic; Grassland; Dairy farming
1. Introduction
Because intensification of agricultural production
processes has led to environmental burdens, discus-
sions about sustainable farming are taking place.
Agriculture today must be environmentally and eco-
logically sound and aligned with public values, e.g.
positive landscape image and appropriate animal wel-
∗Corresponding author. Tel.: +49-228-73-76-02;
fax: +49-228-73-56-17.
E-mail address: iol@uni-bonn.de (G. Haas).
fare. Converting conventional or intensive agriculture
to organic and extensive farming addresses these
concerns.
Efficient methods combining suitable indicators are
needed to comprehend and assess agricultural impacts
on the environment. Life cycle assessment (LCA) is
a method to compile a complete inventory, to evalu-
ate and to assess all relevant environmental impacts.
Initially developed for assessing the environmental
impact of industrial plants and production processes,
LCAs in agriculture have been mainly carried out for
single crops or production processes (Ceuterick, 1996,
0167-8809/01/$ – see front matter © 2001 Elsevier Science B.V. All rights reserved.
PII: S0167-8809(00)00160-2
44 G. Haas et al./Agriculture, Ecosystems and Environment 83 (2001) 43–53
1998; Wegener Sleeswijk et al., 1996; Audsley et al.,
1997).
The central objective of this study was to use
and adapt the LCA method for assessing all rele-
vant environmental impacts on the whole farm level
to compare different farming production systems. A
process-LCA of 18 grassland farms in three different
farming intensities — intensive, organic and exten-
sified (farmed intensively before, but compared to
international farming practice are still producing on
a high intensity level) — was carried out in 1998
(Wetterich and Haas, 1999).
2. Materials and methods
2.1. Farming in the region
The investigation took place near the city of
Kempten, in the Allgäu region, which is located south-
west of the German state Bavaria. The area is part of
a subalpine hilly region. It is a well-known region for
recreation and vacations offering many possibilities
for outdoor activities (e.g. hiking, swimming in lakes
and skiing). Good climate (annual mean temperature
is 6.9◦C and precipitation is 1274mm) and soil con-
ditions (FAO: Eutric Cambisols and Humic Fluvisols,
both partly Gleyic but usually drained) promote an
intensive use of the permanent grassland. Small dairy
farms with an average size of 20ha, 23 dairy cows
and an average annual milk performance of 6060kg
Table 1
Data for analyzed dairy farms of life cycle assessment in the Allgäu regiona
Farming intensity
Intensive Extensified Organic
Characteristics
Mineral N-fertilizing Yes No No
Purchasing fodder Yes Yes Limited
Share of farms in the region 43% 46% 6%
Farming intensity period (years) – 5 (2–7) 13 (3–20)
Farmed grassland area (ha) 32.7 (23–46) 34.7 (17–62) 25.8 (16–34)
Grassland yield (gross — without harvest losses) (t DMha−1) 11.8 (10.9–12.8) 10.5 (90–12.7) 10.7 (88–12.1)
Stocking rate (LUha−1)b2.2 (2.0–2.6) 1.9 (1.6–2.3) 1.9 (1.6–2.1)
Milk performance (annual) (kg per cow) 6758 a (5100–8050) 6390 ab (5500–7640) 5275 b (4800–5500)
aMean of farming system, range in brackets. Values followed by letters indicate mean significant difference (MSD) of 1148.5 kg per
cow; other parameters are not significant.
bLU: livestock-unit (each 500kg liveweight).
per cow are predominant. Grassland is cut for indoor
feeding, ensiling, grass drying or hay and grazed five
times a year.
After a pre-selection of 35 farms with local advi-
sors, six representative farms for each of the three
farming intensities (Table 1) according to the regional
agri-environmental program of Bavaria were selected
and analyzed in detail by practical investigation on
the farms and a farmer questionnaire, as well as by
consultations together with advisors and local experts
(e.g. water authorities).
The questionnaire covered all basic agricultural
production data about farm structure, main produc-
tion processes in detail, performance, yield, quality,
input and output massflow (e.g. fodder, straw, fertil-
izer, cattle, milk, diesel). Questions were answered by
the farmers by interviewing them during an intensive
farm visit of about 4h, on which main farm buildings
(particularly assessment of housing system and condi-
tion), layout of farmstead, equipment, machinery, and
infrastructure in general were examined. In following
visits starting at the beginning of May, 50–60% of
the main and characteristic grassland areas of each
farm were investigated for biodiversity and landscape
image before first cut.
2.2. Methodology of life cycle assessment
The LCA method is internationally standardized
(SETAC, 1993; ISO, 1997) but it needs to be specifi-
cally adapted for agriculture (Geier and Köpke, 1998;
G. Haas et al./Agriculture, Ecosystems and Environment 83 (2001) 43–53 45
Table 2
Impact categories and indicators of life cycle assessment in the Allgäu region
Impact category Environmental indicator
Resource consumption
Energy Use of primary energy
Minerals Use of P- and K-fertilizer
Global warming potential CO2-, CH4-, N2O-emission (in CO2-equivalents)
Soil function/strain
Grassland Accumulation of heavy metals
Of other ecosystems
(acidification, eutrophication) NH3-, NOx-, SO2-emission, N- and P-surplus (in SO2- and PO4-
equivalents)
Water quality
Ground water (nitrate content) N-fertilizing, N-farmgate-balance, potential of nitrate leaching
Surface water (P-eutrophication) P-fertilizing, P-balance, percentage of drained area
Human- and ecotoxicity Application of herbicide and antibiotics, potential of nitrate leach-
ing, NH3-emission
Biodiversity Grassland (number of species, date of first cut), hedges and field
margins (density, diversity, state/care, fences)
Landscape image (aesthetics) Grassland, hedges and field margins (see above), grazing animals
(period, breed, alpine cattle keeping), layout of farmstead (regional
type, buildings, farm garden, trees, orchard)
Animal husbandry (appropriate animal welfare) Housing system and conditions, herd management (e.g. lightness,
spacing, grazing season, care)
Geier, 2000; Haas et al., 2000). According to specific
agri-environmental indicators (OECD, 1997; Rudloff
et al., 1999) the impact on biodiversity, landscape im-
age and animal welfare, topics that have high public
awareness and are governed by the agri-environmental
policies of the European Union have to be taken into
account. Additionally site-specific and regional as-
pects (e.g. typical regional layout of the farmstead)
were included in the framework (Table 2).
The amount of fossil energy used in direct (e.g.
diesel, according to agricultural planning data; KTBL,
1994, 1997) and indirect (e.g. fertilizers) forms was
calculated based on the consumption of primary
energy factors for Germany (Patyk and Reinhardt,
1997). Energy need for grass drying (per t DM)
amounts to 100kWh electricity and 3200 kWh of
natural gas. Purchased fodder was calculated with
678.5MJ t−1fresh matter (average of the 18 analyzed
farms) and purchased concentrates 2500MJ t−1. The
assumptions of emitted climate relevant trace gases
were based on the data from Crutzen et al. (1986);
Boumann et al. (1991); Kirchgessner et al. (1991);
Gibbs and Woodbury (1993); Heyer (1994); Patyk and
Reinhardt (1997); Rück et al. (1997) and Mosier and
Kroeze (1998). The global warming potential (GWP)
was computed according to the CO2-equivalent fac-
tors by IPCC (1996) for CO2:1,CH
4: 21 and N2O:
310, period of time 100 years. The emission of am-
monia was assumed as 28% nitrogen of the stored
and applied farmyard manure (BLBP, 1997a, p. 12).
To calculate the acidification potential of the differ-
ent trace gases the SO2-equivalent factors for SO2:
1, NOx: 0.7 and NH3: 1.89 derived from Reinhardt
(1997) and for calculating the eutrophication poten-
tial the PO4-equivalent factors for NOx: 0.13, N: 0.42
and P 3.06 derived from Heijungs et al. (1992, p. 87)
were used.
The components of the farmgate balance for N and
P were the input by mineral fertilizers, the purchase
of fodder, straw and animals and the output of the
sold milk, animals, silage and hay. N2-fixation was
included in the farmgate N balance and assumed to
be 30kg N t−1(DM) of white clover Trifolium repens
L. after Weissbach (1997). The total yield portion
of white clover was assessed while investigating the
number of species in the grassland areas. The poten-
tial nitrate leaching balance per ha was approximated
by adding atmospheric deposition of 20kg N ha−1
(BLBP, 1997b, p. 93), deducting 28% ammonia losses
of the excrement-N and denitrification losses (calcu-
46 G. Haas et al./Agriculture, Ecosystems and Environment 83 (2001) 43–53
Table 3
Estimation of the impact category biodiversity of life cycle assessment in the Allgäu regiona
Index
54 3 2 1
Grassland
Number of species (flora) ≤22 23–25 26–28 29–31 ≥32
Time of first cut (after) 5 May 10 May 15 May 20 May 25 May
Hedges and field margins
Density (relative frequency) Low Average High
Diversity Low Average High
State/care Poor Average Very good
Fences None Medium density, small fences High density, broad fences
a1: very good; 3: average of region; 5: unsatisfactory.
lated as two-times of the N2O-N emission) of the
farm-gate balance surplus.
Estimation schemes based on self-defined criteria
and assumptions were used in the impact categories
biodiversity (Table 3), landscape image and animal
husbandry. The scientific basis for the indicators
of these impact categories were derived from the
much more detailed but time-consuming methods by
Frieben (1998) for biodiversity and for animal welfare
by Sundrum et al. (1994) and Sundrum (1997). Index
3 was defined as the estimated typical and character-
istic average of all farms in the region, whereas the
range was created by the random sample of the 18
analyzed farms.
Beyond the biotic elements the existence of fences
was evaluated in the impact category of biodiversity
(Table 3), because they create small biotopes; plant
species and fauna were detected, which in a permanent
grassland region only exist underneath a fence. Land-
scape image in the Allgäu region is mainly influenced
by agriculture. Diversity of grassland, especially
flowering plants, as well as structured grassland areas
through hedges and fences cause a pleasant impres-
sion for the people. Care and layout of the farmsteads,
which are usually exposed single locations or groups
of farms in the countryside, and the grazing cattle
were evaluated as characteristic elements of the region
in the landscape image category (Table 2). The pres-
ence of the grazing cattle, and therefore the fences,
depends on the length of the grazing period and typ-
ical look (breed, horned, cow bells). Alpine young
cattle keeping in the summer time was estimated as a
positive indicator, because it ensures the characteristic
alpine mountain meadows in southern Germany.
The farmed area was the geographical coverage.
Considerations were restricted to the year 1997; single
or rare events (for example each 10 years) compared
with other years were excluded. Most impacts were
referenced to the LCA functional unit ha of farmed
grassland. Some abiotic categories also were related
to the produced units of milk (in kg) (see Haas et al.,
2000). The functional unit for the categories of bio-
diversity, landscape image and animal husbandry was
the whole farm.
The impact on human- and eco-toxicity, accumula-
tion of heavy metals and the consumption of mineral
resources as a result of agriculture in the Allgäu
region are low. Therefore, the methods used and re-
sults achieved in these categories are not presented.
In case of no substantial deviation from the normal
distribution and homogeneity of variance, the results
for the abiotic impact categories were tested for sig-
nificance using the PROC GLM procedure for the
analysis of variance with a completely randomized
design of three farming systems with six replications
(farms) by using the SAS statistical package (SAS
Institute Inc., Cary, NC, 1996). Differences between
the means (MSD) were analyzed using the Tukey test
at the alpha 5% level.
3. Results and discussion
3.1. Resource consumption
Intensive farms consume a significant higher
amount of fossil energy caused by the use of grass
drying industrial plants and mineral N-fertilizer
G. Haas et al./Agriculture, Ecosystems and Environment 83 (2001) 43–53 47
Table 4
Inventory of the impact category primary energy consumption of life cycle assessment in the Allgäu regiona
Impact category/indicator Intensive Extensified Organic MSD5%
Fuel and lubricants for grassland farming (GJha−1) 4.482 4.117 3.439 n.s.
Hay drying (indoor) (GJha−1) 0.721 0.320 0.966 n.s.
Grass drying (in industrial plants) (GJha−1) 6.391 a 0.306 b 0.745 b 5.21
Mineral fertilizer (GJha−1) 3.674 a 0.194 b 0 b 1.08
Purchased fodder (GJha−1) 3.836 a 3.724 a 0.790 b 2.88
Energy consumption
Primary energy (area-related) (GJha−1) 19.1 a (10.4–28.7) 8.7 b (5.5–12.2) 5.9 b (3.8–10.6) 7.23
Primary energy (product-related: t milk) (GJt−1) 2.7 a (1.6–3.9) 1.3 b (1.0–1.6) 1.2 b (0.8–1.8) 0.98
aMean and range (in brackets) of farming system. Differences between the means (MSD) were tested using the Tukey test at the alpha
5% level indicated by different letters or as not significant (n.s.).
(Table 4). In contrast, organic farms need only
one-third of area-related energy input and only half of
the product-related energy that intensive farms need.
Besides fuel and lubricants for grassland farming,
additional energy was calculated to compute the total
amount of fossil energy needed, used either for hay
drying (electricity for ventilators, fuel oil for heating
the ventilation air) or for grass drying in small local
industrial plants, managed by farmers’ cooperatives,
producing grass pellets (electricity, natural gas and
diesel). With both drying processes high fodder qual-
ities for high performing dairy cows are achieved.
Five intensive and only one extensified and organic
farm perform grass drying.
Compared with the intensive farms a reduction
of 55 and 69% area-related along with 52 and 56%
product-related energy use for the extensified and
organic farms, respectively, is realized. Because of
lower fodder purchases the organic farms need less
area-related energy compared with extensified farms.
Conventional farms in Germany use approximately
19.4GJ ha−1and organic farms around 6.8GJ ha−1
(65% less, Haas et al., 1995). For mixed farming in the
Hamburg region conventional farms uses 16.3 GJha−1
and organic farms 6.8GJ ha−1(58% less, according
Geier et al., 1998). Although the methods and fig-
ures to quantify the energy used in these studies dif-
fer slightly, the energy use in the Allgäu region with
permanent grassland farming is high compared with
mainly arable farming in the Hamburg region and for
German farms on average. The product-related energy
use calculated by Cederberg and Mattsson (1998) for
Swedish dairy farms is 2.85 and 2.4GJ t−1milk for
conventional and organic farming, respectively, which
is twice as much energy as the organic farms in the
Allgäu region (1.2GJ t−1) need, but different calcula-
tion methods were used.
3.2. Global warming potential
Differences in CO2-emissions are caused by
these different uses of fossil energy. The CH4- and
N2O-emissions are comparably low, but due to the
high GWP of these trace gases their climate relevance
is much higher (Table 5). The area-related GWP
decreases for intensive (9.4t CO2ha−1), extensified
(7.0t CO2ha−1) and organic farms (6.3 t CO2ha−1),
accordingly. For product-related energy use the ex-
tensified farms (1.0t CO2ha−1) cause the significant
lowest GWP, whereas intensified and organic farms
(1.3t CO2ha−1) have the same emissions. Lower
CO2- and N2O-emissions of the organic farms are
compensated by a higher emission of CH4per unit of
produced milk because of lower milk performance.
3.3. Soil functions: eutrophication and acidification
In the Allgäu region eutrophication and acidifi-
cation stress the forest and fen soils and their re-
lated ecosystems. Acidification is almost exclusively
caused by ammonia emission from the cattle keeping
(Table 6). Because of the lower stocking rate and a
lower N-excretion (milk production is lower) the ex-
tensified (119 kg SO2ha−1) and especially the organic
farms (107kg SO2ha−1) emit a lower amount of
ammonia compared with the intensive farms (136kg
SO2ha−1), although it is still too much for sensitive
ecosystems in the region. Eutrophication potential
48 G. Haas et al./Agriculture, Ecosystems and Environment 83 (2001) 43–53
Table 5
Inventory of the impact category global warming potential of life cycle assessment in the Allgäu regiona,b
Impact category/indicator Intensive Extensified Organic MSD5%
CO2-emission 1.280 a 0.666 b 0.428 b 0.45
CH4-emission 5.102 a 4.535 ab 4.114 b 0.77
N2O-emission 3.017 a 1.808 b 1.776 b 0.65
Global warming potential
Area-related (tha−1) 9.4 a (7.5–11.2) 7.0 b (5.7–8.0) 6.3 b (5.6–7.3) 1.66
Product-related (tt−1milk) 1.3 a (1.1–1.7) 1.0 b (0.9–1.2) 1.3 a (1.2–1.4) 0.22
aIntCO
2-equivalents; mean and range (in brackets) of farming system.
bDifferences between the means (MSD) were tested using the Tukey test at the alpha 5% level indicated by different letters.
computed in PO4-equivalents is mainly indicated by
the N- and P-surplus (Table 6). It is significantly
higher for intensive (54.2kg PO4ha−1) compared
with organic farms (13.5kg PO4ha−1), extensified
farms emit 31.2kg PO4ha−1.
3.4. Water quality: N- and P-balances
With farmyard manure (mainly slurry) an average
of 144kg N ha−1is applied by intensive farms com-
pared to 128 and 117kg N ha−1by extensified and or-
ganic farms, respectively (28% of ammonia losses are
already extracted). Intensive farms use 68kg Nha−1
of mineral fertilizer additionally. The average portion
of white clover was 5.8, 10 and 15.5% for intensive,
extensified and organic farms, respectively.
Table 6
Inventory of the impact category eutrophication and acidification of life cycle assessment in the Allgäu regiona,b
Impact category/indicator Intensive Extensified Organic MSD5%
Acidification (in SO2-equivalents)
SO2-emission 1.1 a 0.7 b 0.3 c 0.31
NOx-emission 6.1 a 4.6 a 2.6 b 1.94
NH3-emission 129 a 113 ab 104 b 21.68
Sumb136 a (119–145) 119 ab (96–143) 107 b (94–118) 23.01
Eutrophication (in PO4-equivalents)
Nox-emission 1.13 a 0.86 a 0.48 b 0.36
N (farmgate balance)c33.6 a 13.4 b 13.1 b 15.0
P (farmgate balance)c19.5 16.9 0 n.s.
Sumb54.2 a (17.8–90.1) 31.2 ab (0.6–48.6) 13.5 b (7.4–19.0) 28.3
aIn kgha−1; mean and range (in brackets) of farming system.
bDifferences between the means (MSD) were tested using the Tukey test at the alpha 5% level indicated by different letters or as not
significant (n.s.).
cCalculated eutrophication potential for N and P are based on the farmgate balances for these elements (see Tables 7 and 8), but only
positive N and P farm-balances were considered.
Farmgate balances for intensive farms result in av-
erage 80 and 31kg N ha−1for extensified and organic
farms (Table 7). Only as a rough calculated figure the
nitrate leaching potential is 36kg N ha−1for intensive
farms, whereas on average no nitrate leaching can be
stated for the other farming intensities. In the Allgäu
region the nitrate content in the ground water is gen-
erally low (around 10–25mg nitrate l−1). The soil and
subsoil conditions (humic) are considered to have a
high denitrification potential.
The amount of total P-fertilizing is 34.6, 30.9 and
23.2kg P ha−1and P-farmgate balance is 5.3, 4.5
and −2.3kg P ha−1for the intensive, extensified and
organic farms, respectively (Table 8). In an intensive
case study Pommer et al. (1997) and Neyer (1999)
quantified a tolerable P-input into a lake in the region
G. Haas et al./Agriculture, Ecosystems and Environment 83 (2001) 43–53 49
Table 7
Inventory of the impact category ground water quality (N-balance) of life cycle assessment in the Allgäu regiona
Indicator Intensive Extensified Organic MSD5%
A N-fertilizer 68.1 (35.5–100.7) 0 0
B Purchased fodder, straw and cattle 39.5 ab (23.5–65.2) 45.0 a (9.2–85.6) 11.9 b (2.3–19.6) 28.74
C Symbiotic N2fixation 20.3 b (9.0–28.0) 31.7 b (20.0–48.0) 50.2 a (28.0–62.0) 14.92
D N-export 47.8 a (37.3–62.5) 45.3 a (39.0–56.8) 31.0 b (25.4–38.4) 11.07
T1N-farmgate balanceb80.1 a (40.4–115) 31.4 b (−3.8–66.2) 31.1 b (16.2–44.2) 36.21
E Atmospheric deposition 20.0 20.0 20.0 –
F Ammonia losses 55.9 (48.6–63.4) 49.6 (40.4–60.1) 45.4 (39.6–50.5) n.c.
G Denitrification 8.1 (6.9–9.1) 6.0 (4.8–7.4) 6.2 (5.0–7.0) n.c.
T2Potential NO3-N-leachingc36.0 a (2.9–63.5) 4.3 b (−37.4–25.8) −0.5b(−9.2–6.8) 31.59
Total N-fertilizingd(mineral fertilizer and slurry) 212.5 a (166–255) 128.0 b (104–157) 116.7 b (102–130) 36.21
aIn kg Nha−1; mean and range (in brackets) of farming system. Differences between the means (MSD) were tested using the Tukey
test at the alpha 5% level indicated by different letters; not computed (n.c.) because calculations are based on rough figures.
bN-farmgate balance: T1=A+B+C−D.
cNegative values for potential nitrate leaching are not very meaningful, but do show variation and differences between the systems
(T2=T1+E−F−G).
dValues for the total amount of fertilizing only serve as an additional indicator that is not part of the farmgate balance.
of about 0.41kg P related to 1 ha farmed grassland in
the catchment area, whereas the measured input was
1.6kg P ha−1, causing clear P-eutrophication, which
is stated for several lakes with grassland farming in the
catchment area in the Allgäu region. Besides the use
of mineral P-fertilizer, the P-import of purchased fod-
der — straw and cattle, amount to 1kg P ha−1max-
imum — in intensive (8.2kg Pha−1) and extensified
farms (8.5kg P ha−1) doubles significantly compared
with the organic farms (3.8kg P ha−1; Table 8), where
additional purchase of fodder is limited.
Table 8
Inventory of the impact category surface water quality (P-balance) of life cycle assessment in the Allgäu regiona
Indicator Intensive Extensified Organic MSD5%
A Mineral P-fertilizer 5.4 (0–18.0) 4.6 (0–15.4) 0
B Purchased fodder, straw and cattle 8.2 ab (4.8–12.2) 8.4 a (1.9–14.2) 3.8 b (2.9–6.1) 4.41
C P-export 8.3 a (7.1–9.9) 8.5 a (7.3–10.5) 6.1 b (5.0–7.6) 1.72
T3P-farmgate balanceb5.3 (−3.3–17.1) 4.5 (−5.4–12.7) −2.3 (–4.6–0.5) n.s.
Total P-fertilizingc(mineral fertilizer and slurry) 34.6 a (27.0–48.6) 30.9 ab (23.5–40.7) 23.2 b (20.0–26.0) 8.90
aIn kg Pha−1; mean and range of farming system. Differences between the means (MSD) were tested using the Tukey test at the
alpha 5% level indicated by different letters or as not significant (n.s.).
bP-farmgate balance: T3=A+B−C.
cValues for the total amount of fertilizing only serve as an additional indicator that is not part of the farmgate balance.
3.5. Biodiversity
In general biodiversity of grassland species on all
farms is low compared with results gained 30–40 years
ago (Kohler et al., 1989; Abt, 1991; Schulz, 1992).
Currently the number of plant species in organic and
extensified grassland is slightly higher (29.0 and 26.8
in average of all main fields per farm, respectively)
compared with the intensively used permanent grass-
land (24.7), a result confirmed by Frieben and Köpke
(1996). One of the organic farms that converted only
50 G. Haas et al./Agriculture, Ecosystems and Environment 83 (2001) 43–53
Table 9
Inventory of the impact category biodiversity, landscape image and animal husbandry of life cycle assessment in the Allgäu regiona
Impact category/indicator Intensive Extensified Organic
Biodiversity 3.7 3.3 2.4
Grassland 3.6 (3.0–4.0) 3.5 (2.5–5.0) 2.1 (1.5–4.0)
Hedges and field margins 3.8 (1.5–4.8) 3.0 (1.3–4.0) 2.7 (1.3–3.8)
Landscape image (partly including biodiversity) 3.2 3.0 2.0
Grazing cattle 2.7 (1.8–3.5) 3.0 (1.3–5.0) 1.8 (1.0–2.5)
Farmstead 2.6 (1.0–3.8) 2.5 (1.8–3.3) 1.5 (1.0–2.5)
Animal husbandry 3.1 3.3 2.0
Housing system 3.2 (2.1–3.6) 3.3 (1.7–4.8) 2.5 (1.8–3.4)
Herd management 3.0 (2.0–4.0) 3.3 (1.8–4.5) 1.5 (1.0–2.3)
aEstimation index: 1: very good, 3: average of the region, 5: unsatisfactory; mean of farming system, range in brackets.
three years ago to organic agriculture has a very low
number of species (23) compared to the rest of the or-
ganic farms (on average 30.2 plant species), as well as
a very early first cut (around 15 May). The earliest first
cut is carried out by extensified farms between 5 and
15 May, whereas intensified farms cut the first time
in spring for silage around 10 May until 20 May. The
organic grassland is cut around 20 May until 25 May
(except one farm, see above). Cutting for grass drying
by the intensive farms of about an average of 6 ha each
starts around the beginning of May. Hedges and field
margins are primarily influenced by the farmers’ atti-
tude (e.g. personality, ecological understanding), indi-
cated by the wide range of the indices. The presence
of these indicators is predominantly independent of
the farming system (Table 9).
3.6. Landscape image
In the impact category of landscape image the exten-
sified farms show the shortest and the organic farms,
the longest grazing season. Only the five organic farms
do not dehorn the calves. No differences for breed (i.e.
share of Brown Swiss) and alpine cattle keeping were
found between the farming intensities. Nevertheless,
there were some differences between the farms. The
more positive scoring of organic farms for the indica-
tor farmstead (Table 9) could not be explained by aim-
ing for an attractive farmstead for direct marketing.
Although there was usually one farm in each farming
system that was exemplary for at least one indicator
in the landscape image impact category, organic farms
on average show a higher scoring (Table 9).
3.7. Animal husbandry
Animal housing systems also are slightly more
positively scored in organic farms whereas herd man-
agement of the organic farms is clearly more posi-
tive compared to extensified and intensive farming
(Table 9). In extensified farms grazing is strongly
reduced or all year inside keeping and feeding is
practised, which was negatively scored. More appro-
priate housing systems would result in a more positive
scoring for all farming intensities.
3.8. Impact of farming systems
Compared with extensified or organic farms, inten-
sive farms show negative environmental impacts in
the impact categories of energy consumption, global
warming potential, soil functions of other ecosystems
(acidification), surface water (P-eutrophication), bio-
diversity and landscape image (Fig. 1). Extensified
and organic farms reduce these negative effects in
the energy consumption, global warming potential
and ground water impact categories, because they
renounce mineral nitrogen fertilizer. The impact on
animal husbandry of extensified farms is slightly
more negative compared with the intensive farms.
Because fodder production is not as high as by inten-
sive farming, extensified farms try to compensate by
intensifying other production processes (e.g. earliest
first cut of grassland, shortest or no grazing season to
avoid grazing losses from indoor feeding).
Organic farms have clear positive or comparatively
fewer negative effects on surface water, biodiversity,
G. Haas et al./Agriculture, Ecosystems and Environment 83 (2001) 43–53 51
Fig. 1. Inventory (schematic) of selected impact categories and indicators of life cycle assessment of the farming systems intensive,
extensified and organic in the Allgäu region. Netline outside (=100%, positive)/centre of net (=0%, negative): estimation indices=1/5;
energy use=3.5/30GJha−1;CO
2-equivalentsha−1=5.5/12t; SO2-equivalentsha−1=90/160 kg; farmgate surplus=20/120kg N ha−1(ground
water) and −5/20kg Pha−1(surface water).
landscape image and animal husbandry. P-eutrophica-
tion of the lakes, diminishing biodiversity and char-
acter of the landscape are the central environmental
problems in the region. The impact in these categories
does not directly relate to the stocking rate or the use
of mineral fertilizers. Organic agriculture shows inher-
ent ecological advantages of the production system,
which in most of the indicators were significant com-
pared with intensive farming.
4. Conclusions
Using the process-LCA method differences among
the agricultural production intensities according to
their environmental impact were identified and com-
prehensively evaluated for their various effects on the
environment. Therefore, LCA can be an important
and efficient tool for ecological weak-point analyzes
for farmers and advisors (Haas and Wetterich, 1999)
as well as for politicians creating agri-environmental
programs (Haas and Wetterich, 2000). Efficient mea-
sures can be derived to establish an environmentally
sound agricultural production system.
The study confirmed the suitability of LCA for
comparing farms and farming systems, but further
development of the LCA-methodology in agricul-
ture is required. In the future the basis of evaluation
should be reference data, limiting values, critical load
limits or goals of environmental quality, if reliable
data exist. The estimations made in the biotic and
aesthetic subranges are more or less subjective, al-
though the determined differentiation and order of
rank can be reproduced. The definitions and clas-
sifications chosen need to be standardized. Experts
and local people should achieve consensus if further
LCAs on a broader base will be undertaken in the
region. This is obvious for the criteria and evaluation
in the impact category of landscape image. However,
because several impact categories and indicators must
be monitored, only pragmatic and resource efficient
approaches can be used.
Acknowledgements
The authors wish to thank Dr. Uwe Geier and Mrs.
Denise Short for critical review and proof reading of
the manuscript.
References
Abt, K., 1991. Qualitative und quantitative Aspekte des Agrars-
trukturwandels im württembergischen Allgäu — Konsequenzen
52 G. Haas et al./Agriculture, Ecosystems and Environment 83 (2001) 43–53
für Natur- und Umweltschutz. Verhandlungen der Gesellschaft
für Ökologie 20, 375–383.
Audsley, A., Alber, S., Clift, R., Cowell, S., Crettaz, R., Gaillard,
G., Hausheer, J., Jolliett, O., Kleijn, R., Mortensen, B., Pearce,
D., Roger, E., Teulon, H., Weidema, B., van Zeijts, H.,
1997. Harmonisation of environmental life cycle assessment for
agriculture. Final report, Concerted Action AIR3-CT94-2028,
European Commission DG VI, Brussels, Belgium.
BLBP (Bayerische Landesanstalt für Bodenkultur und Pflan-
zenbau), 1997a. Datenvorgabe (Basisdaten) für die Erstellung
des Vergleiches der Nährstoffzu- und -abfuhren nach §5
Düngeverordnung und Handformular zur Berechnung der
Nährstoffzu- und -abfuhren, Stand Februar 1997, Eigenverlag,
Freising/München, Germany.
BLBP (Bayerische Landesanstalt für Bodenkultur und Pflan-
zenbau), 1997b. Boden-Dauerbeobachtungs-Flächen, Bericht
nach 10jähriger Laufzeit 1985–1995, Teil II. Schriftenreihe der
Bayerischen Landesanstalt für Bodenkultur und Pflanzenbau
May 1997, Eigenverlag, Freising/München, Germany.
Boumann, A.F., Van den Born, G.J., Swart, R.J., 1991. Land
use related sources of CH4and N2O. Expert report document
12/1-a, 25–26 November 1991, Enquete-Commission of the
German parliament on Protecting the Atmosphere of the Earth,
Bonn, Germany, pp. 207–267.
Cederberg, C., Mattsson, B., 1998. Life cycle assessment of
Swedish milk production — a comparison of conventional
and organic farming. In: Ceuterick, D. (Ed.), Proceedings of
the International Conference on Life Cycle Assessment in
Agriculture, Agro-Industry and Forestry, Brussels, Belgium,
3–4 December 1998, pp. 161–167.
Ceuterick, D. (Ed.), 1996. In: Proceedings of the International
Conference on Application of Life Cycle Assessment in
Agriculture, Food and Non-Food Agro-Industry and Forestry,
Brussels, Belgium, 4–5 April 1996.
Ceuterick, D. (Ed.), 1998. In: Proceedings of the International
Conference on Life Cycle Assessment in Agriculture,
Agro-Industry and Forestry, Brussels, Belgium, 3–4 December
1998.
Crutzen, P.J., Aselmann, I., Seiler, W., 1986. Methane production
by domestic animals, wild ruminants, other herbivorous fauna,
and humans. Tellus 38B, 271–284.
Frieben, B., 1998. Verfahren zur Bestandsaufnahme und Bewertung
von Betrieben des Organischen Landbaus im Hinblick auf
Biotop- und Artenschutz und die Stabilisierung des Agrarö-
kosystems. Ph.D. Thesis, Verlag Dr. Köster, Berlin, Germany.
Frieben, B., Köpke, U., 1996. Effects of farming systems on
biodiversity. In: Isart, J., Llerena, J.J. (Eds.), Proceedings of
ENOF (European Network on Research in Organic Farming)
Workshop on Biodiversity and Land Use: The Role of Organic
Farming, Bonn, Germany, pp. 11–21.
Geier, U., 2000. Anwendung der Ökobilanzmethode in der Land-
wirtschaft — dargestellt am Beispiel einer Prozess-Ökobilanz
konventioneller und organischer Bewirtschaftung. Ph.D. Thesis,
Verlag Dr. Köster, Berlin, Germany.
Geier, U., Köpke, U., 1998. Comparison of conventional and
organic farming by process-life cycle assessment — a case study
of agriculture in Hamburg. In: Ceuterick, D. (Ed.), Proceedings
of the International Conference on Life Cycle Assessment in
Agriculture, Agro-Industry and Forestry, Brussels, Belgium,
3–4 December 1998, pp. 31–38.
Geier, U., Frieben, B., Haas, G., Molkenthin, V., Köpke, U.,
1998. Ökobilanz Hamburger Landwirtschaft — Umweltrelevanz
verschiedener Produktionsweisen, Handlungsfelder Hamburger
Umweltpolitik. Verlag Dr. Köster, Berlin, Germany.
Gibbs, M.J.J., Woodbury, W., 1993. Methane emissions from live-
stock manure. In: Van Amstel, A.R. (Ed.), Methane and Nitrous
Oxide. RIVM Report, National Institute of Public Health
and Environmental Protection, Bilthoven, The Netherlands,
pp. 81–91.
Haas, G., Wetterich, F., 1999. Ökobilanz der Umweltwirkung
landwirtschaftlicher Betriebe im Allgäu. Z. für Angew.
Umweltforschung 12, 368–377.
Haas, G., Wetterich, F., 2000. Agrarumweltprogramm mit
Ökobilanz im Allgäu zielorientiert gestalten. Berichte über
Landwirtschaft 78, 92–105.
Haas, G., Geier, U., Schulz, D.G., Köpke, U., 1995. A comparison
of conventional and organic agriculture. Part I. Climate relevant
carbon dioxide emission from the use of fossil energy. Berichte
über Landwirtschaft 73, 401–415 (in German).
Haas, G., Wetterich, F., Geier, U., 2000. Life cycle assessment
framework in agriculture on the farm level. J. LCA, in press.
Heijungs, R., Guinée, J.B., Huppes, G., Lankreijer, R.M., Udo de
Haes, H.A., Wegener Sleeswijk, A., 1992. Environmental life
cycle assessment of products. Centre of Environmental Science,
Leiden, The Netherlands.
Heyer, J., 1994. Methan. Studie im Auftrag der Enquetekommi-
ssion des Deutschen Bundestages Schutz der Erdatmosphäre.
Studienprogramm, Band 1, Landwirtschaft, Teilband I.
Economica Verlag, Bonn, Germany.
IPCC (Intergovernmental Panel on Climate Change), 1996. Climate
change 1995 — the science of climate change. University Press,
Cambridge, UK.
ISO (International Standardization Organization), 1997. Environ-
mental management — life cycle assessment — principles and
framework — ISO-norm 14040:1997. European Committee for
Standardization, Brussels, Belgium.
Kirchgessner, M., Windisch, W., Müller, H.L., Kreuzer, M., 1991.
Release of methane and of carbon dioxide by dairy cattle.
Agribiol. Res. 44 (2/3), 91–102.
Kohler, A., Abt, K., Zelesny, H., 1989. Das Grünlandgebiet des
Allgäu aus der Sicht der Landschaftsökologie. Informationen
für die Landwirtschaftsberatung in Baden-Württemberg Nr. 6,
49–71.
KTBL (Kuratorium für Technik und Bauwesen in der Land-
wirtschaft), 1994. KTBL-Taschenbuch Landwirtschaft 1994/95,
17th Edition, Germany.
KTBL (Kuratorium für Technik und Bauwesen in der
Landwirtschaft, Ed.), 1997. Datensammlung Betriebsplanung
1997/98, 15th Edition, Germany.
Mosier, A., Kroeze, C., 1998. A new approach to estimate emi-
ssions of nitrous oxide from agriculture and its implications
for the global N2O budget. Newsletter of the International
Geosphere–Biosphere Program (IGBP) No. 34, pp. 8–13.
G. Haas et al./Agriculture, Ecosystems and Environment 83 (2001) 43–53 53
Neyer, H., 1999. Influence of landuse on the phosphorus export
from grassland. Zeitschrift für Kulturtechnik und Landen-
twicklung 40, 49–52 (in German).
OECD (Organisation for Economic Co-operation and
Development), 1997. Environmental Indicators for Agriculture.
Paris, France.
Patyk, A., Reinhardt, G.A., 1997. Düngemittel — Energie- und
Stoffstrombilanzen. Verlag Vieweg, Braunschweig, Wiesbaden,
Germany.
Pommer, G., Neyer, H., Jordan, F., 1997. Phosphorus burden of a
prealpine lake with grasslands in the catchment basin. Agribiol.
Res. 50 (4), 325–332 (in German).
Reinhardt, G.A., 1997. Bilanzen über die gesamten Lebenswege.
In: Kaltschmitt, M., Reinhardt, G.A. (Hrsg.), Nachwachsende
Energieträger — Grundlagen, Verfahren, ökologische Bilanz-
ierung. Verlag Vieweg, Braunschweig, Wiesbaden, Germany,
pp. 84–95.
Rück, F., Homevo-Agossa, C., Stahr, K., 1997. Significance
of gaseous losses on the nitrogen balance in subalpine
grasslands. In: Jarvis, S.C., Pain, B.F. (Eds.), Gaseous Nitrogen
Emissions from Grasslands. CAB International, Oxon, New
York, pp. 73–75.
Rudloff, B., Geier, U., Meudt, M., Schick, H.-P., Urfei, G., 1999.
Development of indicators for the assessment of agricultural
impacts on the environment. Federal Environmental Agency,
Texte 42/99, Berlin, Germany, 252 pp. (in German).
Schulz, H., 1992. Langjährige Veränderungen der Grünland-
bestände im Württembergischen Allgäu. In: AG Grünland und
Futterbau in der Gesellschaft für Pflanzenbauwissenschaften,
36. Jahrestagung, Stuttgart-Hohenheim, Tagungsband, Ger-
many, 27–29 August 1992, pp. 14–24.
SETAC (Society of Environmental Toxicology and Chemistry),
1993. Guidelines for Life-Cycle Assessment: A Code of
Practice. Brussels, Belgium.
Sundrum, A., 1997. Assessing housing conditions in terms of
animal welfare — possibilities and limitations. In: Sorensen,
J.T. (Ed.), Proceedings on Livestock Farming Systems — More
Than Food Production, Foulum, Denmark, EAAP (European
Association for Animal Production) Publication No. 89,
pp. 238–246.
Sundrum, A., Andersson, R., Postler, G. (Eds.), 1994. Tierger-
echtheitsindex — 200 1994 — ein Leitfaden zur Beurteilung
von Haltungssystemen. Verlag Köllen, Bonn, Germany.
Wegener Sleeswijk, A., Kleijn, R., van Zeijts, H., Reus,
J.A.W.A., Meeusen van Onna, M.J.G., Lenemann, H., Sengers,
H.H.W.J.M., 1996. Application of LCA to agricultural
products. Report 130, Centre of Environmental Science, Leiden
University, Leiden, The Netherlands.
Weissbach, F., 1997. Zur Schätzung der Stickstoffbindung
durch Weissklee auf Dauergrünland. In: AG Grünland und
Futterbau in der Gesellschaft für Pflanzenbauwissenschaften,
41. Jahrestagung, Aulendorf, Tagungsband, Germany, 28–30
August 1997, pp. 250–253.
Wetterich, F., Haas, G., 1999. Ökobilanz Allgäuer
Grünlandbetriebe — Intensiv, Extensiviert, Ökologisch. Verlag
Dr. Köster, Berlin, Germany.
