Coefficients for N and P content of cereals and temporary
grasses for use in the Swedish nutrient balances
Ylva, Andrist Rangel1,*
Jan, Eriksson2
Karin, Fägerlind1
Josefine, Liew1
Gerda, Ländell 1
Gunnel, Wahlstedt1
1
Statistics Sweden
SE-701 89
Örebro, Sweden
*[email protected]
2
Swedish University of Agricultural Sciences
Department of Soil and Environment
P.O. Box 7014
SE-750 07
Uppsala, Sweden
ABSTRACT
Gross Nutrient Balances (GNB) for nitrogen (N) and phosphorous (P) are key agri-environmental
indicators (AEI) within the EU and are relevant to at least one of the indicators for agricultural
statistics as defined in the Global Strategy. Therefore, there is an increased need to obtain
harmonised statistics related to crop nutrients. Statistics Sweden has calculated and published
national soil surface nutrient balances intermittently since 1991. In 2003, the method was partly
adjusted to follow the one recommended by the OECD and Eurostat to increase comparability
between countries. In spite of changes in crop breeding and farm management, the coefficients for
N and P content in crops and crop residues have not been revised. The objective of this action is to
obtain updated and well-defined coefficients for N and P content in harvested cereals, temporary
grasses and crop residues and to estimate the amount of crop residues in Sweden. The N and P
concentrations in grain and straw of cereals used in the official Swedish nutrient balances have been
compared with concentrations reported in scientific literature for crops grown in Sweden or under
conditions similar to those in Sweden as regards climate, soil type, production systems,
management practices, etc. Comparisons have also been made with coefficients reported from the
Swedish national environmental monitoring programme, and the Swedish variety trials as well as
with coefficients used in other calculations, e.g. the OECD/Eurostat GNB, the national greenhouse
gas inventory and the data modelling of N and P leaching from Swedish arable soils. N and P
concentrations have been determined by analysing samples collected as part of the Swedish national
environmental monitoring programme. For estimation of crop residues removed from the field, a
sample survey of 3 000 agricultural holdings was conducted during the autumn/winter 2012/2013.
A sensitivity analysis of the effect of using a range of different N and P coefficients for harvested
products and crop residues on the Swedish GNB has been conducted. Results from the action show
that for grains of the three main cereal crops in Sweden, i.e. winter wheat (Triticum aestivum),
spring barley (Hordeum vulgare) and oats (Avena sativa), there are discrepancies in the coefficients
used in the Swedish nutrient balances and what is reported and/or used by others. The N and P
content in crop residues of these three cereal crops varied considerably. For example, N in straw of
winter wheat and oats used in the Swedish nutrient balances were in line with the concentrations
used in the OECD/Eurostat GNB and by the Swedish Board of Agriculture. However, they were
higher than those used by the Swedish Environmental Protection Agency for the greenhouse gas
inventory. The sensitivity analysis provides a basis for decisions on which variables it is worth
investing resources in to obtain more precise coefficients and hence a more accurate estimation of
the output. The outcome of this action can serve as a basis for recommendations on sound
methodologies to obtain better practices in terms of obtaining N and P crop content coefficients,
considering factors such as scientific ground, cost-efficiency and practical feasibility. Ideally,
harmonised coefficients should be used by all reporting systems within each country. Differences in
coefficients between countries should be supported by well-documented and confirmed evidence of
differences in crop products in terms of N and P content.
Keywords: Nutrient balance, GNB, Nitrogen, Phosphorous, Crop, Ley, Straw, Crop residues,
Sensitivity analysis
2
1. Introduction
Nutrient balances, i.e. a book-keeping of nutrient inputs and outputs to agricultural land, have
in later years become increasingly important as an indicator of the environmental performance and
leaching potential of farming management on different scales (Oenema et al., 2003). From first
being used as a tool to understand nutrient cycling, nutrient balances are now being used as a policy
instrument, and to evaluate how well environmental targets are met (Öborn et al., 2003). As
agriculture has been estimated to contribute 70-90 percent and 60-80 percent of the total load of
anthropogenic diffuse losses of nitrogen (N) and phosphorus (P), respectively, into the Baltic Sea
(HELCOM, 2011), there is an increased focus on actions to reduce the losses of nutrients from
agricultural land.
There are three basic types of nutrient balances (Oenema et al., 2003; Öborn et al., 2003; Hoang &
Alauddin, 2010):
(i)
(ii)
(iii)
Farm gate or black-box balance, which records nutrients of all kinds of products that enter
and leave the farm via the farm gate (note that the “farm gate” can here be the borders of a
region or a country).
Soil surface balance, which counts all nutrients that enter the soil via the surface and leave
the soil via crop uptake.
Soil system balance, which includes all nutrient inputs and outputs, including gains and
losses within and from the soil.
Gross Nutrient Soil Surface Balances/Budgets (GNB) for nitrogen (N) and phosphorous (P) have
been identified as key agri-environmental indicators within the EU (Eurostat, 2011). Nutrient
balances are also relevant for one of the indicators for agricultural statistics as defined in the Global
Strategy, namely no. 28, “Change in effect of inputs on the environment” (World Bank, FAO and
UNSC, 2011). Therefore, there is an increased need to obtain harmonised statistical data related to
crop nutrients. Nutrient output via harvested crops often accounts for the largest item on the output
side of the balance/budget. Thus the coefficients used to estimate the output items often have a
significant effect on the magnitude of any surplus or deficit.
Statistics Sweden has calculated and published national soil surface nutrient balances intermittently
since 1991. In 2003, the method was partly adjusted to follow the method recommended by the
OECD and Eurostat to increase comparability between countries. In spite of changes in crop
breeding and farm management, the coefficients for N and P content in crops and crop residues
have not been revised. Neither have the statistics on the utilisation of crop residues been updated
since 1997.
The general objective of this action is to obtain updated and well-defined coefficients for N and P
content in harvested cereals, temporary grasses and crop residues and to estimate to what extent the
crop residues are utilised in Sweden.
Specific objectives are:
I.
To compare the N and P concentrations in grain and straw of cereals used in the official
Swedish nutrient balances with coefficients reported from field investigations as well as
with those presently used in Sweden by, for example, national authorities, the agricultural
advisory service, non-governmental organisations and research institutes, for nutrient
balance calculations and data modelling.
3
II.
To quantify the areas (hectares) where crop residues have been and not been removed from
the field of main crops in Sweden and, if possible, update currently available coefficients on
N and P for crop residues.
III.
To conduct a sensitivity analysis of the effect on the Swedish GNB of using a range of
different N and P coefficients for harvested products and crop residues, respectively.
IV.
To make recommendations on sound methodologies for obtaining best practice in terms of
obtaining N and P crop content coefficients, considering factors such as scientific ground,
cost-efficiency, practical feasibility, etc.
Resent calls for the harmonisation of coefficients used in the production of official statistics related
to plant nutrients and in other reporting systems/bodies, such as the green house gas inventory, have
led to the realisation of this project. The approach, main results and outcome will be presented in
the following sections. The focus of this paper is on nitrogen. For a more comprehensive report, see
Andrist Rangel et al. (2013).
2. Harmonisation of coefficients for N and P in cereals and temporary grasses
The nutrient content in harvested crops depends on various factors, such as location, year and
fertilisation regime (Haak, 1998; Vos & van der Putten, 2000; Mattsson, 2005; Saleque et al., 2006;
Wang et al., 2007; Peltonen-Sainio & Jauhiainen, 2010; Neacşu, 2011; Swanston et al., 2012).
Hence, there is a range of different coefficients for N and P in harvested crops and crop residues
published in scientific literature and in reports and databases from environmental monitoring,
experimental trials, etc.
2.1 Approach and data collection
Coefficients for N and P content in crop products from the three main cereal crops in
Sweden, i.e., grain and straw of winter wheat (Triticum aestivum), spring barley (Hordeum vulgare)
and oats (Avena sativa) as well as temporary grasses currently available and used in Sweden, were
investigated by means of a desk study. In addition, data on N and P content in crop samples of
temporary grasses (n=100) collected parallel to the Swedish national environmental monitoring
programme was obtained from the Swedish University of Agricultural Sciences (SLU). The
samples had been selected to represent five geologically different regions in Sweden. For P, the data
was already available, whereas for N, data was obtained after elemental analysis (LECO total
nitrogen analyzer) of achieved crop samples conducted by a laboratory at the Department of Soil
and Environment at SLU.
2.3 Results
Apart from the official Swedish nutrient balance calculations, four main uses of the
coefficients and two main field investigations were identified (Table 1). In addition to this, data
from various experimental trials, e.g., different levels of fertilisation, types of cropping systems,
etc., was found (data not shown).
4
Table 1: Main users and field investigations of N and P coefficients in crops identified in this
action.
Institution
Usage* /field
investigation#
Geographic level
Reference
Designation in
this paper (and
colour in Tables
2 and 3)
Statistics Sweden
Official statistics on
nutrient balances
National/
Regional (agricultural
production areas,
catchments)
Haak, 1988 in
Claesson and
Steineck, 1991
SCB
Swedish Board of
Agriculture
*In the agricultural
advisory service, for
calculation of N, P and K
balances, leaching, climate
footprint, etc., at farm
level
Farm
Swedish Board of
agriculture, 2013
STANK in MIND
SMED/ Swedish
EPA
*Green house gas
inventory report to
UNFCCC
National
Swedish EPA, 2012
GHG
SMED /Swedish
EPA
*Modelling of N and P
leaching losses from
arable land
National/Regional
Statistics Sweden
*Calculation of
OECD/Eurostat GNB for
N and P
#National
SLU
SLU
environmental
monitoring of arable soils
and cereal crops.
(Light brown)
(Grey)
(Light blue)
Johnsson et al. 2008,
Blombäck et al.
2011
N and P modelling
National/
Regional (NUTS2)
OECD/Eurostat
GNB Excel
spreadsheets
OECD/Eurostat
National/Regional
Eriksson et al., 2010
Environmental
monitoring
(Dark blue)
(Purple)
(Green)
#Variety trials
Larsson, 2011
Variety trials
The span of coefficients found for N and P are presented in Tables 2 and 3. The colour indicates the
use/source of the coefficient. The results show that for grains of winter wheat, spring barley and
oats, there is a variation in the coefficients that is reported and/or used. There is also a tendency for
SCB coefficients to be at the higher end of the range of grain N concentrations. For temporary
grasses, the presented data in Tables 2 and 3 shows the maximum, median and the minimum of the
concentrations measured in the samples collected parallel to the environmental monitoring
programme. Here, the SCB coefficients are well in line with the median, both for N and for P.
There is a fairly large variation in the compiled concentrations of N in crop residues of the studied
crops. For straw of winter wheat and oats, the N concentrations used in the SCB nutrient balances
are in line with the concentrations used in the OECD/Eurostat GNB and by the Swedish Board of
Agriculture in STANK in MIND, but higher than the ones used by the SMED/Swedish EPA in the
calculations for the greenhouse gas inventory (Table 2). The SCB coefficients for winter wheat and
oat straw are also higher than the concentrations found in field trials (data not shown). For spring
barley, the SCB coefficients are higher than what is reported and used elsewhere. For P
concentrations in straw, only a few experimental results have been found, and the figures reported
vary considerably. The P concentrations used in the SCB balances are the same as the ones used by
the Swedish Board of Agriculture (Table 3). Compared to these, coefficients used in the
OECD/Eurostat balances are considerably higher.
5
Table 2: Nitrogen concentrations (%), in grain and straw of winter wheat (Triticum aestivum),
spring barley (Hordeum vulgare) and oats (Avena sativa) at 15 percent moisture content and for
samples of temporary grasses in dry matter (DM). Range of coefficients among main users and
reporters (for references and colour labels see Table 1).
Levels used in the
sensitivity analysis
High
Medium
Low I
Low II
Nitrogen (%)
Grain
Straw
0.7
1.9
0.7
1.79
1.8
0.6
1.7
0.43
1.66
0.43
High
1.7
1.7
0.8
Spring barley
Medium
1.6
0.7
0.7
Low
1.5
0.65
High
1.8
1.8
0.7
0.7
0.7
Oats
Medium
1.65
0.62
Low
1.45
0.62
High
3.97*
Temporary
Medium
2.22*
2.4
grasses
Low
1.1*
* Whole crop samples. % DM. Max, median and min for samples of temporary grasses, n=100, collected parallel to the
environmental monitoring programme. (Previously unpublished data, see Section 2.1).
Winter wheat
Table 3: Phosphorus concentrations (%), in grain and straw of winter wheat (Triticum aestivum),
spring barley (Hordeum vulgare) and oats (Avena sativa) at 15 percent moisture content and for
samples of temporary grasses in dry matter (DM). Range of coefficients among main users and
reporters (for references and colour labels see Table 1).
Levels used in the
sensitivity analysis
High
Medium
Low I
Low II
Phosphorus (%)
Grain
a
Strawa
0.37
0.32
0.35
0.1
0.1
0.32
0.1
0.1
0.31
0.31
0.1
0.1
High
0.35
0.35
0.35
0.32
Spring barley
Medium
0.35
0.35
0.35
0.1
0.1
Low
0.34
0.34
0.1
0.1
High
0.35
0.32
Oats
Medium
0.34
0.1
0.1
Low
0.33
0.33
0.33
0.1
0.1
High
0.48*
Temporary
Medium
0.29*
0.26
grasses
Low
0.14*
* Whole crop samples. % DM. Max, median and min for samples of temporary grasses, n=100, collected parallel to the
environmental monitoring programme. (Previously unpublished data, see Section 2.1).
Winter wheat
6
3. Crop residues
Crop residues need to be accounted for both in nutrient balance calculations and in the
calculation of nitrous oxide (N2O) emissions in the greenhouse gas inventory. Until now, the
available official statistics on crop residues in Sweden have been based on a one-time survey from
1997. The data is being used both in the official Swedish nutrient balances (Statistics Sweden
2011), the OECD/Eurostat balances (OECD, 2013) as well as in the GHG inventory for Sweden
(Swedish EPA, 2012). With increased focus on more efficient use of natural resources parallel to
greater demand for energy, it is likely that the use of crop residue has changed since the late 1990s.
Therefore a new survey on the utilisation of straw and tops was conducted within the framework of
this action.
3.1 Approach and data collection
The existing sample survey on “Cultivation measures in agriculture” conducted every
second year by Statistics Sweden was used to collect information on crop residues for the main
crops in Sweden. The survey was carried out on a sample of 3000 agricultural holdings during the
autumn/winter 2012/2013 by adding a number of additional questions on crop residues to the
ordinary questionnaire. The data collected included: areas (hectares) from which the crop had been
harvested as green fodder; areas in which the crop residues had been incorporated in the soil at
tillage; and areas from which crop residues had been removed from the field. The type of usage was
also surveyed. Reprinted categories for use were: Fodder for direct use; fodder industry; litter;
biogas; energy (heating); and other uses.
3.2 Results
Straw was removed from the field on 40 percent of the area on which cereals were grown.
This is an increase compared to earlier results from 1997. For 73 percent of the cereal area where
crop residues had been removed, the straw was used as litter. Use of straw as feed was second most
common, with a share of 13 percent. The use of straw for heating had increased compared to earlier
results, to comprise about 9 percent of the cereal area where crop residues had been removed from
the field. The complete results on crop residues have been published as official statistics together
with the other results from the survey (Statistics Sweden, 2013).
4. Sensitivity analysis
The nutrient balances consist of several in- and outputs (Table 4). Hence, the effect of
changing the coefficients related only to one of the items in the balances, namely ‘crop products’, is
unclear. Sensitivity analysis was chosen as a tool to evaluate this effect.
4.1 Approach
Sensitivity analysis was conducted in order to quantify the effect on the Swedish GNB of
using a range of different N and P coefficients for winter wheat, spring barley, oats and temporary
grasses. The coefficients were allowed to vary both for harvested crop product (grain) and crop
residues (straw). All other coefficients were fixed. Data on quantities was the same as that used in
the calculations of the official Swedish nutrient balances for 2009 (Statistics Sweden, 2011).
7
Table 4: In- and outputs for the nitrogen and phosphorus soil surface balances used in the
sensitivity analysis.
Nutrient inputs
Mineral fertiliser
Soil amendments
Stable- and grazing manure
Seed
Atmospheric deposition
Sewage sludge
Biological nitrogen fixation (only applicable for N)
Nutrient outputs
Crop yield
Harvested/removed crop residues
Three different levels of N and P concentrations were set for each of the crops: High, Medium and
Low. Coefficients for the respective levels were chosen from the range of concentrations found in
the preceding screening (see Section 2.3). The resulting matrices of data, serving as input to the
sensitivity analysis, can be seen in Tables 2 and 3 (fixed coefficients not shown).
Programming for the sensitivity analysis was performed using the Statistical Analysis Software
SAS. The calculation model used in the official Swedish nutrient balances 2009 was used to
compute each combination of alternatives for winter wheat, spring barley, oats and temporary
grasses. Factors other than N and P coefficients for grain and straw were kept unchanged. The
results were exported to Excel, where the sensitivity analysis was performed. By compiling pivot
tables, the analyst could see the effects on the nutrient balances, both in tables and graphs, by
changing the nutrient levels to High, Medium or Low for the crop of interest. Presentations can be
made for both national and regional levels. The regional levels are either eight agricultural
production areas (PO) (Fig 1) or catchments (eight in total).
4.2 Results - N balances
In total, when analysing the sensitivity of the N balances at national level, 108 different
combinations of varying N content in winter wheat, spring barley, oats and temporary grasses had
been computed (all other factors fixed). Accordingly, 108 different balances were obtained (Figure
2). The span of N balances in kg per ha of agricultural land ranged between 14 and 50 kg N/ha.
Hence, all the balances at national level indicate a surplus of N. The surplus consists of ammonia
losses, leaching losses, denitrification as well as build-up in the soil. There was a clear pattern with
three different groups in the data, with most of the balances around 16, 36 and 48 kg/ha
respectively.
8
PO 1
Plain districts in Southern Götaland
2
Central districts in Götaland
3
Plain districts in Northern Götaland
4
Plain districts in Svealand
5
Forest districts in Götaland
6
Forest districts in Central Sweden
7
Lower parts of Norrland
8
Upper parts of Norrland
Figure 1: Sweden is divided into eight agricultural production areas (PO).
Figure 2: Soil surface N balances for Sweden 2009, resulting from varying N concentration in
grain and straw in winter wheat, spring barley, oats and temporary grasses. All other in- and
outputs fixed.
9
In the next step, only one crop at a time was allowed to vary in terms of N coefficients and the
analysis was done at regional level (Figures 3 and 4). The surplus/deficit of the balances varies
among the production areas (PO) as well as their sensitivity to the change in the crop N coefficient.
For example, the Plain districts in southern Sweden (PO 1) make up the production area that is most
sensitive to a change in the coefficient for winter wheat, whereas it is the least sensitive to changes
in temporary grasses N content (Figures 3 and 4). This is explained by differences in relative crop
areas among the production areas. In PO 1, winter wheat is the dominating crop, whereas it has the
smallest area of temporary grasses among all the production areas (Statistics Sweden, 2010). In
Lower and Upper parts of Norrland (PO 7 and 8), the areas of winter wheat are insignificant, and
hence there is no change in the balances when the corresponding N coefficient is changed. In
general, the sensitivity analysis shows rather limited effects on the N balances when only the N
content in winter wheat is changed, using the range of concentrations found in the screening (Table
2). For temporary grasses on the other hand, the span in concentrations was much greater. This, in
combination with large areas of temporary grasses in several of the production areas, resulted in the
sensitivity analysis showing large differences in N balances; from -5 to 60 kg N/ha.
Figure 3: Soil surface N balances for agricultural production areas (PO 1-8) in Sweden 2009,
resulting from varying N concentrations in grain and straw of winter wheat. All other in- and
outputs fixed. The category “SCB” denotes the official Swedish soil surface balances for 2009.
10
Figure 4: Soil surface N balances for agricultural production areas (PO 1-8) in Sweden 2009,
resulting from varying N concentrations in temporary grasses. All other in- and outputs fixed. The
category “SCB” denotes the official Swedish soil surface balances for 2009.
The strong influence of the temporary grasses on the total balance also explains the outcome of
groups of balances in Figure 2. The common factor for the balances in the lowest group (14-18 kg
N/ha) is that the coefficients for temporary grasses are at High for all the combinations. Hence,
large amounts of N are removed from the field when the temporary grasses are harvested, which
results in a small surplus of N in the soil surface balances. In the next group (34-37 kg/ha), the
coefficients for temporary grasses are at Medium, and accordingly at a Low level in the last group
with balances between 46 and 50 kg N/ha.
5. Recommendations
To keep the N and P content in grains and crop residues updated, the main “sample
programmes” which would be suitable for benchmarking are the Swedish environmental monitoring
of crops and the variety trials. Both these programmes seem to have fairly secure funding in the
immediate future, and are carried out on a regular basis. Although the number of samples for wheat,
barley and oats taken in the environmental monitoring programme will probably decrease in
forthcoming rounds, there will hopefully still be enough samples for capturing the trends of whether
the nutrient contents in grain crops are decreasing, increasing or remaining steady. A benefit of
using the environmental monitoring programme as the first hand option for benchmarking is that
both N and P have been analysed up to the last sampling series. However, in the ongoing sampling
series, only P is being analysed due to financial restrictions. Additional funding could enable N
analysis to be carried out, as the crop samples are archived. Since the locations for samples are
chosen to be representative for Swedish agriculture, due to a complex system of grid sampling
(Eriksson et al., 2010), the results are likely to give the best available overall picture of Swedish
farming today. Sampling of temporary grasses is not included in the regular programme; hence data
on temporary grasses would need to be obtained from an alternative source. The Nordic Feed
11
Evaluation System (NorFor, 2012) database could be one option, where coefficients for different
feedstuff for cattle are available for four countries (Denmark, Iceland, Norway and Sweden).
The variety trials are carried out every year, and the results are published a few months after
harvest. The experimental yields are higher than the officially reported “Standard yields”. This may
cause a dilution effect of the N concentration in grain, which, on the other hand, is counteracted if
more N fertilisers are applied. A drawback of the variety trials is that no P analyses are performed.
However, many varieties are analysed, and mean values are provided for a five-year period, which
reduce the effect of weather and wind during a certain year. If used together with the environmental
monitoring, the variety trials can contribute valuable information about the current state of crops in
terms of N content.
The updated official statistics on the utilisation of crop residues for 2012 produced within this
action present data on areas. However, to fully make use of this data in the nutrient balances and in
the greenhouse gas calculations, the areas need to be converted into quantities. For this, data on
crop yields combined with coefficients for the ratio between grain yield and straw and/or other crop
residues is needed. Another factor to consider is the stubble height. These factors would also need
to be reviewed and revised. This lies outside the scope of this action, but is a forthcoming action to
take in order to completely provide updated and harmonised coefficients for crop products in
nutrient balances and GHG calculations.
In terms of cost efficiency and practical feasibility, the sensitivity analysis is a useful tool to
evaluate the effect of changing one or several factors in a model. Here we have demonstrated one
way of conducting a sensitivity analysis in order to update crop N and P coefficients in nutrient
balances. The analysis provides a basis for decisions on which variables are worth investing
resources in to obtain more precise coefficients and hence a more accurate estimation of the output.
7. Conclusions
The outcome of this action can serve as a basis for recommendations on sound
methodologies to obtain better practices in terms of obtaining N and P crop content coefficients,
considering factors such as, scientific ground, cost-efficiency and practical feasibility. Ideally,
harmonised coefficients should be used by all reporting systems within each country. Differences in
coefficients between countries should be supported by well-documented and confirmed evidence of
differences in crop products in terms of N and P content.
7. Acknowledgement
This work has been conducted within the frame of an action, supported by Eurostat via an EU
grant.
12
REFERENCES
Andrist Rangel Y, Eriksson J, Fägerlind K, Liew J, Ländell G, Wahlstedt G (2013). Estimating the
content of nitrogen and phosphorus in agricultural products in Sweden. Statistics Sweden. PM
RM/L Lantbruksstatistik 2013:1 (submitted)
Blombäck K, Johnsson H, Lindsjö A, Mårtensson K, Persson K, Schmieder F. (2011) Läckage av
näringsämnen från svensk åkermark för år 2009 beräknat med PLC5-metodik – Beräkningar av
normalläckage av kväve och fosfor för 2009, SMED Rapport 57 2011 (in Swedish)
Claesson S, Steineck S (1991).Växtnäring hushållning – miljö. Speciella skrifter 41, SLU, Uppsala.
(in Swedish)
Eriksson J., Mattsson L., Söderstöm M. (2010) Current status of Swedish arable soils and cereal
crops. Data from the period 2001-2007. Naturvårdsverket, Rapport 6349. (in Swedish with
English summary)
Eurostat (2011) Analysis of methodologies for calculating greenhouse gas and ammonia emissions
and nutrient balances. Eurostat Methodologies and Working papers. ISSN 1977-0375
Haak E (1998) Influence of weather and N-fertilization on DM-yield and nutrient uptake in the
fertility experiment R3-9003, 1985-1992. Swedish University of Agricultural Sciences,
Department of Soil Sciences, Division of Soil Fertility, Report 200.
HELCOM (2011) The Fifth Baltic Sea Pollution Load Compilation (PLC-5). Balt. Sea Environ.
Proc. No. 128
Johnsson H, Larsson M, Lindsjö A, Mårtensson K, Persson K, Torstensson G (2008). Läckage av
näringsämnen från svensk åkermark Beräkningar av normalläckage av kväve och fosfor för 1995
och 2005. Naturvårdsverket rapport nr 5823. 152 pp. (in Swedish with English summary)
Larsson S (2011a) Sortvalstabeller 2011. Resultat från sortförsök 2006-2010. Sortprovningen, SLU.
(in Swedish)
Mattsson L (2005) Halmskörden, hur stor är den? Swedish University of Agricultural Sciences,
Department of Soil Science, Division of Soil Fertility (in Swedish, not officially published)
Neacşu A (2011) Grain protein concentration and its stability in a set of winter wheat cultivars,
grown in diverse environments and management practices. Romanian Agricultural Research 28,
29-36.
NorFor (2012) Nordic Feed Evaluation System, search for feedstuff, online:
http://feedstuffs.norfor.info/ [2012-05-08]
Öborn I, Edwards AC, Witter E, Oenema O, Ivarsson K, Withers PJA, Nilsson SI, Stinzing R
(2003) Element balances as a tool for sustainable nutrient management: a critical appraisal of
their merits and limitations within an agronomic and environmental context. European Journal
of Agriculture 20, 211-225.
OECD (2013) Homepage - OECD/Eurostat Gross Nutrient Balance Excel spreadsheets.
http://stats.oecd.org//Index.aspx?QueryId=48675# [2013-06-29]
13
Oenema O, Kros H, de Vries W (2003) Approaches and uncertainties in nutrient budgets:
implications for nutrient management and environmental policies. European Journal of
Agronomy 20, 3-16.
Peltonen-Sainio P & Jauhiainen L (2010) Cultivar improvement and environmental variability in
yield removed nitrogen of spring cereals and rapeseed in northern growing conditions according
to a long term dataset. Agricultural and Food Science 19, 341-353.
Statistics Sweden (2010) Production of cereals, dried pulses oilseed crops, potatoes and temporary
grasses in 2009. Final statistics. JO 16 SM 1001 (in Swedish with English summary)
Statistics Sweden (2011) Nitrogen and phosphorus balances for agricultural land and agricultural
sector in 2009. MI 40 SM 1102 (in Swedish with English summary)
Statistics Sweden (2013) Cultivation measures in agriculture 2012. Set-aside, ley, spring barley,
oats, winter grain and utilisation of straw and tops. MI 1001 SM 1302 (in Swedish with English
summary)
Swanston JS, Smith PL, Agu RC, Brosnan JM, Bringhurst TA, Jack FR (2012) Variation, across
environments within the UK, in grain protein and grain hardness, in wheat varieties of differing
distilling quality. Field Crop Research 127, 146-152.
Swedish Board of Agriculture (2013) STANK in MIND product list, updated May 2013.
Swedish EPA (2012) National Inventory Report Sweden 2012.
Vos J, van der Putten PEL (2000) Nutrient cycling in a cropping system with potato, spring wheat,
sugar beet, oats and nitrogen catch crops. I. Input and offtake of nitrogen, phosphorus and
potassium. Nutrient Cycling in Agroecosystems 56, 87-97.
Wang X, Cai D, Hoogmoed WB, Perdok UD, Oenema O (2007) Crop residue, manure and fertilizer
in dryland maize under reduced tillage in northern China: I grain yields and nutrient use
efficiencies. Nutrient Cycling in Agroecosystems 79, 1-16.
World Bank, FAO and UNSC (2011) Global Strategy to Improve Agricultural and Rural Statistics.
Report no. 56719‐GLB, World Bank, Washington DC. [www.fao.org/economic/ess/esscapacity/ess-strategy/en/]
14