Subscriber access provided by Karolinska Institutet, University Library
Article
Variability of greenhouse gas footprints of field tomatoes grown for processing: inter-year and inter-country assessment Wan Yee Lam, Rosalie van Zelm, Ana Benítez-López, Michal Kulak, Sarah Sim, John M. Henry King, and Mark A.J. Huijbregts Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b04361 • Publication Date (Web): 01 Dec 2017 Downloaded from http://pubs.acs.org on December 5, 2017
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 25
Environmental Science & Technology
1
Variability of greenhouse gas footprints of field tomatoes grown for processing:
2
inter-year and inter-country assessment
3 4
Wan Yee Lam1*, Rosalie van Zelm1, Ana Benítez-López1, Michal Kulak2, Sarah Sim2, J. M. Henry King2,
5
Mark A. J. Huijbregts1
6
1
7
P.O. Box 9010, 6500 GL Nijmegen, The Netherlands
8
2
9
brook, Bedfordshire MK44 1LQ, United Kingdom
10
*
Department of Environmental Science, Institute for Water and Wetland Research, Radboud University,
Unilever Safety and Environmental Assurance Centre, Unilever R&D, Colworth Science Park, Sharn-
Corresponding author
11 12
ABSTRACT: Our study provides an integrated analysis of the variability of greenhouse gas (GHG) footprints of
13
field grown tomatoes for processing. The global farm-specific dataset of 890 observations across 14 countries over
14
a three-year period (2013-2015) was obtained from farms grown under Unilever’s sustainable agricultural code. It
15
represents on average 3% of the annual global production of processing tomatoes: insights can be used to help
16
inform corporate sourcing strategies and certification schemes. The median GHG footprint ranged from 18 in Chile
17
to 61 kg CO2-eq. per tonne of tomatoes in India; lower than results reported in other studies. We found that foot-
18
prints are more consistent within countries than between them. Using linear mixed effect models, we quantified the
19
relative influence of environmental conditions and farm management factors. Key variables were area of production
20
and the method of fertilizer application. GHG footprints decreased with increasing area of production to a threshold
21
of 17.4 ha. Farms using single fertilizer application methods in general had a larger GHG footprint than those using
22
a combination of methods. We conclude that farm management factors should be prioritized for future data collec-
23
tion and more stringent guidance on acceptable practices is required if greater comparability of outcomes is needed
24
either within a scheme, such as the Unilever’s sustainable agriculture code or between schemes.
25 ACS Paragon Plus Environment
Environmental Science & Technology
Page 2 of 25
26
INTRODUCTION
27
Greenhouse gas (GHG) calculators and footprinting can be used by companies to inform management strategies
28
within agricultural supply chains1. The impacts of the agricultural phase in the life cycle of bio-based products are
29
highly variable and subject to many influencing factors, especially in open-field cultivation systems2-5. Variability
30
in GHG footprints (kg CO2-eq per tonne) of crop production is directly related to variation in factors such as,
31
fertilizer use, machine use, irrigation and yield2, 3, 5. These sources of variability are interrelated and influenced by
32
environmentally related factors, such as climate, soil properties and elevation as well as other farm related factors
33
such as area of production and fertilizer application methods6, 7.
34
Previous life cycle assessment (LCA) studies investigating variability of agricultural production focused on factors
35
directly used in GHG footprint calculations as the major contributors to variability in GHG footprints2, 3, 5, 8, 9. For
36
field tomato production, Clavreul et al2 conducted intra- and inter-year variability analysis of GHG footprints of
37
cultivation using data from 189 farms from 2012-2015 in the Extremadura region in Spain and Portugal. The GHG
38
footprints, ranging from 29 to 89 kg CO2-eq per tonne of tomatoes, were found to be most sensitive to the variability
39
in yield, followed by farm practices such as the extent of pump irrigation and choice and amount of fertilizer used.
40
Pishgar-Komleh3 quantified variability of GHG footprints of field tomato production using data from 204 farms in
41
Iran and obtained GHG footprints ranging from 100 to 400 kg CO2-eq per tonne of tomatoes. They also found the
42
variability of GHG footprints to be mainly driven by the variability in yield, followed by fertilizer application.
43
While Clavreul et al2 attributed the importance of inter-year variability in GHG footprints to variability in weather
44
conditions, there was no formal quantification of the relationship between climatic and soil conditions and the GHG
45
footprint of the tomatoes.
46
Linear mixed models (LMM), also known as linear mixed-effects models, are able to incorporate a wide variety of
47
correlation patterns in their random effects structure and this flexibility provides accurate estimates of the fixed
48
effects in the presence of correlated errors due to data hierarchy and repeated measurements 10, 11. This approach is
49
particularly suitable for datasets that have several observations nested within each country, and repeated measure-
50
ments from farms in different years; and offers a systematic approach to analyze the importance of other environ-
51
mentally and farm related factors, such as soil, climate, and fertilizer application method, in contributing to the
52
variability in GHG footprints of crop production. Analyzing the variability of the GHG footprints at the farm level ACS Paragon Plus Environment
2
Page 3 of 25
Environmental Science & Technology
53
and understanding the drivers of the variability is necessary to identify possible areas of GHG reduction and to
54
enable more targeted GHG mitigation strategies within agricultural supply chains.
55
The goal of this study was to: 1) assess the variability (between farms, countries and years) of the GHG footprint
56
of commercially grown field tomatoes globally and 2) understand the relationship between GHG footprints of com-
57
mercially grown field tomatoes and environmental factors, such as climate and soil characteristics, and farm related
58
factors, such as production area and fertilization method. The dataset represents farms that were compliant with the
59
Unilever’s Sustainable Agriculture Code (SAC) and covers 14 countries that represent approximately 80% of global
60
production of field-grown tomatoes and includes the top five producing countries namely, the United States of
61
America (USA), China, Italy, Spain and Turkey12. Firstly, we assessed the variability in GHG footprints and quan-
62
tified, with a partial correlation analysis, the relative importance of factors that are used in standard GHG footprint
63
calculations of tomato production, namely (i) tomato yield, (ii) emissions from fertilizer production and field ap-
64
plication and (iii) emissions from energy use. Secondly, we used LMM to quantify the relative influence of envi-
65
ronmental conditions and farm management factors on the variability of GHG footprints for global tomato produc-
66
tion. The first analysis was conducted using field tomato production data from 890 farm-specific observations
67
across 14 countries and measured over three years (i.e. 2013 -2015)13. The second analysis was based on a subset
68
of 719 observations with unique geolocations and complete description of farm management factors.
69
ACS Paragon Plus Environment
3
Environmental Science & Technology
Page 4 of 25
70
MATERIALS AND METHODS
71
GHG footprint. Following the system boundaries illustrated in Figure 1, the GHG footprint of field-grown tomato
72
production on a specific farm x in a specific year j (GHGtomato in kg CO2-eq tonne-1 tomato produced) including
73
GHG emissions from energy use by machinery and irrigation (GHGenergy in kg CO2-eq ha-1), GHG emissions from
74
fertilizer production and field nitrous oxide emissions from application of nitrogen fertilizers and crop residues
75
(GHGfertilizer in kg CO2-eq ha-1) per unit of tomato produced (Yield in tonne ha-1) was defined as (Equation 1):
76
𝐺𝐻𝐺𝑡𝑜𝑚𝑎𝑡𝑜,𝑥,𝑗 =
GHG𝑒𝑛𝑒𝑟𝑔𝑦,𝑥,𝑗 + 𝐺𝐻𝐺𝑓𝑒𝑟𝑡𝑖𝑙𝑖𝑧𝑒𝑟,𝑥,𝑗 Yield𝑥,𝑗
(1)
77
Data for GHG footprint calculations (Table S1) were collected from the Cool Farm Tool14, but footprints were
78
calculated outside of the data collection software14. Amount of energy and fertilizer consumption were provided in
79
the data collection sheet as aggregate values by their respective types (MJ of electricity, diesel and petrol for energy
80
consumption and kg of different types of fertilizers for fertilizer consumption), without further breakdown of the
81
consumption in each agricultural process illustrated in Figure 1. As specific land use history for the farms was not
82
available in the extracted dataset, biogenic GHG emissions from land use change were considered following the
83
approach originating from Mila I Canals15 and recommended by Nemecek et al16. Upon analysis of the historical
84
land cover data (FAOSTAT17 over the past 20 years), no land use change arising from tomato production was found
85
in the sample of countries considered. Thus, emissions from land use change were considered to be zero in this
86
analysis16. GHG emissions from pesticide production were excluded as many farms did not provide sufficient in-
87
formation on type and amount of pesticide use. This is likely to have limited impact in the calculation of GHG
88
footprints, as previous studies have indicated pesticide production and application represents less than 5% of the
89
GHG footprint of field grown tomatoes2, 8. GHG emissions from capital goods production were also omitted as
90
their contribution to the GHG footprint of agricultural products is typically low18. Temporary carbon sequestration
91
by tomato plants was also not taken into account due to regular harvest of the tomato crop compared to perennial
92
crops, as well as the short-lived nature of the tomato-based products8. Carbon dioxide (CO2), methane (CH4) and
93
nitrous oxide (N2O) emissions were summed using Global Warming Potentials (GWP) of 1, 30 and 265 CO2-
94
equivalents, respectively, representative of a 100-year time horizon19. The GHG emission factors, expressed as kg
95
CO2-eq per unit of material or energy consumed, were derived from secondary sources, mainly ecoinvent version ACS Paragon Plus Environment
4
Page 5 of 25
Environmental Science & Technology
96
3.220 (Table S5). See Section S2 of Supplementary Information (SI) for more details about the overall GHG calcu-
97
lation methodology.
98 99 100
Figure 1. System boundaries for greenhouse gas footprinting from cradle to farm gate (solid lines). Emissions from pesticides production and capital goods production were excluded (dotted lines).
101 102
We calculated annual GHG footprints of 890 farm-specific observations from 14 countries: Australia, Chile, China,
103
Egypt, Greece, India, Israel, Italy, New Zealand, Poland, Portugal, Spain, Turkey, and the USA for the three-year
104
period 2013-2015. The summed production volume of the farms relative to the annual global tomato production in
105
each year was about 3%12. The farms in this study applied conventional farming methods (not organic) and were
106
sampled randomly for self-assessment according to the scheme rules21 from farms operating in compliance with
107
Unilever’s Sustainable Agriculture Code (SAC)13. In certain countries, farms were concentrated in specific regions,
108
these included: California in the USA, Extremadura region in Spain, and Xinjiang region in China. Farm data were
109
collected via spreadsheets from the Cool Farm Tool14 in an unaudited format and they were processed and cleaned
110
prior to analysis (Section S1 of SI). A data quality score for each observation, ranging from 1 to 12, with lower
111
scores representing more unique and complete information, was applied in accordance to the criteria as specified
112
in Section S3 of SI. Only farms that have unique and complete information for area, yield, fertilizer and energy
ACS Paragon Plus Environment
5
Environmental Science & Technology
Page 6 of 25
113
consumption. were given data quality scores less than or equal to 7, and were included in the cleaned dataset of 890
114
observations (Section S3 of SI).
115 116
Variability and correlation analysis. A dataset of 890 farm observations was used for the analysis of variability
117
of GHG footprints, as well as for the relative contribution to variability of the factors used in the GHG footprint
118
calculations; i.e. GHGenergy, GHGfertilizer and Yield. Variability in GHG footprints was quantified using the inter-
119
quartile range and coefficient of variation at four spatial and temporal scales namely, 1) within each country in each
120
year, 2) within each country in all years, 3) within all countries in each year, 4) within all countries in all years
121
(overall dataset). Only countries with data for at least 10 farms per year were included in the variability analysis at
122
level 1; and only countries with data for at least 10 farms over the three years were included in the variability
123
analysis at level 2 (See S1 of SI for number of farm observations in the 14 countries by year). This helps to ensure
124
that the analysis at the country level is based on a sufficient number of observations within a single country. At
125
level 3 and 4, the full dataset across all 14 countries was used as appropriate. We also quantified the variability of
126
GHG footprints 5) between each year in each country and 6) between each country covering all years. We then
127
calculated the Spearman’s rank and partial correlation coefficient between the GHG footprint and Yield, GHGfertilizer
128
and GHGenergy at the first four spatial and temporal levels (ppcor package16, R 3.3.1 software17). The rank correlation
129
provided an indication of the relative influence of each factor on the variability of the GHG footprints while con-
130
sidering the interaction between them.
131 132
Linear mixed model. We used LMM to assess variations in GHG footprints as a function of a set of environmental
133
conditions and farm management factors at the global level. In this analysis, we only included observations with
134
unique geocoded locations and for which data on farm management factors were available (N = 719, See S3 of SI).
135
For each observation we obtained information from the Cool Farm Tool14 on two farm management factors, namely
136
the area of production and fertilizer application method (Table 1). Farmers employ a variety of single and multiple
137
fertilizer application methods, and thus we classified farms accordingly into single fertilizer application methods,
138
i.e. ‘incorporation’, ‘apply in solution’, ‘subsurface drip’, and ‘broadcast’, and multiple methods, i.e. ‘incorporate-
139
subsurface drip’, ‘incorporate-broadcast’, ‘broadcast-apply in solution’, ‘incorporate-apply in solution’, and ‘com-
ACS Paragon Plus Environment
6
Page 7 of 25
Environmental Science & Technology
140
bination of 3 unique methods’ (9 types of fertilizer application methods in total). In addition, farm-specific envi-
141
ronmental characteristics for climate conditions and soil type were obtained from spatially-explicit maps using
142
ArcGIS22 (See Table 1 for the spatial resolution) (See S5 for detailed procedure of farm geolocation and spatial
143
data extraction). Climate parameters were obtained for the growing season of open-field tomatoes in each country,
144
e.g. mid-April to mid-October for Spain (See Section S4 of SI for the country-specific growing seasons). Soil
145
parameters were obtained for the topsoil fraction (top 30 cm of soil) in which the roots of field tomatoes are typically
146
concentrated 23, 24 and where soil parameters influence growth of tomatoes 25, 26. We chose extreme climate param-
147
eters to capture extreme events, such as droughts which could have a large impact on GHG footprints due to in-
148
creased irrigation demands.
149 150
Table 1. Fixed effect variables for model-building Fixed effect variables Climate conditions during growing season
Grid size
Source
50km x 50km
27
5km x 5km
28
Minimum daily temperature (oC) Maximum daily temperature (oC) Monthly rainfall driest month (mm month-1) Monthly rainfall wettest month (mm month-1) Rain day frequency driest month (days month-1) Maximum monthly potential evapotranspiration (mm day-1) Maximum monthly cloud cover (%)
Soil properties of topsoil fraction Organic carbon (percentage by weight) Clay content (percentage by weight) Soil bulk density (kg dm-3) Cationic exchange capacity of clay fraction in topsoil (cmolc kg-1) Soil pH (-log(H+))
ACS Paragon Plus Environment
7
Environmental Science & Technology
Mean elevation
Farm management factors
Page 8 of 25
1km x 1km
29
Farm-specific
Primary dataset
Area of production (hectares) Fertilizer application method (9 types)
151 152
In our linear mixed model (Equation 2), 𝑌𝑖 is the response variable (GHG footprint), 𝑋𝑖𝑗 and 𝛽𝑗 represent the fixed
153
effect variables (environmental conditions and farm management factors in Table 1) and their coefficients,
154
𝑍𝑖𝑘 and 𝑏𝑖𝑘 represent the random effect variables and their coefficients, and 𝜀𝑖 is the error term. i, j and k represent
155
each farm data point, each fixed effect variable and each random effect variable, respectively. To account for vari-
156
ability between the countries, years and farms that may be influenced by factors other than the fixed variables, we
157
included Country, Year and FarmID as random variables. We chose a random intercept structure of (1|Country/Far-
158
mID)+(1|Year) (denoted in lme430 syntax) as farms were nested within countries and both farms and countries were
159
represented by three years of data.
160
𝑌𝑖 = 𝑋𝑖𝑗 𝛽𝑗 + 𝑍𝑖𝑘 𝑏𝑖𝑘 + 𝜀𝑖
161
Prior to model building, we assessed the normality of the response variables and the homogeneity of response
162
variables across the explanatory variables31. Both the GHG footprint and area of production were log10-transformed
163
to correct for skewness and area was included as a quadratic term. We assessed multicollinearity between explan-
164
atory variables using variance inflation factors (VIF)32. All VIFs were lower than 10, so all variables were retained
165
for model selection. Each explanatory variable was standardized before model fitting. Models were fitted using all
166
possible variable combinations with the packages MuMIn33 and lme430, and ranked according to the corrected
167
Akaike’s Information Criterion (AICc)34. We selected models with delta AICc less than 2 and conducted model
168
averaging based on Akaike weights35, 36. The goodness of fit of the averaged model was assessed using the weighted
169
marginal and conditional R2 37, which represent the respective explained variance by the fixed effect variables and
170
the full model (fixed + random effect variables). The explained variance by the random effect variables was
171
obtained by subtracting marginal R2 from the conditional R2. We attributed a fraction of the marginal R2 to each
172
fixed effect variable in proportion to its sum of squares obtained from the analyis of variance (ANOVA) of the
173
model. Likewise, the explained variance by each random variable was attributed in proportion to its variance ACS Paragon Plus Environment
(2)
8
Page 9 of 25
Environmental Science & Technology
174
obtained from the summary of the model. The relationships of GHG footprints with fixed effect variables found to
175
be important at the global level were further examined at the country level using Spearman’s rank correlation co-
176
efficient and variability analysis for continuous and categorical fixed effect variables, respectively. As in the case
177
for variability analysis, only countries with data for at least 10 farms were included for the analysis at the country
178
level.
179 180 181 182
ACS Paragon Plus Environment
9
Environmental Science & Technology
Page 10 of 25
183
RESULTS
184
GHG footprints and variability analysis. The annual mean GHG footprints of tomatoes weighted by production
185
volume within all 14 countries in the years 2013, 2014 and 2015 (level 3) are 63, 50 and 47 kg CO2-eq per tonne
186
of tomatoes, respectively. This represents on average a 25% decrease in GHG footprints within the sampled dataset
187
over the three years. The GHG footprint shows large variability between countries and farms, and a slightly lower
188
variability between years (Figure 2 and Section S6 of SI). The median GHG footprint within each country in all
189
years (level 2) ranges from 18 to 61 kg CO2-eq. per tonne of tomatoes, with Chile having the smallest median and
190
India the largest median GHG footprint. The weighted mean GHG footprint within each country in all years (level
191
2) ranges from 19 to 59 kg CO2-eq. per tonne of tomatoes, with Chile showing the smallest and China the largest
192
weighted mean GHG footprint. The coefficient of variation of GHG footprints within each country in all years
193
(level 2), ranges from 33% in Portugal to 159% in USA; this is larger than the coefficient of variation of GHG
194
footprints between each year in each country (level 5), that ranges from 8% in Greece to 50% in USA. See Section
195
S7 of SI for variability diagrams of yield, fertilizer and energy consumption.
2013
GHG footprints (kg CO2 eq per tonne tomato)
120
2015
100 80 60 40 20 0
196
2014
Overall Australia Chile (n=890) Countries (n=24) (n=100)
China (n=212)
Greece (n=158)
India Italy Portugal Spain USA (n=106) (n=41) (n=41) (n=122) (n=67)
197
Figure 2. GHG footprints shown for each country per year (level 1) and overall by year 2013, 2014 and 2015 (level
198
3). n refers to the number of observations for each country. Only 9 countries (i.e. Australia, Chile, China, Greece,
199
India, Italy, Portugal, Spain and USA) have data for at least 10 farm-year combinations in at least one year.
200
Amongst them, only China, Greece, Spain and USA have data for at least 10 farms in each year from 2013 to 2015.
201
The variability diagrams show the 10th percentile, first quartile, median, third quartile and 90th percentile. Farm-
202
year combinations outside of this range are not presented. ACS Paragon Plus Environment
10
Page 11 of 25
Environmental Science & Technology
203
Correlation analysis. Figure 3 shows that variability in GHGenergy, GHGfertilizer and Yield contribute fairly equally
204
to the variability in GHG footprints within all 14 countries in each year (level 3) and within all 14 countries in all
205
years (level 4). This is different when looking at the variability within each country in all years (level 2) (Figure
206
3a). In this case, only 10 countries (i.e. Australia, Chile, China, Egypt, Greece, India, Italy, Portugal, Spain and
207
USA) have data for at least 10 farms over the three years of the dataset and were included in the analysis. In Chile
208
and the USA, variability in GHGenergy contributes most to the variability in the GHG footprint. Yield, on the other
209
hand, is an important driver of variability in the GHG footprints in India, Australia and, to a lesser extent, Italy.
210
Between the different years, most countries have only 2 years’ worth of data with at least 10 farms in each year for
211
comparison. Only China, Greece, Spain and USA have data for at least 10 farms in each year from 2013 to 2015.
212
The relative contribution of the different sources of variability remains relatively consistent over the years for farms
213
within each country and for farms within all countries, i.e. they remain in the same region of the tri-plots. Exceptions
214
occur for China in 2013 and USA in 2015. China moves from the higher contribution of yield and energy in 2013
215
to the middle region in 2015. USA moves from the middle region in 2013 to the higher contribution of yield and
216
energy in 2015. The median GHG footprints of field tomato production in China in 2013 and USA in 2015 are also
217
notably higher than those in the same country in other years (Figure 2).
218
ACS Paragon Plus Environment
11
Environmental Science & Technology
219
a) All years
Page 12 of 25
b) 2013
220 221 222
iv GHGenergy
GHGfertilizer
iv
GHGenergy
GHGfertilizer
223 224
ii
i
ii
iii
i
iii
225 226
Yield
227
Yield
c) 2014
d) 2015
228 229
iv
iv
230
GHGenergy
GHGfertilizer
GHGenergy
GHGfertilizer
231 232
ii
i
iii
ii
i
iii
233 234
Yield
Yield
235
Figure 3. Relative contribution of sources of variability to variance of GHG footprint by country for a) all years
236
b) 2013, c) 2014 and d) 2015. Figure 3a covers the analysis within each country in all years (level 2) and within all
237
countries in all years (level 4) while Figure 3b, 3c and 3d cover the analysis within each country in each year (level
238
1) and within all countries in each year (level 3). The regions in the tri-plots correspond to the following situations:
239
i, all three factors are relatively equally contributing; ii, Yield is the most contributing; iii, GHGfertilizer is the most
240
contributing; iv, GHGenergy is the most contributing. The abbreviations are: Overall, All 14 countries in each year
241
or all years; AU, Australia; CL, Chile; CN, China; EG, Egypt; ES, Spain; GR, Greece; IN, India; IT, Italy; PT,
242
Portugal; USA, United States of America. Only 10 countries (i.e. Australia, Chile, China, Egypt, Greece, India,
243
Italy, Portugal, Spain and USA) have data for at least 10 farm-year combinations over the three years of the
244
dataset. Amongst them, only China, Greece, Spain and USA have data for at last 10 farms in each year from 2013
245
to 2015.
ACS Paragon Plus Environment
12
Page 13 of 25
Environmental Science & Technology
246
Linear mixed model. The marginal and conditional R2 of the best averaged model are 29.6% and 64.2% respec-
247
tively and includes area of production (13.8% explained variance), method of fertilizer application (10%), rain day
248
frequency driest month (3%) and minimum temperature (2.9%) as fixed effects variables. At the global level, the
249
relationship between GHG footprints and area of production is non-linear, with GHG footprints decreasing with
250
increasing area of production up to a threshold of 17.4 ha, and then increasing for farms with larger areas of pro-
251
duction (Figure 4). Among the fertilizer application methods those farms using single fertilizer application methods
252
in general have a larger GHG footprint than those using a combination of methods (Figure 5). Amongst the nine
253
fertilizer application methods, ‘apply in solution’ is associated with the largest GHG footprints, while the combi-
254
nation ‘incorporate-apply in solution’ results in the lowest GHG footprints. Fertilizer application methods other
255
than these two extremes appear to be similar in their fitted values when considering the 90% confidence interval.
256
See S9 of SI for country-specific variability of GHG footprints with area of production (Spearman’s correlation
257
coefficient) and method of fertilizer application (median GHG footprints for each method of fertilizer application).
258
Regarding the random effect variables, these explain 34.6% of the total variance of GHG footprints, with ‘country’
259
being the most important (33.8% explained variance), followed by ‘farm’ (0.5%) and then ‘year’ (0.3%). 35.8% of
260
the variability of GHG footprints remains unexplained by the model.
GHG footprints (kg CO2 eq per tonne tomato)
500
50
5 0.2
261
2 Australia India Portugal 90% CI
20 200 Farm area (hectare) Chile China Italy New Zealand Spain Turkey Fitted
2000 Greece Poland USA
262
Figure 4. Modeled relationship between GHG footprints and Area of production for the global dataset of 719
263
observations in logarithmic scale on both axes, holding other factors constant at their median values. The bold
264
line represents the fitted value and the grey dashed lines represent the 90% confidence interval. The data points
265
represent the 719 farm-year combinations used for model-building. ACS Paragon Plus Environment
13
GHG footprints (kg CO2 eq per tonne tomato)
Environmental Science & Technology 65 60 55 50 45 40 35 30 25 20 15
Single
Double
Page 14 of 25
Triple
Fertilizer application method
266 267
Figure 5. Modeled relationship between GHG footprints and fertilizer application method at the global level,
268
holding other factors constant at their median values. The squares represent the fitted values and the lines with
269
caps represent the 90% confidence interval. Farms that used a single, double and triple number of fertilizer
270
application methods are grouped as such; N refers to the number of observations for each fertilizer application
271
method.
272
ACS Paragon Plus Environment
14
Page 15 of 25
Environmental Science & Technology
273
DISCUSSION
274
This study is, according to the authors’ knowledge, the first to provide an integrated analysis of the variability of
275
GHG footprints of global production of a crop, namely commercial field grown tomatoes for processing. In addition
276
to quantifying the relative importance of GHGenergy, GHGfertilizer and Yield on the variability of GHG footprints of
277
field tomato production, we further explored the relationships of GHG footprints with additional farm and environ-
278
mental factors not normally considered in the GHG footprint calculations, using linear mixed effect models. In the
279
sections below we highlight the implications for GHG footprint reduction strategies within the field tomato supply
280
chains.
281
GHG footprints of field tomato production. The weighted mean GHG footprints of field tomato production in
282
this study range from 19 to 59 kg CO2-eq. per tonne of tomatoes in Chile and China respectively. This is comparable
283
to the values ranging from 29 to 89 kg CO2-eq per tonne of tomatoes reported previously for Spain and Portugal in
284
2012-20152 using the same data platform14. Compared to the range of GHG footprints of field tomatoes reported in
285
other literature, i.e. 100-400 kg CO2-eq per tonne in Iran in 2014-20153, 82 and 130 kg CO2-eq per tonne in Italy
286
in 200735 and 20118 respectively, the GHG footprints in our study and that of Clavreul et al2 are lower. The com-
287
paratively large footprints of Pishgar-Komleh3 could be explained by the fact that natural land was used as a refer-
288
ence state and carbon dioxide emissions from the conversion of natural land to agriculture were included3. The
289
differences may also be explained by the fact that the sample of farms in this and Clavreul’s study2 complied with
290
the Unilever Sustainable Agriculture Code (SAC)13. The SAC requires improvements in GHG management prac-
291
tices and the findings may be indicative of improved performance. Indeed, the 25% decrease in the annual weighted
292
mean GHG footprint over the period 2013 to 2015 is also suggestive of the potential effectiveness of the SAC in
293
reducing the GHG footprint of tomato production through education and enforcement of sustainable practices
294
within supply chains13. The efficacy of sustainability codes and certification schemes is often neglected due to their
295
principal focus on management practices rather than performance38. However, due to crop rotation and supply
296
chain sourcing practices it is not possible to ascribe the reduction in GHG footprint compared to non-certified
297
sources solely to compliance with the SAC.
298
Another reason for the differences in GHG footprints compared to earlier studies may be from the use of different
299
emission factors (e.g. emission factors from earlier versions of ecoinvent (v239, v2.28 and v3.13) instead of v3.240) ACS Paragon Plus Environment
15
Environmental Science & Technology
Page 16 of 25
300
and changes in the IPCC recommended GWP values for GHG (e.g. GWP values of N2O of 298 CO2-equivalents
301
from IPCC 2007, 296 CO2-equivalents from CML2001 and 265 CO2-equivalents from IPCC 201319). Indeed, the
302
use of higher GWPs (296 CO2-equivalents for the GWP of N2O for a 100-year time horizon) by Clavreul et al2 from
303
within the Cool Farm Tool14 resulted in slightly higher GHG footprints than those from the same countries in our
304
study, i.e. Portugal and Spain (38 to 41 in our study vs 53 kg CO2-eq. per tonne of tomatoes in Clavreul et al.2).
305
However, the dataset for Spain and Portugal from Clavreul et al.2 also included an additional year (i.e. 2012) com-
306
pared to our study. In the study by Pishgar-Komleh et al.3, the use of higher GWP of N2O, coupled with the ex-
307
ceedingly high level of nitrogen input (up to 3000 kg N ha-1) could explain the higher GHG footprints in Iran when
308
compared to our study (range: 5 to 623 kg N ha-1, median: 210 kg N ha-1) and Clavreul et al.2 (range: 66 to 505 kg
309
N ha-1, median: 188 kg N ha-1). Lower yield (71 t ha-1 to 74 t ha-1) in the studies by Manfredi et al.8 and Theurl et
310
al.35, on the other hand, could explain the higher GHG footprints than those in our study (range of yield: 49 to 132
311
t ha-1, median=85 t ha-1) and Clavreul et al. (29-148 t ha-1, median= 86 t ha-1) despite the lower nitrogen inputs (130
312
to 143 kg N ha-1) in that study.
313
Variability of GHG footprints between farms, years and countries. The wide variability of GHG footprints
314
within our study shows the potential for further reduction in the GHG impact of tomato production. In the correla-
315
tion analysis, we found that the relative contribution to variability of GHG footprints by each factor, i.e. GHGenergy,
316
GHGfertilizer and Yield, is different between the different countries. On the other hand, it is consistent within the
317
same countries and years except for USA in 2015 and China in 2013, where the contribution by GHGenergy and
318
Yield, respectively are the most important. This suggests that footprints are more consistent within countries than
319
between them. One reason could be that tomato production tends to be concentrated in certain regions in countries
320
that make up a large proportion of the dataset, they include: Xinjiang in China, California in USA and Extremadura
321
region in Spain and Portugal. The farms within the same countries are hence likely to experience similar climate
322
and soil conditions as well as more similar farm management practices. The fact that farms in Australia, India and
323
to a smaller extent Italy can be found within the yield side of the tri-plot, (i.e. Region ii in Figure 3a) suggests that
324
not all farms are operating at the most efficient levels. Farms in India have the lowest median yields (See Section
325
S7 of SI) despite levels of fertilizer application and energy consumption that are comparable to, or higher than those
326
in other countries (See Section S7 of SI). Farms in Australia, on the other hand, have the largest median yield.
327
However, the large variability in yields (See Section S7 of SI) within the country implies that some farms had much ACS Paragon Plus Environment
16
Page 17 of 25
Environmental Science & Technology
328
lower yields compared to their counterparts. It is therefore important to look into the reasons for the low yields so
329
as to improve the production as well as environmental performance of these farms.
330
The shape of the curve describing the relationship between farm area and GHG footprint suggests that production
331
area may need to reach a certain minimum size before achieving economies of scale41 through the use of machinery
332
and irrigation practices and through more effective ‘learning-by-doing’42. Most countries display negative relation-
333
ships between GHG footprints and the area of production (S9 of SI). India, with the lowest median area of produc-
334
tion (0.6 ha) amongst the different countries, is also the country with the lowest median yields and highest median
335
GHG footprints (S7 of SI). Only USA has large positive relationships between area of production and both GHG
336
footprint and GHGenergy, suggesting that exceedingly large farms (median area of production = 150 ha) may not
337
always be performing optimally. Nevertheless, compared to other countries, USA has one of the lowest median
338
GHG footprints (S7 of SI).
339
In the linear mixed model, farms that adopted ‘apply in solution’ as their fertilizer application method may have
340
the highest GHG footprints due to large wastage from fertilizer flows that did not get retained in the root system13,
341
i.e. higher fertilizer dose is needed to achieve the same uptake by roots and yield leading to larger GHG footprints.
342
The combination ‘incorporate-apply in solution’, however, results in the lowest GHG footprints. This may be due
343
to the synergistic effect such that the initial incorporation of solid fertilizers help build up a strong root system that
344
could better absorb the liquid fertilizers from the application of fertilizers in solution 43. In Australia, farms using
345
‘incorporate-subsurface drip’ as method of fertilizer application have higher yields, lower GHGenergy and lower
346
GHG footprints than farms that use ‘subsurface drip’ (S9 of SI). This explains the large variability of yields within
347
the country, and suggests that a shift from purely liquid based methods to combination of solid and liquid-based
348
methods may help to increase yields and lower GHG footprints.
349
The GHG contribution from energy was most important to the variability in the GHG footprints of production in
350
USA in 2015. The state of California, where the majority of USA farms were located, experienced its fourth con-
351
secutive dry year in 2015, with more than 60% of the land experiencing exceptional drought 44. We noticed a shift
352
in the dataset from ‘subsurface drip’ in 2013 to ‘broadcast’ and ‘apply in solution’ in 2015 as the most commonly
353
used methods of fertilizer application in USA. This could indicate that in the face of drought, farmers switched to
354
more water intensive irrigation methods45 to reduce water deficit of the crop. The result was larger GHG median ACS Paragon Plus Environment
17
Environmental Science & Technology
Page 18 of 25
355
footprints since farms consumed more energy to operate irrigation pumps46. Overall, the farms in the USA were
356
successful in responding to changing weather conditions as the variability in yields in 2015 remained similar to
357
previous years (See S7 of SI).
358
Farms in China, however, were less successful in responding to drought conditions, which were the strongest in
359
2013 in Northern Xinjiang40, 41, where the majority of the tomato farms in China were located47, 48. High variability
360
in yield in 2013 (most important contributing factor for the variability of GHG footprints in China in 2013) occurred
361
despite higher energy consumption (S7 of SI), suggesting that such interventions may not always have produced
362
higher yields in times of drought. Indeed, we saw a shift in the most commonly used methods of fertilizer applica-
363
tion among the sample farms from ‘broadcast-apply in solution’ in 2013 to ‘incorporate-apply in solution’ and
364
‘incorporate-broadcast’ in 2014 and 2015 (See S9 of SI), with the earlier method associated with higher volume of
365
water use45. Further examination of factors influencing the differences between China in 2013 and USA in 2015
366
may provide guidance for future drought intervention practices.
367
Inherent differences between countries (‘country’ as a random effect variable) that are not captured by the fixed
368
effect variables explained 33.8% of variability in GHG footprints in the global dataset. This could suggest a diver-
369
gence of GHG footprints between countries due to their unique political and economic situations, e.g. country-
370
specific fertilizer policies 49-52, legislative limits 49, 51-55, subsidies or taxes. Moreover, farmers from the same country
371
may learn more easily from each other
372
persist unless changes are made at the country-level through policy improvements and technological transfers. Dif-
373
ferences between farms and years are less pronounced, and may suggest that most of these differences have been
374
captured by the fixed effect variables.
375
Implications for sustainable sourcing. Corporate and governmental policies for sustainable sourcing often pro-
376
mote certain sets of management practices without quantification of their actual impacts 57. In Unilever, the Sus-
377
tainable Agriculture Code (SAC) provides a mechanism for monitoring quantitative farm-level data over time13.
378
The findings of this study provide some evidence of a reduction in GHG emissions over time but repeated sampling
379
of the same farms over an extended number of years is required to fully understand the benefits of the scheme. The
380
large variability of GHG footprints within this study for sustainably sourced tomatoes partially reflects the range
381
of management practices that are acceptable in the Unilever SAC. If greater comparability of outcomes is required,
56
leading to a convergence of practice. Such differences are expected to
ACS Paragon Plus Environment
18
Page 19 of 25
Environmental Science & Technology
382
either within a scheme such as the Unilever SAC, or between schemes then more stringent guidance on acceptable
383
practices would be required. However, highly prescriptive approaches to certification could hinder adoption by
384
farmers and reduce the push for continuous improvement across the sector.
385
Implications for development of data collection platforms and GHG calculators. As the energy and fuel con-
386
sumption were reported as single figures, the impact of specific management practices, e.g. irrigation, harvesting,
387
tilling, etc. could not be quantified. Moreover, information for factors that have significant influence on variability
388
in GHG footprints58, such as genetic resources (varieties), farmer’s knowledge and habits (e.g. on fertilizer appli-
389
cation, planting dates, pest and disease control), were not available. We also relied on village or city names for
390
geolocation of farms, leading to uncertainty in the extraction of spatial-temporal parameters. Data collection plat-
391
forms, such as the Cool Farm Tool14, could seek to facilitate the gathering of this information. This would improve
392
the identification of drivers of GHG variability and allow development of more specific GHG management strate-
393
gies. Indeed, in the latest online version of the CFT 14, the user is able to input information regarding the energy
394
consumed for each type of agricultural process, including machine usage and irrigation. However, there is a trade-
395
off between obtaining more data for detailed analysis and increasing the burden on farmers for further data collec-
396
tion and reporting57. Based on this study, we suggest prioritizing data collection related to types and quantities of
397
farm management practices rather than aggregate energy consumption. The data collected should include the types
398
and number of passes for soil preparation activities or the types of irrigation and the amount of water use.
399 400
Data quality issues related to the CFT were discussed by Keller (2016)57 and Clavreul et al (2017)2. However, there
401
was no methodological guideline regarding how to assess the quality of the data before this analysis. The
402
methodology developed in this study, specifically the assignment of a data quality score based on the general
403
criterion of uniqueness and completeness of the observation, could be considered by the developers of data
404
collection platforms and for future data analysis by others 1.
405 406 407 408 ACS Paragon Plus Environment
19
Environmental Science & Technology
Page 20 of 25
409
ASSOCIATED CONTENT
410
Supporting Information. Data processing and cleaning methodology; Detailed methodology for derivation of
411
emission factors and calculation of GHG footprints; Data inclusion criteria for the different types of analysis;
412
Literature review of the typical growing period of open-field tomato proudction in different countries; Methodology
413
of the geocoding of farm locations and extraction of spatial-temporal parameters from spatial maps; Results of the
414
variability of GHG footprints; Results of the variability of yield, fertilizer application and energy use; Results of
415
the linear mixed model analysis; Results of the variability of GHG footprints with area of production and method
416
of fertilizer application supplied as Supporting Information are available free of charge via Internet at
417
http://pubs.acs.org.
418 419
AUTHOR INFORMATION
420
Corresponding Author
421
* Address: P.O. Box 9010 (mailbox no 89), NL-6500 GL Nijmegen, Phone: +31 (0)24 - 356 20 60; e-mail: lam-
422
[email protected] 423
Funding Sources
424
The study is part of the RELIability of product Environmental Footprints (RELIEF) project, which is funded by the Eu-
425
ropeans Union’s Horizon 2020 research and innovation program under the Marie Sklodowska-Curie grant agreement
426
No. 641459.
427
Notes
428
The authors declare no competing financial interest.
429
ACKNOWLEDGMENTS
430
The authors would like to thank Isabela Butnar, Lau Tamberg, Valerio Barbarossa, Zoran Steinmann and Mirza Čengić
431
for their support in data extraction and technical advice. We would also like to thank three anonymous reviewers for
432
their valuable input.
433
ACS Paragon Plus Environment
20
Page 21 of 25
Environmental Science & Technology
434
___________________________________________________________________________________
435
REFERENCES
436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481
1. Keller, E.; Chin, M.; Chorkulak, V.; Clift, R.; Faber, Y.; Lee, J.; King, H.; Milà i Canals, L.; Stabile, M.; Stickler, C.; Viart, N., Footprinting farms: a comparison of three GHG calculators. Greenhouse Gas Measurement and Management 2014, 4, (2-4), 90-123. 2. Clavreul, J.; Butnar, I.; Rubio, V.; King, H., Intra- and inter-year variability of agricultural carbon footprints – A case study on field-grown tomatoes. Journal of Cleaner Production 2017, 158, 156-164. 3. Pishgar-Komleh, S.; Akram, A.; Keyhani, A.; Raei, M.; Elshout, P.; Huijbregts, M.; van Zelm, R., Variability in the carbon footprint of open-field tomato production in Iran-A case study of Alborz and East-Azerbaijan provinces. Journal of Cleaner Production 2017, 142, 1510-1517. 4. Anton, A.; Torrellas, M.; Nunez, M.; Sevigne, E.; Amores, M. J.; Munoz, P.; Montero, J. I., Improvement of agricultural life cycle assessment studies through spatial differentiation and new impact categories: case study on greenhouse tomato production. Environ Sci Technol 2014, 48, (16), 9454-62. 5. Heidari, M. D.; Mobli, H.; Omid, M.; Rafiee, S.; Jamali Marbini, V.; Elshout, P. M. F.; Van Zelm, R.; Huijbregts, M. A. J., Spatial and technological variability in the carbon footprint of durum wheat production in Iran. The International Journal of Life Cycle Assessment 2017, 22, (12), 1893-1900. 6. Reidsma, P.; Ewert, F.; Lansink, A. O.; Leemans, R., Adaptation to climate change and climate variability in European agriculture: The importance of farm level responses. European Journal of Agronomy 2010, 32, (1), 91-102. 7. Robert, M.; Thomas, A.; Bergez, J.-E., Processes of adaptation in farm decision-making models. A review. Agronomy for Sustainable Development 2016, 36, (4), 64. 8. Manfredi, M.; Vignali, G., Life cycle assessment of a packaged tomato puree: a comparison of environmental impacts produced by different life cycle phases. Journal of Cleaner Production 2014, 73, 275-284. 9. Bystricky, M.; Baumgartner, D. U.; Gaillard, G.; Guggenberger, T.; Herndl, M., Identifying hotspots to improve the environmental performance of farms: Main drivers and their variability. 10th International Conference on Life Cycle Assessment of Food 2016 Book of Abstracts. Retrieved from: http://www.lcafood2016.org/wp-content/uploads/2016/10/LCA2016_BookOfAbstracts.pdf#344 on 13 December 2016. 2016. 10. Garson, G. D., Hierarchical Linear Modeling: Guide and Applications. SAGE Publications. California, USA: 2012. 11. Seltman, H. J., Experimental design and analysis. Retrieved from: http://www.stat.cmu.edu/~hseltman/309/Book/ on 2 February 2017. 2012. 12. World Processing Tomato Council, World production estimate as of 27 October 2016. Retrieved from: https://www.wptc.to/pdf/releases/WPTC%20World%20Production%20estimate%20as%20of%2027%2 0October%202016.pdf on 1st Jan 2017. 2016. 13. Smith, G.; Pendlington, D.; Vis, J. K.; Studley, S.; King, V., Sustainable Agriculture Code. Retrieved from: https://www.unilever.com/Images/sac-2015_tcm244-427050_en.pdf on 19th June 2016. 2015. 14. Hillier, J.; Walter, C.; Malin, D.; Garcia-Suarez, T.; Mila-i-Canals, L.; Smith, P., A farm-focused calculator for emissions from crop and livestock production. Environmental Modelling & Software 2011, 26, (9), 1070-1078. 15. Milà i Canals, L.; Rigarlsford, G.; Sim, S., Land use impact assessment of margarine. The International Journal of Life Cycle Assessment 2013, 18, (6), 1265-1277. 16. Nemecek, T.; Bengoa, X.; Lansche, J.; Mouron, P.; Rossi, V.; S., H., Methodological Guidelines for the Life Cycle Inventory of Agricultural Products. Version 2.0, July 2014. World Food LCA Database (WFLDB). Quantis and Agroscope, Lausanne and Zurich, Switzerland. 2014. 17. FAOSTAT, Food and Agriculture Organization of the United Nations Statistics Division. Retrieved from: http://faostat3.fao.org/home/E on 12 February 2016. 2015. ACS Paragon Plus Environment
21
Environmental Science & Technology
482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532
Page 22 of 25
18. Frischknecht, R.; Althaus, H. J.; Bauer, C.; Doka, G.; Heck, T.; Jungbluth, N.; Kellenberger, D.; Nemecek, T., The environmental relevance of capital goods in life cycle assessments of products and services. International Journal of Life Cycle Assessment 2007, 12, 7-17. 19. IPCC, Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press: Cambridge, United Kingdom and New York, NY, USA, 2013. 20. Weidema, B. P.; Bauer, C.; Hischier, R.; Mutel, C.; Nemecek, T.; Reinhard, J.; Vadenbo, C. O.; Wernet, G., The ecoinvent database: Overview and methodology, Data quality guideline for the ecoinvent database version 3, www.ecoinvent.org. 2013. 21. Unilever, Unilever Sustainable Sourcing Programme for Agricultural Raw Materials Scheme Rules version 4.7. Retrieved from: https://www.unilever.com/Images/sac-scheme-rules-01082017_tcm244502686_en.pdf on 9 November 2017. 2017. 22. ESRI, ArcGIS Desktop: Release 10.3.1. . Redlands, CA: Environmental Systems Research Institute. 2015. 23. Machado, R. M. A.; do Rosário, M.; Oliveira, G.; Portas, C. A. M., Tomato root distribution, yield and fruit quality under subsurface drip irrigation. Plant and Soil 2003, 255, (1), 333-341. 24. Patanè, C.; Cosentino, S. L., Effects of soil water deficit on yield and quality of processing tomato under a Mediterranean climate. Agricultural Water Management 2010, 97, (1), 131-138. 25. Monte, J. A.; Carvalho, D. F. d.; Medici, L. O.; Silva, L. D. B. d.; Pimentel, C., Growth analysis and yield of tomato crop under different irrigation depths. Revista Brasileira de Engenharia Agrícola e Ambiental 2013, 17, 926-931. 26. Zegbe, J. A.; Behboudian, M. H.; Clothier, B. E., Partial rootzone drying is a feasible option for irrigating processing tomatoes. Agricultural Water Management 2004, 68, (3), 195-206. 27. Harris, I.; Jones, P. D.; Osborn, T. J.; Lister, D. H., Updated high-resolution grids of monthly climatic observations – the CRU TS3.10 Dataset. International Journal of Climatology 2014, 34, (3), 623-642. 28. Wieder, W. R.; J. Boehnert; G.B. Bonan; M. Langseth, Regridded Harmonized World Soil Database v1.2. Data set. Retrieved from: http://daac.ornl.gov from Oak Ridge National Laboratory Distributed Active Archive Center, Oak Ridge, Tennessee, USA. 2014. 29. USGS, Global Multi-resolution Terrain Elevation Data (GMTED2010). Retrieved from: http://topotools.cr.usgs.gov/gmted_viewer/ on 30th October 2016. 2010 30. D. Bates; M. Maechler; B. Bolker; S. Walker, Fitting Linear Mixed-Effects Models Using lme4. Journal of Statistical Software, 67(1), 1-48. Accessed from: https://cran.rproject.org/web/packages/lme4/ on 11 August 2016. 2015. 31. Zuur, A. F.; Ieno, E. N.; Elphick, C. S., A protocol for data exploration to avoid common statistical problems. Methods in Ecology and Evolution 2010, 1, (1), 3-14. 32. Craney, T. A.; Surles, J. G., Model-Dependent Variance Inflation Factor Cutoff Values. Quality Engineering 2002, 14, (3), 391-403. 33. K. Bartoń, MuMIn: Multi-Model Inference. R package version 1.15.6. Retrieved from: https://cran.rproject.org/web/packages/MuMIn/index.html on 25 August 2016. 2016. 34. Burnham, K. P.; Anderson, D. R., Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach. Springer New York: 2007. 35. Symonds, M. R. E.; Moussalli, A., A brief guide to model selection, multimodel inference and model averaging in behavioural ecology using Akaike’s information criterion. Behavioral Ecology and Sociobiology 2011, 65, (1), 13-21. 36. Mazerolle, M. J., AICcmodavg: Model selection and multimodel inference based on (Q)AIC(c). R package version 2.1-0. Retrieved from: https://cran.r-project.org/package=AICcmodavg on 3 April 2017. 2016. 37. Nakagawa, S.; Schielzeth, H., A general and simple method for obtaining R2 from generalized linear mixed-effects models. Methods in Ecology and Evolution 2013, 4, (2), 133-142. 38. Barry, M.; Cashore, B.; Clay, J.; Fernandez, M.; Lebel, L.; Lyon, T.; Mallet, P.; Matus, K.; Melchett, P.; Vandenbergh, M.; Vis, J. K.; Whelan, T., Toward sustainability: the roles and limitations of ACS Paragon Plus Environment
22
Page 23 of 25
533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582
Environmental Science & Technology
certification, Final Report. Prepared by the Steering Committe of the State-of-Knowledge Assessments of Standards and Certification, Washington, DC 2012. 39. Theurl, M. C.; Haberl, H.; Erb, K.-H.; Lindenthal, T., Contrasted greenhouse gas emissions from local versus long-range tomato production. Agronomy for Sustainable Development 2014, 34, (3), 593-602. 40. Wernet, G.; Bauer, C.; Steubing, B.; Reinhard, J.; Moreno-Ruiz, E.; Weidema, B., The ecoinvent database version 3 (part I): overview and methodology. The International Journal of Life Cycle Assessment 2016, 21, (9), 1218-1230. 41. Nuppenau, E.-A., Modelling economies of scale, energy use and farm size to reduce GHG: On contrasting "High-Tec"-agriculture with labour intensive farming. European Association of Agricultural Economists 114th Seminar, April 15-16, 2010, Berlin, Germany. Retrieved from: http://purl.umn.edu/60912 on 18th Dec 2016. 2010. 42. Argote, L.; Epple, D., Learning Curves in Manufacturing. Science 1990, 247, (4945), 920-924. 43. Mahler, R. L., Fertilizer Placement. University of Idaho, College of Agriculture, Cooperative Extension System, Idaho Agricultural Experiment Station. Retrieved from: http://www.cals.uidaho.edu/edcomm/pdf/cis/cis0757.pdf on 18 Aug 2017. 1985. 44. Folger, P.; Cody, B. A., Drought in the United States: Causes and Current Understanding. Congressional Research Service. Retrieved from: http://www.rnrf.org/R43407_6_25_15.pdf on 5th December 2016. 2015. 45. Reddy, A. R. G.; Santosh, D.; Tiwari, K., Effect of Drip Irrigation and Fertigation on Growth, Development and Yield of Vegetables and Fruits. Int. J. Curr. Microbiol. App. Sci 2017, 6, (2), 14711483. 46. Saadi, S.; Todorovic, M.; Tanasijevic, L.; Pereira, L. S.; Pizzigalli, C.; Lionello, P., Climate change and Mediterranean agriculture: Impacts on winter wheat and tomato crop evapotranspiration, irrigation requirements and yield. Agricultural Water Management 2015, 147, 103-115. 47. Yuan, W.; Cai, W.; Chen, Y.; Liu, S.; Dong, W.; Zhang, H.; Yu, G.; Chen, Z.; He, H.; Guo, W.; Liu, D.; Liu, S.; Xiang, W.; Xie, Z.; Zhao, Z.; Zhou, G., Severe summer heatwave and drought strongly reduced carbon uptake in Southern China. 2016, 6, 18813. 48. Li, Y.; Chen, C.; Sun, C., Drought severity and change in Xinjiang, China, over 1961–2013. Hydrology Research 2016, 48, (5), 1343-1362. 49. Li, Y.; Zhang, W.; Ma, L.; Huang, G.; Oenema, O.; Zhang, F.; Dou, Z., An analysis of China’s fertilizer policies: impacts on the industry, food security, and the environment. Journal of environmental quality 2013, 42, (4), 972-981. 50. van Zanten, B. T.; Verburg, P. H.; Espinosa, M.; Gomez-y-Paloma, S.; Galimberti, G.; Kantelhardt, J.; Kapfer, M.; Lefebvre, M.; Manrique, R.; Piorr, A.; Raggi, M.; Schaller, L.; Targetti, S.; Zasada, I.; Viaggi, D., European agricultural landscapes, common agricultural policy and ecosystem services: a review. Agronomy for Sustainable Development 2014, 34, (2), 309-325. 51. Agriculture Victoria, Guide to Victorian Fertiliser Regulations. Agricultural and Veterinary Chemicals (Control of Use) (Fertilisers) Regulations 2016. Economic Development, Jobs, Transport and Resources. Accessed from http://agriculture.vic.gov.au/__data/assets/pdf_file/0010/321121/Guide-toVictorian-Fertiliser-Regulations.pdf on 3rd March 2017. 2016. 52. United States Environment Protection Agency, Agriculture: Nutrient Management and Fertilizer. United States Environment Protection Agency Website. Accessed from: https://www.epa.gov/agriculture/agriculture-nutrient-management-and-fertilizer on 3rd March 2017. 2017. 53. Amery, F.; Schoumans, O. F., Agricultural phosphorus legislation in Europe. Institute for Agricultural and Fisheries Research (ILVO). Retrieved from: http://library.wur.nl/WebQuery/wurpubs/fulltext/300160 on 3 March 2017. 2014. 54. van Grinsven, H. J. M.; ten Berge, H. F. M.; Dalgaard, T.; Fraters, B.; Durand, P.; Hart, A.; Hofman, G.; Jacobsen, B. H.; Lalor, S. T. J.; Lesschen, J. P.; Osterburg, B.; Richards, K. G.; Techen, A. K.; Vertès, F.; Webb, J.; Willems, W. J., Management, regulation and environmental impacts of nitrogen fertilization ACS Paragon Plus Environment
23
Environmental Science & Technology
583 584 585 586 587 588 589 590 591 592 593
Page 24 of 25
in northwestern Europe under the Nitrates Directive; a benchmark study. Biogeosciences 2012, 9, (12), 5143-5160. 55. Monteny, G. J., The EU Nitrates Directive: A European Approach to Combat Water Pollution from Agriculture. The Scientific World Journal 2001, 1, 927-935. 56. Foster, A. D.; Rosenzweig, M. R., Learning by Doing and Learning from Others: Human Capital and Technical Change in Agriculture. Journal of Political Economy 1995, 103, (6), 1176-1209. 57. Keller, E. J. GHG management in agri-food supply chains: A focus at the farm level. University of Surrey (United Kingdom), 2016. 58. Kulak, M.; Nemecek, T.; Frossard, E.; Gaillard, G., How Eco-Efficient Are Low-Input Cropping Systems in Western Europe, and What Can Be Done to Improve Their Eco-Efficiency? Sustainability 2013, 5, (9), 3722.
594 595
ACS Paragon Plus Environment
24
Page 25 of 25
Environmental Science & Technology
TOC Art 74x36mm (120 x 120 DPI)
ACS Paragon Plus Environment