Subscriber access provided by Oakland University Libraries
Article
What is the contribution of city-scale actions to the overall food system’s environmental impacts? Assessing water, GHG, and land impacts of future urban food scenarios Dana Boyer, and Anu (Anuradha) Ramaswami Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b03176 • Publication Date (Web): 13 Sep 2017 Downloaded from http://pubs.acs.org on September 17, 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 27
1 2 3 4 5 6 7 8 9 10 11
Environmental Science & Technology
What is the contribution of city-scale actions to the overall food system’s environmental impacts? Assessing water, GHG, and land impacts of future urban food scenarios Dana Boyer,1* Anu Ramaswami1 *Corresponding author,
[email protected] 1 Center for Science, Technology, and Environmental Policy, Hubert H. Humphrey School of Public Affairs, University of Minnesota, Minneapolis, Minnesota 55455, United States
ABSTRACT
12 13
This paper develops a methodology for individual cities to analyze the in- and trans-boundary
14
water, greenhouse gas (GHG) and land impacts of city-scale food system actions. Applied to
15
Delhi, India, the analysis demonstrates that city-scale action can rival trans-boundary action,
16
though no single city-scale action can rival in all three environmental impacts. Particularly
17
improved food waste management within the city (7% system-wide GHG reduction) matches the
18
GHG impact of pre-consumer trans-boundary food waste reduction. The systems approach is
19
particularly useful in illustrating key tradeoffs/co-benefits. For instance, multiple diet shifts
20
achieving GHG reduction are accompanied by an increase in water and/or land impacts. Vertical
21
farming technology (VFT) with current application for fruits and vegetables can provide modest
22
system-wide water (4%) and land reductions (3%), though implementation within the city may
23
raise questions of constraints in water stressed cities—with such a shift in Delhi increasing
24
community-wide direct water use by 16%. Improving the nutrition status for the bottom 50% of
25
the population to the median diet, is accompanied by proportionally smaller increases of water,
26
GHG and land impacts (4%, 9%, and 8%)—increases that can be offset through simultaneous
27
city-scale actions, e.g. improved food waste management and VFT.
28 29 30
INTRODUCTION
31 32
The global food supply is heavily dependent on energy, water, and land resources. Seventy to
33
85% of global consumptive water loss is attributed to food production, with 19-29% of global
34
greenhouse gas (GHG) emissions emitted from food system activities spanning production 1 ACS Paragon Plus Environment
Environmental Science & Technology
Page 2 of 27
35
through consumption.1,2 Similarly, 12% of global ice-free land area is for agricultural cultivation,
36
with a further 28% for pasture.3 With such high resource use, the environmental impact of the
37
food system has received substantial attention at national and global scales.4 Attention to these
38
interactions from a city perspective, however, has received little attention,5 with the important
39
question being: the extent to which city-scale actions shape the larger food system’s
40
environmental impact, particularly in terms of GHG, water, and land.
41 42
Despite historic city involvement in food system planning,6–8 with the rise of industrialization
43
and refrigerated freight cities became seen as lesser players in shaping the food system9—with
44
the food system being associated predominately with agriculture, rather than expanded to include
45
not only production, but consumption, transport, retail and food waste management activities,
46
acknowledging the linkages between production and demand. Recent conceptualizations of the
47
food system that include cities as demand centers,10 embed the city within the larger food system
48
that extends well beyond the urban boundary.5
49 50
Facilitated through programs such as the Milan Food Pact,11 the UN FAO’s Food for the Cities12
51
and C40’s Food System Network,13 the concerns of urban food systems are many and diverse. At
52
the individual city level, municipalities have begun developing objectives and initiatives aimed
53
at addressing a multiplicity of urban food concerns, ranging from improving nutrition, health of
54
diet, food access/equity, environmental sustainability, management of food waste, and increasing
55
local production.14–16 With the city embedded in the larger trans-boundary food system (see
56
Figure 1), city-scale food system actions are likely to create changes in the environmental impact
57
both locally, and beyond the city boundary. City-scale food system actions are those initiatives
58
taken by actors within the city, and could include, changes in consumer behaviors, diets and
59
purchasing, urban production, and infrastructure changes of water, energy, and waste
60
management that interact with the food system. Cities, however, do not yet have analytical tools
61
to assess the extent to which city-scale food system actions affect GHG, water and land impacts
62
through the whole of the food system. Such analysis is essential to understand how the multitude
63
of urban food system objectives may coincide, or conflict, with concerns of environmental
64
sustainability and resource constraints facing the food system both within, and beyond, city
65
boundaries. 2 ACS Paragon Plus Environment
Page 3 of 27
Environmental Science & Technology
66 67
Conceptually, city-scale food system actions can be expected to have large impact on the global
68
food system. Urban areas are expected to hold two thirds of the 2050 projected population17 and
69
currently responsible for substantial environmental impact, with urban energy use accounting for
70
75% of the global GHG emissions18 and large water demand.19 Individual cities have also
71
reported the embodied energy of the food supply. Denver, for example, reports food related
72
emissions as 10% of its total community-wide infrastructure GHG emissions20 while Delhi
73
reports a comparably high 15%.21 Similarly, San Francisco reports that food accounts for 19% of
74
total GHG emission footprint of households.22
75
metabolism reports found that, across accounting methods, the food sector, on average, had the
76
third largest mass and carbon flow by sector, suggesting consistently high impact of urban food
77
use. Environmental resource-use footprinting in terms of water has also found the food sector to
78
make large contribution.24–26 The embodied land of city food supply specifically has not been
79
reported in the scientific literature in a manner disaggregated from ecological footprints, with the
80
exception of some grey literature analyses of smaller UK cities.27
A recent analysis23 of 100 cities’ urban
81 82
Thus city-scale action could be highly impactful in shaping the overall environmental impacts of
83
the food system on water, GHG and land. Recent advances in urban footprinting methodology
84
by Ramaswami et al. 2017 has provided a first framework to capture both in- (i.e. transport
85
within the city, in-boundary waste management, urban agriculture) and trans-boundary (i.e.
86
agricultural production beyond city boundary) impacts of community-wide urban food demand
87
on water and GHG.5 This paper builds upon that framework to incorporate land impacts to assess
88
the nexus between water, energy and land, and focuses on conducting a scenario analysis to
89
assess the impact of a range of city-scale actions on the overall environmental footprints of the
90
food system.
91 92
The key question that this study asks is: what is the contribution of city-scale actions (i.e.
93
actions taken within individual cities) to the overall food system’s environmental impacts
94
within, and outside, the city boundary? The environmental impacts of focus are water use,
95
energy use/GHG and land. The actions include those of multiple actors within the community of
3 ACS Paragon Plus Environment
Environmental Science & Technology
Page 4 of 27
96
consumers (households), producers, and community-wide infrastructure providers (i.e. waste
97
managers).
98 99
Applied to Delhi, India, this study examines the actions of various diet changes, greater equity of
100
household food consumption, increased urban agriculture, improved cooking fuel and food waste
101
management, each of which is compared with the reference trans-boundary intervention of
102
reducing food waste along the pre-consumer supply chain. Thus, actions occurring within the
103
city boundary are compared with pre-consumer actions typically occurring outside of the city
104
boundary at a first order level. It should be noted that the motivation for these city-scale actions
105
is not solely mitigation of environmental impact, and may include other societal priorities such
106
as improved food access, nutritional equity, and support of local food production.
107 108
The development of the linked water-energy/GHG-land footprinting along with scenario analysis
109
of city-scale actions is applied to Delhi, India to develop the methodology, though is
110
generalizable to any city where the requisite data are available. Benefits of the methodology
111
include: ability to compare the environmental impact both within, and beyond the city boundary
112
of city-scale action against supply chain and/or production level interventions; informing of
113
policy and action in data poor environments; determining largest city policy levers that can
114
provide water, GHG and land benefits; and identifying instances of complementary or
115
conflicting sustainability outcomes (i.e. equity versus multiple environmental impacts).
116 117
METHODS
118 119
Delhi (in the year 2011; population of 16.4 million) serves as a case study to develop and
120
implement the methodology due to a uniquely large amount of food-related data in the public
121
domain (See SI). Furthermore, the authors were successful in obtaining city food supply chain
122
data, not readily available in many cities, allowing for use of spatially explicit resource intensity
123
factors.5 This allows scenario assessment to determine the extent of impact that city-scale actions
124
have on the larger food system, without the impediment of limited data.
125
4 ACS Paragon Plus Environment
Page 5 of 27
Environmental Science & Technology
126
The trans-boundary framework of community-wide food supply to a city assessing linked water-
127
GHG/energy-land impacts is shown in Figure 1. This figure illustrates the different components
128
of the food system, as they intersect with the city (illustrated by the dotted square that includes
129
home, businesses and industries) allowing differentiation between in- versus trans-boundary
130
impacts.
131 132
Baseline
133 134
Total annual community-wide food use for homes, business, and industry is calculated with data
135
first reported by Ramaswami et al. 2017.5 This work further expands the food use analysis,
136
delineating use by households of different socio-economic-class (SEC), as determined by the
137
Indian Government by monthly per capita expenditure. The authors obtained important data on 5 ACS Paragon Plus Environment
Environmental Science & Technology
Page 6 of 27
138
all components of the food system of agricultural production (in- and outside of the city), freight
139
transport, use (household, commercial, industrial) and food waste management (see SI for
140
detailed methods). Taking such a community-wide footprinting approach20,28,29 allows
141
visualization of potentially beneficial synergies between sectors—greater potential than can be
142
realized by looking at each sector in isolation.
143 144
India-specific water consumptive loss (blue and green)30 and GHG31 resource intensity factors of
145
production literature studies were supplemented with second order impacts from spatially
146
explicit electricity-use and diesel-use for irrigation and on farm machinery.5 Thus, the water
147
embodied in electricity production (power plant cooling) and fuel is included in addition to water
148
use in crop irrigation. The GHG of production does not include emissions as a result of land use
149
change.
150 151
The land footprint of production was calculated in line with methods described by various past
152
studies32,33 as a function of community-wide agri-food demand and crop specific yields. India-
153
specific crop yields were obtained from FAOSTAT34 and Government of India reports.35,36
154
International agri-food imports are relatively small for India, and verified with Delhi’s food
155
supply chain data.37 Potential uncertainties include knowledge of multi-cropping and lack of
156
country specific animal land area requirements, further discussed in the SI. However, most land-
157
footprinting estimates in the literature face similar constraints.33
158 159
Methods for determining resource impact of freight transport, use (home use, commercial
160
preparation, industrial processing), and food waste management (solid and waste water) are
161
detailed in the SI.
162 163
Scenario analysis
164 165
The scenario analysis quantifies changes in the annual GHG, water and land footprints of Delhi’s
166
community-wide food demand from the baseline case, with several city-scale food system
167
actions, motivated by a variety of goals described below and summarized in Table 1. Detailed
168
methods for each scenario are included in the SI. 6 ACS Paragon Plus Environment
Page 7 of 27
Environmental Science & Technology
169 170
(a) City equity & health scenario action: Addressing hunger and food security is an explicit
171
objective of the Sustainable Development Goals.38 This is particularly relevant in India, a
172
country home to nearly 25% of the world’s undernourished population.39 In Delhi’s baseline case
173
specifically, the top 5% of the population consumes 2.5 times by mass the quantity of food
174
consumed by the lowest 5%. Similarly, the highest 5% consumes much greater quantities of
175
fruits and vegetables than the lowest 5%, resulting in greater dietary diversity (92 kg versus 46
176
kg per capita per year).5 This scenario therefore assesses the resulting impact of the lower 50%
177
of the population by SEC consuming the same diet as the median SEC population—a diet
178
deemed sufficient in terms of calories and protein.40
179 180
(b, c, d, e) City diet change scenario action: Much research exists to recommend “sustainable
181
diets”—i.e. a diet that meets the nutritional needs of the population while minimizing
182
environmental impact.41,42 This research, however, generally speaks only to the Western context,
183
assuming a high baseline meat and dairy consumption and availability of alternatives.43,44
184
Applicability of these recommendations is questionable in developing country settings, with a
185
research need to examine “sustainable diet” shifts in developing countries, conscious of local
186
food system conditions and limitations.44 A decrease in rice consumption is of potential interest
187
in the Indian context due to: a) in-country concern over high white rice consumption linked to
188
diabetes;45 b) rice’s large environmental impact;31 and c) available alternatives within the Delhi
189
context (wheat, sorghum, millet). Actions in the diet change scenarios therefore explore the
190
impact of a 100% shift from rice to wheat, sorghum, and millet, on a calorie-to-calories basis. A
191
shift from meat to pulses is also assessed to provide quantitative comparison with North
192
American and European analyses.
193 194
(f, g, h) Urban agriculture scenario action: Urban agriculture is promoted globally for a variety
195
of reasons ranging from various environmental benefits to mitigation of urban malnutrition and
196
livelihood generation.6,7 Benefits are claimed for both conventional (soil-based) farming methods
197
and vertical technology (VFT) (i.e. hydroponics, aero-ponics). VFT, in particular, is promoted
198
for high water savings and its independence from land/soil quality.46,47 VFT has received some
199
attention in India,48,49 where water scarcity and land degradation can present a challenge for 7 ACS Paragon Plus Environment
Environmental Science & Technology
Page 8 of 27
200
conventional production methods. While types of urban agriculture vary greatly (i.e. community
201
gardens v. residential plots v. rooftop etc.), In Delhi, urban agriculture currently takes the form of
202
large agricultural lands along the Yamuna River cutting through the middles of the city. Three
203
urban agriculture scenario actions therefore assess: 1) a doubling of current soil-based urban
204
practices (some fruits, vegetables, milk, grains); 2) conversion of current in-boundary agriculture
205
production to VFT (without any increase in production quantity); and 3) conversion of all viable
206
crops within the whole of Delhi’s food supply (in- and trans-boundary) to VFT, to be grown
207
within Delhi.
208 209
(i) City food preparation scenario action: Fumes of dirty cooking fuels, (i.e. firewood and cow
210
dung) have large impact on human health, causing respiratory issues, particularly for women and
211
children.50,51 In Delhi, 22% of household cooking energy is provided by dirty fuels, with use
212
concentrated in the lower SECs.52 Thus the Government of India provides household subsidies to
213
allow greater access of liquefied petroleum gas (LPG) as a cleaner alternative.53 This scenario
214
therefore determines the impact of a shift of all current dirty household cooking fuels to LPG.
215 216
(j, k) City food waste management scenario action: While India is estimated to have lesser
217
consumer level food waste than developed countries, organic kitchen scraps contribute over 50%
218
of total municipal solid waste generated and 80% of household waste.54,55 This results in
219
accumulation of food waste on the streets, in landfills and associated methane emissions. This
220
scenario action explores the impact change with alternative food waste management options of:
221
j) composting and k) anaerobic digestion (AD),56 producing organic fertilizer and in the case of
222
AD, electricity generation. Land change of this scenario was not assessed due to data limitation.
223 224
l) Trans-boundary reference scenario action: Pre-consumer food waste is a particular problem in
225
India due to poor food transport and storage infrastructure, such as lack of refrigeration.57–59 This
226
provides much scope for waste reduction along the supply prior to the city boundary, estimated
227
as high as 40-50% in developing countries.60 This scenario analyzes the impact if the percentage
228
of pre-consumer food waste is decreased to levels of international best practice. This scenario
229
provides a point of comparison of how a lever outside of the city boundary compares with that of
230
city-scale actions. 8 ACS Paragon Plus Environment
Page 9 of 27
Environmental Science & Technology
231 232
If the impact of city-scale actions ((a) through (k)) rival those of trans-boundary reduction of
233
agri-food waste (action l), it is of high significance, indicating the role a city can play in shaping
234
the larger food system.
235 236 Table 1 | Summary of city-scale scenario actions to be analyzed for change of water, GHG and land impacts. Category Scenario a. Lower 50% of the population by socio-economic-class (SEC) City equity & health consuming same diet as median SEC population (deemed sufficient in (a) terms of calories and protein) b. 100% rice calories to wheat c. 100% rice calories to sorghum City diet change (b, c, d, e) d. 100% rice calories to millet e. 100% meat calories to pulses f. Doubling of current urban agriculture production Urban agriculture g. Shift of all viable in-boundary production to VFT (holding production (f, g, h) levels constant) h. Shift of all viable crops to VFT (to be grown in Delhi) City food preparation i. Conversion of all cooking fuels in Delhi to LPG (i) City food waste j. All household organics to composting management k. All household organics to anaerobic digestion (j, k) Trans-boundary reference scenario: Pre-consumer food l. Decrease of pre-consumer food waste to international best practice waste (l)
237 238 239 240 241 242 243 244 245 246 9 ACS Paragon Plus Environment
Environmental Science & Technology
247
Page 10 of 27
RESULTS
248 249
Baseline
250 251
1. Baseline system-wide impacts
252
Figure 2 illustrates the whole system-wide GHG (16.9 mil t CO2e), water (17,592 mil m3), and
253
land (4.8 mil ha) impacts of Delhi’s baseline food system.
254
Different portions of the system are illustrated by the various colors, with the hatched bar portion
255
signifying trans-boundary impact and the solid bar portions indicating in-boundary impact. The
256
figure illustrates that for both water and land, total impact is dominated by the production stage
257
(>99% for both). GHG impact exhibits a more even distribution throughout the whole of the
258
supply chain, with only 50% attributed to the production stage, and substantial in-boundary
259
emissions from households mainly from cooking-related emissions (12%), commercial (5%)
260
food preparation (cooking fuel emissions), industrial processing (5%) and food waste emissions
261
from both solid waste landfill methane emissions (6%) and liquid waste (1%).
262 263 10 ACS Paragon Plus Environment
Page 11 of 27
Environmental Science & Technology
264
2. Distribution of agricultural production impacts by food types
265
Figure 3 delineates the production impacts into specific food categories. The vertical bars of the
266
figure (Figure 3a-e) show the percentage contribution of each food category to the total impact in
267
terms of mass, nutritional intake, GHG, water and land by each of the respective food categories.
268
The horizontal bars (Figure 3f) on the far right illustrate the percentage of each food group (and
269
associated impact) occurring in- versus trans-boundary. This helps to demonstrate which food
270
categories are exhibiting high impact as a result of high quantity of consumption in Delhi versus
271
high resource requirement of production. For instance, wheat (burnt orange) contributes 34% of
272
total calories, versus a proportionally lesser 11% of GHG impact. Milk (light blue) clearly
273
dominates in terms of GHG, with 36% of the total environmental impact of production, with
274
substantial contributions to the water (25%) and land (21%) production footprints as well. Meat
275
(light purple) makes very little contribute across all footprints due to low consumption.
276 277
11 ACS Paragon Plus Environment
Environmental Science & Technology
Page 12 of 27
278 279 280
3. Delineation of in-versus trans-boundary impact
281
Figure 3f illustrates the in- versus trans-boundary distribution of food production by mass to
282
meet Delhi’s food demand with the horizontal bars on the right of the figure. This provides
283
understanding into the distribution of localized versus trans-boundary impact. Currently 8% of
284
food by mass is produced within the city, with substantial in-boundary reliance on meat and milk
285
products, provisioning 45% and 15% of total Delhi demand and lesser contributions to rice,
286
vegetable and fruit. Thus, while there are some levels of urban production, the majority of
287
resource impact occurs beyond the city boundary. This baseline analysis is also important to 12 ACS Paragon Plus Environment
Page 13 of 27
Environmental Science & Technology
288
inform reasonable localization strategies—currently one of the most commonly promoted food
289
system interventions.6,7
290 291
4. Impact distribution by socio-economic class
292
Figure 4 illustrates the large differences that exist between SECs in terms of water, GHG and
293
land impact of food demand. Each of the pie slices represents 10% of the population, with the
294
wedge size proportionate to the contribution of that class to the whole of Delhi’s food resource
295
impact. Thus an “equitable” society in terms of resource use would have uniformly sized slices.
296
The lowest (5th) and highest classes (100th) are outlined to illustrate the large difference between
297
the lowest and highest classes. For instance, in the case of GHG, the wealthiest class is
298
responsible for 17% of the impact, while the lowest class only 6%.
299
300
Scenario analysis
301 302
1. System-wide impact
303
Figure 5 summarizes the change in full system-wide GHG, water, and land impacts with each of
304
the city-scale food system actions described in Table 1. The bottom axis indicates the percentage
305
change of the total food system’s (in- plus trans-boundary) land (green), water (blue) and GHG
306
(red) footprint with each of the food system interventions ((a) through (l) as described in Table
307
1). Second order impacts are indicated with hatching. 13 ACS Paragon Plus Environment
Environmental Science & Technology
Page 14 of 27
308 14 ACS Paragon Plus Environment
Page 15 of 27
Environmental Science & Technology
309
City equity & health: When the bottom 50% of the population consumes the diet of the median
310
class (Figure 5a), system-wide GHG, water and land impacts increase by 4%, 9%, and 8%,
311
respectively.
312 313
City diet change: A 100% shift of rice to wheat reduces total system-wide water and GHG
314
impacts of 3% each with a slight increase in land footprint (