Science of the Total Environment 599600 (2017) 474482

Contents lists available at ScienceDirect

Science of the Total Environment

journal homepage: www.elsevier.com/locate/scitotenv

Virtual water trade of agri-food products: Evidence from

italian-chinese relations
Lucrezia Lamastra a, Pier Paolo Miglietta b,, Pierluigi Toma b, Federica De Leo b, Stefania Massari b
a
Institute of Agricultural and Environmental Chemistry (ICAA), Universit Cattolica del Sacro Cuore, 29122 Piacenza, Italy
b
Department of Economics and Management, University of Salento, 73100 Lecce, Italy

H I G H L I G H T S G R A P H I C A L A B S T R A C T

Water footprint and virtual water can be

used as quantitative measurement of
the appropriation of water by humans
activities.
International trade and virtual water
use are two closely inter-connected ac-
tivities.
China imports water from Italy trough
animal products, instead Italy import
water from China trough crop products.
The water ow involved in the Italy-
China trade of agri-food products im-
plies a global water loss of 129.29 mil-
lion m3 at a global scale.
The agri-food policies should therefore
include an analysis of the effects on
water resources.

a r t i c l e i n f o a b s t r a c t

Article history: At global scale, the majority of world water withdrawal is for the agricultural sector, with differences among
Received 9 March 2017 countries depending on the relevance of agri-food sector in the economy. Virtual water and water footprint
Received in revised form 19 April 2017 could be useful to express the impact on the water resources of each production process and good with the ob-
Accepted 19 April 2017
jective to lead to a sustainable use of water at a global level. International trade could be connected to the virtual
Available online xxxx
water ows, in fact through commodities importation, water poor countries can save their own water resources.
Editor: D. Barcelo The present paper focuses on the bilateral virtual water ows connected to the top ten agri-food products traded
between Italy and China. Comparing the virtual water ow related to the top 10 agri-food products, the virtual
Keywords: water ow from Italy to China is bigger than the water ow in the opposite direction. Moreover, the composition
Water footprint of virtual water ows is different; Italy imports signicant amounts of grey water from China, depending on the
Virtual water different environmental strategies adopted by the two selected countries. This difference could be also related to
Trade the fact that traded commodities are very different; the 91% of virtual water imported by Italy is connected to
Agri-food crops products, while the 95% of virtual water imported by China is related to the animal products. Considering
China
national water saving and global water saving, appears that Italy imports virtual water from China while China
Italy
exerts pressure on its water resources to supply the exports to Italy. This result at global scale implies a global
water loss of 129.29 million m3 because, in general, the agri-food products are traded from the area with
lower water productivity to the area with the higher water productivity.
2017 Elsevier B.V. All rights reserved.

Corresponding author.
E-mail address: pierpaolo.miglietta@unisalento.it (P.P. Miglietta).

http://dx.doi.org/10.1016/j.scitotenv.2017.04.146
0048-9697/ 2017 Elsevier B.V. All rights reserved.
 L. Lamastra et al. / Science of the Total Environment 599600 (2017) 474482 475

Fig. 1. EU-China Trade in goods () 20132015.

Source: Elaboration from European Commission, 2016.

1. Introduction
Freshwater availability is not unlimited and water scarcity is grow-
ing because of economic and population growth, climate change and
the increasing water demand (Miglietta et al., 2016).
Humans are currently consuming 54% of all of the earth's accessible
freshwater available in rivers, lakes and underground aquifers
(UNWATER, 2016). The present estimates indicate that, by 2025,
water stress, a condition in which water is not sufcient to meet agricul-
tural, industrial or domestic needs (Lamastra, 2015), will be a reality for
almost half of the world's population. Added to the problems of local
and regional water shortages, there is the problem of water pollution
which renders enormous volumes of water unsuitable for civil and
non-civil use. Pollution threatens the quality of water resources and is
also linked to demographic growth and to the access to the market of
large swathes of populations previously excluded from mass consump-
tion, which has resulted in an increase in production and in the manage-
ment of its waste.
2. Background
Between 1900 and 2010, global agriculture water withdrawal con-
sumption increased from 600 km3/year to about 2700 km3/year (FAO,
2016a) and the forecast for 2050 predicts an increase of about 20% if
there is not an improvement in the efciency of the agricultural produc-
tion systems (UNWATER, 2016). An important indicator which captures
this efciency is water productivity. It dened as the ratio of agricultural
output to the total amount of water consumed (crop per drop) (Kijne
et al., 2003; Molden et al., 2010),
In this context, it is important to have a way to express the impact on
the water resources of each production process and consumer good
with the objective to guide and urge for the most sustainable use of
water. Water footprint and virtual water concepts were born to set a
quantitative measurement of the appropriation of water by human ac-
tivities, both representing the volume of freshwater virtually
contained in a product. The water content, in this case, represents all
the water used and polluted throughout the production process. How-
ever, the water footprint, differently by the virtual water, expresses
and distinguishes the different types of water used and it is spatiotem-
poral explicit (Bonamente et al., 2016; Lamastra et al., 2014).
International trade and water use are two closely inter-connected
activities. The increasing importance of international trade, especially
in agricultural products, placed freshwater issues in a global context
that should be analyzed and regulated by sustainable policies. The
effects of international trade in virtual water are very important, espe-
cially for water-scarce countries which, through commodities importa-
tions, can save their own water resources. According to Hoekstra
(2010), if water-intensive commodities were traded from countries
with high water productivity to countries with low water productivity
there would be a reduction of 5% in the global water use. In fact, if one
country exports a product to another country, it also exports virtual
water.
About 81% of virtual water is traded by China, the United States and
the European Union which compose the world's trading centre (Chen
and Chen, 2013).
Commodity trade plays a signicant role in redistributing water re-
sources between nations and it is conrmed by the fact that one-third
of global withdrawal represents the total virtual water embodied in in-
ternational trade (Chen and Chen, 2013).
This paper addresses the issues of freshwater scarcity, its relation
with trade and the need for international regulation and focuses on
the bilateral virtual water trade of agricultural products with the case
study of Italy and China.
In Italy, the distribution of water withdrawal is 44.07% for the agri-
cultural sector, 35.67% for the industrial sector and 20.06% for the mu-
nicipal sector (Gleick and Ajami, 2014). In China, instead, 64.61% is
used for the agricultural sector, 23.21% for the industrial sector and
12.19% for the municipal sector (FAO, 2016b).
China is one of the most important trading partner for the EU and
Italy, especially for imports. Europe and Italy, in fact, have a trade
decit with China as shown in Figs. 1 and 2. The trade ow between
Italy and China is characterized by industrial products, but there is
not a suitable procedure to calculate their virtual water content.
For these reasons the research focuses on agricultural products that
also contribute to the global virtual water ow more than the indus-
trial ones.
The aim of this research is to assess the bilateral virtual water trade
in agricultural products between Italy and China measuring the amount
of water saved on a national and global level with a distinction of green,
blue and grey water.

Fig. 2. Italy-China Trade in goods () 20132015.

Source: Elaboration from Eurostat, 2016.
476 L. Lamastra et al. / Science of the Total Environment 599600 (2017) 474482

Table 1
Top 10 imported products of Italy, 20102015.

Virtual Volume of trade (tonnes) Virtual water ow (millionm3)

water
Product
content
code
in 2010 2011 2012 2013 2014 2015 Average 2010 2011 2012 2013 2014 2015 Average
(HS)
China
3
(m /t)

Dried leguminous
vegetables, shelled,
1 0713 2458 39,835 54,751 38,366 46,309 41,835 42,021 43,853 97.91 134.58 94.30 113.83 102.83 103.29 107.79
whether or not skinned or
split
Pigs', hogs' or boars'
bristles and hair; badger
2 0502 hair and other brush 14,163 778 739 550 548 679 639 656 11.02 10.47 7.79 7.76 9.62 9.05 9.28
making hair; waste of
such bristles or hair
Vegetables, uncooked or
3 0710 cooked by steaming or 704 13,274 13,655 11,643 12,430 10,077 11,232 12,052 9.34 9.61 8.20 8.75 7.09 7.91 8.48
boiling in water, frozen
Vegetable saps and other
mucilages and thickeners
4 1302 derived from vegetable 9779 851 825 688 614 677 587 707 8.32 8.07 6.73 6.00 6.62 5.74 6.91
products, whether or not
modied
Other oil seeds and
5 1207 oleaginous fruits, whether 3251 759 3,856 924 1888 1,220 1056 1617 2.47 12.54 3.00 6.14 3.97 3.43 5.26
or not broken
Other nuts, fresh or dried,
6 0802 whether or not shelled or 5521 716 783 739 972 984 952 858 3.95 4.32 4.08 5.37 5.43 5.26 4.74
peeled
Vegetables provisionally
preserved, e.g. by sulphur
dioxide gas, in brine, in
sulphur water or in other
7 0711 499 10,306 11,009 7607 9335 8500 8369 9188 5.14 5.49 3.80 4.66 4.24 4.18 4.58
preservative solutions,
but unsuitable in that
state for immediate
consumption
Dried apricots, prunes,
apples, peaches and other
8 0813 edible fruits, mixtures of 8395 140 152 137 227 728 910 382 1.18 1.28 1.15 1.91 6.11 7.64 3.21
edible and dried fruits or
of edible nuts
Soya beans, whether or
9 1201 3136 145 73 107 141 843 3890 867 0.45 0.23 0.34 0.44 2.64 12.20 2.72
not broken
Dried vegetables, whole,
cut, sliced, broken or in
10 0712 704 2803 3275 3,103 2499 2253 1821 2626 1.97 2.31 2.18 1.76 1.59 1.28 1.85
powder, but not further
prepared

3. Materials and methods The volume of surface and groundwater consumed to make a prod-
uct is called blue water and it is equal to (Hoekstra and Chapagain,
The virtual water ow is dened as the amount of water embodied 2011):
in exported or imported products from a country to another country.
The following formula indicates how to calculate the amount of the vir-
tual water ow referred to a product: WFproc;blue Blue W:Evap: Blue W:Incorp:
Lost Return Flow volume=time
Virtual water flowprod Trade flowprod  Virtual Water contentprod
The volume of freshwater required to assimilate the pollution of a
The concept of virtual water is different from that of the water foot- production process is called grey water and is equal to (Hoekstra and
print. The former refers only to the water volume embodied in a product Chapagain, 2011):
and the latter is a multidimensional indicator which considers where
the water footprint is located, which source of water is used and when
the water is used, but in this context virtual water content of a product L
WFproc;gray volume=time
can be considered as its water footprint (Hoekstra and Chapagain, cmax cnat
2011). The water footprint is composed by the green, blue and grey
water footprints of direct and indirect use of a producer or consumer.
L = pollutant load [mass/time]
The volume of rainwater consumed to make a product is called
cmax = maximum acceptable concentration [mass/volume]
green water and it is equal to (Hoekstra and Chapagain, 2011):
cnat = natural concentration in the receiving water body [mass/
volume]
WFproc;green Green Water evaporation The national water saving Sn[p] indicates the amount of water saved
Green Water incorporation volume=time by a nation as a result of trade in a product p and it is equal to (Hoekstra
 L. Lamastra et al. / Science of the Total Environment 599600 (2017) 474482 477

104,313 109,609 106,801 127,402 115,929 90,572 109,104 1,552.80 1,631.64 1,589.84 1,896.51 1,725.72 1,348.25 1,624.13
Average

171.58
133.20

68.76

16.08
3.93
3.63

3.28

0.92
0.74
137.44
90.78

44.91

16.18

12.65
2015

6.68
5.24

1.38
3.49
125.29
103.64

56.71

15.35
2014

6.03
8.57

5.61

1.38
0.69
169.08
140.87

72.63

13.12
2013

4.29
4.79

0.94

0.93
0.15
Virtual water ow (millionm3)

138.40
148.72

89.55

19.60
2012

3.21
0.86

0.44

0.60
0.09
Fig. 3. Structure of Italian virtual water imports referred to Italian Top 10 products,
20102015.
221.02
142.77

74.36

18.50
2011

1.93
1.68

0.03

0.29
0.01 and Chapagain, 2011):
238.27
172.44

Sn p Ti pTe p  WFprod p volume=time

74.41

13.74
Average 2010

1.43
0.66

0.02

0.93
0.02

where:
13,699 20,100

26,918 26,757
8,753

7477

1098
5555

12,512 3245

1837
19,061 4045

Ti[p] = volume of imported product p [p.units/time]

Te[p] = volume of exported product p [p.units/time]
WFprod[p] = water footprint of product p in the nation considered
2015

7011

4883

1867
8013

2758

If Sn is a positive number the nation saves water, on the contrary if Sn

is a negative one the nation loses water.
15,639

25,544

13104

The global water saving Sg[ne, ni,p] corresponds to the sum of the na-
2014

6391

6166

1686

5545

2769
3764

tional savings of all the nations and it is also equal to (Hoekstra and
Chapagain, 2011):
21,257

21,832
2013

8625

7897

1199
7322

1857
926

806

Sg ne ; ni ; p Tne ; ni ; p

 WFprod ni ; pWFprod ne ; p volume=time
22,441

32,615
Virtual water Volume of trade (tonnes)

2012

7060

9737

1314

1198
898

434

508

where:
T[ne, ni, p] = volume of product trade [p. units/time].
11,274
21,543

30,778
2011

8085

2567

WFprod[ni, p] = water footprint of product p in the importing nation.

538

586
33

34

WFprod[ne, p] = water footprint of product p in the exporting nation.

If the importing nation cannot produce p, it is necessary to consider
12,154
26,021

22,854
in Italy (m3/t) 2010

8091

1012

1855

the difference between the global average and the WF in the exporting
400

18

98

country (Hoekstra and Chapagain, 2011).

If Sg[ne, ni,p] is negative there is a global water loss.
The data sources of this analysis are elaborated over the period
content

14,886
19,604

20102015 and consist of:

6627

9197

3579

1011
601

654

500
183

the volume of agri-food products trade between Italy and China for
Fruit juices (and grape must) and vegetables juice, no spirit
Olive oil, virgin and its fractions rened but not chemically

products with code at a four-digit level (International Trade Centre,

Milk and cream, not concentrated nor containing added

2016);
the virtual water content of crops and derived crop products
(Mekonnen and Hoekstra, 2010a);
Raw hides & skins of bovine or equine animals

Fresh strawberries and other fruits

sugar or other sweetening matter

Top 10 imported products of China, 20102015.

Bovine or equine leather

Beer made from malt

Wine of fresh grapes
Cheese and curd
Swine leather

modied
Product

4104
4107
4101

1509

2204
0406
0810

0401

2009
10 2203
code
(HS)
Table 2

Fig. 4. Structure of Chinese virtual water imports referred to Chinese Top 10 products,
1
2
3

5
6
7

20102015.
478 L. Lamastra et al. / Science of the Total Environment 599600 (2017) 474482
Fig. 5. Composition of virtual water ow of Italian Top 10 imports from China, 2010
2015. (For interpretation of the references to color in this gure legend, the reader is re-
ferred to the web version of this article.)
the virtual water content of animals and derived animal products
(Mekonnen and Hoekstra, 2010b, 2012).
According to the Harmonized Commodity Description and Coding
System, also known as the Harmonized System (HS) of tariff nomencla-
ture, Mekonnen and Hoekstra analyzed the water content of products
coded at a six-digit level. In our study this water content has been
brought to a four-digit level by calculating the average of water contents
at the correspondent six-digit level (Jiang and Marggraf, 2015).
Virtual water ow between China and Italy has been analyzed by
selecting the Top 10 agri-food products imported by Italy and China.
The ow has been calculated considering for each selected product the
virtual water content and the amount of tonnes traded between Italy
and China in the period 20102015. This time span, in fact, highlights
the intensication of Italian-Chinese trade relations. The Top 10 prod-
ucts that Italy imports are represented in Table 1, while the Chinese
Top 10 products are represented in Table 2.
4. Results and discussion
4.1. Virtual water trade ow
Comparing global virtual water ow Italy and China have a different
structure. Italy imports 101,416.4 million m3/yr and exports
39,259.9 million m3/yr of water, so Italy is a net importer; China, on
the contrary, imports 119,231.6 million m3/yr and exports
Fig. 6. Composition of virtual water ow of Chinese Top 10 imports from Italy, 2010
2015. (For interpretation of the references to color in this gure legend, the reader is re-
ferred to the web version of this article.)
Fig. 7. Composition of virtual water ow of Italian Top 10 imports from China in %, 2010
2015. (For interpretation of the references to color in this gure legend, the reader is re-
ferred to the web version of this article.)
142,697 million m3/yr, so it is net exporter (Mekonnen and Hoekstra,
2011).
As can be seen in these tables, Italy imports a lot of crop products and
derived products that have a high water content as vegetable extracts
(9779 m3/t), nuts (5521 m3/t), dried fruits (8395 m3/t), dried legumes
(2458 m3/t) and soya beans (3136 m3/t). Dried legumes are the most
imported products with 42,021 t in 2015 and an average virtual water
ow of 107.79 million m3. Bristles and hair of pig, hog, and boars are
in second place due to their water content of 12,153 m3/t, even if Italy
imports just 656 t on average. China imports more animal products
than Italy. Bovine/equine leather, swine leather and raw hides and
skins of bovine/equine animals are in rst place with an average virtual
water ow respectively of 1624.13 million, 171.58 million and
133.20 million of m3. It is important to emphasize the presence of
some typical Made in Italy products among the Chinese Top 10 prod-
ucts and, in particular, olive oil, wine and beer. Between 2010 and
2015, Italy exported to China on average 7477 t of olive oil, 26,757 t of
wine and 4045 t of beer and in that period wine exportation increased
by 17.78%, the amount of beer increased from 98 t to 19,061 t, while
olive oil exports, instead, decreased by 40%.
As shown in Figs. 3 and 4, the structure of the virtual water ow be-
tween Italy and China, referred to the selected products, is very differ-
ent. In Italy about 91% of water import is related to crop products and
in China, instead, about 95% of water import is related to animal
products.
It is important to distinguish between green, blue and grey compo-
nents of the virtual water ows. The compositions of virtual water
Fig. 8. Composition of virtual water ow of Italian Top 10 imports from China in %, 2010
2015. (For interpretation of the references to color in this gure legend, the reader is re-
ferred to the web version of this article.)
Table 3
Italian national water savings/losses.

Volume of Italian IMPORT (tonnes) Volume of Italian EXPORT (tonnes)

Product code (HS) Virtual water content in Italy (m3/t)
2010 2011 2012 2013 2014 2015 2010 2011 2012 2013 2014 2015

1 0713 1,876 39,835 54,751 38,366 46,309 41,835 42,021 8 229 210 357 392 239
2 0502 14,703 778 739 550 548 679 639 1 12
3 0710 507 13,274 13,655 11,643 12,430 10,077 11,232 0 5 1 8
4 1302 16,259 851 825 688 614 677 587 274 624 496 456 304 336
5 1207 4,365 759 3,856 924 1,888 1,220 1,056 6 0 63
6 0802 6,216 716 783 739 972 984 952 189 151 79 88 142 139
7 0711 642 10,306 11,009 7,607 9,335 8,500 8,369 0 85 8 6 1
8 0813 2,825 140 152 137 227 728 910 1 0 3 43 83
9 1201 1,336 145 73 107 141 843 3,890
10 0712 512 2,803 3,275 3,103 2,499 2,253 1,821 4 17 64 36 12 7
1 4104 14,886 486 658 17 92 33 73 104,313 109,609 106,801 127,402 115,929 90,572

L. Lamastra et al. / Science of the Total Environment 599600 (2017) 474482

2 4107 19,604 70 119 190 98 72 88 12,154 11,274 7,060 8,625 6,391 7,011
3 4101 6,627 34 47 16 94 0 26,021 21,543 22,441 21,257 15,639 13,699
4 1509 9,197 0 1 0 8,091 8,085 9,737 7,897 6,166 4,883
5 2204 601 38 0 8 15 8 22,854 30,778 32,615 21,832 25,544 26,918
6 0406 3,579 5 400 538 898 1,199 1,686 1,867
7 0810 654 47 58 42 59 43 31 1,012 2,567 1,314 7,322 13,104 8,013
8 0401 1,011 18 33 434 926 5,545 12,512
9 2009 500 1,235 296 115 162 22 3,328 1,855 586 1,198 1,857 2,769 2,758
10 2203 183 3,141 3,123 3,165 3,248 3,371 3,224 98 34 508 806 3,764 19,061

Table 3 (continued)

IMPORT-EXPORT (tonnes) ITALIAN NATIONAL WATER SAVINGS/ LOSSES (mln m3)

2010 2011 2012 2013 2014 2015 2010 2011 2012 2013 2014 2015 Average

1 39,827 54,522 38,156 45,952 41,443 41,782 74.72 102.28 71.58 86.21 77.75 78.38 81.82
2 678 627 9.97 9.22 9.59
3 13,274 13,650 12,429 11,224 6.73 6.92 6.30 5.69 6.41
4 577 201 192 158 373 251 9.38 3.27 3.12 2.57 6.06 4.08 4.75
5 3,850 1,888 993 16.81 8.24 4.33 9.79
6 527 632 660 884 842 813 3.28 3.93 4.10 5.49 5.23 5.05 4.51
7 10,306 7,522 9,327 8,494 8,368 6.62 4.83 5.99 5.45 5.37 5.65
8 151 137 224 685 827 0.43 0.39 0.63 1.94 2.34 1.14
9
10 2,799 3,258 3,039 2,463 2,241 1,814 1.43 1.67 1.56 1.26 1.15 0.93 1.33
1 103,827 108,951 106,784 127,310 115,896 90,499 1,545.57 1,621.84 1,589.59 1,895.14 1,725.23 1,347.17 1,620.76
2 12,084 11,155 6,870 8,527 6,319 6,923 236.89 218.68 134.68 167.16 123.88 135.72 169.50
3 25,987 21,496 22,425 15,545 13,699 172.22 142.45 148.61 103.02 90.78 131.42
4 9,737 7,896 6,166 89.55 72.62 56.71 72.96
5 22,816 30,778 21,824 25,529 26,910 13.71 18.50 13.12 15.34 16.17 15.37
6 533 1.91 1.91
7 965 2,509 1,272 7,263 13,061 7,982 0.63 1.64 0.83 4.75 8.54 5.22 3.60
8
9 620 290 1,083 1,695 2,747 570 0.31 0.15 0.54 0.85 1.37 0.29 0.49
10 3,043 3,089 2,657 2,442 393 15,837 0.56 0.57 0.49 0.45 0.07 2.90 0.15

479
 480
Table 4
Chinese national water savings/losses.

Volume of Chinese import (tonnes) Volume of Chinese export (tonnes)

Product code (HS) Virtual water content in China (m3/t)
2010 2011 2012 2013 2014 2015 2010 2011 2012 2013 2014 2015

1 0713 2,458 8 229 210 357 392 239 39,835 54,751 38,366 46,309 41,835 42,021
2 0502 14,163 1 12 778 739 550 548 679 639
3 0710 704 0 5 1 8 13,274 13,655 11,643 12,430 10,077 11,232
4 1302 9,779 274 624 496 456 304 336 851 825 688 614 677 587
5 1207 3,251 6 0 63 759 3,856 924 1,888 1,220 1,056
6 0802 5,521 189 151 79 88 142 139 716 783 739 972 984 952
7 0711 499 0 85 8 6 1 10,306 11,009 7,607 9,335 8,500 8,369
8 0813 8,395 1 0 3 43 83 140 152 137 227 728 910
9 1201 3,136 145 73 107 141 843 3,890
10 0712 704 4 17 64 36 12 7 2,803 3,275 3,103 2,499 2,253 1,821
1 4104 14,326 104,313 109,609 106,801 127,402 115,929 90,572 486 658 17 92 33 73

L. Lamastra et al. / Science of the Total Environment 599600 (2017) 474482

2 4107 18,885 12,154 11,274 7,060 8,625 6,391 7,011 70 119 190 98 72 88
3 4101 6,316 26,021 21,543 22,441 21,257 15,639 13,699 34 47 16 94 0
4 1509 4,405 8,091 8,085 9,737 7,897 6,166 4,883 0 1 0
5 2204 692 22,854 30,778 32,615 21,832 25,544 26,918 38 0 8 15 8
6 0406 5,877 400 538 898 1,199 1,686 1,867 5
7 0810 997 1,012 2,567 1,314 7,322 13,104 8,013 47 58 42 59 43 31
8 0401 1,664 18 33 434 926 5,545 12,512
9 2009 1,301 1,855 586 1,198 1,857 2,769 2,758 1,235 296 115 162 22 3,328
10 2203 152 98 34 508 806 3,764 19,061 3,141 3,123 3,165 3,248 3,371 3,224

Table 4 (continued)

Import-export (tonnes) Chinese national water savings/losses (mlnm3)

2010 2011 2012 2013 2014 2015 2010 2011 2012 2013 2014 2015 Average

1 39,827 54,522 38,156 45,952 41,443 41,782 97.89 134.02 93.79 112.95 101.87 102.70 107.20
2 678 627 9.60 8.88 9.24
3 13,274 13,650 12,429 11,224 9.34 9.61 8.75 7.90 8.90
4 577 201 192 158 373 251 5.64 1.97 1.88 1.55 3.65 2.45 2.86
5 3,850 1,888 993 12.52 6.14 3.23 7.29
6 527 632 660 884 842 813 2.91 3.49 3.64 4.88 4.65 4.49 4.01
7 10,306 7,522 9,327 8,494 8,368 5.14 3.75 4.65 4.24 4.18 4.39
8 151 137 224 685 827 1.27 1.15 1.88 5.75 6.94 3.40
9
10 2,799 3,258 3,039 2,463 2,241 1,814 1.97 2.29 2.14 1.73 1.58 1.28 1.83
1 103,827 108,951 106,784 127,310 115,896 90,499 1,487.43 1,560.83 1,529.79 1,823.84 1,660.33 1,296.49 1,559.78
2 12,084 11,155 6,870 8,527 6,319 6,923 228.21 210.66 129.74 161.03 119.33 130.74 163.29
3 25,987 21,496 22,425 15,545 13,699 164.13 135.77 141.64 98.18 86.52 125.25
4 9,737 7,896 6,166 42.89 34.78 27.16 34.94
5 22,816 30,778 21,824 25,529 26,910 15.79 21.30 15.10 17.67 18.62 17.70
6 533 3.13 3.13
7 965 2,509 1,272 7,263 13,061 7,982 0.96 2.50 1.27 7.24 13.02 7.96 5.49
8
9 620 290 1,083 1,695 2,747 570 0.81 0.38 1.41 2.21 3.57 0.74 1.27
10 3,043 3,089 2,657 2,442 393 15,837 0.46 0.47 0.40 0.37 0.06 2.41 0.13
 L. Lamastra et al. / Science of the Total Environment 599600 (2017) 474482
ows referring to Italian and Chinese Top 10 products are shown in
Figs. 5 and 6.
Comparing the water ows, it is important to underline that China
imports on average about 2026.26 million of m3 of water from Italy
and exports on average about 158.83 million m3 of water. Therefore,
the virtual water ow from Italy to China is bigger than the water
ow in the opposite direction.
Comparing the green, blue, grey percentage on the annual water
ow (Figs. 7 and 8) it is evident that green water has the largest share
of total virtual water ow, in particular, it is on average 72% for Italian
imports and 86% for Chinese imports. Blue water takes the smallest
share and accounts for the 1% for Italy and 6% for China. The main differ-
ences are in the grey water ows: in fact, the 26% of water imported by
Italy is grey, while just the 8% of the water imported by China is grey.
Considering that grey water is the freshwater volume needed to assim-
ilate pollution, this aggregated results suggest that, in general, Italian
production has fewer environmental impacts than Chinese production.
This aspect has to be considered by Chinese environmental regulation,
but also by enterprises, that have to promote and adopt more environ-
mental practices in order to reduce pollution and water consumption.
4.2. Water saving through trade
The national water savings have been calculated for Italy and China
taking into account the Top 10 products exchanged. The results
(Tables 3 and 4) show that Italy has a national water loss in the period
20102015 on average of 1891.15 million m3 and China has a national
water savings on average of 1761.85 million m3. At global scale, there
is a global water loss on average of 129.29 million m3.
Three specic products affect the results more than the others.
Bovine/equine leather (HS code 4104) trade produces the biggest
loss, as we can see from Tables 3 and 4. The production of one tonne
of bovine/equine leather in Italy requires 14,886 m3 of water
(12,895 m3 green, 827 m3 blue and 1164 m3 grey), whereas in China
are required 14,326 m3 of water (13,328.3 m3 green, 583.2 m3 blue
and 414 m3 grey) (Mekonnen and Hoekstra, 2010b). Since Italy
imported on average 226.5 t and exported on average 109,104 t, the bal-
ance is 108,878 t. The result shows that Italy loses water on average for
1620.76 million m3 and China, instead, saves water for
1559.78 million m3 and so, there is a global water loss of
60.97 million m3. There is a loss because the product is traded from
the area with lower water productivity to the area with higher water
productivity and it is important to underline that, in this case, consider-
ing the grey water footprint Chinese production is also cleaner than the
Italian one.
The second product, highlighted in Tables 3 and 4, is olive oil (HS
code 1509) that requires 9197 m3/t of water (8569 m3 green, 612 m3
blue, 16 m3 grey) in Italy and 4405 m3/t of water (3738 m3 green
67 m3 blue, 601 m3 grey) in China. Italy imported just one tonne of oil
in 2011 but exported on average 7477 t (Mekonnen and Hoekstra,
2010a). This trade produces a global water loss of 38.01 million m3.
Even if Italy consumes a larger quantity of water to produce olive oil,
its production is cleaner than the Chinese (16 m3 in Italy vs 601 m3 of
grey water in China).
The third products are dried vegetables (HS code 0713). Italy im-
ports on average 43,853 t and exported on average 239 t, so the balance
is 43,614 t. Dried vegetables require 1876 m3/t (1848 m3 green, 28 m3
grey) of water in Italy and 2458 m3/t (1739 m3 green, 719 m3 grey) of
water in China (Mekonnen and Hoekstra, 2010a). Also in this case, con-
sidering the grey water footprint, Italian production is more water
friendly than the Chinese one. The results are a national water savings
on average of 81.82 million m3 for Italy and a national water loss of
107.20 million m3 for China, which produces a global water loss of
25.38 million m3 in the period considered.
According to the theory of international commerce, countries have
to try to gain prots by exporting commodities that are produced
481
through their abundantly available resources, importing products
whose raw materials are scarce. The large volume of international virtu-
al water ows and associated external water dependencies can repre-
sent the global approach to water scarcity issues (De Leo and
Miglietta, 2012; De Leo et al., 2013). In the case of China-Italy trade
there is a global water loss, that has to be considered to increase, at a
global level, the water productivity. Since environmental sustainability
of countries depends on water resource savings, it becomes essential
to consider virtual water as an alternative source and an opportunity
deriving from the growing liberalization of international commerce
(De Leo and Miglietta, 2015).
5. Conclusions
In this paper is shown that, from a national point of view, Italy im-
ports, through agri-food products, virtual water from China, more
than it exports. China, on the other hand, exerts pressure on its water re-
sources to supply the exports to Italy, being a net exporter of agricultur-
al products. We also demonstrate that the agricultural trade between
the Italy and China implies a global water loss.
These empirical results have no intentions of evaluating the good-
ness of the revealed virtual water ows. Italy does not import water in
virtual form because it intends to save domestic water resources but,
by importing virtual water, through agri-food trade, Italy saves domes-
tic water resources. This result does not imply that saving water re-
sources has to be an incentive for creating dependence or
externalizing negative effects of water use in other Countries without
paying.
The international trade of agri-food commodities is very complex
and overcoming the mere water strategy approach, depends on a lot
of other factors such as availability of land, labour, knowledge and cap-
ital, competitiveness, national subsidies and taxes (Hoekstra and
Chapagain, 2007).
Policy-makers and water managers cannot identify the best choices
and smart strategies focusing just on water scarcity, virtual water and
water footprint trade (Wichelns, 2015) because international virtual
water trade cannot be explained solely on the basis of territorial water
abundances or shortages (Yang et al., 2006).The study, however, dem-
onstrates that, since agri-food commodity inuences the water appro-
priation in a country, agri-food trade policy should, therefore, include
an analysis of the effects on water resources. In fact, the assessment of
bilateral relations between Italy and China suggests that trade manage-
ment represents an option not only to increase economic value but also
to contribute to the mitigation of global water loss.
In recent years, the scientic and civil interest in sustainable water
use has grown signicantly, but the weak state of transparency and in-
formation, regarding the direct and indirect water use, hinders an effec-
tive contribution in reducing water pressure, achieving a sustainable
water allocation at a relevant scale. We are able to quantify and map vir-
tual ows and footprints, but we are still far away from understanding
how to manage footprints with a sustainable approach (Hoekstra
et al., 2015).
References
Bonamente, E., Scrucca, F., Rinaldi, S., Merico, M.C., Asdrubali, F., Lamastra, L., 2016. Envi-
ronmental impact of an Italian wine bottle: carbon and water footprint assessment.
Sci. Total Environ. 560:274283. http://dx.doi.org/10.1016/j.scitotenv.2016.04.026.
Chen, Z.M., Chen, G.Q., 2013. Virtual water accounting for the globalized world economy:
national water footprint and international virtual water trade. Ecol. Indic. 28:
142149. http://dx.doi.org/10.1016/j.ecolind.2012.07.024.
De Leo, F., Miglietta, P.P., 2015. The contribution of environmental indicators to the sus-
tainable performance of countries. Int. J. Sustain. Policy Pract. 11:1226. http://dx.
doi.org/10.18848/2325-1166/CGP/v11i04/55359.
De Leo, F., Miglietta, P.P., 2012. Water footprint and virtual water trade of olive oil. Pro-
ceedings of the 18th IGWT Symposium Technology and Innovation for a Sustainable
Future: a Commodity Science Perspective. Rome, 2428 September 2012. Communi-
cation Unit of ENEA - IGWT - AISME, Rome:pp. 112 http://dx.doi.org/10.13140/2.1.
4295.2483.
482 L. Lamastra et al. / Science of the Total Environment 599600 (2017) 474482
De Leo, F., Miglietta, P.P., Toma, P., 2013. Il virtual water trade balance come indicatore di
sostenibilit del settore vitivinicolo italiano. In: Guido, G., Massari, S. (Eds.), Lo
Sviluppo Sostenibile. Ambiente, Risorse, Innovazione, Qualit. Franco Angeli, Mila-
no:pp. 485496 http://dx.doi.org/10.13140/RG.2.1.4438.2240.
European Commission, 2016. Trade Policy. http://ec.europa.eu/trade/policy/countries-
and-regions/countries/china/ (accessed 20.10.16).
Eurostat, 2016. International trade in goods. http://ec.europa.eu/eurostat/data/database
(accessed 20.10.16).
FAO, 2016a. Aquastat. http://www.fao.org/nr/water/aquastat/data/query/index.html?
lang=en (accessed 20.10.16).
FAO, 2016b. Acquastat. http://www.fao.org/nr/water/aquastat/countries_regions/chn/
index.stm, (accessed 20.10.16).
Gleick, P.H., Ajami, N., 2014. The World's Water Volume 8: The Biennial Report on Fresh-
water Resources. Island press, Washington.
Hoekstra, A.Y., 2010. The relation between international trade and freshwater scarcity.
World Trade Report 2010 on Trade in Natural Resources: Challenges in Global Gov-
ernance https://www.wto.org/english/res_e/reser_e/ersd201005_e.pdf.
Hoekstra, A.Y., Chapagain, A.K., 2011. The Water Footprint Assessment Manual: Setting
The Global Standard. Earthscan, London.
Hoekstra, A.Y., Chapagain, A.K., 2007. The water footprints of Morocco and the
Netherlands: global water use as a result of domestic consumption of agricultural
commodities. Ecol. Econ. 64:143151. http://dx.doi.org/10.1016/j.ecolecon.2007.02.
023.
Hoekstra, A.Y., Chapagain, A.K., Zhang, G., 2015. Water footprints and sustainable water
allocation. Sustain. For. 8:20. http://dx.doi.org/10.3390/su8010020.
International Trade Centre, 2016. Title - Trade Map - Trade statistics for international
business development. http://www.trademap.org/Bilateral_TS.aspx (accessed
20.10.16).
Jiang, W., Marggraf, R., 2015. Bilateral virtual water trade in agricultural products: a case
study of Germany and China. Water Int. 40:483498. http://dx.doi.org/10.1080/
02508060.2015.1022848.
Kijne, J.W., Barker, R., Molden, D.J., 2003. Water Productivity in Agriculture: Limits and
Opportunities for Improvement. IWMI and CABI Publisher, Wallingford (UK).
Lamastra, L., Suciu, N.A., Novelli, E., Trevisan, M., 2014. A new approach to assessing the
water footprint of wine: an Italian case study. Sci. Total Environ. 490:748756.
http://dx.doi.org/10.1016/j.scitotenv.2014.05.063.
Lamastra, L., 2015. The virtual water in a bottle of wine. In: Antonelli, M., Greco, F. (Eds.),
The Water We Eat. Springer International Publishing, Cham (CH), pp. 209228.
Mekonnen, M.M., Hoekstra, A.Y., 2012. A global assessment of the water footprint of farm
animal products. Ecosystems 15:401415. http://dx.doi.org/10.1007/s10021-011-
9517-8.
Mekonnen, M.M., Hoekstra, A.Y., 2011. National Water Footprint Accounts: The Green,
Blue and Gray Water Footprint of Production and Consumption - Value of Water Re-
search Report Series 50. UNESCO-IHE, Delft (NL).
Mekonnen, M.M., Hoekstra, A.Y., 2010a. The Green, Blue and Gray Water Footprint of
Crops and Derived Crop Products - Value of Water Research Report Series 47.
UNESCO-IHE, Delft (NL).
Mekonnen, M.M., Hoekstra, A.Y., 2010b. The Green, Blue and Gray Water Footprint of
Farm Animals and Animal Products - Value of Water Research Report Series 48.
UNESCO-IHE, Delft (NL).
Miglietta, P.P., De Leo, F., Toma, P., 2016. Environmental Kuznets curve and the water
footprint: an empirical analysis. Water Environ. J. 31:2030. http://dx.doi.org/10.
1111/wej.12211.
Molden, D., Oweis, T., Steduto, P., Bindraban, P., Hanjra, M.A., Kijne, J., 2010. Improving ag-
ricultural water productivity: between optimism and caution. Agr. Water. Manage.
97:528535. http://dx.doi.org/10.1016/j.agwat.2009.03.023.
UNWATER, 2016. Agriculture is the biggest water user, with irrigation accounting for 70%
of global water withdrawals. http://www.unwater.org/statistics/statistics-detail/en/
c/246663/ (accessed 20.10.16).
Wichelns, D., 2015. Virtual water and water footprints do not provide helpful insight re-
garding international trade or water scarcity. Ecol. Indic. 52:277283. http://dx.doi.
org/10.1016/j.ecolind.2014.12.013.
Yang, H., Wang, L., Abbaspour, K.C., Zehnder, A.J.B., 2006. Virtual water trade: an assess-
ment of water use efciency in the international food trade. Hydrol. Earth Syst. Sci.
10:443454. http://dx.doi.org/10.5194/hess-10-443-2006.