New patterns in China’s water footprint: Analysis of spatial and structural transitions from a regional perspective

Journal Pre-proof New patterns in China's water footprint: Analysis of spatial and structural transitions from a regional perspective Yiling Xiong, Xin Tian, Shangwei Liu, Zhipeng Tang PII:

S0959-6526(19)33812-0

DOI:

https://doi.org/10.1016/j.jclepro.2019.118942

Reference:

JCLP 118942

To appear in:

Journal of Cleaner Production

Received Date: 7 May 2019 Revised Date:

3 September 2019

Accepted Date: 17 October 2019

Please cite this article as: Xiong Y, Tian X, Liu S, Tang Z, New patterns in China's water footprint: Analysis of spatial and structural transitions from a regional perspective, Journal of Cleaner Production (2019), doi: https://doi.org/10.1016/j.jclepro.2019.118942. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Ltd.

New patterns in China’s water footprint: Analysis of spatial and structural transitions from a regional perspective

Yiling Xiong1, Xin Tian1,2*, Shangwei Liu1, Zhipeng Tang3

1.School of Environment, Beijing Normal University, Beijing 100875, China 2.State Key Laboratory of Water Environment Simulation, Beijing Normal University, Beijing 100875, China 3.Key Laboratory of Regional Sustainable Development Modeling, Institute of Geography Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China.

* Corresponding Author: Xin Tian, [email protected], +86-10-58800397, Beijing Normal University, Beijing 100875, China

1

Abstract

2

China faces a major challenge to support the operation and development of an

3

enormous economy with limited water resources. Decoding the way that the final

4

demand influences water consumption patterns is helpful for understanding the

5

discrepancy between water supply and demand. Especially in the current time, when

6

China has experienced socioeconomic transition and its government has proposed a

7

series of water conservation policies over the years, exploring the transition in the

8

water consumption pattern is necessary for further water conservation in China.

9

Meanwhile, there exist distinct economic levels and stages of development among

10

provinces, making it necessary to develop more refined water management strategies

11

in the various provinces. This study comprehensively investigates the trend, spatial

12

distribution and structure of the water footprint in mainland China. Furthermore, it

13

explores the effects of socioeconomic factors on transitions in water footprint

14

patterns with the aid of multi-regional input–output analysis, structural

15

decomposition analysis, and linear regression. The results show that the water

16

footprints of almost all the provinces were decoupled from economic growth from

17

2007 to 2012. Several northern provinces dramatically increased their water

18

footprints, further enlarging the difference between north and south. The north also

19

showed a wider urban–rural gap in per capita water footprint than did the south.

20

From the viewpoint of structural transitions, household consumption (as the primary

21

contributor) remained stable while an obvious increase in the water footprint was

22

induced by fixed capital formation, in which agriculture and strategic emerging

23

manufacturing played important roles. From the perspective of driving forces, water

24

intensity improvement and changes in the structure of the final demand saved water.

25

However, these effects were counteracted by the change in the final demand level

26

and the composition of the final products. By identifying China’s water footprint

27

patterns and its transitions, our results clarify China’s strategic water demand pattern

28

and provide refined insights for future water conservation.

2

29

Keywords: water footprint, China, multi-regional input–output analysis, structural

30

decomposition analysis, socioeconomic transition

3

31

1. Introduction

32

As a strategic resource, water is of vital importance for regional stability and

33

security, and water shortages are one of the major challenges for China’s

34

development (Cheng et al., 2009; Jiang, 2009). This is evinced in two aspects. First,

35

China’s renewable internal freshwater resources per capita have been grossly

36

inadequate, at only approximately one-third of the global level in the past five

37

decades (World Bank, 2014). Second, the increased population and the economic

38

expansion have placed growing pressures on water resources in recent years. China’s

39

water use increased by 51.2 billion m3 from 2000 to 2017 (NBS, 2017, 2000).

40

Therefore, the balance of China’s water resources between supply and demand has

41

become a pressing issue. It is crucial to understand the effects of China’s current

42

water consumption on its water resources.

43

China has been experiencing rapid industrialization and urbanization over

44

recent years (Huang, 2018; Ren et al., 2019), during which socioeconomic transition,

45

characterized by changes in patterns of production and consumption, may have had

46

considerable impacts on water consumption patterns. The effect of socioeconomic

47

factors on water consumption is also influenced by water conservation policies

48

implemented by the Chinese government. From the viewpoint of production patterns,

49

China is entering the late period of industrialization with significant transitions in its

50

industrial structure (Huang, 2014), which may alter the structural features of water

51

consumption. Changes in technology and the production structure may also

52

influence water consumption in production processes. In addition, the government

53

has laid emphasis on improving water productivity and promoting water

54

conservation in production processes, proposed a policy regarding water-saving

55

technologies (NDRC, 2005) and focused on improving water-use efficiency in key

56

fields such as agricultural irrigation (NDRC, 2017) and industry (The State Council,

57

2000). From the perspective of consumption patterns, transitions in residential

58

lifestyles lead to higher demand and a more diverse demand structure in the process

59

of urbanization, which may induce changes in household water consumption. 4

60

Meanwhile, the government has also paid increasing attention to water conservation

61

in consumption patterns. To achieve a water-saving society (NDRC, 2012, 2007), a

62

more integrated management system is being established involving total quantity

63

control and quota management of water use (DRCSC, 2012; NDRC, 2007) and

64

strengthening compensable use of water resources (The State Council, 2006). More

65

comprehensive measures have also been gradually applied to effectively encourage

66

water saving, such as by means of price adjustments (The State Council, 2008). In

67

this context, the need to understand the changes in China’s water consumption under

68

the dual influences of socioeconomic transition and government policies is urgent.

69

The transition in water consumption patterns may present diverse features and

70

trends in different provinces within China, owing to the uneven spatial distributions

71

of developmental levels and water resources. There is a mismatch between water

72

resources and economic levels in China. For example, coastal regions presented a

73

higher gross domestic product (GDP) than western and central regions, ranging from

74

28.4 thousand yuan/capita (Gansu) to 129.0 thousand yuan/capita (Beijing) in 2017

75

(NBS, 2017). At the same time, water resources ranged from 83.5 m3/capita (Tianjin)

76

to 13138.8 m3/capita (Qinghai) among thirty provinces in 2017 (NBS, 2017), with a

77

pattern of richer water resources in the middle-south and southwest. For this reason,

78

it is important to identify key regions with serious water issues and to develop more

79

refined water management plans. Meanwhile, owing to imbalances in the economic

80

development levels and positions among provinces, the flow of products and

81

services has increased within China. Water flows across provinces through being

82

embodied in products and services, which allows a province’s water demand to be

83

partially removed from dependency on local water resources (Allan, 1998, 1993). In

84

spite of reduced dependency, the total volume of water consumed to produce all the

85

products and services to satisfy provincial demands, i.e., the water footprint, reflects

86

the spatial feature of the demands for water resources (Hoekstra et al., 2009). Hence,

87

given the strategic value of water resources and the tremendous cost of water

88

diversion projects, it is necessary to reveal regional strategic demands for water and 5

89

to further develop distinct regional water management strategies from a water

90

footprint perspective.

91

There have arisen a number of studies on water footprints. To quantify the

92

impacts of human activities on water resource in quantity and quality, scholars

93

explored different types of water footprints, such as blue water and grey water (Liao

94

et al., 2019; Wang and Yang, 2018). The water footprint can be quantified by two

95

types of approaches. One is a “bottom-up” approach based on production processes

96

and is widely applied in accountings of water footprints of specific products such as

97

agricultural products (Chapagain and Hoekstra, 2002) and livestock products

98

(Chapagain and Hoekstra, 2003). The other is a “top-down” approach represented by

99

an input–output model, which has been growing in popularity and has been used in

100

combination with water consumption data for recent years. In comparison to the

101

bottom-up approach, the input–output method better reflects the network of

102

relationships among sectors and covers both direct and indirect water consumption

103

(Feng et al., 2011). Furthermore, a multi-regional input–output model enables

104

decoding both local and external footprints of the region. These give input–output

105

analysis the advantage of calculating the water footprint along with the full supply

106

chain across sectors and regions.

107

Scholars identify water footprint features, such as sectoral structures, sources,

108

and trends, by means of input–output analysis. On the one hand, there are numerous

109

scholars who have devoted to investigating water footprint traits of single regions in

110

China. These studies involve various spatial scales such as city (Han et al., 2015;

111

Zhang et al., 2011), province (Dong et al., 2013; Liu et al., 2017; Okadera et al.,

112

2015), basin (White et al., 2015) and nation (Wang and Yang, 2018), and have

113

tended to focus on regions with serious water shortage such as North China (S. Liu

114

et al., 2018; Wang et al., 2013; Zhao et al., 2010) and Northwest China (Shi and

115

Zhan, 2015). On the other hand, there are also some studies focusing on multiple

116

regions. However, they pay more attention to how water flows across regions (Feng

117

et al., 2014; Lenzen et al., 2013; Zhang et al., 2016; Zhao et al., 2015), while just a 6

118

few studies identify the spatial distribution and structure of water footprints inside

119

river basins (Feng et al., 2012; Okadera et al., 2014) or across thirty provinces within

120

China (Dong et al., 2014; Zhang and Anadon, 2014). Beyond that, studies further

121

explain the features of water footprints by identifying the driving factors. From the

122

viewpoint of economic growth, there was a positive linear relationship between per

123

capita water footprint and per capita GDP across countries (R. Wang et al., 2016).

124

From the perspective of comparative advantages, China’s agricultural virtual water

125

flow relied to a large extent on land scarcity across provinces rather than water

126

scarcity (Zhao et al., 2019), which explained why the water footprint was generally

127

higher in relatively developed areas. Using structural decomposition analysis (J. Liu

128

et al., 2018) or index decomposition analysis (Xu et al., 2015), socioeconomic

129

driving factors have also been identified in the temporal variations (Distefano et al.,

130

2018; Qian et al., 2018; X. Wang et al., 2016) and spatial differences (Sun, 2019).

131

In spite of a surging numbers of studies on water footprints, there are still

132

several shortcomings in existing research: 1) Numerous studies have explored the

133

features of water footprints of single regions on various spatial scales, and studies on

134

multiple regions pay more attention to virtual water flows, whereas identifying

135

spatial differences of features of water footprints is inadequate; 2) existing studies on

136

multi-regional water footprints are also lacking in elaboration, and the trend and

137

spatial distribution of water footprints of fine categories can be further unveiled, and;

138

3) studies involving multiple regions are mostly quiescent identifications of a certain

139

year. However, socioeconomic transition and policy implementation have changed

140

water production and consumption patterns in China since 2007, which may lead to a

141

transition in the trend, spatial distribution, and structure of China’s water footprint.

142

We investigated China’s provincial water footprints in 2007 and 2012 from the

143

viewpoints of trends, spatial distribution and sectoral structure. Meanwhile, we

144

explored the changes in water footprint as well as the factors leading to the changes,

145

which are necessary for the understanding of China’s strategic water demands and

146

for further water conservation from regional perspective. Section 2 introduces the 7

147

methodologies applied in this study and presents the data sources. Section 3 shows

148

the main results, demonstrating the trends, structures, and driving factors of

149

provincial water footprints. In section 4, we discuss the implications of the results,

150

and section 5 concludes the paper.

151

2. Methodology and data

152

2.1 Methodology

153

An integrated analysis framework combining multi-regional input–output

154

analysis, structural decomposition analysis, and elasticity was developed to identify

155

the new water footprint patterns (Fig. 1).

156 157 158

Fig. 1. Analysis framework of China’s water footprint

2.1.1 Water footprint accounting

159

A consumption-based method was applied to quantify water footprints by

160

means of the multi-regional input–output (MRIO) model (Wiedmann, 2009;

161

Wiedmann et al., 2007). As an extension of the basic input–output model (Miller and

162

Blair, 2009), the raw balance in MRIO can be expressed as:

163

∑

∑

8

,

(1)

164

where

165

reflects local intermediate inputs among sectors in region r;

represents direct

166

input coefficient from region r to region m; and

express the final

167

demand from region r to itself and from region r to region m, respectively.

168

Transforming this equation into matrix form, we obtain:

and

169

denote the total output of regions r and m, respectively;

and

,

(2)

170

where

171

coefficient matrix, and the vector of final demand, covering multiple regions. In

172

these vectors (matrix), each element represents a sectoral vector (matrix) of one

173

province or one interprovincial path from one province to another province.

174 175

,

, and

, respectively, denote the vector of total output, the direct input

With further extension by environmental factors, the following equation is obtained: ( − )

176

,

(3)

177

where

178

the direct water intensity, denoting the consumption volume of unit total output; is

179

the identity matrix, and;

180

reflecting the production structure.

represents each province’s sectoral water consumption;

⋮

⋯ ⋱ ⋯

⋮

is regarded as

denotes the Leontief inverse matrix,

181

The provincial water footprint is the volume of water consumed to meet all the

182

final demands of a province, including direct household consumption and virtual

183

water embodied in commodities that are both locally produced and imported. Note

184

that virtual water in merchandise exported to meet other provinces’ final demand is

185

not included in the local water footprint. Rather, this portion is regarded as an extra

186

burden on the local water resource. A province r’s water footprint

187

expressed as:

188

∑

∑

9

, ,

,

is

(4)

189

where m and k represent provinces,

190

consumed by residents in province r, whereas the sum of the other two terms is the

191

volume of water embodied in products and services, which is quantified by the

192

MRIO model.

is the volume of water directly

193

Note that in equation (4), the sum of the first two terms represents the

194

consumption volumes from local water resources, which is called the internal water

195

footprint of region r. The last term represents the volumes consumed by outside

196

production activities; thus, it is regarded as the external water footprint of region r.

197

In this study, China’s total water footprint is the domestic water footprint,

198

because the external footprint doesn’t cover international imports. Nevertheless, it

199

needs to be noticed that China’s virtual water import from other nations cannot be

200

ignored, especially as China is currently playing a more important role in

201

international trade (Tian et al., 2018).

202

2.1.2 Structural decomposition analysis

203

We further allocated water footprint total variation into the contributions of

204

several driving factors, which allowed decryption of the effects on water

205

consumption of technological change, industrial transition, and consumption pattern

206

shifts during the study period (Yamakawa and Peters, 2011). Note that the water

207

directly consumed by residents,

208

analysis (SDA) as

209

premise of SDA (Wachsmann et al., 2009). In this context, per capita water

210

footprints

" "

211

, is excluded in the structural decomposition

is not calculated by the input–output model, which is the

is decomposed as follow: # $,

(5)

212

where

213

final demand category of each province;

214

the structure of each province’s final demand categories; and

215

demand level, that is, the provincial per capita final demand. By means of SDA, five

is the composition of final products, expressing sectoral structure in each #

10

is the final demand structure, denoting $

represents the final

216 217

factors’ contributions to the variation in per capita water footprints ∆ quantified. ∆

"

∆

# $

∆

∆

# $

# $

∆

# $

"

are

#∆ $

218

(6)

219

The first term is the per capita water footprint variation induced by water intensity

220

change, the second represents the contribution of production structure change, the

221

third term denotes the variation caused by change in the composition of final

222

products, and the last two terms reflects the effect of change in final demand

223

structure and final demand level, respectively. For the decomposition with five

224

factors, there exist 5! equally valid results on account of different computing orders

225

(Dietzenbacher and Los, 1998). We choose their mean values as the final results.

226

2.1.3 Elasticity

227

We used elasticity to measure the influence of the economic level on the degree

228

of water demand. Based on the data of provinces, a unary linear regression was

229

applied to calculate the elasticity (ε) of the water footprint (Stern, 2010):

230

231 232

'

($ ( )*+′) ($ ( ,-.′)

%∆)*+′ %∆,-.′

,

(7)

where CWF′ and GDP′ are the per capita water footprint and per capita GDP, respectively.

233

The decoupling index (DI) is also a kind of elasticity measure, which was

234

defined as the ratio of the percentage variation of CWF to that of GDP during the

235

study period. We calculated each province’s DI to determine the degree of change in

236

the water footprint with consideration of economic growth. Given that the changes in

237 238 239 240

GDPs of provinces from 2007 to 2012 were all positive, i.e., ∆GDP > 0, DI < 0 means there is a strong decoupling between the water footprint and the economy,

0 < DI < 0.8 represents a weak decoupling, 0.8 ≤ DI ≤ 1.2 reflects a coupled relationship, and 1.2 < DI denotes expansive negative decoupling (Tapio, 2005). 11

241

2.2 Data sources

242

Two types of data were obtained to calculate the water footprints of thirty

243

provinces in mainland China (excluding Tibet): multi-regional input–output tables

244

and water consumption data.

245

We used China’s multi-regional input–output tables for 2007 and 2012

246

compiled by the Institute of Geographic Sciences and Natural Resources Research

247

(Chinese Academy of Sciences). The MRIO table for 2007 was further adjusted to a

248

constant 2012 price, eliminating the impact of inflation by the double deflation

249

method (Liu and Peng, 2010). However, this method ignores the difference among

250

various products produced by one sector because we can only obtain each sector’s

251

price index rather than each product’s (Mi et al., 2017a, 2017b). Price indexes from

252

2007 to 2012 were collected from provincial statistical yearbooks (NBS, 2013,

253

2008).

254

For water withdrawal and water use data, water uses in agriculture and industry

255

for each province in 2007 and 2012 were obtained from China Environmental

256

Statistics Yearbooks (NBS and MEP, 2013, 2008). Provincial water withdrawals in

257

industry with detailed sectoral data for 2007 were acquired from China's first

258

national pollution source census (MEP, 2011; Zhang and Anadon, 2014). Water uses

259

in construction and service industries were estimated by allocating water uses in

260

production activity based on the proportions of intermediate inputs from the “water

261

production and supply” sector to different economic sectors in the input–output

262

tables. Water uses in production activities for 2007 and 2012 were obtained from

263

China Urban–Rural Construction Statistical Yearbooks (MOHURD, 2013, 2008),

264

from which direct water uses in both rural and urban households were also obtained.

265

Water consumption data were estimated based on water withdrawal or water use

266

data. Water consumption in agriculture for each province was calculated as the water

267

consumption multiplied by the water consumption rate, which was obtained from

268

provincial Water Resource Bulletins (PWRB, 2012, 2007). In a similar manner, total

269

water consumptions for industries for each province were also estimated. Sectoral 12

270

water consumption data for industry were calculated as the water withdrawal minus

271

wastewater discharge, which was also obtained from the first national pollution

272

census (MEP, 2011; Zhang and Anadon, 2014). This sectoral water consumption

273

structure for industry was further applied to proportionally distribute the gross water

274

consumption in industry for 2007 and 2012 into sectoral water consumptions in

275

industry. For construction, service, and household, water consumptions were

276

estimated by multiplying the water uses by their consumption rates (Zhang and

277

Anadon, 2014). This study only covered blue water to determine the total volume of

278

surface or groundwater within this country needed to satisfy domestic demand.

279

3. Results

280

3.1 Overall trend in China’s water footprint and transition in spatial

281

distribution

282

China’s water footprint increased from 178.5 m3/capita (232.3 billion m3) in

283

2007 to 193.1 m3/capita (259.6 billion m3) on average in 2012, showing an annual

284

growth rate of 1.6%. On the one hand, provinces tended to demand more water as

285

their economic levels increased. Elasticity was determined as 0.4 in 2012, indicating

286

that the water footprint would increase by 40% if the economic level doubled (Fig.

287

2a). On the other hand, elasticity decreased from 2007 to 2012, indicating that the

288

growth rate of the per capita water footprint became slower than that of economic

289

growth. Meantime, there appeared to be decoupling effects between water footprints

290

and economic growth in almost all the provinces, which explains why the growth in

291

China’s water footprint was relatively slight (Fig. 2b).

292

With regard to spatial distribution, northern China presented a higher water

293

footprint than the south, and this gap expanded from 2007 to 2012. Water footprints

294

of fifteen northern provinces (regional classification details are provided in the

295

Supporting Information [SI]) increased by 20.4 m3/capita (10.8%) on average, which

296

was double the increment and double the growth rate of provinces in the south. As a

297

result, these fifteen northern provinces had an average water footprint of 209.5

298

m3/capita, 28.5 m3/capita higher than provinces in the south in 2012. 13

299

Provinces with the highest per capita water footprints in 2007 were

300

concentrated in the northwest and in the three most developed provinces (Beijing,

301

Shanghai, and Tianjin). However, in 2012, northern China was further highlighted,

302

as not only northwestern provinces (Xinjiang, Ningxia, and Qinghai) but also Inner

303

Mongolia and Heilongjiang had high per capita water footprints. This transition in

304

spatial distribution was the result of opposing changes in provincial water footprints.

305

On the one hand, a few provinces dramatically decreased their water footprint per

306

unit of GDP. For example, Tianjin had the most prominent economic growth among

307

the thirty provinces from 2007 to 2012, but it reduced its water footprint by 260.7

308

m3/capita (−54.9%), with its water footprint per unit of GDP dropping markedly, by

309

72.9%. Hainan, Shanghai, and Beijing also reduced their per capita water footprints

310

by 49.2%, 47.8%, and 38.6%, respectively. On the other hand, several provinces had

311

remarkably increased water footprints with economic growth. For instance, the water

312

footprint of Inner Mongolia grew dramatically to 352.5 m3/capita (+139.5%) from

313

2007 to 2012. Although Inner Mongolia also experienced significant economic

314

growth, its water footprint per unit of GDP decreased by only 21.7%. The water

315

footprints of Heilongjiang, Xinjiang, and Jiangsu also increased during this period,

316

by 57.8%, 38.6%, and 38.3%, respectively.

14

317 318 319

Fig. 2. Spatial distribution and overall trend in China’s water footprint from 2007–2012. (a) Elasticity across thirty provinces; (b) Decoupling index of each province during the study period.

320 321

3.2 Main sources of China’s water footprint

322

With regard to final demand categories, household consumption played the sole

323

prominent role in the water footprint, accounting for an average of 65.0% in 2012.

324

Fixed capita formation was the second main component, contributing an average of

325

21.8% in 2012. Inventory change and government consumption contributed less to

326

the water footprint. During the study period, the water footprint increased for all the

327

final demand categories except inventory change.

328

3.2.1 Structural transition of household consumption induced water footprint

329

As the primary source of the water footprint, average household consumption

330

was 125.4 m3/capita (168.7 billion m3) in 2012, having increased by only 0.2%

331

annually from 2007 to 2012. Most of the provinces followed the trend that provinces

15

332

with higher income levels usually showed larger water footprints related to

333

household consumption, except certain provinces in the northwest (Fig. 3).

334 335 336

Fig. 3. Trends in household consumption related per capita water footprints.

337

To appropriately reflect various facets of residential lifestyles, household

338

consumption was further decomposed into seven categories. From 2007 to 2012, the

339

structure of the water footprint related to household consumption generally remained

340

stable, in which food and beverages was the primary contributor while transportation

341

made the least contribution, and the proportions were relatively even across other

342

categories (Fig. 4a). Specifically, food and beverages generated a 91.7 m3/capita

343

(73.1%) average water footprint in 2012. Owing to this category’s leading role in

344

household consumption, food and beverages resulted in the elasticity of the

345

household consumption induced water footprint and determined its spatial pattern,

346

presenting enormous water footprints in certain northwestern provinces. Direct

347

household consumption contributed 9.5 m3/capita (7.6%) to the water footprint

348

related to household consumption in 2012. Direct household consumption showed a 16

349

similar elasticity to food and beverages. However, in contrast to the overall spatial

350

pattern of water footprint related to household consumption, southern China showed

351

a water footprint 4.9 m3/capita higher than the average of the north in 2012 for direct

352

household consumption. Provinces with the highest water footprints related to direct

353

household consumption were relatively developed and/or were located in water-rich

354

regions, such as Shanghai, Guangdong, Zhejiang, Jiangsu, and Hubei. This spatial

355

distribution might be explained by the differences in household water use habits of

356

southern and northern China, which are affected by various factors such as climate

357

and income level (Zhao, 2015; Zhong et al., 2018). Accommodation and food service

358

activities and shelter and household equipment had water footprints of 7.2 m3/capita

359

(5.7%) and 5.4 m3/capita (4.3%), respectively, in 2012. The north and south

360

presented close water footprints, with a difference of no more than 0.8 m3/capita.

361

Inner Mongolia, Heilongjiang and eastern coastal provinces showed the highest

362

water footprints related to these two categories. Apart from this, relatively developed

363

provinces were also observed to have high water footprints related to

364

accommodation and food service activities, while shelter and household equipment

365

also led to high water footprints in several central provinces such as Hubei (7.7

366

m3/capita), Jiangxi (7.4 m3/capita) and Chongqing (6.8 m3/capita). Note that the

367

water footprint of accommodation and food service activities grew at 80% of the rate

368

of economic growth in 2012, obviously higher than other categories, which may

369

explain this difference.

17

370 371 372 373 374 375 376

Fig. 4. Structure of household consumption related per capita water footprint. (a) Structures of household consumption related water footprints in 2007 (left bar) and 2012 (right bar); (b) Structures of rural (left bar) and urban (right bar) household consumption related water footprints in 2012. Top six provinces with highest/lowest per capita water footprints related to household consumption in 2012 are presented in red/green.

377

Urban and rural households showed distinct patterns of water footprints (Fig.

378

4b). Firstly, in terms of quantity, urban households generated over twice the water

379

footprint of rural households, generating 169.7 m3/capita (121.4 billion m3) and 75.1

380

m3/capita (47.2 billion m3), respectively, in 2012. From a per capita perspective, this

381

gap narrowed from 2007 to 2012, with an increasing trend in the water footprint

382

related to rural household consumption (+7.0 m3/capita) and a decreasing trend in

383

that related to urban household consumption (−19.8 m3/capita). However, the

384

magnitude of the urban–rural gap was enlarged in terms of the total amount owing to

385

a transition in population structure, with a growing urban population and a declining

386

rural population in the context of urbanization. Secondly, from the viewpoint of

387

structure, the water footprint related to rural households included larger proportions

388

of food and beverages and direct household consumption than did urban households. 18

389

From 2007 to 2012, rural households narrowed the urban–rural gaps in the

390

proportions of shelter and household equipment, accommodation and food service

391

activities, and other services but expanding the urban–rural gap in the proportion of

392

the direct household consumption related water footprint. Thirdly, from the regional

393

perspective, the gap in the water footprint between urban and rural households was

394

wider in the north than in the south. On average, the water footprint of urban

395

households was 2.5-fold that of rural households in the north but 2.1-fold higher in

396

the south. Note that the elasticity of rural households with lower income levels was

397

larger than that of urban households with higher income levels, which to some extent

398

explained the larger gaps in developing provinces, such as Qinghai, Shaanxi, Hebei,

399

and Xinjiang in the north.

400

3.2.2 Structural transition of fixed capital formation induced water footprint

401

The water footprint related to fixed capital formation increased significantly

402

during the study period, from 27.2 m3/capita to 42.1 m3/capita (i.e., an annual growth

403

rate of 9.1%), obviously higher than that for other categories. The growth rate was

404

higher in the north (11.2%), which further widened the gap in the water footprint

405

between north and south (Fig. 5). In 2012, northern China had a water footprint of

406

45.8 m3/capita related to fixed capital formation, which was 16.1% higher than in the

407

south. Compared with household consumption, the elasticity of fixed capital

408

formation induced water footprint was lower, which obviously declined from 2007 to

409

2012 (Fig. 6). This was reflected by the fact that the water footprint related to fixed

410

capital formation decreased in several relatively developed provinces, such as

411

Shanghai (−32.6 m3/capita), Beijing (−11.1 m3/capita), and Tianjin (−7.2 m3/capita).

19

412 413 414 415 416 417

Fig. 5. Distribution and structure of per capita water footprint related to fixed capital formation from 2007–2012. The two bars for each province represent different years, 2007 (left) and 2012 (right). The six provinces with the highest/lowest per capita water footprints related to fixed capital formation in 2012 are presented in red/green.

418

With regard to sectors, construction was the major source of the water footprint

419

induced by fixed capital formation, contributing 49.9% (10.1 m3/capita) in 2012, and

420

showing an annual average growth rate of 7.9%. Several northern provinces

421

consumed the most water through fixed capital formation in construction, such as

422

Ningxia (49.5 m3/capita), Inner Mongolia (44.8 m3/capita), Xinjiang (44.2 m3/capita),

423

and Qinghai (36.7 m3/capita) in 2012. Construction also showed less elasticity (0.3

424

in 2012), which was reflected by the fact that provinces positioned in middle

425

economic levels tended to demand more water through fixed capital formation in

426

construction. As the second-largest contributor, agriculture had a water footprint of

427

21.0 m3/capita (24.1%) in 2012, increasing by 11.7% annually. However, a

428

downward tendency was observed in its water footprint related to fixed capital

429

formation with an increase in the economic level, whereas the opposite of that was

430

seen in other sectors. Notwithstanding, several northern provinces had remarkably

431

increased water footprints related to fixed capital formation in agriculture from 2007 20

432

to 2012. Beyond that, strategic emerging manufacturing also played an increasingly

433

important role, with the water footprint increasing from 4.4 m3/capita to 7.2

434

m3/capita during the study period. The elasticity of the strategic emerging

435

manufacturing related water footprint declined considerably during the study period.

436

During this period developing provinces obviously increased their water footprints

437

related to strategic emerging manufacturing especially in Inner Mongolia (+11.1

438

m3/capita), while it decreased in several relatively developed provinces such as

439

Shanghai (−11.7 m3/capita) and Beijing (−2.1 m3/capita). In addition, the difference

440

between northern and southern water footprints related to fixed capital formation

441

was caused mainly by agriculture and strategic emerging manufacturing, rendering a

442

difference (north minus south) of 3.8 m3/capita and 1.6 m3/capita, respectively, in

443

2012.

444 445 446

Fig. 6. Trends in per capita water footprints related to fixed capital formation in 2012.

447

3.3 Effects of transition in production and consumption pattern on water

448

footprint

449

Socioeconomic driving factors were classified into consumption pattern and the

450

production pattern. Of the 30 provinces in mainland China, 27 provinces increased 21

451

their per capita water footprints through changes in their consumption patterns and

452

reduced them through changes in their production patterns from 2007 to 2012 (Fig.

453

7).

454 455 456 457

Fig. 7. Effects of socioeconomic factors on variations in per capita water footprints.

3.3.1 Effects of transition in production pattern on provincial water footprints

458

The production pattern covers the water intensity and the production structure.

459

Water intensity generally determined the overall effects of the production pattern,

460

while change in the production structure had a weaker impact on most provinces.

22

461 462 463 464 465

Fig. 8. Sectoral and regional contribution to the variations in per capita water footprints from

466

Water intensity was the primary driving factor significantly reducing the water

467

footprints of almost all of the regions and sectors (Fig. 8). With regard to sectors,

468

agriculture was the major contributor, inducing more than 30% of the declines in

469

water footprints related to water intensity improvement in the 26 provinces. Notably,

470

agriculture contributed over 60% of the water savings through water intensity in

471

one-fifth of the provinces, such as Xinjiang (118.1%), Qinghai (74.6%), and Ningxia

472

(69.7%) in the northwest. At the same time, food processing and tobacco, and

473

construction, also accounted for over 10% of water conservation induced by water

474

intensity in 27 and 20 provinces, respectively. With regard to regional contributions,

475

the local contribution accounted for over half of the water savings in 21 provinces, of

476

which northwestern provinces had the highest local contributions. In contrast,

477

relatively developed provinces benefited more from external regions, such as Beijing

478

(82.7%), Tianjin (77.5%), and Shanghai (71.3%).

production patterns. Xinjiang is excluded from this figure. Details of regional contribution are provided in SI.

23

479

Change in production structure had a relatively weak influence for the

480

neutralization of different sectors and different regions (Fig. 8). From the perspective

481

of sectoral contribution, food processing and tobacco reduced its water footprint in

482

28 provinces, with the greatest reductions seen in several relatively developed

483

provinces such as Tianjin (−107.2 m3/capita), Beijing (−27.0 m3/capita), and

484

Shanghai (−21.2 m3/capita). Other light manufacturing also saved water in 80% of

485

the studied provinces. Nevertheless, construction and other services led to growths in

486

the water footprints in at least 25 provinces. From the perspective of regional

487

contributions, on the one hand, over 70% of provinces saved water through external

488

regions; 12 provinces, however, counteracted this positive external effect by their

489

local contributions. On the other hand, for external contributions, neutralization of

490

the southern and northern contributions was observed in 16 provinces.

491

3.3.2 Effects of transition in consumption pattern on provincial water footprints

492

The consumption pattern is reflected by the final demand structure, the

493

composition of final products, and the final demand level. Rising final demand levels

494

and changes in the composition of final products increased the water footprints in

495

most provinces, while the final demand structure reduced the water footprints in 26

496

provinces (Fig. 9).

497

Change in the final demand level played a significant role in the increases in the

498

water footprints of northern provinces such as Ningxia (+259.6 m3/capita), Qinghai

499

(+221.0 m3/capita), Inner Mongolia (+185.8 m3/capita), and Tianjin (+140.1

500

m3/capita), offsetting the positive effect of water intensity. Additionally, the sectoral

501

and regional contributions of the change in the final demand level presented similar

502

structures to that of changes in water intensity, probably implying the rebound effect

503

of a growing demand level on efficiency improvement (J. Liu et al., 2018).

504

For the two types of structures in the consumption pattern, from the perspective

505

of sectoral contributions, their opposing effects were induced by agriculture, food

506

processing

507

manufacturing. However, strategic emerging manufacturing increased the water

and

tobacco,

other

light

24

manufacturing,

and

resource-related

508

footprint through changes in both the final demand structure and the composition of

509

final products in over 20 provinces, in which several northern and central provinces

510

saw increases, such as Xinjiang (+27.8 m3/capita), Inner Mongolia (+9.0 m3/capita),

511

Hubei (+8.4 m3/capita), and Heilongjiang (+7.3 m3/capita). From the regional

512

contribution perspective, external regions triggered rises in water footprints through

513

changes in the composition of final products and declines in water footprints through

514

changes in the final demand structure in over 80% of the provinces. Especially,

515

northern regions played a more prominent role than the south, whether in the growth

516

in the water footprint through changes in composition of final products or in the

517

reduction in the water footprint by change in the final demand structure.

518 519 520 521 522

Fig. 9. Sectoral and regional contributions to the variations in per capita water footprints in terms of the consumption pattern. Xinjiang is excluded from this figure. Details of regional contribution are provided in SI.

25

523

4. Discussion

524

4.1 New patterns in China’s water footprint

525

China has experienced a socioeconomic transition, and in recent years its

526

government has been promoting a series of policies regarding water saving. In this

527

context, there have appeared transitions in the trend, spatial distribution, and internal

528

structure of China’s water footprint from 2007 to 2012. In general, most provinces

529

enjoyed economic booms along with lower rates of growths or even declines in their

530

water footprints. With regard to the spatial pattern, northern China presented a higher

531

water footprint than southern China and this gap widened from 2007 to 2012. The

532

north was also observed to have a wider urban–rural gap in the water footprint

533

related to household consumption than in the south. With regard to the transitions in

534

structural pattern, fixed capital formation contributed more in China’s water footprint

535

while the household consumption induced water footprint remained stable in total

536

volume. Nevertheless, the per capita water footprint related to rural household

537

consumption exhibited clear growth and a diversified structure.

538

4.2 Mechanism of the decoupling of water footprint from rapid economic

539

growth

540

In almost all the provinces, the increments in water footprints from 2007 to

541

2012 slowed, decoupling from economic growths. Under the marked impetus of

542

water conservation policies, socioeconomic driving factors played important roles in

543

reducing the growth in the water footprint.

544

Firstly, water use efficiency has been significantly improved, curbing the rapid

545

growth of water demanded by expanding production. As the results show, almost all

546

of the sectors in each province lessened their water intensity, especially agriculture

547

which was the most water-intensive sector. On the one hand, a series of policies

548

focusing on upgrading technology and equipment indeed promote improvement in

549

the water use efficiency of each sector. The government has been devoted to a

550

program for water-saving transformation in irrigated areas in the past few decades.

551

In this context, water conservation by agriculture is strongly promoted in China, with 26

552

the water-saving capacity of irrigation improving by 21 billion m3 and the average

553

actual water per unit area of irrigation decreasing by 19.0% from 1998 to 2015

554

(NDRC, 2017). Meanwhile, the government has also required existing enterprises to

555

eliminate obsolete technology and facilities and to implement technical

556

transformations (NDRC, 2007). On the other hand, through the flow of goods and

557

services across regions, improvement in water use efficiency in key regions, such as

558

major grain-producing provinces, have rendered further decreases in the water

559

footprints of import-dominated provinces.

560

Secondly, the structures have been optimized to further reduce the water

561

footprint. With regard to the production pattern, changes in the production structure

562

of several industries (such as food processing and tobacco and other light

563

manufacturing) play important roles in water conservation. The government pays

564

more attention to water saving by all sectors, especially highly water-intensive

565

industries rather than only agriculture. On the one hand, the government accelerates

566

industrial restructuring and limits industries with high water consumption and low

567

efficiency (NDRC, 2007). On the other hand, highly water-intensive industries have

568

been encouraged to carry out a water-saving transformation of their production

569

processes (NDRC, 2007). With regard to the consumption pattern, the change in the

570

final demand structure also has positive effects on reducing the water footprint.

571

According to the results, consumption presented a higher water footprint than did

572

investment, because household consumption had a great demand for water-intensive

573

products such as food. However, among the three drivers of China’s economic

574

development, investment has played a significant role in China’s growing economy

575

since the reform and open-up policy, and China further transitioned into an

576

investment- and export-dominated development pattern after entering the World

577

Trade Organization in 2001 (Liu and Yan, 2015). In comparison with investment,

578

consumption with higher demands for water always played a relatively minor role.

27

579

4.3 Diverse water management strategies based on new spatial patterns of water

580

footprints

581

The spatial distribution of China’s water footprint highlights the leading

582

positions of several northern provinces. They experienced diverse trends in their

583

water footprints and presented prominent water footprints in 2012. It is necessary for

584

different water management strategies to be applied in these provinces.

585

First, several provinces induced enormous water footprints, ranking within the

586

top six in both 2007 and 2012, such as Xinjiang, Ningxia, and Qinghai in the

587

northwest. It seems that their enormous water footprints are not just a short-term

588

effect induced by rapid socioeconomic development, although the development did

589

result in higher water footprints in general. The positions of these provinces in the

590

country and their natural conditions lead to their high water footprints. Specifically,

591

compared with relatively developed provinces, the provinces rely more heavily on

592

local supply of water-intensive products such as agricultural products. However,

593

drought conditions and irrigation-dominated production mode cause much water to

594

be embodied in their agricultural products. Therefore, agriculture is the key to

595

relieving the water footprint of the northwestern provinces and ensuring regional

596

security. On the one hand, water-use efficiency and the production structure of

597

agriculture should be further improved to achieve food self-sufficiency with lower

598

water consumption, ensuring local food security. On the other hand, these provinces

599

are also in heavy water stress, showing a great conflict of local severe water scarcity

600

with the enormous demand for water (details are provided in SI). It is essential for

601

these provinces to plan their development based on their natural water endowment.

602

Especially that other provinces also imposed obvious extra burdens on them (Cai et

603

al., 2019; Qian et al., 2018) (details are provided in SI). Thus, adjusting trade

604

patterns, limiting the development water-intensive industries and transferring more

605

food supply source to regions with richer water resources and less water-intensive

606

production processes can lessen water scarcity and optimize water-use patterns in

607

China. 28

608

Second, the water footprints of several northern provinces were at average (or

609

even lower) levels in 2007 but grew dramatically during the study period, such as in

610

Inner Mongolia and Heilongjiang. For Inner Mongolia, a rise in the water footprint

611

was observed in both household consumption and fixed capital formation, which

612

may have been mainly caused by its socioeconomic development. In the process of

613

urbanization, residents generate higher demands for all products and services with

614

growth in the income level, especially for sufficient food. Massive investments in

615

buildings and infrastructure also incur high water demands. Thus, it is a focal point

616

for Inner Mongolia to lessen the impact of socioeconomic development on water

617

demand. Means of controlling the soaring demand by all sectors and optimizing the

618

consumption structure of water-intensive products should receive increased attention.

619

At the same time, Inner Mongolia needs to balance the surging demand for water

620

with its serious water-deficiency (details are provided in SI) by adjusting industrial

621

developing orientation and trade mode. The growth in Heilongjiang’s water footprint

622

was caused mainly by the heavy investment in agriculture. As Heilongjiang is the

623

major grain-producing province, this may be the result of national plans to increase

624

the capacity of grain production and ensure food security (NDRC, 2009a, 2009b).

625

Hence, ways of saving water in agriculture while ensuring agricultural production

626

capacity will be a future focus for Heilongjiang. Additionally, although Heilongjiang

627

is experiencing lighter water stress and also consumes much less scarce water than

628

the above four provinces, it still needs to further adjust its trade partner to reduce

629

consuming external scarce water (details are provided in SI).

630

4.4 Emerging demands for water under the transition in structural pattern

631

The government has proposed water-saving policies consecutively and has

632

made steady progress on water conservation in key fields. However, there still exists

633

the emergent or potential growth in water footprint in the context of socioeconomic

634

development and transition.

635

First, rural households are generating higher water demands. Over the period

636

from 2007 to 2012, the per capita water footprint related to rural household 29

637

consumption displayed obvious growth, in contrast to that for urban households. In

638

almost all of the categories of residential consumption, rural household consumption

639

also presented higher growth rates of water footprints than urban household

640

consumption when the economic levels increased. Rural households might emerge to

641

be a considerable driver of the water footprint with the continuous increases in

642

income levels but inadequate water-saving measures, such as relatively poor

643

infrastructure and obsolete appliances, weaker awareness of the importance of saving

644

water, and less improved water resource management (Li et al., 2019).

645

Second, the increasing diversity of the residential consumption pattern has a

646

potential impact on water consumption during the process of urbanization. With food,

647

for example, urban residents have more options for dining in restaurants rather than

648

at home. In this case, residents may appear to consume less water in direct household

649

consumption, whereas actually this consumption is transferred to other categories

650

(such as food service activities). Hence, water saving by households does not mean

651

just saving water in direct household consumption. Additionally, it should be further

652

explored whether this transfer induced by structural diversification may be a

653

potential source of growth in the water footprint.

654

Third, the growing demands for water also occur in both emerging and basic

655

industries. On the one hand, emerging strategic manufacturing significantly

656

increased the water footprint during the study period. In the process of

657

industrialization, China has upgraded the internal structure of its investment in

658

secondary industry, which displays an obvious tendency to involve a high degree of

659

processing and to be technology intensive (Huang, 2014). On the other hand, heavy

660

investment in agriculture has also induced an increase in the water footprint.

661

Although industrialization results in a descending contribution by primary industry

662

in the economic structure, agriculture as the fundamental and the most

663

water-intensive industry contributes a large proportion of the water footprint.

664

Beyond that, the scale of agricultural production may expand to ensure national food

665

security (NDRC, 2009b). 30

666

5. Conclusion

667

This study provided a comprehensive analysis framework for exploring the

668

changes in traits and driving factors of multi-regional water footprints in China.

669

Empirical research was carried out to investigate the transitions in China’s water

670

footprint pattern from 2007 to 2012, which was conducive to future water

671

conservation by extrapolating on the current situation of China’s water footprint and

672

identifying its major contributors and potential or emerging sources. Three

673

transitions in its traits were investigated. China’s water footprint decoupled from

674

economic growth in terms of the overall trend, and the elasticities of specific

675

categories declined. With regard to the spatial pattern, the north (especially Inner

676

Mongolia, Heilongjiang and several northwestern provinces) presented high water

677

footprints in 2012, widening the difference in the per capita water footprint with

678

southern China. With regard to structure, the contribution of primary contributors

679

was apt to flatten or even decrease, while secondary contributors usually played

680

increasingly important roles in the structure of the water footprint. These minor

681

contributors, such as rural household consumption and investments in agriculture

682

and strategic emerging manufacturing, may potentially generate higher water

683

footprints, and should be paid more attention in the future.

684

Policies and socioeconomic transition process played important roles in the

685

transitions in China’s water footprint. Water use efficiency remarkably improved in

686

almost all the provinces and sectors under the implementation of policies on

687

upgrading technology and optimizing industrial structure. Notwithstanding, the

688

spatial distribution of China’s water footprint was still not conducive to lessen the

689

water stress of provinces with heavy water scarcity. The unreasonable distribution

690

might be attributed to long-standing location discrepancies, local demands for

691

socioeconomic development, and national food security demands. The key provinces

692

experiencing severe water stress should appropriately expand imports and adjust

693

trade partners, while those in lighter water stress should be on alert to emerging

694

demands. 31

695 696

Acknowledgments

697

This work was jointly supported by the National Natural Science Foundation of 682

698

China (No. 71704012), the foundation for Innovative Research Groups of the 683

699

National Natural Science Foundation of China (No. 51721093), and the National 684

700

Key Research and Development Program (No. 2016YFC0503005).

701 702

References

703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732

Allan, J.A., 1998. Virtual water: a strategic resource global solutions to regional deficits. Ground Water 545–546. Allan, J.A., 1993. Fortunately There are Substitutes for Water Otherwise Our Hydropolitical Futures Would be Impossible. Priorities for Water Resources Allocation and Management. ODA, London, the United Kingdom. Cai, B., Zhang, W., Hubacek, K., Feng, K., Li, Z., Liu, Yawen, Liu, Yu, 2019. Drivers of virtual water flows on regional water scarcity in China. Journal of Cleaner Production 207, 1112–1122. https://doi.org/10.1016/j.jclepro.2018.10.077 Chapagain, A., Hoekstra, A., 2003. Virtual water flows Between Nations in Relation to Trade in Livestock and Livestock Products. Value of Water Research Report Series No.13, UNESCOIHE Inst. for Water Educ., Delft, Netherlands. Chapagain, A.K., Hoekstra, A.Y., 2002. Virtual water flows: A Quantification of Virtual Water Flows Between Nations in Relation to International Crop Trade, Value of Water Research Report Series. UNESCO-IHE, P.O. BOX 3015, 2601 DA DELFT, THE NETHERLANDS. Cheng, H., Hu, Y., Zhao, J., 2009. Meeting China’s Water Shortage Crisis: Current Practices and Challenges. Environmental Science & Technology 43, 240–244. https://doi.org/10.1021/es801934a Dietzenbacher, E., Los, B., 1998. Structural Decomposition Techniques: Sense and Sensitivity. Economic Systems Research 10, 307–324. https://doi.org/10.1080/09535319800000023 Distefano, T., Riccaboni, M., Marin, G., 2018. Systemic risk in the global water input-output network. Water Resources and Economics 23, 28–52. https://doi.org/10.1016/j.wre.2018.01.004 Dong, H., Geng, Y., Fujita, T., Fujii, M., Hao, D., Yu, X., 2014. Uncovering regional disparity of China’s water footprint and inter-provincial virtual water flows. Science of The Total Environment 500–501, 120–130. https://doi.org/10.1016/j.scitotenv.2014.08.094 Dong, H., Geng, Y., Sarkis, J., Fujita, T., Okadera, T., Xue, B., 2013. Regional water 32

733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776

footprint evaluation in China: A case of Liaoning. Science of The Total Environment 442, 215–224. https://doi.org/10.1016/j.scitotenv.2012.10.049 DRCSC, 2012. State Council’s opinion on implementing the most stringent water resources management system. Development Research Center of The State Council. Feng, K., Chapagain, A., Suh, S., Pfister, S., Hubacek, K., 2011. COMPARISON OF BOTTOM-UP AND TOP-DOWN APPROACHES TO CALCULATING THE WATER FOOTPRINTS OF NATIONS. Economic Systems Research 23, 371–385. https://doi.org/10.1080/09535314.2011.638276 Feng, K., Hubacek, K., Pfister, S., Yu, Y., Sun, L., 2014. Virtual Scarce Water in China. Environmental Science & Technology 48, 7704–7713. https://doi.org/10.1021/es500502q Feng, K., Siu, Y.L., Guan, D., Hubacek, K., 2012. Assessing regional virtual water flows and water footprints in the Yellow River Basin, China: A consumption based approach. Applied Geography 32, 691–701. https://doi.org/10.1016/j.apgeog.2011.08.004 Han, M.Y., Chen, G.Q., Mustafa, M.T., Hayat, T., Shao, L., Li, J.S., Xia, X.H., Ji, X., 2015. Embodied water for urban economy: A three-scale input–output analysis for Beijing 2010. Ecological Modelling 318, 19–25. https://doi.org/10.1016/j.ecolmodel.2015.05.024 Hoekstra, A.Y., Chapagain, A.K., Aldaya, M.M., Mekonnen, M.M., 2009. Water footprint manual : state of the art 2009. Water Footprint Network, Enschede, The Netherlands. Huang, Q., 2018. China’s Industrial Development and Industrialization Process during the 40 Years of Reform and Opening-up. Institute of Industrial Economics 5–23. https://doi.org/10.19581/j.cnki.ciejournal.2018.09.011 Huang, Q., 2014. “The New Normal”, the Late Stage of Industrial Growth. China Industrial Economics 5–19. https://doi.org/10. 19581/ j. cnki . ciejournal. 2014. 10. 001 Jiang, Y., 2009. China’s water scarcity. Journal of Environmental Management 90, 3185–3196. https://doi.org/10.1016/j.jenvman.2009.04.016 Lenzen, M., Moran, D., Bhaduri, A., Kanemoto, K., Bekchanov, M., Geschke, A., Foran, B., 2013. International trade of scarce water. Ecological Economics 94, 78–85. https://doi.org/10.1016/j.ecolecon.2013.06.018 Li, H., Ding, Y., Li, Y., Guo, D., 2019. Analysis of status quo and problems of water conservation in China in new situation. South-to-North Water Transfers and Water Science and Technology 17, 202–208. https://doi.org/10.13476/j.cnki.nsbdqk.2019.0027 Liao, X., Chai, L., Xu, X., Lu, Q., Ji, J., 2019. Grey water footprint and interprovincial virtual grey water transfers for China’s final electricity demands. Journal of Cleaner Production 227, 111–118. Liu, J., Zhao, X., Yang, H., Liu, Q., Xiao, H., Cheng, G., 2018. Assessing China’s “developing a water-saving society” policy at a river basin level: A structural decomposition analysis approach. Journal of Cleaner Production 190, 799– 33

777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820

808. https://doi.org/10.1016/j.jclepro.2018.04.194 Liu, Q., Peng, Z., 2010. China’s Constant Price Input-output Table and Relative Analysis. China Statistics Press, Beijing. Liu, R., Yan, W., 2015. Performance Evaluation of Final Demands and Imbalance of Its Structure. Nanjing Journal of Social Sciences 22–30. https://doi.org/10.15937/j.cnki.issn1001-8263.2015.08.004 Liu, S., Han, M., Wu, Xudong, Wu, Xiaofang, Li, Z., Xia, X., Ji, X., 2018. Embodied water analysis for Hebei Province, China by input-output modelling. Frontiers of Earth Science 12, 72–85. https://doi.org/10.1007/s11707-016-0614-z Liu, S., Wu, Xudong, Han, M., Zhang, J., Chen, B., Wu, Xiaofang, Wei, W., Li, Z., 2017. A three-scale input-output analysis of water use in a regional economy: Hebei province in China. Journal of Cleaner Production 156, 962–974. https://doi.org/10.1016/j.jclepro.2017.04.083 MEP, 2011. Dataset of China Pollution Source Census. Ministry of Environment, China Environmental Science Press. Mi, Z., Meng, J., Guan, D., Shan, Y., Liu, Z., Wang, Y., Feng, K., Wei, Y., 2017a. Pattern changes in determinants of Chinese emissions. Environmental Research Letters 12, 074003. Mi, Z., Meng, J., Guan, D., Shan, Y., Song, M., Wei, Y., Liu, Z., Hubacek, K., 2017b. Chinese CO2 emission flows have reversed since the global financial crisis. Nature Communications 8. https://doi.org/10.1038/s41467-017-01820-w Miller, R.E., Blair, P.D., 2009. Input–Output Analysis: Foundations and Extensions, Second Edition. Cambridge University Press, The Edinburgh Building, Cambridge CB2 8RU, UK. MOHURD, 2013. China Urban–Rural Construction Statistical Yearbook. Ministry of Housing and Urban–Rural Development, China Planning Press. MOHURD, 2008. China Urban–Rural Construction Statistical Yearbook. Ministry of Housing and Urban–Rural Development, China Planning Press, Beijing. NBS, 2017. China Statistical Yearbook. National Bureau of Statistics, China Statistics Press. NBS, 2013. Provincial Statistical Yearbook. China Statistics Press. NBS, 2008. Provincial Statistical Yearbook. China Statistics Press. NBS, 2000. China Statistical Yearbook. National Bureau of Statistics, China Statistics Press. NBS, MEP, 2013. China Statistical Yearbook on Environment. National Bureau of Statistics, Ministry of Environmental Protection China Statistics Press. NBS, MEP, 2008. China Statistical Yearbook on Environment. National Bureau of Statistics, Ministry of Environmental Protection China Statistics Press. NDRC, 2017. Implementation plan for continuous water-saving renovation of large and medium-sized irrigation districts nationwide. National Development and Reform Commission. NDRC, 2012. The 12th Five-Year Plan for Water-saving Society Construction. National Development and Reform Commission. 34

821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864

NDRC, 2009a. Outline of the medium and long-term planning for national food security (2008-2020). National Development and Reform Commission. NDRC, 2009b. National plan to increase grain production capacity of 100 billion jin (2009-2020). National Development and Reform Commission. NDRC, 2007. The 11th Five-Year Plan for Water-saving Society Construction. National Development and Reform Commission. NDRC, 2005. China Water Conservation Technology Policy Outline. National Development and Reform Commission. Okadera, T., Geng, Y., Fujita, T., Dong, H., Liu, Z., Yoshida, N., Kanazawa, T., 2015. Evaluating the water footprint of the energy supply of Liaoning Province, China: A regional input–output analysis approach. Energy Policy 78, 148– 157. https://doi.org/10.1016/j.enpol.2014.12.029 Okadera, T., Okamoto, N., Watanabe, M., Chontanawat, J., 2014. REGIONAL WATER FOOTPRINTS OF THE YANGTZE RIVER: AN INTERREGIONAL INPUT–OUTPUT APPROACH. Economic Systems Research 26, 444–462. https://doi.org/10.1080/09535314.2014.934324 PWRB, 2012. Water Resource Bulletin. Provincial Water Resource Bureau, China Water Power Press, Beijing. PWRB, 2007. Water Resource Bulletin. Provincial Water Resource Bureau, China Water Power Press, Beijing. Qian, Y., Dong, H., Geng, Y., Zhong, S., Tian, X., Yu, Y., Chen, Y., Moss, D.A., 2018. Water footprint characteristic of less developed water-rich regions: Case of Yunnan, China. Water Research 141, 208–216. https://doi.org/10.1016/j.watres.2018.03.075 Ren, G., Song, Y., Jiang, J., 2019. Research on the Chinese Urbanization Process in the Past 40 Years from the Reform and Opening-up. Social Sciences in Ningxia 23–31. Shi, C., Zhan, J., 2015. An input–output table based analysis on the virtual water by sectors with the five northwest provinces in China. Physics and Chemistry of the Earth, Parts A/B/C 79–82, 47–53. https://doi.org/10.1016/j.pce.2015.03.004 Stern, D.I., 2010. Between estimates of the emissions-income elasticity. Ecological Economics 69, 2173–2182. Sun, S., 2019. Water footprints in Beijing, Tianjin and Hebei: A perspective from comparisons between urban and rural consumptions in different regions. Science of The Total Environment 647, 507–515. https://doi.org/10.1016/j.scitotenv.2018.07.343 Tapio, P., 2005. Towards a theory of decoupling: degrees of decoupling in the EU and the case of road traffic in Finland between 1970 and 2001. Transport Policy 12, 137–151. https://doi.org/10.1016/j.tranpol.2005.01.001 The State Council, 2008. Notice on Promoting Water Price Reform to facilitate water conservation and protection of water resources. Office of the State Council. The State Council, 2006. Water permit and water resource fee collection regulations. The State Council, 2000. Notice on Printing and Distributing “the Opinions on 35

865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908

Strengthening Industrial Water Saving Work” (State Council Bulletin). Office of the State Council. Tian, X., Joseph, S., Geng, Y., Qian, Y., Gao, C., Bleischwitz, R., Xu, Y., 2018. Evolution of China’s water footprint and virtual water trade- A global trade assessment. Environment International 121, 178–188. Wachsmann, U., Wood, R., Lenzen, M., Schaeffer, R., 2009. Structural decomposition of energy use in Brazil from 1970 to 1996. Applied Energy 86, 578–587. https://doi.org/10.1016/j.apenergy.2008.08.003 Wang, H., Yang, Y., 2018. Trends and Consumption Structures of China’s Blue and Grey Water Footprint. Water 10, 494. https://doi.org/10.3390/w10040494 Wang, R., Hertwich, E., Zimmerman, J.B., 2016. (Virtual) Water Flows Uphill toward Money. Environmental Science & Technology 50, 12320–12330. https://doi.org/10.1021/acs.est.6b03343 Wang, X., Huang, K., Yu, Y., Hu, T., Xu, Y., 2016. An input–output structural decomposition analysis of changes in sectoral water footprint in China. Ecological Indicators 69, 26–34. https://doi.org/10.1016/j.ecolind.2016.03.029 Wang, Z., Huang, K., Yang, S., Yu, Y., 2013. An input–output approach to evaluate the water footprint and virtual water trade of Beijing, China. Journal of Cleaner Production 42, 172–179. https://doi.org/10.1016/j.jclepro.2012.11.007 White, D.J., Feng, K., Sun, L., Hubacek, K., 2015. A hydro-economic MRIO analysis of the Haihe River Basin’s water footprint and water stress. Ecological Modelling 318, 157–167. https://doi.org/10.1016/j.ecolmodel.2015.01.017 Wiedmann, T., 2009. A review of recent multi-region input–output models used for consumption-based emission and resource accounting. Ecological Economics 69, 211–222. https://doi.org/10.1016/j.ecolecon.2009.08.026 Wiedmann, T., Lenzen, M., Turner, K., Barrett, J., 2007. Examining the global environmental impact of regional consumption activities — Part 2: Review of input–output models for the assessment of environmental impacts embodied in trade. Ecological Economics 61, 15–26. https://doi.org/10.1016/j.ecolecon.2006.12.003 World Bank, 2014. Renewable internal freshwater resources per capita (cubic meters). World Bank Open Data. Xu, Y., Huang, K., Yu, Y., Wang, X., 2015. Changes in water footprint of crop production in Beijing from 1978 to 2012: a logarithmic mean Divisia index decomposition analysis. Journal of Cleaner Production 87, 180–187. https://doi.org/10.1016/j.jclepro.2014.08.103 Yamakawa, A., Peters, G.P., 2011. STRUCTURAL DECOMPOSITION ANALYSIS OF GREENHOUSE GAS EMISSIONS IN NORWAY 1990–2002. Economic Systems Research 23, 303–318. https://doi.org/10.1080/09535314.2010.549461 Zhang, C., Anadon, L.D., 2014. A multi-regional input–output analysis of domestic 36

909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937

virtual water trade and provincial water footprint in China. Ecological Economics 100, 159–172. https://doi.org/10.1016/j.ecolecon.2014.02.006 Zhang, Z., Yang, H., Shi, M., 2016. Spatial and sectoral characteristics of China’s international and interregional virtual water flows – based on multi-regional input–output model. Economic Systems Research 28, 362–382. https://doi.org/10.1080/09535314.2016.1165651 Zhang, Z., Yang, H., Shi, M., 2011. Analyses of water footprint of Beijing in an interregional input–output framework. Ecological Economics 70, 2494–2502. https://doi.org/10.1016/j.ecolecon.2011.08.011 Zhao, D., Hubacek, K., Feng, K., Sun, L., Liu, J., 2019. Explaining virtual water trade: A spatial-temporal analysis of the comparative advantage of land, labor and water in China. Water Research 153, 304–314. https://doi.org/10.1016/j.watres.2019.01.025 Zhao, W., 2015. Empirical study on the water consumption behavior and its influences on the domestic water use: based on the data collected in Beijing. Journal of Arid Land Resources and Environment 29. https://doi.org/10.13448/j.cnki.jalre.2015.130 Zhao, X., Liu, J., Liu, Q., Tillotson, M.R., Guan, D., Hubacek, K., 2015. Physical and virtual water transfers for regional water stress alleviation in China. Proceedings of the National Academy of Sciences 112, 1031–1035. https://doi.org/10.1073/pnas.1404130112 Zhao, X., Yang, H., Yang, Z., Chen, B., Qin, Y., 2010. Applying the Input-Output Method to Account for Water Footprint and Virtual Water Trade in the Haihe River Basin in China. Environmental Science & Technology 44, 9150–9156. https://doi.org/10.1021/es100886r Zhong, F., Guo, A., Jiang, D., Yang, X., Yao, W., Lu, J., 2018. Research progress regarding residents’ water consumption behavior as relates to water demand management: a literature review. Advances in Water Science 29. https://doi.org/10.14042/j.cnki.32.1309.2018.03.017

938

37

Highlights:

Transitions in distribution and structure of China’s water footprint is identified.

Water footprint decoupled by efficiency improvement and structural optimization.

North-south gap enlarged with growingly higher per capita water footprint in north.

Household consumption, as primary source of water footprint, remained stable.

Growth of water footprint was mainly induced by fixed capital formation.