Features and evolution of international fossil fuel trade network based on value of emergy

Applied Energy 165 (2016) 868–877

Contents lists available at ScienceDirect

Applied Energy journal homepage: www.elsevier.com/locate/apenergy

Features and evolution of international fossil fuel trade network based on value of emergy Weiqiong Zhong, Haizhong An ⇑, Wei Fang, Xiangyun Gao, Di Dong School of Humanities and Economic Management, China University of Geosciences, Beijing, China Key Laboratory of Carrying Capacity Assessment for Resource and Environment, Ministry of Land and Resources, Beijing, China Lab of Resources and Environmental Management, China University of Geosciences, Beijing, China

h i g h l i g h t s
Number of trade relations and trade quantities follow power law distribution.
The pattern of top relations is diversified.
The trade density of fossil fuel is increasing.
Coal is the ‘‘cheapest” fuel measuring by ‘‘energy cost” and is most widely traded.
Countries with more than 20 trade relationships tend to have hierarchy structure.

a r t i c l e

i n f o

Article history: Received 20 July 2015 Received in revised form 8 December 2015 Accepted 17 December 2015 Available online 12 January 2016 Keywords: Fossil fuel International trade Emergy Complex network

a b s t r a c t Fossil fuel is crucial to the development of modern society. The major types of fossil fuel are coal, crude oil and natural gas. The uneven distribution of the production and consumption of fossil fuel makes the fossil fuel flows between countries by international trade. This study aims to quantitatively analyse the features and evolution of the international trade of fossil fuel by complex network and emergy. We transform the trade quantity of coal, crude oil and natural gas into emergy by transformity and the sum of the three emergies is the emergy of fossil fuel. The complex network models of the integrated fossil fuel trade as well as the trade of coal, crude oil and natural gas are built up based on the value of emergy. We analyse the trade relationships, trade quantity, trade density, and hierarchy structure of the networks. We find that the number of trade relationships and the trade quantities follow the power law distribution; countries with many export relationships tend to have many import relationships; the centralization of trade quantity is becoming more intense for fossil fuel, crude oil and coal, but less intense for natural gas; the pattern of top relationships is diversified; the trade density of fossil fuel is increasing; and countries with more than 20 trade relationships tend to have a hierarchy structure. Our findings implicate that as the hierarchy structure is becoming more ordered, the statuses of the countries are clearer, and thus it is easier for policy makers to identify the roles of their own countries or the roles of other countries. Coal is the ‘‘cheapest” fuel measuring by ‘‘energy cost” and is the most widely traded type of fossil fuel. When two countries exchange fossil fuel and money in the international trade, they should look further into the energy cost of them and reconsider the effectiveness of the trade. Our study can also reveal the trade strategy of the countries. Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction The international trade of fossil fuel is an integrated system with three major commodities: coal, crude oil and natural gas. ⇑ Corresponding author at: School of Humanities and Economic Management, China University of Geosciences, Beijing, China. Tel.: +86 01082323783; fax: +86 01082321783. E-mail address: [email protected] (H. An). http://dx.doi.org/10.1016/j.apenergy.2015.12.083 0306-2619/Ó 2015 Elsevier Ltd. All rights reserved.

According to the statistics of U.S. Energy Information Administration, the three major types of fossil fuels account to 86% of the world total primary energy consumption in 2012.1 There are numerous countries and complicated relationships in the international trade of fossil fuel which form a huge and complex system. A better understanding of the characteristics of this integrated

1

http://www.eia.gov/totalenergy/.

W. Zhong et al. / Applied Energy 165 (2016) 868–877

complex system can help us understand the international fossil fuel market [1]. Previous studies on international fossil fuel trade focused on energy security [2], trade patterns [3] and political factors [4]. This study aims to quantitatively analyze the features and evolution of international fossil fuel trade by combining network analysis and emergy transformity. It provides a new perspective for the study of international trade of fossil fuel. Complex network modeling has the advantage of analyzing the complex system of international trade. In 2003, Serrano et al. [5] introduced complex network model into the study of international trade. Then Garlaschelli et al. [6] studied the fitness-dependent topological properties of the international trade network. The study of Fagiolo et al. [7,8] provided a detailed quantitative analysis of the trade links and the role of the countries topologically and dynamically. Vidmer et al. [9] applied link prediction algorithms to predict the future evolution of the international trade network. In recent years, some scholars used complex network to analysis the international trade of energy. For example, Geng et al. [10] studied the structure and the integration of the international natural gas market by complex network. Üster et al. [11] designed an integrated large-scale mixed-integer nonlinear optimization model to analyze the natural gas transmission network. Zhong et al. [12] constructed weighted and unweighted complex network models to study the evolution of communities in the international oil trade. Ji et al. [13] introduced a global oil trade core network to analyze the overall features, regional characteristics and stability of the oil trade. Zhong et al. [13] and Zhang et al. [14] introduced complex network to analyze the competition between countries in the oil trade. However, as far as we know, most of the previous studies on international energy trade are based on single commodity. Our study provides an integrated view of the international trade of fossil fuel by considering the trade quantity of coal, crude oil and natural gas together in the model, and reveals features of the integrated system. A unified unit to measure the commodities of coal, crude oil and natural gas is needed because they are in different forms and qualities. Traditionally, money is applied to measure the integrated trade volume, however the fluctuating price and exchange rate [15] will affect the results. The unit of ‘‘joule” can be used to measure energy content of the fuels, however it only measures the ability to cause work. Exergy is another concept which measure the maximum useful work of the fuels [16]. These methods cannot reflect the ‘‘cost” of the energy which means how much energy is needed in order to produce a certain amount of fossil fuel. The main idea of Emergy is ‘‘energy cost” which regards the difference of energy quality and the accumulative cost of energy [17]. It measures the values of resources in common units of the solar energy used to make them (in unit of seJ) [18,19]. Transformity (in unit of seJ/J) can be used to transform the trade quantity of coal, crude oil and natural gas into emergy. The sum of the three emergies can be used to measure the emergy flow of fossil fuel. If a country exports fossil fuel, it not only exports the energy currently existing in the commodities, but also exports the energy consumed in forming, mining and producing the commodities. If a country imports fossil fuel, it also imports the embodied ‘‘energy cost” in the commodities. As far as we know, most of the previous studies of international energy trade use money, energy or exergy as trade quantity. Our study goes further in considering the accumulative amount of solar energy (Emergy) as trade quantity. In this study, we design the integrated complex network model of fossil fuel as well as the single commodity network models of coal, crude oil and natural gas based on the emergy flows among countries. The characteristics of the international fossil fuel trade can be reflected by network analysis. Section 2 introduces the data and the process of modeling. Four indexes of network analysis are

869

introduced: degree and strength are indicators of the individual countries, and network density and hierarchy structure are indicators of the whole network. Section 3 is the analysis of trade relationships, trade quantity, trade density, and hierarchy structure of the network. Section 4 is the discussion and conclusion remarks. 2. Data and method 2.1. Data and transformity The data of international trade of coal, crude oil and natural gas is from the website of UN Comtrade which contains all the export and import flows among 226 countries. The trade volumes are measured by kilogram. We selected the annual data of all the available countries from 2000 to 2013. We transformed the trade quantities of the three fuels into emergy and the sum of them is the emergy of fossil fuel. The description of the data source, the energy content of the commodities and the transformity of coal, crude oil and natural gas are shown in Table 1. In our data, only crude oil is included in the HS Code 2709, and there are several categories of coal in the HS Code 2701. We use the average energy content and the average emergy transformity of crude oil and coal in our study. The total emergy in fossil fuel trade increased during 2000–2008 as the world economy grew, and the total emergy declined in 2009 after the US mortgage subprime crisis.2 The majority of fossil fuel trade emergy was contributed by crude oil, coal contributed the least emergy, and natural gas contributed a little more than coal (please see Fig. 1). 2.2. International trade network model The complex network model G = (V, E) contains the nodes V and the edges E, where V = {vi:i = 1, 2, . . ., n}, n is the number of nodes, E = {ei:i = 1, 2, . . ., m}, and m is the number of edges. In our model, the nodes are the countries, the edges are the trade relationships, the directions of the edges are the directions of the emergy flows, and the weights of the edges are the value of emergies. We constructed network models of the integrated fossil fuel trade as well as the single commodities based on the transformed data. An example of the integrated fossil fuel trade network in 2012 is shown in Fig. 2. We filtered the network with trade quantity in order to make it more readable by showing the top 50 countries in trade quantity in the network. The size of the node is the total trade quantity of the country. The larger the node is, the more emergy the country has trade in this year. The width of the edge is the value of the emergy of this trade link. The wider the edge is, the higher value of emergy this trade link has. 2.2.1. Degree: the range of the direct impact Degree is the number of direct trade relationships of a country. It reflects the range of a country’s direct impact. The out-degree is the number of export links a country has with others, and the indegree is the number of import links. The higher value of outdegree or in-degree indicates a wider range of the country’s direct impact. These values are computed by [21] out

ki ðtÞ ¼

n X dij ðtÞ

ð1Þ

j¼1 in

ki ðtÞ ¼

n X dji ðtÞ

ð2Þ

j¼1 2 The U.S. subprime mortgage crisis was a nationwide banking emergency that coincided with the U.S. recession of December 2007–June 2009 (explanation from Wikipedia).

870

W. Zhong et al. / Applied Energy 165 (2016) 868–877

Table 1 Data description, energy content and transformity. Commodity

HS code

Description

Energy content [20]

Transformity [19]

Coal Crude oil Natural gas

2701 2709 271,111 271,121

Coal; briquettes, ovoids and similar solid fuels manufactured from coal Petroleum oils and oils obtained from bituminous minerals, crude Natural gas, liquefied (LNG) Natural gas in gaseous state (NG)

2.094E4 J/g 4.337E4 J/g

8.17E4 seJ/J 1.48E5 seJ/J

3.883E7 J/m3

1.71E5 seJ/J

Note: Data source is http://comtrade.un.org/. In the data source, the unit of the commodities is kilogram, thus we convert the units by 1 kg of NG = 1400 L of NG, 1 kg of LNG = 2.35 L of LNG, and 1 L of LNG = 600 L of NG. The geobiosphere baseline is 15.2E24 seJ/yr. The transformity of coal is the average of hard coal and soft coal according to [19].

where if country i exports oil to country j during year t, a link from i to j is drawn, and dij(t) = 1. Otherwise, no link is drawn, and dij(t) = 0. out

The out-degree ki ðtÞ of country i in the year t is the sum of dij(t), in

and the in-degree ki ðtÞ of country i in the year t is the sum of dji(t). If the network has a degree distribution that can be fit with a power law distribution (4), it implies that the network is a scalefree network, where c is the power law index and k is the degree of the nodes [10]. c

PðkÞ  k

ð3Þ

2.2.2. Strength: the quantity of emergy The total trade quantity of emergy of a country can be measured by strength in the network. The out-strength sout i ðtÞ and in-strength sin ðtÞ of country i reflect a node’s importance in the network coni sidering both relationships and quantities of emergy. The higher in the value is, the more important the country is. sout i ðtÞ and si ðtÞ are computed by [21]

sout i ðtÞ ¼

Crude oil

Natural gas

n X dji ðtÞ  wji ðtÞ

ð5Þ

j¼1

where wi,j(t) is the weight of dij(t), which is the total amount of emergy that country i exports to country j during the year t.

2.50E+25

Total Emergy (seJ)

ð4Þ

j¼1

sin i ðtÞ ¼ Coal

n X dij ðtÞ  wij ðtÞ

2.00E+25

2.2.3. Network density: the tightness of relationships among countries Network density can be used to measure the tightness of the trade relationships among the countries in the fossil fuel trade network. It equals to ‘‘total number of relationships that actually exist” divided by ‘‘maximum number of relationships that theoretically can exist‘‘. If the number of actual relationship is m, the number of nodes is n, then the network density is [10]:

1.50E+25 1.00E+25 5.00E+24 0.00E+00

Year Fig. 1. Total emergy in the 4 types of trade networks.

D¼

2m nðn  1Þ

Fig. 2. Filtered network model of international fossil fuel trade in 2012.

ð6Þ

871

W. Zhong et al. / Applied Energy 165 (2016) 868–877

fossil fuel are shown in Table 2. We can see that the top 10 countries were mainly from North America, Europe and East Asia area. From 2000 to 2009, the USA was the country with the largest number of import relationships. However, in 2010 and 2013 China became No. 1, and in 2011 and 2012 India was No. 1 in import relationship. The top 10 countries in number of export relationships of fossil fuel are shown in Table 3. An interesting phenomenon is that many importing countries were also with high out-degrees. The USA ranked No. 1 in the number of export relationships through the whole observation period. The rank of China was also increasing, and became No. 2 since 2006.

2.2.4. Hierarchy structure: the order of the trade network Clustering coefficient of a country is the probability of trade relationship existing between the countries connecting to this country in the network. It reflects the connectivity of the neighboring countries of this country. If a country’s neighbors are closely related, the country has a higher clustering coefficient; on the contrary, if a country’s neighbors are loosely related, the clustering coefficient of this country is lower. If nodes with the same degree have similar clustering coefficient, the hierarchy structure of the network is more ordered because similar roles have similar connectivity. The clustering coefficient Ci of node i with degree ki is computed by:

C i ¼ ni =ki ðki  1Þ

3.2. Trade quantity

ð7Þ The trade quantities of emergy are carried by the trade links (edges), thus we analyzed the accumulative distributions of the weights of the edges in the 4 types of networks each year. The results of 3 years (2000, 2006, and 2013) are shown in Fig. 4. We should focus on the gaps of the curves which were moving toward up left corner. This implies that the trade of coal, crude oil and fossil fuel were becoming more concentrated from 2000 to 2013. However, the tendency of natural gas was in opposite direction. It was less concentrated. The proportions of edges shouldering 80% of the trade quantity are shown in Fig. 5. A small number of trade links shoulder a large part of the trade quantities. This phenomenon is less obvious in the network of crude oil, and is more obvious in the network of natural gas. We can see that less than 8% of the trade relationships contain up to 80% of the trade quantity of emergy in the fossil fuel trade network. The top 10 countries in importing emergy of fossil fuel are shown in Table 4. We can see that the top 10 countries are mainly

where ni is the number of the edges among the neighbors of node i. 3. Results and analysis 3.1. Trade relations Degree is the number of edges of a node in the network. It is an index measuring how many countries have trade relationships with a given country. It indicates the activeness of a country in the network. Countries with higher degrees possess important roles, because they have wider range of trade, and their impacts can directly reach more partners. The trade relationships in the 4 types of networks each year follow power law distribution. A small number of countries own many trade partners and most of the countries own a few trade partners (the figures of 2000, 2006 and 2013 are shown in Fig. 3). The top 10 countries in number of import relationships of

ln (p(k))

0

0 4

6

0

y = -0.7363x - 2.1871 R² = 0.7843

-4

2

-2

4

0

6

y = -0.6854x - 2.2784 R² = 0.7389

-4

-2

ln (k)

2

ln (k)

0 -2

ln (k)

ln (p(k))

ln (p(k))

0

4

6

y = -0.6164x - 2.5196 R² = 0.7035

-4

-6

-6

-6

-8

-8

-8

2000

2

2013

2006 Fig. 3. Power law of the number of trade partners.

Table 2 Top 10 countries in in-degree in fossil fuel trade. Year

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

Rank 1

2

3

4

5

6

7

8

9

10

USA USA USA USA USA USA USA USA USA USA China India India China

Italy France Germany China France France China Germany Germany China USA USA China Netherlands

France UK France Germany Germany Germany Germany China China India India China Netherlands USA

Germany Italy Spain Spain China China France UK France Germany UK UK USA Germany

UK Spain Italy Italy Spain UK UK France India UK France Germany France South Korea

Spain Germany Netherlands France UK Spain Spain Italy Canada Spain Italy France Germany India

China China China UK Italy Netherlands Canada Spain Netherlands France Germany Italy South Korea France

Netherlands Netherlands South Korea Canada South Korea Italy Netherlands Canada Spain South Korea Canada Netherlands Italy UK

South Korea South Korea UK Netherlands Canada Canada South Africa Netherlands UK Canada Spain South Korea UK Canada

Singapore Thailand Canada Belgium Netherlands South Korea Italy India Italy Netherlands South Korea Spain Japan Italy

872

W. Zhong et al. / Applied Energy 165 (2016) 868–877

Table 3 Top 10 countries in out-degree in fossil fuel trade. Year

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

Rank 1

2

3

4

5

6

7

8

9

10

USA USA USA USA USA USA USA USA USA USA USA USA USA USA

South Africa UK South Africa UK UK South Africa China China China China China China China China

Russia Russia UK Russia Russia China South Africa South Africa South Africa UK UK South Africa South Africa UK

UK South Africa Russia South Africa China Germany UK UK UK South Africa South Africa Russia UK South Africa

Germany Germany Germany Germany South Africa UK Germany Germany Germany Nigeria Russia UK Russia Russia

China China Italy China Germany Russia Russia India Russia Russia Germany Germany Germany Netherlands

France Australia China Australia Australia Italy Netherlands Russia France Germany Nigeria Colombia France Germany

Netherlands Italy Australia Indonesia Italy Ukraine Italy Italy UAE UAE Ukraine Netherlands Netherlands France

Australia France UAE Italy Netherlands Australia UAE France Australia Italy Italy Ukraine Italy Italy

Italy Netherlands Indonesia Netherlands France France Indonesia Netherlands Netherlands Australia Netherlands Italy Ukraine Ukraine

Fig. 4. Accumulative distribution of the weights (emergy).

from North America, East Asia and Europe. The USA ranked No. 1 in most of the years, and China became No. 1 importing country in 2013. Japan ranked No. 2 in most of the years except in 2012. The top 10 countries in exporting emergy of fossil fuel are shown in Table 5. Russia replaced Saudi Arabia and became the No. 1 exporting country since 2001, and Saudi Arabia had been No. 2 ever since. Top 10 trade relationships in value of emergy in fossil fuel trade are shown in Table 6. We can see that the trade relationship with highest value of emergy is from Canada to the USA. In the early years, the flows from North America and South America to the

USA were the top flows with high emergy. However, as the imports of the USA and some other developed countries decreased and the imports of some developing countries increased, the ranks of top 10 relationships were changing. 3.3. Trade density The trade density of the fossil fuel trade was increasing from 2000 to 2013 (please see Fig. 6). The trade densities of coal and crude oil were at the same level, while the trade density of natural gas was much lower than the others. This may due to the

873

W. Zhong et al. / Applied Energy 165 (2016) 868–877

Fossil

Coal

Oil

second most, and the natural gas was the least (please see Fig. 7 (a)). Although the number of countries was increasing slightly, the number of trade links among them was increasing obviously. As a result the trade density was increasing. Consistent with the feature of the number of countries, the number of trade links of coal, crude oil and natural gas had the same features. The number of trade links of coal was the most, the number of trade links of crude oil was the second most, and the number of trade links of natural gas trade was much less than the others (please see Fig. 7(b)).

Gas

14%

Proporon

12% 10% 8% 6% 4% 2% 0%

3.4. Hierarchy structure

Year Fig. 5. Proportion of edges shouldering 80% of emergy.

restriction of transportation. This is because the majority of natural gas was transported by pipeline and LNG tankers. The cost of transportation made it is harder for natural gas to be widely traded between distant countries. To look further into the trade density, we plotted the number of countries and the number of trade relationships of the 4 types of networks. There were around 200 countries participating in the fossil fuel trade during the observation years. The total numbers of countries were slightly increasing in the four types of networks during the observation period. The numbers of countries in the three types of single commodities were similar. The crude oil trade was the

Degree indicates the direct impact of a country. Countries with higher degree play important roles, because they directly affect more countries. Clustering coefficient reflects the connectivity of the neighboring countries of this country. If nodes with the same degree have similar clustering coefficient, the hierarchy structure of the network is more ordered because similar roles have similar connectivity. We plotted the degree and the clustering coefficient of all the countries in scatter diagrams chronologically in Fig. 8. The abscissa is degree and the ordinate is the value of clustering coefficient [22]. From the distribution of the points we can see that countries with higher degree tend to have lower clustering coefficient. The centrality of the points in Fig. 8 reflects the hierarchy structure of the network. We can see that the hierarchy structure of fossil fuel trade network is not obvious because nodes have similar degree

Table 4 Top 10 countries in in-strength in fossil fuel trade. Year

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

Rank 1

2

3

4

5

6

7

8

9

10

USA USA USA USA USA USA USA USA USA USA USA USA USA China

Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan Japan China Japan

South Korea South Korea South Korea South Korea South Korea France France Italy France China China China Japan USA

Germany France Germany Germany France Italy South Korea France China South Korea Italy South Korea India India

Italy Germany France France Germany South Korea Italy South Korea South Korea India South Korea India South Korea South Korea

France Italy Netherlands Italy Italy Germany Netherlands China Italy Italy India Italy Italy Germany

Netherlands Netherlands Italy Netherlands Netherlands Netherlands China Germany India France Netherlands France Netherlands Italy

Spain Spain Spain China China China Germany India Germany Germany France Netherlands Germany Croatia

China UK UK Spain Spain Belgium Belgium Netherlands Netherlands Netherlands Germany Germany UK France

UK Ukraine Ukraine Belgium UK Spain UK Spain Spain UK UK UK Spain UK

Table 5 Top 10 countries in out-strength in fossil fuel trade. Year

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

Rank 1

2

3

4

5

6

7

8

9

10

Saudi Arabia Russia Russia Russia Russia Russia Russia Russia Russia Russia Russia Russia Russia Russia

Russia Saudi Arabia Saudi Arabia Saudi Arabia Saudi Arabia Saudi Arabia Saudi Arabia Saudi Arabia Canada Saudi Arabia Saudi Arabia Saudi Arabia Saudi Arabia Saudi Arabia

Norway Norway Norway Canada Canada Canada Canada Canada Saudi Arabia Norway Norway Qatar Norway Qatar

Canada Canada Canada Norway Norway Norway Norway Algeria Norway Canada Canada Norway Indonesia Norway

Algeria UK UK Algeria Algeria Algeria Algeria Norway Qatar Australia Algeria Canada Australia Indonesia

UK Venezuela Australia Iran Iran UAE Nigeria Iran Algeria Indonesia Qatar Indonesia Qatar Australia

UAE Australia Mexico Australia Nigeria Iran UAE Nigeria UAE Algeria Indonesia Australia Canada Canada

Iran UAE Algeria UAE UAE Venezuela Iran UAE Australia Iran Australia Nigeria UAE UAE

Indonesia Iran UAE Mexico Australia Australia Australia Australia Nigeria Nigeria Nigeria Algeria Nigeria Kazakhstan

Australia Mexico Indonesia UK Venezuela Nigeria Venezuela Indonesia Iran UAE Iran Kazakhstan Kazakhstan Netherlands

874

W. Zhong et al. / Applied Energy 165 (2016) 868–877

Table 6 Top 10 trade relationships in value of emergy in fossil fuel trade. Rank

Exporter

Importer

Fossil emergy (seJ)

Rank

Exporter

Importer

Fossil emergy (seJ)

2000 1 2 3 4 5 6 7 8 9 10

Canada Saudi Arabia Venezuela Mexico UAE Saudi Arabia Nigeria Australia Norway Norway

USA USA USA USA Japan Japan USA Japan UK Germany

1.09E+24 4.45E+23 4.43E+23 4.17E+23 3.80E+23 3.45E+23 3.03E+23 3.00E+23 2.95E+23 2.69E+23

2001 1 2 3 4 5 6 7 8 9 10

Canada Saudi Arabia Venezuela Mexico UAE Saudi Arabia Australia Norway Nigeria Norway

USA USA USA USA Japan Japan Japan UK USA Germany

1.24E+24 5.14E+23 4.65E+23 4.42E+23 3.68E+23 3.49E+23 3.08E+23 3.06E+23 2.80E+23 2.78E+23

2002 1 2 3 4 5 6 7 8 9 10

Canada Saudi Arabia Mexico Venezuela UAE Saudi Arabia Norway Australia Norway Russia

USA USA USA USA Japan Japan Germany Japan UK Ukraine

1.19E+24 4.77E+23 4.73E+23 4.55E+23 3.45E+23 3.39E+23 3.26E+23 3.06E+23 3.05E+23 2.88E+23

2003 1 2 3 4 5 6 7 8 9 10

Canada Saudi Arabia Mexico Venezuela UAE Norway Saudi Arabia Russia Norway Australia

USA USA USA USA Japan UK Japan Ukraine Germany Japan

1.46E+24 5.55E+23 4.98E+23 4.55E+23 3.62E+23 3.54E+23 3.53E+23 3.25E+23 3.22E+23 3.14E+23

2004 1 2 3 4 5 6 7 8 9 10

Canada Venezuela Mexico Saudi Arabia UAE Norway Saudi Arabia Russia Nigeria Australia

USA USA USA USA Japan UK Japan Ukraine USA Japan

1.52E+24 5.17E+23 4.97E+23 4.84E+23 3.77E+23 3.67E+23 3.59E+23 3.54E+23 3.48E+23 3.31E+23

2005 1 2 3 4 5 6 7 8 9 10

Canada UAE Venezuela Mexico Saudi Arabia Saudi Arabia Norway Netherlands Australia Nigeria

USA Japan USA USA USA Japan UK Belgium Japan USA

1.94E+24 6.10E+23 5.12E+23 4.88E+23 4.64E+23 4.14E+23 3.97E+23 3.92E+23 3.58E+23 3.58E+23

2006 1 2 3 4 5 6 7 8 9 10

Canada Mexico Venezuela Saudi Arabia Saudi Arabia Norway Netherlands Belgium UAE Australia

USA USA USA USA Japan UK Belgium France Japan Japan

1.74E+24 5.03E+23 4.67E+23 4.52E+23 4.36E+23 4.24E+23 4.16E+23 3.88E+23 3.85E+23 3.72E+23

2007 1 2 3 4 5 6 7 8 9 10

Canada Algeria Belgium Mexico Venezuela Saudi Arabia Norway Netherlands Australia Saudi Arabia

USA Italy France USA USA USA UK Belgium Japan Japan

1.86E+24 5.67E+23 5.28E+23 4.65E+23 4.61E+23 4.53E+23 4.39E+23 4.09E+23 4.01E+23 3.79E+23

2008 1 2 3 4 5 6 7 8 9 10

Canada Belgium Saudi Arabia Norway Netherlands Venezuela Australia Mexico Saudi Arabia UAE

USA France USA UK Belgium USA Japan USA Japan Japan

2.55E+24 5.07E+23 4.80E+23 4.50E+23 4.25E+23 4.22E+23 4.10E+23 3.87E+23 3.83E+23 3.66E+23

2009 1 2 3 4 5 6 7 8 9 10

Canada Norway Venezuela Australia Saudi Arabia Mexico Russia Russia Saudi Arabia UAE

USA UK USA Japan Japan USA Netherlands Italy USA Japan

8.58E+23 4.02E+23 3.94E+23 3.71E+23 3.68E+23 3.42E+23 3.39E+23 3.29E+23 3.27E+23 2.91E+23

2010 1 2 3 4 5 6 7 8 9 10

Canada Algeria Mexico Norway Australia Russia Venezuela Saudi Arabia Saudi Arabia Nigeria

USA Italy USA UK Japan Netherlands USA Japan USA USA

1.07E+24 4.81E+23 4.55E+23 4.55E+23 4.19E+23 3.84E+23 3.58E+23 3.54E+23 3.47E+23 3.23E+23

2011 1 2 3 4 5 6 7 8 9 10

Canada Mexico Norway Australia Saudi Arabia Saudi Arabia Venezuela Qatar Saudi Arabia UAE

USA USA UK Japan USA Japan USA Japan China Japan

1.15E+24 4.63E+23 4.38E+23 3.94E+23 3.83E+23 3.73E+23 3.41E+23 3.33E+23 3.23E+23 3.09E+23

2012 1 2 3 4 5

Canada Australia Norway Saudi Arabia Saudi Arabia

USA Japan UK USA Japan

6.91E+23 4.37E+23 4.32E+23 4.27E+23 3.82E+23

2013 1 2 3 4 5

Canada Australia Mozambique Hungary Qatar

USA Japan South Africa Croatia Japan

7.90E+23 4.76E+23 4.28E+23 4.26E+23 4.25E+23

875

W. Zhong et al. / Applied Energy 165 (2016) 868–877 Table 6 (continued) Rank

Exporter

Importer

Fossil emergy (seJ)

Rank

Exporter

Importer

Fossil emergy (seJ)

6 7 8 9 10

Saudi Arabia Russia Mexico UAE Netherlands

China Netherlands USA Japan Belgium

3.46E+23 3.44E+23 3.07E+23 3.01E+23 2.84E+23

6 7 8 9 10

Saudi Arabia Norway Saudi Arabia Norway Saudi Arabia

USA UK Japan Germany China

3.96E+23 3.77E+23 3.66E+23 3.65E+23 3.46E+23

4. Discussion and conclusion

Fossil

Coal

Crude oil

Natural gas

In this paper, we constructed the integrated trade network of fossil fuel based on emergy value, as well as the single commodity networks of coal, crude oil and natural gas. The trade quantity of coal, crude oil and natural gas were transformed into emergy and the sum of them was the emergy of fossil fuel. We studied trade relationships, trade quantity, trade density and hierarchy structure of the networks. These indexes reflected the individual and entire features of the fossil fuel trade network. Our observation period was from 2000 to 2013, we looked further into the evolution of these features over time. Our conclusions and discussions are as follows:

0.07 0.06

0.04 0.03 0.02 0.01 0

(1) The numbers of trade relationships of the single countries follow power law distribution. A small number of countries own many trade partners and most of the countries own a few trade partners. The top 10 countries with the largest number of import or export relationships are mainly in North America, Europe and East Asia area. Countries with high in-degree also tend to have high out-degree, for example the USA, China and Germany. The number of trade relationships can reflect a country’s activeness in international trade. If a country is active, although it is a net importing country, it will still have many export flows with small value of emergy to other countries especially to its neighboring countries. Also, these countries tend to have many big energy companies, which not only target to domestic market, but also target to the world market. (2) A small number of trade links shoulder most of the trade quantities. Less than 8% of the trade relationships contain up to 80% of the trade quantity of emergy in the fossil fuel trade network. The centralization of trade quantity of fossil fuel was becoming more intense, however, natural gas had

Year Fig. 6. Trade density of the 4 types of networks.

appear to have various clustering coefficient. However, we can still find some clues of hierarchy structure when we observe the low degree and high degree separately. Take the year 2013 as an example. The R2 of linear regression of all the nodes was 0.0649 (please see Fig. 9(a)), which indicates that there was no linear relation between degree and clustering coefficient. However, as we deleted the nodes with low degree, the R2 of linear regression is increasing. When there were nodes with degree above 17, the R2 of linear regression was 0.6174 (please see Fig. 9 (b)), which indicates that there was linear relation between degree and clustering coefficient. We recorded the R2 when we deleted nodes with degree from 1 to 50 in the year 2000, 2006 and 2013 (please see Fig. 10). The R2 reached 0.6 when there were nodes with degree more than about 20. It was faster for R2 to reach 0.6 from 2000 to 2013, which means the hierarchy structure was becoming more ordered.

Fossil

Crude oil

Coal

Fossil

Coal

Natural gas

Crude oil

Natural gas

Total number of edges

3000

200 150 100 50

2500 2000 1500 1000 500

Year

Year

(a)

(b)

Fig. 7. Total number of countries and trade links of the 4 types of networks.

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

0 2001

0 2000

Total number of nodes

250

2000

Density

0.05

876

W. Zhong et al. / Applied Energy 165 (2016) 868–877

Fig. 8. Scatterplots of degree and clustering coefficient of fossil fuel trade network.

k>0

k>17 0.8

Clustering coefficient

Clustering coefficient

1.2 1 0.8

y = -0.002x + 0.5058 R² = 0.0649

0.6 0.4 0.2

0.6

y = -0.0029x + 0.5629 R² = 0.6174

0.4 0.2 0

0 0

100

200

0

50

100

Degree (k)

Degree (k)

(a)

(b)

150

200

Fig. 9. The linear regression of degree and clustering coefficient in fossil fuel trade network in 2013.

2000

2006

2013

0.8 0.7 0.6

R2

0.5 0.4 0.3 0.2 0.1 0 1

6

11

16

21

26

31

36

41

46

51

Degree Fig. 10. R2 of the linear regression of degree and clustering coefficient in fossil fuel trade (the abscissa is the threshold of degree, countries with degrees equal to or higher than the threshold is remain in the regression).

an opposite tendency. The centralization of trade quantity of natural gas was less intense. The increasing of production, especially the development of unconventional gas (such as shale gas in the USA), reshaped the supply pattern of natural gas and made it less centralized to a small number of trade relationships. (3) The trade density of fossil fuel is increasing, and coal is the ‘‘cheapest” energy measured by ‘‘energy cost” which is being most widely traded. Due to the globalization of fossil fuel trade, more countries were participating in the world fossil fuel trade, and more relationships among countries were built up. In the traditional study of single fuel, we can only analyse the features of one part of the fossil fuel market, our study can reveal the tendency of the integrated market and it also easily to compare the features of different fuels. In our results, an interesting phenomenon is that although crude oil contributes the most to the total emergy of fossil

W. Zhong et al. / Applied Energy 165 (2016) 868–877

fuel trade, the number of countries and the number of relationships of crude oil is not the largest. Coal contributes the least emergy to the total, however, it has the largest number of countries and relationships. This phenomenon indicates that when considering the ‘‘energy cost” of the geobiosphere and the producing process, coal is the ‘‘cheapest” energy that is being traded most widely. This conclusion is based on the concept of emergy and cannot obtained by the traditional study based on money or exergy. When we look further into the calculation of emergy values of the three types of fuels, we can see that the energy content of coal is not that small, however, when transformed into emergy, its emergy value is much less than crude oil and natural gas. This means that coal consumed less energy in the geobiosphere process and the producing process. This is the reason for the small contribution of coal in the total emergy of fossil fuel. Another point is that the energy cost in the trade quantity should be considered when making international trade policies. Emergy flows contain the energy cost in the producing process of the exporting countries. At the same time, the money of a country also contains emergy in it. Both of the energy costs in fossil fuel and money can reflect the energy cost of this country. Thus when two countries exchange fossil fuel and money in the international trade, they should look further into the energy cost of them and reconsider the effectiveness of the trade.

877

relationships. Mexico was one of the top 10 countries only in 2001–2003, however, it was on the list of top 10 relationships in most of the years. This is because Indonesia equally exported its fossil fuel to Southeast Asian countries such as India, China, Japan and South Korea. Although the value of the single emergy flow was not high, the total amount of its export was large. On the contrary the fossil fuel export of Mexico was concentrated on the link with the USA. The emergy values of its trade links with other countries were much lower. Due to the limit of data, the observation period of this paper is only from 2000 to 2013. In the future, we can expand the observation period to several decades and introduce more indexes of network analysis in order to find more features of the international fossil fuel trade network. Acknowledgements This research is supported by grants from the National Natural Science Foundation of China (Grant No. 71173199) and the Humanities and Social Sciences Planning Funds project under the Ministry of Education of the PRC (Grant No. 10YJA630001). The authors would like to express their gratitude to Mark T. Brown who helped a lot during their work. References

(4) The evolution tendency of the trade density of natural gas is denser, and according to the accumulative distribution of trade quantity the trade of natural gas is becoming more diversified. This implicates that more trade relationships of natural gas will be built and the trade volume will not be concentrated in a few trade links. Thus, for exporting countries it is an opportunity to extend their sales markets and enhance their status. For importing countries, it is also an opportunity to seek for more importing sources and increase the importing volume of the existing trade links. More pipelines are needed to be built, and the progress of the technology of liquefaction and regasification will extend the trade scale of LNG. (5) Countries with more than 20 trade relationships tend to have a hierarchy structure, which means countries with similar roles tend to have similar connectivity. This phenomenon was becoming more pronounced during the observation period. Another interesting finding is that countries with more trade relationships tend to have lower connectivity among its neighboring countries. This is because countries with high degree have wider trade ranges. Their trade partners are located all over the world, thus the probability of building up trade relationships between neighboring countries is lower. As the hierarchy structure of the international fossil fuel trade network is becoming more ordered, the statuses of the countries are clearer. It is easier for policy makers to identify the roles of their own countries or the roles of other countries. The impact of a country is spreading faster when this country has better connectivity. (6) Our results can also reveal the trade strategy of the countries. For example, in the early years, USA and Japan were the main exporting target of Saudi Arabia. In recent years, the emergy amounts of Saudi Arabia’s trade flow to China were increasing fast due to the booming of Chinese economy and China’s rocketing demand for fossil fuel. Another example is that although Indonesia was one of the top 10 countries in exporting fossil fuel, it was not on the list of top 10

[1] Ji Q, Guo JF. Oil price volatility and oil-related events: an internet concern study perspective. Appl Energy 2015;137:256–64. [2] Mityakov S, Tang H, Tsui KK. Geopolitics, global patterns of oil trade, and China’s oil security quest. HKIMR working paper; 2011. No. 32. [3] Zhang HY, Ji Q, Fan Y. What drives the formation of global oil trade patterns? Energy Econ 2015;49:639–48. [4] Kashcheeva M. Political limits on the world oil trade: firm-level evidence from US firms. IDE discussion paper; 2013. No. 401. [5] Serrano MA, Boguna M. Topology of the world trade web. Phys Rev E 2003;68. [6] Garlaschelli D, Loffredo MI. Fitness-dependent topological properties of the world trade web. Phys Rev Lett 2004;93. [7] Fagiolo G, Reyes J, Schiavo S. World-trade web: topological properties, dynamics, and evolution. Phys Rev E 2009;79. [8] Fagiolo G, Reyes J, Schiavo S. The evolution of the world trade web: a weightednetwork analysis. J Evolut Econ 2010;20:479–514. [9] Vidmer A, Zeng A, Medo M, Zhang Y-C. Prediction in complex systems: the case of the international trade network. Physica A 2015;436:188–99. [10] Geng JB, Ji Q, Fan Y. A dynamic analysis on global natural gas trade network. Appl Energy 2014;132:23–33. [11] Uster H, Dilaveroglu S. Optimization for design and operation of natural gas transmission networks. Appl Energy 2014;133:56–69. [12] Zhong WQ, An HZ, Gao XY, Sun XQ. The evolution of communities in the international oil trade network. Physica A 2014;413:42–52. [13] Ji Q, Zhang HY, Fan Y. Identification of global oil trade patterns: an empirical research based on complex network theory. Energy Convers Manage 2014;85:856–65. [14] Zhang HY, Ji Q, Fan Y. Competition, transmission and pattern evolution: a network analysis of global oil trade. Energy Policy 2014;73:312–22. [15] An HZ, Gao XY, Fang W, Huang X, Ding YH. The role of fluctuating modes of autocorrelation in crude oil prices. Physica A 2014;393:382–90. [16] Qi H, An HZ, Hao XQ, Zhong WQ, Zhang YB. Analyzing the international exergy flow network of ferrous metal ores. PLoS ONE 2014;9. [17] Brown MT, Odum HT, Jorgensen SE. Energy hierarchy and transformity in the universe. Ecol Model 2004;178:17–28. [18] Brown MT, Herendeen RA. Embodied energy analysis and EMERGY analysis: a comparative view. Ecol Econ 1996;19:219–35. [19] Brown MT, Protano G, Ulgiati S. Assessing geobiosphere work of generating global reserves of coal, crude oil, and natural gas. Ecol Model 2011;222:879–87. [20] Bastianoni S, Campbell DE, Ridolfi R, Pulselli FM. The solar transformity of petroleum fuels. Ecol Model 2009;220:40–50. [21] Garlaschelli D, Loffredo MI. Structure and evolution of the world trade network. Physica A 2005;355:138–44. [22] Duan WQLB. Research on the measurement and evolution model of world trade networks. Beijing: Guangming Daily Press; 2011.