当前位置：和泉文库 > 交通运输 > 上海交通大学：《物流地理学 Geography of Logistics》课程教学资源（教学资料）The complex network of global cargo ship movements

上海交通大学：《物流地理学 Geography of Logistics》课程教学资源（教学资料）The complex network of global cargo ship movements

文件格式：PDF，文件大小：1.13MB，售价：5.37元

文档详细内容（约12页）

Downloaded from rsif.royalsocietypublishing.org on November 19,2012 JOURNAL THE ROYAL FirstCite J.R.Soc.Interface nterface doi10.1098/sif.2009.0495 e-publishing Published online The complex network of global cargo ship movements Pablo Kaluza,Andrea Kolzsch,Michael T.Gastner and Bernd Blasius* Institute for Chemistry and Biology of the Marine Environment,Carl von Ossietzky Universitit,Carl-von-Ossietzky-Strafe 9-11,26111 Oldenburg,Germany Transportation networks play a crucial role in human mobility,the exchange of goods and the spread of invasive species.With 90 per cent of world trade carried by sea,the global network of merchant ships provides one of the most important modes of transportation. Here,we use information about the itineraries of 16 363 cargo ships during the year 2007 to construct a network of links between ports.We show that the network has several fea- tures that set it apart from other transportation networks.In particular,most ships can be classified into three categories:bulk dry carriers,container ships and oil tankers.These three categories do not only differ in the ships'physical characteristics,but also in their mobility patterns and networks.Container ships follow regularly repeating paths whereas bulk dry carriers and oil tankers move less predictably between ports.The network of all ship movements possesses a heavy-tailed distribution for the connectivity of ports and for the loads transported on the links with systematic differences between ship types The data analysed in this paper improve current assumptions based on gravity models of ship movements,an important step towards understanding patterns of global trade and bioinvasion. Keywords:complex network;cargo ships;bioinvasion;transportation 1.INTRODUCTION human livelihoods,health and local economies (Mack The ability to travel,trade commodities and share et al.2000).The financial loss owing to bioinvasion is information around the world with unprecedented effi- estimated to be $120 billion per year in the USA alone (Pimentel et al.2005). ciency is a defining feature of the modern globalized economy.Among the different means of transport, Despite affecting everybody's daily lives,the ship- ocean shipping stands out as the most energy efficient ping industry is far less in the public eye than other sectors of the global transport infrastructure.Accord- mode of long-distance transport for large quantities of goods(Rodrigue et al.2006).According to estimates, ingly,it has also received little attention in the recent as much as 90 per cent of world trade is hauled by literature on complex networks (Wei et al.2007;Hu ships (International Maritime Organization 2006).In Zhu 2009).This neglect is surprising considering the 2006,7.4 billion tons of goods were loaded at the current interest in networks (Albert Barabasi 2002: world's ports.The trade volume currently exceeds 30 Newman 2003a;Gross Blasius 2008),especially air- trillion ton-miles and is growing at a rate faster than port (Barrat et al.2004;Guimera Amaral 2004; the global economy (United Nations Conference on Hufnagel et al.2004;Guimera et al.2005),road (Buhl Trade and Development 2007). et al.2006;Barthelemy Flammini 2008)and train The worldwide maritime network also plays a crucial networks (Latora Marchiori 2002;Sen et al.2003). role in today's spread of invasive species.Two major In the spirit of current network research.we take here pathways for marine bioinvasion are discharged water a large-scale perspective on the global cargo ship from ships'ballast tanks (Ruiz et al.2000)and hull network (GCSN)as a complex system defined as the fouling (Drake Lodge 2007).Even terrestrial species network of ports that are connected by links if ship traffic passes between them. such as insects are sometimes inadvertently transported in shipping containers(Lounibos 2002).In several parts Similar research in the past had to make strong of the world,invasive species have caused dramatic assumptions about flows on hypothetical networks levels of species extinction and landscape alteration, with connections between all pairs of ports in order to thus damaging ecosystems and creating hazards for approximate ship movements (Drake Lodge 2004; Tatem et al.2006).By contrast,our analysis is based on comprehensive data of real ship journeys allowing *Author for correspondence (blasiusaicbm.de). us to construct the actual network.We show that it Electronic supplementary material is available at http://dx.doi.org/ has a small-world topology where the combined cargo 10.1098/rsif.2009.0495 or via http://rsif.royalsocietypublishing.org. capacity of ships calling at a given port (measured in Received 11 November 2009 Accepted 21 December 2009 This journal is 2010 The Royal Society

The complex network of global cargo ship movements Pablo Kaluza, Andrea Ko¨lzsch, Michael T. Gastner and Bernd Blasius* Institute for Chemistry and Biology of the Marine Environment, Carl von Ossietzky Universita¨t, Carl-von-Ossietzky-Straße 9-11, 26111 Oldenburg, Germany Transportation networks play a crucial role in human mobility, the exchange of goods and the spread of invasive species. With 90 per cent of world trade carried by sea, the global network of merchant ships provides one of the most important modes of transportation. Here, we use information about the itineraries of 16 363 cargo ships during the year 2007 to construct a network of links between ports. We show that the network has several features that set it apart from other transportation networks. In particular, most ships can be classified into three categories: bulk dry carriers, container ships and oil tankers. These three categories do not only differ in the ships’ physical characteristics, but also in their mobility patterns and networks. Container ships follow regularly repeating paths whereas bulk dry carriers and oil tankers move less predictably between ports. The network of all ship movements possesses a heavy-tailed distribution for the connectivity of ports and for the loads transported on the links with systematic differences between ship types. The data analysed in this paper improve current assumptions based on gravity models of ship movements, an important step towards understanding patterns of global trade and bioinvasion. Keywords: complex network; cargo ships; bioinvasion; transportation 1. INTRODUCTION The ability to travel, trade commodities and share information around the world with unprecedented effi- ciency is a defining feature of the modern globalized economy. Among the different means of transport, ocean shipping stands out as the most energy efficient mode of long-distance transport for large quantities of goods (Rodrigue et al. 2006). According to estimates, as much as 90 per cent of world trade is hauled by ships (International Maritime Organization 2006). In 2006, 7.4 billion tons of goods were loaded at the world’s ports. The trade volume currently exceeds 30 trillion ton-miles and is growing at a rate faster than the global economy (United Nations Conference on Trade and Development 2007). The worldwide maritime network also plays a crucial role in today’s spread of invasive species. Two major pathways for marine bioinvasion are discharged water from ships’ ballast tanks (Ruiz et al. 2000) and hull fouling (Drake & Lodge 2007). Even terrestrial species such as insects are sometimes inadvertently transported in shipping containers (Lounibos 2002). In several parts of the world, invasive species have caused dramatic levels of species extinction and landscape alteration, thus damaging ecosystems and creating hazards for human livelihoods, health and local economies (Mack et al. 2000). The financial loss owing to bioinvasion is estimated to be $120 billion per year in the USA alone (Pimentel et al. 2005). Despite affecting everybody’s daily lives, the shipping industry is far less in the public eye than other sectors of the global transport infrastructure. Accordingly, it has also received little attention in the recent literature on complex networks (Wei et al. 2007; Hu & Zhu 2009). This neglect is surprising considering the current interest in networks (Albert & Baraba´si 2002; Newman 2003a; Gross & Blasius 2008), especially airport (Barrat et al. 2004; Guimera` & Amaral 2004; Hufnagel et al. 2004; Guimera` et al. 2005), road (Buhl et al. 2006; Barthe´lemy & Flammini 2008) and train networks (Latora & Marchiori 2002; Sen et al. 2003). In the spirit of current network research, we take here a large-scale perspective on the global cargo ship network (GCSN) as a complex system defined as the network of ports that are connected by links if ship traffic passes between them. Similar research in the past had to make strong assumptions about flows on hypothetical networks with connections between all pairs of ports in order to approximate ship movements (Drake & Lodge 2004; Tatem et al. 2006). By contrast, our analysis is based on comprehensive data of real ship journeys allowing us to construct the actual network. We show that it has a small-world topology where the combined cargo capacity of ships calling at a given port (measured in *Author for correspondence (blasius@icbm.de). Electronic supplementary material is available at http://dx.doi.org/ 10.1098/rsif.2009.0495 or via http://rsif.royalsocietypublishing.org. J. R. Soc. Interface doi:10.1098/rsif.2009.0495 Published online Received 11 November 2009 Accepted 21 December 2009 1 This journal is q 2010 The Royal Society Downloaded from rsif.royalsocietypublishing.org on November 19, 2012

Downloaded from rsif.royalsocietypublishing.org on November 19,2012 2 Complex network of ship movements P.Kaluza et al. a b >4 2 the 20 most central ports 1 Panama Canal 11 Santos 2 Suez Canal 12 Tianjin 3 Shanghai 13 New York and New Jersey 4 Singapore 14 Europoort 5 Antwerp 15 Hamburg 6 Piraeus 16 Le Havre 111 111 journeys 7 Terneuzen 17 St Petersburg 8 Plaquemines 18 Bremerhaven <10205010020050010002000>5000 9 Houston 19 Las Palmas 10 Ijmuiden 20 Barcelona Figure 1.Routes,ports and betweenness centralities in the GCSN.(a)The trajectories of all cargo ships bigger than 10000 GT during 2007.The colour scale indicates the number of journeys along each route.Ships are assumed to travel along the shortest (geodesic)paths on water.(b)A map of the 50 ports of highest betweenness centrality and a ranked list of the 20 most central ports. gross tonnage (GT))follows a heavy-tailed distribution. to the port authorities.This technology is primarily This capacity scales superlinearly with the number of used to avoid collisions and increase port security,but directly connected ports.We identify the most central arrival and departure records are also made available ports in the network and find several groups of highly by Lloyd's Register Fairplay for commercial purposes interconnected ports showing the importance of as part of its Sea-web database (www.sea-web.com). regional geopolitical and trading blocks. AIS devices have not been installed in all ships and A high-level description of the complete network. ports yet,and therefore there are some gaps in the however,does not yet fully capture the network's com- data.Still,all major ports and the largest ships are plexity. Unlike previously studied transportation included,thus the database represents the majority of networks,the GCSN has a multi-layered structure. cargo transported on ships. There are,broadly speaking,three classes of cargo Our study is based on Sea-web's arrival and ships-container ships,bulk dry carriers and oil departure records in the calendar year 2007 as well as tankers-that span distinct subnetworks.Ships in Sea-web's comprehensive data on the ships'physical different categories tend to call at different ports and characteristics.We restrict our study to cargo ships travel in distinct patterns.We analyse the trajectories bigger than 10000 GT that make up 93 per cent of of individual ships in the GCSN and develop techniques the world's total capacity for cargo ship transport. to extract quantitative information about characteristic From these,we select all 16 363 ships for which AIS movement types.With these methods,we can quantify data are available,taken as representative of the that container ships sail along more predictable,fre global traffic and long distance trade between the 951 quently repeating routes than oil tankers or bulk dry ports equipped with AIS receivers (for details see elec- carriers.We compare the empirical data with theoreti- tronic supplementary material).For each ship,we cal traffic fows calculated by the gravity model obtain a trajectory from the database,i.e.a list of Simulation results,based on the full GCSN data or ports visited by the ship sorted by date.In 2007, the gravity model,differ significantly in a population- there were 490 517 non-stop journeys linking 36 351 dis- dynamic model for the spread of invasive species tinct pairs of arrival and departure ports.The complete between the world's ports.Predictions based on the set of trajectories,each path representing the shortest real network are thus more informative for international route at sea and coloured by the number of journeys policy decisions concerning the stability of worldwide passing through it,is shown in figure 1a. trade and for reducing the risks of bioinvasion Each trajectory can be interpreted as a small directed network where the nodes are ports linked together if the ship travelled directly between the 2.DATA ports.Larger networks can be defined by merging tra- jectories of different ships.In this article,we An analysis of global ship movements requires detailed aggregate trajectories in four different ways:the com- knowledge of ships'arrival and departure times at their bined network of all available trajectories and the ports of call.Such data have become available in recent subnetworks of container ships (3100 ships),bulk dry years.Starting in 2001,ships and ports have begun carriers(5498)and oil tankers(2628).These three sub- installing the automatic identification system (AIS) networks combinedly cover 74 per cent of the GCSN's equipment.AIS transmitters on board of the ships total GT.In all four networks,we assign a weight w automatically report the arrival and departure times to the link from port i to j equal to the sum of the J.R.Soc.Interface

gross tonnage (GT)) follows a heavy-tailed distribution. This capacity scales superlinearly with the number of directly connected ports. We identify the most central ports in the network and find several groups of highly interconnected ports showing the importance of regional geopolitical and trading blocks. A high-level description of the complete network, however, does not yet fully capture the network’s complexity. Unlike previously studied transportation networks, the GCSN has a multi-layered structure. There are, broadly speaking, three classes of cargo ships—container ships, bulk dry carriers and oil tankers—that span distinct subnetworks. Ships in different categories tend to call at different ports and travel in distinct patterns. We analyse the trajectories of individual ships in the GCSN and develop techniques to extract quantitative information about characteristic movement types. With these methods, we can quantify that container ships sail along more predictable, frequently repeating routes than oil tankers or bulk dry carriers. We compare the empirical data with theoretical traffic flows calculated by the gravity model. Simulation results, based on the full GCSN data or the gravity model, differ significantly in a populationdynamic model for the spread of invasive species between the world’s ports. Predictions based on the real network are thus more informative for international policy decisions concerning the stability of worldwide trade and for reducing the risks of bioinvasion. 2. DATA An analysis of global ship movements requires detailed knowledge of ships’ arrival and departure times at their ports of call. Such data have become available in recent years. Starting in 2001, ships and ports have begun installing the automatic identification system (AIS) equipment. AIS transmitters on board of the ships automatically report the arrival and departure times to the port authorities. This technology is primarily used to avoid collisions and increase port security, but arrival and departure records are also made available by Lloyd’s Register Fairplay for commercial purposes as part of its Sea-web database (www.sea-web.com). AIS devices have not been installed in all ships and ports yet, and therefore there are some gaps in the data. Still, all major ports and the largest ships are included, thus the database represents the majority of cargo transported on ships. Our study is based on Sea-web’s arrival and departure records in the calendar year 2007 as well as Sea-web’s comprehensive data on the ships’ physical characteristics. We restrict our study to cargo ships bigger than 10 000 GT that make up 93 per cent of the world’s total capacity for cargo ship transport. From these, we select all 16 363 ships for which AIS data are available, taken as representative of the global traffic and long distance trade between the 951 ports equipped with AIS receivers (for details see electronic supplementary material). For each ship, we obtain a trajectory from the database, i.e. a list of ports visited by the ship sorted by date. In 2007, there were 490 517 non-stop journeys linking 36 351 distinct pairs of arrival and departure ports. The complete set of trajectories, each path representing the shortest route at sea and coloured by the number of journeys passing through it, is shown in figure 1a. Each trajectory can be interpreted as a small directed network where the nodes are ports linked together if the ship travelled directly between the ports. Larger networks can be defined by merging trajectories of different ships. In this article, we aggregate trajectories in four different ways: the combined network of all available trajectories and the subnetworks of container ships (3100 ships), bulk dry carriers (5498) and oil tankers (2628). These three subnetworks combinedly cover 74 per cent of the GCSN’s total GT. In all four networks, we assign a weight wij to the link from port i to j equal to the sum of the <10 20 50 100 200 500 1000 2000 >5000 journeys 0 1 2 3 >4 betweenness (×104) the 20 most central ports 1 Panama Canal 2 Suez Canal 3 Shanghai 4 Singapore 5 Antwerp 6 Piraeus 7 Terneuzen 8 Plaquemines 9 Houston 10 Ijmuiden 11 Santos 12 Tianjin 13 New York and New Jersey 14 Europoort 15 Hamburg 16 Le Havre 17 St Petersburg 18 Bremerhaven 19 Las Palmas 20 Barcelona (a) (b) Figure 1. Routes, ports and betweenness centralities in the GCSN. (a) The trajectories of all cargo ships bigger than 10 000 GT during 2007. The colour scale indicates the number of journeys along each route. Ships are assumed to travel along the shortest (geodesic) paths on water. (b) A map of the 50 ports of highest betweenness centrality and a ranked list of the 20 most central ports. 2 Complex network of ship movements P. Kaluza et al. J. R. Soc. Interface Downloaded from rsif.royalsocietypublishing.org on November 19, 2012

Downloaded from rsif.royalsocietypublishing.org on November 19,2012 Compler network of ship movements P.Kaluza et al.3 available space on all ships that have travelled on the The degree distribution P()shows that most ports link during 2007 measured in GT.If a ship made a jour- have few connections,but there are some ports linked ney from i to j more than once,its capacity contributes to hundreds of other ports (figure 2a).Similar right- multiple times to wir skewed degree distributions have been observed in many real-world networks (Barabasi Albert 1999). While the GCSN's degree distribution is not exactly 3.THE GLOBAL CARGO SHIP NETWORK scale-free,the distribution of link weights,P(w),fol- lows approximately a power law P(w)oc w with The directed network of the entire cargo fleet is notice- u=1.71+0.14 (95%CI for linear regression, ably asymmetric,with 59 per cent of all linked pairs of figure 26,see also electronic supplementary material). ports being connected only in one direction.Still.the By averaging the sums of the link weights arriving vast majority of ports (935 out of 951)belongs to one at and departing from port i,we obtain the node single strongly connected component,i.e.for any two strength si(Barrat et al.2004).The strength distri- ports in this component,there are routes in both direc- bution can also be approximated by a power law tions,though possibly visiting different intermediate P(s)cs”with)=1.02±0.17，meaning that a ports.The routes are intriguingly short:only few steps small number of ports handle huge amounts of cargo in the network are needed to get from one port to (figure 2c).The determination of power law relation- another.The shortest path length l between two ports ships by line fitting has been strongly criticized (e.g. is the minimum number of non-stop connections one Newman 2005;Clauset et al.2009),therefore,we must take to travel between origin and destination.In analysed the distributions with model selection by the GCSN,the average over all pairs of ports is extre- Akaike weights (Burnham Anderson 1998).Our mely small,(=2.5.Even the maximum shortest results confirm that a power law is a better fit than path between any two ports (e.g.from Skagway, an exponential or a lognormal distribution for P(w) Alaska,to the small Italian island of Lampedusa)is and P(s),but not P(k)(see electronic supplementary only of length ax=8.In fact,the majority of all poss- material).These findings agree well with the concept ible origin-destination pairs (52%)can already be of hubs-spokes networks (Notteboom 2004)that connected by two steps or less. were proposed for cargo traffic,for example,in Asia Comparing these findings to those reported for the (Robinson 1998).There are a few large,highly con- worldwide airport network (WAN)shows interesting nected ports through which all smaller ports transact differences and similarities.The high asymmetry of their trade.This scale-free property makes the ship the GCSN has not been found in the WAN,indicating trade network prone to the spreading and persistence that ship traffic is structurally very different from avia- of bioinvasive organisms (e.g.Pastor-Satorras tion.Rather than being formed by the accumulation of Vespignani 2001).The average nearest-neighbour back-and-forth trips,ship traffic seems to be governed degree,a measure of network assortativity,addition- by an optimal arrangement of unidirectional,often cir- ally underlines the hubs-spokes property of cargo cular routes.This optimality also shows in the GCSN's ship traffic (see electronic supplementary material). small shortest path lengths.In comparison,in the Strengths and degrees of the ports are related accord- WAN,the average and maximum shortest path lengths ing to the scaling relation (s()6+01(95%CI for are (=4.4 and ax=15,respectively (Guimera et al. standardized major axis regression;Warton et al.2006). 2005),i.e.about twice as long as in the GCSN.Similar Hence,the strength of a port grows generally faster to the WAN,the GCSN is highly clustered:if a port Xis than its degree(figure 2d).In other words,highly con- linked to ports Yand Z,there is a high probability that nected ports not only have many links,but their links there is also a connection from Yto 2.We calculated a also have a higher than average weight.This obser- clustering coefficient C (Watts Strogatz 1998)for vation agrees with the fact that busy ports are better directed networks and found C=0.49,whereas equipped to handle large ships with large amounts of random networks with the same number of nodes and cargo.A similar result,(s())k501,was found for links only yield C=0.04 on average.Degree-dependent airports(Barrat et al.2004),which may hint at a gen- clustering coefficients C reveal that clustering eral pattern in transportation networks.In the light of decreases with node degree(see electronic supplemen- bioinvasion,these results underline empirical findings tary material).Therefore,the GCSN-like the that big ports are more heavily invaded because of WAN-can be regarded as a small-world network pos- increased propagule pressure by ballast water of more sessing short path lengths despite substantial and larger ships (Williamson 1996;Mack et al.2000; clustering (Watts Strogatz 1998).However,the aver- see Cohen Carlton 1998). age degree of the GCSN,i.e.the average number of links A further indication of the importance of a node is its arriving at and departing from a given port (in-degree betweenness centrality (Freeman 1979;Newman 2004) plus out-degree),(=76.5,is notably higher than in The betweenness of a port is the number of topologi- the WAN,where ()=19.4 (Barrat et al.2004).In cally shortest directed paths in the network that pass the light of the network size (the WAN consists of through this port.In figure 16,we plot and list the 3880 nodes),this difference becomes even more pro- most central ports.Generally speaking,centrality and nounced,indicating that the GCSN is much more degree are strongly correlated (Pearson's correlation densely connected.This redundancy of links gives the coefficient:0.81).but in individual cases other factors network high structural robustness to the loss of can also play a role.The Panama and Suez canals,for routes for keeping up trade instance,are shortcuts to avoid long passages around J.R.Soc.Interface

available space on all ships that have travelled on the link during 2007 measured in GT. If a ship made a journey from i to j more than once, its capacity contributes multiple times to wij. 3. THE GLOBAL CARGO SHIP NETWORK The directed network of the entire cargo fleet is noticeably asymmetric, with 59 per cent of all linked pairs of ports being connected only in one direction. Still, the vast majority of ports (935 out of 951) belongs to one single strongly connected component, i.e. for any two ports in this component, there are routes in both directions, though possibly visiting different intermediate ports. The routes are intriguingly short: only few steps in the network are needed to get from one port to another. The shortest path length l between two ports is the minimum number of non-stop connections one must take to travel between origin and destination. In the GCSN, the average over all pairs of ports is extremely small, kll ¼ 2:5. Even the maximum shortest path between any two ports (e.g. from Skagway, Alaska, to the small Italian island of Lampedusa) is only of length lmax ¼ 8. In fact, the majority of all possible origin –destination pairs (52%) can already be connected by two steps or less. Comparing these findings to those reported for the worldwide airport network (WAN) shows interesting differences and similarities. The high asymmetry of the GCSN has not been found in the WAN, indicating that ship traffic is structurally very different from aviation. Rather than being formed by the accumulation of back-and-forth trips, ship traffic seems to be governed by an optimal arrangement of unidirectional, often circular routes. This optimality also shows in the GCSN’s small shortest path lengths. In comparison, in the WAN, the average and maximum shortest path lengths are kll ¼ 4:4 and lmax ¼ 15, respectively (Guimera` et al. 2005), i.e. about twice as long as in the GCSN. Similar to the WAN, the GCSN is highly clustered: if a port X is linked to ports Y and Z, there is a high probability that there is also a connection from Y to Z. We calculated a clustering coefficient C (Watts & Strogatz 1998) for directed networks and found C ¼ 0.49, whereas random networks with the same number of nodes and links only yield C ¼ 0.04 on average. Degree-dependent clustering coefficients Ck reveal that clustering decreases with node degree (see electronic supplementary material). Therefore, the GCSN—like the WAN—can be regarded as a small-world network possessing short path lengths despite substantial clustering (Watts & Strogatz 1998). However, the average degree of the GCSN, i.e. the average number of links arriving at and departing from a given port (in-degree plus out-degree), kkl ¼ 76.5, is notably higher than in the WAN, where kkl ¼ 19.4 (Barrat et al. 2004). In the light of the network size (the WAN consists of 3880 nodes), this difference becomes even more pronounced, indicating that the GCSN is much more densely connected. This redundancy of links gives the network high structural robustness to the loss of routes for keeping up trade. The degree distribution P(k) shows that most ports have few connections, but there are some ports linked to hundreds of other ports (figure 2a). Similar rightskewed degree distributions have been observed in many real-world networks (Baraba´si & Albert 1999). While the GCSN’s degree distribution is not exactly scale-free, the distribution of link weights, P(w), follows approximately a power law PðwÞ / wm with m ¼ 1:71 + 0:14 (95% CI for linear regression, figure 2b, see also electronic supplementary material). By averaging the sums of the link weights arriving at and departing from port i, we obtain the node strength si (Barrat et al. 2004). The strength distribution can also be approximated by a power law PðsÞ / sh with h ¼ 1:02 + 0:17, meaning that a small number of ports handle huge amounts of cargo (figure 2c). The determination of power law relationships by line fitting has been strongly criticized (e.g. Newman 2005; Clauset et al. 2009), therefore, we analysed the distributions with model selection by Akaike weights (Burnham & Anderson 1998). Our results confirm that a power law is a better fit than an exponential or a lognormal distribution for P(w) and P(s), but not P(k) (see electronic supplementary material). These findings agree well with the concept of hubs – spokes networks (Notteboom 2004) that were proposed for cargo traffic, for example, in Asia (Robinson 1998). There are a few large, highly connected ports through which all smaller ports transact their trade. This scale-free property makes the ship trade network prone to the spreading and persistence of bioinvasive organisms (e.g. Pastor-Satorras & Vespignani 2001). The average nearest-neighbour degree, a measure of network assortativity, additionally underlines the hubs –spokes property of cargo ship traffic (see electronic supplementary material). Strengths and degrees of the ports are related according to the scaling relation ksðkÞl / k1:46+0:1 (95% CI for standardized major axis regression; Warton et al. 2006). Hence, the strength of a port grows generally faster than its degree (figure 2d). In other words, highly connected ports not only have many links, but their links also have a higher than average weight. This observation agrees with the fact that busy ports are better equipped to handle large ships with large amounts of cargo. A similar result, ksðkÞl / k1:5+0:1, was found for airports (Barrat et al. 2004), which may hint at a general pattern in transportation networks. In the light of bioinvasion, these results underline empirical findings that big ports are more heavily invaded because of increased propagule pressure by ballast water of more and larger ships (Williamson 1996; Mack et al. 2000; see Cohen & Carlton 1998). A further indication of the importance of a node is its betweenness centrality (Freeman 1979; Newman 2004). The betweenness of a port is the number of topologically shortest directed paths in the network that pass through this port. In figure 1b, we plot and list the most central ports. Generally speaking, centrality and degree are strongly correlated (Pearson’s correlation coefficient: 0.81), but in individual cases other factors can also play a role. The Panama and Suez canals, for instance, are shortcuts to avoid long passages around Complex network of ship movements P. Kaluza et al. 3 J. R. Soc. Interface Downloaded from rsif.royalsocietypublishing.org on November 19, 2012

Downloaded from rsif.royalsocietypublishing.org on November 19,2012 4 Complex network of ship movements P.Kaluza et al. (a)101 (b) 10-3 102 10-6 10-7 米 103 10r1 (M)d 10-8 10r2 109 10- 104 1010 10- 4 10 103 1011 10310510610 10 102 103 104 105 105 107 108 degree,k link weight.w (c) 105 (d) 米 106 108 107 107 80 米 108 02 10 106 10 109 10—8 10-9 105 10-10 1010 104105105107108 104 104 105 106 107 108 1 10 100 1000 node strength,s degree,k Figure 2.Degrees and weights in the GCSN.(a)The degree distributions P(k)are right-skewed,but not power laws,neither for the GCSN nor its subnetworks.The degree k is defined here as the sum of in-and out-degree,thus k=1 is rather rare (asterisk,all ship types;square,container ships;circle,bulk dry carriers;triangle,oil tankers).()The link weight distributions P(w)reveal clear power law relationships for the GCSN and the three subnetworks,with exponents u characteristic for the movement pat- terns of the different ship types(asterisk,.u=1.71±0.14；square,.μ=1.42±0.l5:circle,μ=1.93±0.1l;triangle,.μ=1.73± 0.25).(c)The node strength distributions P(s)are also heavy tailed,showing power law relationships.The stated exponents are calculated by linear regression with 95%confidence intervals(similar results are obtained with maximum likelihood estimates,see electronic supplementary material)(asterisk,n=1.02+0.17;square,n=1.05+0.13;circle,n=1.13+0.21;triangle,n= 1.010.16).(d)The average strength of a node(s()scales superlinearly with its degree,(s(,indicating that highly connected ports have,on average,links of higher weight. South America and Africa.Other ports have a high cen- has a higher mean degree and fewer journeys per link trality because they are visited by a large number of (C=0.43,()=44.61.(J)=4.65).For the oil tankers ships (e.g.Shanghai),whereas others gain their status we find intermediate values (C=0.44,()=33.32. primarily by being connected to many different ports (J)=5.07).Note that the mean degrees (of the sub- (e.g.Antwerp). networks are substantially smaller than that of the full GCSN,indicating that different ship types use essentially the same ports but different connections. 4.THE NETWORK LAYERS OF DIFFERENT A similar tendency appears in the scaling of the link SHIP TYPES weight distributions (figure 26).P(w)can be approxi- mated as power laws for each network,but with To compare the movements of cargo ships of different different exponents u.The container ships have the types,separate networks were generated for each of smallest exponent (u=1.42)and bulk dry carriers the the three main ship types:container ships,bulk dry car- largest (u=1.93)with oil tankers in between (u= riers and oil tankers.Applying the network parameters 1.73).In contrast,the exponents for the distribution of introduced in the previous section to these three subnet- node strength P(s)are nearly identical in all three sub- works reveals some broad-scale differences (table 1) networks,n=1.05,n=1.13 and n=1.01,respectively. The network of container ships is densely clustered These numbers give a first indication that different C=0.52,has a rather low mean degree,()=32.44, ship types move in distinctive patterns.Container and a large mean number of journeys (i.e.number of ships typically follow set schedules visiting several times any ship passes)per link,()=24.26.The bulk ports in a fixed sequence along their way,thus provid- dry carrier network,on the other hand,is less clustered, ing regular services.Bulk dry carriers,by contrast, J.R.Soc.Interface

South America and Africa. Other ports have a high centrality because they are visited by a large number of ships (e.g. Shanghai), whereas others gain their status primarily by being connected to many different ports (e.g. Antwerp). 4. THE NETWORK LAYERS OF DIFFERENT SHIP TYPES To compare the movements of cargo ships of different types, separate networks were generated for each of the three main ship types: container ships, bulk dry carriers and oil tankers. Applying the network parameters introduced in the previous section to these three subnetworks reveals some broad-scale differences (table 1). The network of container ships is densely clustered, C ¼ 0.52, has a rather low mean degree, kkl ¼ 32:44, and a large mean number of journeys (i.e. number of times any ship passes) per link, kJl ¼ 24:26. The bulk dry carrier network, on the other hand, is less clustered, has a higher mean degree and fewer journeys per link (C ¼ 0:43, kkl ¼ 44:61, kJl ¼ 4:65). For the oil tankers, we find intermediate values (C ¼ 0:44, kkl ¼ 33:32, kJl ¼ 5:07). Note that the mean degrees kkl of the subnetworks are substantially smaller than that of the full GCSN, indicating that different ship types use essentially the same ports but different connections. A similar tendency appears in the scaling of the link weight distributions (figure 2b). P(w) can be approximated as power laws for each network, but with different exponents m. The container ships have the smallest exponent (m ¼ 1.42) and bulk dry carriers the largest (m ¼ 1.93) with oil tankers in between (m ¼ 1.73). In contrast, the exponents for the distribution of node strength P(s) are nearly identical in all three subnetworks, h ¼ 1.05, h ¼ 1.13 and h ¼ 1.01, respectively. These numbers give a first indication that different ship types move in distinctive patterns. Container ships typically follow set schedules visiting several ports in a fixed sequence along their way, thus providing regular services. Bulk dry carriers, by contrast, 10–1 (a) (b) (c) (d) 10–2 10–3 P(k) 10–4 104 105 106 107 108 10–4 10–3 10–2 10–1 1 10 102 degree, k link weight, w 102 104 10–4 10–5 10–6 10–7 10–8 10–9 10–10 10–11 10–5 10–6 10–7 10–8 10–9 10–10 105 106 107 104 105 106 107 108 1 10 103 107 106 10 104 5 1 100 1000 10 node strength, s degree, k 108 104 10–11 10–10 10–9 P(w) 10–8 10–7 10–6 10–5 10–10 10–9 10 P –8 (s) 10–7 10–6 10–5 105 106 107 108 s(k) Figure 2. Degrees and weights in the GCSN. (a) The degree distributions P(k) are right-skewed, but not power laws, neither for the GCSN nor its subnetworks. The degree k is defined here as the sum of in- and out-degree, thus k ¼ 1 is rather rare (asterisk, all ship types; square, container ships; circle, bulk dry carriers; triangle, oil tankers). (b) The link weight distributions P(w) reveal clear power law relationships for the GCSN and the three subnetworks, with exponents m characteristic for the movement patterns of the different ship types (asterisk, m ¼ 1.71+0.14; square, m ¼ 1.42+0.15; circle, m ¼ 1.93+0.11; triangle, m ¼ 1.73+ 0.25). (c) The node strength distributions P(s) are also heavy tailed, showing power law relationships. The stated exponents are calculated by linear regression with 95% confidence intervals (similar results are obtained with maximum likelihood estimates, see electronic supplementary material) (asterisk, h ¼ 1.02+0.17; square, h ¼ 1.05+0.13; circle, h ¼ 1.13+0.21; triangle, h ¼ 1.01+0.16). (d) The average strength of a node ks(k)l scales superlinearly with its degree, ks(k)l / k1.46+0.1, indicating that highly connected ports have, on average, links of higher weight. 4 Complex network of ship movements P. Kaluza et al. J. R. Soc. Interface Downloaded from rsif.royalsocietypublishing.org on November 19, 2012

点击进入文档下载页（PDF格式）

共12页，试读已结束，阅读完整版请下载

您可能感兴趣的文档

点击购买下载（PDF）

下载及服务说明

购买前请先查看本文档预览页，确认内容后再进行支付；
如遇文件无法下载、无法访问或其它任何问题，可发送电子邮件反馈，核实后将进行文件补发或退款等其它相关操作；
邮箱：

文档浏览记录