Temporal accessibility in archipelagos: inter-island shipping in the Canary Islands

Journal oftransport Geography 10 (2002) 231 239 Viewpoint Temporal accessibility in archipelagos: inter-island shipping in the Canary Islands Jose Angel Hernandez Luis * www.elsevier.com/locate/jtrangeo Departamento de Geografıa, Universidad de Las Palmas de Gran Canaria, Gran Canaria, 35003 Canary Islands, Spain Abstract The aim ofthis paper is to analyse the current conditions oftemporal daily accessibility among the seven Canary Islands by means ofsea transport for both passengers and vehicles. To do this a method is proposed that can be applied to other areas sharing similar characteristics ofterritorial remoteness and fragmentation. Time available daily at the destination is also studied in relation to administrative or commercial working hours. Ó 2002 Published by Elsevier Science Ltd. Keywords: Temporal daily accessibility; Shipping lane; Land corridor; Intermodality 1. Introduction For financial reasons and because ofthe lack ofa hinterland with sufficient population, it has not been possible to provide a wide range ofservices for the local population on each ofthe Canary Islands. Such services include, for example, a large hospital offering multiple specialities and services, or a large shopping centre with a great variety ofshops. Transport toward those islands that do possess such essential services therefore becomes critical. The barrier formed by the sea prevents the use of private vehicles except via ferries and inter-island transport is restricted to ships or aeroplanes. Inter-island public transport faces similar challenges to those on the mainland but with the major difference oflack ofcompetition from private car and lorry transport. However the planning ofthe number oftrips, timetables, availability ofseats or cargo capacity are the main problems faced by sea and air transport. The greater the number oftrips and ship and air transport capacity, the greater accessibility will be between the nodes on the different islands. Even so this will also depend on timetable planning that may or not facilitate intermodality with land transport together with an adequate availability oftime at a destination. * Tel.: +34-928-45-2789; fax: +34-928-45-2775. E-mail address: jahdez@cicei.ulpgc.es (J. A. Hernandez Luis). Because ofthe high cost ofair travel, short-sea shipping is the principal mode for achieving social and financial integration, both between the islands and with the outside world. Perhaps the most important factors currently creating more travel between the islands are the progressive specialisation ofthe space economy and its effect on increased mobility, together with the booming tourist industry and the rising income per capita. An additional factor is the appearance of new ship owners due to the complete liberalisation ofshortsea transport in European Union (EU) waters since the beginning of1999. 1 This has resulted in an increase in daily trips, an improvement in timetables and reduced fares. The consequence is that whereas in 1994 only two shipping companies were operating a fleet offerries capable oftransporting both vehicles and passengers, in 1999 there were three. One ofthe two existing companies has also improved its services by adding three more routes. In 1998 there were nearly 3.7 million inter-island trips by sea, or 2.24 trips per inhabitant, as well as 1.9 million inter-island air transport trips. Inter-island mobility still does not provide relative territorial continuity because ofthe slowness ofthe vessels as well as the lack oftimetable synchronisation to provide optimum intermodality amongst three or 1 CEE Ruling nr. 3577/92 ofthe Council, 07/12/92, article 6, point 2 which led to the implementation ofcabotage for intra-eu shipping services. Since responsibility was transferred to the Canary Islands in shipping transport matters, in accordance with Royal Decree of24/02/ 95, the entry into service ofnew shipping lines has accelerated. 0966-6923/02/$ - see front matter Ó 2002 Published by Elsevier Science Ltd. PII: S0966-6923(02)00014-5

232 J.Á. Hernandez Luis / Journal of Transport Geography 10 (2002) 231 239 more islands. Therefore it may be construed that there is still a significant field ofpotential demand that sea transport does not satisfy because of the slow operating speed at sea and incomplete intermodality. As in other archipelagos in developed countries, the tendency over the last few years in the Canary Islands has been to de-centralise shipping routes that have traditionally been concentrated on each island s main, capital, city in favour of links through those regional ports that are nearest to the adjoining islands (Knowles, 1996). This allows improved temporal accessibility since the average speed by road is greater than the 20 knots or less ofa conventional ship. Accessibility is undoubtedly improved although the pressure on the roads is increased, especially taking into account that one ship can carry nearly 250 cars (or 100 cars and 40 trailers). This aggravates the traffic problem on land and particularly at certain times ofthe day, for example, in the morning, when coinciding with the rush hour for internal traffic movements. 2. The regional context of the Canary Islands The Canary Islands is a group ofseven main or large islands and seven small islands. All the large islands are inhabited but only one ofthe small ones and that has just 600 inhabitants (Fig. 1). Another fundamental characteristic is that all the islands with the exception ofthe two central islands Tenerife and Gran Canaria are chronically deficient in services because oftheir small population. This leads to greater demand for travel between these islands and the two central ones. At the same time, travel for employment is of great significance which leads to a very high incidence ofday trips. The availability ofsufficient time at the destination is a key factor for the functional integration of the islands. Since the 1960s, growth in the tertiary economy on the Canary Islands has been progressively linked with tourism and this has contributed to an increase in the standard ofliving ofthe inhabitants. This also led to an increase in inter-island mobility, both for work and leisure. In the same way the tourists have also demanded more excursions to and from adjacent islands on day trips. The population ofthe seven main islands is quite uneven ranging from 714 000 on Gran Canaria and 666 000 on Tenerife to just 8000 on the island of Hierro (Table 1). At the same time there is a very high concentration ofpopulation in each ofthe islands capital cities reaching 53% in the case ofgran Canaria. This fact will be taken into account when evaluating growing interest in the use ofregional ports as against ports in the capital cities. The number ofvisitors to the islands has increased continuously since the 1960s and annually over 10 million people make this their holiday destination including Spanish tourists from the mainland. This has had an influence on the demand for maritime transport. As an example, there is a (return) flow of1.3 million passengers between the islands oftenerife and La Gomera, despite the latter s population being less than 20 000 (see Table 1). Gross added value tends to be concentrated in the capital cities especially on the central islands oftenerife and Gran Canaria where high density ofpopulation, economic specialisation and central place and administration functions have generated gross added value four percent higher than population share (Table 1). This confirms the important role played by island capital cities as points ofdeparture and destination. At the same time the majority ofgovernment functions, national health service facilities and commerce are to be found in these capital cities. This explains why the population has a higher average income in these capital cities in comparison with the rest ofthe territory and consequently greater potential for mobility (Hernandez, 1996). 3. Methodology Accessibility is a key factor in regional development policy and this is well documented (Brookfield, 1980; Dunbar, 1981; Biehl et al., 1986; Rutz and Coull, 1996; Hoyle, 1999). However, it is often not the determining factor for promoting a region since development also depends on a large range ofparameters including physical conditioning factors, population size and characteristics and state investment policy. There are two main types ofaccessibility: positional and personal (Gutierrez, 1988). Temporal accessibility falls into the first category and may also be translated into financial accessibility, in the sense that time spent travelling may be considered as not productive in working hours. In this way, and through surveys on salaries for different economic activities, as well as the transport companies fares, it is possible to consider financial accessibility between two or more nodes ofone network 2 (Hernandez, 1996). Temporal accessibility has been researched extensively, for example, in Hagerstrand s (1970) work on time geography, as well as by Nutley (1983), Rodrıguez (1989), and Lopez (1990). These, among others, refer 2 Included in the Libro Blanco de los Transportes en Canarias [Canary Islands White Book on Transport] (Gobierno de Canarias, 1998) are some isochrone maps for shipping and air transport, from Tenerife and Gran Canaria to the rest of the archipelago. There are also some isocost maps that only take into account the costs ofbuses and taxis plus the air or maritime fares leaving out the costs related to time consumption in working hours. These are, at the beginning ofthe day, quite considerable since the main users are people in the administration and private enterprise.

J.Á. Hernandez Luis / Journal of Transport Geography 10 (2002) 231 239 233 Fig. 1. Canary Islands: Population and Gross Added Value (1996).

234 J.Á. Hernandez Luis / Journal of Transport Geography 10 (2002) 231 239 Table 1 Population and economic characteristics ofthe Canary Islands Islands Population (1996) % ofpopulation in capital city Arrival oftourists (1997) Annual income, Euros per capita (1991) Gran Canaria 713 768 53.27 2 733 978 6461 57.47 Tenerife 665 611 32.09 3 157 343 6870 36.96 La Palma 81 507 21.82 117 007 5703 33.32 Lanzarote 77 379 52.23 1 466 570 7950 42.21 Fuerteventura 42 938 44.67 958 975 8027 37.69 La Gomera 17 008 35.12 69 754 6127 49.31 El Hierro 8338 49.57 8709 6487 52.67 Canary Islands 1 606 549 8 512 336 6692 Source: Instituto Canario de Estadıstica. a Gross added value. % ofgav a generated in the capital city (1991) directly or indirectly to the problems oftemporal accessibility by public transport, particularly in mainland areas. However, mainland inhabitants can always fall back on the alternative ofprivate transport when the public service is deficient or does not supply their needs. In an archipelago there is no private road transport alternative for reaching another island and the planning ofthe timetable, availability ofseats and the possibility ofbeing able to travel at a certain time are ofvital importance (see Lutter et al., 1992; Hernandez, 1996; or Cozar et al., 1998). This paper concentrates on temporal accessibility in relation to inter-island modes oftransport because ofits vital importance for the Canary Islands. 3 The capital cities ofthe islands contain the greatest concentrations ofpopulation, in Gran Canaria and Tenerife accounting for 53% and 32%, respectively. At the same time it is also necessary to identify the economic strength of these cities, 4 in terms ofthe per capita income ofthe inhabitants as this will influence who undertakes the most inter-island travel for different reasons (higher cultural level or more qualified employment). 5 Due to these factors, each of the seven Canary Islands capital cities has been taken as the most frequent points of departure and destination for inter-island shipping trips. 3 According to Murillo et al. (1995), 45% ofround trips between Tenerife and Gran Canaria are carried out in only 1 day, whereas with the other islands they decrease to 23.2%. Ifwe add to this the round trips of 2 or 3 days, the percentages rise to 94.6 and 87.4, respectively. These percentages are extremely high, reflecting the incidence of temporal accessibility within insular modes oftransport. This is particularly so when the trips are repetitive. Therefore, according to MECSA (1993), the frequency of the trip between Tenerife and Gran Canaria by Jet-foil is 7% once a week by 7% rising to 53% when the frequency is between once a week and once a month. 4 According to Caixa (1999), 59.4% and 37.5% ofthe financial activity is concentrated in the capital cities ofgran Canaria and Tenerife, respectively, in spite of the important tourist activity in the south ofboth islands. 5 According to the Canary Island Government (Gobierno de Canarias, 1995), Gran Canaria and Tenerife surpass the average income for their islands by 5% and 8%, respectively. The total time spent in travelling door to door from origin to destination was calculated including the time spent in overland travel, that spent checking-in and boarding at the port, the actual time spent at sea and that in disembarking and collecting luggage. It is also necessary to calculate the time available to the traveller for undertaking the activities at the destination that are the primary reason for their trip. If the travellers do not have enough time to complete their business in one day, assuming that they have travelled on the first available ferry of the day, the cost of the trip will increase considerably. They will either have to add the cost ofan overnight stay the hotel bill plus meals 6 or use air transport ifit is available. Similar studies ofdaily accessibility have been carried out by, for example, Lutter et al. (1992). Available time is defined as the time between the arrival at the destination (such as the office in a particular street or a private house), and the latest time to commence the return trip to the ferry terminal for check-in and boarding. In any case, the more frequent trips are usually subject to government and commercial hours ofwork. Such hours vary between countries, but in the Canary Islands they are: government offices (08:00 15:00 h) commercial enterprises (09:00 13:00 and 16:00 20:00 h). If, for example, the reason for the trip is government business and the traveller arrives at their final destination at 09:30 h and has to return no later than 17:00 h, they will lose 90 min out ofthe normal working day. 4. Evolution of the accessibility times by sea in the Canary Islands The analysis assumes that departure and destination points are located within an island s capital city. This 6 MECSA (1993) indicates that the reason for the trip work/ business between Tenerife and Gran Canaria by Jet-foil and by plane rose to 57% and 56%, respectively. Leisure was relegated to a mere 17% on average for both modes of transport.

J.Á. Hernandez Luis / Journal of Transport Geography 10 (2002) 231 239 235 Table 2 Temporal maritime accessibility between island capital cities in the Canary Islands, 30 June 1994 (in min) a Travelling to and from ports Check-in and boarding Time of sea trip Disembarking and luggage S/C de Tenerife Las Palmas de G.C. (Jet-foil) 20 30 80 10 140 S/C de Tenerife Las Palmas de G.C. (Ferry) 20 30 240 10 300 S/C de Tenerife S.S. de La Gomera (Hydro-foil direct) a 60 40 35 10 155 S/C de Tenerife S.S. de La Gomera (Ferry direct) a 60 40 80 10 190 S/C de Tenerife Valverde (Ferry direct) a 60 40 270 10 380 S/C de Tenerife S/C de La Palma (Ferry) 20 30 495 10 555 Las Palmas de G.C. Puerto del Rosario (Jet-foil direct) a 100 40 80 10 230 Las Palmas de G.C. Puerto del Rosario (Ferry) 20 30 480 10 540 Puerto del Rosario Arrecife (Ferry direct) a 70 40 35 10 155 Puerto del Rosario Arrecife (Ferry) 20 30 180 10 240 Mean 45.0 35.0 197.5 10.0 288.5 Source: Timetables ofshipping and bus companies. a (1) Only direct links have been considered with no stopovers whatsoever. (2) Temporal accessibility refers to passage. However, the accessibility of a trailer is similar to that shown here since the travelling to and from ports is similar to the times used by the buses. (3) Considered here is bus transport between Santa Cruz de Tenerife and Los Cristianos (or vice versa); Morro Jable and Puerto del Rosario (or vice versa); Corralejo and Puerto del Rosario (or vice versa); and Playa Blanca Arrecife (or vice versa). From there on the time left for the departure of the ship corresponds to checking-in and boarding times. Total allows seasonal comparisons or time series analyses over several years to be undertaken. Travelling times vary significantly due to changes in ship type, and improvement in ports or road networks. In this case study two key dates have been chosen: 1994 when only two shipping lines were operating in the Canary Islands with the capability ofcarrying passengers and motor vehicles, one ofwhich only operated two routes; and 1999 when there were three shipping lines. In 1994 there were 10 different ferry services on 7 routes among the islands, with 2 operators on 1 ofthem (Table 2). In 1999 there were 13 different ferry services on 10 routes with 2 operators working 7 ofthem (Fig. 2, and Table 3). Two ofthe new routes provided shorter and more direct sea journeys by ferry from Los Cristianos in Tenerife to Santa Cruz de La Palma and by fast ferry from Santa Cruz de Tenerife to Agaete in Gran Canaria. The third new service was a direct ferry from Las Palmas de Gran Canaria to Morro Jable on Fuerteventura. Whereas the time spent in checking-in and boarding and disembarking and luggage collection has not changed, the balance between time to travelling to and from ports and time at sea has changed. Since 1973, new shorter routes between adjacent local ports have in some cases replaced longer distance routes between the main, capital city, ports (Fig. 2). For example, the route Los Cristianos in southern Tenerife to La Gomera replaced the earlier route from Santa Cruz de Tenerife in northern Tenerife to La Gomera. El Hierro now has a direct ferry to Los Cristianos instead of being connected to Santa Cruz de Tenerife via La Gomera. The route from Corralejo in northern Fuerteventura to Playa Blanca in southern Lanzarote helps provide a much faster door-to-door journey time between the respective capitals ofpuerto del Rosario (Fuerteventura) and Arrecife (Lanzarote) than the allwater route. Similarly, the fast ferry route between Agaete on Gran Canaria and Santa Cruz de Tenerife provides a much shorter door-to-door journey between the respective capitals oflas Palmas de Gran Canaria and Santa Cruz de Tenerife than the all-water route. Minimising the sea journey, by using nearer local ports, and maximising the length ofland journey, reduces total travelling time and also enables ferry frequency to be increased. Contrary to this trend, a long distance ferry route has opened between the two islands ofgran Canaria and Lanzarote, with an average crossing time of 840 min (Table 3). Between 1994 and 1999, crossing times have been reduced on only two routes. The direct ferry from Los Cristianos in southern Tenerife to Valverde on El Hierro has cut the time at sea by 15% from 270 to 230 min. Consequently, the door-to-door journey between the two capitals ofsanta Cruz de Tenerife and Valverde dropped by 11% from 380 to 340 min. The introduction of a direct ferry from Los Cristianos to Santa Cruz de La Palma has cut the time at sea between Tenerife and La Palma by 46% from 495 to 230 min. As a result, the door-to-door time between the two capitals ofsanta Cruz de Tenerife and Santa Cruz de La Palma decreased by 61% from 555 to 340 min. Journey times for inter-island maritime transport are still high and still limit territorial integration in the archipelago. Table 4 shows a simulation oftemporal accessibility that could be achieved using fast ferries travelling at speeds of35 38 knots. These ships carry

236 J.Á. Hernandez Luis / Journal of Transport Geography 10 (2002) 231 239 Fig. 2. Canary Islands: Passenger Shipping Routes 1973 and 1999.

J.Á. Hernandez Luis / Journal of Transport Geography 10 (2002) 231 239 237 Table 3 Temporal maritime accessibility between island capital cities in the Canary Islands at 30 June 1999 (in min) a Travelling to and from ports Check-in and boarding Time of sea trip Disembarking and luggage S/C de Tenerife Las Palmas de G.C. (Jet-foil) 20 30 80 10 140 S/C de Tenerife Las Palmas de G.C. (Ferry) 20 30 240 10 300 S/C de Tenerife Las Palmas de G.C. (Fast-ferry direct) a 60 20 65 10 155 S/C de Tenerife S.S. de La Gomera (Hydro-foil direct) a 60 40 35 10 155 S/C de Tenerife S.S. de La Gomera (Ferry direct) a 60 40 80 10 190 S/C de Tenerife Valverde (Ferry direct) a 60 40 230 10 340 S/C de Tenerife S/C de La Palma (Ferry) 20 30 420 10 480 S/C de Tenerife S/C de La Palma (Ferry direct) a 60 40 230 10 340 Las Palmas de G.C. Puerto del Rosario (Ferry) 20 30 490 10 550 Las Palmas de G.C. Puerto del Rosario (Jet-foil direct) a 100 40 80 10 230 Las Palmas de G.C. Puerto del Rosario (Ferry direct) a 100 40 195 10 345 Las Palmas de G.C. Arrecife (Ferry) 20 30 840 10 900 Puerto del Rosario Arrecife (Ferry) 20 30 180 10 240 Puerto del Rosario Arrecife (Ferry direct) a 70 40 35 10 155 Mean 49.3 34.3 228.6 10.0 322.9 Source: Timetables ofshipping and bus companies. a (1) Only direct links have been considered with no stopovers whatsoever. (2) Temporal accessibility refers to passage. However, the accessibility of a trailer is similar to that shown here since the travelling to and from ports is similar to the times used by the buses. (3) Considered here is bus transport between Santa Cruz de Tenerife and Los Cristianos (or vice versa); Morro Jable and Puerto del Rosario (or vice versa); Corralejo and Puerto del Rosario (or vice versa); and Playa Blanca Arrecife (or vice versa). From there on the time left for the departure of the ship corresponds to checking-in and boarding times. Total both passengers and vehicles (cars and trailers) to and from the ports on each island that are nearest to each other. Such high speed services are planned by the government and the shipping companies themselves. 7 In 1999 only one fast ferry was operating between the two central islands oftenerife and Gran Canaria with a travelling time ofjust over one hour including berthing operations. The rest ofthe trips in Table 4 are authentic simulations, except for the existing short ferry route between Lanzarote and Fuerteventura. This simulation is compared with the 1999 network (Table 3). Fast ferries would cut journey times between Santa Cruz de Tenerife and the capital cities on La Gomera El Hierro and La Palma by 15 min (10%), 130 min (38%) and 125 min (37%), respectively and between Las Palmas de Gran Canaria and Puerto del Rosario (Fuerteventura) by 105 min (30%). With optimum timetable planning, where perfect intermodal exchange is guaranteed between on-shore and maritime transport from the most eastern point to the most western point ofthe Canary archipelago (or vice versa), a trailer could leave at 06:00 h from the capital city on Lanzarote (the most eastern island), to board at 7 Co-operation agreement between the Ministry of Public Works and the Canary Islands Government in matters of roadways, Las Palmas de Gran Canaria, 16/04/97; Preview ofthe Regional Plan for Transport in the Canary Islands, Las Palmas de Gran Canaria, 1998; Regional Plan for the Canary Islands Road Network (1994, 2007), Las Palmas de Gran Canaria, 1994; and the General Guiding Plan for Infrastructures on the Canary Islands, Santa Cruz de Tenerife, 1996. 07:00 h and reach the capital cities on La Palma or El Hierro (the most western islands), at 18:00 h, after crossing the islands offuerteventura, Gran Canaria and Tenerife (about 450 km in a straight line). Currently it would be impossible to carry out this trip in one day, due to the slow ship speeds and poor intermodal connections. It is therefore only possible at present to cross from one island to another from east to west or viceversa and territorial integration is still limited in spite of improvements since 1994. 5. The availability of time as a territorial integrating factor for insular spaces One ofthe main requirements for adequate territorial integration in archipelagos is the level oftemporal accessibility between islands. Inter-island travel is ofvital importance, particularly for those islands that lack certain services due to small populations, and these trips have to be repeated quite often over the course of the year. Timing therefore plays an important role. Taking all ofthis into consideration, the availability oftime usually in one day for undertaking any activity in a specific place on the island visited may easily become a problem ofeconomic accessibility. This is because ifthe return trip cannot be completed by ferry within one day, the costs increase considerably by either having to use air transport, ifavailable, or having to pay for a hotel room and losing part of the next working

238 J.Á. Hernandez Luis / Journal of Transport Geography 10 (2002) 231 239 Table 4 Simulation of temporal accessibility between inland capitals on the Canary Islands using fast ferries and regional ports (in min) a Travelling to and from ports Check-in and boarding Time of sea trip Disembarking and luggage S/C de Tenerife Las Palmas de G.C. (Fast-Ferry) 60 20 65 10 155 S/C de Tenerife S.S. de La Gomera (Fast-Ferry) 60 30 40 10 140 S/C de Tenerife Valverde (Fast-Ferry) 60 30 110 10 210 S/C de Tenerife S/C de La Palma (Fast-Ferry) 60 30 115 10 215 Las Palmas de G.C. Puerto del Rosario (Fast-Ferry) 100 30 100 10 240 Puerto del Rosario Arrecife (Ferry comarcal) a 70 40 35 10 155 Mean 68.3 30.0 77.5 10.0 185.8 Source: Timetables ofshipping and bus companies. a (1) Only direct links have been considered with no stopovers whatsoever. (2) Temporal accessibility refers to passage. However, the accessibility of a trailer is similar to that shown here since the travelling to and from ports is similar to the times used by the buses. (3) We have kept the traditional ferry between Fuerteventura and Lanzarote due to the short trip involved (35 min). (4) Considered here is bus transport between Santa Cruz de Tenerife and Los Cristianos (or vice versa); Morro Jable and Puerto del Rosario (or vice versa); Corralejo and Puerto del Rosario (or vice versa); and Playa Blanca Arrecife (or vice versa). From there on the time left for the departure of the ship corresponds to checking-in and boarding times. Total Table 5 Availability ofcommercial and administrative time for maritime passengers in the capital cities ofla Gomera and Tenerife islands, 30 June 1999 (in min) a Government time Commercial time S/C de Tenerife S.S. de La Gomera (Hydro-foil) 375 330 S/C de Tenerife S.S. de La Gomera (Ferry Fred. Olsen) 270 240 S/C de Tenerife S.S. de La Gomera (Ferry Trasmediterranea) 285 215 S/C de Tenerife S.S. de La Gomera (maximum time) 375 330 S. S. de La Gomera S/C de Tenerife (Hydro-foil) 255 135 S. S. de La Gomera S/C de Tenerife (Ferry Fred. Olsen) 335 345 S. S. de La Gomera S/C de Tenerife (Ferry Trasmediterranea) S. S. de La Gomera S/C de Tenerife (maximum time) 335 345 Source: Timetables ofshipping and bus companies. a (1) From then on the rest ofthe time to departure ofthe ferry corresponds to check-in time, boarding and land transport from the place of destination to the bus terminal. (2) Government hours (08:00 15:00 h); commercial hours (09:00 13:00 and 16:00 20:00 h). day. Inter-island ferries and air transport do not operate overnight in the Canary Islands. Due to the population s dependence on inter-island transport there could be a case for a transport service that maximises temporal accessibility. Aerotaxis, or similar systems, are impractical because oftheir high cost. The maximum availability oftime for passengers in certain places on the destination island is a very important requirement ofinter-island transport systems. This is particularly the case between Tenerife and Gran Canaria, whose economies exhibit complementarity in employment, business, commerce and health. Administrative and commercial time available are considered separately in the case study chosen ofworking time available in the capital cities ofla Gomera and Tenerife (Table 5). In spite ofthe possibility ofcombining ship operators on the outgoing and incoming routes in order to increase the time available on the island ofdestination, the temporal accessibility ofthe trip reduces the time available to a large extent. Taking the case ofa person leaving Tenerife using the first available ferry to La Gomera and returning on the last ferry back to Tenerife. Only 69% of commercial working hours (330 of480 min) and 89% (375 of420 min) ofgovernment working time can be used on La Gomera. In the reverse direction, time available in Santa Cruz de Tenerife is 72% ofcommercial working hours and 80% ofgovernment working hours. However on other routes not all the inter-island services show such a relatively high ratio of available time, for example, from Gran Canaria to Lanzarote, the trips are not daily and it is therefore not possible to complete the return trip by ferry in one day. 6. Conclusions Efficient sea transport is a requirement for the social and economic development ofarchipelagos. Air transport is much faster but more expensive and not affordable for most island residents. A series of parameters can

J.Á. Hernandez Luis / Journal of Transport Geography 10 (2002) 231 239 239 be considered to optimise inter-island public transport, for example, by improving temporal accessibility and the availability oftime during government and commercial working hours on the destination island in order to avoid increasing costs by having to stay overnight. Inter-island transport problems in the Canary Islands are characteristic ofthose archipelagos whose islands are too distant from each other for the construction of fixed links. Temporal accessibility can be improved by minimising sea distances and introducing fast ferries to shorten overall journey times. This increases the time available for business on other islands within a day s return travel. References Biehl, D. et al., 1986. The contribution ofinfrastructure to regional development. Final report, Commission ofthe European Communities, Luxemburgo. Brookfield, H.C., 1980. The transport factor in island development. In: Shand, R.T. (Ed.), The Island States ofthe Pacific and Indian Oceans: Anatomy ofdevelopment. Development Studies Centre Monograph No. 23. Australian National University, Canberra, pp. 201 238. Caixa. 1999. Anuario Comercial de Espa~na, Servicio de Estudios de la Caja de Ahorros y Pensiones de Barcelona, Barcelona. Cozar, R. et al., 1998. El Transporte Marıtimo de Pasajeros en la Bahıa de Cadiz. Servicio de Publicaciones de la Universidad de Cadiz, Cadiz, 84 pp. Dunbar, A.C., 1981. Transport and development: Inter-island shipping in Vanuatu, PhD thesis, Australian National University, Canberra (unpublished). Gobierno de Canarias. 1995. Estimacion de la Renta Insular y Municipal, Instituto Canario de Estadıstica, Las Palmas de Gran Canaria. Gobierno de Canarias. 1998. Libro Blanco de los Transportes en Canarias, Consejerıa de Turismo y Transportes, Las Palmas de Gran Canaria. Gutierrez, J., 1988. Accesibilidad y transporte rural. Una perspectiva social. Revista del Ministerio de Turismo, Transportes y Comunicaciones 24, 27 40. Hagerstrand, T., 1970. What about people in regional science? Papers, Regional Science Association 24, 7 21. Hernandez, J. A., 1996. El transporte marıtimo en el contexto socioeconomico de Canarias. (Ed.) Consejerıa de Turismo y Transportes del Gobierno de Canarias, Las Palmas de Gran Canaria, 144 pp. Hoyle, B.S., 1999. Islands, transport and development. In: Biagini, E., Hoyle, B.S. (Eds.), Insularity and Development: International Perspectives on Islands. Pinter, London. Knowles, R.D., 1996. Fixed links and short sea crossings. In: Hoyle, B.S. (Ed.), Cityports, Coastal Zones and Regional Change. Wiley, Chichester. Lopez, E., 1990. Una vision del papel territorial de la oferta de transporte publico de pasajeros por carretera en la Comunidad Autonoma andaluza. Fundamentos para su reforma. In: In Estudios Geograficos, 198, pp. 65 81. Lutter, H. et al., 1992. Accessibility and Peripherality ofcommunity Regions: The Role ofroad-, Long-distance Railway- and Airport Networks, Commission ofthe European Communities, Brussels. MECSA. 1993. El transporte de viajeros entre Gran Canaria y Tenerife, Ministerio de Obras Publicas y Transportes, Madrid. Murillo, C. et al., 1995. Coste de la insularidad en Canarias, Ed. Consejerıa de Economıa y Hacienda, Gobierno de Canarias, Las Palmas de Gran Canaria, 129 pp. Nutley, S., 1983. Transport Policy Appraisal and Personal Accessibility in Rural Wales. Geo Books, Norwich. Rodrıguez, I., 1989. Accesibilidad de la poblacion en el medio rural: el caso de San Martın de Valdeiglesias. In: Actas XI Congreso Nacional de Geografıa, Madrid, vol. I, pp. 206 211. Rutz, W.O.A., Coull, J.R., 1996. Inter-island passenger shipping in Indonesia: development ofthe system, present characteristics and future requirements. Journal of Transport Geography 4 (4), 275 286.