International Journal

on Marine Navigation

and Safety of Sea Transportation

Volume 1

Number 2

June 2007

163

Worldwide and Route-specific Coverage of

Electronic Navigational Charts

E. Vanem, M.S. Eide & R. Skjong

DNV Research & Innovation, Det Norske Veritas, Hovik, Norway

G. Gravir

DNV Energy Region Nordic and Eurasia, Det Norske Veritas, Hovik, Norway

A. Lepsoe

DNV Maritime Technology and Production Centre, Det Norske Veritas, Hovik, Norway

ABSTRACT: This paper presents the results from a recent study on the coverage of Electronic Navigational

Charts (ENC). Global traffic data has been evaluated in relation to the coverage of ENC and eleven specific

ship routes, representative for global merchant shipping, have been analyzed in further detail. Overall, the

ENC coverage was found to be extensive, with 82 – 94% coverage for SOLAS ships, and 28 – 100% coverage

along selected routes. Furthermore, it has been demonstrated how the coverage of ENCs could be taken into

account when assessing the effect of ECDIS for safer ship navigation and associated cost effectiveness.

1 INTRODUCTION

1.1 Background and motivation

Electronic Chart Display and Information Systems

(ECDIS) represent a means for increasing the

navigational safety of ships. Formal Safety

Assessments (IMO, 2002) have been carried out on

various shiptypes, e.g. large passenger ships

(Norway 2005), oil tankers, bulk carriers, product

tankers (Denmark & Norway 2006) and LNG vessels

(Vanem et al. 2007a), and ECDIS emerged as a cost-

effective risk control option for all these shiptypes.

The possibility of formulating mandatory carriage

requirements for ECDIS has been on the agenda of

the Maritime Safety Committee (MSC) and the sub-

committee on safety of navigation (NAV) at the

International Maritime Organization (IMO).

However, many delegates expressed the view that

sufficient coverage of ENC would be a prerequisite

for such mandatory carriage requirement.

In order to investigate the actual coverage of ENC

in more detail, this paper compares global ship

traffic densities with actual ENC coverage. In this

way, the extent of holes in the global ENC coverage

and its effect on the cost-effectiveness of ECDIS

may be assessed. This will allow for a more accurate

evaluation of ECIDS as a risk control option. This

has been done for various shiptypes and

representative shipping routes for the present

situation as well as for the anticipated ENC coverage

in 2010.

1.2 ENC and ECDIS

ENCs are vector charts compiled from a database of

individual geo-referenced objects. IMO offer the

following definition for ENC (IMO 1995): ENC

means the database, standardized as to content,

structure and format, issued for use with ECDIS on

the authority of government-authorized hydrographic

offices. The ENC contains all the chart information

necessary for safe navigation, and may contain

supplementary information in addition to that

contained in the paper chart (e.g. sailing directions)

which may be considered necessary for safe

navigation. Being a database, ENC content may be

continuously retrieved by special operational

functions in ECDIS to give warnings of impending

danger related to the vessel’s position and its

movements.

164

The IMO ECDIS Performance Standards (IMO

1995) defines ECDIS equipment as follows:

Electronic chart display and information system

(ECDIS) means a navigation information system

which, with adequate back-up arrangements, can be

accepted as complying with the up-to-date chart

required by regulation V/20 of the 1974 SOLAS

Convention, by displaying selected information from

a system electronic navigational chart (SENC) with

positional information from navigation sensors to

assist the mariner in route planning and route

monitoring, and by displaying additional

navigation-related information if required.

1.3 Description of data sources

Two types of data are essentially needed for this

study, i.e. estimates of the distribution of the global

ship traffic and information about the global

coverage of ENCs.

For ship traffic distributions, AMVER and

COADS data has been used (Endresen et al. 2003).

Global traffic distributions were hence based on a

joint dataset containing both COADS and AMVER

data corresponding to a complete year. The ship

traffic density from this dataset is illustrated in

Figure 1.

Fig. 1. Combined AMVER and COADS data for one year

2000/2001

An overview of the worldwide coverage of ENC

is provided by the online catalogue of the

International Hydrographic Organization’s (IHO)

website (http://www.iho.shom.fr). The catalogue

distinguishes between ENCs that are commercially

available and ENC that will be available in the near

future. The coverage may be investigated for

different usage bands, i.e. overview, general,

coastal, approach, harbour and berthing. There are

known to be some gaps in the IHO ENC catalogue,

thus the results based on this source will tend to be

somewhat conservative. The coverage of

commercially available ENCs with resolution

coastal or better according to this catalogue is

illustrated in Figure 2.

Fig. 2. Global coverage of commercially available ENCs of

coastal or better resolution according to the IHO catalogue

2 GLOBAL ENC COVERAGE

2.1 Suitable ENC coverage

In order to estimate the global coverage of ENC, it is

necessary to regard ENCs as suitable or not. For the

purpose of this study an ENC is assumed suitable if

it contains sufficient level of detail for safe

navigation for the specific area it covers. It is further

assumed that ENCs labelled coastal or better will be

suitable for navigation in waters within 20 nautical

miles from the shore. I.e., suitable ENC coverage

will be assumed for all parts of a voyage closer to

shore than 20 nautical miles where officially

approved ENCs of scale coastal or larger are

available. In open waters further away from shore,

general or overview ENCs are deemed sufficient.

2.2 ENC coverage for SOLAS ships

Worldwide coverage of ENC has been mapped to

global ship traffic distributions in order to

investigate the percentage of the global traffic that

operates in areas with sufficient ENC coverage, i.e.

available ENC coverage of resolution coastal or

better for all stretches of the trade closer than 20

nautical miles to shore. World traffic patterns have

been collected for one complete year 2000/2001, and

these traffic distributions have been utilized. The

world traffic pattern is assumed unchanged in 2007

and 2010 compared to this dataset.

The ship traffic is reported on a global grid with a

1x1 degree grid cell resolution. The size of the

observation grid cells introduces considerable

uncertainties into this part of the study, and two

different approaches have been employed in order to

estimate which part of the traffic takes part in areas

close to shore. The first approach is to count the

traffic in all the grid cells intersecting with the 20

nm band. The second approach is to count only the

traffic in the grid cells whose centre point intersects

165

with the 20 nm band. These counting approaches

will be referred to as intersecting and centre-

intersecting, respectively. It is noted that the

intersecting counting approach will regard notable

more traffic to be within the coastal bands.

Two similar approaches have been employed in

order to estimate whether there is ENC coverage

within a cell or not, i.e. where there are ENC coverage

intersecting with any part of a cell and where there are

ENC coverage intersecting with the centre of the grid

cell respectively. Thus, there will be four estimates of

the ENC coverage based on the possible combinations

of the two counting approaches. The most

conservative approach is to use the intersecting traffic

grid cell counting approach for traffic estimation

coupled with the ENC centre-intersecting grid cell

counting approach for ENC coverage estimation.

Accordingly, a conservative estimate of the

percentage of world shipping traffic within 20 nm to

shore having sufficient ENC coverage is 82% for

2007. The most optimistic estimate is 94%. Thus,

the percentage of the current world shipping trade

having suitable ENC coverage along their voyages

are between 82% and 94%. For 2010, this is

anticipated to increase to 85% and 96% respectively.

Comparing the results for the anticipated ENC

coverage in 2010 to the current estimates, a slight

increase is expected. This is attributable to ongoing

or planned activities at various national hydrographic

offices. However, the increase is insignificant, and

this is explained by the fact that ENC coverage is

already quite extensive along coastal areas that carry

a great portion of the world ship traffic.

Estimates of the ENC coverage were also broken

down on four major shiptypes according to the

differences in trading patterns among these. The

most conservative estimates are presented in Table

1, for the current situation as well as the anticipation

for 2010. It can be seen that the variation between

the different shiptypes are not significant, and all the

four shiptypes that were investigated are associated

with ENC coverage around the global average and

well above 80%. Container ships were found to have

the highest coverage of more than 90%.

Table 1. Percentages of world ship traffic within 20 nm to

shore with sufficient ENC coverage – major shiptypes.

________________________________________________

Ship type 2007 2010

________________________________________________

Bulk carrier 82.4 84.4

Tanker 84.9 86.8

Container ship 90.4 91.4

General cargo ship 86.4 87.9

________________________________________________

3 SPECIFIC REPRESENTATIVE ROUTES

In order to evaluate the effect of holes in the ENC

coverage, particular routes and shiptypes have been

selected for more detailed investigation. These are

representative of the global traffic of merchant

shipping, in terms of reflecting both the most

common shiptypes and the busiest waters. For the

purpose of this study, eleven routes were selected as

representatives for the world seaborne trade, i.e.

three typical oil tanker routes, three bulk carrier

routes, two container vessel routes, one general

cargo route, one LNG carrier route and one chemical

tanker route. The selected routes, which are

indicated on a world map in Figure 3, are:

− Oil tankers:

1 Dammam, Saudi Arabia – Yokohama, Japan

2 Yanbu, Saudi Arabia – Galveston, TX, USA

3 Yanbu, Saudi Arabia – Barcelona, Spain

− Container vessels:

4 Singapore, Singapore – Rotterdam, Holland

5 Hong Kong, China – Long Beach, CA, USA

− Bulk carriers:

6 Newcastle, Australia – Qinhuangdao, China

7 Vitoria, Brazil – Hamburg, Germany

8 Vancouver, Canada – Salvador, Brazil

− General cargo vessels:

9 Helsinki, Finland – Cadiz, Spain

− Chemical tankers:

10 Rotterdam, Holland – Savannah, GA, USA

− LNG carriers:

11 Point Fortin, Trinidad & Tobago – Everett,

MA, USA

Fig. 3. Specific routes selected to represent global shipping

166

4 THE EFFECT OF LACK OF ENC COVERAGE

Previous studies have developed comprehensive risk

models based on Bayesian Networks and spreadsheet

models for accidents related to navigational failure

(Norway 2005). The risk models developed in these

previous studies were utilized also in the current

study in order to assess the effect of holes in the

ENC coverage along ship trades. In these risk

analyses, the frequency of grounding due to

navigational error (powered grounding) was

estimated based on the definition of three types of

waters, i.e. open waters, coastal waters and narrow

waters. The effect of ECDIS and hence of the extent

of ENC coverage will be different for these types of

waters. For the purpose of this study, the types of

water are defined in the following way:

− Open waters: > 5 nm from shore

− Coastal waters: 2 - 5 nm from shore

− Narrow waters: < 2 nm from shore

In order to investigate the ENC coverage along

the selected routes, the IHO global ENC catalogue

has been used to assess the availability of suitable

ENCs together with detailed route descriptions and

estimates of the time of voyage for each of the

selected routes. An additional three day in port has

been assumed for each trade.

The effect of holes in the ENC coverage along a

route will be less risk reduction attributable to

ECDIS. In areas where suitable ENCs are not

available, no benefits from ECDIS are assumed. I.e.

no risk reduction is ascribed to ECDIS in raster

mode. For the purpose of this study, it is assumed

that the effect of such holes is proportional to the

ratio of the route in coastal and narrow waters where

ENC coverage is insufficient, i.e. the ratio of the

route closer to shore than 5 nautical miles where

suitable ENC is not available. Thus, the net risk

reducing effects of ECDIS, ΔR

NET

will be reduced

accordingly.

( )

ENCNoENC

ENC

ECDIS

NET

×∆=∆

(1)

In equation (1), ΔR

NET

denotes the net risk

reducing effect of ECDIS for the selected route,

ΔR

ECDIS

denotes the risk reducing effect of ECDIS

for areas where suitable ENCs are available (about

38% according to previous studies (Denmark &

Norway 2006), assuming dual ECDIS). ENC

denotes the distance along the route in coastal or

narrow waters with suitable ENC coverage and

(ENC + No ENC)

is the total distance along the

route in coastal or narrow waters. These distances

have been investigated for the 11 selected routes.

It is noted that this study only accounts for the

effect of grounding risk reduction. It is

acknowledged that also other navigational risks may

be reduced, e.g. related to collision, and the

estimates of risk reduction used in this study should

therefore be regarded as conservative.

It is considered out of scope of this paper to

present the investigation of each of the eleven routes

in detail. The investigation of one of the routes will

be explained in more detail as a proxy, and it is

noted that the remainder of the routes are

investigated in a similar manner. Hence, in the

following, the investigation of the route between

Yanbu, Saudi Arabia and Barcelona, Spain will be

outlined.

4.1 Examining the Yanbu – Barcelona trade

The route between Yanbu in Saudi Arabia and

Barcelona, Spain, covers about 2100 nautical miles,

from the Red Sea, through the Suez Canal and past

the south tips of Sicily and Sardinia to the west coast

of Spain. 575 nautical miles of this route is closer to

shore than 20 nautical miles (27%), 187 nautical

miles is closer to 5 nautical miles (9%) and 96

nautical miles is closer than 2 nautical miles (4%).

The route is illustrated in Figure 4, and the ENC

coverage for this route is illustrated in Figure 5. The

characteristics of the route together with the ENC

coverage are presented in Table 2. The voyage

excluding time in port is estimated to take about 6

days.

The characteristics of this route have been used to

obtain the probability of critical course per year. The

corresponding annual grounding frequencies for

ships sailing this trade are presented in Table 3,

including estimates with and without ECDIS. It is

noted that for this particular route, the ENC coverage

is already quite extensive, and there are no

anticipated increase in ENC coverage within 2010.

Fig. 4. Route from Yanbu to Barcelona

167

Fig. 5. ENC coverage along the route Yanbu – Barcelona

Table 2. Route characteristics and ENC coverage for the route

Yanbu – Barcelona.

________________________________________________

Total < 20 nm < 5 nm < 2 nm

________________________________________________

Distance

nm 2154 575 187 96

% 100 27 9 4

ENC coverage

2007 95% 94% 98%

2010 95% 94% 98%

________________________________________________

Table 3. Annual grounding frequencies (per shipyear) and

frequency reduction for the route Yanbu – Barcelona.

_________________________________________________

Without 100% ENC Actual ENC

ECDIS coverage coverage

_________________________________________________

Frequency 7.2x10

-2

4.6x10

-2

4.8x10

-2

Frequency reduction 38% 36%

Groundings averted 2.8x10

-2

2.6x10

-2

_________________________________________________

4.2 Estimated grounding risk reduction in light of

actual ENC coverage

Similar exercises have been performed for the 10

other routes as well, and the ENC coverage,

grounding frequency reduction and the expected

number of averted groundings per shipyear are

summarized for all of the selected routes in Table 4

(only for the current situation).

Table 4. Estimated grounding frequency reduction and

groundings averted due to ECDIS with current ENC coverage.

__________________________________________________

ENC Grounding Groundings

Coverage frequency averted

(< 5 nm) reduction (per shipyear)

__________________________________________________

Dammam - Yokohama 41% 15% 7.2x10

-3

Yanbu - Galveston 57% 22% 1.8x10

-3

Yanbu - Barcelona 94% 36% 2.6x10

-2

Singapore - R

otterdam 63% 24% 1.5x10

-2

Hong Kong - Long Beach 100% 38% 3.1x10

-3

Newcastle - Qinhuangdao 28% 11% 1.3x10

-3

Vitoria – Hamburg 65% 25% 8.7x10

-3

Vancouver - Salvador 49% 19% 7.9x10

-3

Helsinki - Cadiz 100% 38% 1.2x10

-2

Rotterdam - Savannah 100% 38% 8.9x10

-3

Point Fortin - Everett 100% 38% 8.1x10

-3

__________________________________________________

The following general observations can be made

based on the examination of the selected routes:

− 4 of the 11 selected routes already have 100%

ENC coverage in coastal areas

− 6 of the 11 selected routes sees no anticipated

changes in the ENC coverage from 2007 to 2010

− The estimated grounding frequency reductions

due to ECDIS, in light of actual ENC coverage,

are between 11 and 38% for the selected routes

− The different routes have ENC coverage between

28% and 100% for stretches closer to shore than 5

nm. The global ENC coverage for ship traffic

closer to shore than 10 nm was estimated to be

between 84 – 96%.

4.3 Cost effectiveness of ECDIS in light of actual

ENC coverage for selected routes

The cost effectiveness corresponding to

implementing ECDIS on a newbuilding expected to

sail its entire life-time on each of the selected routes

have been estimated and the corresponding GCAF

(Gross Cost of Averting a Fatality) and NCAF (Net

Cost of Averting a Fatality) (see definitions in

Norway (2000)) values are presented in Vanem et al.

(2007c). In estimating the cost effectiveness, the

effect of reduced probabilities for oil spills are taken

into account based on the recently proposed CATS

approach, as described in Vanem et al. (2007b).

Based on these results, the following general

observations can be made, all of which are equally

true for 2007 as for 2010:

− GCAF > USD 3 million for all routes. This is due

to the generally low fatality rates in grounding

accidents for cargo ships, and hence a somewhat

limited effect of ECDIS in terms of saving lives.

− NCAF < 0 for all routes except one, indicating

that ECDIS is a cost effective risk control option.

− For cargo ships, the most important effect of

ECDIS is the environmental and property

protection in case of grounding.

− NCAF > USD 3 million for the route with poorest

ENC coverage only. Hence, ECDIS will only

cease to be cost effective on particular routes with

poor ENC coverage.

5 GENERIC COST EFFECTIVENESS OF ECDIS

IN LIGHT OF ENC COVERAGE

The cost effectiveness of ECDIS, taking the actual

ENC coverage into account, has been estimated for

particular routes. However, in order to formulate

recommendations for the world fleet, some globally

applicable estimates are required. Thus, the

168

arithmetic average reduction in grounding risk for all

routes will be assumed to represent the global risk

reduction from ECDIS implementation, i.e. for the

current situation: 9.1 x 10-3 groundings averted per

shipyear (this would increase to 1.0 x 10-2

groundings averted per shipyear for the anticipated

ENC coverage in 2010).

Different shiptypes are associated with very

different accident costs, and some global average

will be needed. It was found that the accident costs

are considerably higher for oil tankers than for other

cargo ships, mainly due to the high costs associated

with major oil spills, and that the accident cost

generally increase with ship size. Hence, a simple

average accident cost per GT will be assumed for oil

tankers and other cargo ships respectively. The

following average accident costs were derived based

on the cost model established by Spouge (2002), but

adopting the CATS approach to account for

prevention of oil spills:

− Oil tankers: 720 USD/GT

− Other cargo ships: 120 USD/GT

The expected number of fatalities in a grounding

accident is generally a function of crew size and

shiptype. The crew size is generally a function of the

size of the ship, but an average crew size of 25 has

been assumed for all ships for the purpose of

obtaining average estimates. According to the risk

models utilized in this study, the corresponding

average fatality rate per grounding accident,

applicable to all shiptypes, is 0.01 fatalities per

grounding event.

An average expected lifetime of 25 years is

assumed for all vessels. All estimates are assumed to

be valid for all SOLAS ships larger than 500 GT.

Ships smaller than this is considered out of scope of

this study. Based on these assumptions as well as

estimates related to the cost of ECDIS acquisition,

installation and maintenance, generic cost

effectiveness estimates for new and existing cargo

ships may be obtained.

5.1 Cost effectiveness for newbuildings

GCAF values associated with implementing ECDIS

on newbuildings are USD 30 million. This would be

reduced to USD 27 million for the anticipated

coverage in 2010. The NCAF value will generally be

a function of shiptype and size due to large

variations in accident costs.

For oil tankers, ships of 500 GT are associated

with an NCAF of USD 8.2 million. It can be shown

that NCAF will be less than USD 3 million for all

ships greater than 630 GT and negative for ships

larger than 700 GT. Hence, ECDIS have been

assessed to be cost effective for all new oil tankers

larger than 630 GT.

For other cargo ships, ships of 500 GT are

associated with an NCAF of USD 26 million. It can

be shown that NCAF will be less than USD 3

million for all ships greater than 3800 GT and

negative for ships larger than 4200 GT. Hence,

ECDIS have been assessed to be cost effective for all

new cargo ships, other than oil tankers, larger than

3800 GT.

5.2 Cost effectiveness for retrofit on existing ships

For existing ships, the cost effectiveness achievable

from implementing ECDIS will be a function of the

ship age. However, it can be shown that GCAF will

never be less than USD 3 million, which has been

used as cost effectiveness criteria in FSA

applications at IMO.

The NCAF value corresponding to implementing

ECDIS on existing cargo ships will generally be a

function of the shiptype, the ship size and the ship

age. The size of ships that correspond to NCAF <

USD 3 million and NCAF < 0 have been calculated

for various ship ages and are summarized in Tables 5

(for oil tankers) and 6 (for other cargo ships)

respectively.

Table 5. Oil tanker sizes corresponding to NCAF < USD 3

million and NCAF < 0.

______________________________________________

Ship age Ship size (GT)

_____________________________

NCAF < USD 3M NCAF < 0

______________________________________________

Newbuilding 630 700

5 years 720 780

10 years 870 920

15 years 1200 1200

20 years 2000 2100

24 years 9300 9300

______________________________________________

Table 6. Other cargo ship sizes corresponding to NCAF <

USD 3 million and NCAF < 0.

______________________________________________

Ship age Ship size (GT)

_____________________________

NCAF < USD 3M NCAF < 0

______________________________________________

Newbuilding 3800 4200

5 years 4300 4700

10 years 5200 5500

15 years 7000 7300

20 years 12,000 13,000

24 years 56,000 56,000

______________________________________________

6 CONCLUSIONS

The coverage of ENC in coastal waters have been

investigated and compared to global ship traffic data.

It was found that the actual worldwide coverage of

169

suitable ENC lie between 82% and 94% for SOLAS

ships. A selection of specific trades was made, and

the ENC coverage along these routes varied from

28% to full coverage. In light of this, the cost

effectiveness of ECDIS as a risk control option for

new and existing cargo ships has been evaluated.

The Gross Cost of Averting a Fatality will exceed

USD 3 million for all cargo ships. However,

considering the Net Cost of Averting a Fatality,

ECDIS emerged as cost effective for many

combinations of ship types, sizes and ages. In

general, there are major differences between oil

tankers and other types of cargo ships. This is mainly

due to the high costs ascribed to major oil spills.

Indeed, for cargo ships, averting oil spills was found

to be the most important aspect of averting

grounding accidents in terms of significant

contributions to accident cost savings.

The cost effectiveness in terms of NCAF as a

function of ship size and age has been evaluated, and

recommendations regarding carriage requirements

for ECIDS may be based on these results. Thus,

based on the analysis presented herein and IMO

criteria, it may be recommended that ECDIS be

made mandatory for:

− All new oil tankers of 500 gross tonnage and

upwards.

− All new cargo ships, other than oil tankers, of

3000 gross tonnage and upwards.

− All existing oil tankers of 3000 gross tonnage and

upwards.

− All existing cargo ships, other than oil tankers,

10,000 gross tonnage and upwards.

ACKNOWLEDGEMENTS

This study has been carried out with the support

from the Finnish Maritime Administration,

Norwegian Mapping Authority – Hydrographic

Service, Swedish Maritime Administration and

National Survey and Cadastre (Denmark). Results

from the study have been reported to IMO (NAV

53).

REFERENCES

Denmark and Norway (2006). "FSA Study on ECDIS/ENCs",

Submitted by Denmark and Norway, MSC 81/24/5, IMO.

Endresen, Ø, Sørgård, E, Sundet, JK, Dalsøren, SB, Isaksen,

ISA, Berglen, TF and Gravir, G (2003). "Emission from

international sea transportation and environmental impact",

Journal of Geophysical Re-search, Vol 108, 4560.

IMO (1995). "Performance standards for electronic chart

display and information systems (ECDIS)", IMO resolution

A.817(19).

IMO (2002). “Guidelines for Formal Safety Assessment (FSA)

for use in the IMO rule-making process”. MSC/Circ.1023 -

MEPC/Circ.392, IMO.

Norway (2000). “Formal Safety Assessment – Decision

parameters including risk acceptance criteria”, Submitted by

Norway, MSC 72/16, IMO.

Norway (2005). "FSA – Large Passenger Ships - Navigational

Safety", Submitted by Norway, NAV 51/10, IMO.

Spouge, JR (2002). "A Simple Model of the Costs of Ship

Accidents Rev 1", DNV Job No C383184/5.

Vanem, E, Antão, P, Castillo, F and Skjong, R (2007a).

"Formal Safety Assessment of LNG Tankers", Proc 10th

International Symposium on Practical Design of Ships and

Other Floating Structures, PRADS 2007.

Vanem, E, Endresen, Ø and Skjong, R (2007b). "CATS – Cost-

Effectiveness in Designing for Oil Spill Prevention", Proc

10th International Symposium on Practical Design of Ships

and Other Floating Structures, PRADS 2007.

Vanem, E, Gravir, G and Eide, MS (2007c). "Effect of ENC

Coverage on ECDIS Risk Reduction", DNV Report no.

2007-0304, Det Norske Veritas. http://research.dnv.com

/skj/FSA-ENC/ENC.pdf