July 15, 1999
§ 3. The shipping industry and ballast water: Metric tonnes, dollars, and things getting through the cracks
Because the ballast water of transoceanic shipping is the major vector for current invasions of the Great Lakes, the following subsections provide a detailed review of the global shipping industry, the characteristics of ships which affect the water they carry, how much they carry, what comes into the Great Lakes, where it comes from, the Great Lakes and Seaway system, the costs of shipping, the overall threat from ships, the current Great Lakes exchange regime, and technological options for managing ballast water.
About 98% of world trade, 4.5 billion metric tonnes per year,(59) and 80% of world commodities,(60) are carried by sea. The major bulk cargoes are coal, iron ore, crude oil, grain, rice, steel, timber, bauxite, phosphates and refined products of oil. These relatively "low value" cargoes are fuel and raw materials for industry, and food for people. In addition, all variety of higher value manufactured products are carried by ships, many of these in the newer form of containerized cargo which is "intermodal" between sea and land. Shipping is a truly global trade, with vessels of a multitude of nations calling at most ports around the world. The international character of shipping is also reflected in the fact that it is not uncommon to see a vessel owned by a company in one nation, registered in another, chartered for use by a company in another, and actually sailed by officers and crewmembers from several other nations. Many vessels operate under "flags of convenience," which simply means that they register with nations which offer economical terms and conditions, but which sometimes also have doubtful regulatory standards. (There has been a long-term decline in the number of US-owned or controlled vessels actually registered in the United States.) It is a highly sophisticated, but also highly varied industry. Some vessels are operated by major companies with large fleets, long-term international reputations to uphold, and high internal standards for maintenance and safety. Other vessels are "bare boat" chartered to temporary owners with little interest in the maintenance of the vessel or owned by one-vessel corporations which anticipate bankruptcy. Some vessels, "liners," operate on regular routes and schedules. Others are "tramps," which does not necessarily mean that they are poorly maintained, chasing the cargo wherever it takes them.
Shipping is one of the most highly regulated industries in the world. Nations control shipping both as "flag states," exercising legal authority over vessels registered under their flag wherever in the world, and as "port states," exercising legal authority over vessels of any nation operating in their waters. In both Canada and the United States, shipping is considered a matter of special concern to the federal governments and has been subject to regulation long before other areas of the national economy. In addition, shipping is regulated by international organizations, both governmental and commercial. The International Maritime Organization in London, an agency of the United Nations, provides a central forum for negotiation of "conventions" or treaties on safety and environmental protection. International insurance consortiums such as Lloyds of London and "classification societies" such as the American Bureau of Shipping and Det Norske Veritas set industry safety standards which are often adopted by government agencies. International commodity exchanges, such as the Baltic Exchange (which operates in London), and other international "conferences" (standing industry consortiums) negotiate standard cargo rates for liners, subject to some governmental controls.(61)
Many governments (including both the United States and Canada) have policies to promote shipping under their flag for purposes of both national security and economic protectionism. These may include cabotage protection (restricting domestic coastal trade to the ships of the coastal nation), cargo preferences (requiring government or government subsidized cargo, such as cargo paid for by foreign aid, to be carried on flag vessels), and outright subsidies. Despite both industry and governmental efforts to create market stability and promote the shipping of flag states, the prices for ships, charter rates, and freight rates are strongly affected by changes in economic conditions around the world and are highly volatile. (For example, the "Asian crisis" of 1998 drove down the prices of ships by 8% to 10%, and even more for new construction, depending on type. Some of that many have been due to the fact that currencies fell in Japan and Korea, where many ships are built. During the same time, however, daily charter rates for certain classes of bulk carriers declined by halves and thirds, and freight rates for carrying grain on some vessels declined by almost half.(62)) Some industry representatives argue that they are losing economic viability due to excessive governmental regulation, and they particularly object to the long-term trend of industry standards being replaced by legal conventions and domestic legislation.(63) Generally, however, the shipping industry is characterized by oversupply of ships and declining profits to vessel owners and operators, due to government policies subsidizing shipping, increases in efficiency through containerization,(64) and perhaps also the market power of the shipping conferences.(65)
Bulkers, tankers, and container ships. The main categories of major commercial vessels are bulk carriers ("bulkers"), tankers, and container ships. More than 70% of the vessels entering the St. Lawrence Seaway and the Great Lakes from foreign ports are bulkers.(66) The bulkers are still the most efficient way to carry large "low value" dry bulk cargoes such as grain and ore. However, the fact that these cargoes gather relatively little money for each tonne makes the trade sensitive to relatively small variations in freight rates and other costs.
Ships great and small. How much ballast a ship carries varies with its type and size. Ship sizes are most often identified by "deadweight tons" (DWT), in English long tons in the US and in metric tonnes(67) in the rest of the world. This is a rough measure of their cargo carrying capacity (after deduction for fuel, stores, etc.). The tankers and bulkers are commonly grouped in rough size categories which relate to where they are small enough to operate, although that is also determined by their specific physical dimensions. The "handysize" tankers and bulkers which are small enough to fit through the St. Lawrence Seaway,(68) also sometimes called "small handys," are anywhere from 10,000 to 35,000 DWT. Other ships in world trade include a somewhat larger "handymax" ship from 35,000 to 50,000 DWT, a "Panamax" ship which can just fit through the Panama Canal from 50,000 to 80,000 DWT, a "Capesize" bulker which must go around Cape Horn from 80,000 to 150,000 DWT, a "Suezmax" tanker for the Suez Canal at about 140,000 DWT, a "very large crude carrier" (VLCC), a tanker over 300,000 DWT, and an "ultra large crude carrier" (ULCC) or "mammoth" tanker at 500,000 DWT.(69)
Ballasting up. Ships use ballast water to maintain the essential operational and safety parameters for overall immersion, trim, stability, and hull integrity. It is dangerous for a ship to be either too low or too high in the water. Regulatory "load lines" or "marks" (literally welded-on marks on the outside of the hull) control the commercial tendency to load on as much cargo as possible. When running without cargo, or a light load, ballast is taken on to compensate according to the judgement of the master and loading guidance for the specific ship. The master is often balancing commercial profit and safety. A ship not ballasted sufficiently is unsafe. But one ballasted too much is slowed down and uses more fuel. Ballast may be carried in either dedicated or "segregated" ballast tanks, used exclusively for that purpose or in the cargo tanks of tankers or bulkers. Ballasting cargo tanks is less common than it once was, and is now typically to be found only when the vessel is putting on extra "heavy ballast" or "storm ballast" in order to deal with bad sea conditions. A vessel reports itself as "BOB" for "ballast on board," meaning that it is either "in ballast" (full up) or "with ballast" (partially ballasted), or "NOBOB" for "no ballast on board," meaning that it has pumped down the ballast tanks as far as possible. But "NOBOB" is almost never literally true, because most ships carry unpumpable slop and sediment in the range of a few hundred tonnes on the bottom of the tanks. For the obvious commercial reasons, ships will run with maximum cargo and minimum ballast as much as possible. Container ships generally carry less ballast and, unlike the tankers and bulkers, some container ships may carry a substantial portion of their ballast water as permanent or semi-permanent ballast which does not need to be discharged, because it is used to trim a loaded vessel for purpose of stability (to offset the relatively high weight of containers on deck). Through careful planning, some of the newer container ships can significantly minimize the total amount of ballast that they need to discharge in port.(70) Most tankers and bulkers, however, must discharge large quantities of ballast in order to take on their cargo.
Typical loads. Getting an idea of the amounts of ballast water to be dealt with is important to understanding the practicalities of various management options. Unfortunately, this is not as straightforward as one might think. Keeping in mind that it varies according to the specific type, size, and design of a vessel, one rough rule of thumb is that the basic ballast capacity of a bulker or tanker, using only dedicated tanks, runs around one third of its DWT.(71) The ballast capacity of dedicated tanks on a handysize bulker or tanker coming into the Great Lakes through the Seaway runs from about 1,000 to 10,000 metric tonnes.(72) The typical ballast carried by a vessel running "in ballast" or "with ballast," depending on sea conditions, is less than this, anywhere from 15% to 30% of DWT.(73) Estimates of average or typical loads on the BOB vessels entering the St. Lawrence Seaway and the Great Lakes from overseas ports - mainly bulkers - vary quite a bit from a low of 3,115 metric tonnes(74) to highs of 7,013(75) and 7,500(76) metric tonnes. Most of the vessels, however, are running with cargo in a NOBOB status, with tanks pumped down as far as possible. These have varied between 75% and 95% of the vessels entering the system in recent years,(77) but a study conducted in 1990 estimated the NOBOBs to be only 52% of the total number.(78) They carry some amount of unpumpable slop and sediment, estimated to range from 59 to 468 metric tonnes and average 157.7 metric tonnes per vessel.(79)
Moving the water around. Ballast pumping rates - which also bear directly on the costs of many of the technological options for managing exotics in ballast - tend to vary according to the basic category of the vessel and the purpose for which the ballast is used. Bulkers (such as the third party foreign vessels entering the Great Lakes), which are using ballast to replace cargo, have typical rates of 2,000 to 10,000 m3/hour. Tankers, which use it for the same purpose, have typical rates of 5,000 to 20,000 m3/hour. Container ships, which are often using it for trim and stability in both unloaded and loaded conditions, and are therefore moving less total water around, have typical rates of 1,000 to 2,000 m3/hour.(80) (These are combined rates, including more than one pump, and smaller vessels with less total ballast tend to be on the lower end of the scale.)
Speeds. The speed of vessels, which affects the survivability of the organisms in the tank, the scouring action of fouling on the outside of the hull, and the time available for exchange in the open ocean, has been slowly increasing over the years. Typical maximum speeds for modern vessels (which require maximum fuel consumption) range from 12 to 14 knots (14 to 16 miles per hour), with older and larger vessels tending toward the low end of the range. The passage from Rotterdam to Montreal is 3,138 nautical miles (3,612 statute miles and 5,805 kilometers), or 10 days at 13 knots. (Passage up through the St. Lawrence Seaway and the Great Lakes to the western end at Duluth or Thunder Bay is slower going and may take another 7-9 days.) During the crossing, a handysize vessel may come up in weight from its original loading (typically right "down to the marks," the regulatory loadlines on the outside of the hull showing the maximum safe immersion for the sea and season) due to burning off something around 40 tonnes of fuel per day.(81) (Larger vessels would burn correspondingly larger amounts.) Some vessels will make intermediate stops at ports along the North American coast before entering the Great Lakes, during which they may mix North American coastal water with unpumpable slop and sediment they carried across the Atlantic while loaded with cargo (and therefore unable to exchange ballast).
Counting metric tonnes. The total amount of overseas ballast entering US coastal ports (non-Great Lakes) each year has been estimated at 44.7 million metric tonnes (or 11.5 billion gallons).(82) For Canada, one set of somewhat mixed estimates puts the corresponding figure for overseas ballast entering Canadian coastal ports each year at 49.7 million metric tonnes (or 13.1 billion gallons).(83) There have been a number of attempts to estimate the amount of overseas ballast water entering the Great Lakes. Unfortunately, different methods and assumptions have resulted in widely different estimates. There is a significant difference between what the vessels are carrying when they enter the St. Lawrence Seaway and what they eventually discharge, with lake water added in along the way, at Great Lakes ports. The first amount is relevant to what has to be exchanged or treated aboard the ship before it shows up in the Great Lakes. The second amount is relevant to what would have to be treated in port upon arrival - which also varies significantly between different ports inside the Great Lakes. Both figures will also vary with changes in the trade, ranging from about 400 to 600 vessels each year, and how many vessels have ballast on board (BOB) because they are not loaded with incoming cargo.
Based on a comparison of the studies, I would suggest a round figure of 720,000 metric tonnes being carried upon entry into the Seaway in an average year. This figure is based on two of the studies, one conducted in 1990 and another in 1995. Both estimates in those studies may be slightly high for different reasons, but are probably reasonably close to current years, during which trade has increased somewhat.(84) A notional profile of the vessel traffic upon entry through the Seaway, averaging out several assumptions in the various studies, and not reflecting any one actual year, might be:
|500 Transoceanic entries||Total ballast||=||720,000 MT|
|100 BOBs (20%)||x||6,560 MT/BOB||=||656,000 MT|
|400 NOBOBs (80%)||x||160 MT/NOBOB||=||64,000 MT|
Both the BOBs and NOBOBs engage in considerable cross-ballasting and movement of additional water in the course of stopping at an average of 2.3 ports inside the Seaway and the Great Lakes.(85) The amount of overseas water ending up at the ports, mixed with water from intermediate ports, has been separately estimated to be as much as 4,406,498(86) to 5,700,000(87) metric tonnes, although both of these estimates are likely to be somewhat high.(88) According to the study producing the lower estimate, "nearly 45% of all ballast loading [inside the system] occurred in Hamilton and Cleveland…while almost 41% of all deballasting occurred in Duluth-Superior and Thunder Bay. Additionally, results indicate that of the 4.4 million tonnes of ballast discharged during 1995, vessels entering NOBOB accounted for nearly 3.7 million (84%)."(89) This last point is of considerable significance because the NOBOBs do not exchange their water on the open ocean under the current regime.
In terms of the biological material in the water, the exact amount of the water is less important than where it comes from and how it gets here. It is primarily a problem with what are called the "third party" foreign vessels - vessels which are neither US nor Canadian - entering the Great Lakes through the St. Lawrence Seaway from transoceanic ports, although some Canadian vessels also participate in this trade. There is no bar to US transoceanic vessels entering the Great Lakes. But it happens to be very rare, perhaps in part because of the limitations on the size of a vessel which can fit through the Seaway.
The Erie Canal. There are two physical connections between the Great Lakes and the Atlantic Ocean. The Erie Canal (or "New York State Canal System," as it is now called) opened in 1825 and connected the Hudson River and New York to Lakes Ontario and Erie through a 524-statute mile system of interlocking canals and locks. (The Welland Canal, connecting Lakes Ontario and Erie, bypassing Niagara Falls, was also opened in 1829.) Although the Erie Canal was commercially significant at the time, it is much less so now after the opening of the St. Lawrence Seaway. The Erie Canal can accommodate barges and small vessels, but the locks limit it to vessels within the dimensions of 325 feet in length by 45 feet beam by 12 feet draft by 15.5 feet air draft (clearance under bridges). Although it is still used by recreational and fishing vessels from the open ocean, it is not used by major transoceanic commercial carriers.
The St. Lawrence Seaway. The St. Lawrence Seaway System was opened in 1959, as a joint project of Canada and the United States, to open up the Great Lakes to modern transoceanic trade. It is administered by the Canadian St. Lawrence Seaway Management Corporation (a non-profit corporation which replaced the Canadian Seaway Authority in 1998) and the US St. Lawrence Seaway Development Corporation. The Seaway Management Corporation and Canadian government agencies exercise authority over the entrance to the system via the Canadian waters of the Gulf of St. Lawrence, the St. Lawrence River, five Seaway locks from Montreal to Lake Ontario, and the Welland Canal. The US St. Lawrence Seaway Development Corporation and other US government agencies exercise authority over two locks between Massena, New York, and Lake Ontario. Each nation has exclusive territorial jurisdiction over the waters of the St. Lawrence within their boundaries (as a matter of international law, these are "internal waters," not "high seas" or "territorial seas" subject to rights of innocent passage) and vessels transiting the system must comply with the domestic laws of both nations as they pass through. The similarity of the maritime laws and traditions of the two nations, the close working relationship between the maritime agencies on the Great Lakes, and joint boarding programs set up in cooperation with both the maritime agencies and the two national Seaway administrations, all make this much less burdensome and confusing to vessel operators than it might otherwise be. The maximum Seaway dimensions are 225.5 meters (740 feet) length, by 23.7 meters (78 feet) beam, by 8.0 meters (26 feet, 3 inches) draft, by 35.5 meters (116.5 feet) "air draft" (clearance under bridges). This opens up the Great Lakes to major transoceanic vessels, but only to the relatively smaller and older handysize vessels in the world fleets.
The Great Lakes and Seaway maritime industry. Like Caesar's Gaul, the Great Lakes and Seaway maritime industry is divided into three distinct tribes, (1) US domestic "lakers," (2) Canadian "lakers" and "salty lakers," and (3) third party "salties," which are our primary concern. The largest, although not the most numerous, ships sailing the lakes belong to a fleet of about 70 large US domestic bulk carriers called "lakers," (of which, perhaps 60 or so are actually sailing in a typical year), most of which belong to companies who are members of the US Lake Carriers' Association (LCA). These are restricted to the inside of the lakes, and therefore are not a source of transoceanic ballast, although they may contribute (along with more than 2.2 million US recreational vessels operating on the Great Lakes) to the spread of exotics already introduced. There is a comparable Canadian fleet of about 70 vessels, generally smaller in size, operating both inside and outside the Great Lakes, most of which are members of the Canadian Shipowners Association (CSA).(90) These are also called "lakers," but they are more difficult to define because a good number of them are "salty lakers" which operate on the ocean as well, sometimes seasonally, often making regular runs through the St. Lawrence Seaway. Although these vessels operate on the ocean, they are primarily in Canadian coastal and Great Lakes trade. Those among them that do make transoceanic voyages become subject to the same requirements for ballast exchange applicable to the third party transoceanic fleet.
The third party salties. And then there is a "third party" fleet, neither US nor Canadian, of transoceanic vessels, "salties," trading between the Great Lakes and the rest of the world. This includes a wide variety of vessels, anywhere from 200 to 300 ships running under 35 to 45 different flags, making 400 to 600 round trips (or a little more) per year. Of this, there is a core group of about 25 vessels which make regular runs into the Great Lakes. These are "liners" if they publish definite schedules. Many of the core group are owned and managed by major corporations in North America.(91) Some of the other vessels in the overall lot are what the public associates with the term "tramp" vessel (which technically just means a vessel without a regular repeat cargo run). That is to say that some of them are vessels frequently changing ownership or charter, operating under "flags of convenience" from countries with doubtful enforcement of marine safety laws, sometimes under "bare boat charter" for only one specific cargo run, sometimes owned by a "shell corporation" whose assets are little more than the one vessel.
The only thing the vessels of this third party fleet have in common is that almost all of them are handysize bulkers. On average, these handysize bulkers are old. Vessels in the area of 25 years of age are quite common, and they are becoming older. (That is not old in comparison to US lakers, which have a median age of just about 25 years. But the fact that lakers run in fresh water, which is much less corrosive, and the very high standards of maintenance on both US and Canadian lakers, make that a meaningless comparison.) There were 13,000 handysize vessels in the world in 1996,(92) but there had been only 89 new vessels of this class built in a ten-year span from 1984 to 1993.(93) More recent market reports indicate the size of the handysize fleet is continuing to decline.(94) Some of the third party vessels coming into the lakes are in fact "rust buckets" on their last legs, and with little remaining capital value.
They generally carry steel and high value cargoes such as heavy machinery and other manufactured goods into the Great Lakes. Steel, weighing in at about 5 million metric tonnes (about 10% of total tonnage) is valued at about $2 billion per year.(95) Although the foreign steel is of great concern to those in the domestic raw steel industry, such as the US lakers, this high quality feed to US and Canadian industries helps them to remain competitive in world markets. Also, despite the objections by domestic suppliers and vehement complaints about dumping of foreign steel, these imports are actually subsidized by a discount of 50 cents per tonne on the toll for steel, granted by both Seaway agencies in 1994.(96) The most important outgoing cargo is low value grain (low value only in the sense of a low price per tonne) from the North American heartland. Most of this is shipped out of Duluth and Thunder Bay, the major US and Canadian ports on the western end of Lake Superior.
Vessel operators always try to run with cargo both ways. Anywhere from 75% to 95% of them, more towards 95% in a typical year, if there is such a thing for this trade, run "in cargo" rather than "in ballast" on the way in. These report themselves as "no ballast on board" or "NOBOB" upon entry, but they actually carry a substantial amount of slop in the bottom of their tanks. Some of these also discharge part of their cargo at Montreal before proceeding to the first lock of the Seaway System, often because they have to lighten up to meet the Seaway draft limits of 8.0 meters (26 feet, 3 inches). After discharging that partial cargo in Montreal, they may then have ballast back on a few hundred or thousand metric tonnes to adjust trim.
The triangle trade. The third-party ships come into the lakes from every continent and climatic zone in the world. In 1994, when US and Canadian marine safety officers combined forces to prevent the illegal discharge of a load of freshwater ballast from the motor vessel Pal Wind (which has since changed ownership and name at least twice), they discovered from the ship's log that the fresh water was from the mouth of the Congo River. However, the major overseas markets are Western Europe, the Baltic, the Mediterranean, and the Middle East.(97) (The Mediterranean, Black, and Baltic Seas have also been heavily impacted by invasions of exotics, including organisms from North America.(98)) Perhaps 70% of the vessels follow a typical pattern of "triangle trade," carrying grain from the Great Lakes to ports in the Mediterranean, then to ports in Northern Europe, often with Mediterranean ballast in their tanks, to pick up steel, machinery, and other high value cargoes for the Great Lakes.(99) Transit time across the North Atlantic, sometimes with intermediate stops on the US or Canadian seaboard, may run from around ten days to two weeks. That relatively short transit time, from the heavily contaminated waters of Europe, increases the probability that exotics will survive the voyage.(100) Although the Seaway is only open from late March to late December, the North Atlantic in their operating area of latitudes is frequently cold and stormy, especially during the last runs before the end of December. Exchange of ballast on the open ocean, under the limitations of current tank and piping designs, can be dangerous. That was very well illustrated when the Flare, a typical old handysize bulker, broke in half off Newfoundland on January 16, 1998.
Tonnage carried. Total cargoes carried through the Seaway (both ways, including Canadian and third party vessels, and both the St. Lawrence and Welland Canal sections) ranged from 40 to 50 million metric tonnes per year in recent years. While the recent average annual variance in US and Canadian tonnage is in the area of 2%, the average annual variance for the Seaway is double that at about 4%.(101) (Actual changes in individual years are sometimes much more dramatic.) Although cargo levels have improved in recent years, they are still significantly below the historic high level of 74 million metric tonnes in 1979.(102) Overall cargo topped 51 million metric tonnes in 1998 for the first time in a decade.(103)
Paying for the Seaway. Although it was originally supposed to be self-supporting, the Seaway has never paid for itself. In the words of a supporter, "The early high expectations for the Seaway have not yet been realized."(104) It has instead become another publicly supported transportation system (not in principle different from subsidies to airports and trains, or to other marine transportation infrastructures provided by US and Canadian navigation projects and marine services). The approximate cost of the original navigation project was just over $470 million, of which Canada paid $337 million and the US about $134 million. (Associated US projects to deepen interior Great Lakes navigation channels to accommodate Seaway traffic, which also benefited domestic shipping, added $257 million to that original bill.)
The US spent nearly $14 million, in 1979-1980, to conduct the Seaway Navigation Season Extension Demonstration Program, in order to prove the concept of all-winter navigation. That was a failure, and there is no serious advocacy for all-winter navigation in either the US or Canada at the present time. In fact, both nations, and their marine industries, are struggling with the need to reduce or reimburse government expenditures on current icebreaking (which are funded separately from the Seaway appropriations). The US Merchant Marine Act of 1970 relieved the US SLSDC of the requirement that it pay interest on its construction debt, and the US Department of Transportation Fiscal Year 1983 Appropriations Act cancelled the SLSDC's remaining $110 million construction debt. The US Water Resources Development Act of 1986 established a Harbor Maintenance Trust Fund, funded by a tax on foreign cargoes handled at all US ports, which now supports 90% of the SLSDC's operations. The SLSDC continued to collect tolls, but they were then rebated to users by the US Treasury. US tolls were completely cancelled in 1994.(105)
Canada has raised its tolls in recent years in an attempt to recover costs. In 1986, however, Canada appropriated $175 million from general revenue for a seven-year project to rehabilitate the Welland Canal. The Canada Marine Act of 1998(106) replaced the St. Lawrence Seaway Authority with the St. Lawrence Seaway Management Corporation, a nonprofit corporation.(107) The Canadian Management Corporation is continuing to collect tolls, and announced at the beginning of 1998 that "Toll revenues were the highest in Seaway history, at $80 million, thanks to the substantial increase in steel movements."(108) However, the Canadian government still owns the assets of the Seaway and has committed to pay for all renewal and maintenance of the 40-year old infrastructure beyond $15 million per year.(109) It does not appear that the system will ever pay for itself on the Canadian side either.
In order to put the discussion of the costs of technological options for handling ballast water into context, the following figures are offered as general parameters. Each vessel's value is affected not only by type, size, and age, but also by its design, special features such as cargo handling equipment, history of maintenance, and changing market needs. Markets are highly volatile. New construction prices slumped from 15% to 30% in 1998, in part because of the decline of currencies in the Asian countries where almost all bulkers and tankers are built, but also because of real declines in raw materials and decline in demand due to general over-capacity.(110) It costs tens of millions of US dollars for a new ship, depending on size and design. A new handysize tanker may go for around $20 million. It is rather difficult to find public reports of typical prices for new handysize bulkers of the size which fit through the Seaway because very few of them are being built, but a current new construction price for a rather small handysize bulker of $13 million(111) suggests that prices for average range handysize bulkers would be a little less than that $20 million benchmark for tankers if there were a current market for them. Average new prices for the next largest ships, the handymax bulkers, ranged from $26 to $19 million from the third quarter of 1997 to the third quarter of 1998 (a decline of 27%).(112) The ranges for larger ships during the same period were as follows: Panamax bulkers, $27 to $20 million. Panamax tankers, $36 to $30 million. Capsize tankers, $42 to $35 million. Suezmax tankers, $51 to $45 million. VLCC tankers, $82 million to $70 million.(113) Older ships can be had for much less. A handysize or handymax bulker around 10 to 20 years old may go for something in the range of $4 million to $10 million.(114) (Old ships at the end of their useful life, anywhere from 20 to 30 years, may obtain something around $1 million in scrap value.) A statistically average size bulker in the world fleet is around 49,500 DWT and is 14.5 years old.(115) This "average" large handymax (or near small Panamax) would be near the upper end of that $4 to $10 million range. According to an index of average values in a "representative fleet," the mean average value of a bulker in world trade is currently $8.6 million, down from $9.7 million one year ago, and $13 million for a tanker, down from $20.5 million one year ago.(116)
Ships are commonly chartered for 6-12 months or voyages, with rates calculated by the day. Depending on the same variables of size, age, etc., charter rates run anywhere from $5000 to $30,000 per day, with extreme fluctuations due to markets. Handymax bulker charter rates ranged from $8,000 to $7,000 per day in 1997-1998.(117) Rates for the smaller handysize bulkers fitting inside the Seaway would be a little less. For a typical Capesize bulker, it ranged from $15,000 to $10,000 in 1997-1998.(118) Tanker charters depend on type of cargo and route as well as size. Charter rates for Suezmax crude carriers varied from $22,000 to $17,000, and for very large crude carriers (VLCC), $33,000 to $27,000, during 1998.(119) According to an index of average rates in a "representative fleet," the mean average rate for a bulker in world trade is currently $6,700 per day, down from $7,044 one year ago, and $14,983 for a tanker, down from $18,350 one year ago.(120)
To this one must also add thousands of dollars per day for crew, fuel, other operational expenses, and in-port fees. Visits to a drydock can add tens of thousands to a hundred thousand or more per visit, not counting the actual repair or maintenance costs.
Although ships have been subjected to extensive regulation for safety for about a century, and have been regulated to prevent spills by oil and chemicals for several decades, there are still no international legal controls on exotics in ballast water and other shipborne sources for aquatic organisms. This is odd, in a sense, because ships have long been recognized as vectors for the transmission of diseases and terrestrial pests.(121) Some of the marine biologists have pointed out that a ship can well be thought of as a "biological island" carrying its organisms around the world.(122) The variety of means by which ships can spread organisms include ballast water, fouling on the outside of the hull, poorly maintained marine sanitation devices, water on decks, water in anchor lockers, or fouling on anchors or chains. ("Bilge" water, commonly confused with ballast water, is small quantities of water forming by condensation or leaking from decks into lower interior spaces, and is not a significant threat.) Ballast water remains the greatest concern because of the huge quantity of water transported and the wide variety of organisms it carries.
Ballast. Ballast water began to be commonly used in ships with the introduction of steel construction in the mid 1800s.(123) Since that time, the threat of transport of exotics has grown as (a) total world trade has increased, (b) ships have become larger, and (c) ships have become faster. Large transoceanic vessels could not enter the Great Lakes until after the opening of the St. Lawrence Seaway in 1959, although some much smaller traffic from ports around the world could make it in from the Hudson River after the opening of the Erie Canal in 1825. Since 1960, however, introductions attributable to transoceanic ballast water have accounted for both a clear majority of all recent introductions (perhaps 60% of introductions between 1960-1997) and a dramatic surge in introductions since that time. "In fact, almost one-third of the exotic species in the Great Lakes have been introduced in the last 30 years, and this surge corresponded with the opening of the St. Lawrence Seaway."(124) The sea lamprey may have come up the Erie Canal. Most of the more recent invaders of note, the zebra mussel, the European ruffe, the round goby, the tubenose goby, and the spiny water flea, are attributable to ballast water.(125) Some of the researchers have calculated that, at any given time, there may be "somewhere in excess of 3,000 species" (presumably not including bacteria and viruses) "in motion in the ballast water of ocean-going ships around the world."(126) Recent studies in the Great Lakes have documented a wide variety of organisms, including pathogens, in ballast water and sediment.(127) (Because of the low water quality in various ports around the world where international shipping ballasts up, they may also contain toxic chemicals, although the total levels of discharge are likely to be much less than discharges from terrestrial sources in the Great Lakes.) Somewhere around 720,000 metric tonnes of foreign ballast, most of it an organic soup, are discharged into the Great Lakes each year.(128) Although other vectors from both ships and terrestrial sources need to be more closely examined, it is clear that prevention of ballast water invasions takes the highest priority.
Hull fouling. In addition to ballast water, which has been well documented as a threat, the next most likely shipborne source of exotics is probably hull fouling. This became less of a problem with the common use of steel construction and anti-fouling paints. Unfortunately, the most common anti-fouling paints use organotin tributyltin (TBT), which is toxic to marine organisms (and depends on that toxicity for its effectiveness). These are likely to be banned in the near future, in favor of copper-based coatings and silicon-based paints, which make the surface of the ship slippery so that sea life will be easily washed off.(129) (Copper is also toxic to marine life, but the idea is that it will work mainly by denying the organisms an easy surface to attach to rather than by dissolving toxins into the water.) It is generally believed that the increased speeds of ships (which adds to the ballast water threat) also decreases the threat from hull fouling by helping to wash off creatures not firmly attached, and it may also be much less of a potential vector for the transmission of freshwater organisms.
Marine sanitation devices (MSDs). Marine sewage is a potential source for exotic pathogens. All ships which operate in Canadian or US waters are required to use shipboard marine sanitation devices (MSDs) which are designed to treat the sewage to basic land-side standards. However, perhaps in part because these vessels are not require to use MSDs in international waters, many of these MSDs on third party foreign vessels entering the Great Lakes and other Canadian or US waters are very poorly maintained.
The exchange requirement. The primary defense against exotics in all current ballast water regimes, both voluntary and regulatory, is the requirement for an open ocean exchange. Under the US mandatory regime establishing the exchange requirement for the Great Lakes in 1993, the shipping industry is invited to submit alternative measures to the US Coast Guard for approval, but no one has yet done so. Exchange was something that mariners had done on their own in the past, occasionally, for the purpose of cleaning excessive loads of sediment out of their tanks. Exchange was therefore seized upon by those who developed the first voluntary guidelines in Canada and Australia as a practical measure which could be immediately adopted at very little cost to the industry.
The US mandatory regime which went into effect in the Great Lakes in 1993,(131) generally based on voluntary guidelines previously developed by Canada,(132) requires that the exchange (1) is carried out beyond the exclusive economic zone (EEZ, extending 200 nautical miles from the "baseline" which approximates the shoreline of a nation) in a depth of at least 2000 meters, and (2) achieves a resulting level of salinity in the ballast water equal to or exceeding 30 parts per thousand (ppt). Neither the selection of the EEZ 200-mile line nor the regulatory standard of 30 ppt salinity were based on hard science, and there are problems with both of these regulatory standards.
Defining "open ocean." The 200-mile line only roughly approximates defining the distinction between the coastal and open ocean environments, and may be a particularly poor line to use over the wide continental shelf of Canada off the Gulf of St. Lawrence. The additional requirement of 2000 meters is intended to assure that the vessel is off the continental shelf, but it may allow a vessel to exchange in an area where currents carry the contaminated water into the coastal environment. On the other hand, it could well be that the 200-mile and 2000 meter requirement is too restrictive, depending on the actual currents and coastal environments. A report just submitted to the US ANS Task Force seems to indicate that, with a more precise evaluation of actual currents along the coastline, areas much closer to shore may be safe for exchange.(133) Unfortunately, this analysis does not include Canadian waters. The issue is important, because ships often need more time during the transoceanic passage, or more sheltered waters, in order to conduct their exchange safely.
Salinity. Salinity has proven to be a convenient standard for enforcement purposes, but a poor standard for ensuring that an effective exchange has actually taken place. The salinity of seawater varies in various parts of the ocean from 30 to 39 ppt, but stays fairly close to a mean of 35.3 ppt in the middle of the North Atlantic.(134) Thus, a reading of 30 ppt or more from an exchange in the North Atlantic nominally indicates that 84.98% (30/35.3) or more of a tank previously carrying fresh water has been exchanged. However, an analysis of salinity readings(135) taken by the US Coast Guard indicated that a substantial number of vessels begin with high salinity water (probably from the Mediterranean, a common area for the delivery of grain from the Great Lakes). This clearly undermines the validity of the salinity standard as a guarantee of exchange. Also, it should be kept in mind that a 100% exchange, not 85%, is the goal. The regulatory level of 30 ppt, or a nominal 85% exchange, was purely a practical accommodation for the shipping industry because of the difficulty that many vessels have in accomplishing a 100% exchange, and not because of any scientific basis for saying that 30 ppt salinity, or 85% exchange, provides a critical level of protection.
The logic of exchange. What is the purpose of requiring an exchange of ballast in the open ocean? Contrary to what is often assumed, the main idea is not to salt up the tanks to kill or inhibit the reproduction of fresh water organisms in the ballast. Salting up the tanks is undoubtedly a useful attack against some freshwater organisms. But that is not an effect which can be relied upon, and is at best a secondary purpose of the exchange requirement. There are many organisms in a variety of taxa - the sea lamprey(136) and the cholera bacterium(137) come immediately to mind - which can make the transition from salt to fresh water quite nicely. Moreover, many exclusively freshwater organisms can live in a dormant form while exposed to salt water and become active again when exposed to fresh water. "A surprisingly diverse group of [freshwater] taxa, representing protozoans and 11 animal phyla, possess resting stages which may be capable of surviving extended saltwater immersion (although experimental data for most of these taxa are lacking)."(138) One of the seminal researchers in the field, Dr. James Carlton of Williams College, christened this the "Malinska Effect" after finding the freshwater calanoid copepod Eurytemora affinis still doing quite nicely after living for two weeks in the 30 ppt exchanged water of the motor vessel Malinska.(139) The sediments in the ballast tanks give many organisms a place in which to find shelter.(140)
It does appear, however, that the small pelagic organisms peculiar to the highly saline, high ultraviolet, and highly oligotrophic environment in the open ocean are not suited to reproduction in the low saline, low ultraviolet, and less oligotrophic environment of coastal areas - which are subject to invasion by creatures across the ocean from other coastal areas. (One might presume, this simple sailor would think, that anything floating around live for long periods on the surface of the open ocean which is going to thrive in the coastal areas has already had thousands of years to move in on ocean currents.) Although there can be a good number of organisms in the open ocean, marine biologists have advised that "the probabilities of reciprocal introductions" between the open ocean and coastal environments "are virtually non-existent."(141) So exchange is in fact "contrary ballasting" between two distinct ecological zones, using the fact that open ocean is a natural barrier to invasion (to anything that has not already had plenty of opportunity via natural causes).
Effectiveness of the exchange regime. Compliance with the Great Lakes regime has been generally good. US Coast Guard enforcement statistics collected since the beginning of the regime in 1993 have indicated a steady decline in the number of "problem vessels" having difficulty meeting the regulatory standard of 30 ppt salinity.(142) In essence, the marine community has easily adapted to the regulatory requirement. But the problem, discussed above, is that the salinity standard does not mean much in terms of actual effectiveness of the regime. In addition to the problem of the regulatory salinity standard, there are three substantive limitations to the current exchange regime. These are (1) safety, (2) the problem of residual slop and sediment in the "empty" tanks of "NOBOB" vessels, and (3) the lack of effectiveness of exchange in actually flushing out the organisms. All three of these defects are directly related to the fact that ballast tanks on vessels currently in service were simply not ever designed for the purpose of exchange.
The design of ballast tanks. Conventional ballast systems are built with only one two-way pipe end in the bottom of each ballast tank. Unless some other provision is made for flushing the water through the tank, the only way it can be exchanged is by pumping down a full tank and refilling it. When conducting a pump-down and pump-up or "sequential" exchange, the vessel operators typically do one set of side-by-side (port and starboard) tanks at a time in order to avoid endangering stability. (Even if it were feasible within the limits of the pumping system to pump down all tanks at once, lightering the whole ship at once would create a dangerous instability. That is why the ballast tanks are there in the first place.) Recent stress studies by one marine engineering firm indicate that some ships should instead bracket the tanks port and starboard (pumping one tank in the middle of the vessel on one side and two tanks near the ends on the other side) in order to reduce overall hull stress.(143) But these are all make-do measures to compensate for the limitations in existing tank and pumping design. The tanks could be fitted with a second set of pipes allowing water to be discharged and refilled simultaneously in what is called a "flow-through" or "flush" exchange. The International Maritime Organization has recommended that some existing vessels conduct a flow-through exchange by pumping water up out the top of the tank through the manholes or the vent pipes on the deck.(144) However, while preserving overall hull integrity, this make-do procedure can create dangerous over-pressurization of the tanks.(145) In order to be done safely, it requires some change in the existing piping. There is no general principle of marine engineering or naval architecture which requires there to be only one pipe end connecting to each tank.(146) Before exotic species became an issue in the late 1980s, there was simply no need for another pipe end to be built into the ballast system for each tank.
In addition, getting a good flush of the tank is impeded by a large number of structural members typically lining the sides and bottoms of a ballast tank.(147) This is, to a large extent, a constraint of naval architecture. The hull the of vessel must be supported by a semi-rigid internal structure of steel framing in order to supply overall strength to the hull and prevent the tendency of tubular shells to buckle. The exact design of that internal structure - and thus the degree to which it impedes cleaning out the tanks - varies somewhat among types and specific designs of vessels. Some vessels could accomplish much more effective exchanges, whether sequential or flow-through, with relatively small modifications. Others would require major work or could not be safely modified for that purpose.(148)
These two design problems interact. The structures inside the tanks serve to trap water and sediment when the water rises and falls during a pump-up pump-down cycle. How efficient a process can be created by the addition of piping for a flow-through exchange depends a great deal on the specific configuration of the tank and whether or not the piping dictating the pattern of flow is strategically located. Computer modeling conducted by Brazil indicates that relatively slight variations in the placing of inlets and outlets can cause significant differences in the effectiveness of the flushing action.(149) In addition, there is no design or safety constraint which prevents the installation of small, relatively cheap plastic or non-marine steel piping systems inside the tank for the purpose of cleaning off the internal structure.
Those are only a few of the technical changes which could be made. There is a whole range of available technologies for treating ballast water, including filtering, heat, ultraviolet light, biocides, and shore side treatment of water, some of which may well be economically feasible for particular types of vessels and trades. (See § 3.7 below.) None of them require invention of new technology. However, in the absence of a legal regime requiring such changes, and thus creating a level playing field, there is little incentive for any shipping company to make the required investment.
Breaking ships in half. The sequential method of exchange, dictated by the piping systems on most existing vessels, creates some amount of unavoidable hull stress because of the change in buoyancy in one section of the vessel at a time. The degree to which this creates a safety problem varies with the general design of the ship, the strength of the structural members, the size of the ship, the length-to-breadth ratio, the age of the ship, its maintenance history, and other stresses which may be created by high seas or distribution of the weight of the cargo in the vessel. It is important to understand that hull stress is a chronic problem, particularly with older bulk carriers - related to age, maintenance, cargo loading, and sea conditions - regardless of whether or not those vessels are required to conduct ballast exchanges. Figures from the International Association of Classification Societies (IACS), which has expressed strong concern about hull stresses incurred by improper cargo loading practices, show that around the world from 1983 to 1997 there were 73 bulk carriers lost or written off due to structural failure, and another 40 suffering serious damage.(150) During that same period, there were no losses associated with vessels conducting ballast exchange on the way into the Great Lakes. In 1998, however, a handysize vessel named the Flare, bound for Montreal after the Seaway was closed for the season (and thus not subject to the US mandatory exchange regime), broke in half with the loss of most of the crew. It was an older vessel with a history of prior problems, and was lost in high seas. But it appears that a ballast exchange may have contributed to the disaster. We know that hull stress is a problem. What we do not know is how much, and under what specific circumstances, ballast exchange will contribute to that problem.(151) Most importantly, it does not appear that the masters of the individual vessels are always aware of their safe parameters. The better-managed shipping companies are working to correct that problem.
The fundamental problem to be dealt with, as far as ballast exchange is concerned, is that this is something the ships, ballast tanks, pumps, and piping systems were simply not designed for.(152) In order to make exchange both safe and effective, there must be some changes in those systems.
The infamous NOBOBs. The "NOBOBs" are vessels entering the lakes reporting "no ballast on board" because they contain no pumpable ballast in their tanks, but which carry a considerable amount of unpumpable slop still in these tanks. This is a gaping hole in the protection provided by our current regulatory regime, and is likely to be just as large a problem for any expansion of an exchange regime to areas where vessels make more than one port stop along the coast.(153) Although the concept was suggested some time ago,(154) little serious consideration has been given to the idea of requiring some sort of partial exchange, what is known informally as a "swish and spit," to help clean the slop out of the bottom of the NOBOBs. The NOBOBs typically come across the ocean at or close to their marks (literally, the marks on the outside hull which designate the safe loading limits), with some cargo that is offloaded at Montreal or an earlier Canadian port in order to come up to the Seaway draft limits. They do come up a little along the voyage as they burn off fuel during transit, typically about ten days long. This may provide a few hundred metric tonnes of clearance, but only near the end of the transit. However, a ship would not necessarily have to forego a great deal of cargo in order to add enough margin for a swish and spit, particularly when that is compared to the total cargo being carried.
The effectiveness of exchange. A 1990 study conducted for the Canadian Government(155) after the promulgation of the voluntary guidelines in 1989 confirmed the feasibility and relative effectiveness of mid-ocean exchange as a control measure for vessels entering the St. Lawrence Seaway, but warned that it was far from completely effective. Based on a sampling of 12 vessels following the voluntary Canadian guidelines for exchange, the Canadian Government found that:
Although the absence of live freshwater zooplankton from most saltwater ballast samples indicated ballast exchange to be very useful, their presence in a few cases indicates exchange to be less than 100 percent effective. We calculated effectiveness of ballast exchange using ships originating in foreign freshwater ports and exchanging ballast water in mid-ocean.... Four vessels (33%) carried zooplankton that could live in the Great Lakes. Thus effectiveness of ballast water exchange was 67 percent.(156)The Australians have found exchange to be less effective on the larger vessels calling at their ports, especially for removal of the dinoflagellate cysts which are of great concern to them. Some of their tests "showed that among 32 vessels which explicitly claimed to have exchanged ballast water in mid ocean, 14 were still found to contain significant amounts of sediments, including dinoflagellate cysts."(157) In other words, to make the same calculation, although the basis is not exactly the same, the effectiveness on these larger vessels is 56%. A follow-up study on vessels entering the Great Lakes conducted by the Canadians in 1996, after the promulgation of the 1993 US mandatory regime, confirmed that a large range of invertebrates and bacteria are carried in both exchanged water and NOBOB slop.(158)
The bottom line. The bottom line on the effectiveness of the current exchange regime is whether or not exotic species are getting through. This is difficult to judge, because of the delay between the introduction of an exotic and its detection. But the available evidence does strongly suggest that the existing regime is not effective. The Canadian voluntary guidelines for exchange on vessels entering the Great Lakes, supposedly observed by most of the industry, were first put out in May 1989.(159) In the nine years before then, 1979-1988, there were six new invasions documented.(160) In the nine years since, 1989-1998, there were also six new invasions documented, exactly the same number.(161) Four of these were detected during the period 1990-1991, and it is certainly conceivable that they could have been introduced before 1989 but not detected for several years. That is less likely for the last two introductions documented in 1995 and 1998. (With respect to both sets of statistics, one should remember that many other organisms, especially small ones, may have been introduced but have not yet been detected.) Even more revealing, there are two fresh-and-salt water exotics, the Chinese mitten crab(162) and the European flounder,(163) which were discovered in the Great Lakes in 1994. Because these two species live in fresh water, but require salt water to reproduce, they had to be relatively recent introductions. A careful evaluation of the age of the crab put its introduction to the lakes at no earlier than "late 1989,"(164) after the promulgation of the first Canadian guidelines in May of 1989 (and it may, of course, been much later). The flounder was younger, and "was probably released into Thunder Bay either at the end of the first year (autumn 1993) or at the beginning of the second year of life (spring 1994), probably the later,"(165) either of which was after the beginning of the US Coast Guard mandatory regulations in April 1993. This, combined with the documentation of live organisms in tanks which have been exchanged (and the problem of the NOBOB tanks), makes it clear that much is slipping through the cracks in the current exchange regime.
There is no lack of known technical means to deal with the problem of exotics in ballast water. (In fact, it is beginning to appear that the great variety of technical options is an impediment to selection and implementation of something.) The problem, however, is that it costs something to install those means on ships, especially if they are to be retrofitted on old vessels.(166) Most of the following options are not exclusive. There are many logical combinations and permutations on the theme. All of the following cost estimates should be regarded as very rough, and there are a number of debatable assumptions that go into them. Much more lengthy and detailed discussion of technological options may be found in other sources.(167) Many assumptions go into any cost estimate, and all of the following estimates are debatable. They should be taken as indications of orders of magnitude and benchmarks. Another problem is that many of the estimates bandied about in the literature and informal discussions fail to specify the size and type of the vessel being considered, and are therefore comparing apples and oranges. A study conducted by Pollutech Environmental for the Canadian Coast Guard(168) is one of the few to put the estimates in terms of dollars per metric tonnes, and thus provide a consistent standard for comparison. (The figures given here are in US dollars, rounded off, from these estimates made in 1992.)
Open ocean exchange. It is important to clearly distinguish between open ocean exchange as a current practice, without any retrofitting or redesign of vessels, and open ocean exchange as it could be conducted, in top-down flow-through mode, with such changes.
Filtering. This includes a number of different filtering technologies, including (1) screen filters, (2) media filters (probably only practical for shoreside use), and (3) hydrocyclones.
Ultraviolet light (UV). This is electromagnetic radiation in wavelengths between 4,000 angstroms (near visible) and 40 angstroms (near x-rays), typically produced by lamps which either illuminate a transparent tube, through which the water flows (and may re-circulate), or are set closely together to form a barrier which the water must pass through. Effective penetration is usually measured in fractions of an inch.
Biocides. This includes a large range of chemicals with very different effects.
Heat. Target temperatures of anywhere from 35° to 70° C (95° to 158° F) have been proposed. But there is no consensus on how hot is hot enough.
Ultrasound. Although theoretically interesting, this is not a proven technology in any large-scale application, and therefore neither data on effectiveness nor cost estimates are available. It may also create safety concerns aboard ship, in the form of both crew exposure and increase in the corrosion of steel (caused by the same cavitation effects that kill organisms). I note it here mainly because there has been a recent revival of interest in ultrasound, which was fairly well dismissed by early scoping studies. I suggest that it also serves as a good example of many other ideas which periodically resurface, but are then dismissed again as something which obviously "needs more work."
Shoreside treatment. This is not a separate technology except to the extent that shoreside treatment allows for the use of media filtration and settling tanks which cannot be fitted aboard vessels. But it does create a significantly different management scenario.
Getting serious about the options. It should be obvious, from this quick review of the leading options, that questions of effectiveness and cost depend on what the standard is. How good does the exchange have to be, how low a number of microns must be filtered, how high does the dose of chemicals or UV have to be, or how high does the temperature have to be? Without a performance standard (which should be improved over time) it is difficult to compare these options in a meaningful way. All of the cost estimates are very rough, and should be qualified with two general observations: (1) Costs for installation of systems in new construction will always be significantly lower than for retrofitting existing vessels, sometimes by as much as an order of magnitude. (2) Costs are likely to decline, in general, as systems come into commercial production and the natural forces of competition come into play. In order to encourage such competition, however, it is important that government policy prescribe standards for bottom-line performance rather than particular technological approaches.
The fact that there is still some work to be done to perfect various options, or that there are other interesting but unproven technologies to consider, should not be allowed to obscure the fact that technologically and economically feasible means to deal with ballast do in fact exist at the present time. Both retrofitting for top-down flow-through exchange and shoreside treatment present reasonable means to significantly reduce the threat of new invasions in the immediate future. They may not be the ideal or ultimate technologies, and further work on other options should certainly continue. But that is not a valid excuse for the lack of current action. Moreover, it is illogical to put off changes in new vessel construction in hope of having some wonderfully cheap alternative come on line in the future, because any future reductions in the cost of other options are highly unlikely to overcome the increase in the cost of retrofitting old vessels.