Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
5 Ship Time Costs and Their Impacts How the increasing cost of ship time will affect the types of science done aboard ships. One of the most serious issues facing federal agencies that support shipborne science, ship operating institutions, and science at sea itself is the increasing cost of operating research vessels. Higher ship costs will almost certainly force significant changes in the way U.S. academic research ships are scheduled and used. This issue has been studied in recent years by committees convened by the University-National Oceano- graphic Laboratory System (UNOLS), federal agencies and their advisory boards, and independent commissions (e.g., U.S. Commission on Ocean Policy, 2004; Betzer et al., 2005; McNutt et al., 2005; Collins et al., 2006; UNOLS Fleet Improvement Committee, 2009). Trends relating to the cost of ship time for the UNOLS fleet are examined in this chapter. These include major cost factors and trends, the relationship of research ship scheduling to operational costs, the potential for expeditionary planning, future trends in fleet composition, and ship layups. The impacts of these cost trends are also examined in the context of their effect on research pro- posals and awards, the efficiency of ship operations, and their potential to alter the present operating model. 61
62 SCIENCE AT SEA SHIP TIME COST TRENDS The primary expenses of research ship operation are crew costs, fuel costs, maintenance and overhaul, technical and shore support, and con- sumables. These costs for the UNOLS fleet between 2000 and 2008 are shown in Figure 5-1. Crew and fuel costs are the two largest single com- ponents of total research vessel operating costs, accounting for approxi- mately 50 percent of total operating costs in this period, although the impact of fuel costs on total costs more than doubled over nine years. While âall other costsâ also appears to be a significant factor, it is driven by fleet indirect costs, which are proportional to direct costs. Indirect costs make up between 37 and 46 percent of the categoryâs costs, while the rest of the category (food, insurance, equipment and supplies, travel, shore facility support, and miscellaneous costs) has individual costs of approximately $3 million or less per year. The increase in overall UNOLS fleet costs from 2000 to 2008 was not $30,000,000 $25,000,000 Crew $20,000,000 All Other Costs Dollars $15,000,000 Fuel & Maintenance Lube Oil $10,000,000 & Overhaul Shore Support Staff $5,000,000 $0 2000 2001 2002 2003 2004 2005 2006 2007 2008 Year Figure 5-1â Major cost factors FigureUNOLS fleet, 2000-2008. The categories for the 5-1.eps listed are crew salaries and benefits (dashed gray line), fuel and lube oil (solid gray line), maintenance and overhaul (small dashed black line), shore support staff (long dashed black line), and all other costs (solid black line). The category of all other costs includes food, insurance, equipment and supplies, travel, shore facility support, indirect costs, and miscellaneous costs. This figure includes both estimated and actual costs from ship proposals. In several cases, total operating costs for individual ships are missing. This is most often the case with Local class vessels, and it is not expected to significantly impact the total costs. The 2008 costs associated with Marcus Langseth are not included (data from the UNOLS Office, 2009).
SHIP TIME COSTS AND THEIR IMPACTS 63 $90,000,000 6000 $80,000,000 5500 5000 $70,000,000 4500 Dollars $60,000,000 Days 4000 $50,000,000 3500 $40,000,000 Total Operating Costs 3000 $30,000,000 Total Operating Days 2500 $20,000,000 2000 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 Year Figure 5-2â UNOLS fleet total Figure 5-2.eps operating costs (black) versus number of ship days (gray) (data from the UNOLS Office, 2009). due to an increase in the total number of operating days for the fleet. In fact, ship operating days declined 13 percent from 2000 to 2008, while total costs increased 75 percent (Figure 5-2), meaning that the average cost per ship day doubled in that same period. Should long-term ship costs continue to increase at rates comparable to those of the last decade, it is very likely to pose severe problems for research ship operators and federal funding agencies alike. Crew Costs Crew salaries and benefits are consistently among the greatest cost drivers for the academic fleet (an average of $23.3 million per year from 2000 to 2008; data from the UNOLS office, 2009). Crew sizes on Ocean and Global class vessels are regulated by the U.S. Coast Guard (46 CFR Â§188-196), so there is little room for cost savings through personnel reduc- tions. In addition, UNOLS vessels must comply with new environmen- tal, safety, and security regulations. These new measures have required increased crew training and increased staffing requirements (UNOLS Fleet Improvement Committee, 2009). Operating institutions also face salary competition, especially for marine engineers, from other industries (i.e., cruise ships, offshore oil and gas).
64 SCIENCE AT SEA $5.00 $4.00 Dollars per Gallon $3.00 $2.00 $1.00 $0.00 1994 1996 1998 2000 2002 2004 2006 2008 Year Figure 5-3â U.S. retail diesel prices, 1994-2009 (data from the Energy Information Figure 5-3.eps Administration, August 2009). Fuel Costs Fuel consumption is proportional to the size and speed of the ves- sel. Research vessels, which are equipped with diesel electric drives and typically operated at low speeds, do not lend themselves to significant future efficiencies. The type of research done aboard ship can also affect fuel costs on individual programs. For example, a surveying cruise that maintains a sustained ship speed will use more fuel than a research cruise with short transits between stations. Due to recent market volatility in the price of crude oil (Figure 5-3), fuel costs for the fleet have escalated over the past few years. Between 2005 and 2008, fuel expenses doubled from $9.3 million to $18.6 million per year (data from the UNOLS office, 2009). In 2009, fuel prices dropped substantially from their 2008 highs, but the unstable nature of recent oil prices suggests that fuel costs may remain a significant aspect of total operating costs in the future. However, future market controls or carbon emission legislation could have significant and as yet unknown impacts on the price and rate increases of fuel. Ship Schedules and Fleet Management Research ship schedules are managed through UNOLS with the goal of federal oversight specifically to seek maximum efficiency. Ship day rates (the metric used by supporting federal agencies) are influenced by
SHIP TIME COSTS AND THEIR IMPACTS 65 scheduling because fixed operating costs affecting the daily rate are not proportional to the number of days each vessel is used each year (e.g., costs of full-time crew, scheduled maintenance, shore support, regulatory compliance). Full schedules (referred to as âefficientâ schedules) lower the daily at-sea rate by combining these fixed costs with incremental opera- tional costs over a greater number of days. The present ship scheduling process produces a one-calendar-year schedule, beginning six to eight months ahead. There is an attempt to build full schedules for each ship from an ad hoc collection of cruises that federal agencies indicate are likely to be funded. However, it is nearly universal for ship schedules to change based on the differing funding and decision time scales among supporting agencies, often with late notice compared to the National Science Foundation (NSF) funding time frame (Rose Dufour, personal communication, 2009). This leads to schedules that are not finalized until after the beginning of the operating year, often with some remaining uncertainty. In addition, some flexibility is built into schedules to account for episodic events of scientific interest (e.g., volcanic eruptions, harmful algal blooms). Similarly, flexibility is needed for cruises that are rescheduled or canceled on short notice due to lost or damaged equipment or societal events (e.g., Indian Ocean piracy). Recent years have seen some scheduling delays in an effort to pre- vent ships from being idled (discussed later in this chapter). Dividing scheduled projects among several ships of the same class aims to reduce unnecessary expenditures for partial layups, but also results in an increase in ship rate for the ships that have fewer days scheduled. Operational costs are also impacted by the geographic distribution of the fleet, which does not mirror the requested locations for use. Regional class ships are more closely tied to the location of their operating institu- tion than are larger vessels. The aging Intermediate class presents some scheduling problems because these ships tend to work closer to their home port. Wecomaâs Oregon location leads to a shorter operating season. The close proximity of Endeavor and Oceanus (Rhode Island and Cape Cod, respectively) to each other presents difficulty in creating full, more efficient schedules. This issue is exacerbated by the lack of capabilities on these ships, which makes them less desirable for research cruises. Expeditionary Scheduling One alternative to ad hoc annual scheduling is expeditionary schedul- ing. In this type of scheduling, an announcement is made that a ship will be operated in a specific region during a given time window. Proposals are then sought to use that opportunity. UNOLS does not consider any of its ships presently to be funded in expeditionary mode (Mike Prince,
66 SCIENCE AT SEA personal communication, 2009). With a few exceptions, noted below, expeditionary scheduling may not be desirable for the UNOLS fleet. Within each class, ships are somewhat interchangeable and it may not be advantageous to fix the operating region for future years. The Atlantis (with Alvin) is generally tied to the annual window of opportunity on the Juan de Fuca Ridge and often spends the rest of the year near the East Pacific Rise. If the ship were scheduled in an expedi- tionary mode, a community workshop or panel could decide on other regions far enough in advance to allow for proposals to be submitted, reviewed, and awarded. At present very few proposals for Alvin are sub- mitted or funded for work beyond the Juan de Fuca Ridge or East Pacific Rise. When such proposals are submitted, many state that either Alvin or Jason can be used, allowing schedulers to send ships equipped with Jason to these regions. As another example, the seismic vessel Marcus Langseth could possibly benefit from some expeditionary scheduling. Currently, the schedule is dominated by previously funded seismic work. Future yearsâ schedules may require orderly movement from one area to the next to lower transit costs and time expended. As with Atlantis, a community workshop could provide recommendations for the most efficient use of the ship. Fleet Composition and Science Impacts The ship replacement and retirement plan outlined in the 2009 UNOLS Fleet Improvement Plan will reduce the academic research fleet by nearly 40 percent by 2025 (Figure 5-4; UNOLS Fleet Improvement Committee, 2009). The projected retirement of three Global class ships reduces overall ship sizes and could produce overall fleet economies. However, Global class vessels are presently the most heavily subscribed. Chapters 2 and 3 conclude that there will be increased demand for the large research vessels with their deck loading, berthing, and sea state capacities. The new and planned Ocean class ships are significantly less capable than the Global class in terms of deck loads and berthing. Accommodating heavy deck loads and large science parties on Ocean class vessels would require scheduling extra legs, leading to more time in port and a greater number of ship days per research mission. In addition, the current Ocean class ship, Kilo Moana, has a day rate that is comparable to the Global class (Figure 5-5). Thus, if day rates for the planned Ocean class vessels are similar to Kilo Moana, total operating costs for 2025 will not decrease. Furthermore, with the planned addition of Ocean class vessels, there will be fewer vessels able to support the widest-ranging, most resource- intensive marine science research programs of the future and the decrease
SHIP TIME COSTS AND THEIR IMPACTS 67 BUILT/ 2008 2009 2020 2022 2025 2023 2024 2021 2010 2016 2012 2013 2015 2018 2019 2014 2017 2011 SHIP/CLASS* Conv GLOBAL CLASS Melville 1969 Knorr 1970 Thomas G. Thompson 1991 Roger Revelle 1996 Atlantis (Submersible Support Ship) 1997 Marcus G. Langseth (Seismic Ship) 2008 Total Global Ships 6 6 6 6 6 6 6 5 4 4 4 4 4 4 3 3 3 3 OCEAN CLASS Kilo Moana 2002 ARRV 2014 OC #1 2014 OC #2 2015 Total Ocean Ships 1 1 1 1 1 1 3 4 4 4 4 4 4 4 4 4 4 4 INTERMEDIATE CLASS Seward Johnson 1985 Wecoma* 1976 Endeavor* 1976 Oceanus* 1976 New Horizon 1978 Total Intermediate Ships 5 5 5 2 2 2 2 2 1 0 0 0 0 0 0 0 0 0 REGIONAL SHIPS Point Sur 1981 Cape Hatteras 1981 Atlantic Explorer 2006 RC #1 2013 RC #2 2015 RC #3 2017 Total Regional Ships 3 3 3 3 1 2 2 3 3 4 4 4 4 4 4 4 4 4 REGIONAL/COASTAL SHIPS Robert Gordon Sproul 1981 Pelican 1985 Walton Smith 2000 Hugh R. Sharp 2005 Total Small Regional/Coastal Ships 4 4 4 4 4 4 3 3 2 2 2 2 2 2 2 2 2 2 LOCAL SHIPS Urraca 1986 Savannah 2001 Blue Heron 1985 Clifford Barnes 1966 Total Local Ships 4 3 3 3 3 3 2 2 1 1 1 1 1 1 1 1 1 1 Total Ships 23 22 22 19 17 18 18 19 15 15 15 15 15 15 14 14 14 14 * Ships approaching the end of their projected service life will be evaluated to assess their condition. Figure 5-4â The UNOLS fleet projected service life time line (adapted from Figure 5-4.eps UNOLS Fleet Improvement Committee, 2009; used with permission from UNOLS).
68 SCIENCE AT SEA $50,000 $40,000 Global (6) Ocean (1) $30,000 Intermediate Regional and (5) Regional/ $20,000 Coastal (6) $10,000 $0 Figure 5-5â The average daily operating rate by UNOLS class, 2008. The number of ships in each class is in parentheses after the class name. For example, the In- termediate class has five ships (data from the UNOLS Office, 2009). Figure 5-5.eps Re-sized and fixed type in overall fleet size will create greater difficulty in scheduling multiship operations. An additional consideration is the cost per science berth. Global ves- sels, although the most expensive ships of the fleet with the highest day rates, are in high demand and have heavy usage (UNOLS Fleet Improve- ment Committee, 2009). Thus, as discussed in Chapter 4, the Global class ships are less expensive than the Ocean and Intermediate classes when their day rate is divided by the number of science berths aboard. This economy of scale suggests that there are advantages to building larger ships with more science berths and more deck and payload space. Scheduling General Purpose Ships with Specialized Facilities The specialized ships (Atlantis and Langseth) create their own sched- uling niches because they fulfill research missions that cannot be accom- modated by general purpose vessels. The specialized facilities on some of the general purpose Global class vessels (i.e., long coring facility on Knorr, sonar system on Revelle) also attract science missions. Increasing sched- uling efficiency for the general purpose vessels may lead to difficulty in scheduling cruises that require specialized facilities. Some redundancy may have to be built into these ship schedules.
SHIP TIME COSTS AND THEIR IMPACTS 69 Idle Periods And Layups All research ships have idle periods when they are not carrying out work at sea. Required maintenance, training, and inspection activities are part of this idle time and are included in the shipâs normal sched- ule. Factors that affect idle periods on research ships include planned major maintenance periods, variations in funding levels for major field programs, uncertainties in federal research budgets, inconsistent busi- ness relationships with nonfederal users, and seasonal demands. Some excess ship capacity is occasionally used in each class to handle planned maintenance periods, which require fairly long downtimes. Global class ships carry larger crews, have a wider variety of maintenance and repair equipment, and often have longer transits, so there is opportunity to carry out more routine maintenance at sea than on smaller vessels. Large field programs, such as the Joint Global Ocean Flux Study (JGOFS), Climate Variability and Predictability (CLIVAR), and Global Ocean Ecosystem Dynamics (GLOBEC), also affect idle periods (Mike Prince, personal communication, 2009). These types of programs require ships to be in specific locations at specific times and can also require mul- tiship operations, increasing the difficulty of creating efficient schedules for other funded programs. At-sea support required for major programs in the future, such as the Ocean Observatories Initiative (OOI), may place even heavier demands on the fleet in certain locations in the future. Federal budget uncertainties in recent years have made efficient ship scheduling and full utilization even more difficult. Some federal agencies that support ship use are forced to withdraw their already scheduled requests or do not place ship time requests until after annual research ship schedules are determined. Nonfederal users can help strengthen ship schedules, but are subject to restrictions including Navy approval. Ship schedulers tend to seek out these partners only when openings appear in schedules, creating an inconsistent and unsatisfactory business relationship. Additionally, outside users are not held to a contract, which allows them to withdraw at late dates and create further holes in ship schedules. Seasonal demand also affects lay-ups. Peak ship demand during late spring to fall time frame normally comes close to fully utilizing the fleet (Figure 5-6), but winter demands are reduced especially for those ships operating in areas with rougher winter weather.
70 SCIENCE AT SEA 500 450 2004 400 2005 Operating Days 350 2006 300 250 200 150 100 50 0 Jan Feb Mar Apr May June July Aug Sep Oct Nov Dec Figure 5-6â The UNOLS fleet seasonal utilization trends from 2004 to 2006 Figure 5-6.eps (adapted from UNOLS Fleet Improvement Committee, 2009; used with permis- sion from UNOLS). THE IMPACTS OF INCREASING SHIP COSTS Agency Proposals Requiring Ship Support The most direct impact of rising ship costs on science has been a decline in the total number of funded ship days from 2004 onward (Figure 5-2). NSFâs annual budgets, including those for divisions that support research requiring the oceanographic fleet, have been funded at levels below the inflation rate. This is a serious mismatch to ship costs, which at the same time have risen at more than three times the inflation rate. Despite the rising number of days at sea requested in proposals (Fig- ure 5-7), NSF could not maintain its non-ship-related research without reductions in the number of ship days funded each year (UNOLS Fleet Improvement Committee, 2009). An initial agency response was to defer ship time into future years to meet shrinking federal budget levels, with the hope that future budget increases would provide relief. In 2005, more than 500 ship days were deferred to future years. Proposal success rates also factor into ship demand. The success rate of proposals in the NSF Ocean Sciences Division (OCE) is higher than the overall NSF success rate (Figure 5-8). However, the rate of successful proposals with ship time has declined, while the overall proposal success rate within OCE has increased. For example, 33 percent of 1349 proposals submitted to OCE in 2008 were funded, whereas only 21 percent of 255 proposals with ship days were successful. This leaves an obvious imbal- ance in the number of ship days funded versus the number of ship days requested, as shown in Figure 5-7. While some OCE proposals related to the UNOLS fleet but not including ship time (including those that sup-
SHIP TIME COSTS AND THEIR IMPACTS 71 Days Requested 14000 Ship Days Funded 12000 Available Ship Days 10000 Ship Days 8000 6000 4000 2000 0 2001 2002 2003 2004 2005 2006 2007 Figure 5-7â UNOLS ship time demand versus days funded, 2001 to 2007. The Figure 5-7.eps number of days requested is shown in gray, the number of funded ship days is shown in white, and the number of available ship days is a black line (adapted from UNOLS Fleet Improvement Committee, 2009; used with permission from UNOLS). 35% 30% Percent Awarded 25% 20% All OCE OCE with Ship Overall NSF 15% 2002 2003 2004 2005 2006 2007 2008 2009 Year Figure 5-8â The success rate of proposals submitted to the National Science Figure 5-8.eps Foundation (NSF) overall, to the Ocean Sciences Division (OCE), and to OCE that have requested ship use (data from NSF, 2009). port ship operations, marine technicians, and other facilities) have very high success rates, the low rate of successful proposals with ship time is detrimental to maintaining the UNOLS fleet.
72 SCIENCE AT SEA Efficiency of Ship Operations One common metric of efficiency in ship operation is the fullness of each ship schedule. Projected reductions in fleet capacity due to ves- sel retirements implies that if the present level of ship days funded per year is maintained, the fleet would be fully supported by approximately 2011 (Figure 5-9). A more difficult measure of efficiency is the productiv- ity of each ship day. Anecdotal accounts suggest that funding agencies and the research community are more effectively linking and combining programs to make the best use of ship time. The rise of multidisciplinary, multi-investigator programs discussed in Chapter 2 implies continued progress in this measure of efficiency. It is worth noting that greater efficiency in operations and schedul- ing does not imply cost decreases. Working efficiently may actually mean working more expensively. For example, costs per day could escalate quickly on a ship that is simultaneously operating several AUVs while also involved in other types of sampling or data collection. More deck, bridge, and technical support personnel may be required to assist with the various operations, contributing to a higher daily operating rate. In addi- 7000 Ship Days Funded 6000 5000 Ship Days Available 4000 3000 2000 1000 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 Figure 5-9.eps Figure 5-9â The difference between UNOLS funded ship days and the ship days available for use, 2000-2025. Funded ship days are shown in gray, while avail- able ship days are shown with a black line. This graph assumes that all ships are capable of being fully utilized to support science operations. Realistically, some of the older and less capable vessels are likely to continue to be undersubscribed (adapted from UNOLS Fleet Improvement Committee, 2009; used with permis- sion from UNOLS).
SHIP TIME COSTS AND THEIR IMPACTS 73 tion, multi-investigator cruises may lead to more idle time for scientists onboard, increasing science team costs. Alterations to the Present UNOLS Model The present model of UNOLS funding and operation is sensitive to the robustness of fiscal support. If future federal budgets supporting ocean science increase at the same rate as ship operation costs, the com- munity can retain the flexible scheduling and excess capacity that has worked in the past. If not, UNOLS and the ocean research community will be required to use existing resources more effectively (Mike Prince, personal communication, 2009). Because the scheduling process already attempts to maximize efficiency, further costs savings may require a lon- ger time horizon for planning. Although budget issues in the past few years have caused ship scheduling to occur later in the year, there is growing pressure to schedule the Global class ships significantly earlier than is the current practice. Another trend may be toward increasing the flexibility of cruise tim- ing, especially if it requires specialized equipment supported by only one ship or involves work in remote locations. For these areas, already-funded research projects may be deferred to a later year when there is enough demand. This can cause difficulties for research programs that require repeat surveys, recovery of deployed instruments, or significant interna- tional cooperation. A third option is to average the fleet costs over a multiyear period. This would stabilize the day rate for a certain amount of time and increase the possibility of working with nontraditional funding sources. Finally, it may be worthwhile to investigate the possibility of home-porting ships in common locations to take advantage of cost savings and provide geo- graphic proximity to research areas. CONCLUSIONS Due to insufficient funds to support research on increasingly expen- sive ships, the number of ship days requested is rapidly outpacing opera- tional days. Crew and fuel costs are likely to continue as significant factors in total operational fleet costs. The push for more efficient ship scheduling may lead to longer lead times for research projects and reductions in the ability of the future fleet to accommodate late-breaking scientific and funding opportunities. Present trends in science and technology indicate further growth in major research programs requiring significant ship resources. The increasing cost of ship time and economies of scale may
74 SCIENCE AT SEA lead to greater use of Global class UNOLS vessels, which are capable of simultaneously carrying out multiple science operations. Complex pro- grams are less likely to require multiple legs, lowering operational costs, if put on the largest ships of the fleet. The reliance on Ocean class vessels in the current fleet renewal strategy probably will not lead to a future fleet with reduced operational costs, but may lead to a fleet with fewer capabilities.