Planning

In this chapter, I'll try to describe those elements that occur before the cruise, at least those that are not instrumentation specific. These are described under the general title of planning, but these pre-test activities are wide ranging in scope. Despite the breadth of planning activities, this entire chapter can be summed up by a single rule:

This rule may be short, but it is anything but simple. It requires years of experience to know what to plan and how to anticipate the unexpected. I'll describe those items which I have come to think of as important, but in the end, you'll have to learn for yourself. Still, I hope you will consider my comments carefully, because planning is the heart of conducting a good experiment.

Planning needs to begin with the scientific ideas and goals that you want to accomplish on your cruise. These should be expressed within a science plan for the experiment. An explicit elaboration of these goals is important in nearly all experiments.

You may enter into collaborations with other scientists, or you may be on your own, but you will inevitably be working within a team. Your role in the team, what you will contribute and what you expect from others, should be explicitly agreed to before the experiment. A discussion and agreement, prior to the experiment, on the roles everyone will play will go a long way towards avoiding misunderstandings during and after the experiment.

Coordination of action during the test is critical. A written test plan can provide the common ground to insure that everyone understands what is going on, and works together towards the common goal of a successful experiment. The test plan should cover a variety of topics including organization, test conduct, communications, and navigation.

If your test involves multiple platforms, or relies on some form of data exchange with shore facilities, then communications planning should be a high priority. Communications always seems to be the first thing to break down under field conditions. While this may be surprising at first, there are good psychological as well as physical reasons for this that will be discussed.

I have never been on a cruise where we didn't need to know where we were and what time it was. With modern satellite-based systems this should be easy. Still, mistakes are constantly made and it has been my experience that the navigation and timing on research cruises are generally inadequate. In this day and age navigation and timing are not hard to do, but they do deserve careful planning and attention. I know that providing for navigation and timing sounds incredibly mundane, but it can turn out to be of critical importance.

Finally, I'll discuss the truly trivial matters of what to pack and how to get it onto the ship. This should not require a Ph. D., but a few tricks that I have developed over the years should make this a more enjoyable experience. (Actually this is a relative statement. Packing and loading a ship is always a crummy job, but someone's got to do it.)

Scientific Planning and Collaborations

Scientific planning and goals

In planning an experiment the very first thing is to settle upon a set of experimental goals. Without a clearly written set of goals, experiments grow and transform in obscure ways, which can lead to a failure to accomplish anything. A clear vision of the reasons for an experimental program is needed, and this vision needs to be shared among all of the participants.

Sometimes, due to politics or funding constraints, the participants and instruments will be settled upon before the goals of the experiment are set. This is unfortunate, because invariably, one could choose a better set of collaborators to perform the desired measurements than those selected by a sponsor. When this happens to you, as it likely will at some point during your career, my advice is to make the best of it. Make the effort to determine exactly what expertise, instrumentation and research interests each party can bring to the experiment. Meshing these into a single unified whole may be difficult, but it is well worth the effort. If you fail to try to accommodate investigators selected by your sponsor, then the slighted investigators will complain to the sponsor, who will then make their unhappiness evident to you. Unless the stipulated investigators are total bozos, try to get along.

The more normal and productive course is to select investigators and capabilities based on the stated scientific objectives and goals. While it is important to select people that can work together, the usual criteria are based almost entirely on expertise and capability.

[Keeping this in mind will help you when you are deciding on future research areas. When adding capability, it may be better to complement existing capabilities within other research groups rather than duplicating those capabilities. There are all kinds of interesting areas in my own field that I avoid, because I know that other groups around the country provide a sufficient capability in those areas to meet potential sponsor's needs. I think it better to move into areas where I see a niche that is not filled. The great thing about oceanography is that there are lots of such niches to be filled and lots of areas where our ignorance exceeds our wisdom. It may be a funny reason to recommend a science, but I think it is our ignorance within oceanography that makes it such an interesting field of study.]

Collaborations, contracts and agreements

Typically, the investigators you will be working with are respected, intelligent scientists with solid research programs. While some investigators prefer to work alone, most will welcome collaborations with others. By eliminating duplication of efforts, successful collaborations allow limited research dollars to be stretched, benefiting everyone involved. In contrast, unsuccessful collaborative efforts can degenerate into the worst kind of bedlam, engendering bad feelings between those involved, hindering the advance of science. I have been involved in both good and bad collaborations over the years and there are a few rules that I have learned.

First, collaborations among newly introduced scientists are often tentative and prone to failure. When I was younger I assumed that everyone shared my enthusiasm for collaboration and my views on sharing data. I quickly found out that some scientists were wary of my openness. They didn't know me and thus had no reason to trust me. In a competitive world such as science, where money depends on both your ideas and your success in selling them, there is a natural tendency to be wary of strangers. I have learned that it is best to take collaborations slowly, stressing the advantages of the collaboration to all parties concerned.

The second key is to form an explicit contract describing what each party will do for, and receive from, the other parties in the collaboration. This agreement should be reached as early in the collaboration as possible. I am not suggesting that everyone hire lawyers, but everyone should clearly understand what is involved in the collaboration. A colleague of mine once collaborated with a scientist at another institution. Going into the collaboration they agreed that both of their names would appear on any papers resulting from their joint work. The first paper, which my colleague wrote, went smoothly. A disagreement then arose over the authorship of the second paper. My colleague took the typical view that the person who wrote the paper should be first author. His collaborator, who was a good scientist but not much of a writer, felt that he should be the first author on the second paper, no matter who wrote it. He reasoned that he had been the second author on the first paper, so he should be the first author on the second paper. While most would view this as an unreasonable request, the collaborator didn't feel it was and so a problem developed in the collaboration. My colleague would have been better off if he had known and discussed his collaborator's expectations before he entered into the collaboration.

A good collaboration should be like a partnership, with all involved contributing something to, as well as taking something from the partnership. On one occasion I suggested to a colleague the use of one of my current meters as part of his measurement platform, a towed catamaran system with lots of other instruments. At first he was nervous that I was trying to get some free credit for latching on to his already successful program. After convincing him that I was not interested in taking credit for his accomplishments, he agreed to the scientific utility of adding my current meter. The next hurdle came several weeks later after a planning discussion, when I brought up the topic of data sharing. My colleague had somehow gotten the impression that I was going to provide my current meter without any strings attached. I corrected him, suggesting that we fully share all of the data from his instruments and my current meter, with the proviso that I would not use his data for anything other than the work with the current data. Furthermore I suggested that all of our initial work and publication be joint, at least on those aspects of the problem again involving my instrument. I thought this was a reasonable and fair proposition. Again he was wary, but after thinking about my proposal for a few days he accepted. From these rather tentative beginnings our work together turned out to be an extremely pleasant collaboration. The key was that it was based on trust, mutual respect, and a clear understanding of what was expected of each participant.

In contrast, I entered into another collaboration with a scientist to share certain data for an intercomparison of systems. It later turned out that the scientist that I reached an agreement with was but one of three scientists who were working with these data, and the other two were not thrilled to have me as a working partner. (I think they were concerned about spreading the credit for their work to thinly.) It eventually worked out because they wanted my data as much as I wanted theirs, but the tensions ran high for quite some time. I had to endure a number of angry phone calls and long discussions about my role in the collaboration. While some good work has come out of this effort, in retrospect, I doubt if I would have entered into this collaboration knowing the difficulties involved. Entering into a collaboration is just like buying real estate - it is best to know exactly what you are getting into before you sign on the dotted line.

Science Plan

Any experiment that involves more than one investigator needs a clear and detailed science plan. Such a document is normally required by sponsors, but is useful even when it is not required. A typical plan would consist of an introduction outlining the scientific problems to be investigated. A background section would detail pertinent knowledge in the field and point out the needs for additional data that this program will supply. The scientific objectives would then be described in clear and concise terms. The next section usually describes the general methodology of the program, listing investigators, instruments, platforms and program organization.

This document is not the same as a test plan, which contains detailed information regarding the conduct of the experiment. It instead provides an overview of the program goals and approaches that should be useful to both your sponsors and fellow investigators. The process of planning an experiment can be likened to the planning of a symphony orchestra's performance. In this analogy, the science plan is like the evening's musical program that is decided upon by the artistic director of the company. It is the document that provides the overall direction and sets the common goal for the performance. In contrast, the test plan is like the musical score. It specifies in great detail how the performance is to be carried out, who is to play which notes, and exactly at what time. This analogy can be carried further. The director is the chief scientist, the orchestra members are the other scientists, the custodial and support staffs are the ship's crew, and the paying audience are the sponsors! (I can hear my colleagues now, quipping that my role in this orchestra would be to play the Bassoon, spelling it either with an 'ass' or a 'uff'.) In fact, the only major difference that I can see between an oceanographic experiment and a symphony orchestra is that the orchestra's test plan has usually been written by a guy who's been dead for several hundred years.

The science plan should act to document the scientific intent of your program providing a focus for subsequent test planning. Getting all of the co-investigators to agree to a science plan is an important step in getting everyone to work together to meet the goals of the experiment. Experiments conducted without a science plan are likely to degenerate into a set of separate investigations, with everyone making their own measurements in support of their own research. While such disjoint experiments can yield good results, they are never as productive as a carefully coordinated program. The ideal is to get everyone thinking and working collectively and towards a common goal.

Assigned Responsibilities

It has been my experience over the years that technicians should not be sent to sea to run scientific equipment by themselves. Technicians may be trained to operate equipment, but only a scientist can do science; and science is the ultimate goal of measurements at sea. This observation may seem a bit out of place, here in the chapter on planning, but it is but one element of a larger point I want to make about assigned responsibility.

I was once on a short cruise investigating near-surface internal waves. The scientist responsible for the CTD measurements on this cruise decided to attend a professional meeting and sent a well-trained technician in his place. In the end I was impressed with the technician's knowledge and technique with the system. His sole problem was that he didn't understand the scientific objectives of the experiment. The technician had been given enough cable to do CTD profiles to a depth of 1000 m, and so that is what he intended to do. Never mind that the mixed layer depth was less than 50 m. Never mind that the internal waves of interest were essentially confined to the upper 100 m or so. He had 1000 m of cable and he was going to use it. Given the fixed speed of the profiling this was a ridiculous waste of time, but the technician didn't seem to mind. On this experiment, the time necessary for a single deep cast could have been better spent performing multiple casts in the important near-surface zone. Any scientist would have recognized this tradeoff and modified the sampling strategy accordingly. The technician stubbornly stuck to his strategy despite my protests to the contrary.

(The one good that came from this debacle was that I got to shrink a large number of styrofoam cups to hand out as souvenirs to friends and acquaintances. In case you are ever involved in a cruise where deep casts are to be made, be sure to take along a mesh laundry bag and a bunch of 8 oz styrofoam cups. When the cups are sent down to depth in the laundry bag that is attached to the instrument, the pressure crushes them surprisingly evenly, shrinking them to the size of a thimble! You can even write cute messages on them before they are sent down, and your writing comes back as small as the shrunken cup. I use a full size cup, along with a shrunken cup, in talks that I give to elementary school kids to get them to think about what it is like on the bottom of the ocean. As a hint, it is best to send the cups down stacked. They are a bit more difficult to get apart this way, but they don't distort as much.)

The larger point here is that each instrument system, whether it be a simple CTD or a complex manned submersible bristling with sensors, should be the responsibility of a single scientist. This is not to say that one scientist cannot handle more than one instrument. And it is not to say that some instruments aren't so complex that they require several scientists for their care and feeding. The point is that in the end, a single person has to take responsibility for each instrument and measurement that is taken. By identifying a single scientist as responsible for an instrument, you are assuring yourself that either that scientist will make certain the instrument is used with the proper care to make the desired measurements, or that scientist's reputation will suffer. Without such an assignment of responsibility, the details that are the difference between good and bad measurements are too often lost between the cracks. Henry W. Menard, a famous geological oceanographer, said [Menard, 1969], "Experience shows that we rarely get first-class records unless the scientist who will eventually study them is on board the collecting ship or at least personally involved in the expedition."

Graduate Students At Sea

If my opposition to technician's working alone at sea isn't bad enough, I'm going to compound the effrontery and suggest that the same prohibition should stand for the majority of graduate students. In general, I don't think graduate students should be sent to sea without the presence of their major advisor, or at least someone on their committee or within their department. As this book is directed in large part to graduate students, I have probably offended you at this point. Before passing judgment though, please allow me to explain.

I just recently returned from a deployment on the Research Platform Flip. There were two pairs of graduate students on Flip that had been left there by their major professors. One pair worked hard to keep their equipment going, they showed significant interest in learning about the other research projects, and were always happy to lend a hand when some work needed to be done. Their professor had trained them well.

The problem was with the other two students, who could have been the first two student's evil twins. Unfortunately, problems with their equipment left them with little to do. They responded in the worst possible way by doing nothing during the cruise. They showed no interest in anyone else's work. They rarely offered to help anyone even though they were constantly in the way. To make matters worse, one of them, who had never been to sea before, repeatedly violated the rules of life on Flip. He managed to flood the scientist's sleeping area not once, but twice, by forgetting to turn off a toilet valve. Despite an explicit proscription, he repeatedly took showers during meal times, making it difficult for the rest of us to wash our hands on the way to the mess. When you're on a platform as small as Flip for any length of time, these kinds of transgressions take on major proportions. By the time we got off Flip, all of us were either mad at, or disgusted with, these two.

I later found out that the novice student was sent to sea by his professor to satisfy his department's cruise requirement. Prior to this experience, I was all for cruise requirements in general, but as far as I can see this particular student gained nothing from his experience, while hindering the work of the rest of us. I don't think this is what the department had in mind when they passed this requirement.

Flip is a 110 m long cross between a spar buoy and ship. Flip consists of a 90 m long tube containing a large floodable sea water tank with 20 meters of a ship's bow welded to one end. Flip is deployed by towing it out to sea in the horizontal position - it has no propulsion system of its own. Once Flip has arrived at the desired location, the sea water tank is flooded and Flip begins to sink. When sufficient water has entered the tank, the aft end of Flip actually sinks completely, leaving the forward compartments sticking straight up out of the water.

This unique design makes Flip an incredibly stable platform from which to work. As the crest of a wave passes a normal ship, the ship heaves upwards in response to the force exerted by the extra volume of water it displaces. The ship is accelerated by this force, proportional to the volume of water displaced, acting on the mass of the ship. When Flip is in the horizontal position, its wetted area is roughly its length times its beam or width. When it is in its vertical position though, its wetted area is reduced to pi times the square of the radius of its body. Its mass is actually larger in this vertical position, because of the water it has taken on as ballast, and its wetted area is much smaller, making it extremely stable.

This is not my only example of a wasted graduate cruise. When I was in graduate school, a professor in my department sent a student on a cruise to the equatorial Pacific by herself, despite the fact that the student was working on a computational degree. Upon the student's return, she was required to give a seminar on her experience. I remember her slide show vividly. She began with pictures of the boat. (Her terminology, not mine.) Then there were some pictures of her clowning around with the new friends she made on board. Then there was a picture of her sunbathing on the steel beach above the wheelhouse. Each picture was accompanied by a small story. Finally, she showed a picture of a Rosette sampler dangling from a winch. She clicked by this last picture quickly though, commenting that occasionally they would stop to lower this instrument over the side. When someone asked her about the instrument she admitted that she had no idea what the thing was. As far as I could see, her trip was an all-expenses-paid cruise and she had not cluttered her mind with any of the research work that was going on around her.

Now I know that there are countless counter examples to the last two stories. Many students are mature enough to go on a major cruise without guidance. But many more are not, and I don't think valuable cruise time or space should be devoted to the unprepared or unmotivated. This is not a diatribe against grad students. I was once one myself. Instead it is a plea that students be properly supervised on cruises to allow them to get the most out of the experience.

Test Plan

Arguably, the single most important document to the success of an experiment is the check from the sponsor for the full amount of the cost of the experiment. Given that the check is in the mail, as it always it seems to be, the next most important document is the test plan. As I said earlier, without a science plan the test may degenerate into a series of independent tests, one per investigator. In contrast, a test executed without a test plan often degenerates into a dangerous babble of disagreements, arguments, and name calling. A test without a test plan is not a pretty sight.

There are a wide variety of levels of detail found within a test plan. A good plan will include all of the information for a participant to know what, where and when things are supposed to happen during the test. It should provide a guide for individual investigators of their responsibilities during the test. It should provide sufficient information to insure the coordination of all of the participants in the test, especially across platforms, and from land to sea. In most cases it is a supplement to the communications that are a necessity in any test, providing a reference point about which all of the participants can communicate.

I have been spoiled during my years at APL because of the complete test plans that our test group requires that we produce. A typical APL test plan contains an incredible amount of detail. They typically begin with a brief statement of the scientific objectives of the experiment. The second section will contain an overview of operations, beginning with the general and becoming more specific. This overview will list and describe the platforms (buoys, ships, aircraft, or satellites) involved, along with the major instrumentation. It will also describe the general location and timing of the experiment. Following sections, one per platform, then go into more detail about the instrumentation and duties of the scientists on each platform. Finally a number of sections at the end of the plan detail the navigation requirements, any peculiar safety issues, contingency plans in case of bad weather, instrument failures, etc., and an overall communications plan. For an example of the level of detail specified, the communications chapter will provide call signs for all participants, primary and backup frequency allocations, pertinent phone numbers of land-based participants, and details of communications schedules which each platform is expected to follow.

Such a test plan will specify how a test is to be conducted while acknowledging the flexibility that may be required in the field. A typical plan may specify a number of run 'modules', specifying specific tracks and instrument deployment schemes to perform a particular measurement. A prototype schedule of such modules may be given, but the flexibility of final module scheduling is left to the ship's chief scientist. A good test plan allows the experiment to proceed in an orderly fashion while providing the scientists the freedom to make changes if necessary.

Putting together a test plan of this detail is a lot of work. I have been out on experiments that do not have this sort of detail in their test plans. Some have worked well and others have not. It seems to me that a high degree of planning before the test removes considerable risk in actually implementing the test. With a detailed and well- thought-out test plan, individuals can follow the plan without much effort. Without such a plan the success of the experiment relies on individual initiative and numerous decisions made in the field by the participants. While each individual may make good decisions for their own work, there still may be gaps left in the program that could have been avoided through better coordination.

As an example, I was involved in a minor capacity in a multi-national experiment to study radar backscatter from the ocean surface. The planning for this experiment was undertaken by a committee of the senior investigators, but the test plan they produced lacked detail in some areas. After reading over the plan, I asked the principal investigator on the U. S. side whether anyone was bringing a rain gauge to the experiment. He said that he was certain that our foreign colleagues were providing this capability, even though it was not spelled out in the plan. Well, sure enough, when we got out to sea, there was no rain gauge. In the end, we had to rely on hourly notes from the watch that recorded whether it was raining or not. Meanwhile, two low-cost, easy- to-use rain gauges were sitting 4000 km away, packed in boxes, back in my lab.

I hope I have convinced you of the importance of the test plan. To a certain extent though, this is obvious. The real problem lies in figuring out the right things to do in the test plan. While each cruise and experiment will be different, some of these topics are discussed in general terms in the following sections.

Coordination

Chain of command

The chain of command within the experiment as a whole, and on each platform participating in the experiment, needs to be understood by all participants. While democracy may be a good organizing principle for a country, it would be a disaster on an experiment. Inevitable disagreements occur and decisions need to be made quickly. Democracies would never do for work at-sea. Instead most at-sea expeditions are run as a benevolent dictatorship. (Although some of my colleagues would argue with my use of the word 'benevolent' in this context!) The chief scientist is in charge and his word is the law. Despite this power, a good chief scientist will gather everyone's opinions and weigh the good of the experiment before coming to decisions. Hence my preference for, and use of, the term 'benevolent'.

Actually, that is not quite the case on an APL expedition. The responsibilities for the cruise are split amongst several people, but this division of responsibility is explicit. As for all cruises, the captain of the vessel is in charge of all issues of safety and navigation of the vessel. The chief scientist is in charge of all scientific matters. And an APL deck chief is in charge of all deck operations. The chief scientist may request a buoy deployment, but the captain or the deck chief can veto the deployment on safety grounds. The system works because the individual's responsibilities are well delineated and separate. As a working scientist on board a vessel, it is your responsibility to respect this chain of command, and to address your requests, complaints or suggestions to the chief scientist. Don't directly ask the captain or the deck chief to do things. By respecting the chain of command you'll keep confusion down to a minimum and will keep the chief scientist informed.

I have been on cruises where scientists bypass the chain of command and the results can be a mess. I have seen scientists that have been banned from the bridge for making requests directly to the crew, countermanding standing orders of the chief scientist and captain! The worst case I ever ran across was on a large experiment that I was running, involving several aircraft, a helicopter and three ships. In the initial planning stages, the scientists on one of the aircraft, a four-engine Navy P3, had requested some flight time at altitudes of 300 feet in order to test out a new infrared remote sensor. Since this system was not directly germane to the sponsor's program, and because the helicopter was scheduled to operate from 200 to 600 foot altitudes, their request was denied. The test plan was quite explicit in the restriction that the helicopter was to stay below 900 feet and the P3 was to stay above 1200 feet.

On the first day of testing, the experiment had been underway for about an hour, when I heard the P3, communicating directly with the helicopter pilot, request that the helicopter climb to an altitude of 1500 feet. The helicopter pilot, not wanting to play chicken with a four-engine fixed-wing aircraft readily agreed. I was livid, and the tracking range safety officer was also rather upset. While the helicopter-mounted cameras got some good footage of the P3 flying underneath them, just above the surface, the entire episode was a violation of our safety rules and a waste of valuable research time.

The P3 pilot, who had been briefed on the restrictions, was ordered to report to the operations center after he landed. The pilot, concerned that he had done something wrong, mentioned his orders to the scientists on board. The scientists told him he needn't bother going to the operations center, as the scientists had been the ones to direct him to fly the lower course. So, the pilot never showed up. The next morning, when the P3 had taxied out to the end of the runway to take off, the range safety officer refused to give his permission for a takeoff. Instead, the crew was forced to taxi back to the control tower, and the pilot had to deplane and report to the control center. There he received the chewing out that he had avoided the previous day. I made certain that the scientists responsible also heard about it, as did the sponsors. Such unilateral breaches of an agreed upon test plan should not be tolerated. These scientists violated safety rules and based decisions on an unwarranted belief in the utility of their data in comparison with data from other platforms.

Not all cases are this outrageous but the lesson should be clear. A strict hierarchy needs to be set up to execute an experiment and this hierarchy should be respected by all of the participants.

Ship resource allocation

During the earliest stages of planning, participating investigators should relay their requirements and desires for on board space and other shipboard resources to the chief scientist. Space on most research vessels is at a premium, so the fight for space can be brutal. Your documentation of space requirements should include the number of people participating in the cruise broken down by sex; the size, weight and deployment needs of any in-water instrumentation; the size and type (wet or dry lab) of laboratory space needed; special needs such as refrigerator space or explosives storage; and any additional storage needs. [As an aside, laboratory space on most modern day cruises is measured in linear bench space, using the non-SI unit, IBM-PC (or PC for short). Thus I have been known to tell someone that I need three PC of space on a particular cruise. For those of you not familiar with this measure of length, a PC is just enough space to locate an IBM PC, monitor and keyboard with tie downs and a chair for a scientist to work at the system.]

The chief scientist has the responsibility of getting information on the ship for the other investigators (you don't want ten groups all calling the marine department asking for layout drawings of the ship). He then matches the stated space requirements with what is actually available on the ship, and using a shoehorn and cattle prod, develops a preliminary space allocation plan. Sharing this plan with the investigators prior to the experiment and allowing them to comment on the preliminary plan before making it final can save a lot of problems during the onload.

Coordination of multiple groups

In a typical experiment there will be multiple research groups, with differing interests, on each platform. (I use the term platform here, instead of ship, because I have worked from ships, aircraft, towers, or piers. In this context a platform is anything that holds instruments.) This means that a primary responsibility of the platform scientist is to coordinate the research activities of the various groups. This requires a detailed understanding of each group's research interests and how these interests fit into the context of the overall experiment. This type of coordination is best done well before the cruise, so that the each group has time to consider, and eventually agree to, the chief scientist's view of their role in the experiment. Reaching an agreement prior to the cruise is key because it will be the chief scientist that runs the operations on the platform. Don't delude yourself into thinking that disagreements regarding operations, which occur before the experiment, will get settled to your satisfaction in the field. It usually doesn't work out that way.

At the same time, the platform scientist has the duty to weigh the oftimes competing interests of the groups on board, insuring that each receives their fair share of the data collection opportunities. A perfect example of this occurred on an APL cruise that I went on with Gene Terray. Prior to this cruise a deal was struck with Gene to provide an opportunity for his graduate student to field a new buoy system that he had been developing. APL had agreed to these deployments as partial compensation for Gene's contributions to the APL experiment. While this new buoy was not of direct interest to the sponsors of this experiment, it was still the responsibility of the chief scientist on the cruise to see that sufficient opportunities were available for the testing of the buoy. Because the balancing of competing interests at sea often takes on the air of a high wire act, it is best to reach a consensus on the approach to be taken prior to sailing.

Coordination of multiple platforms

As difficult as coordinating the activities of a number of different groups on a research vessel is, coordinating activities between multiple platforms is even worse. The problems are inevitably exaggerated by the difficulties of communications. If there is a dispute on board a ship, all of the parties can sit down together and argue for their interests. Communications between platforms are usually limited to radio, a fact that makes open-ended discussions difficult. For this reason, if no other, the chief scientist on each platform must be responsible for communications with the other platforms in the experiment. While all of the platform chiefs will confer on the conduct of the overall experiment, there is still a need for an overall head, or test scientist, to resolve disputes that might occur.

The keys to coordination of multiple platforms are planning and communications. Noting that communications would be almost unnecessary in a perfectly planned experiment, it should be clear that detailed planning of the coordination of platforms should be carried out prior to the experiment. Communications, which are inevitably needed when the pre-test planning proves to be less than perfect, are discussed in more detail in a following section.

Coordination of preparation and onload

Some final thoughts on the pre-test coordination of preparation and onload are important. I spend considerable time in this book discussing individual preparation for at-sea work. It is sometimes equally important in large experiments to coordinate the preparations between multiple groups. This is especially true when data are to be shared by these groups. When planning your test, be sure to allow time, money, and support for the coordination of various group's systems. This can mean testing joint data acquisition systems, fabrication of brackets to hold another group's instrument on your platform, or just planning on how data will be shared after the experiment.

Coordination and planning of the onload are also important. The chief scientist on the platform should take the lead in keeping all of the participants informed with regard to the onload schedule. In return, individual investigators need to let the chief scientist know about any special requirements that they may have for onload, such as crane support, support for extra heavy loads, welding services to attach buoy cradles to the deck, or any other support which will require the arrangement of special services. The onload schedule should be planned to avoid common problems. Will dock services, such as a forklift, be available? Does the onload occur over a weekend when needed personnel may be off duty? Are sufficient crew scheduled to aid with the installation of equipment? Asking lots of these kinds of questions, along with careful planning, are the keys. When nearly all has been said and done, you should always ask one more question, namely, "Are there any questions that I didn't ask, but should have?" You'll know you're ready when the answer to the last question is no.

Communications

Communications is a very important part of any experiment. Unfortunately it is usually the first thing to fail and these failures often have more to do with people than hardware. For this reason it is of critical importance to include some detailed planning for communications in your pre-test planning.

Communications planning should begin by identifying the needs particular to each experiment. For example, all experiments will require communications between personnel on the ship, many experiments will require communications with researchers or support personnel on land, and a few large experiments will require communications between research vessels in order to coordinate their activities. In this section, I'll discuss some of the communications technologies available for platform-to-platform or platform-to-land communications and discuss some of the limitations that you should expect in current systems. I'll then discuss communications planning before tackling the more difficult subjects of those aspects of group psychology that affect all forms of communications at sea.

Marisat, cellular phone and fax, telemetry, ...

Communications technology is rapidly evolving. At the time of this writing there are a number of forms of ship communications, but the range of available choices should explode in the next few years as new commercial satellite systems become available. For example, just on the horizon is the Iridium system with numerous low power satellites in low Earth orbit designed to provide universal, inexpensive communications. I hope that such systems, and the competition they engender, will revolutionize communications for oceanographers. Until then we are somewhat limited in our available choices.

At the present time, communications to vessels more than a few hundred kilometers from shore is restricted to either Marisat satellite communications or older and less reliable HF systems. Marisat is an international consortium that provides voice, fax and data communications through a satellite network. The system uses relatively inexpensive earth terminals that are common on most research vessels, but can be rented for those vessels that don't have a permanent installation. Users of the system pay a per-minute and per-call fee that can be rather substantial for long conversations. Marisat is fairly reliable but not perfect. You should expect some difficulties in establishing connections and will almost certainly experience some lost connections at times. These are worst on temporary installations where the antennas are typically smaller, and in large sea states where ship motions can affect the link. While we regularly use Marisat for fax transmission, we have had mixed success with data transmissions. At this time I would not suggesting relying on Marisat data transmissions as your sole option for relaying data.

Despite its problems, Marisat is still preferable to HF radio systems, which are dependent on the vagaries of ionospheric propagation. These may be useful for the most cost-sensitive experiments, but the unreliability and low bandwidth of the link usually makes regular communications difficult. HF propagation links depend on ionospheric conditions, which vary with time of day, season and phase of the 11-year sunspot cycle.

On the other hand, HF radio can be an interesting diversion on long cruises. Members of my group at APL often take an HF amateur radio rig on board for communications with other amateur operators. Sometimes, they can arrange communications with families back home through a cooperating amateur who will use a device called a phone patch which routes the transmissions into the phone system. This is a bit of a kludge, but it can save considerable money in reduced Marisat charges. As a licensed amateur radio operator I will caution though that operators are required to be licensed and that these communication links are legally restricted to non-business uses.

Closer to shore, the communications channel of choice is the cellular telephone. With a decent antenna mounted high on the ship reliable cellular communications can be maintained out to 100 km. In the Gulf of Mexico, cellular sites maintained on oil production platforms extend cellular ranges to 240 km. Fax and data modems are readily available for use with cellular systems. Cellular systems are so popular that they are being used by numerous institutions for telemetry and control of remote buoys in the coastal environment.

One problem with cellular phones that you should be aware of is that standard modem communications over a cellular phone range from poor to mediocre. On a recent experiment we used a standard high-speed modem connected to a cellular phone to check our email and exchange small data files with colleagues back home. While the modem would tell us that it had linked at 9600 baud, the actual baud rate was closer to 1200 baud, and that is when it worked at all. The modem industry has responded to these problems by creating new, more robust data exchange protocols. I have not had a chance to try out these new protocols, but I think that cellular modems have a long way to go before they are as easy to use as their wired cousins.

Finally, a few words about data telemetry. Having even a limited real time data link to remote instruments is a big plus. Telemetry systems on surface buoys are often one- or two-way radio links, while many subsurface buoys make use of acoustic modems. In either case, care should be taken to understand the characteristics of the telemetry link, including useable range, dependence on environmental factors such as sea state, and transmission error correction schemes and resultant error rates. Radio telemetry systems that work well over land, will often have reduced performance over water due to multipath interference caused by the proximity of the antennas to the sea surface. When using a new system for the first time, be sure to test it over water beforehand, and to include contingency plans in case the system fails or works only at reduced ranges.

For even more remote operation, Service ARGOS is a satellite-based system for locating and acquiring data from buoys at sea. ARGOS transmitters, officially called PTTs, are at the time of this writing down to as little as $550 per unit. You also pay a daily fee to Service ARGOS for location or data information. Data can be transmitted from your buoy in small, 32-byte packets, so data rates are extremely slow. Still, for longer scale drifter studies ARGOS is hard to beat. In addition to data transmission, the ARGOS system can provide location information for buoys, although the precision is not particularly high (typically 1.5 km). The navigational fixes can be useful for tracking a drifting buoy or finding a buoy that has gotten away from you. Given the cost of a lost buoy, the inclusion of an ARGOS PTT is cheap insurance.

Communication schedules

After having decided on the physical system(s) used for communications, you must plan for how those channels will be used. I strongly suggest that communication schedules be developed and maintained throughout an experiment. A communications schedule is nothing more than an agreement to establish contact with other platforms or land-based investigators at particular times. Such scheduling will in all likelihood be flexible, but once a scheduled is agreed to, it should be kept.

On experiments with two ships, we regularly schedule a major communication to discuss the day's events prior to breakfast each morning. By having a fixed schedule, those scientists interested in contributing to the discussion can join the chief scientist in the conversation. Early scheduling of the daily planning discussion also has the advantage that any major changes to the schedule can be communicated to the other scientists on board at breakfast, before everyone scatters to work on their equipment.

While the communication systems on a modern R/V (Research Vessel) should be monitored all of the time, it is usually inconvenient to track down someone on the cruise when an outside call occurs. Scheduled communications thus are more efficient for those items that can wait until the appointed time. There will always be emergencies requiring immediate communications, but schedules act to reduce the chaos.

Group psychology

Anyone that has attended a college sporting event can attest to the 'us versus them' mentality that is generated by such competitions. In the best sporting traditions, team conflict raises the level of competition without invoking the worst in people. In other cases, competition acts to reduce people to their basest self, sometimes with dire consequences. If you think I exaggerate the possibility, just try attending a Washington Redskins game in a Dallas Cowboys jersey.

I bring this up, because group conflicts inherently arise in the high-pressure world of working at sea. People seem to have a natural tendency to form groups and the rigors of at-sea work seem to enhance this tendency. Some of these groupings, such as scientists versus crew, are a natural extension of cultural or economic groupings that occur in society. But there are other groups whose formation are unique to the at-sea experience, such as watch versus watch, ship versus ship, or ship versus land. And just as in the sporting world, these groupings can be both good and bad.

A healthy competition amongst watches to make the best measurements, can make a cruise more enjoyable. A competition between those same watches where the goal of each watch is to minimize their own workload can be destructive. On one cruise I was on, such a competition began, with each watch working hard to arrange to have the rosette sampler ready to come on deck just as they were going off duty. The downward spiral of this childish game led to unnecessary delays in sampling, as well as to hard feelings on the part of the watches that "lost" the competition.

I have seen similar problems arise between ships and between ship and shore. There are several reasons for this. Typical communication channels are very narrow, making it hard to fully discuss an idea or plan. People often transmit instructions, directions or requests without providing sufficient justification or rationale. This void can lead to doubt in the recipient's mind - doubt about the utility of the request, doubt about whether the requesting party understands the difficulties involved in meeting the request, doubt whether the competing interests have been properly weighed, and sometimes even doubt about the sanity of the requestor. These doubts may be compounded in others on board the ship. Because only one person can be on the radio at a time, the dynamics of the off-ship groups are typically not fully communicated to the others on ship. This is all quite general and obtuse, so let me provide a specific example of what can happen.

This story comes from what I like to call my 'Cruise from Hell', which actually took place north of Hawaii several years ago. I had signed on to help out a colleague who needed a senior scientist to participate in the cruise, providing some quality control and on-site interpretation of the data. The cruise took place on a modified 150 ft mud boat, which was our first mistake. (Mud boat refers to a class of ships that were designed with large holds for transporting mud produced by oil-drilling platforms back to shore.) It turns out that on this experiment we were looking for big waves and unfortunately for us we found them. I began the cruise by spending the first three days in my bunk praying for an early death. By the time I got my sea legs, morale was already starting to fade. We were having problems with key instruments, the ship's crew was rather unhelpful, our test engineer injured his back and had to spend several days in his bunk, and the cook was the worst I have ever seen. To top all this off, the scientists back on land kept requesting status reports and data that we just couldn't provide.

In our communications with our colleagues on land we tried to explain the difficulties that we were facing, including trying to correct instrument problems caused by poor installation, software bugs and unexpected equipment failures. They would commiserate with our problems, but then call back several hours later asking for another status report. From my own experience back on shore, I knew that sitting around waiting for status reports from a ship at sea is a very frustrating business. Still, they were driving us crazy with their incessant requests. I finally told them that we would no longer answer the phone if they called. Instead, we would call them back when we got our systems reasonably on line. After a few choice invectives the next day when they called, they got the message and stopped bothering us.

After the initial crises had passed, morale continued to sink because of the poor working conditions and the constant pressure from land to acquire data. I fought back by adding more and more humor to my daily cruise summaries, which we were faxing back to land. With each passing day I would add increasingly outrageous paragraphs about the sighting of sea monsters, the problems we were having with the radiation from passing UFOs affecting our instruments, and other such sundry items. On board the ship, my reports became one of the highlights of each day, with a copy eagerly passed around amongst the scientists and technicians before it was sent off. It had become a struggle of us against them and everyone appreciated that I was fighting back.

The nadir of the cruise came a few days before the end of the test, after a particularly nasty stretch of time when we had lost the data from a large group of drifting GPS buoys that finally gave up the ghost. On that day I wrote an especially long and incredibly bizarre report and faxed it back to land. My friend and colleague, who was running the test from the land site, happened to be hosting a visit from the sponsor and a group of interested scientists on that day. My fax came in, and he decided to read it out loud to his visitors. He was embarrassed to find out that in my frustration I had written a long and comical report that contained absolutely no useful scientific information whatsoever.

After the experiment, I rightfully caught a little flack about this incident. It was unprofessional and I knew better. Still, the temptation to succumb to the dark side of an 'us versus them' mentality can be quite strong. The solution is simple but not always easy. The people on each end of the conversation need to attempt to communicate better. One suggestion to improve communications is to actually listen to the other party. (What a radical idea!) Each person should also have a common understanding of the goals of the experiment, the plan that is to be followed, and the difficulties of working at sea. Few things are worse than having an inexperienced person at the other end of the conversation when you are at sea. By trying to put yourself into the shoes of the other person, and by showing a lot of patience, problems of mis-communication can be minimized, if not eliminated.

Communications etiquette

In a typical scientific conversation there is a lot of give and take. Scientific discussions are often complex, involving detailed technical arguments put forward to persuade others of our views. This usually works well when we talk to each other in person, but often breaks down when the medium is radio. There are some simple solutions, though, that can help you minimize communications problems at sea.

1. Try to listen to the other person. I know I am repeating what I just said above, but it can't be stressed enough.
2. Gather your thoughts before speaking. If you are uncertain of a point, tell the person that you'll contact them after thinking about the problem. Few things are more annoying then listening to a rambling discussion over the radio. Also be aware that the person on the other end of the radio is likely to be as busy as you, so don't waste each other's time.
3. Be extra polite and courteous. There are a lot of forces that act to make life at sea difficult. Don't let those spill over into communications with those not on the ship.
4. Remember that virtually no ship communications are private. Usually, there are others listening to both ends of the conversation so don't say anything that you don't want everyone to know. This specifically means that you should be wary of discussing personnel problems or other private matters over an open communications link. On one cruise the officer on watch overheard a fisherman call his wife to tell her that the fishing was particularly good and that he would stay out another night. This conversation was immediately followed by a call to the fisherman's girlfriend telling her that he would be right over. My guess is that there were several hundred people that heard this conversation, and there is no telling if one of these people might have been a friend of the man's wife.

Chain of command

Although this was discussed previously, I feel compelled to reiterate here that respecting the chain of command is critical for successful communications at sea. The platform scientist should be the focal point for all communications and he should in turn pass on these communications to the other members of the scientific party. Real problems arise when only the chief scientist knows what is going on, but these problems are easily corrected. The chief scientist just needs to maintain a constant dialog with the others on board.

If you are the chief scientist it's a good idea to use printed notices on a central ship bulletin board (or bulkhead) for the promulgation (great word isn't it?) of cruise plans and information. There are two reasons for this. First, you won't always be available to the scientific party when they have a question about the schedule, but the bulletin board will be. Second, given the typical watch schedule, it is usually hard to get everyone together at one time. Thus printed instructions and plans are a handy communications tool.

On experiments that do not involve watches, I suggest you schedule daily experimenter's meetings. For planning-intensive experiments with complex and variable daily schedules it is best to have this meeting prior to each day's work. For more laid back experiments, you might want to consider meetings at the end of the day designed to review daily progress. These experimenter's meetings are a good forum for monitoring the progress of the experiment and getting everyone working together. They need not, and should not, be any longer than is necessary to conduct the business at hand. A typical daily experimenter's meeting, excluding the first few days of an experiment that are always hectic, may last only 10 minutes. Still, it is a great way to keep everyone up-to-date on the progress of the experiment.

Navigation and Timing

Well we know where we're goin'
But we don't know where we've been
And we know what we're knowin'
But we can't say what we've seen
And we're not little children
And we know what we want
And the future is certain
Give us time to work it out

We're on the road to nowhere
Come on inside
Takin' that ride to nowhere
We'll take that ride...

These are the opening words to what I have come to think of as the oceanographer's theme song, The Road To Nowhere by the Talking Heads. As best I can figure, David Byrne, the author of these words, must have been an oceanographer in some past life. To my mind, his words capture that uncertain, lost feeling that you can have at sea facing the unknown, yet struggling to understand. I particularly like the key line "We're on the road to nowhere" because it all looks the same at sea. Lewis Carroll eloquently made a similar point in his epic poem, "The Hunting of the Snark":

The Bellman himself they all praised to the skies--
Such a carriage, such ease and such grace!
Such solemnity, too! One could see he was wise,
The moment one looked in his face!

He had bought a large map representing the sea,
Without the least vestige of land:
And the crew were much pleased when they found it to be
A map they could all understand.

"What's the good of Mercator's North Poles and Equators,
Tropics, Zones, and Meridian Lines?"
So the Bellman would cry: and the crew would reply
"They are merely conventional signs!

"Other maps are such shapes, with their islands and capes!
But we've got our brave Captain to thank:
(So the crew would protest) "that he's bought us the best--
A perfect and absolute blank!"

As Lewis Carroll might attest were he alive today, joint navigation and timing are of critical importance in most voyages and particularly so on oceanographic experiments. Here I am referring to those data that tell you where and when particular measurements were made. If you are out on a ship by yourself, then maybe you need only know where data are acquired to the nearest kilometer and the time of day to the nearest hour. If, on the other hand, your data will be merged with data from other instruments, which may be on other platforms, the need for accurate joint timing and navigation becomes more acute. I use the word joint here to indicate that often the requirement is not to have high absolute accuracy in timing and navigation, but instead to have high relative accuracies between platforms. I'll say more on this subject later.

Navigation used to be complicated with lots of choices. Now you only need to remember two systems: GPS and DGPS. GPS stands for Global Positioning System, a rather amazing technology that utilizes a constellation of satellites broadcasting to small receivers on the Earth. For less than $400 one of these receivers will tell you where you are on the Earth to within 60 m rms error. This is cool stuff.

For those cases where 60 m accuracy is not sufficient, Differential GPS (DGPS) comes to the rescue. It turns out that the GPS system was actually designed to achieve accuracies of just a few meters. Unfortunately, the accuracy of the system is purposely degraded by the U. S. Department of Defense using a system called Selective Availability. This system, which works in part by adding clock noise to the satellite broadcasts, reduces the accuracy of the system for the average user, but has no effect on special coded receivers used by the military. Differential GPS gets around this added noise through the use of a base station, which is set up at a fixed site. The GPS base station receives the information from each satellite, computes its position relative to the satellites according to the GPS system, and then broadcasts correction information to any nearby DGPS receivers. The nearby DGPS set computes its location based on the GPS signals and the corrections it receives from the base station. In this case, nearby means closer than about 200 km, a distance dependent on satellite visibility and ionospheric variations.

There are several interesting facts about DGPS. The first is that the U. S. Coast Guard is going around putting up DGPS base stations and beacons along the coast of the U. S. To my simple way of thinking this seems like one government agency paying lots of money to undo something that another agency is intentionally doing. The designers of the system have always claimed that the development of DGPS base systems was foreseen, but I still don't understand the reasoning behind Selective Availability. Even more curious is the fact that during the war in the Persian Gulf, the U. S. military turned off the Selective Availability, making the full resolution of the system available to all users. They did this because they didn't have sufficient military GPS receivers to go around. Instead they bought lots of commercial sets for their personnel in the desert. I'm certainly no expert on military affairs but if I was the boss I think I'd turn off Selective Availability, except in time of war.

There are other, even more fantastic forms of DGPS. The system I have referred to here is called pseudo-range DGPS, and is the simplest form. Another, more sophisticated form of DGPS, called carrier-phase DGPS, can provide centimeter-level accuracy. This is seriously cool stuff. At this time, carrier phase techniques are quite expensive, but they are beginning to find application in multi-antenna systems that can determine not only the location but also the orientation of a ship. This is more accurate on long time scales than the standard gyrocompass, and so can be used to reduce compass errors in ADCP measurements.

At the same time that GPS provides location, it can also provide very precise time information. The trouble is that in my experience very few oceanographers take advantage of this capability. I cannot tell you how many times I have seen an otherwise reputable scientist rely on the clocks in their computers (which they set from their wristwatches) for timing on experiments. Wristwatch time just doesn't cut it when data from several systems need to be integrated together after the experiment. This is something that I can safely say you won't fully appreciate until the fourteenth time in a post-test analysis where the only explanation for a discrepancy is a possible several minute timing error in one of the data acquisition systems. The point here is that synchronized timing is often needed between data acquisition systems within a lab, between labs, and between platforms. While some details of clocks and how they drift are described in Chapter 6's description of data acquisition systems, it is good to think about the overall need for time synchronization during the planning of an experiment.

To give you some idea of what I mean, I'll relate two brief stories of timing problems that I have personally experienced. On one cruise we had difficulty with the interface between a gyrocompass and our Acoustic Doppler Current Profiler (ADCP). It seems that the scientist responsible for the system had no idea that stepper interfaces were notorious for drifting due to missed steps. After we realized that we had a problem, several of us scrambled to come up with a workable solution. We could post-test calibrate the ADCP data using the compass data from our meteorological system. In investigating this solution, we discovered that the programmer of the meteorological data acquisition system had utilized the system clock in the IBM PC, which was drifting like mad. This same system had a GPS time code going into it, but the careless programmer had decided to discard the GPS time data in favor of a clock that drifted due to his overuse of interrupts. In the end, the data were important enough that several weeks of effort had to be devoted to correcting the clock drift in the meteorological data stream by performing short-time correlations of the high frequency portion of the compass signal with the stepper output. Once the two data streams were synchronized then the met compass was used to correct the ADCP data.

My favorite timing story took place during a cruise on a Russian research vessel. When we first set up our equipment we installed a GOES satellite clock to synchronize our own systems. Our offer to provide these time signals to our Russian colleagues was initially turned down. Our Russian colleagues explained to us that this was a very modern research vessel and each lab contained a video screen that displayed a clock derived from a satellite time code entering into the central computer. I was a very impressed with this setup, at least until we noticed that the video clock in our lab was about 5 seconds behind our satellite clock. When we brought this discrepancy to our colleagues' attention their initial reaction was that our satellite clock must be wrong! It took several days to determine that the problem was in the central computer. When the computer experienced a large load, such as when acquiring CTD data, the updating of the clock was delayed, sometimes for several seconds. Thus not only was the video clock late, but it was late by an amount that depended on the computer load. By this time our colleagues were convinced that they should use our clock to time their instruments. When the signal was run to a forward lab, the scientists there at first objected that it differed significantly from their clocks. It turned out that their entire lab had been synchronized to a wall clock that differed from real time by about 15 minutes! To this day I kid my friends who were in that lab about the forward lab in the ship being in a different time zone then the rest of us.

One final comment about time, and that is units. When you make a scientific measurement of length you use meters. When you measure mass you use kilograms. Likewise there is only one way to measure time on a scientific experiment, and that is UTC. (UTC is essentially French for Greenwich Mean Time or GMT. It has always struck me as funny how the French have a different word for everything.) Local time is not acceptable. The use of local time is certainly a sign that timing has not been taken seriously on an experiment.

What to pack

Packing for an experiment should be a nerve racking experience. You read that sentence correctly. While packing, you should be constantly worrying whether you are taking the right stuff. After all, when you get out on that ship, the nearest Radio Shack might as well be on Mars. (Actually I think several of the people who work at my local Radio Shack spend their spare time on another planet, but that's another story.)

In this section I'll provide a few hints to help insure that you take the things that you need. The key in all of this is to focus your planning on your possible needs at sea.

Packing lists

When I first went to sea, I just packed up those things that I thought I'd need, and I went. I was saved from disaster in these initial expeditions by colleagues from APL, who were better prepared than I and could afford to lend me supplies, tools and equipment. It slowly sunk into my skull that I needed a system. We're not talking quantum physics here. I just started making a list of the things that I use on an experiment. I began by listing all of the items that I had taken on my previous experiments. To this I added all of the stuff that I had borrowed or used from others. I then added the new items that I knew I would need for my upcoming experiment. I worked and worked on the list, making sure that it included every item that I might possibly need. I used this list to guide my packing for my next experiment.

Things went extremely well on my first experiment using my list, in large part because I had planned and packed so well. Still there were some items I had overlooked, that I ended up needing during the experiment. I kept a list at the end of my cruise notebook of all of these items. I also added items to this list that I saw other scientists using on our cruise that I thought might be handy on future tests. After the experiment ended I added these items to my Master packing list, which gets longer and longer after each experiment.

My packing list was in fact too long for the next experiment that I went on because of the limited nature of my participation in that particular experiment. I thus added the new items that I was going to take on that cruise to my Master list, duplicated the Master list and then edited the duplicate list deleting those items that I didn't need. I now do this on every experiment. In this way my Master list grows and grows, reflecting the items that I have used on nearly all of my experiments. When an individual experiment arises I add the new items to the Master list, make a copy and then delete the items I don't need from the copy, which is now my specific packing list for that experiment. This forces me to go over every item I have ever found useful before every experiment helping to insure that I don't forget anything. This sounds tedious, and it is, but the system works well for me.

The problem generally is starting such a system. Here is where I can help. I have included a commented packing list as Appendix C of this book. This should help you get started with your own Master list, and may give you some good ideas about things that I have found useful on my experiments. I will also admit to an ulterior motive for including this list - it is a blatant attempt to get this book onto the New York Times bestseller list. After all, what could be more exciting than reading someone else's personal packing list?

Expendables

It's hard to do science without pens and paper. Expendables should be an important part of your packing list. When planning your usage of expendables remember to plan for the unexpected. Someone else may have forgotten to bring rubber bands, which prove essential for the repair of a vital instrument. Everyone on the ship may want copies of your cruise log, straining your meager supply of paper. Your scissors, through an act of spontaneous evolution unimagined by even Darwin himself, may grow a set of legs and walk out of your lab. If I can partake of a Gump-ism, spares are good.

Again, I've included a separate list of expendables that I take on experiments in Appendix C. The quantities shown should be adjusted to meet the needs of each particular experiment, but it can be a starting point for your own list.

Tools

The only tool that some mountain gorillas use is a short, thin stick to get ants out of hollows in the trees. While this seems to work well for the gorillas, you may aspire to a slightly higher level of professionalism in your career. If this is the case then you'll need a lot more tools than the gorilla uses. (Although I don't want to slight the utility of a short, thin stick in many at-sea circumstances.) I haven't included a full list of the tools that I carry into the field, because your list will necessarily be different and I felt I could only pad the length of this book so much. The important thing though is to take what you need, to take what you might need, and to take what you don't think you need. If this sounds like I'm suggesting you over-pack, you're getting the point. Tools typically don't take up a lot of space. They are heavy, so you'll be cussing me during the onload and offload, but when you find that the automotive break adjustment tool in your toolbox is the ideal gizmo for working on your CTD, you'll thank me.

I usually pack a large commercial-quality roll-on toolbox packed with heavy tools. I also take a smaller toolbox with electronic tools, soldering irons, etc. Finally, think about taking some small tool bags for carrying a few tools out onto deck or out on a small boat for repairs of buoys. Let your needs guide what you take, but don't forget anything.

Instrumentation and spares

The next major things to pack are your instruments. My advice here is to pack them well, always take copies of all of the documentation that you have, and take spares, and lots of them.

Documentation and spare parts are the keys to repairing equipment in the field. Anything can be repaired eventually. It is just that without proper documentation, including schematics, equipment repairs may take several orders of magnitude longer. Spare parts are also critical. Try to carry spares for all expendable items such as batteries, o-rings, and seals, as well as any delicate or exposed sensors that might get broken due to handling. Major subsystems and power supplies are other good items to spare. Finally, if you can afford it, spare instruments are great to have along. Even if the spare instrument is not of the same quality as the main instrument, it may make the difference between success and failure.

Personal Gear

Telling you what to pack for your personal gear is a bit more difficult. I have a commented list of the things I take to sea, but your needs will likely differ from mine. Still, it is best to anticipate your needs if at all possible.

• Drugs - I'll admit here and now that I would not have become a sea-going oceanographer if it was not for the ingestion of certain drugs while I am at sea. (No, not those kinds of drugs!) In particular, I use motion-sickness drugs to combat the nausea of sea-sickness. While I won't recommend one drug over another, an overview of available motion sickness drugs, both prescription and over-the-counter, can be found in the section on sea-sickness found in Chapter 4. In addition I always take along some aspirin for headaches, some antacid for heartburn, and some non-drowsy formula cold tablets, just in case.
• Clothing - Use common sense and take clothing that fits the season. Remember that early morning watches can get quite chilly, even in warmer waters, so its always good to take along a jacket or sweatshirt. In colder weather remember that layered clothing preserves warmth by trapping air. It also makes it easier to adjust the amount of insulation you need by removing individual layers. Ships are typically dirty places and major clothing hazards, such as grease, abound. For this reason, I'll take at most one decent set of clothes on a cruise for use in port. The rest of the time I wear old jeans and t-shirts. The idea is to be able to discard your clothes at the end of the cruise if you have to. Remember that you'll be working closely with others so don't take clothing which is too tattered, suggestive, or immodest. A certain relaxed decorum is the accepted norm.
• Shoes - A second pair is a good idea, in case your main pair gets wet.
• Toiletries - Bring your own soap, shaving cream/razor, and other assorted stuff you need to prepare yourself for each day. These are not available on a ship, and its rather embarrassing to try to borrow a colleague's toothbrush.
• Personal safety equipment - As I have discussed before, I like to take my own safety gear on cruises. It is a bit more professional and it avoids the problems of people using my stuff. My typical list includes: steel toe boots, hard hat, personal flotation vest, and work gloves.
• Flashlight - I usually take a small personal flashlight on cruises to get around on deck at night. Research vessels can often be extremely dark at night, so be prepared.
• Alarm clock - Personally I use my watch, but don't forget a more substantial alarm clock if you are heavy sleeper. It is worth mentioning that most motorized AC clocks will not work well on board a ship. While the 60 Hz (or 50 Hz for our more- civilized colleagues) AC power coming out the plug in your home is tuned to the proper frequency within ridiculously high tolerances, the AC power on a ship runs at somewhat arbitrary and chaotically varying frequencies depending on the number of waffles being made for breakfast and other, less discernable, factors. While the ship's power is good enough to bake a casserole, it is not good enough to run the timer on the oven. My advice is to stick to quartz crystal clocks if you want to wake up on time.
• Baseball or knit watch cap - A hat should be a requirement on all cruises. In sunny climates it can keep the sun off, and in cooler climes it can keep you warm. Furthermore, if you end up in the water, a hat can significantly retard the heat loss from your body, increasing your chances for survival. In warm weather I prefer a baseball cap, just be aware that they do have a tendency to blow off your head while under way.
• Sun block lotion - To steal a line from Woody Allen, I don't tan, I stroke. Sun block is a necessity in sunny weather. Look for an SPF rating of 30 or more.
• Rain gear & boots - Somehow I don't mind getting wet as much when I am at sea, but rain gear and boots are still nice to have when the weather turns nasty.
• Washcloth - This may seem like a trivial thing, but I like to use a washcloth in the shower, and many ships don't provide them. Be forewarned.
• Flip flops - Sometimes these don't get out of my duffel bag, but on some cruises they are great for going to and from the shower. Otherwise, it is not a great idea though to wear sandals or flip flops around the ship because of the danger to your feet. They are nice though for use on those ships where the showers are located outside of your room.
• Deck chair - A few years ago I invested $60 in a deck chair that folds into an extremely small space. Now, on every cruise, people make fun of me at the beginning of the cruise when I unfold my deck chair. By the end of the experiment, though, I usually have to fight to get to sit in my own chair. This is one of the best $60 investments I ever made.
• Personal entertainment - This is a must with me, even though I am usually swamped with work on each cruise. I find that some music, books and computer games help fill the time. What you bring along for entertainment is your own business, just remember that not everyone will share your tastes. For example, I usually take along a CD player, an assortment of CDs and a small set of speakers. I always ask permission before playing any music through the speakers and if even one person objects I use my headphones. I also make it a point to share my system with others within my lab, so that we can each listen to music that we enjoy. Computer games are also great, just don't let them get in the way of your work. You should also come prepared to use headphones on your computer games as the noise of most games may drive others crazy.

Items that are usually supplied by the ship include sheets, pillows, blankets, towels, cleaning supplies, toilet paper and laundry soap. You should always check first though and take an item along if you are in doubt. Remember that your personnel storage space may be limited. For that reason you shouldn't carry hard-sided luggage on board, but instead should opt for an inexpensive duffle or sea bag. They can double as a dirty clothes hamper and take up no space when empty.

Travel

Inevitably you'll need to travel to get to your on load site and from your offload site. You should take some care in planning your travel. Questions to ask include:

• When will I need to be on the ship?
• When will the ship return to port?
• How certain is the schedule and what are the possibilities for change?
• Will I need to hand carry equipment or data either to or from the ship?
• How much time will spend in port?
• Can I stay on the ship or do I need a hotel room?
• How much money will I need for living expenses?

Make your travel arrangements well in advance and double check them prior to departure. Allow plenty of time at airports. You don't want to miss a cruise because you got bumped from an aircraft flight. If your trip includes a stop in a foreign country then you should consider a whole host of additional questions:

• Will I need a passport or visa?
• How long before my trip do I need to apply for a visa?
• Do I need any special shots or medical precautions to travel to this area?
• What dress is acceptable in this country?
• Is it better to use traveler's checks or cash?
• What exchange rates can you expect and what is the cost of living?
• What are the customs regulations regarding transport of equipment/data?

It is important to start asking these questions early. Visas can take 6 to 8 weeks. Foreign embassies are usually helpful but unmoved by pleads for special consideration, especially when you could have planned ahead.

Before departing on travel, it is a good idea to leave a complete itinerary and a copy of your passport with your office and home. That way people can get hold of you in transit in case schedules change, or can help out if you happen to lose your passport while overseas.

It is also best to keep reservations open ended and changeable, at least to the extent possible. You don't want to go overboard with this though. I'll usually compare the costs of getting an unrestricted airfare, allowing me to freely make any changes to the schedule, to the costs of changing the schedule on a nonrefundable ticket. I find that it is usually cheaper to buy the nonrefundable, supposedly nonchangeable, ticket and then pay a $50 or $100 fee to make a change later if you must. Just be careful to know the terms of your reservations and don't be afraid to ask questions of your travel agent.

On load

The first step is to pack your equipment carefully. I have been on several experiments where major instruments arrived on the dock broken due to a shipping mishap. Several years ago there was a cute television commercial for a luggage manufacturer, featuring some chimpanzees acting as airline baggage handlers. The commercial was quite funny, but in today's competitive environment, shipping companies have progressed way beyond this common stereotype. In today's fast-paced world of shipping, the chimpanzees are now driving forklifts. Your only hope is to pack well. This means boxes in boxes, lots of padding cut to fit, and solid outer containers. There are lots of good commercial packing containers on the market, so I suggest you invest in some tough containers.

Once your equipment is packed you have to ship it to the onload port. Shipping is always a precarious venture that typically relies on a inordinately large number of things going right for your shipment to arrive safely and on time. As in almost any human endeavor I have found it best to assume that the system will break down at some point. So I routinely expect that my shipments will take longer than I am told. I expect that they might get lost or trapped in customs. I expect that they will be left out in the rain for some period of time. The best you can do is pack well, track you shipment as closely as possible until it reaches its destination, and pray.

Once you and your equipment arrive at the ship, the onload can begin. Well almost. First you'll need to check out the ship. Ask for a brief tour of the vessel, paying particular attention to your assigned room, lab and deck space. You'll also need to check that the proper support is available for your onload of equipment. This might include a forklift, hand trucks, cranes, welding services for deck cradles, etc. Finally, you'll likely need help from colleagues or crew just carrying your equipment on board. This is where you can get off to a good or bad start. Most people will be glad to help you if you ask nicely, work along side them, and reciprocate the favor by helping others.

This is the time you should start building your relationship with your colleagues. Your primary responsibility is to get your equipment on board and get it working. At the same time, this is the responsibility of all of the other scientists and engineers involved with this cruise. Some time spent helping others can more than pay for itself in good will later in the cruise. If you finish your work, or are needed temporarily to help someone else, don't hesitate. I always hate seeing people set up their equipment and then run off when others are working. It is a certain sign that the scientist is more interested in their own work than in the success of the cruise as a whole.

This is also the time to start building your relationships with the ship's crew. Remember each time that you board a new ship that you are walking into someone else's home. You should treat the crew with respect and realize that they also have a lot of duties to get ready for the cruise. Getting off to a bad start with the crew can lead to problems for the entire cruise, so be on your best behavior.

One colleague I know tries to get off to a good start with the crew, at least in warm weather, by taking a cooler of soft drinks on board with him during the onload. Offering a cool drink to the crew after they have been working may seem like a cheap bribe, but it is a small gesture that usually goes over well.

Planning Checklist

• Agree upon experiment objectives and methodologies within a written science plan.
• Come to complete, mutual and explicit agreement on all collaborations.
• Make sure responsibility for each instrument is assigned to a single individual.
• Plan for adequate scientific supervision of technicians and graduate students at sea.
• Write and agree upon a detailed test plan that should address:
  • Scientific overview
  • Experiment overview, including schedule
  • List of participants, platforms and instruments
  • Progressive detail of operations from platform to instruments, including coordination of platforms and groups
  • Communications, including frequencies and schedules
  • Navigation and timing
  • Safety
  • Contingency plans
  • Command organization
  • Ship resource allocation
  • Onload/offload scheduling and resources
• Establish a chain of command for the experiment
• Allocate resources fairly among the various investigators
• Coordinate activities of various platforms, groups, and instruments
• Coordinate onload and offload activities including shipping and ship/pier support
• Establish a communications plan including voice, fax and data; establish frequencies and schedules
• Be sure those communicating understand difficulties and proper communications etiquette
• Plan overall experiment navigation and timing, setting requirements based on overall experimental objectives
• Pack equipment, tools, supplies, and personal gear. Insure everything necessary is packed by using packing lists
• Make and double check all travel arrangements