We arrived in Portugal, the Vilamoura Marina, on 26 June after leaving Galle, Sri Lanka more than 9 weeks earlier. And we made a hurried trip, with little time in the ports along the way as the sailing was slow, up the Red Sea and across the Med. Larry Armi, the chief scientist and my friend from university days at Berkeley, arrived after a few days, and his assistant, Rita, was already here to greet us. After a busy few days spent installing the necessary equipment we made our first research trip on 5 July, accompanied by Prof. Amy Bower of the Woods Hole Oceanographic Institution and Dr. Fatima Sousa of the Oceanographic group at the University of Lisbon. The research was a joint project with these institutions.

Larry Armi, who proposed this particular project, obtained his PhD at the University of California at Berkeley in Mechanical Engineering and while there became a friend of mine and of the family. He did some nice work studying flows with two liquids of different densities, in particular the instabilities and waves that could result. Later, at Woods Hole Oceanographic Institution, he became interested in the flow through and outside the Straits of Gibraltar. After becoming an established scientist he joined the University of California at San Diego and the associated Scripps Institute of Oceanography, and carried out studies on the flow of ocean water in and out of the Straits of Gibraltar, and also some pioneering work on the "Meddies" formed from the Mediterranean outflow. When he heard that I had bought Kialoa he asked if I was interested making a series of measurements off the coast of Portugal to study the formation of these Meddies. I was positive and Larry wrote a proposal which was eventually funded by the National Science Foundation

The Straits of Gibraltar have a complicated flow of water; on the surface North Atlantic water is flowing in, but below the surface Mediterranean water is flowing out. I understand some of this, but not completely as I am a layman when it comes to the fluid dynamics of the oceans. On the assumption that you may be interested, I will try to explain the basic phenomena.

In the Mediterranean the evaporation of water from the surface is so great that the incoming fresh water from the rivers which empty into the Med will not keep it filled, and as a result salt water from the Atlantic Ocean flows in through the Straits of Gibraltar to keep the Mediterranean level with the Atlantic Ocean. By evaporation this salt water is concentrated in the Med, just like in salt ponds such as are at the south end of the San Francisco Bay (and in many other places in the world) . Thus the water of the Mediterranean is more salty, and hence more dense, than the Atlantic Ocean water. Also the Mediterranean water is warmer, due to the heating by the sun. The Atlantic Ocean water is cooled by the circulation of the Gulf Stream from the cold Arctic region.

The Straits of Gibraltar are about 9 miles wide and also relatively shallow so that they tend to keep the waters on the two sides separated. The lighter, less salty and a bit cooler Atlantic water flows in on top while the heavier Med water flows out in a submerged layer below the surface. This has long been known, and submarines can sometimes "hide" in the more dense Med water as the sound they make can be reflected by the surface between the two layers and thus not heard by the microphones carried by surface ships.

The outgoing Med water passes over the bottom of the shallow Straits and then drops as it flows out into the Atlantic, but at a depth of about 1,500 meters it reaches the density of the Atlantic water and then becomes a layer of warmer, saltier water, perhaps 1,000 meters thick, flowing westward away from Gibraltar. Due to the rotation of the earth, this water tends to flow along the northern shores, following the coast of Spain and Portugal, and then off into the North Atlantic Ocean proper.

Something rather peculiar happens during the flow along these coasts. Rotating eddies form from time to time and often grow to a fairly large size as they enter the Atlantic, becoming as much as 100 Kilometers in diameter. These rotating eddies, consisting of discs of Mediterranean water, move slowly across the Atlantic, decaying in speed while mixing with the Atlantic water on the upper and lower sides, and sometimes lasting for up to 2 years. In the oceanographer's lingo they have been dubbed "Meddies".

So the question is: where are these formed, how often, and why? Larry's proposal was to make systematic measurements off the coast of Portugal where it is believed they are formed. We do two things, first we make temperature profile measurements down to 2000 meters depth by dropping probes that have a very small wire connected back to the recording instrument on Kialoa. We go along a straight line, dropping these probes every 1.5 miles or so, and all the data is recorded by a PC computer. These are called "expendable probes", XBT''s in oceanographer lingo, and when they reach the bottom the wire is broken and they are left. We will probably have launched more than 1000 of these by the time we are through.

Second, we drop floats, called RAFOS floats, which contain some rather sophisticated electronic equipment. These floats, which are a glass tube about 5 inches in diameter and 6 feet in length, are carefully weighted so they will sink to the depth of the Mediterranean water and stay there, moving with the water and therefore enabling the motion of the water to be recorded. The way this is done is interesting and is a good example of modern day technology. Several sound sources have been placed out in the Atlantic, anchored to the bottom and suspended at 1 to 2 Kilometers depth. Three times a day, at precisely determined times, these sound sources emit a short burst of low frequency sound. Three have been placed specifically for this project, others have been placed by different projects and can be used as the floats move out into the Atlantic.

These floats hear the sound signal and record the time of arrival. Because of the known sound speed in water, the time of arrival from two sources can be used to locate the position of the float. This information, and also the temperature and pressure, are recorded in the memory of a small computer in the float. After a preset time, which can be up to one year, this computer signals the end of the mission and the weight attached to the float is released so that the float comes to the surface. A small antenna in the part of the float above water broadcasts the results of all these measurements to the Argos Satellite System. Then, if all has worked well, these results, taken over the last year, are obtained for analysis. And the motion of the Mediterranean water, into which they were dropped, can be plotted to see if it has become part of a Meddy.

Typically we go about 20 miles offshore and then do a 30 mile line with 25 temperature profile measurements. From a look at the Med water as shown on the temperature profiles we then drop two of the RAFOS floats at good locations. We are now also recording another line of temperature profile measurements on the way back, at a different place. This takes about 20 hours, and is carried out once a week, weather permitting, leaving Saturday morning.

We also have recovered about 15 of the Rafos floats which have surfaced. Initially several were programmed for a 1 month mission and we recovered them to check out their condition. They were also returned to Woods Hole for refurbishment and re-use. This was interesting, as recovering a 5 inch glass tube out in the middle of the ocean is a bit of a challenge. Again the technique is a good example of modern technology. From the radio signals to the Argos satellite system the position of the probe is determined every few hours with an accuracy of a few hundred meters. This information is telephoned to us from the staff at Woods Hole, and when we are at sea we can call by radio to get the latest positions. On Kialoa we use a standard satellite navigation system, called GPS, which gives our position to within 50 meters. On board we also have a directional receiver which listens to the frequency the probe transmits. When we are within 5 to 10 miles of the float we usually begin to pick up the signal, the receiver indicates the direction of the float and the strength of the signal, and we are able to close in on it. As we get close, sometimes having circled it, it becomes a game to spot it first. Normally someone sees it from about 30 or so meters distance, but the last time I was alone and almost ran over it, accidentally spotting it as it went by about 10 meters from me. We have then been able to bring the boat alongside upwind of the float and someone can lean way, way over from the deck and grab it with one hand. So far we have 100% recovery on the searches.

A serious problem with these floats arose shortly after beginning the project. Several floats came up unexpectedly and we went out and recovered some of these. They had developed a leak where an aluminum end plate was sealed to the glass tube. A small amount of water leaked in and they slowly sank; when they reached about 2000 meters depth the computer ordered the weight dropped and they surfaced. Unfortunately the solution was not easy and we took a vacation for 4 weeks while some changes in the sealant used was made. The first fix was even worse, and 3 of the 4 probes launched leaked, sank and then came to the surface in a few days. The second solution seems better, of 10 or so launched only one has leaked and surfaced so far, hopefully this was an anomaly. As a result the project has been delayed and instead of being completed by the end of 1993 we will finish in March 1994. I hope there are no more delays.

My friend Frank Ansak, who joined in Sri Lanka, has been first mate and general handyman. Rita, who had worked with Larry at Scripps, handles most of the measurements and also communications with Larry and Amy back in the States. She has an apartment here but is on every trip. Cynthia, who is from Sri Lanka, takes care of cooking, and also two or three students from the University of Lisbon have come to help on most of the trips. On some trips Prof. Isabel Ambar, or her collaborator Fatima Sousa, have joined us. Larry has been out three times so far, generally for 2 weeks . (He will come in February with his wife Eva and daughter and I am looking forward to having them.) So we normally have a crew of 7 for each trip. The trips are made mostly under power with only occasional sailing, but we generally use small sails to steady the boat and cut down the rolling. The trips are work, requiring 24 hours straight for the routine measurements. But the results are interesting and will help understand some of the mixing of the Med water with the Atlantic water, and hopefully the formation of the Meddies. One of the floats we launched, which had a 30 day mission, actually got caught up in an eddy of some sort and performed a dozen circles or so as it drifted out a little ways into the Atlantic.

Notes added in 1996: All of the floats we launched came up within one year and I believe most of the data has been analyzed. A preliminary paper has been published and final papers are now in preparation.

After we left a neighboring boat, "Monara", helped in the project by going out once a month to make a series of temperature profiles along the same lines we had been recording. In that way Larry obtained a complete year's worth of data on the profiles of the Mediterranean water off the coast of Portugal. He says the data is most interesting. I am waiting to see the final published paper.

Copyright 1996 by Frank Robben