Phil Richardson Interview




Q. A lot of the material I've been reading about the southern tip of Africa describes how the Agulhas rings introduce warm salty water from the Indian Ocean into the cooler and less salty Atlantic and its influence on the overturning circulation. How significant an input is this, say compared to water from the Mediterranean?

A. Models have shown that it is significant. The Med increases the heat and salt at around 1000 m depth and below which increases the salinity of the deep water flowing southward. I think the leakage salinity is important because when the water eventually reaches the northern North Atlantic the salinity is important for creating deep mixing and the formation of deep water. If the near surface water were fresher (say from melting of glaciers and heavy rains or from fresher MOC) then the deep water formation could be shut down or at least reduced. If the near surface water were salty then the deep water formation could be enhanced in that when cooled the salt results in dense water and therefore deep water--given lots of wintertime cooling. Of course there are ocean-atmosphere feedbacks and this subject is very complicated, but I would say that having more salt flux from the Indian Ocean does affect the MOC and deep water formation, but it is not simple and is being researched especially in light of climate warming.

Q. Is it a major source of salt in the Atlantic? Does this saltier water eventually get channeled up through the Gulf Stream, or does it get mixed into the wider ocean?

A. The warm and salty leakage water disperses into the South Atlantic especially in the very vigorous mixing and stirring environment in the Cape Basin (off Cape Town) where energetic Agulhas current rings and cyclones interact and exchange water. Some rings translate across the mid-Atlantic Ridge carrying Agulhas water a long way. The blended mixture of waters in the Cape Basin is also carried by currents across the mid-Atlantic Ridge. Near the western boundary of the South Atlantic the flow divides, part (~15 Sv) going north toward the equator and eventually into the Caribbean and Gulf Stream and part (~15 Sv) recirculates southward in the Brazil Current. So, yes the leakage does get channeled up through the Gulf Stream but often in complicated paths--western boundary currents, eddies, upwellings, etc. And it can get sidetracked for various amounts of time going around the South Atlantic subtropical gyre and back into the Benguela Current again.

Q. Without this input, would the overturning circulation be effected?

A. People have modelled the ocean circulation with and without the leakage and they find that the leakage does have a strong effect on the MOC. Of course the whole circulation is very complicated and models need to be verified. One way to think of it is that given a MOC, you can get water from the Drake Passage or from the Agulhas leakage. Drake passage water is cooler and fresher than Agulhas water. So, more Agulhas water means more heat and salt flowing through the Atlantic, Drake Passage water means less heat and salt. As the leakage goes north, it looses some heat in Cape Basin during winter due to cold weather, but then it gains heat on its way to the equator and near the equator. So as the MOC goes northward it increases in heat and the heat flux is northward (warm water going north, cold water going south underneath)throughout the Atlantic. And there is more evaporation than precipitation (E-P) in the subtropics so salt increases as the MOC goes north, and part of the warm salty water subducts under the surface and flows equatorward. Near the equator there is lots of upwelling and the cooler upwelled water warms up due to strong heating there. This water is carried north the the North Brazil Current and its current rings toward the Caribbean.

Q. I'm a little confused about why some of the water off the coast of Brazil gets subducted and heads north. Is it due to some properties of the water, or is it because a lot of currents are coming together, and the water has nowhere else to go.

A. The wind stress in the subtropics, both north and south Atlantic subtropics, generates a near surface convergence of water and near the equator divergence. Its a pull push question. Water is subducted because of the convergence and because of buoyancy forcing and it upwells near the equator because of the strong divergence there. So water from the South Atlantic is partly subducted and partly it flows from the Cape Basin and is modified as it goes by stirring and mixing. Most of the MOC flows northward along the coast of South America 5S-10S as a swift western boudary undercurrent--under a near surface layer heading south and driven by local winds--the Ekman flow. Maximum northward currents are located at a depth around 200-300 m but northward flow extends down to around 1000 m. Near the equator a lot of the 200-300 m water upwells and eventually continues north (after going eastward and westward in the equatorial jets) in the North Brazil Current which is surface intensified.  The deeper water--intermediate water--also continues northward along the boundary. The NBC breaks up into rings which carry over half of the water up the coast. (The MOC is driven in part by density of the water--the thermohaline circulation--and in part by the wind. They both are important but the ratios differ in different places.)


Q. Are the properties of the subducted water the same as those of the water that joins the Brazil Current heading south?

A. I think that the water heading westward as the MOC and as the recirculation of the subtropical gyre is blended somwhat by the strong mixing in the Cape Basin. There might be more leakage water on the northern side than southern side of the westward flow, but it is stirred quite well, plus mixed with water heading south from the equator along the west coast of Africa which is recycled north again. Its not easy to use water properties to answer your question as far as I can tell.

Q. Does the subducted north-bound water resurface at the equator?

A. Yes, plus some water from the North Atlantic that has subducted there.

Q. Now for easier questions. My impression has always been that sailing around the southern tip of Africa could be pretty hairy due to all of the powerful currents. Is this true?

A. The major problem is that the Agulhas is swift, around 5 knots, and it feeds directly into winds that are blowing typically from the west and are also strong. This creates huge seas there. I have read that an aircraft carrier had its flight deck dammaged--peeled back. There are stories of super tankers breaking in two down there. The seas are unimpeded by land so the fetch is huge and the seas can grow to be very large. It is like a strong northeastward wind that blows against the Gulf Stream. You don't want to be there.

Q. Is this true? What was it like for you when you were on a ship down there?

A. We mainly worked in the Benguela region north of the Agulhas, but even that was rough when we had one storm come through. Water forced its way through a deck locker into the electronics room and started shorting out the equipment which started smoking (salt water is bad). And the seas were sweeping the deck which made it difficult and dangerous to work out there. But we got off light compared to some of the other cruises farther south. On the way back toward the equator in the southeast trades we had day after day of beautiful weather. Stopped on St Helena for a day. Compared to the Gulf Stream region the south Atlantic seemed quite empty--no jets, no ships sighted, no garbage in the water etc.

Q. Did you wish you were in the Caribbean instead?

A. Yes, although the Caribbean in hurricane season can be deadly. But in winter it is definitely the place to be.

Q. I read with interest about how you used the ship drift data to track Columbus's course. Was that the final word, or does the controversy still thrive? I have a bit of a rooting interest. I took a marine ecology course on San Salvador Island in college. The little monument they have there commemorating the landing is just a little more exciting than that for Plymouth Rock.

A. The controversy rages as far as I can tell although the fighting slowed after the quincentenial. We did some more calculations about the track and can get it to land at Grand Turk, one of the potential landing sites. The track depends on the magnetic variation in the west which is unknown, so depending on what variation you assume you can get the track to end over a wide region, covering all the reasonable landfalls. We noticed in a later work that on the earliest charts the islands are all plotted too far to the north. You can say that is due to Columbus' dead reckoning. If you agree to that you can then estimate the varitation in the west and use that to get a land fall which is near Grand Turk. But still the landfall is not well known despite lots of research. It is fun to read about. We flew over all the islands trying to identify them using Columbus'log. But of course the islands have had most of their trees cut down etc so its not easy matching them up. And there are zillions of islands down there.