CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Main CFD Forum (
-   -   Modelling ocean currents of the past Earth (

pgm February 25, 2005 09:07

Modelling ocean currents of the past Earth

Does anyone know, has someone tried to model the global ocean currents of the past Earth, say 300 million years backward in time upto present date? Provided that the computed shapes of the continents are accurate, then a knowledge of accurately known ocean currents would certainly be helpful in understanding the climate of the past. It might for example explain the sudden extinction of the dinosaurs some 65 million years ago. In any case, the global ocean circulation is the most significant factor which determines the climate. It does, however, seem like a difficult task to do this, since the ocean currents seems to be very sensitive to other effects, like the salt content in the water. In my opinion, the latest ice ages couldn't have occurred unless the Gulf stream got turned off for some reason (what reason?).

Tom February 25, 2005 09:40

Re: Modelling ocean currents of the past Earth
People have looked at ocean simulations from the past (paleo-climate type studies). However since most climate models run with timesteps of the order of 20 minutes performing a simulation for 300 million years involves a serious amount of computer time - unless of course you have a supercomputer more powerful that anything the big meteorological/climate establishments own?

Another important point is that over such a period of time you will have to account for the continental drift and the changes in the bottom topography - all of which are a significant unkown.

pgm February 25, 2005 15:31

Re: Modelling ocean currents of the past Earth
>Another important point is that over such a period of

>time you will have to account for the continental drift

>and the changes in the bottom topography - all of which

>are a significant unkown.

Yes, i'm interested in just that. 250 million years ago the continents were in a big clump known as Pangaea. The ocean currents were completely different at that time than now - and therefore also the climate. And when continents moved apart, the ocean streams changed dramatically again. It has been said that the period about 130 million years ago to 65 million years ago was probably the warmest period ever (perhaps not taking into account the very early times). I would want the investigate the possibility that this was caused entirely by the ocean currents. In other words, the continents were placed so that they allowed for a very efficient global circulation, which was able to transport heat from the tropics to the arctic regions so effectively, that no polar caps existed.

Then about 70-100 million years ago some continents closed up again, and the "Tethys" sea in the tropical region was closed. Also the part between Greenland, Northern Europe and Northern America was perhaps closed (not sure about this, some simulations say there could have been a narrow gap, which was closed, to open up again later), allowing for no heat transport to the arctic region. All this could explain the resulting cold climate and ice ages starting 65 million years ago. Fortunately, we have the carbonate-silicate cycle, which can save us from such a runway glaciation, so eventually the climate became warmer again.

My real problem is that i need to rely on the simulations regarding the continental plates, i have seen a lot of them, and they all dont look the same. I mean just a small detail that two continents get together somewhere could stop an entire ocean stream, which again would affect the climate dramatically. In addition, would it be even remotely reliable to just sketch out the main currents by inspection, and looking at how hypothetical currents could form?

The time behavior of the ocean bottom topography does also seem to be important, since cold currents move along the bottom, and that is too difficult to model (or has anyone attempted such simulations too)?

Tom March 2, 2005 09:45

Re: Modelling ocean currents of the past Earth
Sounds an interesting problem. I suspect, from a climate point of view that the changes in the atmospheric motion, which drives the Ocean, will also be significantly different due to the change in the positions of the continents.

It might be worth starting with a simpler problem of a 2D barotropic Ocean (modelling the bottom topography using Ekman pumping terms) to see what happens. To my knowledge no ones tried to do what your preposing - most climate Ocean models cannot deal with the changing topography.

Knowing how to simulate (i.e. how to move) the continents will be a real challenge (which makes the 2D problem more appealing as a first go).

Good luck with this,


All times are GMT -4. The time now is 16:01.