I have seen th experiment carried out in flow tanks many times: you make a straight channel in the sand, and start the water flowing.
Within a short time, the straight channel will become a seies of loops-which will then get cut off and form “oxbows”.
Is this the result of turbulent flow of the water?
Wouldn’t conservation of energy dictate that a river would flow with the straightest channel possible?
Energy is still conserved, even if the river meanders.
If you let water flow across a flat surface it will start doing meandering movements all by itself (there should be a physical explanation why, but I leave it to someone else). If it flows in a riverbed these movements erode the sides and the curves will get gradually bigger and bigger.
It’s not so much turbulence as speed of flow (though maybe we mean the same thing). On the inside edge of a curve, the water has a shorter path to travel than on the outside edge of a curve. Thus, the inside tends to move a little more slowly than the outside, which tends to speed up. As water slows, there’s less erosion and perhaps even deposition of suspended silt. As it speeds up, there’s more erosion. So a small curve in the river tends to exaggerate into a large curve or an oxbow. Starting from straight, the water just magnifies tiny differences into large curves.
The only thing that would violate conservation of energy would be water moving uphill. While gravity does tend to pull water in a straight line downhill, the positive-feedback loop described above is a lot stronger than gravity.
I like to think of it this way. If there’s any tendency to a curve, it’s the outside of the curve that needs to push the water around the bend, so to speak, so that’s where there’ll be the most friction and thus erosion causing the curve to become more and more pronounced.
Rivers don’t flow on imaginary perfect globes. They follow gravity on top of different substructures, i.e. the dirt and rocks that are underneath. If these erode at different rates the channels will move regardless of what appears to be happening at the surface.
It’s release of energy. Lay your water hose in the driveway and turn the water on full blast. It squiggles and moves all over the place, right? Rivers are doing the same thing, it’s just a lot slower.
In the uplands, the river channel slopes downward towards sea level. For example, in the headwaters, the riverbed may be at elevation 300, whereas the mouth at the ocean is below sea level. The river loses energy as it drops in elevation.
When it gets to the flatlands, the slope is flattened out and the river has to move from side to side to lose energy.
I think it is well understood why once a meander starts, it grows. Now, due to limits of the real world, it it possible to start a channel with no meanders?
Oh, in the experiments is the flow rate of the water high enough to reach turbulent flow, or is it laminar?
Something rivers do ‘like’ (if I may be permitted to anthropomorphise) is a constant drop in feet per mile. (or meters per kilometers for those who ken the metric)
When a river does finally ‘cut’ through the narrow neck of an ox bow, the river bed will rapidly readjust itself up and down river from the breach to even out the now sharper drop through the cut.
BTW, a very small breach in the oxbow is all it takes to shift the river. The shorter path will have a steeper drop (initially) and will erode out very rapidly to capture the main current of the river. If one wants to ‘move’ a river, a small ditch covering the new path is all that is needed to do the job, assuming the new path is shorter than the old one. Once the connection is made, the river will rapidly cut a new channel following the ditch, and the old part will become a ‘cutoff’.
Lots of meanders on relatively flat terrain are the result of a pool that forms first, then drains as flow finds a clear channel. I assume the resulting stream bed ends up where the bottom of the pool had reached the lowest point.
In uneven terrain such as mountains, rivers will flow between the mountains where past events have eroded paths through them.
Since real terrain is not smooth and even, any flow, turbulent or otherwise, will erode the ground unevenly, and then induce differences in flow as the water travels around curves at different speeds.
Now I’m wondering if larger, faster flowing rivers will have a tendency to straighten out when they aren’t blocked by solid rock.
This was one of Albert Einstein’s favorite problems and it’s not at all trivial. See here for his explanation:http://ponce.sdsu.edu/milestone_einstein.html
spenczar’s link answers your first question. Coriolis force will cause a net transverse current in even an idealized perfectly straight & smooth river channel with initially laminar flow. Which will cause assymetrical bank erosion & seed the meander. Turbulent flow accelerates the erosive process, but isn’t necessary.
If your idealized perfectly straight smooth channel is also oriented exactly due East or West, then Coriolis forces would be minimized. But there would still be some net force between the flow adjacent to the North & South banks, which would induce a corkscrew motion to the flow. The inherent frictional assymmetry between the top surface abutting the atmosphere and the bottom surface abutting the riverbed would still cause assymetrical bank erosion & seed the meander, albeit more slowly than if the river course was oriented North/South.