PDA

View Full Version : Hydrographic (?!?) question: artificial river port



feanaaro
09-15-2013, 04:09 PM
If a city (ancient/medieval level technology, though I am not concerned about absolute realism in that regard) along a river, which in that particular tract happen to have a rather tumultuous course, would want to have a calmer artificial port, could it obtain that by channeling waters to an artificial pond/small lake? Perhaps with a mobile dams system? or is that not even necessary (i.e. simply channelling waters to an artificial pond would create a less agitated environment?
If that could be done, what would be the best way (especially in terms of shapes of the channel/port system, which is what would affect the actual map)?
I don't know where to start with this, and I would like some suggestions. Many thanks.

Hai-Etlik
09-15-2013, 05:09 PM
London's artificial wet docks would be an example although they were more designed to provide secure space off of the main river rather than because of the speed of the Thames (So slow it flows backwards at times). I expect any river where the speed of the flow is a significant problem probably isn't going to allow passage of large ships that would need any significant port infrastructure, although I'm not an expert on this.

London's docks are just big, generally rectangular lakes attached to the Thames by canals.

London Docklands - Wikipedia, the free encyclopedia (http://en.wikipedia.org/wiki/London_Docklands)

London is built on a big, flat, easy to dig floodplain. A "tumultuous" river is going to be in terrain that's much less amenable to digging artificial lakes at river level. More rock and much less flat.

It would make more sense for a large river port to be as far up the river as possible before it becomes hard to navigate for large ships, at which point artificial wet docks would make sense to provide a secure, off river place to moor ships, just like in London. On longer rivers, the ports might be staged, with a port for ocean going ships as far along as they could navigate, then ports further up for large river boats. I expect this was the case but I don't know the details off hand. It would be worth doing a bit of research into medieval river travel.

The other thing to consider is bridges. IIRC, London is situated at the point where the Thames first becomes practical to bridge (at the time it was founded). Bridges and large ships don't mix particularly well, particularly with medieval technology. So the first bridge will tend to restrict the location of a port (hence the Port of London). Movable bridges or really tall, wide span bridges aren't possible without relatively modern materials and engineering. Rivers would tend to be crossed with ferries more than bridges.

Throwing in some magic (Dwarves built it), can mitigate some issues but it's usually best to do so in moderation, and knowing that you are doing so.

feanaaro
09-15-2013, 05:19 PM
Thanks Hai-Etlik. Bridges are actually part of the equation.
The point is that this is a huge and very long river (15% longer than the Amazon river, roughly), to the point that it cannot be bridged for a very long span of its course. This city is in a place where the river has some islands that splits it and make it bridgeable, that are themselves the result of a rocky bottom in that area (which I figure could be brought to surface by soil erosion, and the islands would simply be random clumps of slightly harder, thus less eroded, rock). The narrowing of a large river, plus the kind of bottom it would have in that tract, would make (I guess, but I don't really know) for a more turbulent flow. Thus the city has to be there, for otherwise no bridge, but it would use a system of canals and artificial port to avoid the problem of the rougher flow. Possibly, or at least that was my idea, the system could be used to completely sidestep the area with the rocky islands (which is comparably small).
I hope I was able to make my idea clear.

Caenwyr
09-16-2013, 04:23 AM
If the rocky area is relatively small, I'd think the river would flow around it, and not over/through it. There's no reason for a river to start eroding a rocky outcrop if it can avoid it: water generally follows the path of the least resistance ;-).

You could however have rocky islands in the middle of a stream as remnants of e.g. glacial erosion (large rocks deposited there after the retreat of a glacier, and just now being uncovered by the eroding of the river). This would result in rocky islands in a still relatively smooth flowing river, eradicating the need for an artificial port altogether.

There's one restriction to that theory, however. Such huge boulders could only have been taken there through glacial erosion, which means it should be in an area that once sported a glacier, an ice field or an ice cap (any of the three would do). Which in turn means the topography should still reflect that in some way. Check out the various post-glacial landscapes if you want to know if one fits the area your city is in.

feanaaro
09-16-2013, 09:33 AM
If the rocky area is relatively small, I'd think the river would flow around it, and not over/through it. There's no reason for a river to start eroding a rocky outcrop if it can avoid it: water generally follows the path of the least resistance ;-).

The idea is that the river flowed as normal through the path of least resistance, made of regular soil. However, by eroding the soil at a certain point it exposes the rocks underneath, which are not smooth but have "spikes", thus the rocky bottom and the small islands. This entire rocky formation relatively near to the surface intersects the flow of the river more or less perpendicularly, thus the river does not go around it but through it, after having eroded enough soil to uncover it at a certain point.

Caenwyr
09-16-2013, 11:57 AM
The idea is that the river flowed as normal through the path of least resistance, made of regular soil. However, by eroding the soil at a certain point it exposes the rocks underneath, which are not smooth but have "spikes", thus the rocky bottom and the small islands. This entire rocky formation relatively near to the surface intersects the flow of the river more or less perpendicularly, thus the river does not go around it but through it, after having eroded enough soil to uncover it at a certain point.

Actually a river meeting an obstacle perpendicular to its original path would, at first, just pool up behind the obstacle in the form of a lake. Eventually it would, by pooling up, reach a height where it would again be able to flow freely, which would be either:

a low point in the rocky structure, in which case this "overflow" would gradually (in the course of several thousand years) be eroded into a smooth channel, eliminating the lake behind it, or
a low point next to the rocky structure, in which case the river would divert its entire flow to that point and, after enough erosion, eventually drain the lake.


In short: the river would first flow over the rocky structure with a lake behind it, then flow through the structure in a relatively smooth channel, or around the structure altogether if the obstacle proved too great.

Caenwyr
09-16-2013, 12:55 PM
One last remark: I realize you're looking for a way to keep your city concept both alive and realistic, so suppose the following scenario. What if the river only just reached the bedrock, and the bedrock turned out to be both very horizontal and very large in area (so the river can't pool up behind it, nor can it flow around it). That would leave the river with only one choice: flow over the surface of the bedrock.

There's one problem with that scenario, however. A bedrock like that would most probably be sedimented seabed which was subsequently petrified and by some miracle remained horizontal instead of curling up under the geological conditions that are normally bound to happen in the timeframe necessary for a sediment to petrify. Even if you believe this unlikely (but possible) scenario, the bedrock would have no spikes whatsoever, since sand/silt/clay simply doesn't sediment that way. It sediments in smooth, nicely horizontal layers. A river meeting such a bedrock would suddenly meet a strong vertical resistance, and would therefor start eroding its own banks, resulting in a very wide, very shallow river - and that over a pretty long distance (remember, we're talking about a very large area of bedrock). Totally unnavigable, that is.

So yeah, that scenario wouldn't work either. Water's a pain in the ass, it almost never does what you want it to do. Think of choppy waters and it will almost always be a very temporary situation - far shorter than it would take a city to develop on its banks.

feanaaro
09-16-2013, 03:00 PM
Sorry, I wasn't clear. The idea is that the river was flowing normally on the surface stratum of soil, which it slowly eroded, so that at one point, where the bedrock is closer to the surface for any reason, the bedrock gets exposed. I'm not trying to have a river flowing uphill or something.
Besides, scrap the bedrock, if a river course is narrowed by one or more islands at a certain loin, should not it become more turbulent anyway?

Caenwyr
09-16-2013, 03:58 PM
Sorry, I wasn't clear. The idea is that the river was flowing normally on the surface stratum of soil, which it slowly eroded, so that at one point, where the bedrock is closer to the surface for any reason, the bedrock gets exposed. I'm not trying to have a river flowing uphill or something.
Besides, scrap the bedrock, if a river course is narrowed by one or more islands at a certain loin, should not it become more turbulent anyway?
You'd think so, indeed, but actually no. A river's discharge remains roughly the same before, at and after a narrow stretch. Which means it will simply be deeper where it's narrower (if not water would build up behind it, forming a lake).

I think I've found the solution, however. There are rivers that are very large and yet pretty turbulent. All you'd need is a relatively sharp drop (say, a few 100 meters over a distance of a few 100 kilometers). The lower Congo River is an excellent example of that, it turns out. If you want to know more, be sure to check out this website (http://www.thefreelibrary.com/Rough+waters%3A+one+of+the+world's+most+turbulent+ rivers+is+home+to+a...-a0198169413).

feanaaro
09-16-2013, 04:03 PM
Interesting. So we could sum up by saying that turbulence depends exclusively on the sharpness of the drop, while, for a given flow rate, depth is inversely proportional to width.
Is that right? I'm not terribly wed to the turbulence thing, if islands narrowing the river make it deeper that might actually be good for a port in itself, without needing much channeling.

waldronate
09-16-2013, 04:04 PM
I think the point here was that if the river uncovers bedrock, then it will try very hard to go around. If that bedrock is at the mouth of a large valley that's recently (in geologic time) had an uplift, then your scenario is reasonable to a certain point.

A river's gradient profile (steepness over its course) tends to look suspiciously like the positive half of a 1/x function. That is, it starts out steeply in the mountains and becomes essentially flat in its lower reaches for long rivers. The energy of the flowing water in the almost-flat sections goes into side-to-side cutting, making meanders. The gradient also tells you the kinds of things that the river can carry. A steeper gradient lets the river carry larger sediments. For a large river near its mouth, it's carrying little more than very fine clay and dissolved materials. The rest has settled out upstream. One of the reasons why a river tends toward near flatness in its lower reaches is because the dropped sediment becomes new land. Floods raise the river level and increase its power to carry larger sediments. These sediments are carried off to the sides of the river when it overflows its banks, forming natural levees (and when the river hits the completely-flat ocean, it drops anything that's not dissolved pretty rapidly).

If the gradient of the lower part of a river is made steeper by uplift, then the river will start cutting down until it can be as flat as possible. It sounds like this is the sort of thing that you're looking for.

If the river's course is narrowed for whatever reason (landslide, log jam, or what have you), then the river will become more turbulent, which increases its cutting power. Islands in a river are unlikely to appear; if they do, the river will use that extra energy to cut its banks out just as quickly as it can to restore its flow.

Having channels that go around rapids on a lake usually involves a lock system, and for good reason. Without the locks, the bypass channel will be at least as steep as the main river channel, and likely steeper. The river will want to flow down the steeper channel, meaning that the new bypass channel and ponds will eventually capture the main river unless careful steps are taken.

Ancient Portus on the Tiber River is analogous to what you're asking about, I think. The engineers dug out a large harbor, but the reason was more about having a better harbor than bypassing a river, even though there were bypass channels in place. I'm not sure how well it would have worked if the river had been relatively swift-moving at that point.