An apology to our esteemed moderator - it doesn't seem like this belongs 
in sbe, but I don't know where else to bring it. This particular slant on 
hydrology and hydraulics never came up in sci.engr.civil.


Guy Hoelzer  wrote in
news:c097po$1pa7$1{at}darwin.ediacara.org: 

> in article bvsn64$13jf$1{at}darwin.ediacara.org, William Morse at
> wdmorse{at}twcny.rr.com wrote on 2/4/04 10:17 PM:
 
>> Guy Hoelzer  wrote in
>> news:bvrd1q$lol$1{at}darwin.ediacara.org:
 
>>> Erosion, for example, plays a critical role in the process of
>>> optimal drainage system (e.g., stream/river systems that drain
>>> surface water from higher to lower elevations) structural
>>> modification. That is, the shape of streams and rivers is constantly
>>> changing, and the process of erosion helps to ensure that each
>>> internally generated change maximizes the rate of flow.
 
>> No, it doesn't maximimize the rate of _flow_. That would be maximized
>> by straight clear channels.
 
> I am using the term "maximization" in a different way than you.  I
> learned my definition from theoretical physicists, and I have come to
> prefer it strongly to your usage, which I used to share.  I now take
> it to indicate an effect that always increases the value of a
> parameter of interest, rather than a process leading as system to
> attain the maximum possible value of a parameter.  In this instance,
> erosion always results in increasing the rate of flow.  If you could
> manipulate a natural stream system so that it is prevented from
> exhibiting erosion, then it will no longer be able to respond to
> changing conditions (e.g., in the rate of water input at the top) by 
> optimally re-configuring its structure resulting in a lower rate of
> flow than it otherwise could have attained.


First note that, despite my knee-jerk reaction - and regardless of 
whether you and your brain-addled theoretical physicist friends want to 
call "maximize" what normal human beings  would call 
"increase" :-) - 
changes to a stream channel do not affect the _rate_ of flow. To explain: 
the rate of flow (cubic volume per second) is determined primarily by 
hydrology, and is relatively unaffected by the hydraulic effects of 
stream channel changes, except as those influence the time of 
concentration. 

Second we note that, as an overall process in a watershed, erosion moves 
soil from areas of steeper gradient to areas of lesser gradient. 
(Actually I am including deposition with erosion, but in my mind - and I 
think in Tim Tyler's discussion of the subject - the two are linked). 
This process decreases the overall gradient in the watershed, which 
increases the time of concentration, which decreases the rate of flow. 

For a particular segment of a stream, an increase in flow (due to other 
hydrologic changes such as increase in rainfall or decrease in 
infiltration due to urbanization or other change in land use/land cover) 
will increase the velocity in the channel, which will increase erosion, 
which will - at least on steeper slopes -  generally increase the 
hydraulic radius and therefore increase the channel capacity. This will 
tend to concentrate flows into the channel, leading to the mentioned 
increase in stream network order. But once a stream network order is well 
established in a mature landscape, such as the Allegheny Plateau, it is 
not clear to me that a further increase in flow will in fact have any 
change other than to increase the channel capacity, which itself will not 
affect the rate of flow. 

Now we get to the tricky subject of meanders, which I don't begin to 
understand, but which I see as critical to the argument. Unless I am 
mistaken, the net effect of meanders is to decrease the average velocity 
in a stream, which will,albeit only slightly, decrease the rate of flow.
Meanders may be a good example of a dissipative system, and I would be 
happy to see an explanation of how they represent useful work (a la 
Lotka) or self organization (a la Kauffman), but I remain unconvinced 
that they increase the rate of flow. 


Yours,

Bill Morse
---
þ RIMEGate(tm)/RGXPost V1.14 at BBSWORLD * Info{at}bbsworld.com

---
 * RIMEGate(tm)V10.2áÿ* RelayNet(tm) NNTP Gateway * MoonDog BBS
 * RgateImp.MoonDog.BBS at 2/16/04 7:29:44 AM