Kelli Hammond

12-6-00

Geo 306

 

Stream Capture: A Look at Natural Thieves

Introduction:
Stream capture is an interesting phenomenon that can occur when two streams ‘run into each other’ by one of a number of different means. Other terms used in reference to stream capture include stream piracy, river capture, and river piracy (Fairbridge 1968). The Boardman River, near Traverse City, MI., has undergone stream capture. This paper discusses the mechanisms of stream capture and how the Boardman River is an example of it.

Study Area:
Stream capture occurs when two streams (usually one actively eroding low level stream and one higher level stream) come together and the latter stream begins to fully or partly flow out the lower level stream channel. The capturing stream ‘erodes more aggressively than an adjacent stream and captures its discharge by intersecting its channel’ (Summerfield 1991). There are different ways that stream capture can occur including abstraction, headward erosion, lateral planation, and subterranean diversion (Fairbridge 1968).

Stream capture by abstraction is among most common types of capture. As the lower level stream headwardly erodes the elevated divide between the two river systems, the higher level stream will eventually be undercut and will then divert itself into the lower level stream (Easterbrook 1969). Capture by headward erosion is similar to capture by abstraction but it seems a bit simpler. In this case, tributaries of the lower level stream cut back into the divide between the two rivers and eventually almost the whole divide will have been cut through. Due to an increase in gradient, the river on the other side of the divide will now flow out of the channel that originally cut into the divide (Fairbridge 1968). The specifics of lateral planation and subterranean diversion will not be discussed in this paper but they are equally if not more interesting forms of stream capture.

A distinct characteristic of a river that has undergone capture is a feature called the elbow of capture (Summerfield 1991). This ‘elbow of capture’ represents a distinct change in the direction of the channel. The change in channel direction is made when capture has taken place, and it represents the point where one stream took over another. A better way to explain it may be that the elbow of capture is the point where the headwardly eroding stream intersects the other low gradient stream (Summerfield 1991).

Another important surface characteristic of stream capture is called a wind gap. The wind gap represents the abandoned river valley of the captured stream (Summerfield 1991). The wind gap and the elbow of capture occur in the same general location.

To further enhance understanding of stream capture it is necessary to examine a real example of a river system that has undergone stream capture. The Boardman River is a prime example and is further discussed in the remaining portion of this paper.

Methods:
In order to more closely examine the Boradman River as an example of stream capture I studied a glacial sediment (Quaternary structure) map of Michigan and topographic maps of the area being studied (Grand Traverse County area). I first examined the glacial sediment map to look for the different types of sediment and depositional processes in the Boardman River drainage basin. I then examined four topographic maps of the area that the Boardman River flows through: the Mayfield quadrangle, Jacks Landing quadrangle, Traverse City quadrangle, and the Grawn quadrangle. I used these maps to locate the elbow of capture for the Boardman River, and to then determine the amount of sediment eroded out of a section of the glacial moraine in that area. This calculation is important because it gives information on the scale of the stream-capture event. I then took the same measurement for Mitchell Creek (figure 4) in order to compare the two.

In looking at the topographic maps I found a section of the river in the Mayfield quadrangle that appeared to be the ‘elbow of capture’ or an area close to it (figure 1). This area is located a little northwest of the small town called Mayfield (figure 2). It is obvious by looking at this area on the map that a lot of sediment has been removed from the moraine by the river. This is evident because of the steep gradient on either side of the river and its floodplain (figure 1).

In order to calculate an approximate volume of sediment loss I chose a portion of the river beginning at Brown Bridge Pond and ending at the river’s outlet, West Grand Traverse Bay. To calculate the volume of sediment loss for this section of the river it was necessary to further section the river into rectangle like pieces to obtain a more accurate calculation. I sectioned this portion of the river into nine parts and calculated a volume of sediment loss for each section. To get the approximate total volume of sediment loss I added these nine values together. To find each volume I simply multiplied the change in height by the width of the river valley by the length of the section of river. The change in height is found by subtracting the lowest point in the river valley from the highest point on the moraine. Please see figure 3 for a more detailed explanation of this volume calculation.

For comparison reasons I also calculated the approximate volume of sediment loss due to a small creek in the same area as the Boardman River. This creek is called Mitchell Creek. I used the same method for calculation of this value as I did for the Boardman River. I sectioned a portion of Mitchell Creek into three sections and found each volume and then totaled the three volumes for a grand total (figure 5). The importance of comparison is explained in the discussion part of this paper.

Results:
In examining the Quaternary glacial sediments of Michigan map I found that the Boardman River flows through four different types of sediments. These sediments include glacial outwash (sand and gravel size), post glacial alluvium, end moraines of coarse textured till, and sand and gravel sized lacustrine sediments (Farrand 1982). The importance of this information is that the Boardman River was eroding these sediments before and after it was captured. The main point in examining the glacial sediments was to determine that there was indeed a moraine involved in the capture of the Boardman River. This will be further examined in the ‘discussion’ portion of this paper.

I found the approximate volume of sediment loss due to stream capture by the Boardman for a specifically defined portion of the river (Brown Bridge Pond to West Grand Traverse Bay) to be 1.7 x 109 m3. This is a very large amount of sediment. For Mitchell Creek the total volume of sediment loss was approximately 7.2 x 106 m3. Compared to the calculation for the Boardman, this is a significantly lower amount of sediment loss (see next section for interpretation).

Discussion:
The Boardman River was once likely to have been a small stream flowing into West Grand Traverse Bay with few or no tributaries. At one point a long time ago the Boardman River may have been similar to the current Mitchell Creek (figure 4), but since it was captured there is now a very different story. The glacial end moraine south of Traverse City was once a drainage divide between two separate river systems. Through headward erosion of the moraine by the once much smaller Boardman, these two separate systems grew closer and closer together. The river system on the more southern side of the moraine was likely traveling in the east/west direction and therefore had a very minimal gradient. With the continued headward erosion by the Boardman the two river systems eventually met and the east/west flowing system soon found a better ‘way out’ via the Boardman River channel. It only makes sense that the east/west flowing river would want to flow out through the Boardman River channel because, after capture, the gradient is much greater and the water can move much faster, carry more sediment, and has a quicker route to base level (Easterbrook 1969). The local base level for the Boardman River is West Grand Traverse Bay. For the east/west flowing river it was most likely Lake Michigan. By flowing out through the Boardman River channel, a quicker route to base level was found. This was the event of capture.

As a result of this capture new power was given to the Boardman. Much more erosional work could be accomplished. It is evident when one looks at a topographic map of the Boardman River area that there is an enormous river valley. This great river valley was cut when these two rivers became like one and the gradient increased, so more erosion of sediment could be done. Please see figure 1 again for a better picture of this great river valley.

As stated in the results the amount of sediment transported out of the now Boardman River valley is on the order of 109 m3. This is an enormous amount of sediment loss, and is accounted for by the obviously large river valley. When compared to Mitchell Creek, and its sediment loss due to fluvial processes it is evident that the Boardman has transported out much greater volumes of sediment. Before capture, the Boardman River system may have been very similar to Mitchell Creek. This data suggests that if the Boardman had not undergone capture it would not have transported so much sediment.

This example of stream capture demonstrates that in the event of stream capture increased erosion can and will occur. This will result in a larger river valley as seen in the example of the Boardman River. Another affect of stream capture is an increased drainage network system. The Boardman River’s drainage basin was expanded greatly by this event of capture and it has tributaries as far East as Kalkaska and as far South as the Kingsley area (figure 2).

Conclusions:
Some obvious affects of stream capture are increased erosion, increase in discharge of the capturing river, and an increase in the drainage basin of the capturing river. The Boardman River case of stream capture demonstrates an example of each of these affects. There are, of course, other affects of stream capture that are evident in different instances. The once small ‘creek-like’ Boardman River is now a vast and elaborate river with a large drainage basin and a significant river valley. The effects of stream capture on the Boardman are highly evident and have affected it greatly.


References

Easterbrook, Don J. 1969. Principles of Geomorphology. McGraw Hill Book Company. New York. Pg 155-163.

Fairbridge, Rhodes W. 1968. The Encyclopedia of Geomorphology. Encyclopedia of Earth Sciences, Volume III. Reinhold Book Corporation. New York. Pg 1054-1057. 

Farrand, W.R. and Bell, D.L. (1982). Quaternary Geology (map) of Southern Michigan with surface water drainage divides. 1:500,000 scale. Dept. of Natural Resources, Geol. Survey Division, Lansing, Michigan.

Summerfield, Michael A. 1991. Global Geomorphology: An introduction to the study of landforms. Longman Scientific and Technical. England. Pg. 410-411.

FIGURE 1.
Fig2.jpg (2347597 bytes)


Figure2.jpg (795264 bytes)

 

 

 

Figure 3: Calculation of an approximate volume of sediment loss due to stream capture by the Boardman River.   To see map examples of first five ‘sections’ see figure 1. The data from these sections are as follows:   
Note: Volume = h*w*L (m3)

Section 1:

D h=46m (276m-230m, in elevation)

w=2222.5m {on map: 3.5 in *(1ft/12in) * (.3048m/ft) * 25,000 (map scale) = 2222.5m}

L= 2222.5m (same calculation as above)

Total Volume = 2.27 x 108 m3

Section 2:

D h= 51m

w=1270m

L=2540m

Total Volume =1.65 x 108 m3

Section 3:

D h=58m

w= 1587.5 m

L= 1746.25 m

Total Volume = 2.63 x 108 m3

Section 4:

D h= 62m

w=1587.5m

L= 2857.5 m

Total Volume = 2.81 x 108 m3

Section 5:

D h= 60m

w= 1900m

L= 952m

Total Volume = 1.08 x 108 m3

Section 6:

D h= 61m

w= 2381.25 m

L= 3492.5 m

Total Volume = 5.07 x 108 m3

Section 7:

D h= 28m

w= 2857.5 m

L= 1111.25 m

Total Volume = 8.89 x 107 m3

Section 8:

D h= 15 m

w= 952.5 m

L= 5715 m

Total Volume = 8.17 x 107 m3

Section 9:

D h= 8m

w= 635 m

L=2222.5 m

Total Volume = 1.13 x107 m3

Grand Total Volume for whole section of river,

(Brown Bridge Pond to West Bay) = 1.7 x 109 m3



Figure4.jpg (732707 bytes)



 

Figure 5: Approximate volume of sediment loss calculations for Mitchell Creek (see figure 4 for example of sections).

Section 1:

D h= 3 m

w= 317.5m

L= 1587.5 m

Total Volume = 1.5 x 106 m3

Section 2:

D h= 3 m

w= 952.5 m

L= 1905 m

Total Volume = 5.4 x 106 m3

Section 3:

D h= 1m

w= 317.5m

L= 952.5m

Total Volume = 3.02 x 105 m3

Grand Total Volume for highlighted portion of Mitchell Creek (fig. 4) = 7.2 x 106 m3