River flow calculations for gabions 2

[mbeschenbach]

I am working with the native population in Panajachel, Guatemala. I am trying to design a river containment system (dikes) with 30º bank slope, covered with 9" x 2m x 4m gabions o/geotec fabric. The river gradient is 200'/mile, the width 70m, x 2m max depth (20,775 cfs)for max (1) 12hr period once a year. Normal flow is 1-4000 cfs or less. River velocity at full capacity is 21.2fps. Can I use gabion mattresses at that flow?, and if they fail, how would they fail?. Can occasional deadmen be attached to resist movement.?

 

[cvg]

9 inch thick and 21 feet/sec is not going to work. 30 degree slope is too steep. I would stay at 3H:1V or flatter for mattress. Anchors should be installed at least every 10 feet. Recommend you only use gabions where velocity is less than 18 fps and you will probably need 3 feet thick for the high velocities.

Gabions fail in different ways. High velocity flow can lift a section of mattress and move it. Debris can snag a wire and break it and then the riprap gets carried away by the current. Alternatively, one or two missing hog rings or a missing tie wire can cause the same thing. Holes or gaps in the filter fabric can allow soil to be piped out during high flow, creating a void below the gabions. High velocity flows can also just deform the baskets leaving a very rough surface. Scour at the toe of the slope can create a hole and the mattress slides in.

 

[mbeschenbach]

(for calculations sake)In a river with a calculated flow of say 30fps, does the liquid have the same velocities at the gabion boundary layer as the rest of the river or is it reduced?

 

[gbam]

McCaferri (sp) has software that one can use to size gabions and reno mattresses. However, the use of flexible protection like these may not be permanent. Have you considered the wire lifespan due to corrosion and etc... It the baskets fail, the riprap will fail as well. The rock is not sized to withstand those types of forces. Have you considered other options like soil cement or rock revetments (big riprap)?

 

[cvg]

boundary layer velocity is less than the maximum velocity. However, keep in mind that velocity calculated by Mannings equation is mean velocity, not maximum. Hydraulic physical modeling (scale models) are often used to estimate the velocity profile through a channel. However in practice, calculated mean velocities are generally used for design.

 

[mbeschenbach]

If one were to put a skin coat of 2" of concrete w/ 1/4" expansion joints on 18" centers on the top face of the mattress surface, would that eliminate the majority of the shear and raise the capability to carry much higher fluid velocities?.

 

[gbam]

Have you compared costs to gabions/renos with concrete against straight reinforced concrete channel lining? How about soil cement, I presume this will be constructed out of the water? There are a number of other types of bank protection like articulated concrete blocks. Can you provide a section and plan for your design? What is the overall size of the project?

 

[cvg]

how about just doing 4" thick shotcrete with WWF and forget about the gabions? skim coat of concrete on top of gabions sounds like a bad idea.

 

[mbeschenbach]

The project is the construction of a river wall extending approx. 1.6 miles down both sides of a 75 meter wide channel. We plan on using 7 levels of 2x4 meter gabions, as shown in the attached files. 3 levels below the river bed and 4 above. During the rainy season, the river basin gains 1-2.5 meters of fill from landslide and erosion material for its whole length. The majority of that fill is removed by sand and gravel miners during the dry season. Because of that, a solid river bottom is not feasible. Also, during the dry season, small rainfall, thundershowers, erode the banks up to 2 meters. When the heavy rains come they once again level out the whole basin and bring more fill. We have very limited funds, so it is essentially unfeasible to use cement. Right now we are estimating $210/meter construction costs.

 

[mbeschenbach]

Additional file

 

[gbam]

Where is the flood stage and the normal river stage on your section? Have you considered that vegetative lining will help protect the banks if you hydroseed over the mattresses? If your design is for flood stage, I would double check your velocity, 21 fps seems high. Do you have a HECRAS model of the river?

 

[mbeschenbach]

gbam,
thanks for the help...
The river shown in the attached section is depicted at maximum expected 1000yr flood stage (4" rain/hr.), dashed line at Agatha level (1.7"/hr). The typical river level meanders in the river basin at 200cfs in the dry season to 1-4000 cfs between storms in the rainy season. At max expected flood, we are calculating 52,972 cfs. The main culprit is a slope of 200+'/mile. I have not seen HECRAS in operation before, but I reviewed the program online and it seems dooable. I will download the program and see what I can do with it... not sure it I am experienced enough to run it but I will see. the first posting I sent has the calcs for flow.

 

[gbam]

mbeschenbach - After reviewing your last section, I have another question. Was there a riverine hydraulic study conducted including a mobile bed analysis? Maybe that is where you obtained your data.

I really quickly calc'd the shear stress at the toe of slope at: normal (Agatha) stage which is +/-10lbs/ft2 and at max stage which is +/-18 lbs/ft2. With this information you may want to investigate other types of protection. I really don't think the reno's will do the trick. Have you called MacCaferri directly and chatted with them?

 

[mbeschenbach]

gbam,
there has not been any other study done, than by map analysis, transit for gradient, and visual observations and calculations for water flow.
I have contacted MacCaferri. They are looking into it now.

 

[gbam]

Let us know the outcome. I am always interested in these large scale project.