Sediment pond under construction at risk during storm

Pond under construction nearly overflowed by 7:00 am 7/23.

Two sediment ponds are under construction along the SW side of Old Middleton Rd, a bit west of Capitol Drive.  The purpose is to treat stormwater coming mostly from Old Middleton Rd.  There are also plans to widen the road here, with the addition of sidewalks and bike lanes.

Neighbors in the area have been opposed to the widespread disturbance to the wild area from construction.  About 200 trees were cut to make way for the ponds, and another 200 for widening the road.  News story.

The pond nearly overflowed

After the recent rain storms of 7/22 and 7/23, the southern pond under construction could have failed.  By failure, I mean that the pond filled by the morning of 7/23 to within about a foot (in elevation) of overflowing. 

"The National Weather Service indicated that nearly four (4) inches* of rain fell yesterday and they are predicting up to three (3) more inches by 7:00am 7/24. Dane County is currently under a flash flood watch until 10:00am Saturday."  Source.

The pond's spillway hasn't been constructed yet.  If the pond were to overflow, rushing water could cut into the soft bank of loose soil, unprotected by any vegetation.   It's possible erosion would continue till the pond emptied.  Large amounts of sediment would spill into the creek next to the pond.  And from there, it's only half a mile to Lake Mendota.  There would also be a huge gash in the side of the pond to repair.

An outlet pipe also has not been installed.   The pond is close to the water table, and has walls made of clay (plus some clay in the bottom).  So it's going to drain very slowly, and will probably have to be pumped out, in order to finish construction.

A basic precaution overlooked

Runoff from Old Middleton Rd. entered pond just in front of the shovel, running in a pipe under the gravel pad.

One of the basic rules of erosion control--often neglected--is to deflect any runoff coming from outside, so it doesn't get into the construction site.  The rain itself is enough of a problem, without having more water flowing in from the outside.

In this case, Old Middleton Rd has a long slope to the SE of the pond, feeding much stormwater into the unfinished pond.  The pond was designed to handle this runoff, but not yet!  This incoming runoff could have been easily deflected from the unfinished pond.**

This construction site has already experienced some erosion problems.  Prior to 7/22, the silt fence protecting the creek from sediment had leaked.  Then during the rains on 7/22, the silt fence for the south pond collapsed, dumping sediment into he creek.  By  7/24, that had been repaired, plus the addition of a gravel dam.  But the runoff from Old Middleton going into the overfilled pond still had not been deflected, even though there was a flash flood warning in effect for three more inches of rain!

By neglecting this basic precaution of deflecting runoff--even after the pond filled and with more flash floods forecast--the work already done on the pond was jeopardized, and Lake Mendota was put in danger of a large sediment spill.

Deflecting incoming water was not mentioned in the erosion control plan, which illustrates just how inadequate these plans typically are.

Destroying nature to save nature

Sediment ponds serve an important function--protecting the lakes.  But large sediment ponds, with poor erosion control during construction, may not be the best way to protect the lake.

Sediment ponds are supposed to "save nature" by keeping sediment out of the lakes.  But here, with cutting 200 trees plus a close call with a bad sediment spill, we have another example of  "destroying nature to save nature." 

The pond on the left is the one that nearly failed. The notch in the wall, above the sediment fence, is where it would have overtopped.  Sediment would then have been flushed down the creek (center) to L. Mendota.

This is the long slope of Old Middleton Rd, down which the runoff came.  The gray pipe in the lower right corner goes under the gravel pad by the shovel, and into the pond.

This pipe should have been plugged.

Here's where the pipe goes under the gavel pad, taking water to the pond.  Some water flowed past the caterpillar (center).  Water could have been directed that way around the pond.

Could the pond really have failed, emptied, and sent a major sediment spill to L. Mendota?  It probably would have taken 1-2 more inches of rain to overflow--but three more were forecast.  The newly disturbed soil was wet and soft--and the slope from overflow point to creek was relatively steep.  This photos shows two of three gullies where runoff came into the pond.  When overflowing, water would have created one gully much deeper.  We were lucky--but what this incident shows, is that plans were poor, no one noticed the danger, and no one took corrective action.
Photo 7/23, 7:00 am, by Tim Heath.

Who is responsible for this lapse

• Design Engineer: Lisa Coleman, 266-4093, City Engineering, lcoleman@cityofmadison.com . No mention of the need to deflect runoff was made in the erosion control plan.
• Construction Engineer: John Fahrney, City Engineering, jfahrney@cityofmadison.com
• Contractor: Rawson Contractors (Low bid was $1, 399, 029.85), Kenneth Servi, President, kservi@rawsoncontractors.com   http://www.rawsoncontractors.com/
• Permit authority: Tim Troester, (608) 267-1995, ttroester@cityofmadison.com
• Erosion Control Permittee: Signed by Robert Phillips, delegated to John Fahrney, City Engineering,(608) 266-9091, jfahrney@cityofmadison.com
• Inspector on 7/23 who missed last chance to deflect inflow: Stewart Mael, (608) 261-9695, smael@cityofmadison.com
• Grantor of permit for construction of ponds: DNR Water Management Specialist Cami Peterson, 608-275-3208.  No mention was made in the permit of the need to deflect runoff from the ponds under construction.

Slide shows
Failure of the silt fences due to stakes (about 1" x 1") being too weak.

*  The report of 4 inches was for parts of Dane Co.  For Madison, the rainfall was very spotty.  Wunderground indicates that rainfall at zipcode 53705 for 7/22 and 7/23 was only 1.63 in.  The pond is supposed to fill with about 2-3 in of rain, which is what it seems to have done.

**  The runoff into the pond could have been easily deflected to the other side of the street, and downstream of the pond, by building a small berm of asphalt on the street; or around the SE side of the pond to the creek, using a silt sock to deflect the water (or a minimal amount of grading).