Vince--The model that Murat has generated will not account for a blockage, unless a meter happened to be placed very close to the problem.  (I don't think one was.)

On Wednesday night, you used some pretty strong language when you questioned the validity of the model.  What I heard you saying was that the model wasn't a good tool for us to use in our decision making since it didn't correlate to what really happened in your neighborhood.  

Based on this information, it looks like the backups in your area were likely due to a maintenance problem, not to a systemic problem in the sewer infrastructure.  I don't think the model is intended to detect maintenance problems.  (Can Robert or OHM weigh in on this point?)

The repairs of three SS lines would explain the discrepancy between the model results and real-life observations in your neighborhood.  Has it changed your mind about whether or not the model is a valid tool?

My opinions: 

1.  The model should be reviewed by the TOAG, but so far it looks like a tool that we can use for decision making.   

2.  Our team should recommend a more frequent inspection regimen.  One neighborhood shouldn't have three problems at one time!

  

 

Peter, 

You are correct in stating that the model is not the correct tool for evaluating the risk of backups due to blockages or sewer collapses and such.  These maintenance issues or failures would not be detectable by the metering data collected, unless there was a meter upstream of the blockage in the zone of influence of the backup.  If the issue is not detected by the metering, we would not be able to identify it as an issue in the model. Once a blockage is identified or detected, the usual correction is to clear the blockage, and the system returns to operating as normal.  A more frequency inspection program, comprised of sewer video inspection and a manhole inspection program are common ways to address this. Many communities attempt to do this on rotating 5-10 year basis. For a system that has never had it done before, it may be desirable to complete the entire system more quickly, and then do a rotating program.  This would be a fine recommendation for the CAC to consider, and I will add it to the list for further discussion.

The model is intended to identify "systematic problems" that are caused by the capacity of the pipes or the amount of flow generated from the system. Once modeled and identified, these become "known system defects" that are detectable by the model and may require system upgrades to address. That is the nature of the six areas we have identified with the model.

Take care,

Robert

Having 3 sewer lines blocked at one time in a 5 block area is totally unacceptable. Having modeling that does not catch this is also an issue. It is presented as a means to indicate basement backup potential but did not provide this in the Glendale neighborhood due to blockages.

I agree with Peter the model should be looked at by the TOAG with the Glendale map.

The problem we are having is the lack of permanent gauging in the city to detect and prevent sewer backups. It is a much less expensive and better method to deal with backups. The cost of these gauges, which do not need to be accurate to a 1/10 of a inch, have come down and could send data directly or picked up by garbage trucks on weekly rounds, to city hall to be used as an early warning system for repair assignments.

Data could also be used to determine the effects of changes on the system such as new development or city efforts to reduce flows thru a special program, like new low flow toilet subsidies.

Vince,

I totally concur that having multiple lines blocked at the same time in such a small area is unacceptable and that better tools are needed to proactively detect and correct these issues.

Metering is one technique to detect some of these blockages. Some of the issues with metering for detecting blockages is that it can be very expensive to completely meter every pipe in the City, it is reactionary (i.e. it detects an problem after a failure has occurred), and processing and managing the data generated by a large metering network can be onerous.

One challenge with metering to detect blockages is that the meter must be located upstream of the blockage, within the zone of influence of the backup from the blockage.  "Within the zone of influence" is the critical element for a meter to be used for detection.  Standards for sewer pipe design dictate that to maintain adequate scouring or self-cleaning velocities, the smaller pipes upstream in neighborhood are sloped steeper than the larger downstream interceptors. The smaller the pipe, the higher the steepness that is required.  This means that the "zone of influence" is shorter in the smaller pipes in the upstream neighborhoods due to the higher steepness of the pipe, thus requiring many meters to provide full coverage for all of the upstream pipes. We are potentially talking about a very large number of meters to cover every pipe in the City (hundreds or maybe thousands of meters - for reference we had 30 meters in for 3-5 months for this SSWWE project at a cost of about $200,000). This goes beyond the typical application for metering, and is the reason that meters are more applicable to detect issues in the downstream interceptors, where the pipes are flatter and the zones of influence are much longer, thus requiring less metering, and making metering more practical for the interceptors.

Detecting blockages in the smaller upstream pipes is more commonly done through sewer video inspection and manhole inspection. There are processes for inspecting and tracking the condition of sewer pipes over time, so that a proactive replacement and repair process can be put in place.  Just like roads, sewer pipes generally don't fail all at once, although sometimes they do if you have ever tracked some of the sewer sinkholes that have occurred in the region. But more generally, just like roads, they deteriorate over time, and by using a regular inspection program on a rotating basis (every 3-10 years), the condition of a pipe can be tracked so that it can be repaired before a catastrophic failure and blockage occurs.  NASCO's Pipeline Assessment and Certification Program (PACP) provides a very common framework that is used to do this (reference: http://nassco.org/training_edu/pdfs/pacp-macp_overview.pdf ).

For a system like Ann Arbor that does not have a rotating program in place for video inspection, it might be desirable to do "one full lap" fairly quickly (perhaps in a year or two) in the high priority areas. This would establish a baseline condition, identify places where failure is imminent, fix those high priority areas, and then identify the appropriate time frame for a rotating program, based on the condition and risks associated with pipes in specific areas. Such a process is proactive, likely cost effective compared to metering every pipe in the City, and generates valuable video data that can be used to track pipe conditions over time.  This is currently the industry standard process for monitoring and maintaining sewers, and is the program we would generally recommend to address the risk of frequent sewer failures in the upstream pipes.

That being said, a smaller number of permanent meters on the interceptors and perhaps on several key neighborhood connections may also be a good idea.  Permanent metering of the downstream interceptors would provide information on blockages where the meter is in the "zone of influence" of the blockage, but more importantly, it would provide a long-term record of flow data from the system which can be useful for assessing impacts from growth, changes from system deterioration leading to new inflow and infiltration sources, the effectiveness of a DOM program (if continued), and important operational data to understand how the system handles future large storm events.

Hope this helps!  I think the discussion is great, and is starting to get the gears turning on what some of the options are for recommendations to address the issues.

Take care,

Robert