Stormwater Library

resources for stormwater professionals

CSO Tunnel Projects in United States

Across the country, large underground tunnels are commonly used to prevent combined sewer overflows. Some of the larger projects have been taken on in Chicago, Atlanta, Milwaukee, Austin, Boston, Cleveland, Houston, Detroit, Los Angeles, New York, Minneapolis, and San Francisco. A typical tunnel project ranges from 7 feet to 33 feet in diameter and from 2 miles to 33 miles in length. The Chicago Tunnel and Reservoir Plan (TARP) is considered the most successful application of this technology. It covers 109 miles and typical tunnel diameters reach 33 feet. Since 1984, when the first TARP tunnels were built, the number of fish species in the Calumet and Chicago River systems have nearly doubled from 33 to 63.

What needs to be put into perspective here is the cost. Combined sewer systems are an infrastructural relic – many dating back to the beginning part of the last century. The amount of new infrastructure to deal with this problem, while successful in reducing CSO discharges to our nation’s surface waters, is astronomical in size and cost. Take a smaller project like that in the City of Milwaukee. Over 2 billion dollars was spent on developing those CSO tunnels – more than half of that money was from federal grants. Another thing to consider is this, while the systems are typically able to successfully store and treat the overflows, the massive amounts of energy to make this happen begs quantification.

Is a massive concrete hole in the ground the answer? It is working, but is it the answer?

Compost-based BMPs – EPA Presentation

One of the better presentations that I have come across on the use of compost in stormwater BMPs was done by Chris Newman of the U.S. EPA. Mr. Newman provides some background into the EPA’s initiative to promote compost-based BMPs and points audience to its fact sheets database.

The presentation is a very nice primer on the benefits, effectiveness, and limitations of using compost blankets, compost filter socks, and compost filter berms. He also goes into some detail about the quality standards and certification required for the use of compost products.

Some of the key benefits highlighted in the presentation include:

  • ability to prevent rill erosion
  • runoff volume reduction
  • promoting establishment of vegetation
  • improving water quality through adsorption of nutrients and pollutants

A brief survey of some of the companies out there installing compost-based BMPs illustrated how creative designers can be with the products. Of particular note is the use of compost filter socks, traditionally used in the same manner as straw wattles, in the creation of permanent stream banks, or even as an alternative to rip-rap.

Stormwater Factoid: Combined Sewer Systems

There are approximately 770 combined sewer systems serving 40 million people in the United States. Most are concentrated in the Northeast, Great Lakes, and Pacific Northwest.


Properly Designing the Rainbarrel

Here are some things to weigh when considering rainbarrels as part of a stormwater BMP. While rain-harvesting and storage are nice potential side benefits of the barrel, its future success relies on keeping the focus squarely on stormwater capture. If you look around the internet and review municipal rainbarrel programs that are being promoted, they are often discussing the 50- to 60-gallon barrel like it is a viable option.

It simply isn’t. Consisder these conservative assumptions:

Let’s say a typical residential roof is 1200 square feet (it is closer to 1600). And, let’s say that a minor storm passes through – on the order of 1/4″. That would create a runoff volume of approximately 187 gallons. Let’s assume that only 60 percent of that runoff made it to the downspout due to ponding and poor pitching of the roof. That would result in a runoff to the downspouts of approximately 112 gallons. If the homeowner had 2 downspouts and 2 50-gallon barrels, the system is are already overflowing. With 4 downspouts and 4 barrels, the system is more than half full after a very small storm – just one storm. Furthermore, the overflow checks that are typically designed into a rainbarrel are not able to discharge at a rate that can keep up with the runoff rate into the barrel. So, the notion that the barrel is safe from overflowing during a storm is on shaky ground.

Here comes the next issue. The homeowner wants to be able to store this rainwater for future use, but the weather is a little rainy and the need to irrigate is not currently there. A couple of days later, the next small storm comes through and your barrels that should be empty are starting off at 56 percent of capacity. The new 1/4″ storm overflows the barrels. And, this is if 4 barrels are installed – not typically the case.

Where does that leave us? If rainbarrels are to be considered as a viable BMP, then larger ones need to be used. Successful programs like those used in the cities of Pittsburgh and Washington, D.C., use barrels with 3 times the typical volume capacity at 150-gallons. This could provide some storage capacity for future use in more arid regions of the country. In the areas prone to heavier rainfall, use of a couple of larger barrels offers the ability to capture runoff from more substantial events, but still not likely act as a harvesting tool.

The long and short of it is this. Focus should primarily be on stormwater capture, especially of that first flush. Once that water is captured, it will likely need to be emptied onsite through controlled overflow to a drywell or dedicated infiltration area on the property. Homeowners need to be educated on these issues and keep tabs on capacity and maintenance.

A final point should be considered when designing rainbarrels for storwater capture. Is it properly constructed? I run across a lot of make-your-own barrel websites that are not always considering issues such as proper overflowing or covering of the barrel. The last thing a homeowner wants to deal with is a massive buildup of water right next to the foundation, or a mosquito population explosion in the sideyard.

While the DIY approach has some appeal, by and large the focus should be on proper design and engineering. If we can get away from poor planning and design, rainbarrels can be a great addition to the lot-level approach of managing stormwater.

LID Implementation in Low Infiltration Areas

One of the questions that comes up quite often in the research about Low Impact Development is how to implement the design when you have naturally low infiltrating conditions (i.e. low permeable soils or shallow groundwater/bedrock).  In design scenarios like these you are inevitably going to be faced with major drainage issues, ponding, and vector control problems.  Using the State of California as an example, ponding present after 4 days is going to be the limiting factor on your BMP design.

Daniel Apt of RBF consultants presented a nice discussion on this very subject in 2008.  Mr. Apt summarized the measures to help reduce runoff from these sites as follows:

  • Reduce/Minimize Total Impervious Areas
  • Minimize Directly Connected Impervious Areas
  • Limit use of sidewalks
  • Reduce road/driveway length and width
  • Modify/Increase Drainage Flow Paths
  • Maximize overland sheet flow
  • Conserve natural areas
  • Minimize disturbance
  • Preserve infiltratable soils
  • Preserve natural depression areas
  • Preserve vegetation

The presentation goes into the benefits of using measures such as green roofs, rain barrels, cisterns, bioretention strips, and grassed swales as possible ways to working around infiltration issues on a site, while still drastically reducing offsite runoff.

Review the power point presentation in its entirety here.

Stormwater Factoid: City of Philadelphia

The City of Philadelphia owns and maintains over 90,000 trapped inlet catch basins.  Over 13,000 tons of debris are cleaned from the inlets on an annual basis.  The City spends over $46 million each year on operation & maintenance of stormwater BMPs.

Source: Leading Stormwater Management Programs in Pennsylvania


Low Impact Development Project Database

One of the trends that I have been spotting in research studies of LID acceptance into practice has been this concept of “proof of benefit”.  Essentially, the idea is that by having high-profile, high-traffic demonstration projects, stakeholders will see the benefit and will encourage LID’s spread through new development, as well as through retrofits in urban settings.  It’s a great concept, but one that takes a decade to truly take hold in a municipality.

The idea behind this website has been pretty simple.  Take a look at the tools that are available in stormwater BMP design and point out the more novel approaches and interesting research being done in sustainable stormwater engineering.  To help further the development of “proof of benefit” as a way of spreading Low Impact Development in communities, I am proposing the development of an LID project database.  It is possible this has been done to a certain extent.  I claim no mantle of originality to any ideas presented on this site and this is certainly no different.  Simply, what I would like to see is a much more comprehensive look at projects that have been completed, are currently in progress, or are in their infancy.  This will be a laborious process – one that will certainly require input and collaboration.  To this end, I make my formal request to others right now.

The idea is this.  Send me information on any project that is available in the public domain.  In particular, I need to be able to start with the basics such as: project location (street address), general description of the project, type of LID approach being used, status of the project, and if possible, a project contact.  I will most likely expand the database to include other fields as development progresses, but starting off simple will get me to a finished product more quickly.  Ideally, I would like to see this be a living databse that is fed by users as new projects spread through North America.

In addition to serving as a way of finding demonstration projects, I envision a database that is tracking the popularity of different LID approaches and spotting trends in technology – perhaps even helping to shine a light as to where LID is falling behind in acceptance.

Email your project lists and ideas to Matt Baumgardner at