In response to the NY Times article related to the Tibbetts Brook daylighting to boost capacity for sewers and some discussion on Twitter, Adam Broadhead (@losturbanrivers) sent a great 2013 journal article in Water Research, “Captured streams and springs in combined sewers: A review of the evidence, consequences and opportunities” by Broadhead, Horn, Lerner, which addresses the issue with some research. The article is paywalled, but let me know if you’d like a copy and I can email it to you.
It’s more of a deep dive into some of the research, but the general thrust is that water intrusion in systems has reduced capacity, and that the intentional encasement of streams and springs in pipes reduces the capacity of infrastructure which has a significant economic, environmental and social implications for the infrastructure, as it reduced the baseflow reduces the overall effectiveness of gray infrastructure.
The typical mechanism for intrusion into pipes is related to cracks, which is assumed to be residual groundwater entering pipes in ‘dry weather’ times, where there should be no flow into the system. Groundwater intrusion should not be discounted, but there are other sources of intrusion that are typically not considered, specifically “capture of streams and springs” that impact combined systems capacity.
The figure below shows the change in baseflow and runoff response due to the intrusion of the additional water from streams and springs.
The paper continues to identify the issue, also highlighting the lack of research on this topic, and answers some fundamental questions about how this capture occurs, how to identify it, what is the magnitude and impacts, and ways to manage it. Always interested in language, one item of interest explores key terminology – culverting, extraneous water, groundwater infiltration, sewer inflows and the key element, stream and spring capture. The wordplay is compelling, with some uniquely evocative terms emerging such as parasite flow, misconnected surface waters, sewer leakage and illicit connections all telling a story of water that is in a sense, ‘out of place’.
The how and why is interesting. The most basic version is to take a free flowing stream and incorporate it into a pipe (Type A in graphic above). “Urban streams were frequently culverted and buried, especially during the period of rapid urban expansion in the 19th century.” It’s not a stretch to show that the literature confirms that “old sewers were frequently the covered channels of brooks”, as early development merely hid the streams, but didn’t generate as much additonal flow to overwhelm the piped streams. This happened with additional development and expansion of cities and impervious zones. Often the buried streams become the names for the sewers themselves, such as those specifically mentioned in the article like Garrison Creek Sewer in Toronto and Minetta Brook Sewer in New York.
The baseflow in the streams, unlike sewage, is clean, so the incorporation into pipes and transportation to wastewater treatment plants means additional strain on purification infrastructure with water that doesn’t need treatment. This relates to the original conceptual idea of the Tibbetts Brook example today, with a clear path to remove ‘clean’ water that is reducing combined capacity and overall resilience to deal with larger storms.
Additional capture happens by interception (Type B in graphic above). The most visible example is the massive interceptor sewers in London developed by Bazalgette in response to the ‘Big Stink’ in the the 1850s, acting as a divertor to sewage entering the Thames. This model was copied around the globe, with numerous examples of streams disconnected from their outfalls and no longer making it to their original destinations in the name of water quality. Portland has a large, expensive example of this called the Big Pipe. Many other cities have similar interceptor systems.
Another mode of is by directly capturing and draining spring and seeps in combined sewers, in this case through leaky pipes with cracks and joint openings. Beyond being shoddy construction, this was intentional, designed as deliberately leaky to provide drainage in areas of perched or high groundwater. The 3 types are summarized graphically above, showing variations of combined sewers and stream capture typologies.
The connection here to lost rivers is outlined in the article: “Not all streams and springs are fully captured by these modes of entry. London’s lost rivers diverted into the High, Mid and Low Level Interceptors to the WwTW, (wastewater treatment works) such as the Walbrook, Fleet, Tyburn and Westbourne, do still discharge to the River Thames during heavy storm events, where the original courses of the rivers serve as CSOs.” This is also a pattern in the United States (New York) and Asia (Tokyo) where many of the piped streams never make it to their original drainage water bodies.
The 19th Century was a historic time for burial of waterways, as the rate of urbanization outpaced the ability of natural streams to remove wastes. Thus: “Urban streams that had become open sewers were frequently culverted and buried to provide more sanitary conditions, and this concept is a popular narrative predominantly explaining the conversion of many smaller watercourses to combined sewers (type A).” Beyond the main drivers of pollution reduction and removal of the streams to create land for development, the introduction of seeps and springs provided some necessary baseflow to ‘flush’ sewers as a method of ‘self-cleansing’, and thus was in common practice in sewer design.
It is obviously difficult to identify these captured streams, as they exist under the surface and the original hydrology has been erased. This is where hidden hydrology methodology, using mapping and other primary sources to show where routes of surface flows used to run. Often these were parts of combined sewers, but in some cases the streams were just dumped into pipes. While still important, it is less impactful to combined systems and wastewater treatment facilities as they are often just draining into the same waterbodies that the original creek flowed in to.
Urban exploration is another method of finding routes of streams mentioned (which I’ve covered in depth here in many cities). Mapping of sewers and streams supplement this work, with many cities having robust sets of maps dating centuries in the past to fill in gaps of knowledge of what was there and what was replaced. More sophistical modeling can be helpful, but simple cues like place and street names and other subtle clues can also be extra data to be used to pinpoint old routes of waterways. As mentioned:
“Relevant information on lost urban watercourses helps to establish the pre-development hydrology, but the usefulness of historic maps depends strongly on spatial and temporal coverage, with many older towns and cities having altered the hydrological landscape before the first available maps. The smallest streams and springs may also not be marked on maps at certain scales, particularly intermittent and ephemeral channels.”
The ability to quantify these captured streams is equally challenging – there is adequate knowledge of the phenomenon but lacking in specific data on volumes, routes and baseflow contributions to the systems. While even knowing the levels would be helpful, measuring current flows will yield radically different results today versus pre-development conditions. When quantities can be estimated, the economic benefits can be modeled to see impacts, but this is not common. How the water is distributed is also variable and depends on unique qualities of each stream.
The major consequences are two-fold. First, the introduction of clean stream water increases the amount of water handled by treatment plants, which has larger infrastructure costs in terms of facility construction and operations. Second, loss of surface streams has impacts to habitat, less ecological connectivity, and overall less ecosystem services, including amenity value. It can even have secondary impacts on urban heat by reduction of linear corridors of riparian vegetation. While the anecdotal benefits of ‘flushing’ using the streams was developed early-on, it’s not understood if there’s actual value of these approaches.
A summary of the impacts on the industry are included:
- More land and costs needed for wastewater treatment infrastructure
- Additional operational costs and use of chemicals
- External impacts, such as public health impacts of CSOs, impacts due to loss of ecosystem services due to diversion of local streams, and economic losses.
There’s a more detailed cast study from Zurich, Switzerland that’s worth exploring more. In summary, the authors mention the city as a pioneer through “innovative management of capture streams and springs in combined sewers,” typically through separation using daylighting. This was driven by understanding the “lost social ad environmental values of watercourses that had become culverted and had historically been used as wastewater sewers.”
The benefits to the public include amenity spaces, and also more efficient infrastructure through additional capacity. This dual benefit is key to the Stream Concept, and became codified into planning policy and laws. The dramatic reducing in streams due to urbanization is similar to other cities, with development displacing larger areas of open space and burial of streams, many of which were converted into combined sewers between 1850 and 1980 as seen in the figure below.
The transformation of streams from this point in 1980 shows the changes in approach used by Zurich in the Stream Concept. This is outline in the existing condition (1) which includes stream capture in a traditional combined sewer system, a severing of the hydrological system and piping; the first transformation (2) consisting of separation of the combined systems to removed capture streams, and eventually the final phase of the Stream Concept (3) “separating captured streams and springs into daylighted urban watercourses.”
An important aspect which reflects my approach allows for hybrids of ‘daylighting’ without and zero-sum outcome of daylight or nothing, but allow for a continuum of potential options – as I’ve discussed, between art and science (abstraction vs. pure restoration) or more specifically, interventions that can be located in a triad of artistic, design, or engineering. The street streams, per the articles:
“Naturalistic stream channels and riparian corridors are used where possible, but where space is limited, engineered “street streams” are installed. The latter may have a lower ecological potential, but nevertheless offer architectural value in urban areas.”
The article concludes that there is value in disconnecting streams and springs from combined systems (or if we could spin time back, not connecting them in the first place), with economic, environmental and social benefits. The diversion of clean, constantly flowing water out of combined systems provides capacity, and by daylighting (vs. piping) these streams, we have the additional ecosystem benefits. The need for more research is mentioned: “By using daylighted urban streams to convey the clean water baseflow, highly promising social and environmental benefits
have been suggested; an independent peer-reviewed appraisal of this approach would be strongly recommended.” Since this is a 2013 article, I’m curious what additional scholarship has emerged in the last decade.
I also am intrigued by two of the US examples identified in the article were in Portland and Seattle, both of which mention combined sewers with springs running in them. Worthy of more exploration, but both of these do related to a location where a stream was buried and integrated into the pipe infrastructure of the city, which was common in many streams in both cities (for instance Ravenna Creek in Seattle, or Tanner Creek in Portland). Perhaps with the continual increasing impacts of climate change on these systems would drive another look at some daylighting to increase the resilience of the pipes to handle more capacity, while also providing habitat, amenity, recreation, and a range of other essential urban ecosystem services?
Full Citation: A.T. Broadhead, R. Horn, D.N. Lerner, Captured streams and springs in combined sewers: A review of the evidence, consequences and opportunities, Water Research, Volume 47, Issue 13, 2013, Pages 4752-4766, ISSN 0043-1354, https://doi.org/10.1016/j.watres.2013.05.020
Header Image: Figure from the article: Historic loss of Zurich’s streams (water in blue) with increasing urbanisation (grey).