Houston Drainage and Flooding Policy Options

Back in May of 2016, five months after he was sworn in, Mayor Turner announced that former City Council Member and mayoral candidate, Stephen Costello, P.E., would serve as the City of Houston’s Chief Resiliency Officer (CRO).  Mr. Costello’s main objective as the city’s CRO is to make some progress in addressing flooding risks in Houston.

Steve Costello and Mayor Turner. Photo: J. Evans. Reposted from Houstonia.

Since that time Mr. Costello has given out his cell phone number, met with numerous community groups and interested stakeholders. Sources close to his office have indicated that Mayor Turner has approved the appointment of a Flooding and Drainage Task Force who will consider the issues and make recommendations for improving things.  This is consistent with Mayor Turner’s original intent, announced on April 20, 2016.  Here are some ideas the Task Force should consider.

1. Gradually Increase Funding for Drainage System Improvements

Continue retiring our existing infrastructure debt using Rebuild Houston funding. Continue to move to a pay-as-you-go approach. Continue to plan, design, and construct replacements and upgrades to storm sewers to improve service levels. Continue to fix the drainage systems with the lowest level of service first – the so called “worst first.” Gradually increase funding for streets and drainage so that areas with lower levels of service can be addressed.

2. Create Tiered Detention Requirement for Redevelopment

Development in the City of Houston is currently required to provide a volume of detention that is a function of the amount of new impermeable surface area constructed. Under this scheme a 2.0 acre site with no added impermeable surface area must provide 27,150 gallons of detention storage. A 2.0 acre site with 100% added impermeable surface area must provide 325,830 gallons. This reflects the current policy of allowing redevelopment that does not change the runoff volume to not detain very much.

In areas of the city with older public storm sewers that do not meet current design standards, this policy should be adjusted. We should require property owners who are completely reconstructing their facilities to reduce the volume of runoff from their properties to help reduce the burden on these older public systems. This will benefit the private property owner or tenants by reducing the risk of flooding in the vicinity of the property.

This could be implemented using a tiered system based upon the degree to which the site is being reconstructed. For example, if a commercial strip center was repaving their parking lot and updating the center facade and signage, they might have a lower detention requirement than another project that was completely demolishing an existing building and surface parking lot and building a new mid-rise office building with underground parking.

The tiered system might be based on the capital investment per acre or a similar measure of the “degree to which the site is being reconstructed.” The detention requirement would be directly proportional to the capital investment per acre.

3. Define Pervious and Impervious Surfaces More Rigorously

Detention requirements are a function of impervious area, which is intended to account for the volume of stormwater that infiltrates into the ground rather than runoff the site. This seems rational, but we don’t consider how much infiltration any particular surface actually provides. Under the current regulatory scheme ground surfaces with very different infiltration characteristics are lumped into two categories: pervious or impervious. The reality is more nuanced, with different ground surfaces allowing stormwater to infiltrate at various and quantifiable rates. Establishing detention rates using a more nuanced consideration of infiltration rates would be helpful.

4. Create Stormwater Volume Trading Program

Many communities with “one-pipe” sewer systems installed in the late 1800’s or early 1900’s are working to capture, reuse, retain, and detain stormwater volume to reduce the severity and number of combined sewer system overflows (CSOs) under consent decrees signed with the United States Environmental Protection Agency. These communities include Chicago, St. Louis, Portland, Seattle, Philadelphia, New York, Baltimore, Washington DC, and many others (see red dots in the map below).

Cities with populations greater than 50,000 with combined sewer systems. Map from U.S. EPA.

CSOs happen when it rains a lot and the combined system can’t handle the volume of runoff and sanitary wastewater flows. CSOs allow untreated sanitary sewage to be discharged into creeks and rivers.

Every cubic foot of stormwater that can be controlled above ground in a detention basin, a cistern, a green roof, a rain barrel, a bioswale, or some other feature is one less cubic foot that needs to be managed in large, centralized, grey infrastructure facilities – think 40 ft. diameter tunnels. Any storage volume, whether provided by a public entity or a private land owner or developer, helps mitigate the overflow issue.

CSO communities are establishing stormwater volume trading schemes and using the market to create incentives to build “surplus” detention or retention volumes and then allowing the market to sell and purchase these stormwater volume credits.

We should set up this type of stormwater volume trading in Houston to reduce flood risk and flood damages.

5. Use Real-Time Controls

Every hour the Weather Prediction Center of the National Weather Service produces an estimate of the probability of precipitation and the amount of precipitation expected in upcoming 6 and 24-hour time intervals from 1 day to 7 days ahead of time.  These data can now be consumed by stormwater infrastructure control boxes connected to the internet. These real-time weather predictions (which are updated each hour) can now be used to control valves, pumps, gates, and other water system devices to enhance the level of service provided by them. Pumps can be switched on or off, valves can be opened or closed, immediately, based on real time data and decision logic programming.

For example, a cistern could be used for rainwater harvesting to reduce potable water consumption during a period of anticipated moderate to low rainfall, but the same system could be used for detention during a period of anticipated high rainfall.  This is possible by operating our drainage systems using real-time control systems.

We should start deploying these types of systems to help get more from our stormwater infrastructure.

6. Establish Hydrological Basis for Using Green Stormwater Infrastructure

The City’s drainage design criteria imposes a detention requirement that is solely a function of the change in imperviousness from pre-development to post-development. This does not allow the determination of pre-development and post-development hydrology and the direct calculation of the detention volume using a hydrological basis. See this earlier post for an illustration of what this means.

It would be helpful to allow detention volumes to be determined using hydrological principals and calculations and to be derived from the difference between the pre-development and post-development runoff volumes.  This would create more of an incentive to use green stormwater infrastructure (what I call “natural drainage” systems).

7. Try Ground Modifications to Enhance Infiltration

We don’t rely enough on infiltration. We tend to over estimate the volume of infiltration we think we are getting through prairie land (that has been historically farmed). We tend to under estimate the volume of infiltration we think we are getting through our clay soils. We almost always stabilize and compact the soils all over our development sites, which, of course reduces infiltration. Why not create more spongy areas by deep ripping some soils? Why not amend our soils in targeted locations with sand or organic material to enhance infiltration rates? Why not examine soil infiltration rates in our normal pre-design geotechnical investigations?  We could rely more on infiltration to manage stormwater volumes if we actually considered it and measured it.

8. Enhance public education on flooding risks.

We should communicate the risk of flooding more consistently, more effectively, and more loudly. This will help align public expectations with reality. I get the feeling that many people in Houston expect the risk of flooding of anything they own (car, bike, home, etc.) to be nearly zero. I get the feeling that many people in Houston are very disappointed when they find out the risk of car flooding is much, much higher than that. We design new streets to convey stormwater away from homes and businesses. That is worth repeating: our streets are designed to carry stormwater! The storm sewers below our brand new streets are sized so that they can carry a depth of rain that has a 50% chance each year of being exceeded. This means that any particular new street in Houston has a 50% of flooding every year. That occurs when we get about 3 to 4 inches of rain in 24 hours. Better risk communication would not only help our citizens with expectations and management of the impacts of high rain fall, but might help inform a discussion around increased funding for flood risk reduction.

“Harvesting the Value of Water” – New ULI Publication

I’ve been working to get more involved with the Urban Land Institute lately. During the 2016 Fall Meeting in Dallas I was able to connect with Katharine Burgess and Rachel MacCleery, both of ULI’s Center for Sustainability and Economic Performance. They were kind enough to allow me to help review and contribute to a report they were working on about the use of natural drainage systems in real estate that would eventually be called “Harvesting the Value of Water.”

The report, which was formally released in May of 2017, was made possible by a grant from the Kresge Foundation and was prepared with review and input from many ULI members and experts.  I was glad to have a small role in its preparation.

The report first explains why stormwater management issues have recently become more visible. In the late 1800’s and early 1900’s cities in North America were just beginning to construct urban sewer systems. There were two competing approaches: a combined sewer system, which handled both sanitary sewage and stormwater runoff, or a two-pipe system that handled each in its own pipe. Several hundred cities constructed combined sewer systems to serve at least some portion of their area.  These system can sometimes overflow during larger rains.  These overflows are known as Combined Sewer Overflows or “CSOs.”  CSOs discharge untreated sanitary sewage into area waterways, which is . . .um.. . not so good.

The Environmental Protection Agency (EPA) started working with cities with CSO issues back in 1994 and now many of them have entered into legal agreements with EPA and the U.S. Department of Justice to address them.

Initial plans to address CSOs called for large diameter tunnels to be constructed to store the mixture of water during rain events to reduce the number and severity of overflows. These projects had price tags of $4, $6, or $8 billion per city. Several cities figured out that using green stormwater infrastructure (GSI) to manage stormwater above ground at each building or site substantially reduced overall CSO mitigation costs. Some cities added the use of GSI to manage private property and public stormwater runoff volumes into their agreements with EPA. These programs require volume control to a larger degree than that typically required for merely floodplain management. So in some parts of the country, outside of Houston, the use of GSI is required.

In other parts of the country GSI is required by states and local governments to reduce stormwater pollution to estuaries or other waterbodies that don’t currently meet surface water quality standards. The Chesapeake Bay restoration effort is one of the largest examples.

The report also provides a good introduction to GSI techniques, such as green roofs, bioswales, rainwater harvesting, and rain gardens.  Chapter 4 explains how smart GSI implementation can provide private real estate developments higher operating income, faster lease rates, higher occupancy levels, greater lot yields, green marketing benefits, and reduced drainage infrastructure costs.

Chapter 5 provides eleven case studies from around the country that illustrate how GSI has been deployed in various types of real estate development projects. The publication includes a profile of Stonebrook Estates, a 50-acre Houston-area single-family residential development case study.  The development features a GSI system designed by Steve Albert, P.E., who was with Aguirre and Fields at the time.  The balance of the site civil infrastructure – the potable water distribution, the sanitary sewer system, roadway paving, and general grading was designed by R. G. Miller Engineers, Inc. before I started working there.

bioswale with overflow structures for handling larger storms at Stonebrook estates. Photo credit: m. bloom

The last chapter summarizes the stormwater policy landscape and explains the range of methods used by the local governments to either allow, encourage, or mandate the use of GSI in land development. These methods can range from merely enacting a permissive regulatory framework that allows private project sponsors to use GSI when it makes good business sense (a bottom-up, market-based approach); to offering some permitting or financial incentives; or to enacting across the board mandates for stormwater volume controls and retention to achieve CSO mitigation or surface water restoration objectives (a top-down, regulatory approach).

I encourage folks to download the publication and check it out.

The Invisible Development

Can we make our land development projects (hydrologically) invisible to downstream properties?  Think of Claude Rains, in the 1933 film adaptation of the H. G. Wells novel, The Invisible Man.

“If I work in the rain, the water can be
seen on my head and shoulders.
In a fog, you can see me – like a bubble.
In smoky cities, the soot settles on
me until you can see a dark outline.”

— The Invisible Man, 1933

I believe we can, using a natural drainage approach.  To illustrate this we must think through some basic — no actual math required! — hydrology.

Imagine a rectangular area of undeveloped land that has a gradual slope from one corner to another.  Imagine that all rain falling on this land drains to the low corner and rain falling outside of this area drains to some other location.  Like this:

Imagine that before we develop the site – the “predevelopment” condition – we install a flow measuring device to the low point. This allows us to record the stormwater runoff flow rate leaving the predevelopment site.

If we did this, one hour before a 10 minute rain event, for example, the runoff flow rate would be 0 gallons per minute (gpm). As the first drops of the 10 minute rain hit the ground, the flow would be 0 gpm. Minutes and hours later, the flow would reach its peak and then start to decline back down to 0, like this:

The graph above displays the predevelopment hydrograph.  All plots of stormwater runoff flow rate vs. time are known as hydrographs. If we know the history of the flow rate vs. time, we can easily determine the total volume of runoff that left the site as a result of our 10-minute rain.  The total runoff volume, if you recall your calculus, is the area under the curve, like this:

This makes sense because if we multiply the dimensions of the x-axis expressed in minutes by the flow rate expressed in gallons per minute we get gallons because the minutes cancel out:

Minutes x Gallons / Minute = Gallons

If we add some kind of development (buildings, roofs, roads, etc.) to the site, thereby increasing the site impervious, the runoff hydrograph changes.  The added smooth hard surfaces and concrete storm sewers:

  • Reduce resistance to flow;
  • Eliminate nooks and crannies for surface storage;
  • Accelerate the timing of the runoff;
  • Reduce or eliminate water infiltration; and,
  • Reduce or eliminate water transpiration (consumption and release to the atmosphere by plants).

These changes to the drainage area change the hydrograph that would be produced if the exact same 10 minute rain event fell on the post-development site.  The post-development hydrograph might look something like this:

Note the following characteristics:

  • Higher peak flow;
  • Earlier peak flow;
  • Faster decline back to zero flow; and,
  • Larger total volume of runoff.

Well that can’t be good, right?

If we developed this way the bayou receiving this runoff water would see a higher flow rate. This would result in a higher water level, which might cause downstream flooding if that higher water level was higher than the top of the bayou banks.

We mitigate the effect by sizing and constructing detention basins downstream of all new development. The detention volume is generally equal to the “excess volume” produced by the development. The “excess volume” is determined by calculating the difference between the pre- and post-development runoff volumes, like this:

The light blue is the difference between the two ares (volumes).  We can place that volume of stormwater runoff anywhere we’d like on the site. Engineers love making them into the shape of nice, regular, rectangles, like this:

Landscape architects have encouraged us to make them into more natural shapes. Regardless of their shape, they are designed to hold the excess volume and to release that water at a rate that does not exceed the predevelopment peak flow rate, like this:

This prevents downstream flooding, but its not perfect. Can you see a few of the problems with this approach?

The main problem is the volume of runoff is not reduced. With detention, the site discharges at the predevelopment peak flow rate for a longer period of time.  (Compare the horizontal distance of the blue line to the brown line.)  Compare the brown area (predevelopment runoff volume) to the blue area (post-development runoff volume), below:

So how can we deal with this extra volume?

Natural drainage systems (also known as “low impact development”) can help address this. Natural drainage systems are installed to slow the water down, infiltrate water, evaporate water, store water in small or micro scale detention areas, transpirate water, and generally to mimic the predevelopment hydrology. The same rain event falling on the site – mitigated with a natural drainage approach – might produce a hydrograph that looks more like the green hydrograph below: 

So how does the runoff volume comparison look using natural drainage? 

Pretty good, huh?

The natural drainage approach seeks match the predevelopment hydrology.  This means that the downstream folks experience no difference in the timing, rate, or volume of runoff (for a given rainfall event).

Some natural drainage proponents, like me, like to say the development is hydrologically invisible.

Nature Play and Natural Drainage

The Houston Chronicle recently published a great article by Dylan Baddour about recent work by TBG Partners at the Wetland Park in Riverstone, a master planned community in Sugar Land, between the Brazos River and State Highway 6.  It has been planned, designed, financed, and built by the Johnson Development Corporation.

The article was generally about how master planned communities are moving away from formal parks, golf courses, and playgrounds and starting to provide more natural environments for recreation. This could include things like real wood and stones for kids to climb on, mud puddles, native grasses and other plants, and direct access to water to see and touch tadpoles. The National Wildlife Federation has suggested that facilities like this provide opportunities for people to conduct “nature play.”  The Natural Learning Initiative at North Carolina State University has published guidance on how to plan and design nature play facilities.

The story highlighted how TBG Partners landscape architect Meade Mitchell designed the Wetland Park, located just east of LJ Parkway and a bit north of University Boulevard.  The Wetland Park is among the newest amenities and it provides a sharp contrast to a more traditional, hardscape amenity area called Riverstone Club, located a bit to the southwest.  The Club includes tennis courts, a dog park, pools, and gym facilities. See map below.

Aerial view of riverstone with insets and labels prepared by m. bloom
from GOOGLE EARTH image and riverstone landplan dated february 2017.

The Wetland Park includes a mud pie kitchen, rock throwing targets in ponds, a bridge with no handrails to make it easier to watch aquatic life, logs and natural wood placed in ponds for aquatic habitat, and a pathway of partly submerged rocks to encourage wet feet.

When I was a kid I played in Shortridge Memorial Park in the headwaters of the East Branch of Indian Creek just outside of Philadelphia. Building dams, catching crayfish. So I would love to live in a place with these types of amenities.

East Branch of Indian Creek, shortridge memorial park,
Wynnewood, PA. Image from Google Earth.

A close look at the 2017 aerial photography of the development from Google Earth illustrates that a traditional underground storm sewer system serves the residential lots and streets.  It also shows how engineers provided large excavated areas to hold and slowly release stormwater runoff from the development.  These are called detention basins (or ponds) and they hold some water permanently to provide lake features for residents to enjoy.  When it rains, water runs off the roofs, the yards, the driveways, and roads, and drops into concrete storm sewer pipes. The water is rushed quickly to the detention ponds, which fill up to their full capacity to store the excess water. Small discharge pipes then release the water at a rate that is less than or equal to the pre-development runoff rate.  This prevents the development from flooding downstream properties.

This traditional approach of speeding up the runoff along curbs and gutters and inside storm sewers only to slow it down inside large detention basins is not very efficient.

What if we conveyed rain water away from the homes and structures using natural topography?  If the land is too flat, what if we graded and planted the landscape to create drainage corridors that look like creeks or bayous?  What if we preserved and extended the existing wetland areas and creeks into the community so that families could interact with natural creek corridors throughout it? The natural drainage concept does this.

Natural drainage allows rain water to flow and collect in natural creek corridors more slowly. This reduces the volume of detention required to protect downstream properties. Natural drainage corridors bring nature next door.  They provide a drainage infrastructure service as well as a valuable water-based linear amenity.

The natural drainage approach can help avoid regulated wetlands and can help integrate them into the overall land plan in a way that can preserve their function and connection to other waters.  Natural drainage corridors also provide a place for hike and bike trails and enhance community connectedness.  The natural drainage approach fits perfectly with the nature play approach and visa versa.

The natural drainage approach helps enhance the economic performance of real estate projects by providing:

  • Higher operating income;
  • Faster sales;
  • Higher amenity values for more of the community;
  • Greater lot yield (through reduced detention requirements);
  • Green marketing benefits; and,
  • Reduced drainage infrastructure costs.

The Wetland Park in Riverstone is a great step in the right direction and I applaud all who were involved in its creation, but we can do even better using integrated natural drainage and nature play concepts.  We can integrate the natural world with the built environment by combining wetlands preservation, drainage, linear trails, connectivity, and recreational natural play environments all into the same system, all while enhancing the economic performance of the community.