 |
   |
||||
|
|
|
 |
||
|
 |
 |
 |
||
|
in Publications: Font size:    
|
|
Tropical Storm Allison 2001: The Worst Urban Flood in US History
Between June 5 and 9, 2001, tropical storm Allison dumped record rainfall on Harris County (population: 3,600,000), home to the City of Houston (the “ Bayou City,” population 2,000,000), causing the worst urban flood in US history. The deluge tested the coping mechanisms of even flood-hardened, disaster-savvy Houstonians.* The heaviest rainfall occurred on June 8 (Friday evening) and 9 (Saturday). Greater than 70,000 buildings in the Houston area reported flood damage, including neighborhoods that had never before flooded. Basements of hospitals in Houston’s massive Texas Medical Center (TMC)** flooded, disabling emergency power generators, which necessitated evacuation of hundreds of patients, many of them from intensive care units.*** Approximately four thousand laboratory animals at the Baylor College of Medicine perished. The price tag to repair the flood-damages in Harris County was $5 billion, much of which went to repair the damages to components of the Texas Medical Center.
Why does Harris County flood so frequently? Can flooding and watershed management minimize its flooding risk? This essay examines answers provided by Dr. Philip Bedient, Rice University’s (Houston) Professor of Environmental Engineering and specialist in Surface and Ground Water Hydrology, and**** in an academic seminar and Webcast on “Texas Water” organized by the Shell Center for Sustainability at Rice University (March 24, 2004).***** Dr. Philip Bedient: “I’ve been studying flood analysis and management for 29 years and have learned a lot of things during that time, which I want to share with you tonight as a set up, if you will, for the other members of the panel who will come after me. 1. Hydrologic cycle issues: Rainfall, infiltration, and stream flow are hydrologic processes that relate to flooding. [Hydrology is the scientific study of the properties, distribution, and effects of water on the earth's surface, in the soil and underlying rocks, and in the atmosphere.] Rainfall [precipitation], of course, kicks the whole cycle off [ Houston averages greater than 48 inches annually]. Houston, during the decade of the 1990s appears to have had five 100 year storms. I still can’t figure out the probability but I continue to work on the problem. Infiltration involves water passing through the interstices or pores of the soil. Every article I read in the newspaper makes me chuckle because it always says, “Oh, if we could just infiltrate some more water into the soil, we would lessen the flooding problem in the Houston area.” Most of the soils I have seen in the Houston area are either hard clay or hard concrete. So it turns out that infiltration is a minimal issue when it comes to [reducing] runoff in Houston. In Houston, rain hits the ground and a huge volume tends to flow overland as runoff. This is the portion of rainfall that gets into pipe flow and eventually makes its way into streets, down streets, and then into receiving streams, which here we refer to as bayous. The hope is that the bayou is large enough to carry the stream flow out into Galveston Bay. But we have found that high levels in the bayous can affect pipe flows. 2. Watershed development issues: In addition to hydrologic issues, there are four watershed development issues: location of storage areas, structure elevation, upstream development, and redevelopment. [A watershed is the region drained by a river system, and is also known as a drainage basin.] a. Location of local and regional retention or storage areas is critical. Storage areas for rainfall can have a critical affect on flood control. We are going to see in the foreseeable future that the addition of large storage areas back into some of our watersheds, especially in the upper end, such as in the Brays Bayou project, will end up having big effects on areas downstream like the Texas Medical Center. b. Elevation of structures is critical. My favorite joke when I came to Rice University was “How flat is Houston?” Houston is so flat, if you stand on a beer can you can see all the way to Dallas.” That sums it up. Houston is on a really flat area—we are at an elevation of 50 feet and we’re 50 miles inland, so it’s about a foot per mile of elevation. It’s pretty flat. So the elevation of the houses, the buildings, the Texas Medical Center is important issues. c. Upstream development that takes place without detention can lead to increased frequency of high flows downstream in a watershed. I think there’s ample evidence to demonstrate this. d. Redevelopment (increased density of buildings) that is ongoing within the older areas, such as between Rice and downtown Houston can cause subtle but very important impacts on flows. New bigger houses with large roofs on smaller lots cause huge drainage problems. During heavy rainfall it will be interest to see how the properties handle the runoff. e. Impedance by older (built in the 1960s), lower bridges can impact flows. These bridges were built without a sense that flows were going to get higher over time. These bridges start to act as impediments to flow and act as dams. Part of the Brays Bayou Project, a number of bridges will be elevated out of harm’s way, out of the impact with water. The current federal project will enlarge the lower part of Brays Bayou channel, add significant amounts of upstream storage and raise many of the bridges. It’s a big expense—a $450 million project that will take 10 years to complete. The other issue is the presence of concrete channels that are typically designed to increase flow capacity and move waters downstream. Another huge issue in Houston is the presence of “blocked flow paths.” For example, a railroad on a high bank will restrict movement of water during flooding, causing it to move in another direction. This happens all over Houston. The water that cannot get out of the area simply backs into living rooms and bedrooms. These are situations that need to be addressed at the local level. 3. Discussion Now back to hydrologic cycle. Rain is the main input. Rain can come in the form of a “design storm,” which is the standard 100-year storm. In Houston, we define a 100-year storm as about 12 ½ inches in a 24-hour period. But I like to think about the design storm as 8 inches or more falling in a 6-hour period, which usually causes lots of problems in certain areas. But we also most recently have begun to look at historical events. The reasons we have been able to do this and do a better job of this is because of “radar rainfall.” For example, we have a complete record of rainfall of tropical storm Frances, which came before Allison. We have a complete record on all the storms coming through, which allows us to do a better job of estimating the input to the system—much better, I think, than a series of rain gauges. It allows us to plan better for how actual storms impact watersheds. This allows us to get a better sense about whether our designs are in line with where they should be.
a. Hydrographs Hydrologic models convert rainfall to water flows, which is what we do behind the scenes. Rainfall comes into the system and goes through the watershed and produces a hydrograph—a response of water flow as a function of time. If you are not familiar with what a hydrograph is, drive down to the bayou after a 1- or 2-inch rain, park your car on the bridge, step out, and look: you will see the hydrograph come through. The water level comes up and then it goes back down. Our business is to try to figure out a way to control this. We would like to be able to predict the hydrographs and control them. One way to lessen the impact of the hydrograph in terms of the height of the water is to add a retention pond into the system. In many older areas in the lower parts of these watersheds, the land simply doesn’t exist. We know that. It’s very hard to find land! Wee go around looking for parking garages with the idea of using the bottom part of the parking garages as retention ponds. There have been suggestions of using the Rice Football Stadium as a retention pond as a better use of land than what it---now that is a joke, work with me here. Another option is an offsite or diversionary approach where you take a channel and send it over to an area that is nearby with the idea of diverting or “peak shaving” the water flow coming downstream. In the process, you end up with less runoff coming down stream. These are the two ways to control peak flows [retention ponds and peak shaving]. The idea of storage in a system is that you add storage back in—that is, you take that big hydrograph and you shift it in time and you lower it in height. You attenuate it and you lag it. And that’s good! Of course the idea of the Brays Bayou Project is to speed water out of the bottom end, using a larger channel and bridges raised, and slow water in the upper end using retention ponds and diversions. The idea is to separate the flows and add a lot more storage and capacity. It is an expensive undertaking. Now that completes my comments on hydrology and I want to shift gears. b. LIDAR mapping I want to tell you a little about LIDAR [acronym for LIght Detection And Ranging] mapping. You may have heard and read a lot about it in the newspaper, but I suspect that very few people in the audience have actually worked with it unless you’re a consultant. It’s a nasty beast because it gives you a data point every 15 feet on the ground, shot from a specially equipped airplane with a laser-guided system. It’s simply the best topography money can buy and apparently Harris County has been LIDAR-shot not once, but twice, once before Allison and once after Allison. The resolution and accuracy are really quite good. This LIDAR survey information formed the basis for all of the flood plain analyses now being produced, and it’s revolutionizing our ability to map floodplains.****** LIDAR enables you to fly through these watersheds and zoom in and view them in three dimensions. Now that we’ve got the LIDAR maps, we’re going to worry about taking rainfall and converting it to runoff and putting it into the stream and figuring out how high the level of water is going to get at the Texas Medical Center, the freeway crossing, or in the upper watershed. We’re going to take peak flows and put them into channel hydraulics and evaluate the water levels. We can also go further and evaluate flood control options. What if we divert or retain water, enlarge the channels, or raise the bridges? We can evaluate all of this. The technology is amazing. You can actually count the number of houses that are going to flood. The point I’m trying to make is that we can be a lot more accurate than we could even five years ago. The technology has changed things. But mapping is just the beginning. And it’s good that we can map things better. But we have to get the rainfall on the ground under control so that we don’t allow the problem to grow. We should not allow this problem to grow! For example, 105 out of 120 schools flooded in Harris County during Allison. This is an amazing statistic! But you will notice that most of the schools that flooded were NOT in the floodplain. This is not a surprise to those of us who work in this business. In fact, Harris County Flood Control experts have said many times that tropical storm Allison caused flooding in huge numbers of structures that were NOT in the flood plain. Just because we can accurately map all the flood plains does not necessarily mean that we’ve solved the flood problem. Rice University is now located in the 100-year flood plain. It didn’t use to be! Part of that is because a more careful analysis was done that included Harris Gully, which was not included in earlier analyses. Now we include it. The flood plain is getting larger over time. There are many reasons for this. Now we’re going to have to learn to deal with that. Storm water models show us that just because you have a nice pipe system and you have water running in the pipes doesn’t mean you are draining the watershed well. The Houston pipe system is designed to carry about a 2-year storm, maybe a 3-year storm in certain areas, and the bayous are supposed to carry 100-year storms. You put that combination together and everything is supposed to work. But the problem that we’ve seen in some areas is that the bayous are getting up a lot more often and at higher levels than ever before. As that happens, not only are they getting up higher, they are staying up! For example, Brays Bayou during Allison stayed up for eight or nine hours. Because of that all of the pipe systems shut down in and around the area and we got a back up of water, a lot of which came into the Texas Medical Center. In summary, with one inch of rain, the rainfall runs off, goes into to the pipes, goes down to the bayou, and everything is fine. At the second [higher] level of rainfall, the bayou water level goes up. The box culverts are running and there is really no restriction to flow—there’s no downstream backwater. At the third level of rainfall, the bayou is up, things are full up and are beginning to “surcharge.” At the fourth level, the bayous are very high, and water is backing up significantly into low areas. The low areas are not only the ponds, but the streets. We had five feet of water standing on Rice Blvd. We had 3 ½ feet of water standing on Fannin. The water in the streets and basements stays up as long as the bayou stays up, and then WHOOSH, the water’s gone. It drains out very quickly. But the bayou was up for a very long time during Allison. This same pattern is happening in many areas in Houston because the bayou capacity is not what it should be. The Brays Bayou Project is going to have a big impact on this situation. It’s going to drop those water levels 3 to 5 feet in and around the Texas Medical Center. That’s a lot. 1, 2, or 3 feet is a LOT in this part of the world. The result of the Harris County Flood District study was to add three major culverts, that is, to take the water coming into one culvert and divert it to three or four culverts. The idea is to divert some of the flows AWAY from the Texas Medical Center. You may ask, how did the flood-prone situation in Texas Medical Center get to be that way in the first place? Well, the drainage system was started in 1947, the year the culvert was dropped below ground level when they built the Baylor College of Medicine. This system needs to change to keep up with the times. Yes, there is an increase in flow when you divert water like that, but there’s a way to mitigate that, which is, in fact, part of the analysis. A great deal of work has gone into creating a solution to the Texas Medical Center flooding, which is a very, very critical problem. We cannot have the Texas Medical Center undergoing that level of impact ever again.” Editor’s Note: It is amazing the quality of presentations that are available online if one can locate them. Thank you to the Shell Center for Sustainability at Rice University for posting this Webcast online by Dr. Bedient so that other people in the US can learn from Houston’s flooding experiences. Sources: *See one couple’s story at: http://buddyandbobbie.com/TSAllison.html. ** Texas Medical Center is a consortium of health care institutions, including 2 medical schools and 13 hospitals with more than 6041 licensed beds. See the following URL for a list of institutions: http://www.tmc.edu/institutions/. ***Cocanour, Christine S., et. al.: “Lessons learned from the evacuation of an urban teaching hospital.” In Archives of Surgery, Vol. 137, October 2002, pp. 1141-1145. ****Philip Bedient’s biography is available at: http://www.ruf.rice.edu/~envi/xBEDIENT.HTML. *****“ Texas Water: Flooding and Watershed Management: How can the flooding risk in populated areas be minimized?” Rice Webcast Archive. Webcast available at: http://webcast.rice.edu/speeches/20040323eesi.html. ****** Explanation of LIDAR use in Harris County: “ LiDAR: A Vital New Use of Laser Technology: LiDAR stands for Light Detection and Ranging. As part of the Tropical Storm Allison Recovery Project (TSARP), highly detailed ground elevation data for all of Harris County will be acquired through this cutting-edge technology that utilizes the projection of millions of laser signals to the ground from a specially equipped aircraft. Using powerful software, the data from these LiDAR reflections is collected by measuring the time it takes for the aircraft to receive each of the millions of laser reflections. The resulting data is then combined and converted into an image that looks exactly like the terrain below, including buildings, trees, roadways, creeks and bayous. So, what will we do with this new data? In order to identify areas of higher flood risk, engineers need a detailed and accurate representation of the shape of the ground. It is just not economical to obtain such detailed information for an area as large as Harris County using conventional survey methods. But LiDAR makes it possible. The LiDAR data will be combined with surveyed creek and bayou cross sections in order to develop detailed computer simulations to determine an estimate of areas that have a higher risk of flooding.” Available at: http://www.hcfcd.org/lidar.asp?flash=yes. |
Copyright
2007 - SEMP INC., All Rights Reserved.