These are but a few of what the federal House Committee on Transportation and Infrastructure has estimated to be an average of 788 water main ruptures per year in Philadelphia. Luckily, they aren't nearly as dangerous as gas-line breaks, such as the February 2004 explosion at the intersection of Ogontz and Olney avenues, when a Water Department work crew accidentally struck a high-pressure gas main. That break lit up the sky like fireworks, and a blazing four-alarm fire sent residents of two nearby high-rises running for cover.

Illustration By: Bill Westervelt

"If you knew the condition of what's under the city, you probably wouldn't walk on the sidewalk," says Fred Graf, a former senior research engineer for PECO.

Graf has seen all kinds of things buried beneath the streets — from gas lines and water mains to time capsules and mass graves — yet he's never dug up one bit of ground. He has spent his career developing the use of a technology called ground penetrating radar (GPR) to gather images of Philadelphia's underground, which is like taking X-rays of the city. The medical analogy is fitting, because a city is much like a living organism, with its most vital parts invisible to the eye. What's out of sight tends to remain out of mind — that is, until it floods the streets or suddenly bursts into flames.

Strolling down the city sidewalks, we implicitly assume that we are treading on solid earth. In reality, just beneath our feet lies a vast and dark world: 6,000 miles of gas mains; 3,300 miles of water mains; 3,000 miles of sewers collecting nearly 500 million gallons of sewage per day; immense labyrinths of communications cables and fiber optics; electrical conduits; oil plumes; storage tanks; tunnels; abandoned subway stations; graves; hidden waterways; archaeological sites; mines; landfills; and more.

If the integrity of any of these underground structures is compromised, so too is the integrity of the sidewalk. A break in a pipe can set off a chain reaction that ends with the sidewalk opening up, forming a sinkhole and swallowing whatever stands on its surface. On June 28, two cars plummeted 70 feet when an enormous sinkhole appeared on Route 924, north of Pottsville. That same day, two truckers were killed after their rigs fell into a sinkhole on Interstate 99 in New York state.

Every year, the American Society of Civil Engineers (ASCE) releases a report card for each state assessing the condition of its infrastructure. This year, Pennsylvania will not be making the honor roll. Our overall GPA was a D. Our drinking water earned a D-plus for "a $12 billion+ funding gap," "incomplete data at the state level," "increasing violations" and "potential threat to public health, the environment, and the economy."

Our wastewater system fared worse with a D-minus, for, among other things, the short remaining life spans for all aspects of infrastructure and existing pollution of lakes and streams. "The useful life of Pennsylvania's wastewater infrastructure is about to expire," the report card reads. "Treatment plants typically have an expected useful life of 20 to 50 years before they have to be expanded or rehabilitated. Pipes have life cycles ranging from 15 to 100 years. ... And in some Pennsylvania cities, a number of pipes are approaching 200 years old."

Age, however, does not always determine a pipe's life span. "Age doesn't give you a great sense of what's going to break," says Rae Zimmerman, professor and director of NYU's Institute for Civil Infrastructure Systems. "A study in New York showed that age doesn't correlate with breakage — it's the environment."

By "environment," Zimmerman is referring to things like changes in the soil around the pipe due to freezing and thawing, added weight from a new building erected overhead, or accidental blows from construction workers digging underground.

So just how bad are things down there?

The answer is as simple as it is frightening: We don't know.

Some cities have already embraced the technology. "The city of New York employed a kind of robot to move through some of the larger water systems and sense any leaks or imperfections in the walls," says Zimmerman.

And in Pittsburgh, a company called RedZone Robotics developed a robotic system called Responder that maneuvers its way through sewers, operated by a remote joystick, looking for problems in the pipes. Responder is equipped with laser and sonar sensors that scan the interior pipe walls — lasers operate above the water line while sonar can scan through the water. The slightest bit of corrosion or the smallest leak will slightly change the time it takes for a signal to hit the wall and bounce back, so advanced software can construct extremely detailed 3-D models of the pipe walls.

Meanwhile, here at home, the Philadelphia Water Department has spent the past year launching a $6 million pilot sewer assessment program to evaluate the condition of the sewer system. A database was to be developed, along with a ranking system to prioritize maintenance. So far they've videotaped and evaluated only 230 miles of sewers out of 3,000.

Recent rains that have been pounding our region are also pushing our aging infrastructure to its limit. In Yardley, floodwaters washed away the soil supporting a buried sewer pipe, and the clay conduit collapsed. Now, damage to Yardley's sewer system is likely widespread, and as sewage and storm water leak through the cracked pipes, the roadways and ground above may begin to sink. The situation is dire enough that the Yardley sewer authority has called in the Federal Emergency Management Agency.

Philadelphia is no stranger to sinking land.

Back in the 1960s, a collapsed sewer in the Mill Creek neighborhood of West Philly sunk several homes. In 2000, the city bulldozed the last of nearly 1,000 sinking homes on Logan triangle in North Philly. Both areas were built atop unsteady landfills made of ash, cinder and debris — stuff that settles and sinks when oversaturated with water.

"Sinkholes are often alluded to being acts of God," Graf says, "but water does the primary damage. Rain and leaky pipes lead to sinkholes. Everybody likes to blame it on God."

But the blame lies with the pipes, and until we figure out exactly what the situation is below our streets, it is impossible to predict where the next sinkhole might open up or what building might start to go under.

When utilities wait until after something breaks to fix it, the costs, labor and disruptions can be through the roof. Repairs to the Yardley sewer system, for instance, might cost up to $20 million.

"This was a big spill," says Colin Flaherty, spokesman for Insituform Technologies, a pipe repair company based in suburban St. Louis. "But it didn't get much attention. Not as much as Alaska — yet the pipes in Yardley are much worse. Of course, it's not just Yardley. Pipes all over the Philadelphia area, and all over the country, are in bad shape."

Insituform has developed technology that can fix a pipe from the inside before it breaks, without any digging. They fill a tube with a special kind of resin, turn it inside out and propel it through the pipe. Then, they heat the water inside the pipe; the resin expands outward and attaches to the interior walls, then hardens due to the heat. The result is a new pipe inside the old pipe. The company actually used this technique on the sewers beneath the White House, which had been built around the time of the Civil War. For security and aesthetic reasons, the government didn't want anyone digging up the White House lawn.

Even when hits appear less catastrophic, the strain caused to a pipe can result in a delayed break days, weeks, even months later. By that time, the damage has been reburied and can be difficult to find.

To avoid damage to underground utilities, every state in the country has created a One Call Center, a nonprofit organization whose job is to mediate communications between the people who are digging and the utility companies. State law requires anyone planning to dig to notify the One Call Center and receive a ticket. The One Call Center then tells the various utility companies where the digging is to take place and the utility companies are in turn responsible for going out to the site and marking, in color-coded paint or chalk, where their lines are buried. Those who decide not to call are held liable for any damages and can be slapped with fines as high as $25,000. In the Verizon fiasco, some of the blame goes to the contractors who started digging before the utility companies had a chance to mark their lines. Many of the accidents, however, resulted from failures of the utility companies to accurately locate their own lines.

In 2003, PA One Call received more than 500,000 calls, yet there were still roughly 20,000 hits to utility lines in the state.

The following summer, after seeing statistics showing that workplace injuries and fatalities were on the rise, Gov. Ed Rendell launched a "Call Before You Dig" awareness campaign, saying, "I am confident that by increasing use of the PA One Call system we will prevent future injuries and save countless thousands of dollars in damage and utility interruptions."

According to the Office of Pipeline Safety, from 2003 to 2005, Pennsylvania averaged nearly $6 million in property damages each year. The average number of third party damages to utility lines is down to about 1.2 per 1,000 tickets, according to PA One Call executive director Bill Kiger. That's far better than what the federal government deems a good rate: 2.5 per 1,000 tickets. Still, with more than 630,000 calls in 2005, that's more than 750 utility damages.

What's causing the damage?

"Forty to 50 percent of damages in Pennsylvania are a result of no notification," Kiger says. "The rest fall into three main areas. One is simple accidents — they bump a lever and really didn't mean to. Another is not paying attention to the marks. And of course, there are cases when the utility didn't find the line."

The city's entire infrastructure has been built and later repaired in a piecemeal fashion over hundreds of years, and the paper records that show what's where have often been scattered about, lost or are simply inaccurate.

"Philadelphia is an old city," says Lucio Soibelman, an associate professor of civil and environmental engineering at Carnegie Mellon University, "so you have old infrastructure and new infrastructure. You have new pipes that are being mapped with GIS [geographical information systems] technology and you have old things that no one knows is there. This is not something that was designed in a perfect way. There's a lot going on, and a lot of research is needed to find out what is underground."

Until we know precisely where pipes are buried, construction crews can't avoid hitting them. Yes, the statistics show that increasing use of One Call systems has lessened the blow to our city's internal organs. But even in the best case scenario, when every party does what they're supposed to, the ability of a utility company to accurately mark its lines is only as good as the data available.

"Many of the records are old, the references change, and you have a good chance of not knowing what's underfoot in any place in the city when you open up and dig," Graf says.

Today, the utility companies are all converting to GIS — computer programs that take all kinds of information (maps, images, raw data, etc.) and organize it in a user-friendly, searchable way.

The Philadelphia Water Department, for instance, recently completed a massive three-year $7 million data conversion project, translating all their paper documents (more than 215,000 pages) detailing the locations and characteristics of the city's water and sewer infrastructure into a high-tech GIS program. Digitizing infrastructure data is a vast improvement — it's almost hard to believe that anyone in the last decade has been working from paper documents

But what's the use of a sophisticated GIS database when the data going into it aren't reliable? As Graf says, "It's garbage in, garbage out."

The city needs to collect better data, and the necessary technology exists. The old method of finding buried utilities is to use an electromagnetic or radio frequency line locator — basically a metal detector. The problem, though, is that many underground utilities aren't metal. Many gas pipes are plastic, and the sewer system is largely terra-cotta or plastic. Fiber optic cables and many of the newer nonmetallic pipes contain what are called tracer wires so they can be picked up by metal detectors, but older pipes remain invisible.

GPR, on the other hand, can find a much broader range of underground objects. Similar to the way sonar is used to locate objects under water, GPR sends thousands of radar pulses per second into the ground, where the signals are either absorbed or reflected back to be caught by receiving antennae. By timing, in nanoseconds, how long it takes the various signals to bounce back, the GPR software can create an image of the material content beneath the surface.

"It's a very successful technique of locating subtle changes within the ground," Graf says. "GPR was originally used to find reinforcement bars in the walls of nuclear plants. An electrician drilling through the walls could impact the integrity of the structure so it was crucial that he wouldn't hit anything. X-rays worked, but they were tedious and the radiation itself was a problem. GPR worked very nicely."

In addition to being able to locate objects underground, GPR can also locate cavities, the potential seeds of sinkholes, making it possible for us to act before they open up and swallow the world above.

Witten had developed a technology called CART (Computer Assisted Radar Tomography). CART consists of a mobile GPR array of transmitters and receivers plus a laser tracking device that overcomes standard GPR's biggest limitation. It allows for extremely accurate depth perception.

Measuring depth with standard GPR is a tricky process. The strength of a radar pulse weakens when it travels through electrically conductive material like clay or saturated soil, making it harder to get a clear image. If the dirt through which the signals are traveling were perfectly uniform from the surface down to the buried utility, GPR would have perfect depth perception. But it's not. Estimating depth with GPR is more of an art than a science, and the only real way to figure out how deep a pipe is buried is to dig.

Not so with CART. Surveys using CART produce remarkable 3-D images of the complex world beneath the earth's surface, showing precisely where various objects are located on scales of thousands of square meters with a resolution of centimeters.

"The ability to take those readings and convert them into a graphic display is pretty amazing," Kiger says. Virtual-reality technology allows a viewer to fly through the 3-D map, seeing what's buried there from all possible angles.

When NASA's Jet Propulsion Laboratory was asked to employ its state-of-the-art image processing techniques to find incredibly subtle aspects of the 9/11 damage, they used Witten's images. In fact, Witten had already been working in the area in a project for Con Edison before 9/11, so they were able to provide before and after images for comparisons.

"In 1986, they had this roadside incident that injured and killed 300 people," says Kiger. "That was the impetus for Rodic. They also had an earthquake. So the Japanese government footed 60 percent of the bill."

Here in Philadelphia, earthquakes are not a major threat, but the protection of critical infrastructure like water and gas from terrorist attacks is a national priority, and could provide additional incentive and funding for such accurate mapping of the underground. At the moment, securing finances is the main obstacle. "The unfortunate thing is that no one is funding all of this," Kiger says. "Money is in short supply."

But there's no question that such an investment would save money, prevent injuries and reduce service disruptions in the long run. In 2002, for instance, PECO completed a yearlong pilot program testing locating technologies for its underground gas lines and estimated that the company would save $1 million a year with improved utility locating.

"Cities ought to be screaming for support," says Jim Garrett, professor of civil and environmental engineering at Carnegie Mellon. "They should be asking for better ways to collect data, better ways to make decisions, etc. And they can team up — a solution for one city could be a solution for another."

The ASCE report card offers some guidance:

"Focus on technology. State government can play an essential role in promoting research, development, testing and evaluation of new technologies and the dissemination of information about proven technologies. ... By creating research partnerships with universities throughout the state, Pennsylvania may reap additional economic benefits through public-private partnerships and licensing of new technologies."

Kiger is hoping that new Pennsylvania legislation will move us forward in terms of gathering data.

"We are currently in the throes of the reauthorization of a state law that requires subsurface utility engineering, or SUE, and the use of varying levels of technology," says Kiger. SUE is a sort of management process for collecting information prior to an excavation, and if the construction job is complex or expensive enough, it requires the use of locating technology like GPR. "Then, whatever information they gather, they'll have to submit to us at the One Call Center. We become a repository and the utilities can come and extract information to update their own information at virtually no cost."

The bill has passed the state House and Senate and is now back to the Consumer Protection Committee for review. Kiger says it could be approved by the end of the year.

"If we could get everybody calm enough that they could understand that there can be a repository of this information," Kiger says, "wouldn't that make all of our design work, all of our expansion that much quicker and that much simpler for everyone concerned? It would provide a better growth rate in the state and in Philadelphia."

Creating such a repository of information will help, but it will be piecemeal and incomplete.

Experts agree that what Philadelphia needs is a citywide, 3-D digital map of its entire underground, continuously updated in real time, compiled from new data collected in a systematic fashion using the latest and best technology. "More accurate records could avoid a lot of problems," says Graf, "but you're going to get tremendous resistance from utilities. The utilities philosophy is entrenched. They can be resistant to new technologies."

"The goal is intelligent modeling technology," says Jim Garrett, professor of civil and environmental engineering at Carnegie Mellon. "Cities should be driving this more. Owners and managers of infrastructure need to be requesting more from the research and product development communities. They need scientifically collected data. They need to know what's actually out there — not a patchwork of human estimates of what's out there."

(a_gefter@citypaper.net)