[an error occurred while processing this directive] [an error occurred while processing this directive]
ArcNews Online
 

Fall 2004
Search ArcNews
 

E-mail to a Friend

The City of Atlanta, Georgia, Meets 21st Century Water/Wastewater Demands With GIS

Clean and safe water is essential to maintaining natural ecosystems, strong economies, and healthy populations free of waterborne disease. In the United States, most of the major infrastructure systems that provide drinking water and treat wastewater were built during the period following World War II and will reach the end of their useful lives within the next 20–40 years. Population and economic growth, which often brings sewer and storm water overflow problems, are causing further stress on the already overloaded and aging systems. Planning and constructing their replacement are essential for public health, ecology, and the nation's economic welfare.

  click to enlarge
Example of color infrared imagery showing the Georgia Tech area.

Like many metropolitan regions in the United States, the city of Atlanta, Georgia, and its surrounding communities have experienced tremendous population and economic growth over the last 100 years. According to Clean Water Atlanta (www.cleanwateratlanta.org), a century ago the city "used open watercourses to capture storm water and divert it into nearby brooks and creeks. As the city grew, and as sinks, bathtubs, and flush toilets became popular, storm water sewers became conduits for carrying household wastewater as well. These drains were covered and transformed into combined sewers for the collection of storm water and sewage."

These types of combined sewers, very common in major cities, were not designed to remove the waste, which then flowed into surface streams and became a public health hazard. To remedy the problem, modernized sewers were designed and constructed to carry storm water and household waste. They were quite effective until the system again became overloaded as Atlanta continued to experience rapid growth. Storm water runoff from newly built paved surfaces and household wastewater loads is greater than ever.

In support of Atlanta's need to build new and separate storm water and groundwater systems to meet 21st century demands, Paul Beaty, research scientist at the Georgia Institute of Technology (Georgia Tech) Center for GIS, and his team set to work on mapping the impervious surfaces in Atlanta using ArcGIS (ArcInfo) from Esri and ERDAS IMAGINE and OrthoBASE from Leica Geosystems GIS & Mapping, an Esri Business Partner. Faculty, staff, and students at Georgia Tech have been using Esri and Leica Geosystems software products for many years and maintain site licenses. For this project, the team used four Dell PCs running Microsoft Windows 2000 Professional and a network server. The resulting data serves as an important input for the preliminary design and budget of the new sewer and storm water runoff systems as well as a starting point for determining future fees based on each parcel's contribution to storm water runoff.

  click to enlarge
Example of black and white imagery showing the Georgia Tech area.

Beaty and his team of research scientists, an engineer, and undergraduate and graduate students began the project in June 2003 by gathering existing reliable CAD, imagery, and GIS data for the entire city of Atlanta. The CAD data acquired from the city Water Department included polygons, such as building footprints, roads, and pavement edges, that did not snap or close. Using ArcInfo, the CAD data was imported and run through multiple processes to be cleaned and built into GIS data.

"ArcInfo software's batch processing handled the majority of CAD data cleanup," according to Beaty. "There was some manual editing, but it was very minor compared to the automated cleanup by ArcInfo." Manual editing was performed using imagery as a backdrop and reference source. "This methodology saved us from making several errors. We shortened the project time because errors were not created while producing the new data."

Beaty and his team used ERDAS IMAGINE Professional to process three different sets of imagery to identify impervious surfaces: winter 1999 color infrared digital orthophoto quarter quadrangles (DOQQs) (United States Geological Survey), summer 2000 true color imagery (Georgia Aerial Surveys), and 2001 six-inch black-and-white aerial photography (city of Atlanta).

  click to enlarge
This photo shows how the black and white imagery can be used to pick up bright features such as driveways and sidewalks. (Source for all images is City of Atlanta.)

First, they orthorectified the DOQQs. "Then using IMAGINE software's spatial modeler, we ran various indices, such as a vegetation index, across the three image sets," explains Beaty. From these processes, Beaty and his team created six spatial models that were applied to the imagery. Team members interpreted and analyzed the accuracy and precision of each model for each image set and then determined if any subsequent models needed to be developed. "The hardwood forest floor identified in the winter images was initially falsely classified as an impervious surface. Using the summer true color imagery, we were able to accurately identify the forest and classify the natural forest floor as a highly permeable surface."

The CAD data contained many buildings but did not include newer residential subdivisions, commercial buildings, and shopping centers. Beaty explains, "We used the imagery to identify areas that were not available in the CAD data. This was an important step in the project because these areas represent huge amounts of impervious surface." The team used the black-and-white aerial photography, resampled to a one-foot resolution, to identify bright features such as sidewalks.

  click to enlarge
CAD data showing the Georgia Tech area, cleaned and converted into GIS vector data.

After the best model for each data set was created, the project team members compared the imagery with the vector data to see if they could combine the data for an overall view. The result was a model that combined the data into one file for the first time. The data was classified as pervious or impervious for the entire city of Atlanta. The total amount of impervious surfaces was calculated for the entire city and then further classified as residential, commercial buildings, or commercial parking lots. Students performed field studies to assess the quality of the data, which allowed for the calculation of a confidence level in the data.

Using ArcInfo, the impervious surface data was then overlaid with parcel boundaries provided by the city of Atlanta, DeKalb County, and Fulton County. Beaty continues, "We ran the Union and Intersect commands to calculate the total impervious surface area per parcel. Finally, we calculated the percent of the total parcel area that is impervious and added these calculations to a copy of the original parcel database."

  click to enlarge
This image shows edited vector data overlaid with the parcel boundaries.

The team at Georgia Tech delivered the data as shapefiles on CD–ROM and via the institute's FTP site to the city of Atlanta's Water Department. "This data is a great resource for planning the new systems," states Beaty. Currently the city of Atlanta can determine where and how much new sewer infrastructure needs to be built based on current water use and sewer billing records. However, the city will use the impervious surface data as an input to calculate where and how much infrastructure needs to be built for the new storm water runoff system. Just like current water and sewer bills, storm water runoff system fees will be parcel based. Owners will be charged a use fee based on the amount of surface water runoff generated from their property. For example, an empty, undeveloped parcel will not have a storm water runoff use fee, but a parcel that is built up with a house, patio, and driveway will be charged a use fee. This fee model, based on water, sewer, and storm water runoff use, is expected to be implemented in some way throughout the entire United States.

  click to enlarge
Converted CAD data shown with the imagery as the backdrop so that accurate editing could be done.

The American Water Works Association defines the current time period as the "Dawn of the Replacement Era" in the United States. For the first time in the country's history, underground water and sewer pipes will need to be replaced. According to the Environmental Protection Agency (EPA), 100 years ago the average American used only about 10 gallons of water a day to drink, cook, clean, and bathe. Today, Americans use 100 gallons a day per person on average. The pipe systems obviously were not designed for today's larger population, high usage, or built-up areas. The city of Atlanta is diligently working toward compliance with the EPA mandate that service providers separate storm water and groundwater systems to ensure clean water for its residents and downstream neighbors. The EPA mandate has far-reaching effects throughout the United States because many sewer and storm water systems currently are combined.

The expected replacement cost over the next several decades exceeds hundreds of billions of dollars. However, government officials, businesses, and informed citizens agree that the cost of not replacing the infrastructure would be even greater. Clean water and wastewater management are essential for industries that generate billions of dollars a year in income such as recreation and commercial fishing.

For more information on the impervious surface study in the city of Atlanta, contact Paul Beaty, Center for GIS, Georgia Institute of Technology (e-mail: paul.beaty@coa.gatech.edu). Visit Georgia Tech's Center for GIS at www.coa.gatech.edu/cgis. Visit Clean Water Atlanta at www.cleanwateratlanta.org. Visit the American Water Works Association at www.awwa.org.

For more information on spatial solutions, contact Leica Geosystems GIS & Mapping (toll free in the United States: 1-877-463-7327; outside the United States: 404-248-9000; Web: gis.leica-geosystems.com).

[an error occurred while processing this directive]