An Electric GIS
Electric transmission lines are a critical component of the modern electric power system. These high-voltage power lines carry wholesale electricity in bulk from power generators to local distribution systems or industrial consumers. The problem is that there simply aren’t enough of them. The per capita demand for energy in the United States has increased at a rate almost double that of the population growth over the last 20 years as consumers have purchased larger homes and equipped them with more home electronics. U.S. residential electricity use increased 23 percent between 1999 and 2009, and the trend is expected to continue with an additional increase of 20 percent by 2030. Between 2010 and 2030, the U.S. electric utility industry will need to make a total infrastructure investment of $1.5 trillion to $2 trillion--$300 billion of which is needed for transmission.
But even though there is a considerable need, there is increasing public opposition to new transmission lines. Typically, people don’t want transmission lines constructed near where they live, work or play. This reluctance is due to both real and perceived impact on the scenery, property values, land use and safety. Often, transmission line developers find themselves between a rock and a hard place.
Georgia Transmission Corp., a cooperative electric transmission services provider based in Tucker, Ga., recognized the need for an improved siting methodology that would be objective, consistent, standardized and defensible. “We saw an opportunity to improve the way we do business with a GIS-based siting process,” says Jerry Donovan, chief operating officer with Georgia Transmission. Working with professionals from Photo Science, the company took an aggressive path more than 10 years ago to develop an industry-leading software solution. That solution, now known as Corridor Analyst, is viewed as a model for the siting of electric transmission lines.
Ian McHarg, a Harvard-educated landscape architect and renowned land-use planner, developed a technique for siting linear facilities in the late 1960s based on “social values,” which he defined as the benefits and costs to society realized by the construction of the new facility. Many factors went into his broad concept of social values, including historic, water, forest, wildlife, scenic, recreation, residential, institutional and land values. A map transparency was created for each factor with the darkest gradations of tones representing areas with the greatest value and the lightest tones associated with the least significant value. All of the transparencies were then superimposed on one another over the original map. Following the format of each individual layer, the darkest regions showed the areas with the greatest overall social values and the lightest with the least. The social value map was then compared with similar maps constructed for geologic and hazard considerations, and the result was a clear picture of where to situate a controversial stretch of road.
The concepts that McHarg defined back in the 1960s served as the basis for the design of Corridor Analyst. The digital world of GIS greatly exceeds the analysis capabilities pioneered at that time with visual overlays. In fact, the digital environment provides an ideal medium for bringing together a significant amount of location-based data and analyzing a near-infinite number of alternative routes in this process. It also greatly improves productivity and analytical capabilities during the siting phase.
Three different perspectives are analyzed during the development of alternate corridors: engineering, natural environments, and built environments. A significant number of layers exists in each of these three models, and each has additional subclassifications. For example, layers in the natural environment include floodplain, streams/wetlands, protected lands, land cover and species-of-concern habitat. The built environment includes proximity to buildings, proximity to cultural resources, building density, proposed development, other water bodies, major property lines and land use. Avoidance areas--defined by community preferences or possible permitting delays--that are used in the analysis of all three perspectives include airports, schools and churches; EPA superfund sites; and military facilities, among others. In all, some 80 different layers are used in the analysis.
During corridor analysis, a 15- by 15-foot grid is overlaid and used in the development of suitability values covering the extents of the project area. Each feature is ranked in the suitability analysis numerically from 1 to 9 with 1 representing the most suitable and 9 the least suitable areas for transmission line construction. All categories are then weighted and combined into single numerical value for each project cell.
After several years of development, Georgia Transmission was armed with a very effective tool for siting new transmission lines. In 2003, however, the company took the process one step further by conducting an unconventional examination of its siting process and asking a host of industry, public advocacy and regulatory officials to take part. According to the company’s CEO, Mike Smith, the research project resulted in the company having “a more rigorous siting process and some important lessons that can improve the way many utilities make siting decisions.”
Georgia Transmission jointly funded and conducted the study with the Electric Power Research Institute Inc. (EPRI), an organization in Palo Alto, Calif., that conducts research and development relating to the generation, delivery and use of electricity for the benefit of the public. The study team comprised an impressive who’s who of planning experts and stakeholders from around the country, including personnel from both Georgia Transmission and EPRI, legal experts in the National Environmental Policy Act, four university professors, and GIS professionals from Photo Science. “This methodology requires us to determine values for each man-made and natural condition and rank potential impacts,” says John Goodrich-Mahoney, EPRI’s representative on the team. “As part of the process, the many judgments we make are quantified and documented. In the end, this kind of analysis helps make the siting practice and decisions more consistent.”
The study team refined a five-step process for siting. First, the team establishes macro corridors, which define large swaths of land potentially suitable for the new route between the two defined endpoints. Satellite imagery and readily available off-the-shelf data are used at this stage. Project area boundaries are then defined, effectively creating the outside limits of the macro corridors. Siting algorithms are employed to determine alternative corridors within the macro corridors. More-detailed project and geographic data are used at this stage. The project team then identifies alternate routes within each alternate corridor, and the siting process concludes after considerable analysis with the selection of a preferred route. The study took two years to complete. A report documenting the effort (“The EPRI-GTC Overhead Transmission Line Siting Methodology”) was published in 2006.
The first step in siting is the identification of macro corridors. These corridors provide a high-level view of the project area with a goal to eliminate areas from further consideration, thus focusing planners and engineers on project areas that offer the most viable siting alternatives. Readily available and relatively inexpensive geospatial data form the basis for this first analysis. Typically, 30-meter satellite data from Landsat Thematic Mapper are combined with publicly available data (e.g., public lands, airport locations, etc.) and linear infrastructure data to support the identification of macro corridors. The USGS Land Cover Institute maintains a National Land Cover Dataset (NLCD) that is useful at this stage.5
Satellite data were examined in the creation of this dataset and resulted in 21 thematic classes of land cover. These classes are valuable in the siting analysis. They include developed, which is further divided into low-intensity residential, high-intensity residential, and commercial/industrial/transportation classes; forested upland, which includes deciduous forest, evergreen forest, and mixed forest classes; and barren, which includes bare rock/sand/clay, quarries/strip mines/gravel pits and transitional classes. The USGS Web site (www.usgs.gov) includes a full listing of all thematic classes in the NLCD and how they compare and contrast to an Anderson land-use and land-cover classification system.6
Georgia Transmission goes beyond the NLCD in an effort to improve results for its siting activities. The company periodically contracts with a private firm to update the land-cover database for Georgia based on more-recent satellite imagery to provide an improved and more-current snapshot of actual land-cover conditions.
Once the geographical extents of the macro corridors are defined, more-detailed data acquisition takes place. Generally, this definition is based on high-resolution aerial photography. Sometimes, new aerial imagery is acquired specifically for the new project. Other times, existing imagery from either the U.S. Department of Agriculture’s National Agriculture Imagery Program (USDA NAIP) or similar projects can be used. Preference is given to imagery collected within two years of project startup.
More-detailed mapping is generated from this imagery. This mapping typically includes building centroids and land-use/land-cover classifications. More publicly available data are brought into the mix such as the U.S. Fish and Wildlife Service’s National Wetlands Inventory, flood maps and stream centerlines. Many times, information from county tax assessors is incorporated. Local planning departments are consulted to add proposed development activity. State historic preservation officers provide information on historic properties. All of these data are used to identify alternate corridors.
Even more-detailed data are collected in the second phase of data acquisition that is conducted prior to the development of a preferred route. Windshield surveys are conducted to classify the building centroids mapped in earlier phases of the project. These include residential, commercial, industrial, agricultural and unoccupied classifications. Any changes in the landscape not defined on any of the maps existing at this time are also noted. The overall data-gathering and analysis process is much like a funnel--it starts out with a very wide area and a correspondingly coarse filter. As the analysis with Corridor Analyst progresses, the amount of data and the level of detail increase considerably. In the end, the user has a highly defined area with very detailed data. The result is a preferred route with considerable statistics available to the planners and designers.
One new and interesting development is in the research and development phase. This is in the form of a visual index map or a map of the entire study area that highlights areas where overhead transmission lines would be most visible. The current thinking is that separate visual indexes will be created for residences, recreation areas, roads, etc. The analysis would be based on terrain and vegetation cover to provide a line-of-sight summary.
A key component of this methodology is the incorporation of external stakeholder input. Stakeholders, which represent members of community groups, regulatory agencies, conservation groups, other utilities, government agencies, elected officials, and others, help develop the criteria and determine the relative suitability and importance of the parameters used to identify alternate corridors. This is done on a programmatic basis, and the resulting model is used on multiple projects. “We have had great success obtaining objective input on a programmatic basis from regulatory agencies and representative organizations,” says Jesse Glasgow, a Photo Science manager who helps utilities and consultants implement the methodology. “This input can be more constructive than ‘not in my back yard’ (NIMBY)-type input and helps gain buy-in to siting results and control project schedules.”
Stakeholder input is used to create suitability maps for three different perspectives. One perspective is the built environment, which contains mapped features that represent human and cultural resource areas. Another is the natural environment, which includes mapped features that represent plants, animals, and hydrologic resources. Finally, there is the engineering concerns perspective, which addresses physical constraints and contains features for maximizing co-location and minimizing costs and schedule delays. A computer algorithm is applied to evaluate all possible routes to determine the routes most preferred from each perspective. The top 3 percent of all the routes are used for the alternate corridors. The stakeholder input is used to focus the project team within the alternative corridors for more detailed analysis.
To select the preferred route, all top-scoring routes are scrutinized by the project team in a procedure known as expert judgment. The team decides upon a set of issues or risk factors that may be unique to each project. These issues are more subjective such as public concern, maintenance accessibility, and schedule-delay risk. Using the expert judgment model, the project team selects the preferred route.
Finally, detailed land surveys are conducted to identify property boundaries as well as physical and environmental constraints such as wetlands and endangered-species habitat. Route adjustments are made during the design phase based on this information.
There is a significant need for a siting methodology that is objective, transparent, inclusive and consistent in its application. The methodology developed jointly by Photo Science, Georgia Transmission and EPRI meets this need.
In addition to overhead electric transmission lines, this methodology and tool is being applied to siting greenways and water transmission lines. It has also been adopted by power generation utilities for siting power plants.
1. Energy Information Administration, Energy Demand. Annual Energy Outlook 2009 with Projections to 2030, DOE/EIA-0383 (2009).
2. Chupka, M. W., Earle, R., Fox-Penner, P., Hledik, R. Transforming America’s Power Industry: The Investment Challenge 2010-2030 (2008).
3. “Ian McHarg: Overlay Maps and the Evaluation of Social and Environmental Costs of Land Use Change,” John Corbett, University of California, Santa Barbara.
4. Dillon, Barry, “GIS-Based Line-Siting Methodology,” Transmission and Distribution World, February 2005.
5. The U.S. Geological Survey (USGS) Land Cover Institute National Land Cover Dataset (NLCD) is available for download at http://landcover.usgs.gov.
6. Anderson, James, et. al., “A Land Use and Land Cover Classification System for Use with Remote Sensor Data”.
7. News Release, “Georgia Transmission Wins NRECA’s Cooperative Innovators Award for New Transmission Lien Siting Approach,” May 4, 2006.