CHAPTER 8 Mapping GIS provides powerful and cost-effective tools for creating intelligent maps for water, wastewater, and stormwater systems. A sewer system map created by GIS (Borough of Ramsey, New Jersey). 2097_C008.fm Page 137 Monday, December 6, 2004 6:04 PM Copyright © 2005 by Taylor & Francis LEARNING OBJECTIVE The learning objective of this chapter is to understand how to create GIS maps for water, wastewater, and stormwater systems. MAJOR TOPICS • Mapping basics • Map types • Advantages of GIS maps • GIS mapping steps • Mapping case studies LIST OF CHAPTER ACRONYMS AM/FM Automated Mapping/Facilities Management AML Arc Macro Language DRG Digital Raster Graphics (USGS topographic maps) NAD-27 North American Datum of 1927 NAD-83 North American Datum of 1983 QA/QC Quality Assurance/Quality Control SPC State Plane Coordinate (Map Projection System) TIGER Topologically Integrated Geographic Encoding and Referencing System (U.S. Census Bureau Mapping System) UTM Universal Transverse Mercator (Map Projection System) VBA Visual Basic for Applications This book focuses on the four main applications of GIS, which are mapping, monitor- ing, maintenance, and modeling and are referred to as the “4M applications.” In this chapter we will learn how to implement the first m (mapping). LOS ANGELES COUNTY’S SEWER MAPPING PROGRAM In the 1980s, the Sanitation Districts of Los Angeles County, California, envisioned a computerized maintenance management system that would provide decision makers with essential information about the condition of the collection system. A sewer and manhole database was subsequently developed, but investments in GIS technology were deferred until the early 1990s when desktop PCs became powerful enough to run sophisticated GIS applications. In 1993, a GIS needs analysis study was performed, which recommended implementation of a large-scale enterprise-wide GIS. An in- house effort was started to implement the recommendations of the study. Several sections in the Districts formed a project committee to pilot test GIS technology that could be duplicated in all 25 districts. The pilot project developed a mapping appli- cation for a small district that acted as a front-end to the large database of sewerage 2097_C008.fm Page 138 Monday, December 6, 2004 6:04 PM Copyright © 2005 by Taylor & Francis facilities developed in the 1980s and 1990s. At this point, much of the information was nonspatial, including multiple databases in a variety of formats and paper maps. Converting this information into GIS proved to be the most time consuming and costly operation. Creation of the layers for sewers and manholes was the most laborious of all the layers that had to be created. Manhole data were represented by approximately 24,000 points or nodes digitized from the paper maps, using a base map. The GIS layers were created in CAD software and linked to the legacy databases. Once linked, detailed data such as sewer pipe and manhole construction material, size, condition, flow, capacity, and inspection data were available for query and analysis through an intuitive map-driven interface. By 2003, the pilot project had grown to become the first enterprise-wide solution deployed by the Districts. Called “Sewerage Facilities GIS,” the system allowed users to access and view data from the legacy databases by selecting sewers and manholes on a map or using standard queries. The mapping application also provided sewer tracing functionality that proved helpful in delineating study area boundaries for design projects, annotating sewers with flow direction, and tracking potential discharge violations (Christian and Yoshida, 2003). MAPPING BASICS The basic concepts essential for understanding GIS mapping are summarized in the following subsections. Map Types There are two major types of GIS maps: vector and raster. In vector format, objects are represented as points, lines, and polygons. Examples of the vector format are maps of water mains, hydrants, and valves. Scanned maps, images, or aerial photographs are examples of raster format. Raster data are also referred to as grid, cell, or grid–cell data. In raster format, objects are represented as an image consisting of a regular grid of uniform size cells called pixels , each with an associated data value. Many complex spatial analyses, such as automatic land-use change detection, require raster maps. Raster maps are also commonly used as base maps (described later in this chapter). Existing paper maps that are used to create GIS maps are called source maps. Topology Topology is defined as a mathematical procedure for explicitly defining spatial relationships between features. Spatial relationships between connecting or adjacent features, such as a sewer tributary to an outfall or the pipes connected to a valve, which are so obvious to the human eye, must be explicitly defined to make the maps “intelligent.” A topological GIS can determine conditions of adjacency (what is next to what), containment (what is enclosed by what), and proximity (how near some- thing is to something else). Topological relationships allow spatial analysis functions, such as network tracing, that can be used to facilitate development of hydraulic models for water and sewer systems. 2097_C008.fm Page 139 Monday, December 6, 2004 6:04 PM Copyright © 2005 by Taylor & Francis Map Projections and Coordinate Systems Because the Earth is round and maps are flat, transferring locations from a curved surface to a flat surface requires some coordinate conversion. A map projection is a mathematical model that transforms (or projects) locations from the curved surface of the Earth onto a flat sheet or 2D surface in accordance with certain rules. Mercator, Robinson, and Azimuthal are some commonly used projection systems. Small-scale (1:24,000 to 1:250,000) GIS data intended for use at the state or national level are projected using a projection system appropriate for large areas, such as the Universal Transverse Mercator (UTM) projection. The UTM system divides the globe into 60 zones, each spanning 6˚ of longitude. The origin of each zone is the equator and its central meridian. X and Y coordinates are stored in meters. Large-scale local GIS data are usually projected using a State Plane Coordinate (SPC) projection in the United States. A datum is a set of parameters defining a coordinate system and a set of control points with geometric properties known either through measurement or calculation. Every datum is based on a spheroid that approximates the shape of Earth. The North American Datum of 1927 (NAD27) uses the Clarke spheroid of 1866 to represent the shape of the Earth. Many technological advances, such as the global positioning system (GPS), revealed problems in NAD27, and the North American Datum of 1983 (NAD83) was created to correct those deficiencies. NAD83 is based on the GRS80 spheroid, whose origin is located at the Earth’s center of mass. The NAD27 and NAD83 datum control points can be up to 500 ft apart. Coordinates are used to represent locations on the Earth’s surface relative to other locations. A coordinate system is a reference system used to measure hori- zontal and vertical distances on a map. A coordinate system is usually defined by a map projection. The GIS and mapping industries use either latitude/longitude- or geodetic-based coordinate grid projections. Because much of the information in a GIS comes from existing maps, a GIS must transform the information gathered from sources with different projections to produce a common projection. Map Scale Map design addresses two fundamental map characteristics: accuracy and depicted feature types. Both characteristics vary with map scale. Generally, larger scale maps are more accurate and depict more detailed feature types. Smaller scale maps, such as U.S. Geographical Survey (USGS) quadrangle maps, generally show only selected or generalized features. Table 8.1 summarizes the relationships among map scale, accuracy, and feature detail. Data Quality The famous computer industry proverb “garbage in, garbage out” conveys very well the importance of GIS data quality. A GIS map is only as good as the data used to create it. Data quality roughly means how good the data are for a given application. Data quality is important because it determines the maximum potential reliability of the GIS application results. Use of inappropriate data in a GIS map may lead to 2097_C008.fm Page 140 Monday, December 6, 2004 6:04 PM Copyright © 2005 by Taylor & Francis misleading results and erroneous decisions, which may erode public confidence or create liability. Data Errors There are two types of data errors: inherent errors embedded in the source of data and operational errors introduced by users during data input, storage, analysis, and output. Inherent errors can be avoided by using the right kind of data. Operational errors can be prevented by quality control and training. A data conversion team should be aware of sources and magnitudes of data error. For example, spatial information in USGS 1:24,000-scale (7.5-min) topographic maps is certified to have 90% of its features within 50 ft (15 m) of their correct location. 50 ft is large enough to underestimate the runoff from a new development and undersize a detention pond for adequate stormwater management. Map Accuracy A primary factor in the cost of data conversion is the level of positional accuracy. Required map accuracy and resolution depend on the application in which the maps will be used. A 2000 survey conducted by the Geospatial Information and Technol- ogy Association (GITA) indicated that the water utilities were seeking more landbase accuracy of 5-ft compared with other utilities, such as the 50-ft accuracy sought by the gas companies (Engelhardt, 2001; GITA, 2001). The same survey for 2002 indicated that the water industry required the highest accuracy in their GIS projects. Among the water organizations, 27% were using a 6-in. landbase accuracy compared to electric (12%), gas (17%), pipeline (17%), and telecom (0%) organizations (GITA, 2003). These data reveal that a trend toward increasing accuracy may be emerging in the water industry. Engineering applications usually require ±1 to 2 ft accuracy. For planning and regional analysis applications, ±5 to 10 ft accuracy is generally appropriate (Cannistra, 1999). Sometimes relative accuracy (e.g., ±1 ft from the right-of-way line) is more important than an absolute level of accuracy (e.g., ±1 ft from the correct location). For the applications where positional accuracy is less important, supposedly low-resolution data, such as USGS digital orthophoto quadrangles (DOQs), may be acceptable. In other applications where features must be positioned within a foot of their actual position, even the presumably high-resolution data, such as 1-m IKONOS imagery, Table 8.1 Relationships among Map Scale, Accuracy, and Feature Detail Map Scale Minimum Horizontal Accuracy, per National Map Accuracy Standards Examples of Smallest Features Depicted 1 in. = 50 ft ± 1.25 ft Manholes, catch basins 1 in. = 100 ft ± 2.50 ft Utility poles, fence lines 1 in. = 200 ft ± 5.00 ft Buildings, edge of pavement 1 in. = 2000 ft ± 40.00 ft Transportation, developed areas, watersheds 2097_C008.fm Page 141 Monday, December 6, 2004 6:04 PM Copyright © 2005 by Taylor & Francis may not be accurate enough. As a rule of thumb, a database built from a map will have positional inaccuracies of about 0.5 mm at the scale of the map because this is the typical line width of the drawing instrument. This can cause inaccuracies of up to 12 m in a database built from 1:24,000 mapping, such as USGS DRGs (Goodchild, 1998). Precision and accuracy are two entirely different measures of data quality and should not be confused. A GIS can determine the location of a point feature precisely as coordinates with several significant decimal places. However, many decimal places in coordinates do not necessarily mean that the feature location is accurate to a 100th or 1000th of a distance unit. Once map data are converted into a GIS environment, the data are no longer scaled, as the data can be scaled as desired to create any output map scale. However, the spatial data can never be any more accurate than the original source from which the data were acquired. GIS data are typically less accurate than the source, depending on the method of data conversion. Therefore, if data were captured from a source map scale of 1 in. = 2000 ft, and a map was created at 1 in. = 100 ft, the map accuracy of features shown would still be 1 in. = 2000 ft (PaMAGIC, 2001). MAP TYPES Various map types used in GIS are discussed in the following subsections. Base Map The map layers are registered to a coordinate system geodetic control framework and a set of base information, often referred to as a base map. The foundation for a successful GIS mapping project is an appropriately designed base map. The base map is the underlying common geographic reference for all other map layers. The common reference provides registration between various layers and allows them to be overlayed, analyzed, and plotted together. Because the base map serves as the reference layer for other layers, its accuracy can affect the accuracy of other layers. This is especially true if the base map is used to create other layers by on-screen (heads-up) digitization. Selection of an appropriate base-map scale is largely determined by the earlier choice of GIS applications. Each application inherently requires a certain minimum base-map accuracy and certain map features. For engineering and public-works applications, the required map accuracy is in the range of ±1 ft, as dictated by the need to accurately locate specific physical features, such as manholes and catch basins. Planning applications, which most often deal with areawide themes, do not generally require precise positioning. Accuracies of ±5 ft, or perhaps as much as ±40 ft, are often acceptable. Less detailed maps, showing nothing smaller than roads and buildings, for example, may be adequate for many planning applications. Whatever the range of mapping requirements, the base map must be accurate and detailed enough to support applications with the most demanding map accuracies of better than ± 2 ft. Utility asset location also requires mapping that depicts specific small features such as manholes and catch basins. As shown in Table 8.1, these requirements are met by a map scale of 1 in. = 50 ft. 2097_C008.fm Page 142 Monday, December 6, 2004 6:04 PM Copyright © 2005 by Taylor & Francis There are three common types of base maps: digital orthophotos, planimetric maps, and small-scale maps. Digital Orthophotos For laypersons, digital orthophotos (or orthophotographs) are scanned aerial photos. For GIS professionals, they are orthorectified raster images of diapositive transparencies of aerial photographs. Creation of a digital orthophoto requires more than a photo and a scanner, and includes surveyed ground control points, stereo plotters, and a digital elevation model. In fact, the digital orthophoto creation process involves many steps, which are listed below: • Establish ground control • Conduct aerial photography • Perform analytical aerotriangulation • Set stereo models in stereo plotters • Capture digital elevation models • Scan aerial photographs • Digitally rectify the scanned photographs to an orthographic projection • Produce digital orthophotos Digital orthophotos are popularly used as base maps that lie beneath other GIS layers and provide real-life perspectives of terrain and surroundings that are not available in the vector GIS layers. Typical vector data do not show vegetation. The vector layers can show the manhole location but may not include the vegetation hiding the manhole. High-resolution orthophotos with submeter accuracy can guide the public-works crews directly to a manhole hidden behind bushes. Know- ing the land-cover characteristics before leaving for an emergency repair of a broken water main will allow the crews to bring the appropriate tools and equip- ment. Knowing whether the job will be on a busy intersection or in somebody’s backyard will determine the kind of equipment, material, and personnel required for the job. Figure 8.1 shows a water system map overlayed on a digital orthophoto base map with an accuracy of ±1.25 ft. Typical digital orthophotos cost $800 to $1600 per mi 2 . Planimetric Maps Like digital orthophotos, planimetric base maps are also created from aerial photographs. However, instead of scanning the aerial photos, the features are digi- tized from them. Thus, whereas digital orthophotographs are raster files, planimetric maps are vector files. Planimetric maps generally show building footprints, pavement edges, railroads, and hydrography. Parcels digitized from existing maps are often added to the mix. Figure 8.2 shows a sample planimetric map for the Borough of Munhall, Pennsylvania, extracted from the Allegheny County land base. The bor- ough’s sewer lines and manholes are overlayed on the planimetric base map. The cost of planimetric maps depends on the level of detail and, therefore, varies signif- icantly from project to project. The typical cost range is $2,500 to $10,000 per mi 2 . 2097_C008.fm Page 143 Monday, December 6, 2004 6:04 PM Copyright © 2005 by Taylor & Francis Small-Scale Maps Small or rural systems often use small-scale street maps or topographic maps as base maps. Street maps can be created by digitizing the existing maps, obtained from a government agency (e.g., U.S. Census Bureau, USGS, or state department of transportation) or purchased from commercial data vendors. In the U.S., 1:24,000-scale raster topographic map layers called digital raster graphics (DRG) are provided by USGS. Shamsi (2002) provides detailed information about the sources of small-scale maps. Users should be aware of the scale, resolution, accu- racy, quality, and intended use of small-scale base maps before using them. Most maps at scales of 1:100,000 and smaller are not detailed enough to be used as site maps or engineering drawings, but they can be used for preliminary studies and planning projects. Figure 8.3 shows interceptor sewers and pumping stations for the Kiski Valley Water Pollution Control Authority in Pennsylvania, on a base map Figure 8.1 A water distribution system overlayed on a digital orthophoto base map. 2097_C008.fm Page 144 Monday, December 6, 2004 6:04 PM Copyright © 2005 by Taylor & Francis of streets. The 1:100,000-scale solid roads are from the U.S. Census Bureau’s 1990 Topologically Integrated Geographic Encoding and Referencing System (TIGER) data. The 1:24,000-scale dashed roads are from the Pennsylvania Department of Transportation. Unlike double-line pavement edges shown on the planimetric maps, these road layers show the single-line street center lines. The difference in the position of the roads in the two layers can be attributed to the resolution, scale, and accuracy of the two layers. ADVANTAGES OF GIS MAPS The most challenging part of a GIS application project is to obtain the right kind of maps in the right format at the right time. Therefore, maps are the most important component of a GIS. Without maps, you simply have a computer program, not a GIS. Figure 8.2 A sewer system overlayed on a planimetric base map. 2097_C008.fm Page 145 Monday, December 6, 2004 6:04 PM Copyright © 2005 by Taylor & Francis In many water and wastewater systems, there is a backlog of revisions that are not shown on the maps and the critical information is recorded only in the memories of employees. However, there is no longer any excuse to procrastinate because GIS- based mapping is easy and affordable. In the past, users have selected computer-aided drafting (CAD) and automated mapping and facilities management (AM/FM) systems to map their water and sewer systems, CAD being the most common method. Although a map printed from CAD or AM/FM might look like a GIS map on paper, it does not have the intelligence of a GIS map. GIS maps are intelligent because they have attributes and topology. Most conventional CAD maps do not have attributes; they simply print data as labels or annotations. For example, the mapmaker must manually write the pipe diameter next to a pipe or must manually change the pipe color or line type to create a legend for pipe size. This is a very cumbersome process. On the other hand, GIS stores the attributes in a database and links them to each feature on the map. This capability allows automatic creation of labels and legends at the click of a mouse button. In GIS, map labels and legends are changed automatically if an attribute changes. In CAD, the mapmaker must manually delete the old label and retype the new label. Only a GIS map knows the spatial relationships among its features. Called topology , this capability makes the GIS maps intelligent. For example, a GIS map is intelligent enough to know which watershed is adjacent to which. Although both the CAD and AM/FM offer map layers to store different types of objects, only a GIS map has the capability to relate data across layers. The spatial relations among layers allow spatial Figure 8.3 A sewer system overlayed on a streets base map. 2097_C008.fm Page 146 Monday, December 6, 2004 6:04 PM Copyright © 2005 by Taylor & Francis [...]... Information and Technology Association (GITA) USGS GIS Maps for Water Resources USEPA GIS Information Copyright © 2005 by Taylor & Francis www.gita.org water.usgs.gov/maps.html www.epa.gov/epahome /gis. htm 2097_C0 08. fm Page 160 Monday, December 6, 2004 6:04 PM CHAPTER SUMMARY GIS provides powerful capabilities for creating cost-effective and intelligent maps for water, wastewater, and stormwater systems. .. of different organizations and agencies For instance, it is claimed that 80 0 worker-years would be needed to convert England’s sewer system to digital format, including field verification work (Bernhardsen, 1999) Field data collection using mobile GIS and GPS technology that employs handheld devices and tablet PCs is becoming common for collecting attributes (Chapter 7, Mobile GIS) The question of quantity... correct style, orientation, scale, size, and format is a time-consuming process Plotter speed and network speed are often the main bottlenecks Templates for symbology, legends, and layouts should be developed inhouse or obtained from other organizations to expedite map production Some vendors package templates and symbol libraries specific to water, wastewater, and stormwater system with their software... horizontal accuracy of 3 ft • Use as-built drawings, GPS, or traditional land surveying methods to determine structure locations • Use State Plane Pennsylvania South NAD83 projection system for coordinates and NAVD 88 datum for elevations • Manhole inverts and rim elevations to a minimum vertical accuracy of 0.10 ft for overflow structures, sewers with greater than 10-in diameters, and other critical sewers •... editing tool increase the speed, efficiency, and accuracy of data conversion for water, wastewater, and stormwater systems They reduce data-entry errors and can even be used to validate the input data If desired, the applications can be modified to meet the project-specific mapping requirements Copyright © 2005 by Taylor & Francis 2097_C0 08. fm Page 153 Monday, December 6, 2004 6:04 PM Data Processing Data processing... the base map GIS data are stored in various file formats The number of data formats has increased exponentially with the growth in the GIS industry (Goodchild, 2002) According to some estimates, there might be more than 80 proprietary geographic data formats (Lowe, 2002b) Why are there so many geographic data formats? One reason is that a single format is not appropriate for all applications For example,... a nondigital form This information was manually entered into the GIS database Map features and attribute information were gleaned from a variety of sources Table 8. 4 lists recommended sources for specific map features and attributes After reviewing the Borough’s available maps and considering the mapping requirements of the priority applications, the use of digitization was recommended for data conversion... format cannot support both fast rendering in a command and control system and sophisticated hydraulic analysis in a water distribution system Different data formats have evolved in response to diverse user requirements GIS software cannot read all the data formats simply because there are so many formats Disparate data formats should be converted to one of the formats compatible with a particular GIS. .. equally critical GIS component The most difficult job in creating a GIS is the enormous effort required to enter the large amount of data and to ensure both its accuracy and proper maintenance (Walski and Male, 2000) Conventional file-based GIS databases (e.g., ArcInfo 7.x) store geographic data in vendor-specific proprietary format files Attribute Copyright © 2005 by Taylor & Francis 2097_C0 08. fm Page 155... extensions to standard GIS packages to supplement specific mapping needs For example, ESRI’s ArcFM water data model supports water and sewer system mapping in a geodatabase Chester Engineers’ (Pittsburgh, Pennsylvania) water editing tool is an application that allows user-friendly mapping of water and sewer systems for ESRI software users It eliminates attribute entry and editing via command prompts or . CHAPTER 8 Mapping GIS provides powerful and cost-effective tools for creating intelligent maps for water, wastewater, and stormwater systems. A sewer system map created by GIS (Borough. conversion applications like the water editing tool increase the speed, efficiency, and accuracy of data conversion for water, wastewater, and stormwater systems. They reduce data-entry errors and can. create GIS maps for water, wastewater, and stormwater systems. MAJOR TOPICS • Mapping basics • Map types • Advantages of GIS maps • GIS mapping steps • Mapping case studies LIST OF CHAPTER ACRONYMS