Home > Mapping Methods
2.0 Parcel Mapping Methods
One of the important steps in creating digital parcel maps is conversion of traditional paper maps (or analog maps) into digital GIS data layers. Not only do the data have to be transferred into a digital form, but they also have to be vectorized so that GIS software would be able to distinguish between individual elements such as lines, points, and polygons. There are several ways to do this:
- Manual digitizing from analog maps;
- Scanning with further "heads-up" digitizing or automatic
vectorization;
- Coordinate Geometry conversion (COGO);
- Positioning property corners with GPS.
2.1 Manual Digitizing from Analog Maps
This is the most basic method of digitizing traditional paper maps. Manual digitizing is done by placing a paper map on a digitizing board and entering all the elements of the map into the data base by means of a sensitive digitizing puck. An operator enters data by placing the digitizing puck over the points on the map attached to the digitizing board and pressing different buttons on the puck, thus indicating the type of each point. A point can be either an individual element, or a part of a larger element such as a line or a polygon. Lines and polygons are still defined by a set of points entered by an operator and connected by lines (vectors). Therefore, the accuracy of the data depends of the accuracy of the location of the points. In the case with lines and polygons, the more points entered, the smoother the curves of the line will appear (see Figure 2.1).
Figure 2.1: Elements of digital GIS data layer.
Digitizers are capable of providing a very high degree of accuracy in defining
the location of each point (as high as 0.002" to 0.005"). Still, the
accuracy of the data depends to a great extent on the accuracy and skills of
the operator who manually enters the points by means of the digitizer. Usually,
after all the data have been entered, it is necessary to edit them in order to
get rid of operator's mistakes. Such mistakes are usually of two kinds - undershoots
and overshoots. They result from the difficulty for an operator to control the
precise location of a puck on the board and because the width of the lines on
the paper map is sometimes much wider than the accuracy of a digitizer. Those
mistakes can later be noticed when the data layer is displayed at a greater
scale on a computer monitor (see Figure 2.2).
Figure 2.2: Common mistakes of manual digitizing.
Accuracy of the digital data also depends on the accuracy of the analog map. If
lines on an analog map are misplaced, the operator will be able to correct them
only to a certain degree. It means that most of the analog map inaccuracies
will be very accurately reproduced on a digital map.
Manual digitizing has its own advantages and disadvantages. In GIS Guidelines
for Assessors prepared by Urban and Regional Information Systems Association
and the International Association of Assessing Officers (IAAO and URISA 1992,
37-38) the following advantages and disadvantages of manual digitizing are
listed:
|
Advantages: |
|
Disadvantages: |
The second method of parcel data conversion is scanning. In order to be scannable, a parcel map should be in a very good condition with minimum text on it. Hardcopy parcel maps are converted to digital form using a scanner. A scanner recognizes the black or white value of each pixel location on a map. Values are assigned across the whole map creating a raster image of the map (IOAA and URISA 1992, 39). Raster format is not very useful for many GIS applications. A raster image is simply a picture and can not be manipulated by the user in ways that tax maps are often used. In order to be able to edit or update the map it is necessary to convert the raster image into a vector one. There are two methods that could be used for this purpose: automatic vectorization and heads-up digitizing. Depending on the method to be used for vectorization, appropriate scanning resolution should be chosen. Most vectorization programs require that lines be at least three pixels wide in order to be reproduced as vectors. A typical hand drafted line in pencil is about 0.018". In order to automatically vectorize such lines, it is necessary to use at least 200 dpi (dots per inch) resolution, but 300 dpi would be better. A line drafted in ink or fine point pencil is about 0.012" and will require resolution about 400 dpi (Litton 1998, 1). If heads-up digitizing is to be used, scanning at high resolution is not necessary and a good quality map could be scanned at 150 dpi or even 75 dpi depending on the size of the source map and whether you want to zoom in for significant enlargement of the image.
2.2.1 Automatic
Vectorization
This method could be used only if a map is in a very good condition - all lines
are clear, have similar thickness, and clear intersections, and if there is no
annotation or unnecessary features on the map. If this is the case, then the
software can trace all the lines on the map and it would leave little editing
to perform afterwards. This is the easiest and quickest method of parcel data
conversion, but, unfortunately, parcel maps are usually not in pristine
condition and have all types of different features and text on them.
Vectorization software can not identify features (for example, it can not
distinguish a water body from a parcel line), and most types of vectorization
software can not distinguish text from lines. If a map cluttered by many
different features is vectorized, it will require a lot of editing afterwards
to delete all extra features and annotation. Time spent editing such a map
could easily overweigh time spent on map conversion done by some other method.
To minimize editing required after vectorization, it could be necessary to
manually redraw a parcel map on Mylar selecting only parcel lines and thus
excluding all extra features and annotation. This method has certain advantages
and disadvantages.
|
Advantages: |
|
Disadvantages: |
Figure 2.3: Advantages and disadvantages of vectorization
Source: Barnett, Goodwin 1994, Struck 1998.
2.2.2 Heads-up
Digitizing
This method is usually used when the analog map is in a very bad condition and
when the digitizing board is not available. Heads-up digitizing involves
manually tracing the lines on a computer screen over the top of the scanned
raster image. The raster image is used as a background image. Everything that
was said about on board digitizing could be applied to on screen digitizing as
well. There are several advantages of this method over on board digitizing:
1. Several people can do on-screen digitizing simultaneously because this work could be distributed between several computers, whereas it is very seldom that an organization would have several digitizing tables;
2. Heads-up digitizing is not as tiring as a board digitizing because an operator remains in a seated position, whereas on-board digitizing requires an operator to be in a standing position all the time.
Source: Barnett, Goodwin 1994; Struck 1998.
2.3 Coordinate Geometry (COGO)
This is a more accurate method of parcel data conversion to a digital form than any of the methods mentioned above. COGO uses a completely different approach to the process of data conversion compared to digitizing or scanning. The difference is that the analog map is not involved, but land surveys and deeds are used instead. Land surveyors record information about each individual parcel in terms of geometric distances and angles from control points (benchmarks). The COGO procedure involves entering into computer such information as a precise starting coordinate, the direction and distances of each side of the parcel. After that, the computer performs some calculations to create a geometrically accurate parcel polygon. Unfortunately, not all parcels have property boundary descriptions or the information could be wrong. Therefore, it is often necessary to go out in the field and collect the missing information or verify the existing documentation. COGO also has its own pluses and minuses.
|
Advantages: |
|
Disadvantages: |
Figure 2.4: Advantages and disadvantages of COGO
Source: Goodwin 1994; Gresavage and Thomas, Struck 1998
2.4 Positioning Property Corners with GPS
The property corners of many surveyed and platted parcels are usually marked with iron pipes or nails. If coordinates of these markers are somehow obtained and input into GIS software, it would only be necessary to connect the dots in order to get a very accurate parcel map. The resulting accuracy could surpass even that of COGO (Struck 1998, np). Today, with the rapid development of the Global Positioning System (GPS)*, it has become relatively easy to get coordinates for the property corners. In order to obtain coordinates, two different strategies could be used: static positioning or kinematic positioning (Struck 1998, np).
The choice of the strategy depends on the needs and available resources. Static method provides better accuracy (down to one centimeter) but it is more time consuming, as it requires longer standing time at each point in order to get an accurate reading. The kinematic method is less accurate (accuracy +/- 1 to 10 feet) but, on the other hand, it is less time consuming. Satellite readings are obtained almost immediately as you walk from one pipe to another (Struck 1998, np). Advantages and disadvantages of this method are summarized in the table below.
|
Advantages: |
|
Disadvantages: |
Figure 2.5: Advantages and disadvantages of using GPS
2.5 Transferring of Existing CAD Maps
CAD (Computer Aided Design) maps appeared in early 1980s (Nale, 3). CAD enabled users to provide much more precise drawings in comparison with hand drafted maps. CAD uses actual parcel dimensions to generate digital parcels. However, capabilities of CAD systems typically do not go beyond simple maps production - they do not support sophisticated queries or analysis. By 1988, topology and buffer generation, thematic mapping, network analysis, feasibility studies, forecasting, etc. became possible with the help of geographic information systems software (Nale, 3). In the past, many organizations have digitized their parcel maps using some drafting packages such as AutoCAD or MicroStation. Now, when they want to make a transition to GIS, the question is whether they can make use of the CAD drawings. Generally, CAD drawings do not comply with strict specifications that GIS coverages require. For example, lines that should be connected are very seldom "snapped" together, maps are rarely projected into a proper coordinate system. Nevertheless, if the maps are accurate and depict current information, they could be translated into GIS software. Translation is "the process in which graphic images are taken from one software package, formatted into another structure, and input into another software package" (Flodmark, p. 3). After the process of translation from CAD to a GIS format, it will be necessary to edit and project the coverage. Vector information (lines) usually translates well, but annotation and attributes very often get lost or altered in the process of translation (Struck 1998, np). Another reason why existing CAD drawings should not be disregarded is the fact that they are already in the digital form and they only need to be imported and edited.
2.6 Example of Analog and Digital Parcel Maps
|
Figure 2.6 Analog Parcel Map |
Figure 2.7 Digital Parcel Map |
Figure 2.6 illustrates scanned analog map. Apart from parcel lines, it also shows parcel numbers, acreage, parcels dimensions, street names, and other features. Figure 2.7 shows the same parcel map, transferred into digital form by means of digitizing. This map shows only parcel lines. All other parcels characteristics could be added to the attribute table associated with the map and optionally shown on the map.
Each GIS software has its own data format. In some cases this format could be used to transfer GIS data from one computer to another, in other cases it is necessary to create a special export format, and then import it into the target system. Common formats to transfer GIS data are summarized in the table below (Hohl 1998, 180-181).
|
Format |
Description |
|
AutoCAD drawing exchange format (DXF) |
A vector format that has become the de facto standard for transfer of data between different CAD systems. Often used to transfer geometry into and out of GIS systems. DXF is not well-suited for transferring attribute data. |
|
ARC/INFO export format (e00) |
A vector format intended to transfer data, including attributes, between different ESRI systems. Despite being a proprietary format, there are other GIS systems that can read or write e00 formats. |
|
ArcView shape file format |
Openly published, this vector format is available for use by other GIS vendors. It consists of three types of files: main files (SHP), index files (SHX), and dBASE tables (DBF) for storing attributes. |
|
MapInfo interchange format (MIF/MID) |
This format, also formally proprietary, has nonetheless been widely implemented in other GIS systems. MIF files store vector graphical information, while MID files store attribute data. |
|
MicroStation design file format (DGN) |
An openly documented vector format used by Bentley's MicroStation CAD software. MicroStation is the platform on which the Modular GIS Environment (MGE) and MicroStation Geographics GIS packages are built. The format doesn't store attribute data, but can store links to relational database records. MGE and Geographics also have export formats that transfer all files and database tables associated with a project. |
|
Digital line graph (DLG) format |
A vector format used by the U.S. Geological Survey (USGS), and to a lesser extent, other federal and state government agencies. DLG format only supports integer attribute information for spatial objects. |
|
TIGER/Line format |
The format used by the Census Bureau to distribute vector and attribute data from its Topologically Integrated Geographic Encoding and Referencing (TIGER) database. |
|
Spatial data transfer standard (SDTS) format |
A standard format used by the USGS and other federal agencies, designed to support all types of vector and raster spatial data, as well as attribute data. The Topological Vector Profile (TVP) and the Raster Profile are implementations of subsets of the SDTS. |
|
Tagged image file format (TIFF or TIF) |
A raster format frequently used for imagery, GeoTIFF is an extension to TIFF that includes georeferencing information. |
|
Joint photographic experts group (JPEG or JPG) format |
Another raster format commonly used for imagery. |
Figure 2.8: Data formats
2.8 Selection of the Best Conversion Method
The choice of the particular method of parcel data conversion depends on the availability and quality of the source maps, availability of the specialized software, and the requirements for quality and completeness of the data vs. the cost of the project. Before initiating the process of conversion, one must evaluate the availability and quality of existing maps. Some of the questions that have to be asked are (Donahue 1994, 8):
1. Were the maps originally drafted from deeds, surveys, and subdivision plats?
2. Do the original land base and parcel maps meet the GIS accuracy requirements (see Sec. 6.2. Accuracy)?
3. Have the maps been maintained on a regular basis?
4. Do the current maps visually edge match?
5. Are the current maps drafted at an acceptable scale?
If the answer to any of these questions is "no", it might be necessary to recompile the paper maps before starting the process of conversion or use a method that does not require the analog map.
Kevin Struck (1998) in his document also suggested a quick questionnaire for organizations in order to choose the best method:
1. Exactly how much money do you have to spend on the project?
|
Enough |
1 |
Little |
0 |
2. How many employees do you want to allocate to it?
|
Enough |
1 |
Few |
0 |
3. What level of mapping expertise does your staff have?
|
High |
1 |
Limited |
0 |
4. How accurate is your control monumentation and how well does it cover your project area?
|
> 50% |
1 |
< 50% |
0 |
5. Are a significant number (50% +) of your parcels platted or surveyed, and are these records legible, complete, and acceptable by modern surveying standards?
|
Yes |
1 |
No |
0 |
6. Are the parcels for your entire municipality mapped in a legible, complete,
and up-to-date hardcopy format?
|
No |
1 |
Yes |
0 |
7. Are all parcel legal descriptions readily available?
|
Yes |
1 |
No |
0 |
8. What is the quality of the legal descriptions? (Is there a small percentage
of descriptions with indefinite dimensions, vague point of beginnings, and
traverse closure errors?)
|
Good |
1 |
Poor |
0 |
9. Do you have large areas of irregular parcels such as French long lots and
Government Survey Lots?
|
No |
1 |
Yes |
0 |
10. What kind of equipment do you have? Digitizer or Scanner? COGO software?
|
COGO |
1 |
Digitizer or Scanner |
0 |
If you got 7-10 points, the best method for you is COGO or GPS, if you got 0-4
points, you should use digitizing or scanning, if you got 5-6 points, it is not
clear what method is the best for your organization. If the organization
doesn't have enough time, skilled personal or special equipment, the best thing
would be to higher a subcontractor to do the parcel conversion work. For more
information about outsourcing versus in-house operation see the Outsourcing
Section of the report.
| Next Section: Georeferencing | Home |