Chapter 6

WORKING WITH VECTOR DATA

The vector data model is used for representation of geographic phenomena as geometric objects composed of points, lines and areas. Lines are used for roads, railroads, streams, or utility networks, while areas can represent soil types, land use categories, lakes, or zoning in urban areas. Vector data are stored using their coordinates. In GRASS, the vector data model includes the description of topology.

At the time of writing this book, the vector data support is undergoing significant changes. This chapter reflects the GRASS 5.3 vector capabilities that are somewhat restricted. Improved tools with 3D vectors and direct DBMS support are under development for the GRASS 5.7 version.

For the description and import of vector data, please refer to Section 4.2.2.

6.1. DIGITIZING VECTOR DATA

A paper map can be converted to digital form by manual digitizing. In general, there are two ways to digitize a map:

using a digitizing board or

digitizing heads-up (on screen).

In the first case, the map is placed on the digitizer board, which provides a special digitizing mouse. The corners are selected by a mouse click and their respective coordinates are entered using the keyboard. This process is called registering a map. Then the lines and points on the map are digitized using a mouse. The advantage of this method is that the user always sees the entire map. However, the high cost of the equipment and the possibility that

the map could be shifted during the digitizing, if it is not properly mounted, are significant disadvantages. Furthermore, the paper map must be free of distortions to prevent displacements.

On-screen digitizing requires a scanned and geocoded raster map that is displayed in the GRASS monitor. All features will be digitized using the mouse. It is not necessary to register such a map because it is already geocoded. The advantage of this method is the possibility to zoom in and thus achieve an improved accuracy. Apart from an access to a scanner, no additional equipment is needed. The major disadvantage is the more difficult orientation on the map.

The following section deals only with heads-up digitizing using mouse. The installation of digitizing boards is described at the GRASS Web site.

6.1.1 General principles for digitizing topological data

To explain the digitizer module, we consider an example of vectorizing features from a scanned topographic map. We assume that the map was scanned, accurately geocoded and imported into GRASS in raster format (see Section 4.1.4 on geocoding scanned maps). Although it may be possible to automate the vectorization of a raster map using r.line or r.poly (see Section 5.3.1), problems often arise from overlapping lines, dots, map signatures etc. and manual digitizing is necessary.

There are few general recommendations for digitizing map features which can minimize the potential accuracy problems. The recommendations are mostly based on the fact that to make the map readable some features are exaggerated compared to their their size at a given scale:

Line features should be digitized along their center-line, e.g. along the middle of a road. A line label point should be placed on the line;

Area features should be digitized by following the center-line of area boundary lines. An area label point should be placed in the center of the

area;

Point features should be digitized at the center of the object, e.g. a point in the center of a map symbol representing the point feature or at the reference point of such a symbol;

The points defining the line or polygon boundary should be selected at a density that is sufficient for preserving the geometry of the digitized features.

Rules for digitizing in topological GIS. When working in a topological GIS such as GRASS, following certain rules is recommended, in order to benefit from the topological features of the software. The following rules apply to the vector data (from GRASS 5 Programmers Tutorial, Neteler, 2000):

Arcs should not cross each other (i.e., arcs which would cross must be split at their intersection to form distinct arcs);

Arcs which share nodes must end at exactly the same points (i.e., must be snapped together using the snapping function of the digitizing module). This is particularly important since nodes are not explicitly represented in the arc file, but only implicitly as end points of arcs;

Common boundaries should appear only once (i.e., should not be double digitized);

Areas must be explicitly closed. This means that it must be possible to complete each area by following one or more area edges that are connected by common nodes, and that such tracings result in closed areas;

It is recommended that area features and linear features be placed in separate vector map layers. However, if area features and linear features must appear in one layer, common boundaries should be digitized only once. An area edge that is also a line (e.g., a road which is also a field boundary), should be digitized as an area edge (i.e., feature type A) to complete the area. The area feature should be labeled as an area (i.e., feature type A in the dig att file). Additionally, the common boundary arc itself (i.e., the area edge which is also a line) should be labeled as a line (i.e., feature type L in the dig att file) to identify it as a linear feature.

Now we explain the digitizing process in detail.

6.1.2 Digitizing in GRASS

Manual digitizing is done by running v.digit after a GRASS monitor has been started. First select the digitizing device. To use the mouse digitizer, choose none as digitizing tool. After specifying a new or existing vector map, you will get to the vector metadata page containing basic information and the map boundaries. In the case of using an existing map the boundary coordinates are set according to the map; in the case of a new map the current region boundary coordinates are defined. Besides the date, map creator and title, the map scale has to be entered. When working on a new map, we strongly recommend changing the default Maps scale from 1:1 to the correct scale of the map (e.g. 1:24,000). This value is important for the snapping tool as the snapping threshold is calculated from the map scale. A sample metadata screen may look like this: