Digital Mapping Techniques '97
U.S. Geological Survey Open-File Report 97-269

Topological and Thematic Layering of Geological Map Information: Improving Efficiency of Digital Data Capture and Management

By David Viljoen

Geoscience Integration Section

Continental Geoscience Division

Geological Survey of Canada

615 Booth St.

Ottawa, Ontario, Canada, K1A 0E9

Telephone: (613) 995-1207

Fax: (613) 995-9273

e-mail: viljoen@gis.nrcan.gc.ca

Abstract

Geological maps are typically complicated documents containing a variety of information that is both spatial and descriptive, such as lithology, structure, mineralogy, topography, and hydrography. This information can be digitally captured, stored, manipulated, and analyzed in a Geographic Information System (GIS). It can be input and stored as points for features at a specific location (e.g. structural observations), lines (chains) for linear features (e.g. faults), and polygons for areal features (e.g. lithological units).

One of the keys to efficient capture and management of geological map data is understanding how polygons are stored in a GIS. Topological and thematic layering are powerful methods that can be utilized to dramatically decrease the amount of time spent capturing, manipulating, analyzing, and managing geological map information. Topological layering involves separating the lines and polygon labels (area points) required to "build" polygons in a GIS from other line and point features. Thematic layering entails separating spatial information by theme (e.g. topography and geology). It is very useful to separate polygon layers thematically. The most dramatic example of this is separating geological (e.g. lithological units) and hydrographic (e.g. lakes) polygons.

There are presently two dominant pathways for digital data capture: digitizing tablet and scan/vectorize. The layering concepts introduced in this paper can be utilized regardless of the data capture pathway and will improve the efficiency of data capture.

This paper will review how polygons are created and stored in a GIS and compare the "single-layer" map oriented approach and "multi-layer" GIS approach to digitally capturing and managing geological map information. It will highlight the advantages of the "multi-layer" approach in terms of data management, analysis, and visualization.

Introduction

There are many references that describe how GIS store information (Aronoff, 1993; Burrough, 1986), how GIS has been used for addressing geological problems (Van Driel, 1989), and issues related to digital data capture (Wright et al, 1990) but few discuss how to manage geological map information with a GIS. The goal of this paper is to propose a methodology for digitally capturing and managing geological map information more efficiently and effectively.

Geoscientific information can be represented by utilizing four spatial data models: points, lines, polygons, and rasters. A spatial data model is a conceptual model of real-world features and has a graphical representation on maps and a digital representation in a CAD or GIS. Map features are defined by two types of information:

  1. Spatial information defining locations of point features, shapes of linear or polygonal features, and reference coordinate and pixel size for raster (image) information
  2. Descriptive information describing what the point, line, and polygonal features or pixel values represent.

For example, a capital city on a small scale map may be graphically represented by a point positioned somewhere on a map (spatial information) and a red star might be used to differentiate capital cities from others on the map. In a GIS, this red star would be special symbolization associated with the descriptive information for that point.

This paper will review how areal features (polygons) such as lithological units are created and stored in a GIS. It will then demonstrate how effective layering of information can dramatically improve the efficiency of geological map data capture, management, analysis, and visualization.

Polygon topology re-visited

From a GIS point of view, topology defines the relationships between spatial objects (e.g. an area point/polygon label and its enclosing lines as shown in Figure 1) that are unaltered through common geometric transformations. The encoding of topological relationships is essential for capturing and managing polygons in a GIS.


Fig 1.

Figure 1. Representation of areal (polygonal) features in a GIS.


Figure 1 shows the map and GIS view of two adjacent polygons with different lithologies as well as the data required to store these polygons in a GIS. There are three types of data required to store polygons in a GIS: spatial, topological, and attribute. The spatial data describes the shapes of the lines (chains) that are used to define the outline of the polygon and the location of the polygon labels (area points). The topological data describe the relationships between the chains, nodes, and area points. The area point topological data describes the connectivity of arcs used to define the polygons. The chain topological data describe the left/right polygon relationships between the chains and the polygons. The attribute data describe which lithology the polygons represent.

Single-layer approach to data capture and management

Given a simple geological map, as in Figure 2, there might be an inclination to digitize with a map oriented, single-layer approach. That is, digitize the lines and attribute the polygons such that the map view can be easily generated by drawing and symbolizing one data layer. While this may seem to be the simplest and most efficient method, there are many negative implications that will be reviewed later in the Multi-layer vs. Single-layer Approach section that clearly demonstrate that this is not the case.


Fig. 2

Figure 2. Single-layer approach to capturing and storing geological map information.


Multi-layer approach to data capture and management

Figure 3 shows a multi-layer approach to capturing and managing the same map described in the previous section. In this case the data has been layered topologically and thematically. Initially, this method may seem to be more complicated to implement and requires geological interpretations where there is no outcrop (e.g. under water). However, there are significant advantages to this methodology that will be reviewed in Multi-layer vs. Single-layer Approach section.


Fig. 3

Figure 3. Multi-layer approach to capturing and storing geological map information.


Topological Layering

Topological layering involves separating the lines and polygon labels (area points) required to build polygons in a GIS from other line and point features. In this example, the lines representing lithological contacts and fault-contacts are separated from the faults. There are portions of faults that are duplicated in both the geology and fault layer. This does not necessarily have to be the case but it does simplify using the fault layer separately for other applications such as a fault overlay on a satellite image. Data management can be dramatically simplified by separating faults and lithological contacts as shown in Figure 4. In addition, a dangling chain (i.e. a line with one or both nodes not shared with another line or lines) is perceived to be a topological error by GIS. This means that including all faults in a lithological contact layer will display many dangling chains that are not necessarily topological errors. This can dramatically slow the process of finding "true" dangling chains and building polygon topology.


Fig. 4a Fig. 4b

Figure 4. Example of additional complexity caused by including all faults in same data set as lithological contacts. A, Single-layer approach -- all faults and contacts on lithological polygon layer. Each P8 label represents a polygon label. There are many more polygons to manage than the single polygon shown in (b). B, Multi-layer approach -- contacts and fault-contacts on lithological polygon layer. This is a single polygon extracted from a digital geological map. All of the faults shown are managed as a separate dataset.


In general, it is desirable to minimize the number of points, lines, and polygons required to digitally capture and manage a geological map. Topological layering is a major step in this direction.

Thematic Layering

Thematic layering entails separating spatial information by theme (e.g. topography and geology). In the example shown in Figure 3, it was wise of the GIS specialist to recognize that there were two polygon themes in this map: one geological (i.e. lithological units) and the other hydrographic (i.e. lakes). Separating these two themes into individual data sets has several advantages that will be outlined in Multi-layer vs. Single-layer Approach section.

Multi-layer vs. Single-layer approach

The multi-layer approach which entails topological and thematic layering of geological map information reduces the number of features (points, lines, polygons) that the GIS specialist needs to manage, simplifies updating themes, and enhances the potential for map generalization, analysis, and visualization.

Data Management Efficiency

Table 1 clearly demonstrates how the multi-layer approach, highlighted in Figure 3, results in significantly fewer features to manage than the single-layer approach shown in Figure 2. In this example, there are approximately 60% fewer features to manage. This can translate into much more than a 60% savings in terms of capturing and managing this particular map.

Table 1. Comparison of the number of features to manage with the single-layer and multi-layer approaches.

Feature

Single-layer (All)

Multi-layer
(geology+faults+lakes)

% fewer features

Area points

29

8 + 0 + 4 = 12

59%

Chains

58

15 + 2 + 8 = 25

57%

Layer Updates Simplified

It is quite common for a digital geological map to be used as a starting point for a smaller scale, more generalized geology map. This often requires a change in the hydrographic layer used as shown in Figure 5. Table 2 compares the editing and polygon topology building operations that are required to achieve the integration of a smaller scale hydrographic layer. With the single-layer approach, the change in hydrology involves removal of all shoreline chains and extraneous area points, adding new shoreline chains and area points, and re-building polygon topology. With the multi-layered approach, the change is simply accomplished by plotting the new hydrology layer over the existing geology.


Fig. 5

Figure 5. Updating a hydrographic layer in a geological map using single-layer vs. multi-layer approach.


Table 2. Summary of changes to a single-layer vs. multi-layer data set required by a change to a smaller scale, more generalized hydrographic layer

Spatial feature

Operation

Number of operations

Single-layer

Multi-layer

Area points

Delete

5

0

 

Add

1

0

Chains

Delete

14

0

 

Add

6

0

Polygons

Rebuild topology

Yes

No

Impacts on analysis

The multi-layered approach can also have dramatic effects on the efficiency of spatial data analysis. For example, in Figure 6, the user may wish to know the total area of lithological unit V1a. If the data were managed using a single-layer approach, with hydrology and geology on one layer, then the total area of unit V1a will be underestimated since the location of the underwater fault is not used to bound polygon. It is virtually impossible to accurately determine the area of geological units from a single-layered data set. Using the multi-layer approach, however, a much better estimation of area can be achieved. If the geology is not well known under lakes, then the lithology can be coded in the descriptive information to reflect this (e.g. N/A). With the multi-layer approach, it will be possible to compute statistics on the basis of known geology. Other types of analysis could be impacted negatively using the single layer approach such as a point-in-polygon analysis (e.g. lithology for a gravity measurement on a lake) or a polygon-on-raster analysis (e.g. average total field value for a particular lithology).


Fig. 6

Figure 6. Area estimation or overlay analysis using single-layer vs. multi-layer approach.


Impacts on Visualization

The multi-layer approach provides for more flexible visualization of geological map information as shown in Figure 7. In this example, there are three polygonal themes: bedrock geology, Quaternary cover, and lakes. This provides the option of displaying any one of the three polygonal themes separately or the opportunity of overlaying the lakes and Quaternary layers with patterned fills such that the underlying bedrock geology is still visible. With patterned fill overlays, a level of confidence in the geological interpretation can be inferred by whether the geology is overlain by Quaternary sediments or water. With a single layer approach, none of these visualizations would be possible.


Fig. 7

Figure 7. The multi-layer approach provides for more flexible visualization of geological map information. (a) geology as portrayed in original map; (b) same as (a) with Quaternary geology layer removed; (c) same as (b) with lakes removed; (d) geology with patterned fills for Quaternary and lake layers.


Digital geological maps are now often integrated with other data (e.g. shaded relief topography or total field magnetics) to highlight interrelationships. In the case of integration with total field magnetics, water and many sediments are magnetically transparent. This means that an integration of geology with magnetics would be best visualized if the bedrock geology was interpreted with the geologist treating lakes and Quaternary sediments as overlays that might be removed for analysis or visualization.

The multi-layer approach: A few simple guidelines

The steps to follow to implement the multi-layer approach are:

  1. Identify the polygon layers in your map (topological layering). For example, lakes, lithological units, Quaternary units, metamorphic zones would be independent polygon layers.
  2. Identify only the layers of lines required to define the polygon layers in step 1. Shorelines for lakes and islands would only be incorporated to define lake polygons. Lithological contacts should not, for example, terminate at a lake shore line creating a dependency between one polygon layer and another. The attribution of the lines would enable the user to distinguish the various "layers" used to define the polygons. Lithological polygons, for example, might be bounded by approximate, defined, assumed, and fault contacts.
  3. Digitize and manage each polygon layer separately from each other and especially from other "non-topological" data (lines and points that are not required for creating polygons). For example, if a fault is not a boundary between two different lithologies, then it should not included with lithological contacts.

The goal is to minimize the number of objects the GIS user has to capture and manage which also simplifies updating themes, and enhances the potential for map generalization, analysis, and visualization.

Conclusions

An understanding of how polygons are created and stored in a GIS is important in appreciating the benefits of proper data capture and management. The benefits of multi- versus single-layer approach to data management are evident from the smaller number of spatial objects that are required in the multi-layer design, the ease with which changes to individual themes can be made, and positive impacts on analysis and visualization. To implement the multi-layer approach, map features should be layered topologically and thematically. For example, thematic layering ensures that hydrographic features are not on the same layer as geological features. Topological layering ensures that only those features required to define polygons are stored together.

References

Aronoff, Stan (1993) Geographic Information Systems: A Management Perspective, WDL Publications.

Burrough, P.A. (1986) Principles of Geographic Information Systems for Land Resources Assessment, Oxford University Press.

Van Driel, J. Nicholas & Davis, John C. (1989) Digital Geologic and Geographic Information Systems. 28th International Geological Congress, Washington D.C, Short Course in Geology, Volume 10.

Wright, Bruce E. & Stewart, David B. (1990) Digitization of a Geologic Map for the Quebec-Maine-Gulf of Maine Global Geoscience Transect, U.S. Geological Survey Circular 1041, U.S. Government Printing Office, Denver, CO.



Home | Contents | Next


<https://pubs.usgs.gov/openfile/of97-269/viljoen.html>
Maintained by Dave Soller
Last updated 10.06.97