) vector 
) be the 
as well 
regular 
al basic 
ay, sum 
n, on a 
ands by 
rformed 
he new 
> for all 
lanning 
ld be a 
r GIS- 
Work is 
y Dana 
of GIS 
mented 
'd map 
cessing 
ıl and 
which 
. Focal 
is of a 
L 8 
Visions 
and a theme map layers. Incremental functions are for 
computing e.g. slopes and other characteristics based on the 
difference between the values of adjacent pixels. 
Tomlin describes a set of functions and shows also examples of 
*programming" procedures by using these functions. Each 
analysis application is a combination of functions. This 
hierarchy of analysis functions has been discussed more in 
(Artimo, 1996 and 1995). 
The most useful grid analysis functions are perhaps the most 
simple ones. With more complicated functions the 
understanding of the result of the analysis is more difficult. 
When the question is about a land use planner the easyness of 
the planning tool is most important. A slight criticism against 
the implementations of Tomlins map algebra, the GRID- 
modules of ArcINFO and Intergraph MGE is perhaps well- 
founded: the implementations are not easy enough for a non- 
professional user. 
3. LAND USE PLANNING AS AN APPLICATION FIELD 
FOR GIS 
3.1 Land use planning process 
As mentioned earlier in this article we deal with “medium- 
scale” land use planning, it means general/master planning in 
municipalities and regional planning in smaller scales. This 
means scales which vary from 1:10000 to 1:50000. 
The description of the planning process is here given in a very 
general form. The definition of the following stages is made on 
the basis of practical experience and is not a result of deep 
analysis on planning process. 
In land use planning the following stages can be identified: 
-source data collection, inventory, 
-definition of the planning goals and restrictions, 
-analysis of the existing situation, 
-creation of the alternative planning ideas and making 
sketches about alternative plans, 
-evaluation of the alternative plans, 
-making presentation documents for the decision makers, 
-decison of the best draft and finalizing the plan, 
-making documents and illustrations for both professionals and 
inhabitants, 
-updating the plan. 
In the following we deal with the possibilities of applying 
raster-based data processing, especially analysis methods in 
some of the previous stages of the planning process. 
3.1.1 Source data collection, inventory 
Perhaps the biggest development from the point of view of land 
use planning among GIS-technology is the progress in scanning 
techniques. Source data available is no more limited to 
manually digitized vector format maps and data bases but any 
graphical map can be scanned, converted into right coordinate 
system and thus utilized in digital form. Manually drafted 
101 
maps in field inventories, historical maps, plans etc, all 
graphical documents are available in digital raster format. 
On the other hand lots of existing maps - topographical and 
soil maps for example - have been scanned and are delivered in 
CD-ROMs, in raster format. 
If we compare the easyness of digitizing into raster format with 
the manual vector digitizing or vectorizing of scanned maps the 
advantages are obvious. 
In addition to the traditional source material digital satellite 
images and aerial photographs are available and can be used as 
source data when 2d raster format is used. In “traditional” 
planning GIS-environment where all active data were in vector 
format both digital aerial photographs as well as satellite 
images could only be used as passive background maps. When 
processing and analysis is in raster format also these data sets 
can be used more effectively. 
3.1.2 Analysis of the existing situation 
When source data have been collected and the planning goals 
and resrictions are defined the planner can concentrate in 
analysis of the existing situation. Analysis means that source 
data are processed into more understandable form. In this stage 
the map algebra based GRID -tools of GIS -software are most 
useful. Simple map overlay analysis can be made for searching 
suitable areas for building. Digital elevation models can be 
used for analysing for example suitable slope and sight 
conditions. In landscape analysis important areas which should 
be protected can be identified. - All these procedures would be 
most difficult and complicated if data were in vector format. 
3.1.3 Creation of alternative plans, evaluation of plans 
Planning is always a human activity. No computer assisted 
application can replace human planner. Knowledge based and 
intelligent systems can take part in planning process by storing 
human knowledge and applying it in planning situations. 
Certain limited problems can be solved also by optimizing. 
Raster processing is also available in this kind of problem 
solving. Shortest paths along or outside networks can be solved 
by GRID -analyst modules as well as proximal areas. However 
planning itself is mainly ruled by human planners. Tomlin 
makes a distinction between descriptive (answering questions) 
and prescriptive (solving problems) analysis methods. In 
practise descriptive methods are perhaps more simple and thus 
more useful than the prescriptive methods. 
3.1.4 Making presentation documents for the decision makers 
It was previously mentioned that scanning technology is one of 
the greatest advantages in developing planning applications. 
Raster colour output technology is the second great 
development. There is not very much use of advanced 
processing of data if the results are not presented in an 
attractive way for the decision makers. In the days of pen 
plotters the presentation of results was the worst bottle-neck. 
High quality colour outputs produced by ink jet, electrostatic or 
laser plotters give the full benefit of computer assisted 
processing for the users. 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B4. Vienna 1996