International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B4. Istanbul 2004 
  
into the data and using it, the matter which may lead to 
incorrectness of decisions made depending on it, as much as 
to inaccuracies of designs, for instance if a planner is to set a 
sloping of a sewage network to an accuracy of half a meter 
based on a terrain model which may reach an error of one 
meter at some areas, due to the lack of knowledge concerning 
the accuracy and validity of the DTM an error may occur, the 
matter which necessitate planners to have a full information 
concerning the goodness of the data used . 
Some surfaces change rapidly and some times drastically due 
to artificial urban development, as much as due to tectonic 
natural phenomenon, which means an up to date should be 
following to keep its validity, again this fact can not be 
realized by simply using the data. 
As an example of showing the importance of rules defining 
the relationships of the surface data, is that a contour can not 
cross an other, a blunder which is still being created by 
interpolating software while smoothing lines or fitting 
surfaces, other than that is the density of contours in steeper 
areas or the discreetness of the contours among others. 
The paper discuses these facts and more for the different 
surface structures, showing the advantages and disadvantages 
of the different structures, and the rules should be 
encountered in the geodatabases concentrating the quality of 
the surfaced data and its effect on the result. 
2. METHODS OF PRESENTATION &DATA 
QUALITY 
For expressing the quality of the data, some information 
should be tabulated in such a manner that it can be viewed 
with the data itself expressing its accuracy, even if the 
accuracy of the data is changing from one part to the other, 
the stored accuracy information in the geodatabase should be 
able to discriminate between areas of different accuracies 
using specific annotations. 
With a similar way as storing the topology and networks in 
the geodatabase accuracy can be a part of the geodatabase or 
when storing spot heights or contours the accuracy can be 
added to it as a property of the layer or the class. 
With out keeping records of the data accuracy and 
expressing it with the data, the operator has to extract the 
knowledge of the data accuracy from reports or records, and 
the end users may use the data in a wrong manner effecting 
the quality of designs based upon this data. 
2-1Quality of Surfaced data 
Generally speaking the quality of all the surfaced 
topographic data depend on the following factors: 
Source of the data, 
Method of Data Collection, 
Density of the Data, 
Interpolation and processing methods, 
Digital Format or Structure of the Data, 
Source of The Data: The source plays a major roll in the 
quality aspects, different data requisition methods are 
available choosing among which results in a certain quality 
measures, since the source can be using field survey 
measurements for instance directly from the ground or can be 
from photogrammetric stereo image pairs where the variation 
of the scale means a variation in quality as much as the 
overlapping areas between the images. 
69 
Other sources can be satellite stereo images, also radar data 
became an important source or data collected using laser 
beam technology, an other source can also be old and 
existing digitized maps or stored data. 
Thus the quality measures begin at this point, where 
propagation and accumulation of errors begin, in case of the 
data collected using field surveying measurements, starting 
by the quality of the existing height control network, its 
reference, its projection, its adjustment and its density, 
followed by the type of instruments used, the methods of data 
collection, certainly a very important factor is the experience 
of the surveyors in collecting such types of data, and finally 
the processing of the data, altogether with the interpolation 
methods, and the representing structure. 
Method of Data Collection: Talking about the effect of the 
data collection methods on the quality of surfaced data, we 
are to supply an example of using the most usual methods of 
data collection, the Regular networks and irregular scattered, 
If the data is collected directly from the field taking into 
consideration the steepness would provide closer surface to 
the reality than collecting the data using a regular spaced 
network, the type of instrument used such as plane table or 
using a total station or GPS would certainly effect the quality 
of the data. On the other hand if the data is collected using 
photogrammetric workstations, then factors such as the 
length of the base with depend on the overlapping areas, the 
scale of the photography, results of the triangulation and 
adjustment, manual data collection in urban areas, and 
automatic data collection would certainly effect the quality 
of the surface. 
Density of the Data: As matter of fact the density of the data 
should have a limit that complies with the final data format 
and accuracy to be reached, flat areas demand less density 
than others, designing the density should be economically 
worthy. Raster data for example has a regular and equal 
density every where, which means that flat areas are accurate 
while hilly areas may lack the required accuracy. 
For solving this particular problem of data to be collected 
using field survey or manually from stereo pairs should be 
designed with a certain surface modeling factor, this 
modeling factor will determine the number of pints to be 
collected within an area. Because the surface modeling factor 
has an influence on the time and price of data, we should 
consider total quality measures for designing such a factor, 
some of these measures arc: 
» what kind of implementations the data shall be 
created for ? 
> How often the data is updated? 
» Is it economically worth it, to go for such an 
accuracy? 
» Is there a probability of using this data for other 
aspects in the future? 
> What shall be the structure of the data, and what 
kind of limitations may the structure nature cause 
to the data. 
Considering these factors and others, a certain surface 
modeling factor is selected such as, collect data if the height 
difference tends to a certain limits, in other words height 
difference between successive mass points should be limited 
to a certain measure, when ever it is possible (Buildings or 
such obstacles may intrude such a process). 
This can be one modeling rule which regulates the quality of 
the data collected or measured, another rule which if 
combined with the surface modeling factor rule can be the 
maximum horizontal distance between the successive mass 
i 
i 
4 
| 
d 
a 
t 
E 
A