Pitout, Cedric 
  
2 DATA MODEL 
2.1 Context 
First of all, data are divided in two classes. The first one is geometrical data generated by the GIS proper (points, lines, 
polygons with their coordinates), whereas a DBMS (DataBase Management System) manages environmental data. This 
paragraph focuses only on the last type of data. This approach can be applied no matter what GIS and DBMS software 
is used. For this work, ESRI ArcView3.0a© GIS and Microsoft ACCESS970 have been used. 
Before to define a data model, it is necessary to synthesize the available information. The interest is to store a large 
amount of qualitative and quantitative parameters to describe the site. By another way, it does not seem advisable to 
consider an exhaustive list of pollutants at the risk of weigh down the database. For this reason only parameters 
specifics to our case-studies are indexed. Nevertheless, it is important to think of the universality of the GIS since each 
industrial site is particular with very heterogeneous data sets. The database model must also allow later append of new 
fields without important modification of the initial structure. 
2.2 Environmental data and data format 
Data result essentially from piezometers, cores, pits, cuttings, packings. For the two last examples, some problems are 
highlighted like reworked soil samples and the geo-referencing of these samples. Whatever the data source, 
investigations are often the same, nevertheless separated databases are built according to the kind of data : soil samples 
and water samples in our case. To resume, the table 1 proposes the environmental data collating. For the chemical 
databases, the last column indicates the number of pollutants for each medium. 
Some databases have qualitative fields, i.e. the data format is not numeric but text. In this case, data information 
extraction with Syntax Query Language (SQL) could be problematic. Thus, qualitative data are integrated only for 
information, except if a data coding lexicon is created. 
Geologi file 
or element 
Alkaline / earth alkali 
Metal / metalloids 
Mineral 
20 (with 10 in soil 
3 
16 
10 
2 
7 in water. 
18 (with 9 in water and 6 in soil 
Hal 
C 
ical-chemical 
Groundwater level 
» |»€ »€ |»€ |»€ |»€ |»€ |»€ 
ic characteristics 
Piezometer ipment 
» |»€ |»€ |»€ |»€ |» |»€ |»€ |»€ |»€ |»€ 
  
Table 1. Environmental databases 
A major point to take into account is the non-redundancy of records. To avoid this, fields of the table 2 are indexed, and 
used for the common model of all databases except saturation level, pump test and piezometer equipment. For each 
database, the common model is completed by specific parameters or pollutants. 
Nom Stri 10 Name of the object 
Date Date [mm/ Date 
De NGF Real Absolute top elevation 
A NGF Real Absolute bottom elevation 
De Real Relative top elevation 
A Real Relative bottom elevation 
Code Ech String [30 Name of the s e 
Table 2. Common databases model 
  
  
1156 International Archives of Photogrammetry and Remote Sensing. Vol. XXXIII, Part B7. Amsterdam 2000.