ul 2004 
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol XXXV. Part B4. Istanbul 2004 
  
Triangle side data structure 
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
society. TriaSideID | StartV | EndV |LeftTria | RightTria | UpSQuad | DownSQuad Cutflag | CutCoord 
| is that 
ibed by 
objects Edge data structure 
segment . ; : ; = : = ae 
QTPV EdgelD | StartV EndV | AdjSQuad | IsBelong | Attribute Cutflag CutCoord 
hich If the value of IsBelong is —1, the edge length will be is zero so it will be a tine point. 
' spatia 
xample, 
rface or Triangle data structure 
or side TrialD | Vertex[3] | TriaSide [3] | AdjTria[3] | PosiAttr | PosiQTPV | NegaAttr | NegaQTPV | IsBelong 
dy IS a 
point or 
ted by a Side quadrilateral data structure 
ted data 
tions of SQuadID | Vertex[4] [TriaSide[2]| Edge[2] |PosiQTPV|NegaQTPV IsBelong | PartType 
e above 
icted in 
QTPV data structure 
QTPVID | Vertex[6] | Tria[2] | SQuad[3] | AdjPrism[5] GeoObjType | IsBelong | CutFlag | CutMark [5] | PartitionCode 
TIN data structure 
TINID Type | TriaNum | Triangles| PosiGeoObj ! NegaGeoObj | Attibute IsBelong 
  
  
  
  
  
  
  
  
  
  
Stratigraphy or fault interface data structure 
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
   
   
  
| [StraID|'ry pe| MinBox TINNum|TINs PosiGeoObjNum|PosiGeoObjs| N NegaGeoObjNum | NegaGeoObjs | Attribute |IsBelong| 
| Geological objects (stratigraphy, ore deposit, fault, folder, etc.) data structure 
GeoObjID | GeoObjType | MinBox | StratIDs | QTPVs AdjGeoObjs | BoundTINObjs Attribute 
= 
3. MODELING OF SUBSURFACE OBJECTS re og | E A ‘ 
M. *. | S. ty Gnd triangulation 
n 3.1 Stratigraphy Modelling = ; 
[. Y Interfaces cross 
With the differences of modeling data, stratigraphy modeling yr É um 
method can be classified into: modeling using interpolation " d Ut 
points of startigraphy interface and modeling using original 
borehole's captured data. They are discussed as follows. Constmict NTPYs 
location, ; x 
of saving 3.1.1 Modeling Using Interpolation Points of Interface: This iu ns 1. Musee Crete 
rimitives modeling method constructs NTPV. The main process includes Triangulation line 
he data 
gle, side- 
order to 
pological 
interface 
r texture 
three steps. First, borchole data are compiled and divided into 
different layers according to their lithology and height. Second, 
stratigraphy interfaces are carried out curved face interpolation 3.1.2 
and thus, regular multi-DEMs are built. These DEMs should 
have coincident frames of references so as for the convenience 
of later modeling. Finally, grid triangulation should be made to 
Figure 4. Constructing of QTPVs by interfaces crossed grid 
Modeling Using Borehole Captured Data: In this case, 
the constructed model is QTPVs. This modeling is similar to 
Delaunay triangle network. The latter is for constructing 
intersecting points belonged to the same interface into a TIN, 
structures 
omposed 
nt or 
be regular and consequently NTPVs are constructed. with 
corresponding two triangles located in adjacent stratigraphy 
interfaces. If the interfaces are crossed in a grid. then some 
special processes, such as calculation of cross line, and grid 
triangulation and construction of NTPVs, should be done, 
referred to Figure 4. 
41 
while the former is for constructing a series of QTPVs between 
adjacent interfaces. Let the data structure of a borehole curve 
point consists of 3D coordinates and an adjacent attribute code. 
The adjacent attribute code is the down adjacent geological 
body attribute of a point, and is numbered in increasing order 
from the earth's surface to the subsurface. The main steps are as 
follows: 
(1) Create a triangle (up-triangle) using the methods of 
constructing a Delaunay TIN according to the borehole 
location points on the earth surface. This triangle is the up-