3. ARCHAEOLOGICAL GIS 
3.1 Primary Data 
Table.1 Base data used in the constructed GIS 
result of the preceding pollen analyses (Cultural Properties 
Protection Department of Aichi, 1994, 1999) around the target 
area through the intended period. 
  
  
Selecting the tree species 
interpreting the major species 
| POLLEN ANALYSIS DATA e— 
  
  
  
  
  
  
  
  
  
DATA NAME spatial attributes 
: ; name, period, type, area, store tools, 
Ruin point ; 
pottenyetc 
DEM 10, 50, 250mMESH elevation 
Topographic map 1/50000 topography 
Geologic map 1/50000 geography 
Temperature 
Normals for the 1KmMESH Temperature distribution map 
Perind 
Sea-Level point eustatic chanæ of sea level 
Water Hazard 1/50000 micro topography water hazard risk 
Pollen point pollen type proportion 
River line river 
  
  
  
  
  
Locations of ruins were digitised from the Ruin Map of Aichi 
Prefecture (1994, 1995, 1996), published by Cultural Properties 
Protection Department, board of education, Aichi. Attributive 
information for each ruin was collected from the chart attached 
to the map. Further detailed information was collected from the 
individual excavation reports of each ruin. Selection of data 
attributes was optimised for environmental archaeology by 
paying special attention to excavated remainder. Especially, the 
form of the pottery, and types of stone tools were taken into 
account as their combinations and proportions could form an 
important foundation to estimate the life style of the ancient 
people. Such information could be examined and further studied 
by multi-layering of the ancient environmental data sets. The 
use of GIS enables clarification of not only the contents of the 
phenomena but also the spatial relation. As long as spatial 
phenomena, such as cultural diffusion or settlement pattern are 
vital topics in archaeology, this function is unquestionably 
relevant. 
3.2 Secondary Data (VEGETATION) 
The past vegetation was reconstructed through an analysis of 
the preference of vegetation for environments. There are studies 
for ancient vegetation reconstruction using GIS (e.g. Spikins, 
1999). This present study features Warmth Index to estimate the 
basic distribution of the vegetation. WI is a temperature index 
of which parallelism with the distribution of tree species (Kira, 
1950) is widely recognised. WI can be calculated by the 
mathematical expression below. 
WI — Y (tm-5) (1) 
im»5 
where tm — average temperature of the month 
Past WI was calculated from Temperature Normality for the 
Period (1981-2000), by adjusting the difference of the 
temperature (Nogami, 1994). The layers, then, used for the 
consequent processing were: 
1) WE 
2) Topography; 
3) Geography; 
4) Distance from the sea; 
5) Ridge; and 
6) Valley. 
Each of the above layers and preference of each tree species 
were compared, and probable distribution of primary vegetation 
was estimated. The selection of tree species was based on the 
  
  
  
  
Compute the broad tree distribution 
WI DATA calculating the WI 
classify the species to each Wi 
  
  
  
  
  
  
TOPOGRAPHIC DATA y 
tapography , : p 
nda = Estimate the detailed distribution 
uy E scoring the tree rule 
di stance fo sea averiay 
geography 
  
  
   
  
  
  
  
Map of Past Vegetation 
Fig.2 Calculating the most competitive vegetation from layer 
and matrix 
  
  
Polygon A 
Layer2(ridge):yes 
Layer3(valley):no 
Layer4(WI):45-50 
Layer5(topo) hill 
  
  
  
   
Vegetation Preference Matrix 
  
Most Competitive 
vegetation is set to 
polygon 
  
  
  
Fig.3 Calculating the most competitive vegetation from layer 
and matrix 
Primary vegetation was estimated by overlaying the data sets. A 
matrix was used to score the preference of the tree, the most 
competitive vegetation was identified as the primary vegetation, 
and polygons were drawn. When the score is same, the 
vegetation was regarded as a mixed forest. 
  
  
    
-— 
3 
2$ nonsi] 
Tele 
SIN 
Fig.4 Calculated Vegetation of the 10000Years ago 
     
-474— | 
  
  
  
ml 
  
B