ISPRS, Vol.34, Part 2W2, “Dynamic and Multi-Dimensional GIS”, Bangkok, May 23-25, 2001 
312 
Preliminary Statistical / Graphical analysis 
Model Description 
Model Derivation 
Model Evaluation 
—--M ■ ■ = 
Model Validation 
Figure 3-2: Model Development Processes 
c. Processes. 
These are the processes involved in deriving the optimal 
model as well as developing the rainfall estimation system. 
These include GIS analysis, point rainfall estimation, updating, 
and area-average rainfall estimation as indicated in the figure 
below: 
Figure 3-3: Processes involved in the system. 
d. Graphical User Interface (GUI). 
It is a menu-driven application to implement the GIS-based 
rainfall estimation system. It consists of the functions including 
the update of database, point and area-average rainfall 
estimation, display of the outputs such as estimates, relief 
map and isohyet. In this menu-driven application, the user is 
presented with a list of actions from which a choice can be 
selected. By following the instructions given by the menus, the 
system can be easily assessed by the users. In GIS, this 
interface is implemented with the aid of ARC Macro Lanquaqe 
(AML). 
4. IMPLEMENTATION 
4.1 Prototype System Design 
Based on the conceptual framework mentioned, the system is 
designed schematically as shown in Figure 4-1. The whole 
system is composed of four major components namely GIS 
databases, rainfall estimation engine, model development, and 
user interface. All the four components are interrelated such 
that: databases are the foundation to all the other 
components, model development and user interface 
components involve rainfall estimation engine, and user 
interface is based on the optimal model derived in the model 
development component. Each of these components is 
depicted in the following figures. 
Rainfall Estimation 
Figure 4.1 : System Design Schematic 
4.2 Databases Development 
With identified spatial and attribute data in the database 
required for the model development, the steps involved in 
developing the database is indicated in Figure 4-2. 
Figure 4-2 GIS Database Development Steps 
4.3 Rainfall estimation Engine 
Rainfall estimation method involves the regression analysis of 
spatial correlationship of topographic parameters with rainfall 
within the area of interpolation. This requires the processes 
spatial analysis as well as surface analysis in determining area