A work in progress

The basic frame of this simulation was first written in actionscript during Mt.Merapi (Central Java) eruption in November 2010. It was a proof-of-concept design with primary goal to visualize configurable distances using polygonal ring zones. The codes used in this demo applies the same underlying principle. Only that it had been refactored and encapsulated to handle more tasks.

There are more to be done to improve existing approaches. One particular notion in the context of Indonesia regional level, is that the built-in data structure still need to be configured in a way so that current arrangement can cope with all administrative divisions. Deliberately prepared and properly abstracted model will advantage future development when model is exposed to relational join in spatial RDBMS. For most part however, it's all about defining scope and purpose. This showcase is intended to balance that abstraction with minimum effort and to see how the component works inside multi-layering scenario. List of GIS stacks behind the front end should show you how deep the customization can be made to achieve more advanced requirement. It is also important to note, that since requests are now automated, SQL queries can later be constructed to process further analysis.

Ideas? Discussion? reach me at ericherlangga at gmail dot com.  

Overview

For server-side processing and visualisation method, only Desa/Village (the 4th level of Indonesia administrative subdivisions) is available at the moment. It covers the whole islands and represents impact: polygon boundaries that have spatial relationship with current ring or areas that are being affected by each analytical zone footprint.

Visual output resulting from the built-in dataset also differs from that when user select their own custom data. Prepared data model is meant for precision query and coverage consistency. It is also designated to be used for advanced analysis involving thousands/millions records that can not be handled in browser. It makes most out of the power of PostGIS spatial database to conduct real geometry collision tests then delegates the result to OGC standard-compliant mapping server, in this case Geoserver, to get the final rendering. For user custom data, collision tests were done purely in browser using geometry bounding box.

There are many tools available for buffer operation such as Multi-Ring Buffer, Multi-Distance Buffer, InaSAFE plugin package for QGIS, or ArcGIS buffer to perform full-fledged analysis. Aside from being deliverable via web, what makes this tool unique is that both the distance buffering and aggregation are performed simultaneously using drag and drop. It leverages user ability to perceive their data as it rapidly generates answer for simple questions such as how many properties/people are there or how much land is there being affected in each zoning category. Output is then formulated using user-selected well-known statistic aggregate functions: COUNT, SUM, AVERAGE, MIN, and MAX. These functions are expandable.

Descriptions of the data used in the sample and featured layers: