Suceptibility Mapping Technique:
An Application of Remote Sensing and
Geographic Information System (GIS): A Case Study in
Sto. Tomas-Marela River, Mt. Pinatubo,
Zambales, Philippines
Remote Sensing: Lahar Change Detection
Methodology
Visual interpretation, three-date color composite and density slicing comparison techniques were used. The use of moderately scaled AP's (1;18,000.1:25,000) and 1;50,000 topo maps were vital in the interpretation of the satellite images. The generalized schematic diagram of the procedure is presented in Fig.6.1.
Image Interpretation
Table 6.2/Fig.6.2 illustrates the three-date color composite, which presents major area of change and no change. The black and white color indicates no change as shown by some protions of the vegetated area (V). Chromatic colors such as blue, yellow, cyan, magenta, red and green denote areas of change. Change from 1991 to 1995 is indicated by red and cyan. Red shows the highly silted tailings pond used by Dizon Mines, and the hyperconcentrated streamflow lahar channels along the Sto Tomas river. Red also show the water areas in 1991 now replaced by laharic deposits. The adjoiningedge of the major debris flow presents a viable example. Cyan shows the additional land area and vegetation cover that was submerged due to the growth of the lake. Likewise, it is manifested by the watered agricultural area on the lowlands, some vegetation areas under shadows. Cyan along the edge of the lake represents the growth and expansion from 1988 to 1995.
Blue, green yellow and magenta represent areas that changed twice from 1988 to 1995. Blue is typically represented by land or vegetated areas submerged under water. These are the lands is typically represented by land or vegetated areas submerged under water. These are the lands submerged by mapanuepe Lake in 1991 rainy season. Green areas show scarce vegetation or bare soil to healthy cogonal grasses and back to scarce vegetation or bare soil. Both green and yellow represent areas with abrupt changes from 1988 to 1995. These changes are represented by the leading edge of the major lahar debris flows along the Sto. Tomas river in 191 and the thick ash deposits on the scarcely vegetated Mt. Pimmayong. Magenta could be displayed by vegetation I which reflectance response possibly returned to a high value. Other possibilities of magenta could be regrowth areas along the slopes and lowlands, and the pyroclastic flows on the upper reaches of Marella.
The three-data color composite from density sliced SPOT-XS images as shown in Fig.6.2a was alternately used for interpretation. Band 1 of 1991, 1995 and 1988 were combined using RGBcolor assignemtns, respectively. The white (upstream of Sta. Fe River and soil cover in the plain) and back (vegetation in the Sto. Tomas Plain and the vast forest cover) areas show no change from 1988 to 1995. Red represents enhanced features in 1991 image, green in 1995 is shown by the intense yellow from image. More lahar debris flows aggradation from 1991 to 1995 is shown by the intense yellow from the upstream of Marela River to Sto. Tomas (reaching the vicinity of Brgy. San Rafael, going south-southwest). Cyan represents the change from 1988 to 1995 as the entire channel had been filled up by sediments and growth of the lake. Magenta is the change from 1988 to 1991 wherein the plain had been covered by ashfall. The green features in the lower right corner of the image are cloud cover from 1995 image. But the green color at the debouchure of Sto. Tomas within the coastal areas represent lahar flows.
GIS Application: Components of Factor Maps on Lahar Distribution
For any natural hazard zonation evaluation and prediction, several spatial data layers are necessary. For scientific, accurate and near to truth prediction, these layers should be related to geology, geomorphology and topography of the area under study. Selection of appropriate data layers, which are not only meaningful, but also necessary for the specific conditions, is the most important task of any methodology dealing with natural hazard analyses.
For lahar hazard prediction, all types of information related to terrain, structures, slope gradient aspect, drainage, rainfall, seismicity, etc. should be at hand. For this particular study, some information layers are not available. Present study is being carried out at medium scale (1:50,000/1:25,000). Either it is landslide or lahar hazard zonation; an advantage at medium scale is to provide a map, based on minimum input layers indicating different levels of hazard potential. Thus, the map produced can be easily used for various planning purposes. Such fact is relevant in lahar-prone areas like Pampanga, Tarlac and Zambales, all being attack by lahar-related processes after the 1991 eruption. In developing countries like the Philippines, the advent of related processes after the 1991 eruption. In developing countries like the Philippines, the advent of remote sensing and GIS technology is of great help in delineating areas changed by repeated occurrences of natural hazards. Thus, this study makes use of remote sensed data to possibly detect changes and produce scientific based susceptibility map with reasonable accuracy.
For any GIS analysis, all maps entered for overlaying must have the same pixel value. In the study area, all rasterized maps as assigned 20m-pixel value as transformed from the georeferenced 1995 SPOT-XS.
Lahar Susceptibility Mapping
General
Susceptibility means the capability of being affected by natural damaging events or phenomena. Hazard is the probability of occurrence, within a specific time in a particular area, of natural damaging events (Varnes, 1984).
A hazard susceptibility map is a mpa derived from the analysis and interpretation of several sets of hazard-related data (Doornkamp in Peraz, 1992). He emphasized that the primary form of hazard map should show feature or set of features existing on the ground related to a particular hazard. With this, the extent of distribution of a particular process-hazard in a particular factor map is delimited. The output map shows the probable response of a particular map unit to specific hazards, expressed as a relative hazard within a simple homogenous map unit. The map to be produced is related to lahar processes actively in progress commonly during the onset of the rainy season. With the geomorphic parameters considered, the resulting susceptibility map can be used in preparing the lahar hazard zonation map. The outcome will in a way help and aid in planning any rehabilitation related purposes. The susceptibility map at the vicinities of Mt. Pinatubo is derived considering the set-up of the are, seven years after its eruption.
Methodology
The main method applied to generate the map is by deterministic approach and date-driven system. The deterministic approach is subjective with the "concept of possibility" (Kim, 1988) appears appropriate since a limited historical information related to the eruption is available. This is basically a direct mapping methodology wherein the susceptibility classes are determined on the discretion of the researcher. However, this direct mapping approach is only applicable in accessible areas. In here, of sequential satellite imageries and aerial photographs are extensively utilized particularly in inaccessible terrains. Decision rules used in this aspect are based on professional reasoning instead of pure statistical analysis.
For the data-driven technique, weights of parameter maps are calculated base from the density of the process map involved. The output map in this respect is purely statistically generated.
Generated Lahar Susceptibility Map
The susceptibility map created is delimited by the following condtions;
- That the susceptibility boundary is estimated based on direct mapping approach;
- That the map assumes engineering interventions;
- That the map is based on lahar lateral distribution;
- That this map applies to seven years after the eruption.
No prominent differences occurred in generating the map from the two techniques mentioned. Fig. 8.3a shows the susceptibility map generated from aereal density calculation with calculated weights of each parameter maps. From the computer-generated method, Sto. Tomas Plain is designated as low to high. The distributary bars which are elevated to some extent, are classed as low to moderate with the channels as high. In general, areas at the fringe of the lahar filed are grouped as high.
The susceptibility map generated from deterministic approach (Fig. 8.3 g) relies on the familiarity of the area and field observations conducted. The whole Sto. Tomas Plain is regarded as high to moderate where distance from the edge of the dike is one of the criteria considered. Although, distributary bars are generally elevated, those located 5.0 kms. Southward are highly susceptible to lahar deposition. Channels in the entire plain are also ranked as high since lahars directly deposited their loads along the conduits.
The reliability of the susceptibility maps generated is dependent on the [performance/stability of the armored dike constructed in the southern segment of Sto. Tomas River. The bit question is … HOW LONG CAN THE DIKE RESISTS THE DEMANDING FORCE OF INCOMING LAHARS?
Lahar Hazard Mitigation
Lahar produced from the eruption of Mt. Pinatubo is significantly hazardous to life and property at about 60 kms. Away from the volcano (PHIVOLCS, 1993). Human population and properties are always at stake from the devastating effect of lahars. Any structures along the route of lahars can be destroyed by direct impact or by inundation and burial from sediment accumulation.
Two basic strategies are suggested to mitigate the hazard;
- Prevent human population from occupying hazardous areas
- Prevent lahars from encroaching human settlements either by diverting and channeling the lahars along harmless flow paths.
The first strategy is quite to achieve since around Mt. Pinatubo, thousands of human population had been occupying the area for a long time.
The second strategy embarks on the construction of engineering works and structures. Most of the pre-eruption temporary dikes (artificial levees) on the downstream side had been breached in 1991 affecting Brgys. Alusiis and isolating Sta. Fe. When the bridge linking the village with San Rafael was toppled. In 1992, dikes were constructed, however overcome by succeeding lahars. Monitoring the constructed dikes is another aspect to ensure its effectiveness.
Fig. 1.4 Methodology of the study area
Fig. 2.2 Annual Rainfall Data in Zambales
