THE POTENTIAL ROLE OF THE COMMUNITY FOR THE FLOOD RISK ASSESSMENT USING FEATURES EXTRACTED FROM LiDAR DATASETS

Transcription

1 THE POTENTIAL ROLE OF THE COMMUNITY FOR THE FLOOD RISK ASSESSMENT USING FEATURES EXTRACTED FROM LiDAR DATASETS Gus Kali Oguis 1, Dr. Genelin Ruth P. James 1, Cinmayii G. Manliguez 1,2, Christine Lou Adino Lazaga 1, Tisha Mae Lopena 1, Engr. Joseph E. Acosta 1,2 2 Department of Mathematics, Physics, and Computer Science, College of Science and Mathematics, University of the Philippines Mindanao, Mintal, Davao City 8022 Philippines 1 Phil-LiDAR 1.B.13, College of Science and Mathematics, University of the Philippines Mindanao, Mintal, Davao City 8022 Philippines KEY WORDS: DSM, DEM, Feature Extraction. ABSTRACT: goguis@upmin.edu.ph rpjames@up.edu.ph jeacosta@up.edu.ph This paper explores the potential role of the local government unit for the attribution of features extracted from LiDAR Technology in order to complete the database salient to flood risk assessment. The digitized ground features from the activity were utilized as inputs to the various analysis needed in flood hazard mapping and flood inundation modelling of the river in the community. The areas covered for the feature extraction project are the flood plain of Lipadas and Sibulan River basins. Digitization of features such as buildings and road networks were accomplished. Each feature that was digitized were attributed to provide specific information. The potential role of the community were explored in the attribution of features, field data gathering and ground verification. 1. INTRODUCTION Flood risk assessment is basically an analysis of the risk of flooding in relation to its residential, commercial and industrial impact as a whole. Flood Risk Analysis investigates the flood process chain from precipitation, runoff generation and concentration in the catchment, flood routing in the river network, possible failure of flood protection measures, and inundation to economic damage (H. Apel 2004). The main objective of this study is to conduct a viable initial flood risk assessment for flood plains. Hand in hand with the flood modeling, from a different study, this particular study aims to assess risks on Lipadas and Sibulan flood plain. Specifically this study aims to: 1. Digitize features such as buildings, roads, water bodies, and bridges for Lipadas and Sibulan flood plain. 2. Explore potential role of the community in validation and attribution of extracted features from LiDAR Datasets. 2. MATERIALS AND METHODS For the initial stage of the study, the features that are to be digitized are taken from a LiDAR DEM, which has a 1 meter resolution. With the aid of orthophotos and other satellite images, like google earth, the features are digitized from the Digital Surface Model on the specified pre-determined flood plain of Lipadas and Sibulan. Buildings, roads, bridges, and water bodies were manually traced to produce shapefiles that would correspond as a digitized feature. Figure 1 shows the overview flow of the method. 2.1 Digitization of Features The 2 specific objectives are attainable and are divided into two steps, digitization and field validation. The digitization, involves Digital Elevation Model manipulation and can be fully done in a workstation able to run GIS software. The next step, however, would involve several different methods that could go hand in hand. This will explore the role of the community in the attribution of features. 2.2 Attribution of Features For the attribution of features, initially, several methods can be conducted. Open source maps which could provide initial attributes for the digitized features can be acquired. This however, may not be thoroughly complete and perfectly accurate. With this, the verification process would require actual field validation. In order to do so, before

2 going to the field, certain preparations are done. Using GPS devices and with the help of maps, navigating through out the flood plain with carefully planned route is done with ease. Digitized features may contain thousands of features even on a 10 sq.km area, depending upon the building density of the flood plain area. Ideally, visiting all the digitized features on the field should be done for more accurate results, but it is not practical and very exhaustive of the resources available such as time, manpower, and budget. As such, the role of the community, specifically the local government units are highlighted on this step. Tables 1-3 shows the additional attributes to be added and verified on the actual field validation on corresponding features. Table 1. Road attributes to be added on digitized features. Fig.1. Method Overview Road Network Road Name Number of Lanes National Road Provincial Road City / Municipal Road NA PR CM Barangay Road BR Table 2. Water Bodies attributes to be added on digitized features. Water Bodies River / Stream Lake / Pond Sea Dam Fish pen Water body name RS LP SE DM FP 3. RESULTS AND DISCUSSION The flood plain total area for Lipadas and Sibulan are 18.3 sq.km and 4.96 sq.km respectively, a total of sq.km. Figures 2 and 3 shows a sample data for the extracted feature on Lipadas and Sibulan respectively.

3 A total of 9723 and 1606 digitized buildings for Lipadas and Sibulan respectively extracted features for building alone over the span of sq.km is very exhaustive for field validation as stated earlier in the methodology. In the case of roads and water bodies, the attributes needed to be added are easily identified unlike buildings. With this, the role of the community, specifically the local government units, were explored. The local community as whole, readily knows their own vicinity. With this fact, it is a given that the study should reach out to the community. Since the part of the study which was primarily intensive regarding the building attributes, which has more variation regarding the type of building, it is imperative that the study should contact and ask advice from the local government units to help with this matter. Encouraging the community to participate is the key idea. After all, the study is geared to help the society, which directly affects them, it is of no argument that they also help themselves (K. Ronan 2005). Table 3. Building attributes to be added on digitized features. Buildings Building Name Height from ground Residential School Market / Prominent Stores RS SC MK Agricultural & Agroindustrial AG Medical Institutions Barangay Hall Military Institution MD BH ML Sports Center / Gymnasium / Covered Court SP Telecommunication facilities TC Transport terminal (Road, Rail, Air, and Marine) TR Warehouse Power plant / Substation NGO / CSO Offices Police Station Water Supply / Sewerage Religious institutions Bank Factory Gas Station Fire Station Other government offices WH PP NG PO WT RL BN FC GS FR OG Other commercial establishments OC

4 Fig.2. Data sample of extracted features for Lipadas. Fig.3. Data sample of extracted features for Sibulan.

5 4. CONCLUSION Based on the primary objective, with the gathered data, the study was able to conduct an initial step in flood risk assessment. Given the flood plain, and knowing the specific features and attributes within it, it is easily determined that a certain flood could be of any impact to the society. However, whether how much impact a certain flood would cause, would be a different study. This is only the initial step for the completion of the flood risk assessment, actual flood models or a flood inundation maps would be needed. Apart from that, a method for weighing impacts on the affected buildings, roads, and other features by the flood models is also needed. With the help of the community, specifically the Local Government Units, the field validation will be done faster and more efficient. With the same method but a vaster scale, the verified and validated attributes will be used for the initial flood risk assessment. In the future study, the method for verifying the attributes via field validation can be improved. Together with the flood models and flood inundation maps, impact assessment using weights and other methods can be used for more accurate analysis. REFERENCES H. Apel, A. H. Thieken, B. Merz, G. Bl oschl. Flood risk assessment and associated uncertainty. Natural Hazards and Earth System Science, 2004, 4 (2), pp K. Ronan, D. Johnston. Promoting Community Resilience in Disasters. The Role of Schools, Youth, and Families ISBN