Conference PaperPDF Available

ASSESSING THE IMPACTS OF FLOODING IN TAGO RIVER BASIN, MINDANAO, PHILIPPINES THROUGH INTEGRATION OF HIGH RESOLUTION ELEVATION DATASETS, LANDSAT IMAGE ANALYSIS, AND NUMERICAL MODELLING

Authors:

Abstract and Figures

In this paper, we present the methods and results for the assessment of flooding brought by two tropical storms in Tago river basin though integration of high resolution elevation datasets, satellite image analysis and numerical modeling. The datasets utilized in this paper include the 1-m spatial resolution LiDAR-derived Digital Terrain and Surface Models (DTM and DSM), 10-m Synthetic Aperture Radar Digital Elevation Model (SAR DEM), and land-cover map derived from the analysis of Landsat 8 Operational Land Imager (OLI) and Landsat 7 Enhanced Thematic Mapper plus (ETM+) satellite images. Tago River Basin is one of the river basins covered by the Caraga State University (CSU) Phil-LiDAR 1 Project under the " Phil-LiDAR 1. Flood Hazard Mapping of the Philippines Using LiDAR " Program supported by the Department of Science and Technology (DOST). The two tropical storms considered in this work are the tropical storms " Lingling " (locally named Agaton) and " Jangmi " (locally named Seniang) which occurred last January and December 2014, respectively. These storms brought heavy to torrential rains to Tago river basin which caused severe flooding to its low-lying areas. We reconstructed these two events and generate corresponding flood hazard maps to assess the impacts it brought to the existing infrastructures and land-cover of Tago river basin, and to provide useful information to the Local Government Units and the community in the river basin about the negative effects of flooding in their areas that they may use in formulating mitigating measures to reduce the damaging impacts of flooding.
Content may be subject to copyright.
ASSESSING THE IMPACTS OF FLOODING IN TAGO RIVER BASIN, MINDANAO,
PHILIPPINES THROUGH INTEGRATION OF HIGH RESOLUTION ELEVATION
DATASETS, LANDSAT IMAGE ANALYSIS, AND NUMERICAL MODELLING
Meriam Makinano-Santillan, Jojene R. Santillan, Arthur M. Amora, Jennifer T. Marqueso, Linbert C. Cutamora,
Jesiree L. Serviano, Ronald M. Makinano, Lorie Cris A. Asube
CSU Phil-LiDAR 1 Project, College of Engineering and Information Technology, Caraga State University,
Ampayon, Butuan City, Philippines,
Email: meriam.makinano@gmail.com
KEYWORDS: Tago river basin, satellite image analysis, numerical models, flood hazard map, impact assessment.
ABSTRACT: In this paper, we present the methods and results for the assessment of flooding brought by two
tropical storms in Tago river basin though integration of high resolution elevation datasets, satellite image analysis
and numerical modeling. The datasets utilized in this paper include the 1-m spatial resolution LiDAR-derived
Digital Terrain and Surface Models (DTM and DSM), 10-m Synthetic Aperture Radar Digital Elevation Model
(SAR DEM), and land-cover map derived from the analysis of Landsat 8 Operational Land Imager (OLI) and
Landsat 7 Enhanced Thematic Mapper plus (ETM+) satellite images. Tago River Basin is one of the river basins
covered by the Caraga State University (CSU) Phil-LiDAR 1 Project under the „Phil-LiDAR 1. Flood Hazard
Mapping of the Philippines Using LiDAR‟ Program supported by the Department of Science and Technology
(DOST). The two tropical storms considered in this work are the tropical storms “Lingling” (locally named Agaton)
and “Jangmi” (locally named Seniang) which occurred last January and December 2014, respectively. These storms
brought heavy to torrential rains to Tago river basin which caused severe flooding to its low-lying areas. We
reconstructed these two events and generate corresponding flood hazard maps to assess the impacts it brought to the
existing infrastructures and land-cover of Tago river basin, and to provide useful information to the Local
Government Units and the community in the river basin about the negative effects of flooding in their areas that
they may use in formulating mitigating measures to reduce the damaging impacts of flooding.
1. INTRODUCTION
Flood is the most common natural disaster that occurs in the Philippines. It has devastating effects to properties,
livelihood, and lives in the community. With these impacts, the Philippine government initiated a program that
would help the community in mitigating and minimizing such damages through proper planning with the aid of the
high spatial resolution flood hazard maps. This project is the Phil-LiDAR 1 Project or the Hazard Mapping of the
Philippines using LiDAR” program. This program has a component project known as the CSU Phil-LiDAR 1 which
is being implemented by Caraga State University. The CSU Phil-LiDAR 1 project processes and utilizes high-
resolution datasets such as LiDAR-derived Digital Elevation Models to develop flood models and generate flood
hazard maps of the river basins within Caraga Region, Mindanao, Philippines.
Tago River Basin (Figure 1) is one
the twelve project areas of CSU Phil-
LiDAR 1. It has an approximate
drainage area of 1,444 km2 with its
major river passing through the
municipalities of San Miguel and
Tago of Surigao del Sur. The Tago
river basin is one of the areas which
was badly hit by the tropical storms
„Lingling‟ (locally known as
Agaton‟) and „Jangmi‟ (locally
known as Seniang) last January
2014 and December 2014,
respectively. These tropical storms
caused widespread flooding of the
municipalities within the river basin
(NDRRMC, 2014; 2015).
Figure 1. Map showing the location and coverage of Tago River basin.
1606
36th Asian Conference on Remote Sensing 2015 (ACRS 2015), Quezon City, Philippines, Volume 3, pp. 1606-1617, 2015
For Tago river basin, flood hazard mapping and impact assessment of the two tropical storms were conducted
through the integration of high resolution spatial datasets. These datasets are the LiDAR-derived Digital Terrain and
Surface Models (DTM and DSM), Synthetic Aperture Radar Digital Elevation Model (SAR-DEM), and land-cover
map derived from the Landsat 7 Enhanced Thematic Mapper plus (ETM+) and Landsat 8 Operational Land Imager
(OLI) satellite images, among many others. This paper focused on the reconstruction of flooding during the
„Agaton‟ and „Seniang‟ events in which the number of affected buildings and area of inundated land-cover classes
were quantified. The result of this study may help the Local Government Units and the concerned community in
determining the previously flooded infrastructures and vegetation which are expected be flooded again in the near
future when heavy rains brought by storms like „Agaton‟ and „Seniang‟ come to the area. With the knowledge and
information derived from this work, the LGUs and the communities in the river basin can be guided in making
geospatially-informed decision making for managing and mitigating the negatives of future flooding events.
2. METHODOLOGY
2.1 Overview
Figure 2. Steps involved in the flooding impacts assessment of Tago River Basin.
Figure 2 shows the flowchart summarizing the steps involved in assessing the impacts of flooding in Tago river
basin. To assess the impacts of flooding, we need to generate flood hazard maps and determine the possible
exposure datasets within Tago river basin. The exposure datasets considered in this study are the existing building
and land-cover classes within basin‟s flood plain areas. The derivation of the buildings exposure datasets were done
through manual digitization of its footprints by utilizing the 1-m spatial resolution LiDAR-derived DSM. The land-
cover map of the basin was derived through the analysis of Landsat 7 ETM+ and Landsat 8 OLI satellite images.
Flood hazard maps were generated though numerical modeling by utilizing the 1-m spatial resolution LiDAR-
derived DTM and the land-cover information extracted from Landsat satellite images, together with the hydrologic
data. The numerical modeling consisted of the development of the hydrologic model of the river basin and the
hydraulic model of the flood plain areas. These models were used to determine the volume and discharge of water
coming from various sub-basins during rainfall and determine the activity of water on how it overflows from the
rivers into the flood plains, respectively. The outputs of these models are flood hazard maps which are validated for
its accuracy through comparison with the actual flooding data that were gathered in the field.
2.2 Exposure Datasets Derivations
2.2.1 Feature Extraction: The 1-m spatial resolution LiDAR-derived Digital Surface Model of Tago river basin
was used to locate and extract features within the area. These features include large buildings, households and other
man-made structures. The extraction was aided and was validated utilizing high-resolution satellite images from
Google Earth.
2.2.2 Satellite Image Analysis for Land-cover Map Derivation: The analysis utilized Landsat 8 OLI and Landsat
7 ETM+ images. Land-cover information was extracted from these images through Maximum Likelihood
classification. More than one image was utilized so that the missing data in one image due to cloud-cover can be
supplemented data from another image. These images are Landsat 8 OLI images acquired last March 31 and June 3,
2014, and Landsat 7 ETM+ image acquired last August 8, 2012. These images underwent radiometric calibration to
convert image pixel values to top-of-atmosphere (TOA) radiance; and atmospheric correction to correct the pixel
values for atmospheric effects (haze, path, radiance, etc.), and to convert the TOA radiance to surface reflectance.
1607
36th Asian Conference on Remote Sensing 2015 (ACRS 2015), Quezon City, Philippines, Volume 3, pp. 1606-1617, 2015
NDVI and DEM were also incorporated to the Landsat surface reflectance bands as additional data sources for land-
cover classification which has been found to increase classification accuracy as these datasets account for the
rugged topography so as to eliminate the presence or absence of certain classes in some elevation zones, and reduce
the impact of shadows and to enhance the separability among various vegetation classes (Watanachaturaporn et al.,
2008). Clouds and cloud shadows were manually digitized from the images which were used to mask and exclude
clouds and cloud shadows during the supervised classification. The Maximum Likelihood algorithm was used to
individually classify the layer-stacks of surface reflectance bands, NDVI and DEM. There were seven classes
considered in the analysis. These are barren, built-up, cropland, grassland, palm, forest, and water areas. The
missing land-cover information due to cloud and cloud shadow contamination in one image was supplemented
using the land-cover maps derived from the other images. After supplementing the missing data in the land-cover
map, it was further subjected to contextual editing to correct obvious misclassification through visual inspection.
The accuracy of the finalized land-cover map was then assessed using a procedure suggested by Congalton and
Green (2009) and by Congalton (1991).
2.3 Numerical Modeling and Flood Hazard Map Generation
2.3.1 Hydrologic Modeling and Calibration: The hydrologic model of Tago river basin was based on the
Hydrologic Engineering Center Hydrologic Modeling System (HEC HMS), a simulation program designed to
simulate the precipitation-runoff processes of watershed systems. HEC HMS modeling is dependent on the three
components: the basin model, meteorological model, and the set of control specification. For Tago, the basin model,
which is the physical representation of the watershed, was developed by utilizing a 10-m Synthetic Aperture Radar
Digital Elevation Model (SAR DEM) and rivers networks in the delineation of watersheds; and was parameterized
using the information from the land-cover maps that was generated through the analysis of Landsat 7 ETM+ and
Landsat 8 OLI satellite images.
The HEC HMS model setup of Tago consisted of 186 sub-basins, 93 and 94 junctions (Figure 3). The model
simulates flow hydrographs based on rainfall data recorded by ASTI DOST rain gauge located at Barangay Tina in
Municipality of San Miguel. The parameters of the model were calibrated by comparing the simulated flow
hydrographs to the actual measured flow in the river. The station utilized during HEC HMS model calibration was
on Cabtik Bridge in San Miguel, Surigao del Sur (approximately located at the middle portion of the river basin).
Hydrological data necessary for calibration was gathered from this station last 12/16/2014 to 12/23/2014 with the
use of water level and velocity data logging sensors together with the river cross-sectional data.
2.3.2 Hydraulic Modeling: The hydraulic model of Tago river basin was based on the Hydrologic Engineering
Center River Analysis System (HEC RAS), a simulation program designed to perform one-dimensional hydraulic
calculations for a full network of natural and constructed channels (USACE, 2010). The Tago HEC RAS model was
developed by first creating geometric representations of the rivers and the floodplains. These geometries are the
river centerlines, banks and cross-sections and the floodplain boundary (Figure 4). The LiDAR DTM was used as
the primary source of elevation data for the cross-section lines. Parameterization of the HEC RAS model utilized
the land cover information by extracting the Manning‟s roughness coefficients, and these values were assigned to
the cross-section segments. Steady flow simulation module of HEC RAS was used to generate flood depth
estimations during the two storms. This module can do one-dimensional water profile calculations for steady
gradually varied flow. Under the steady flow, the flow or discharge values at the inflow were specified using the
result of the HEC HMS simulation. The HEC RAS model of Tago consisted of 25 junctions and 52 reaches. Among
these reaches, 25 are internal flows and 27 are inflows boundary conditions. The interface of Tago HEC RAS model
is shown in Figure 5.
2.3.3 Flood Depth and Hazard Mapping: The calibrated HEC HMS model was used to simulate discharge
hydrographs for the „Agaton‟ and „Seniang‟ events, with the following simulation periods: January 10-24, 2014 and
December 20, 2014-January 5, 2015, respectively. For generating flood depth and hazard maps for an extreme
rainfall event, the maximum flow rate at inflow and internal flow boundary condition locations were obtained from
simulated discharge hydrographs and used as input for HEC RAS steady flow estimation. With this information set
as boundary conditions, HEC RAS was able to compute the maximum water surface profiles in all the cross-
sections within the model domain for a particular event. The water surface profiles computed by the HEC RAS
model were converted into flood depth maps through GIS post-processing using HEC GeoRAS, an extension of
ArcGIS. The procedures include generating a Triangulated Irregular Network (TIN) of water surface elevation
based on the computed water surface profiles at the cross-sections, converting the TIN into a water surface
elevation (WSE) grid, and then overlaying the WSE grid into the 1-m LiDAR DTM to estimate the flood depths
(i.e., subtracting the WSE grid by the DEM grid). The generated depth grid was then categorized based on its
1608
36th Asian Conference on Remote Sensing 2015 (ACRS 2015), Quezon City, Philippines, Volume 3, pp. 1606-1617, 2015
corresponding hazard level. The categorizations are: low hazard for depths of less than 0.50 m, medium hazard for
depths from 0.50 m to 1.50 m, and high hazard for depths of greater than 1.50 m.
2.3.4 Flood Hazard Map Validation: Flood map validation surveys were conducted last February 2015 to
determine the accuracy and reliability of the generated flood hazard maps. Pre-determined random locations within
the floodplain of Tago river basin were visited to determine whether they were flooded or not during „Agaton‟
(January 2014). The validation procedure consisted of comparing the actual flooding data from the field to the
flooding that was generated by the flood model. The total number of correctly predicted points over the total
number of points collected determines the overall accuracy of the flood hazard map.
Figure 3. The interface of the HEC HMS-based hydrologic model of Tago.
Figure 4. Geometric representations created for Tago river basin.
1609
36th Asian Conference on Remote Sensing 2015 (ACRS 2015), Quezon City, Philippines, Volume 3, pp. 1606-1617, 2015
Figure 5. Interface of Tago river basin HEC RAS model.
2.4 Flood Impact Assessment
Impact assessment was done to determine the
effect of flooding to the community. With the
existing flood hazard maps, the affected number
of buildings and the area of every land-cover
class were determined.
The impact of flooding was assessed by
intersecting the exposure datasets (building
footprints and land-cover) to the generated flood
hazard maps for the two tropical storms „Agaton‟
and „Seniang‟ (see Figure 7). The flooding
impact to land-cover classes was determined by
computing the inundated areas for each class for
both events. The building features were assessed
to the impact of flooding by categorizing every
building according hazard level (low, medium,
high, or not flooded) depending on what hazard
level they have intersected.
Figure 7. Figure showing on how the impact of flooding was
determined for each exposure dataset.
Exposure
Datasets
Hazard
Figure 6. Some of the photos showing the flood map validation surveys in Tago River Basin.
1610
36th Asian Conference on Remote Sensing 2015 (ACRS 2015), Quezon City, Philippines, Volume 3, pp. 1606-1617, 2015
3. RESULTS AND DISCUSSION
3.1 Exposure Datasets
3.1.1 Buildings: The digitized exposure datasets derived using the 1-m spatial resolution LiDAR-derived Digital
Surface Model (DSM) of Tago river basin floodplain areas totaled to 10,366 features. These features were checked
and validated using the high-resolution satellite images from Google Earth. Some of the digitized features of Tago
river basin are shown in Figure 8.
3.1.2 Land-cover: The land-cover map of Tago river basin derived from the analysis of the Landsat satellite images
is shown in Figure 9 with its statistics at Table 1. This land-cover map has an over-all classification accuracy of
92%. This was also utilized in the derivation of the two parameters necessary in the development of the flood
model. These are the runoff potential or the Curve Number (CN), and the Manning‟s roughness coefficient.
Figure 8. Figure showing some of the extracted building features in Tago river basin.
Table 1. Area of land-cover
classes on Tago river basin .
Area (km²)
51.14
4.39
168.35
999.12
146.43
55.30
19.75
1,444.49
Figure 9. Year 2014 land-cover map of Tago river basin derived using satellite
image analysis.
1611
36th Asian Conference on Remote Sensing 2015 (ACRS 2015), Quezon City, Philippines, Volume 3, pp. 1606-1617, 2015
3.2 Accuracy of the Hydrologic Model
The result of the HEC HMS model calibration
is shown in Figure 10, which shows the
observed and simulated flow hydrographs. In
evaluating the model performance before and
after calibration, three measures of accuracy
were used. These are the Nash-Sutcliffe
Coefficient of Model Efficiency (NSE),
percentage bias (PBIAS), and the RMSE-
observations standard deviation ratio (RSR).
These measurements are computed by
comparing the observed and the simulated
hydrographs based on the evaluation guidelines
(Moraisi et al., 2007). Based on the model
performance evaluation, the overall
performance of the hydrologic model before
calibration is already “very good” (NSE = 0.77,
PBIAS = -2.26, and RSR = 0.48). After the calibration, although the model has already been evaluated as “very
good”, the statistics were more improved (NSE = 0.96, PBIAS = -3.96, and RSR = 0.21), still indicating a “very
good” model performance.
3.3 Flood Hazard Maps Generated
The generated flood hazard maps of Tago for “Agaton” and “Seniang” events are shown in Figure 11 and Figure
12, respectively.
Utilizing the flood information gathered from several geographically located points within Tago river basin
floodplain areas, the flooding extent during tropical storm „Agaton‟ was analyzed. The result of the confusion
matrix analysis is shown in Table 2. The analysis indicates that the flooding extent generated by the model is
70.30% accurate. Based on the result, the computed user‟s accuracy for “not flooded” points, only gained 37.84%
accuracy which means that several “not flooded” points in the map were truly flooded. Nevertheless, the result
showed higher user‟s accuracy to the “flooded” points in map which is 89.06%. Also, the producer‟s accuracies for
both “flooded” and “not flooded” points gained high percentages of 71.25% and 66.67%, respectively which imply
that the model was able to correctly predict the flooding situation of majority of the points in Tago river basin.
Figure 11. Flood hazard map of Tago river basin during TS „Agaton‟.
Figure 10. Result of calibrating the Tago HEC HMS model
using the measured discharge data at Cabtik Bridge
1612
36th Asian Conference on Remote Sensing 2015 (ACRS 2015), Quezon City, Philippines, Volume 3, pp. 1606-1617, 2015
Figure 12. Flood hazard map of Tago river basin during TS „Seniang.
Table 2. Result of the flood validation analysis in Tago river basin during tropical storm „Agaton‟.
Actual Flooding Scenario
User's Accuracy
Flooded
Not Flooded
Total
Flood Model
Simulated
Flooding Scenario
Flooded
57
7
64
89.06
Not Flooded
23
14
37
37.84
Total
80
21
101
Producer's Accuracy
71.25
66.67
Sum of Diagonal Values
71
Overall Accuracy
70.30
3.3 Flood Impact Assessment
The results of the flooding impact assessment to the land-cover classes and the existing buildings in Tago river
basin floodplain areas are shown in Table 3 and Figure 13, and in Table 4 and Figure 14, respectively. The flooding
extents of the two tropical storms overlaid in the land cover map are shown in Figure 15 and Figure 16. Maps of the
categorized building according to its hazard level are shown in Figure 17 and Figure 18. These results were derived
from the flooding that has occurred during the two tropical storms „Agaton‟ and „Seniang‟.
Based on the generated flood hazard maps, a much deeper and wider extent of flooding occurred during tropical
storm „Agaton last January 2014 compared to tropical storm „Seniang‟ that happened last December 2014.This
comparison was exemplified with the results of the flood impact assessment wherein there were more land cover
areas and number of building inundated during the hit of tropical storm „Agaton‟. Also, it can be observed that
among the land-cover classes, the cropland areas are the most affected with 64.42% and 34.87% flooded during
„Agaton‟ and „Seniang‟, respectively. These results are to be expected since this land-cover class is situated in lower
elevations compared to other classes. The least affected land-cover class is the forest areas which are usually
situated in higher grounds. For the buildings that were affected, it can be noted that several structures were flooded
in both tropical storms. In totality, 5,449 out of 10,366 or 52.57% buildings were flooded during tropical storm
„Agaton‟and 2,311 or 22.29% buildings were flooded during tropical storm „Seniang‟. Among these affected
buildings, there were 1,559 during „Agaton‟ and 382 during „Seniang‟ that were at high risks which could have had
a flood depth of more than 1.50 meters. All of these results are assumed to be 70.30% accurate based on the results
of the flood validation analysis. This low accuracy could have been caused by the lack of river bathymetric data to
the DTM that was utilized. Nevertheless, this flood hazard map can already be an aid in assessing the impacts that
the two tropical storms brought to the area.
1613
36th Asian Conference on Remote Sensing 2015 (ACRS 2015), Quezon City, Philippines, Volume 3, pp. 1606-1617, 2015
Table 3. Table showing the area and the corresponding percentage of flooded land-cover classes during tropical
storms „Agaton‟ and „Seniang‟ events.
Class Name
Total Area
(km²)
TS Agaton
TS Seniang
Flooded Area
(km²)
Percentage
(%)
Flooded Area
(km²)
Percentage
(%)
Barren
5.07
2.25
44.41
1.54
30.37
Built up areas
143.79
14.64
10.18
9.78
6.80
Cropland
187.34
120.68
64.42
65.32
34.87
Forest
38.98
15.30
39.24
9.72
24.93
Grassland
139.76
12.95
9.27
9.23
6.61
Palm
27.65
12.82
46.37
7.82
28.29
Figure 13. Graph showing the percentages of land-cover classes flooded during tropical storms „Agaton‟ and
„Seniang‟.
Table 4. Table showing the number of buildings flooded during tropical storms „Agaton‟ and „Seniang‟.
Hazard Level
No. of Structures
TS Lingling
TS Jangmi
High (> 1.50m flood depth)
1,559
382
Medium (0.50 m - 1.50m)
2,781
916
Low (< 0.50 m)
1,109
1,013
Not Flooded
4,917
8,055
Figure 14. Graph showing the number of buildings flooded during tropical storms „Agaton‟ and „Seniang‟.
44.41%
10.18%
64.42%
39.24%
9.27%
46.37%
30.37%
6.80%
34.87%
24.93%
6.61%
28.29%
0%
10%
20%
30%
40%
50%
60%
70%
Barren Built up areas Cropland Forest Grassland Palm
Percentage
Land-cover Class
Percentages of Land-Cover Classes Flooded during Tropical Storms "Agaton" and "Seniang"
TS Agaton
TS Seniang
1,559
2,781
1,109
4,917
382 916 1,013
8,055
0
1,000
2,000
3,000
4,000
5,000
6,000
7,000
8,000
9,000
High (> 1.50m flood depth) Medium (0.50 m - 1.50m) Low (< 0.50 m) Not Flooded
No. of Buildings
Hazard Level
No. of Buildings Flooded during Tropical Storms "Lingling" and "Jangmi"
TS Lingling
TS Jangmi
1614
36th Asian Conference on Remote Sensing 2015 (ACRS 2015), Quezon City, Philippines, Volume 3, pp. 1606-1617, 2015
Figure 15. Flooding in Tago river basin during TS „Agaton‟ overlaid in the land cover map.
Figure 16. Flooding in Tago river basin during TS „Seniang‟ overlaid in the land cover map.
1615
36th Asian Conference on Remote Sensing 2015 (ACRS 2015), Quezon City, Philippines, Volume 3, pp. 1606-1617, 2015
Figure 17. Map showing the buildings in the major villages in Tago river basin that are categorized to its
corresponding hazard levels during tropical storm „Agaton‟.
Figure 18. Map showing the buildings in the major villages in Tago river basin that are categorized to its
corresponding hazard levels during tropical storm „Seniang‟.
1616
36th Asian Conference on Remote Sensing 2015 (ACRS 2015), Quezon City, Philippines, Volume 3, pp. 1606-1617, 2015
4. CONCLUSION
In this paper we showed that detailed assessment of flooding in Tago river basin can be done through integration of
high spatial resolution elevation datasets, Landsat image analysis and numerical modeling, and conduct of
geospatial analysis. The information depicted on the assessment would be very helpful to the Local Government
Units and the concerned communities within Tago river basin as aid in determining at risk infrastructures and
vegetation to flooding. With this knowledge, they may able to formulate mitigating measures to lessen or possibly
eliminate the negative impacts of flooding to the area through adaptive strategic planning which may include
relocating vulnerable communities to safe grounds.
5. RECOMMENDATIONS
Since the utilized DTM was not yet integrated with the river bathymetric data, the flood maps generated and the
results of the analysis may not be very accurate. The generated flooding extent could be more accurate when the
DTM that will be utilized will be integrated with river bathymetric data during the hydraulic modeling process. This
has a very big impact to the map‟s accuracy since the volume of water that flows in the river without bathymetric
data could not account the approximate volume capacity that the river could hold before it overflows. Hence, river‟s
without bathymetric data tends to overflow even with low volume of water.
Also, the result of the flooding impact assessment could have been more informative if the buildings features
utilized were attributed with necessary data such as the type of structure (i.e. residential, commercial, government
establishments, etc.), and the number of residents occupying the said building. With this type of information,
planning and decision making in terms of flood disaster management would be more defined and accurate.
The use of DTM integrated with river bathymetry in the numerical modelling, as well as the use of properly
attributed exposure datasets in the impact assessment, will be considered in the next phase of this work.
ACKNOWLEDGEMENTS
This work is an output of the Caraga State University (CSU) Phil-LiDAR 1 project under the “Phil-LiDAR 1.
Hazard Mapping of the Philippines using LiDAR” program funded by the Department of Science and Technology
(DOST). The SAR DEM and the LiDAR DTM and DSM used in this work were provided by the University of the
Philippines Disaster Risk and Exposure for Mitigation (UP DREAM)/Phil-LIDAR 1 Program. We thank all CSU-
Phil-LIDAR 1 technical staff and assistants, as well as the LGUs in the Tago River Basin for their assistance during
the conduct of hydrological measurements and flood map validation surveys.
REFERENCES
Congalton, R.G., 1991. A review of assessing the accuracy of classification of remotely sensed data, Remote
Sensing of Environment, 25-46.
Congalton, R.G., Green, K., 2009. Assessing the accuracy of remotely sensed data: principles and practices, CRC
Press/Taylor & Francis, 183p
Moriasi D. N. et al., 2007. Model evaluation guidelines for systematic quantification of accuracy in watershed
simulations, Transactions of the ASABE, vol. 50, pp.885-900.
NDRRMC, 2014. NDRRMC Updates Sitrep No. 33 re: Effects of Tropical Depression Agaton. National Disaster
Risk Reduction and Management Council. Retrieved February 1, 2014 from http://www.ndrrmc.gov.ph/.
NDRRMC, 2015. SitRep No. 22 re Effects of Tropical Storm SENIANG". National Disaster Risk Reduction and
Management Council. Retrieved January 10, 2015from http://www.ndrrmc.gov.ph/.
Santillan, J. R., 2013. Modeling of Flashflood Events Using Integrated GIS and Hydrological Simulations, Training
Center for Applied Geodesy and Photogrammetry, University of the Philippines, Diliman, Quezon City.
Watanachaturaporn, P., Arora, M.K., Varshney, P.K., 2008. Multisource classification using support vector
machines: an empirical comparison with decision tree and neural network classifiers, Photogrammetric Engineering
and Remote Sensing, (74) 239-246.
1617
36th Asian Conference on Remote Sensing 2015 (ACRS 2015), Quezon City, Philippines, Volume 3, pp. 1606-1617, 2015
... These flood events affected not only the communities living in the flood plain but also the sources of income which include large cropland areas. A preliminary assessment of the impacts of Agaton and Seniang in the Tago River Basin has recently been conducted by Makinano-Santillan (2015) . Using an integrated approach involving the use of LiDAR datasets, land-cover from Landsat images, and one-dimensional (1D) flood models based on HEC HMS and HEC RAS, the study was able to estimate 52.57% and 22.29% of the buildings situated in the floodplains to have been flooded during the two flood events, respectively. ...
... The results of the assessments ashowed an increase in the number of buildings and roads, and increase in areas of inundated land-cover as rainfall events become more extreme (i.e., increase in return periods ). The results of these assessments can be considered more reliable than the assessment conducted earlier (Santillan et al., 2015) due to higher accuracy of the flood maps generated in the present study. ...
Article
Full-text available
In this paper, we present a combined geospatial and two dimensional (2D) flood modeling approach to assess the impacts of flooding due to extreme rainfall events. We developed and implemented this approach to the Tago River Basin in the province of Surigao del Sur in Mindanao, Philippines, an area which suffered great damage due to flooding caused by Tropical Storms Lingling and Jangmi in the year 2014. The geospatial component of the approach involves extraction of several layers of information such as detailed topography/terrain, man-made features (buildings, roads, bridges) from 1-m spatial resolution LiDAR Digital Surface and Terrain Models (DTM/DSMs), and recent land-cover from Landsat 7 ETM+ and Landsat 8 OLI images. We then used these layers as inputs in developing a Hydrologic Engineering Center Hydrologic Modeling System (HEC HMS)-based hydrologic model, and a hydraulic model based on the 2D module of the latest version of HEC River Analysis System (RAS) to dynamically simulate and map the depth and extent of flooding due to extreme rainfall events. The extreme rainfall events used in the simulation represent 6 hypothetical rainfall events with return periods of 2, 5, 10, 25, 50, and 100 years. For each event, maximum flood depth maps were generated from the simulations, and these maps were further transformed into hazard maps by categorizing the flood depth into low, medium and high hazard levels. Using both the flood hazard maps and the layers of information extracted from remotely-sensed datasets in spatial overlay analysis, we were then able to estimate and assess the impacts of these flooding events to buildings, roads, bridges and landcover. Results of the assessments revealed increase in number of buildings, roads and bridges; and increase in areas of land-cover exposed to various flood hazards as rainfall events become more extreme. The wealth of information generated from the flood impact assessment using the approach can be very useful to the local government units and the concerned communities within Tago River Basin as an aid in determining in an advance manner all those infrastructures (buildings, roads and bridges) and land-cover that can be affected by different extreme rainfall event flood scenarios.
... Remote Sensing (RS) and Geographic Information System (GIS) have become effective geospatial tools for assessing hazards and risks associated with flood disasters (Manfré et al., 2012), most especially for the simulation of flood characteristics, and for the assessment of its social, economic and environmental consequences (Albano et al., 2015). These technologies have been used, either on their own or in a synergistic manner, to develop numerical flood simulation models that can aid in reconstructing past flood events for the purpose of mapping inundation levels and extents as well as to identify elements at risks (e.g., Amora et al., 2015;Makinano-Santillan et al., 2015;Santillan et al., 2016); in identifying flood-prone areas for the purpose of planning for disaster mitigation and preparedness (Asare-Kyei et al., 2015;Gashaw and Legesse, 2011;Pradhan, 2010;Samuel et al., 2014); and most especially in flood forecasting and early warning (Mioc et al., 2008;Sharif and Hashmi, 2006), among many other uses and applications. In these applications, RS has become an important source of data/information necessary to build flood models and conduct assessment of flood risks such as topography, land-cover, location of built-up areas and other elements that are at risk to flooding (e.g., roads, buildings, and bridges). ...
Article
Full-text available
Flooding is considered to be one of the most destructive among many natural disasters such that understanding floods and assessing the risks associated to it are becoming more important nowadays. In the Philippines, Remote Sensing (RS) and Geographic Information System (GIS) are two main technologies used in the nationwide modelling and mapping of flood hazards. Although the currently available high resolution flood hazard maps have become very valuable, their use for flood preparedness and mitigation can be maximized by enhancing the layers of information these maps portrays. In this paper, we present an approach based on RS, GIS and two-dimensional (2D) flood modelling to generate new flood layers (in addition to the usual flood depths and hazard layers) that are also very useful in flood disaster management such as flood arrival times, flood velocities, flood duration, flood recession times, and the percentage within a given flood event period a particular location is inundated. The availability of these new layers of flood information are crucial for better decision making before, during, and after occurrence of a flood disaster. The generation of these new flood characteristic layers is illustrated using the Cabadbaran River Basin in Mindanao, Philippines as case study area. It is envisioned that these detailed maps can be considered as additional inputs in flood disaster risk reduction and management in the Philippines.
Article
Full-text available
In mitigating and helping lessen the possible effects and damages of disaster to the communities, the transmission of information or end products derived from remote sensing and other multidisciplinary technologies into the community should be immediate, accessible and comprehensive to aid in better planning and decision-making procedures. In this paper, we share a hazard information dissemination procedure which integrates the use of outputs derived from numerical models, web applications and systems as well as the use of social media and telecommunications to promote the utilization of advanced science and technology outputs that could represent and visualize various flooding scenarios through social media and dynamic communication between stakeholders.
Article
Full-text available
Recently, support vector machines (SVMs) have proven to have great potential for classification of remotely sensed hyperspectral data acquired in a large number of spectral bands. Due to Hughes' phenomenon, conventional parametric classifiers fail to classify such a high dimensional dataset. In the past, neural networks and decision tree classifiers, which are nonparametric in nature, have frequently been used for classification of multispectral remote sensing data. These, however, have marked limitations when applied to hyperspectral data. In this paper, we present the results of applying SVMs for a 16-class classification of an AVIRIS image and compare its performance with decision tree, back propagation (BP) and radial basis function (RBF) neural network classifiers. There are a number of parameters that may affect the accuracy of SVM based classifiers. The best values of these parameters have been selected on the basis of a set of hypotheses and experiments. All the SVM classifications have been performed using an in-house code developed in a Matlab environment. The Kappa coefficient of agreement has been used to assess the accuracy of classification. The differences in classification accuracy have been statistically evaluated using a pairwise Z-test. SVM classification using a polynomial kernel of degree 2 produced an accuracy of 96.94% whereas the accuracies achieved by decision tree, BP and RBF neural network classifiers were 74.75%, 38.03% and 95.30% respectively. This clearly illustrates that the accuracy of the SVM classifier is significantly higher than decision tree and BP neural network classifiers at 95% confidence level. Although, the improvement in SVM classification accuracy with respect to the RBF neural network classifier was not statistically significant, the SVM classifier was more computationally efficient than the RBF classifier for hyperspectral image classification.
Article
Full-text available
Watershed models are powerful tools for simulating the effect of watershed processes and management on soil and water resources. However, no comprehensive guidance is available to facilitate model evaluation in terms of the accuracy of simulated data compared to measured flow and constituent values. Thus, the objectives of this research were to: (1) determine recommended model evaluation techniques (statistical and graphical), (2) review reported ranges of values and corresponding performance ratings for the recommended statistics, and (3) establish guidelines for model evaluation based on the review results and project-specific considerations; all of these objectives focus on simulation of streamflow and transport of sediment and nutrients. These objectives were achieved with a thorough review of relevant literature on model application and recommended model evaluation methods. Based on this analysis, we recommend that three quantitative statistics, Nash-Sutcliffe efficiency (NSE), percent bias (PBIAS), and ratio of the root mean square error to the standard deviation of measured data (RSR), in addition to the graphical techniques, be used in model evaluation. The following model evaluation performance ratings were established for each recommended statistic. In general, model simulation can be judged as satisfactory if NSE > 0.50 and RSR < 0.70, and if PBIAS + 25% for streamflow, PBIAS + 55% for sediment, and PBIAS + 70% for N and P. For PBIAS, constituent-specific performance ratings were determined based on uncertainty of measured data. Additional considerations related to model evaluation guidelines are also discussed. These considerations include: single-event simulation, quality and quantity of measured data, model calibration procedure, evaluation time step, and project scope and magnitude. A case study illustrating the application of the model evaluation guidelines is also provided.
Book
Historical background, fundamental concepts, statistical considerations and a case study emphasize the need for absolute precision in applying remotely sensed data. This book is a complete guide to assessing the accuracy of maps generated from remotely sensed data.
Article
This paper reviews the necessary considerations and available techniques for assessing the accuracy of remotely sensed data. Included in this review are the classification system, the sampling scheme, the sample size, spatial autocorrelation, and the assessment techniques. All analysis is based on the use of an error matrix or contingency table. Example matrices and results of the analysis are presented. Future trends including the need for assessment of other spatial data are also discussed.
NDRRMC Updates Sitrep No. 33 re: Effects of Tropical Depression Agaton. National Disaster Risk Reduction and Management Council
NDRRMC, 2014. NDRRMC Updates Sitrep No. 33 re: Effects of Tropical Depression Agaton. National Disaster Risk Reduction and Management Council. Retrieved February 1, 2014 from http://www.ndrrmc.gov.ph/.
SitRep No. 22 re Effects of Tropical Storm SENIANG". National Disaster Risk Reduction and Management Council
NDRRMC, 2015. SitRep No. 22 re Effects of Tropical Storm SENIANG". National Disaster Risk Reduction and Management Council. Retrieved January 10, 2015from http://www.ndrrmc.gov.ph/.
Modeling of Flashflood Events Using Integrated GIS and Hydrological Simulations
  • J R Santillan
Santillan, J. R., 2013. Modeling of Flashflood Events Using Integrated GIS and Hydrological Simulations, Training Center for Applied Geodesy and Photogrammetry, University of the Philippines, Diliman, Quezon City.