Philippine Journal of Science

147 (2): 327-342, June 2018

ISSN 0031 - 7683
Date Received: 26 Sep 2016

GIS-based Estimation of Catchment Basin Parameters

and Maximum Discharge Calculation using Rational Method
of Luinab Catchment in Iligan City

Florife dela Rama-Liwanag1,2*, Daniel Mostrales1, Kristine Sanchez1,

Ruben Tudio1, Vincent Malales1, and Ma. Teresa Ignacio1

1College of Engineering, Mindanao State University - Iligan Institute of Technology, Iligan City, Philippines
2Civil Engineering Department, College of Engineering and Architecture, University of Science
and Technology of Southern Philippines, Cagayan de Oro City, Philippines

Urban flooding is triggered when surface runoff exceeds the capacity of drainage systems,
which happens when heavy rainfall pours on to drainage areas with limited capacity, or on
medium rainfall that falls on poorly planned or maintained drainage systems. Due to accelerated
population growth and change in land use patterns, human vulnerability to floods has increased,
and it is likely to increase further unless changes are made on the urban drainage system. The
use of surface runoff and drainage modeling would help to identify- areas that are susceptible
to flooding and to determine the dynamic capabilities of urban drainage network. The use of
GIS-based software and hydrologic modeling will provide fast and reviewable assessment of the
existing drainage system of the catchment. GIS generated results – especially in the delineation
of catchment and sub-catchments and in identifying drainage networks – were validated in
the field. Field observations also showed the presence of pollutants and heavy vegetation in the
drainage system. Corrected and updated data were then used to calculate of peak discharges
using the rational method. The use of rational method in the calculation of the design peak
discharges for the catchment resulted in the following values: 21.59, 29.07, 33.82, and 40.31
m3/s for return periods of 2-yr, 5-yr, 10-yr, and 25-yr respectively. The main canal towards the
outlet of the catchment is calculated to have a maximum capacity of 7.35 m3/s. This is only 35%
of the peak discharge of the storm with a two-year return period. Improvement of the drainage
system could be achieved by a) increasing the capacity of main canal and/or b) providing an
additional outlet from identified flood-prone areas. The improvement could be further analyzed
and evaluated in succeeding hydrologic studies.

Key words: geographic information system; LiDAR; sub-catchments; urban drainage; urban
flooding; watershed

INTRODUCTION national highway going to Mandulog Bridge 1 – with runoff

coming from Barangay Luinab, part of Del Carmen and
Flooding is a major problem in Iligan in the Philippines, Upper Hinaplanon, including flows from Circumferential
where low-lying terrain and inadequate drainage lead to Road 3 – clearly affect the drainage of stormwater.
serious flooding in many areas. Low topographic gradients
and ineffective drainage channels along the secondary Urban pluvial (surface water) flooding is flooding in
urban areas caused by intense and/or prolonged rainfall
*Corresponding author: florifedelarama@gmail.com
which overwhelms the capacity of the drainage system

327
Philippine Journal of Science dela Rama-Liwanag et al.: Catchment Parameter
Vol. 147 No. 2, June 2018 Estimation and Maximum Discharge Calculation

(Ochoa-Rodríguez et al.; Simões et al. 2011). Large-scale modeling with GIS system could reduce a huge amount
urbanization and population increases have led to large of work on data formatting and processing, allowing easy
numbers of people, especially the poor and those settling interpretation of model inputs and outputs with a more
and living in floodplains in and around urban areas user-friendly map representation (Zhou 2014).
(Douglas et al. 2008). The increase of population leads
to an increase in impervious surface area, which led to For this study, GIS was used to define the catchment
increased storm water runoff (CNT 2014). basin and to identify, delineate, and generate the existing
drainage network of select barangays of Iligan City with
Urbanization restricts where floodwaters can go by the available Light Detection and Ranging (LiDAR)
covering large parts of the ground with roofs, roads, and dataset. The drainage network was validated with actual
pavements – thus obstructing natural channels – and by field measurements. Additional information was also
building drains that ensure that water moves to rivers more collected to supplement and complete the geodatabase
rapidly than it did under natural conditions (Douglas et al. for the catchment area.
2008). These changes modify the features of both urban
hydrology and hydraulics, and alter the distribution of Here, the focus is on flooding and urban drainage in
water (CNT 2014). Iligan City, Philippines, with the aim of distinguishing
potential urban flood hazards and deriving land cover for
Changes in topography may remove key surface water identifying the hydrologic and hydraulic characteristics
storage areas and thus push urban runoff into previous of the catchment in order to prepare a model to simulate
unflooded locations. Similarly, topographic changes can flooding. GIS technology is introduced to delineate the
alter flow paths by either sending excess surface water variation of topography. GIS hydrological modeling is
to undesired locations, or diverting problematic flows applied to delineate the basic hydrological elements from
to locations where there is scope for better surface water a Digital Elevation Model (DEM).
management (Diaz-Nieto 2008).
Increasingly heavy rainfall events are putting additional Case Study: The Luinab Catchment
strain on the deteriorating drainage systems (CNT 2014). The study area is a 4.531 km2 catchment located within
Water flowing through a series of culverts and concrete Iligan City (Figure 1) being served by the Luinab Creek,
channels cannot adjust to changes in the frequency of heavy a tributary of the Mandulog River Basin comprising
rain, as natural streams do. They are often obstructed by silt residential areas on the lower portion, and woodlands on
and urban debris, particularly when houses are built close the upper portion of the catchment.
to the channels. Such situations frequently arise when poor
people build on low-lying floodplains, over swamps, or
above the tidewater level on the coast (Douglas et al. 2008).
METHODOLOGY
The analyses of flood problem in the catchment area and
the development of the technical solution at the local The methods used in the study involved the following
scale form stepping-stones towards sustainable drainage steps: data collection, preparation, assessment, and
system in our locality. Within the flood management processing of the data collected. Figure 2 shows the
process, data acquisition refers to the compilation of conceptual framework of the study.
existing data and the collection of additional data for
system analysis, modeling, and decision making (Bin LiDAR data collection and Digital Terrain Model
Abdullah 2012), which is of the utmost importance for (DTM)
making cost-efficient investment and operational or Available LiDAR datasets used in the study include the
maintenance decision (Hénonin et al. 2010; Bin Abdullah LiDAR point cloud available in classified LiDAR Aerial
2012). Data acquisition provides a database or references Survey (LAS), 1-m resolution aerial imagery, and DEM.
to be used for future improvement of the drainage system. The LiDAR data was an output of the Philippine DREAM
It provides pre- and post-model geographic information Program of the UP Training Center for Applied Geodesy
systems (GIS) layers of infrastructure features that can be and Photogrammetry (UP TCAGP), a program funded
later used to advance asset management initiatives and to by the Department of Science and Technology (DOST).
increase accessibility to consolidated hydraulic models
that can be used for numerous visualization and analysis Laser points were collected over the study area using an
applications (Muir). Optech Airborne Laser Terrain Mapper (ALTM) Gemini
and Pegasus LiDAR system set to acquire points at
The application of GIS contributes to a simple an average density of two points per m2. The reported
understanding of the catchment area (Hénonin et al. accuracy of the LiDAR data is described at 1 m root-mean-
2010; Bin Abdullah 2012) and integrating drainage

328
Philippine Journal of Science dela Rama-Liwanag et al.: Catchment Parameter
Vol. 147 No. 2, June 2018 Estimation and Maximum Discharge Calculation

Figure 1. Catchment basin boundary.

Figure 2. Conceptual framework of the study.

square error (RMSE) horizontal and 15 cm vertical. The Table 1. DEM metadata (DREAM 2013).
survey was conducted on 27-28 Apr 2013. Table 1 shows
Projection: WGS84 UTM Zone 51
the metadata of the DEM.
Resolution: 1m
Tile Size: 1km by 1km
Watershed Delineation
The DEM was delineated adopting the watershed Date of Acquisition: April 27-28, 2013

329
Philippine Journal of Science dela Rama-Liwanag et al.: Catchment Parameter
Vol. 147 No. 2, June 2018 Estimation and Maximum Discharge Calculation

delineation procedure provided by the UP TCAGP (2015). determined and extracted. The outfall was also determined
Using ArcHydro tool in ArcGIS 10.2 (ESRI 2011), the from the drainage point gathered during delineation. The
DEM was delineated using the threshold value of 1% of delineated catchment area was used to crop out the data
the total catchment area. The steps involved in watershed from other layers used in the study, including buildings,
delineation are presented in Figure 3. road network, soil classification, and land cover map.
To create a depressionless DEM, there is a need to fill
depressions or sinks – if there is any – to avoid false Field Surveys
routings and remove small imperfections in the DEM Field surveys were conducted in Apr 2015, which was the
(Prodanović 2009). To fill the sinks, Fill Sink tool are summer dry season in this part of the Philippines. At that
available in ArcGIS, namely in the ArcHydro Toolbox time, it seldom rained and most of the rivers and canals
and in Hydrology in the Spatial Analysts Tools. were dry. Field surveys were carried out along the entire
hydrologic network of the Luinab Catchment.
To calculate flow direction, a grid must exist that was
coded for the direction in which each cell in a surface Prior to the conduct of field surveys, planning and
drains. To determine the direction where a landscape preparation were conducted. The collected maps were
drains, the direction of flow for each cell in the landscape stored together to gather topographical relationship and
is calculated. The Flow Direction in the ArcGIS grid to determine the sufficiency of the available data to be
processor finds the direction of steepest downward descent used for further data processing.
for every cell in the surface grid.
Upon checking on the existing maps collected from Iligan
To generate a drainage network, it is necessary to City Planning and Development Office (ICPDO), it was
determine the ultimate flow path of every cell on the found out that there were no metadata included with the
landscape grid. Flow accumulation is used to generate a datasets. Metadata provide digital identification and help
drainage network, based on the direction of flow of each support archiving and preservation of the resource. In the
cell. By selecting cells with the greatest accumulated flow, absence of metadata, no reference or justification of the
networks of high-flow cells were created. These high-flow quality of data can be presented.
cells should lie on stream channels and at valley bottoms.
Aside from that, when checking the attribute data of the
To determine the flow accumulation, or the number of up collected map, it was also found that the channel shape,
gradient pixels that slope toward each point in the DEM dimensions, geometry at the nodes, and channel surface
grid, Flow Accumulation in the Spatial Analyst Tools condition were not available for the study area concerned.
was selected. This is the longest process in delineating The above-mentioned missing attributes were necessary in
catchment in terms of processing time. order to provide the hydraulic parameters of the drainage
network that describe the hydraulic and hydrologic
After watershed delineation, the catchment basin was characteristics of the catchment, and will be used for the

Figure 3. Procedure for delineation.

330
Philippine Journal of Science dela Rama-Liwanag et al.: Catchment Parameter
Vol. 147 No. 2, June 2018 Estimation and Maximum Discharge Calculation

hydraulic and hydrologic analysis of the catchment. were processed using the eCognition Developer software
(Trimble Inc. 2010). By utilizing eCognition Developer,
The delineated catchment basin and drainage network the object-based image analysis produced a land use map
were validated and the missing attributes were collected from the 1-m resolution LiDAR datasets.
during field surveys. Prior and after the conduct of field
data gathering, coordination meetings were conducted on A set of land cover classification was employed in the
barangays Hinaplanon and Luinab, Iligan City. study. Five classes were classified, including the built-up
areas, fallow, grassland, road, and trees. Using chessboard
Equipment used included handheld Global Positioning segmentation with 1-m tile size was possible for the first
System (GPS), steel tape, measuring rod, pen, field data level segmentation of layers.
sheets, and camera for documentation.
A rule set was created in processing object-based image
After field surveys, the waypoints gathered using GPS was classification. Using chessboard segmentation with 1-m tile
post processed in ArcGIS 10.2 (ESRI 2011). The waypoints size, the layers were segmented. Then, road and building
gathered, in GPS exchange format, was converted into ESRI objects were defined from the thematic layers. Afterwards,
shapefile. The shapefile is defined at Geographic Coordinate multi-resolution segmentation was conducted for the
System (GCS) 1984 and was projected to Universal creation of objects bigger than 100 pixels, or 100 m2.
Transverse Mercator (UTM) World Geodetic Survey (WGS)
1984 Zone 51N. Afterwards, attributes were added to the Test and Training Area (TTA) masks (Definiens Inc.
shapefile, including channel shape, dimensions, geometry 2003) were created from sample objects of different
at the nodes, and channel surface condition. classes for accuracy assessment. Using an error matrix,
accuracy assessment was conducted on the object-based
The results of field surveys showed that the drainage classification to determine whether the result attained the
networks were composed of concrete pipes and open desired level of accuracy. For this study, with land cover
channels with embankments made from riprap, masonry, classification with 1-m LiDAR DTM and orthophoto,
concrete, and natural soil. the recommended Overall Accuracy and Kappa Index of
Agreement (KIA) is more than 90%.
Land Cover Derivation
Land cover maps were derived using object-based image Hydrological Data Preparation and Analysis
analysis (OBIA). Layers necessary for image processing There were five rain gauge stations (Table 2) surrounding
were prepared, including LiDAR derivatives: intensity, the catchment area. These include the rainfall stations
number of returns, Digital Terrain Model (DTM), and located in Pugaan C3 Bridge, Rogongon, Brgy. Digkilaan,
normalized Digital Surface Model (nDSM); and thematic Abuno and Kapatagan. Rainfall data from these rain gauge
layers: buildings footprint, and road buffer shapefiles. stations were collected, and the assessment showed that
Building footprints and road buffers were manually the recorded coordinates of the rain gauge stations did not
digitized. The road network was acquired as polyline point to their true locations. This led to the identification
shapefile from ICPDO. The polyline was rectified – lines of the correct coordinates of the rain gauges selected.
that did not align were edited individually. Buffers were The downloaded rainfall data from the Department
created in ArcGIS 10.2 (ESRI 2011) to represent the of Science and Technology - Advanced Science and
approximate width of the road feature. The buffer for Technology Institute [DOST-ASTI] (2015) contains
road ranges from 0.5 m to 3.0 m. Roads, whether paved information including the region, province, location, X
or gravel, were considered as one feature class. and Y coordinate, sensor name, and the rainfall intensity
LiDAR derivatives, high-resolution digital orthophoto (mm) and pressure (hPa) in 15 min increments. Rainfall
image, and the building footprints and road network records for these stations were recorded in Figure 4. The

Table 2. List of rainfall stations near Luinab Catchment.

Station Name Latitude Longitude Year of Record Maintained By Elevation Above msl (m)
Pugaan C3 Bridge 8.225317 124.2682 2013 - 2015 PAGASA 22.304
Rogongon 8.232697 124.419372 2013 - 2015 PAGASA 0
Brgy. Digkilaan 8.24628611 124.32335 2003, 2010 -2015 PAGASA 0
Abuno 8.18225282 124.2504003 2011 - 2015 PAGASA 0
Cagayan de Oro 8o29’ 124o38’ 1908 - 1999 PAGASA 6

331
Philippine Journal of Science dela Rama-Liwanag et al.: Catchment Parameter
Vol. 147 No. 2, June 2018 Estimation and Maximum Discharge Calculation

Figure 4. Rainfall records for rain gauge stations near Luinab Catchment.

collected records showed that only three years of rainfall T = duration (min); and
data were available for the four rain gauge stations, for A, b, C = constants obtained from curve fitting.
the years 2012 to 2014.
Philippine Atmospheric Geophysical and Astronomical Constants A,b, and C were constants obtained from the
Services Administration (PAGASA) operates/maintains curve fitting that vary for locations and different return
synoptic stations equipped with automatic rainfall gauge. periods.
Figure 5 indicates location of these synoptic stations and
thiessen polygon for each station, which are recommended RIDF equations, both for short (10 min to 1-hr) and
to be used for runoff analysis. When one of the stations is long (1-hr to 1-day) duration, were derived for all return
located inside or near the target river basin, the Rainfall periods – 2-yr, 5-yr, 10-yr, 25-yr, 50-yr and 100-yr – for
Intensity-Duration-Frequency (RIDF) of this station is the Cagayan de Oro synoptic station (Figure 6).
used to obtain rainfall intensity during the concentration Often, historical rainfall data is not available for analysing
time (DPWH & JICA 2003). rainfall-runoff processes. In other cases, historical rainfall
The study area belonged to the Cagayan de Oro synoptic records are available but with a considerable amount of
station. The RIDF curve for Cagayan de Oro Synoptic missing data. A substantial amount of data is missing from
Station from the report conducted by Asian Development the rainfall information collected. This presents difficulty
Bank [ADB] (2008) was collected for runoff analysis. when analyzing and modeling runoff.

RIDF curves were produced by converting the probable

Hydraulic and Hydrologic Parameter
rainfall magnitudes into rainfall intensities and fitting
Information regarding soil types and their drainage
them to an exponential equation. Two types of exponential
characteristics was collected and tabulated to assign
equation are tried to fit the rainfall intensity duration.
hydraulic parameters and coefficients to the catchment.
These RIDF equations were expressed as follows:
These parameters included the Green-Ampt infiltration
Type 1: R = A/(C + Tb) (1) parameters, permeability, Manning’s coefficient for
Type 2: R = A / (C + T)b (2) channels and overland flow, entrance and exit loss
coefficients.
Where:
R = rainfall intensity (mm/hr);

332
Philippine Journal of Science dela Rama-Liwanag et al.: Catchment Parameter
Vol. 147 No. 2, June 2018 Estimation and Maximum Discharge Calculation

Figure 5. Thiessen polygon of RIDF locations (DREAM 2015).

Calculation of Runoff Capacity computed based from the generated land use map. The
The maximum capacity of the drainage channel in the time of concentration was calculated from the outfall to
outfall portion of the catchment was calculated using the the most remote point of the catchment. The probable
rational method. The composite runoff coefficient was rainfall intensity was identified from the RIDF curve

333
Philippine Journal of Science dela Rama-Liwanag et al.: Catchment Parameter
Vol. 147 No. 2, June 2018 Estimation and Maximum Discharge Calculation

Figure 6. RIDF curve for Cagayan de Oro City (ADB 2008).

for the catchment area. The design peak discharges were Table 4. Runoff coefficients applied for the study.
computed using rational equation. Design hyetographs for
Land Cover Type Runoff Coefficient C
the catchment area were prepared from the RIDF curve
and area reduction curve. Built-up 0.55
Fallow 0.30
Runoff Coefficient. In the Philippines, DPWH was
Grass 0.25
established as the Ministry of Public Works and Highways
(MPWH) at the time when tabulated runoff coefficients Trees 0.55
were published in the “Design Guidelines Criteria and Road 0.75
Standards, Vol. 1 MPWH 1987 (Red Book)”, (Table 3).
The Red Book was used as reference in the assignment
of runoff coefficient values for the study area (Table 4). The use of average values was adopted to simplify the
determination of runoff coefficients. Where a drainage
Table 3. Runoff coefficients (MPWH 1987). area is composed of subareas with different runoff
Surface Characteristics Runoff Coefficient coefficients, a composite coefficient for the total drainage
area is computed by dividing the summation of the
Lawn, gardens, meadows and cultivated 0.05 – 0.25
lands products of the subareas and their coefficients by the
total area:
Parks, open spaces including unpaved 0.20 – 0.30
surfaces and vacant lots
Suburban districts with few buildings 0.25 – 0.35 (3)
Residential districts not densely built 0.30 – 0.55
Residential districts densely built 0.50 – 0.75 Time of Concentration. The time of concentration (tc)
Watershed having steep gullies and heavily 0.55 – 0.70
for the catchment area under consideration is obtained
timbered based from the formula presented in the “Manual On
Watershed having moderate slope, 0.45 – 0.55
Flood Control Planning” by the DPWH & JICA (2003).
cultivated and heavily timbered
Suburban areas 0.34 – 0.45 (4)
Agricultural areas 1.15 – 0.25

334
Philippine Journal of Science dela Rama-Liwanag et al.: Catchment Parameter
Vol. 147 No. 2, June 2018 Estimation and Maximum Discharge Calculation

where: ti = inlet time; time it takes to flow from the Peak Discharge. After the composite runoff coefficient,
nearest point to the inlet point or farthest point of river probable rainfall intensity, and area were calculated,
channel. the peak discharges for different design storms were
calculated using the rational equation presented based on
tf = flow time; time it takes from the inlet point or farthest the Rational Formula:
point of the river chanel to the outlet point or point under
consideration.
(7)
= L/V (5)
L = length of river channel from its outlet point to its in which:
farthest point (m)
Q = the maximum rate of runoff, m3/s
V = flow velocity (m/s)
C = a runoff coefficient that is the ratio between the
Since the sub-catchment area of the farthest point of the runoff volume from an area and the average rate of
channel is less than 2 km2, the inlet time – the time it takes rainfall depth over a given duration for that area
to flow from the nearest point to the inlet point or farthest
I = average intensity of rainfall, mm/hr for a
point of river channel – was calculated by the equation:
duration equal to the time of concentration, tc
A = drainage area, km2
(6)
Calculation of the Outfall Discharge Capacity
The capacity of a channel is the maximum discharge
Table 5. Kraven’s Formula.
that can be safely be transported. It is based from the
Ib > 1/100 Ib< 1/200 Manning’s equation:
Riverbed 1/100 >Ib
gradient (Ib) > 1/200
(steep slope) (mild slope)
Flow velocity (8)
3.5 3.0 2.1
(m/s)

Where:
The flow time – the time it takes from the inlet point or
farthest point of the river channel to the outlet point or n = Manning’s coefficient (dimensionless)
point under consideration – is calculated from Kraven’s A = flow cross sectional area
Formula (Table 5), which gives relations between slope R = hydraulic radius
of water course and flow velocity. S = slope

To determine the length of the main channel from a point The hydraulic radius is the ratio of flow cross sectional
of interest to the most remote part of the catchment area, area (A) and wetted perimeter. Outfall profile was
several “remote points” on the boundary line of the sub- gathered from the available DEM.
catchments were created. The distance of water traverse
was estimated by creating a possible path from the “remote
points” towards the point of interest. Distances were
tabulated, and the longest length was extracted from the RESULTS
tabulation. Elevation of the most remote point and the Based on the methods used, the watershed and
point of interest were extracted from the LiDAR DEM. its characteristics were established, including its
The head was computed by taking the elevation difference catchment basin boundary and drainage line, land cover
of these points. Slope gradient was determined from the characteristics, hydraulic and hydrologic parameters, and
computed head and the length of the channel. the peak discharge.
Rainfall Intensity. The rainfall intensity i is the average The catchment basin was delineated using the available
rainfall rate in mm/hr with the decided safety level indicated DTM, and the catchment boundary and drainage line were
in the form of return period for the considered catchment generated, as shown in Figure 7. It was then validated
area during the concentration time. Using the RIDF curve, from the existing maps and in the field. Results of field
the probable rainfall intensities for short duration rainfall for surveys (Figure 8) showed the presence of false positive
2-yr, 5-yr, 10-yr, and 25-yr return periods were calculated and false negative drainage lines and the difference in area
based from the computed time of concentration.

335
Philippine Journal of Science dela Rama-Liwanag et al.: Catchment Parameter
Vol. 147 No. 2, June 2018 Estimation and Maximum Discharge Calculation

Figure 7. Delineated catchment and drainage line.

Figure 8. Catchment and drainage network and waypoints collected from field survey.

336
Philippine Journal of Science dela Rama-Liwanag et al.: Catchment Parameter
Vol. 147 No. 2, June 2018 Estimation and Maximum Discharge Calculation

between the delineated and the validated catchment, which stream network and the catchment boundary.
led to the adjustment of the catchment area boundary and
drainage line. In addition, during field validation, it was found out
that there were existing false positive and false negative
Initially, the catchment has an area of 5.483 km 2 drainage lines. False positives are delineated drainage
comprising nine (9) sub-catchments. After validation, is lines that were actually non-existent in reality, or
has an area of 4.531 km2, and sixteen (16) sub-catchments, depressed areas where water flows intermittently during
with a difference of 0.95 km2. Figure 9 shows the final rainy season. On the contrary, false negatives were
catchment and drainage network. drainage lines not defined during delineation but existing
on site. These drainage lines are usually man-made canals
On the lower portion the catchment, it was found out that have width less than the DEM resolution and/or
that there was a closed drainage channel where water underground or covered drainage networks. Drainage
also exits at the outfall of the catchment. During the lines were rectified, false negatives were incorporated,
watershed delineation, it was not identified as part of the and false positives were deleted in the final drainage lines.
stream network and the sub-catchment that it serves is not
included in the delineated catchment boundary. However, Drainage networks for the concerned areas were composed
during field validation, it was found out that the covered of reinforced concrete pipe culvert (RCPC) with diameters
channel and the area it serves is part of the catchment. The ranging from 0.3 m to 1.2 m; open channels with
channel and the area it serves were digitized and added embankments made from riprap, masonry, concrete, and
into the final stream network and catchment boundary. natural soil; and intermittent and perennial channels that
are mostly dry during summer.
There is also a portion in the catchment where it was
delineated as within the the catchment boundary. Due to variable topography at the upper portion of the
However, upon field validation, it was found out that catchment, the result of the delineation is more accurate,
due to the installed closed channels, water was being compared to the downstream portion, where there are
redirected directly to the Mandulog River. The stream flat terrain characteristics and man-made drainage lines.
network and the area where the water was redirected
towards the Mandulog River were taken out of the final In the upper portion of the catchment, most of the drainage

Figure 9. Final catchment and drainage network map.

337
Philippine Journal of Science dela Rama-Liwanag et al.: Catchment Parameter
Vol. 147 No. 2, June 2018 Estimation and Maximum Discharge Calculation

lines were natural perennial channels. These channels [PhilGIS], was collected for the study. Upon checking
were shallow and wide, consisted of gravel and loose the available land use map, it was found out that there
soil channel bed and grass on its sides. As the channel are portions where the assigned land use class did not
goes downstream, there were portions that already had represent the current land cover of the area. An updated
concrete sides. In addition, man-made channels were also land cover map based on the 2013 LiDAR data was
constructed. There were also portions along the catchment derived and used in assigning hydrologic parameters of
where the width on the upstream portion is bigger than the study.
the downstream portion.
Using error matrix, the processing of land cover classification
At this time, most of the water flowing in the drainage arrived to an overall accuracy of 0.978 and KIA of 0.969.
channels came mainly from the household waste Catchment basin and drainage network properties were
discharges. Most of the open channels were dry and derived from digital representations of the man-made
clogged with garbage and there were also channels that channels and natural surface stream channel network.
were heavily vegetated and with presence of eroded The land surface properties were calculated based on data
materials, garbage, and stagnant water. The RCPC installed derived from the land cover map in Figure 10.
in the area seems to be in good condition, although there
are some that need some maintenance. The widespread
Hydraulic and Hydrologic Parameter
presence of pollutants in the drainage system reflects a
The catchment basin is composed of Bolinao clay
poor and inadequately maintained drainage network. The
loam. The Philippine Rice Research Institute [PhilRice]
pollutant accumulation, if not dealt with, would lower the
(2013) described Bolinao clay loam as an old soil that
quality of drainage waters. When the rainy season comes,
has undergone extensive weathering but has retained a
this debris will partly block drains and channels and so
high base status in its horizon. This is a representative of
reduce their capacity to evacuate stormwater.
the great soil group Paleustaff that has a base saturation
of <75%. This can be found in areas with pronounced
Derived Land Cover wet and dry seasons. The Bolinao clay loam has high
The land use map sourced by Fuentes (2004), which water retention, moderate to good drainage, moderate
was available in the Philippine GIS Data Clearinghouse permeability, workability, erosion, and stoniness included

Figure 10. Derived land cover map.

338
Philippine Journal of Science dela Rama-Liwanag et al.: Catchment Parameter
Vol. 147 No. 2, June 2018 Estimation and Maximum Discharge Calculation

gravels; limestone; iron (Fe) and manganese (Mn) Table 7. Percent IA and percent DCIA.
concretions. Brakensiek and Onstad (1977) determined
Percent of Percent Percent
the average permeability for clay loam with a value of 0.8. Land Cover Type
Total Area Impervious DCIA
Table 6 presented the parameters considered for the study.
Built-Up 8.3723 8.37 3.14
The equivalent Manning’s coefficient values were Fallow 20.5554 - -
assigned to each type of drainage network material. Frost
Grass 8.6569 - -
(2006) presented loss coefficient values for every entrance
type, and exit loss coefficient with a value of 1.0. These Trees 56.8971 - -
values are used for the study. Road 4.2019 4.20 2.54
Canals 1.3163 1.32 1.32
Table 6. Drainage hydraulic parameters. All Land Use 100.00 13.89 6.99
Description Values Unit Categories

Green-Amp Infiltration
Ave. Capillary Suction 298.8 mm Table 8. Calculated composite C values.
Saturated Hydraulic Conductivity 2 mm/hr Sub-catchment Area (km2) Composite C
Initial Moisture Deficit for Soil 0.267 C01 0.478 0.475
Permeability 0.8 cm/hr C02 0.761 0.531
Manning’s Coefficient C03 0.202 0.468
Concrete pipe 0.013 C04 0.894 0.474
Concrete Canals 0.013 C05 0.407 0.490
Open Channel – regular section – C06 0.226 0.520
dense grass and weeds, depth of
C07 0.092 0.447
flow materially greater than
weed height 0.04 C08 0.268 0.448
Entrance Loss Coefficient C09 0.142 0.480
Inward Projecting 0.8 C10 0.225 0.401
Sharp-Cornered 0.5 C11 0.160 0.449
Slightly Rounded 0.25 C12 0.042 0.433
Bell-Mouthed 0.05 C13 0.122 0.493
Exit Loss Coefficient 1.0 C14 0.097 0.453
C15 0.310 0.488
C16 0.105 0.489
Table 6 shows the assignment of drainage hydraulic
Catchment 4.531 0.482
parameters for infiltration, permeability, Manning’s
coefficient, and entrance and exit loss coefficients for
the clay loam soil.
With a total area of 4.531 km2, the composite runoff
For the overland flow, the impervious surface values range coefficient for the catchment area has a value of 0.482.
from 0.25 to 0.75, while the pervious surfaces values The tabulated values of the composite coefficients for
range from 0.05 to 0.25. The percent impervious area (IA) each sub-catchment are presented in Table 8.
and percent directly connected impervious areas (DCIA)
Based on the calculations, the sub-catchment with
were also calculated per sub-catchment. These parameters
the highest coefficient runoff coefficient value is
shall be used for the assignment of values for the sub-
sub-catchment C10, with the value of 0.401. The sub-
catchment. Table 7 presents the result of the calculation.
catchment with the lowest value of runoff coefficient is
sub-catchment C02 with the value of 0.531.
Calculation of Runoff Capacity
The individual areas for the defined land cover classes The runoff coefficient C is a dimensionless coefficient
were tabulated and assigned with runoff coefficients. relating to the amount of runoff to the amount of
Composite coefficients were calculated for each sub- precipitation received. It is larger value for areas with low
catchment and for the whole catchment basin. infiltration and high runoff (pavement, steep gradient), and
lower for permeable, well vegetated areas. This suggest

339
Philippine Journal of Science dela Rama-Liwanag et al.: Catchment Parameter
Vol. 147 No. 2, June 2018 Estimation and Maximum Discharge Calculation

that at sub-catchment C10, a greater portion of the area is development and applications. This data need was
fallow and grass land, and only a small portion consists highlighted in this study.
of concrete roads and built up areas. On the other hand,
sub-catchment C02 is located in the upper portion of the Given that the latest database available in the local
catchment that contains areas of dense tree cover. government for drainage modeling were outdated during
the conduct of the study, information used for the study
The outfall of the catchment area, which serves as the were collected from several data sources and organized
outlet of water from the highest elevation and the most in a geodatabase.
remote point within the catchment area, was chosen
as the point of interest for time of concentration (Tc) The use of GIS technology in delineating catchment
computation. There are fourteen (14) hydraulically network increased the accuracy of determining the flow
“remote points” created within the catchment boundary paths and the catchment and sub-catchment boundaries.
lines. These points are located along the upper portion of The output derived from this study is very much helpful
the catchment area. in the preparation of the hydrologic and hydraulic data to
be used for the modeling and analysis of catchment. With
The possible path that water flows from these “remote the use of GIS and LiDAR datasets, the study was able
points” towards the outlet were traced based from the to define the catchment basin, and identify and generate
LiDAR DTM. Line segments were created for each the drainage network of select Barangays of Iligan City,
path, ending where the common path these lines pass. namely Barangays Luinab, Del Carmen, Hinaplanon,
The lengths of each route were computed by summing parts of Sto. Rosario, Puga-an, Upper Hinaplanon, and
the length of the lines that water passed through from its Mandulog.
location towards the point of interest. From the fourteen
(14) “remote points” created, there are eighteen (18) The field surveys conducted validated the results of
possible routes that water flow from point of interest to delineation and established the catchment boundary and
the most remote part of the catchment area. Out from these drainage lines. Actual field measurements supplemented
routes, it was determined that the maximum length L of the missing data needed for the calculation of peak
main channel from point of interest to the most remote discharge and flood modeling.
point within the catchment is 5,064.33 m. From this With the use of GIS, field measurements and information
longest path, the time of concentration was calculated. from existing maps were organized and converted into
The time of concentration for the catchment was attributes. This allowed the organization of the catchment
determined to be 42.66 min, with inlet flow time equal to and drainage lines characteristics.
18.50 min and flow time equal to 24.16 min. Land cover map was derived and used to identify the
Incorporating the computed time of concentration of 42.66 effects of imperviousness in the hydrologic behavior of
min, the values of probable intensity i for storms 2-yr, 5-yr, the catchment. With the land cover characteristics, the
10-yr, and 25-yr have values of: 58.49, 78.68, 91.51, and pervious and impervious layers were converted into values
109.14 mm, respectively. and used for the calculation of peak discharge.

The peak discharge calculated from the rational equation The use of published hydrologic information such as
and tabulated in Table 9. The values of the maximum the RIDF curve proved to be useful in this study, which
peak discharge are 20.76, 27.92, 32.47, and 38.73 m3/s for supplemented the lacking rainfall data. Series of processes
storm durations 2-yr, 5-yr, 10-yr, and 25-yr, respectively. from sound scientific methodologies were collected,
conducted, and developed results that suit the study needs.
The outfall discharge was calculated using Manning’s
equation, and arrived to 7.35 m3/s. Employing detailed land use/land cover classification
from the 2013 LiDAR data was very helpful in calculating
the peak discharge for the catchment area. The percent
impervious areas (IA) and directly connected impervious
areas (DCIA) was estimated based from the generated
CONCLUSIONS
land use/ land cover map from 2013 LiDAR data and the
The study assessed the prevailing status of the drainage available high-resolution aerial imagery.
network of the selected catchment area. Flood simulation
required detailed physical representation of the surface The peak discharge calculated from the rational equation
areas within the drainage system to accurately describe and tabulated in Table 10. The values of the maximum
flow paths. If such detailed data is not available, it will peak discharge are 20.76, 27.92, 32.47, and 38.73 m3/s for
restrict the analyses of the drainage network for future storm durations 2-yr, 5-yr, 10-yr, and 25-yr, respectively.

340
Philippine Journal of Science dela Rama-Liwanag et al.: Catchment Parameter
Vol. 147 No. 2, June 2018 Estimation and Maximum Discharge Calculation

Table 9. Maximum peak discharge. BIN ABDULLAH AF. 2012. A methodology for processing
raw Lidar data to support urban flood modeling
Return Period Maximum Peak Discharge (Q) (m3/s)
framework [Dissertation]. Delft, the Netherlands: Delft
2-YR 20.76 University of TechnologyCRC Press/Balkema. The
5 - YR 27.92 Netherlands. ISBN 978-0-415-62475-6.
10 - YR 32.47
BRAKENSIEK DL, ONSTAD CA. 1977. Parameter
25 - YR 38.73 estimation of the Green and Ampt Infiltration
Equation. Water Resources Research. doi: 10.1029/
WR013i006p01009
The outfall discharge was calculated using Manning’s
equation, and arrived to 7.35 m3/s. [CNT] Center for Neighborhood Technology. 2014.
The Prevalence and Cost of Urban Flooding: A Case
Using rational method calculation, it was determined Study of Cook County, IL. Retrieved from http://
that Luinab catchment, with the probable intensity i www.cnt.org/sites/default/files/publications/CNT_
for storms 2-yr, 5-yr, 10-yr, and 25-yr have values of PrevalenceAndCostOfUrbanFlooding2014.pdf
58.49, 78.68, 91.51, and 109.14 mm, respectively. The
maximum peak discharge values are 20.76, 27.92, 32.47, Definiens Inc. 2003. eCognition object oriented image
and 38.73 m3/s for storm durations 2-yr, 5-yr, 10-yr, and analysis user guide. Definiens Inc., Munchen, Germany
25-yr, respectively. The main canal towards the outlet of DIAZ-NIETO J, BLANKSBY J, LERNER DN, SAUL
the catchment is calculated to have a maximum capacity AJ. 2008. A GIS approach to explore urban flood risk
of 7.35 m3/s. This is only 35% of the peak discharge of management. Proceedings of the 11th International
the storm with 2-yr return period. Improvement of the Conference on Urban Drainage; 2008 August 31 -
drainage system could be achieved by a) increasing the September 5; Edinburgh, Scotland, UK. Retrieved from
capacity of main canal and/or b) providing an additional https://web.sbe.hw.ac.uk/staffprofiles/bdgsa/11th_
outlet from identified areas that are prone to flooding. The International_Conference_on_Urban_Drainage_CD/
improvement could be further analyzed and evaluated in ICUD08/pdfs/653.pdf
succeeding hydrologic studies.
[DPWH] Department of Public Works and Highways,
[JICA] Japan International Cooperation Agency. 2003.
Manual on flood control planning. DPWH Publishing,
Manila, Philippines. Retrieved from http://www.jica.
ACKNOWLEDGMENT
go.jp/project/philippines/0600933/04/pdf/Manual_on_
This work was supported by the Department of Science and FC_ Planning.pdf
Technology of the Philippines through the Phil-LiDAR 2
[DOST-ASTI] Department of Science and Technology-
project of MSU – IIT. The LiDAR dataset for this research
Advanced Science and Technology Institute. 2015.
was kindly provided for by the Philippine DREAM
Daily Rainfall Data 2010 - 2015.
Program, road network from PhilGIS, and drainage network
from Iligan City Planning and Development Office. Special DOUGLAS I, ALAM K, MAGHENDA M, McDONNELL
thanks are given to the project partners in University of Y, MCLEAN L, CAMPBELL J. 2008. Unjust waters:
the Philippines – Diliman Phil-LiDAR 2 (Melanie Gaspa climate change, flooding and the urban poor in
and Dominic Aloc), research co-workers Anacita Tahud, Africa. Environment and Urbanization 20(1): 187.
Al Mon Dahan, Oliver Arriesgado, and Civil Engineering http://www.sagepublications.com/. International
students of MSU – Iligan Institute of Technology. Institute for Environment and Development. doi:
10.1177/0956247808089156
[ESRI] Environmental Systems Research Institute.
2011. ArcGIS Desktop: Release 10. Redlands, CA:
REFERENCES Environmental Systems Research Institute.
[ADB] Asian Development Bank. 2008. Philippines:
FROST W. 2006. Minor Loss Coefficients for Storm
Master plan for the Agusan River basin. ADB Technical
Drain Modeling with SWMM. KCI Technologies, Inc.
Assistance Consultant’s Report. 42p. Prepared by
Retrieved from https://www.chijournal.org/Journals/
CTI Engineering International Co. Ltd, Halcrow, &
PDF/R225-23. doi: 10.14796/JWMM.R225-23
Woodfiled Consultants, Inc. Philippines. Retrieved
from http://www.adb.org/sites/default/files/project- FUENTES X. 2004. Land Cover 2004-07. [PhilGIS]
document/68233/36540-phi-tacr.pdf Philippine GIS Data Clearinghouse. Retrieved from
http://philgis.org/thematic-maps/land-cover

341
Philippine Journal of Science dela Rama-Liwanag et al.: Catchment Parameter
Vol. 147 No. 2, June 2018 Estimation and Maximum Discharge Calculation

HÉNONIN J, RUSSO B, ROQUETA DS, SANCHEZ-

DIEZMA R, STO. DOMINGO ND, THOMSEN F,
MARK O. 2010. Urban flood real-time forecasting
and modeling: a state-of-the art review. Proceedings
of MIKE by DHI Conference; 2010 September 6-8;
Copenhagen, Denmark.
[MPWH] Ministry of Public Works and Highways. 1987.
Design Guidelines Criteria and Standards, Vol. 1.
MUIR R. (undated). GIS applications in urban drainage
master planning. Retrieved from http://xpsolutions.
com/assets/dms/cs-gisappsinurbandrainagemp.pdf
OCHOA-RODRÍGUEZ S, SMITH KM, AIVAZOGLOU
M, PINA R, MIJIC A. (undated). Urban pluvial
flooding and climate change: London (UK), Rafina
(Greece) and Coimbra (Portugal). Retrieved from
https://www.imperial.ac.uk/grantham/our-work/
impacts-and-adaptation/ipcc-working-group-ii/water-
security-and-flood-risk/urban-flooding/
[PhilRice] Philippine Rice Research Institute. 2013.
“Simplified Keys to Soil Series Pangasinan”. Soil
Series Guidebook ISBN 978-971-9081-81-4:46p.,
August 2013. 17p. Retrieved from http://www.philrice.
gov.ph/wp-content/uploads/2015/08/Simplified-Keys-
to-Soil-Series-Pangasinan.pdf
PRODANOVIĆ D, STANIĆ M, MILIVOJEVIĆ V,
SIMIĆ Z, ARSIĆ M. 2009. DEM based GIS algorithms
for automatic creation of hydrological models data.
Journal of the Serbian Society for Computational
Mechanics 3(1): 64-85. Retrieved from http://www.
sscm.kg.ac.rs/jsscm/downloads/Vol3No1/DEM_
Based_GIS_ Algorithms_for_Automatic_Creation_of_
Hydrological3.pdf.udc: 556.09:528.9]:004.6
SIMÕES N, OCHOA S, LEITÃO JP, PINA RA, SÁ
MARQUES S, MAKSIMOVIĆ Č. 2011. Urban
drainage models for flood forecasting: 1D/1D,
1D/2D and hybrid models. Proceedings of the 12th
International Conference on Urban Drainage; 2011
September 11-16; Porto Alegre, Brazil.
Trimble Inc. 2010. eCognition Developer. Trimble Germany
GmbH, Trappentreustr. 1, D-80339 München, Germany.
[UP TCAGP] UP Training Center for Applied Geodesy
and Photogrammetry. 2015. Flood Forecasting and
Flood Hazard Mapping for Mandulog River Basin,
Disaster Risk Exposure and Assessment for Mitigation
(DREAM), DOST Grants-in-Aid Program, 56p.
ZHOU Q. 2014. A review of sustainable urban
drainage systems considering the climate change
and urbanization impacts. Licensee MDPI, Basel,
Switzerland. Retrieved from www.mdpi.com/2073-
4441/6/4/976/pdf

342