Flood hazard comparison based on geomorphic flood index

and hydraulic HEC-RAS (Case study in Ciliwung Watershed,

Jakarta)

A P Pratiwi1, F I W Rohmat2, E O Nugroho2, M Farid2 and M S B Kusuma2

1
Water Resource Management Master Program, Faculty of Civil and Environmental
Engineering, Bandung Institute of Technology, Bandung 40135, Indonesia
2
Water Resources Engineering Research Group, Faculty of Civil and Environmental
Engineering, Bandung Institute of Technology, Bandung 40135, Indonesia

*Email: anneputri17@gmail.com

Abstract. The Jakarta Special Capital Region is a highly vulnerable area to floods due to its
location on wetlands laced by 13 major rivers and borders the Java Sea, with more than 40% of
its land below sea level and groundwater extraction leading to the ground sinking. The city has
experienced several major flood disasters, and climate change has increased the risk, frequency,
and severity of flooding in Jakarta. The present study aims to identify areas with flood potential
in the Jakarta Special Capital Region (Provinsi DKI Jakarta) using the Geospatial Flood Index
(GFI) method recommended by the National Disaster Management Agency (BNPB). The GFI
method is an alternative rapid assessment that utilizes the Digital Elevation Model for National
Spatial (DEMNAS) data with an 8-meter spatial resolution and employs ARCGIS and QGIS
software to identify areas with high potential for flooding and the extent of inundation. To assess
the effectiveness of this method, a comparison is made with the Hydraulic HECRAS model for
the section from the Automatic Water Level Recorder (AWLR) MT. Hartono to the Manggarai
Flood Gate, considering flood return periods Q50 and Q100. The modeling results indicate that
the inundation area estimated by the conventional GFI modeling is 150% larger than that the
HECRAS Hydraulic model predicted.

1. Introduction
Floods are the stream of the river that overflows beyond the river's capacity and, thus, will pass through
the riverbank and inundate the surrounding area. The causes of flooding are triggered by several factors,
including natural factors such as climate change that increase the frequency, magnitude, and extreme
events, high rainfall intensity, and human factors such as urbanization that increases runoff, changes in
land use, and enhancement of settlement construction in flood-prone areas [1].
Floods in the Jakarta Metropolitan Area occur every year (see Figure 1) and inundate the entire area
of Jakarta City. The floods in 2020 were the most enormous ever. The high precipitation in the Ciliwung
Cisadane River system allegedly caused this incident. Besides, it was influenced by the small flow
capacity of the Ciliwung Cisadane River system and the slight infiltration in the Jakarta Metropolitan
Area, so it was unable to accommodate water runoff from the primary Ciliwung Cisadane River system
or its tributaries.
 Another factor that causes flooding in the Jakarta Metropolitan Area is that the topography of Jakarta
is primarily flat, with an altitude between 0-1000 m above sea level and an average slope between 0-8
%. This kind of topography strongly supports the occurrence of soil erosion processes, which carry
sediment from the upper part, which in turn sedimentation is deposited in river streams and causes silting
of the river so that the continuous capacity of the river for rainwater will cause flooding [2].
Disaster risk mapping is a spatial approach to determine the nature and level of disaster risk by
analyzing potential hazards and evaluating existing exposure and vulnerability conditions that may
endanger people, property, services, livelihoods, and the environment on which they depend [[3].
Disaster risk assessments include hazard, vulnerability, and capacity assessments [[4]. Mapping areas
with a flood hazard level must be done so the government can make the right policies to overcome them
[4].

Figure 1. Flood Frequent in Ciliwung Watershed

2. Study Area
The study area is located in the Ciliwung River System watershed. The Ciliwung River watershed covers
an area of 76 square kilometers, with a total area of 392.3 square kilometers, encompassing ten cities or
regencies, namely West Jakarta, Central Jakarta, South Jakarta, East Jakarta, North Jakarta, Bogor,
Bogor City, Cianjur, Depok city, and Sukabumi in the upstream area.
The morphology of the Ciliwung River watershed in the upstream region in the West Java province
exhibits mountainous and hilly terrain. Meanwhile, in the DKI Jakarta province, the Ciliwung River
watershed has a flat and gentle morphology. The low slope of the terrain in the DKI Jakarta province
increases the risk of flooding, particularly in densely populated and office-dense areas affected by
floods.
This research specifically focuses on the section of the Ciliwung River from the Automatic Water
Level Recorder MT. Haryono to the Manggarai Flood Gate, covering a distance of 7.9 kilometers. This
river section passes through three cities: Central Jakarta, South Jakarta, and East Jakarta. The study area
of this research is depicted in Figure 1.
 Central Jakarta

Manggarai Flood Gate

East Jakarta

South Jakarta AWLR MT. Haryono

Figure 2. Location Study

3. Defining and Terms

3.1. Flood
A flood is overflowing a stream or another body of water or accumulating water over areas not usually
submerged. Floods include river (fluvial) floods, flash floods, urban floods, pluvial floods, sewer floods,
coastal floods, and glacial lake outburst floods.” These various classes of floods are generated by
different mechanisms [5].
Suppose an event is submerged by water in an area that threatens and disrupts the life and livelihood
of the community, resulting in human casualties, environmental damage, property losses, and
psychological impacts. In that case, the flood is considered a disaster [4].
Inland flooding in Jakarta is primarily triggered by intense rainfall associated with tropical
depressions and continuous low-level wind convergence resulting from cross-equatorial northerly
surges. The situation is further aggravated by obstructed waterways, deforestation, and inadequate
drainage and flood control provisions, contributing to the worsening of inland flooding [6].

3.2. Hazard
A process, phenomenon, or human activity that may cause loss of life, injury or other health impacts,
property damage, social and economic disruption, or environmental degradation. Hazard may be caused
by natural, anthropogenic, or socio-natural origin. Natural hazards are associated with natural processes
and phenomena. Anthropogenic or human-induced hazards are induced entirely or predominantly by
human activities and choices. Floods are sociological since they combine natural and anthropogenic
factors, including environmental degradation and climate change [7]. Determines the area where certain
natural events occur with a particular frequency and intensity, depending on the vulnerability and
capacity of the area, which can cause a disaster [[8].

3.3. Hydraulic Modeling with HEC-RAS

HEC-RAS is one of the programs used to model river flow. This program was developed by the
Hydrologic Engineering Center (HEC). The equations used are as follows:
a) Mass Equation

𝜕𝐻
+ ∇. ℎ𝑉 + 𝑞 = 0 (1)
𝜕𝑡
Where:
• V (u, v) is the velocity vector (m/s)
• H is the water surface elevation (m)
• h is the depth (m)
• q is the discharge (m³/s)
• The vector components ∇ = 𝜕/𝜕𝑥, 𝜕/𝜕𝑦 are then decomposed as follows:
𝜕𝐻 𝜕(ℎ𝑢) 𝜕(ℎ𝑣)
+ + +𝑞 =0 (2)
𝜕𝑡 𝜕𝑥 𝜕𝑦

b) Mass Equation
The equation used is as follows:

𝜕𝑢 𝜕𝑢 𝜕𝑣 𝜕𝐻 𝜕2𝑢 𝜕2𝑢
+𝑢 +𝑣 =𝑔 + 𝑉𝑡 ( + ) − 𝑐𝑓 𝑢 + 𝑓𝑣
𝜕𝑡 𝜕𝑥 𝜕𝑦 𝜕𝑦 𝜕𝑥 2 𝜕𝑦 2
𝜕𝑢 𝜕𝑢 𝜕𝑣 𝜕𝐻 𝜕2𝑢 𝜕2𝑢
+𝑢 +𝑣 =𝑔 + 𝑉𝑡 ( + ) − 𝑐𝑓 𝑢 + 𝑓𝑢 (3)
𝜕𝑡 𝜕𝑥 𝜕𝑦 𝜕𝑦 𝜕𝑥 2 𝜕𝑦 2

Where:
• g is gravity (m/s²)
• 𝑉𝑡 is the eddy viscosity
• 𝑓 is the Coriolis effect
• 𝑐𝑓 is the bed friction

3.4. Geomorphic Flood Index

The Flood hazard index employed in this research utilizes the Geomorphic Flood Index method (GFI).
GFI represents a cost-effective and highly efficient approach for assessing areas susceptible to flooding,
using digital elevation model (DEM) data as the fundamental basis for the analysis[9]. The GFI
methodology was initially developed as a QGIS plugin [10]. The description of this method is pictured
in Figure 2. This technique involves the comparison of the water depth variable (hr) at each location
with the elevation differential (H) measured in meters. The value of hr is computed based on the
contribution of the accumulated flow area (Ar) in square meters from the nearest point within the
hydrologically interconnected river or drainage network[11]. By considering the estimated water level
(hr) in the closest segment of the river or drainage network, the GFI method identifies the nearest river
or drainage as a potential source of risk in terms of flood risk evaluation. In this research, the data used
comprises DEMNAS data provided by the Geospatial Information Agency, with a resolution of 8 meters.
There are several main processes in the GFI (Geomorphic Flood Index) process, namely Fill Sinks, Flow
Direction, and Flow Accumulation. Fill Sinks is a process of filling gaps where there is no outflow of
water from these gaps. The Fill Sink method has an advantage in addressing high-value anomalies in
valley areas. These gaps can be eliminated using GIS features[12]. Flow direction determines the flow
direction from each raster cell. The Flow Direction grid indicates the flow direction from each cell to
the nearest area with the steepest slope in the DEM [13]. Flow Accumulation generates flow
accumulation as the sum of all cells that flow into each cell downstream in the output raster. The flow
accumulation results can create a flow network by identifying cells with high flow accumulation
values[13].
 Figure 3. GFI Representation Of The Parameter h
and hr In Cross Section [14]

4. Materials

4.1. Flood Event Data

The data used in this study is flood event data from DSDA DKI Jakarta from the flood event in February
2017 event. The following is the data on the height of floodwaters in the Ciliwung watershed.

Table 1. Plotting Flood Verification

Plot City Inundation Depth (m)
1 South Jakarta 0.7
2 South Jakarta 0.4
3 South Jakarta 0.4
4 South Jakarta 0.5
Source: DKI Jakarta Water Resources Agency, 2022 [15]

Meanwhile, the data used for GFI (Geomorphic Flood Index) and Hydraulic HEC-RAS analysis are as
follows and are presented in Table 2.
Table 2. Type of Data
No Data Type Year Source
1 Precipitation Data 2003-2021 BMKG
BBWS Ciliwung Cisadane
2 Ciliwung River Discharge 2015-2021 AWLR MT Haryono
3 Administrative Boundary - BIG
4 Terrain Data 2020 BIG
5 Landcover 2021 Esri
7 Topographic Data of Ciliwung - Measurement Topographic in 2021 data by Tri
River Nugraha [16]

5. Methods
Analysis of flood inundation using the Geomorphic Flood Index (GFI) method involves two approaches.
The conventional method is outlined in the Technical Module for Flood Disaster Risk by BNPB
(National Disaster Management Agency). In the conventional method, the hr value is calculated as a
function of the contributing area (Ar) (Accumulated Flow) at the nearest point of the river network that
is hydrologically connected to the point under examination. On the other hand, the GFI method involves
conventional rainfall weighting factors, where the value of V represents the multiplication relationship
between the amount of rainfall and the accumulation value at each designated grid. The description of
the GFI method is attached in the following Figure 4.
 PRECIPITATION DEM

Precipitaion DEM Modification

Return Period with Topograhic
T50 Measurement in
2021

FILL SINK H value

FLOW FLOOD PRONE AREA

FLOW DIRECTION
ACCUMULATION DEPTH
INNUDATION
(Wd) FLOOD PRONE AREA
WITH GFI
FLOW ACCUMULATION WITH
MODIFICATION
PRECIPITATION PERIOD

GFI Modification

Figure 4. GFI Modelling Method

Flood inundation analysis is conducted after supporting data such as river geometry, topography, and
upstream flood discharge are available. This analysis compares the flood inundation results obtained
from the GFI analysis with the data generated from the HEC-RAS 2D software. The following is a
diagram of the flood inundation analysis method using the HEC-RAS 2D hydraulic model, visualized
in Figure 5.

START

Data:
- Precipitation Data
- Landcover
- Actual Discharge Record
- River Stream
- DEM Modification with Topograhic
Measurement in 2021

Hydrologic Analysis

Precipitation Return Period

T50

Discharge Return Period Q50

Hidraulic Analysis Hecras

2D Flow Area

Flood
Flood Height
Innudation Map

HAZARD

Figure 5. Hydraulic HEC

RAS Modelling Method
6. Results and Discussion

6.1. Hidraulic HEC RAS Modelling

The flood hazard assessment step in Hydraulic HEC RAS Modelling for the Q50 method is pictured in
Figure 6.

Central Jakarta
Manggarai Flood Gate

East Jakarta

South Jakarta

AWLR MT. Haryono

Figure 6. Potential Inundation Height by HEC-RAS Hydraulic Model Q50 [16]

6.2. GFI Modelling

The flood hazard assessment steps in the GFI method are

Figure 7. GFI Process

6.2. GFI Modelling
The flood hazard results in GFI modeling are pictured in Figure 8.

Central Jakarta
Manggarai Flood Gate

East Jakarta

South Jakarta

AWLR MT. Haryono

Figure 8. Potential Inundation Height by GFI Conventional Method

Four verification points were used to find out the comparison of the accuracy levels of each model. The
4 points come from the flood height of the Jakarta flood event in February 2017. (Source: DKI Jakarta
DSDA Data Portal, 2022)

Central Jakarta
Manggarai Flood Gate

East Jakarta

South Jakarta

AWLR MT. Haryono

Figure 9. Calibration Map

 Table 3. Flood Height Verification
Plot
Parameter
1 2 3 4
Existing Flood Inundation Height (m) 0.7 0.4 0.4 0.5
Flood Hecras Modelling Feb 2017 1.88 1.71 1.81 No flood
Flood Flood Hecras Modelling Q50 1 2 2 No flood
Height (m) GFI Conventional 3.2 3.02 3.1 No flood
GFI Return Period 50 Years 8.93 8.61 No flood No flood

Table 4. Flood Height Error

Plot
Parameter
1 2 3 4
Flood Hecras Modelling Feb
168% 289% 277%
2017
Error
Flood Hecras Modelling Q50 16% 76% 72% -
Value
GFI Conventional 250% 116% 120%
GFI Return Period 50 Years 257% 262% -

Table 5. Flood Inundation Area Comparison

Inundation Depth Hazard Category Area (Km2)
HEC RAS Q50 GFI GFI Return
Conventional Period 50 Years
0 - 0.75 m Low 0.04 0.73 1.17
0.75 - 1.5 m Middle 0.05 0.47 1.43
> 1.5 m October 0.42 0.22 3.33
TOTAL 0.51 1.42 5.93

The analysis results above indicate that the floodwater elevation values from the HEC-RAS modeling
are the closest to reality compared to the GFI method. This suggests that additional approaches may be
necessary to ensure that the floodwater elevation values closely match the existing conditions in
determining floodwater elevation. The results of the analysis indicate that the inundation generated by
GFI has a larger area than the inundation produced by HEC-RAS. However, if you observe the flood
locations closely, the flood locations between GFI and HEC-RAS differ. This is because, in the HEC-
RAS modeling, the flood area is primarily concentrated within the designated river channel area. On the
other hand, in the GFI modeling, the flood inundation distribution is focused on the river area and
follows the conditions of the analyzed DEM (Digital Elevation Model). This is an advantage of the GFI
model, as it can detect potential flood inundation areas that are not covered by river modeling.

7. Conclusions
The inundation area revealed by the GFI analysis is notably more extensive than the present conditions
and the hydraulic HEC-RAS analysis. Conversely, the elevation values obtained from the HEC-RAS
analysis closely align with the existing conditions, in contrast to the conventional GFI and modified GFI
analysis outcomes. Notably, the GFI analysis results hold the potential for detecting areas at risk of
fluvial flood inundation, underscoring the imperative for further comprehensive studies in this regard.

Acknowledgments
The objective of this study is to compare flood inundation methods in Jakarta. The authors thank the
Water Resources Agency of Jakarta for providing the necessary data for this research. Additionally, the
authors express gratitude to the Faculty of Civil and Environmental Engineering, Institut Teknologi
Bandung, and the Water Resources Engineering Research Group of the same faculty for their support
in the publication of this study.
References
[1] O'Donnell E C and Thorne C R 2020 Drivers of future urban flood risk Philos. Trans. Royal
Soc. 378 p 20190216.
[2] Nugroho P C, Sutjiningsih D and Kurniawan L 2019 Flood hazard assessment based on analysis
of geomorphic flood index and history of flood events (case study in Kemuning Watershed,
Sampang) In IOP Conference Series: Earth and Environmental Science vol 366 (IOP
Publishing) p 012033.
[3] Disaster Risk Reduction [Online] www.preventionweb.net
[4] Twigg J Disaster Risk Reduction
[5] Lechowska E 2018 What determines flood risk perception? A review of factors of flood risk
perception and relations between its basic elements Nat. Hazards 94 1341-66.
[6] Bennett William G, et al. 2023 Modelling compound flooding: a case study from Jakarta,
Indonesia Nat. Hazards 118 277–305.
[7] Wijayanti P, Zhu X, Hellegers P, Budiyono Y and van Ierland E C 2017 Estimation of river flood
damages in Jakarta, Indonesia Nat. hazards 86 1059-79.
[8] Albano R, et al. 2020 Large scale flood risk mapping in data scarce environments: An application
for Romania Water 12 p 1834.
[9] Manfreda S and Samela C 2019 A digital elevation model based method for a rapid estimation of
flood inundation depth J. Flood Risk Manag. 12 p e12541.
[10] Samela C, Albano R, Sole A and Manfreda S 2018 A GIS tool for cost-effective delineation of
flood-prone areas Comput. Environ. Urban Syst. 70 43-52.
[11] Nugroho P C, Sutjiningsih D and Kurniawan L 2019 Flood hazard assessment based on analysis
of geomorphic flood index and history of flood events (case study in Kemuning Watershed,
Sampang) In IOP Conference Series: Earth and Environmental Science vol 366 (IOP
Publishing) p 012033.
[12] Anggraini N and Julzarika A 2019 Deteksi Tinggi Vegetasi di Delta Mahakam dengan
Penginderaan Jauh OLDI (Oseanologi dan Limnologi di Indonesia) 4 175-86.
[13] Khoirunisa N, Ku C Y and Liu C Y 2021 A GIS-based artificial neural network model for flood
susceptibility assessment Int. J. Environ. Res. Public Health. 18 p 1072.
[14] Samela C, Troy T J, Sole A and Manfreda S 2016 A New Geomorphic Index for the Detection of
Flood-Prone Areas at Large Scale Convegno Nazionale di Idraulica e Costruzioni
Idrauliche 35.
[15] Dinas Sumber Daya Air Provinsi DKI Jakarta Peta Jaringan Saluran Drainase
[16] Kesuma T N A, Kusuma M S B, Farid M, Kuntoro A A and Rahayu H P 2022 An assessment of
flood hazards due to the breach of the manggarai flood gate GEOMATE J. 23 104-11.