EFFECT OF LAND SUBSIDENCE ON FLOOD INUNDATION IN JAKARTA, INDONESIA

Idham Riyando MOE1, Shuichi KURE2, Nurul Fajar JANURIYADI3, Mohammad

FARID4, Keiko UDO5, So KAZAMA6 and Shunichi KOSHIMURA7

1Member of JSCE, Department of Civil and Environmental Engineering, Tohoku University (468-1 Aoba, Aoba-ku, Sendai-shi, Miyagi, 980-0845, Japan)

E-mail:idham.moe@gmail.com 2Member of JSCE, Associate Professor, Department of Environmental Engineering, Toyama Prefectural University

(5180 Kurokawa, Imizu-shi, Toyama 939-0398, Japan) 3Member of JSCE, Graduate School of Environmental Studies, Tohoku University

(468-1 Aoba, Aoba-ku, Sendai-shi, Miyagi, 980-0845, Japan) 4 Researcher, Water Resources Engineering Research Group,

Faculty of Civil and Environmental Engineering, Bandung Institute of Technology (Jalan Ganesha 10 Bandung 40132, Indonesia)

5Member of JSCE, Associate Professor, International Research Institute of Disaster Science, Tohoku University (468-1 Aoba, Aoba-ku, Sendai-shi, Miyagi, 980-0845, Japan)

6 Member of JSCE, Professor, Department of Civil and Environmental Engineering, Tohoku University (6-6-06 Aoba, Aoba-ku, Sendai-shi, Miyagi, 980-8579, Japan)

7Member of JSCE, Professor, International Research Institute of Disaster Science, Tohoku University (468-1 Aoba, Aoba-ku, Sendai-shi, Miyagi, 980-0845, Japan)

Floods are considered to be one of the major natural disasters in Indonesia. Jakarta in Indonesia has experienced many floods in the past, such as those in 1996, 2002, 2007 and 2013. In this paper, land subsidence problems contributing to flooding in Jakarta were described and historical and future land subsidence situations in Jakarta were reconstructed and projected using a simple linear extrapolation method. Also, those land subsidence impacts were analyzed by using a flood inundation model. As a result of analysis, it was found that the land subsidence in Jakarta contributed by 17.6% to increase of the total flood inundation volume from 1983 to 2013. Also, it was estimated that the land subsidence situations for the future period of 2050 would increase the flood inundation volume by 10.3% compared to those of 2013. However, impacts from land use/cover changes on the flood inundation were found to be much greater than those from land subsidence in Jakarta. It should be noted that the land subsidence affects flooding from the coast when high waves and surges occur but those effects were not considered in this paper. Key Words: Jakarta, land subsidence, flood inundation model, land use/cover change

1. INTRODUCTION

The city of Jakarta in Indonesia, officially known as the Special Capital Region of Jakarta, is the political, economic and cultural center of Indonesia. However, many social problems and human activities in Jakarta contributed to increase of flood frequency and magnitude in Jakarta year by year1). In addition, Jakarta is located in the lowland area, so that the city is inescapable from the threat of floods any time. Jakarta has experienced with a number of floods such as those in 1996, 2002, 2007, 2013 and 2014. As reported by Kure et al.1), those floods resulted in not only human casualties but also terrible economic damage in Jakarta. Flood prevention and mitigation for Jakarta is one of the main priority actions to be taken in Indonesia.

Today, Indonesia is one of the Asian countries

where infrastructures are being developed rapidly with the economic growth. This leads to uncontrolled urban growth and the houses are cramped and huddled. Farid et al.2) pointed out that the urbanization of the Ciliwung River basin is contributing to increase of the flood flow of the river. Moe et al.3) evaluated that the flood peak and flood inundation volume will increase with the expansion of the urbanized area in and around Jakarta. Also, land subsidence is one of the factors contributing to flooding in Jakarta’s lower area4). Increase of the ground water extraction due to urbanization was reported as a main factor of the land subsidence in Jakarta4). Park et al.5) emphasized that all areas including roads and buildings in the northern part of Jakarta are below mean sea level and those areas are vulnerable to flood inundation.

Many researches have been conducted to

I_283

土木学会論文集Ｇ（環境）, Vol.72, No.5, I_283-I_289, 2016.

understand the city’s flood problems, gradually improving the management of water resources of Jakarta. However, flooding in Jakarta is complex because it is related to land use change, land subsidence and other social factors, and no quantitative evaluation has been made to study the city’s flood problem.

The main objective of this paper is to quantitatively evaluate the effects of land subsidence on flooding in Jakarta for the historical and future periods by reconstructing and projecting the city’s land subsidence situations.

2. STUDY AREA AND DATA USED

Jakarta is the largest metropolitan city in Indonesia and its development is progressing rapidly. There are thirteen rivers flowing into Jakarta, and the main and the longest river is the Ciliwung River which passes through the city center from upstream region located in the border of Cianjur and Bogor cities as shown in Figure 1. The Ciliwung River has a catchment area of 382.6 km2 with the river length of 117 km. The Ciliwung River brings the largest flood damage to Jakarta compared to other rivers that flow into Jakarta.

The target area selected in this study includes Jakarta and the Ciliwung River basin totally covering 1346.6 km2 as shown in Figure 1. The digital elevation data6) from Japan Aerospace eXploration Agency (JAXA) with a resolution of 30 m was employed in this study in order to precisely represent the land subsidence situations in Jakarta although the previous studies3),7) used the resolution of 90 m for the flood inundation simulation. The elevation values in Figure 1 are based on the above mean sea level.

The cross section data of rivers and the drainage system of the Ciliwung River in 2011 were obtained from the project authority of JICA. For rainfall data, radar rainfall information was provided by the BPPT (Badan Pengkajian dan Penerapan Teknologi: Agency for the Assessment and Application of Technology), and water level data of the Ciliwung River and flood inundation map of Jakarta were provided by the BPBD (Badan Penanganan Bencana Daerah: Jakarta Disaster Management Agency). Also, land use maps in 1983, 1996 and 2002 were provided by the PU (Ministry of Public Works). 3. LAND SUBSIDENCE PROBLEMS

Land subsidence is a natural and/or human made phenomenon affecting many urban areas all over the world. Jakarta is located in the low land area in the northern coast of West Java, Indonesia. The city has

a population of about 10.2 million in 2015. Land subsidence is a well-known problem seriously affecting Jakarta, as also being reported by other studies4),5). According to those studies, land subsidence in Jakarta has spatial and temporal variations with the rates of about 1 – 15 cm/year, and a few locations have the subsidence rates up to 20 – 28 cm/year over the period of 1982-2010.

Land subsidence in the certain areas1) of the city has spatial relation with flooding during rainy seasons. Land subsidence theoretically leads to expanding inundated areas and deepening water depth in flooded areas. In the northern part of Jakarta, especially coastal areas affected by land subsidence, sea level rise and high tide will contribute directly to increasing water depth and causing flooding to become higher. Also, land use/cover change and shortage of capacity of rivers and small canals in the urban area also affect subsidence areas during heavy rainfall, resulting in the expansion of flood areas.

In the last three decades, the development of Jakarta and its surrounding areas has brought about the rapid growth of many sectors, such as industry, trade, and real estate, leading to many environmental problems including land subsidence4),5). Kure et al.1) and Moe et al.3) emphasized that the urban flooding in the northern part of Jakarta is highly related to urbanization and land subsidence. According to Atlas8), the total land subsidence of up to about 4.0 m occurred in the lowland areas from 1974 to 2010, and those areas are located below mean sea level. Thus, such areas are easily inundated by intensive rainfall4).

As such, land subsidence has been contributing to flooding in Jakarta, and its effects on the floods should be evaluated quantitatively. In this paper, a future land subsidence situation of Jakarta is projected using a simple linear extrapolation method. Also, a physically based rainfall runoff and flood inundation model is applied to the target area in order to represent the flood inundation situation in Jakarta. And then, the effects of land subsidence on flood inundation are evaluated based on the flood model application under the projected land subsidence situation of Jakarta in the future. Details of the methods are explained in the below sentences.

4. METHODOLOGY (1) Land subsidence for the historical and future periods

In order to evaluate the situation of land subsidence in Jakarta for the historical and future periods, accumulated land subsidence from 1974 - 2010 reported by another study4) was extracted at each grid point to make a spatially interpolated land subsidence accumulation map over Jakarta by using

I_284

the inverse distance weighted method. Then, those accumulated values were adjusted to obtain the year average subsidence rate [m/year] by dividing with the total year of accumulation from 1974 – 2010 under the assumption that the land subsidence speed is constant in time. The computed map of land subsidence rate in Jakarta is shown in Figure 2.

The digital elevation model shown in Figure 1 by JAXA with the resolution6) of 30 meters was selected as the base data. The data was observed by ALOS equipped with the Panchromatic Remote-sensing instrument for Stereo mapping (PRISM), to acquire elevation from 2006 to 2011 that was the operational period of ALOS9). Therefore, we assume that the data provided from ALOS represents the Jakarta’s land surface elevation situation in 2009 as the base year.

From the base data of 2009, we reconstructed and projected the land elevation in Jakarta for the historical and future periods by using a linear extrapolation based on the subsidence rate at each grid as shown in Figure 2. Figure 3 shows a projected land subsidence situation in Jakarta in 2050 compared to 2009. It was found that more than 4 m land subsidence from 2009 can be found in the northwestern part of Jakarta in 2050.

However, it should be emphasized that the land subsidence situation for the future period derived from the linear extrapolation was one of the worst scenarios based on the assumption that there would be increase of ground water withdrawal and high-rise buildings in Jakarta keeping land subsidence speed constant. Murakami et al. 10) and Yasuhara et al.11) developed an observational prediction method of land subsidence based on the observed data at the target area. According those studies, time series of the land subsidence at several areas were successfully reproduced by an exponential equation, and they

pointed out that land subsidence rate was affected by ground water level. As such, the land subsidence situation projected in this paper should be considered as the worst scenario. This kind of the linear extrapolation for the projection of the land subsidence was also applied to Jakarta by Takagi et al.12)

(2) Flood inundation modelIn this study, a physically based rainfall-runoff

and flood inundation model3),7) was employed in

Fig. 2 The annual land subsidence rate in Jakarta. Fig. 3 Projected land subsidence situation in Jakarta in 2050 compared to 2009.

Fig. 1 Study area and the location of Jakarta City.

#Manggarai Gate

±0 10 205 Kilometers

#

I_285

order to simulate the flood inundation situations in Jakarta. The model consists of rainfall-runoff module at each subbasin, hydrodynamic module in the river and canal networks, and flood inundation module for the floodplains.

For the rainfall runoff simulation, a physically based distributed rainfall-runoff model13) was employed because this model can simulate the hortonian overland flow in the urban areas, and the subsurface flow and saturation overland flow in mountainous areas depending upon the soil and geological characteristics and the rainfall intensity in a hillslope. Radar rainfall information provided by BPPT was used as the input for the simulation. Soil parameters used in the simulation were calibrated based on 4 land cover classes of urban, forest, cropland, and paddy field in the target area as shown in Table 13). Calibrations for the parameters were conducted based on the comparison between observed and simulated river discharge at the stations of Ciliwung River.

The Saint-Venant equations for the conservation of continuity and momentum were used in a flood routing in rivers and a drainage system. For the flood inundation simulation of the flood plain, the unsteady two-dimensional flow equations consisting of the continuity equation and momentum equation were numerically solved. The manning’s roughness coefficients of the river beds were set from 0.03 to 0.1 for river sections, and the said coefficients of the land surface were set as 0.1 for all the floodplains through the calibration. For the details of the

simulation, see the reference3). In this study, the 30 m resolution DEM was used for the flood inundation simulation instead of the 90 m resolution DEM used in the previous studies3),7) in order to represent the land subsidence situations in the target area more precisely. (3) Model application Results

The flood inundation model was applied to the flood event from 14 - 18 January in 2013. The event resulted in severe damage in Jakarta1). It should be noted that the land use condition in 2002 provided by PU was used for the model parameter calibration and simulation. It meant we assumed that the land use conditions in 2013 was the same as those in 2002 without any city development after 2002. This is because Jakarta had fully urbanized by 2002 except upstream mountainous regions as shown in Figure 6.

Figure 4 shows the observed inundation area and the simulated maximum inundation depth of the target area. The observed flooded area data was provided by the BPBD. The simulated inundation area in the target area matched relatively well with the observation results, namely, at the north-west and

Fig. 4 Comparison between the simulated (left) and observed (right) inundation areas.

#

±0 10 205 Kilometers

±0 10 205 Kilometers

Table 1 Calibrated soil parameters Urban Forest Cropland Paddy Field

Soil depth [cm] 0.5 20 15 5Saturated hydraulicconductivity [mm/h]

5 270 170 50

Effective porocity 0.1 0.5 0.4 0.2Surface roughness 0.1 0.6 0.3 0.2Runoff coefficients 0.8 0.4 0.5 0.6

I_286

north-east of Jakarta and the upstream area of the

Fig. 5 Flood inundation simulation result based on the land subsidence in 1983 (upper left) and 2050 (upper right) and

differences between 2050 and 1983 (lower).

Fig. 6 Land use/cover maps3) in 1983 (left), 1996 (middle), and 2002 (right)

I_287

Manggarai gate point. However, there are discrepancies between the reported and simulated inundation areas, especially in the lower part of the Manggarai gate point. This is because the inundated water from the upstream is flowing down to the lower part of the Manggarai gate area. In the real situation in Jakarta, there are river banks and which makes it difficult for the inundated water to across the river banks and flow down to the downstream area. In addition, a tidal effect and local ponds might affect flood inundations at the north central coast, so that only observation shows the inundations in the area. Those inundation behaviors and local situations should be taken into account in the model simulation to get a better simulation result.

(4) Limitation

As the limitations of the present study, it should be noted that artificial structures such as river gates and pumping stations in the floodplains were not taken into account in the model simulation in this paper.

Therefore, the results presented in the paper should be considered as initial assessment results of the land subsidence. Also, it should be noted that the land subsidence affects flooding from the coast when high waves and surges occur but those effects were not considered in this paper. 5. LAND SUBSIDENCE EFFECT

Impacts of land subsidence on flood inundation in Jakarta are analyzed by the calibrated model with the land surface elevation data reconstructed and projected from 1983 through 2050. In addition to the

land subsidence, the impacts of land use/cover change are analyzed for the historical period in order to compare both the impacts from the land use/cover change and land subsidence on the flood inundation.

The reconstructed and projected land surface elevations considering the land subsidence are used for the elevation data of the flood inundation model for each year. Other conditions such as rainfall runoff simulations and model parameters are exactly the same as the calibration simulation of the 2013 flood event. Figure 5 shows the flood inundation simulation results in 1983 and 2050 and the difference between 1983 and 2050. It can be seen from those figures that the flood inundation increases in 2050 and decreases in 1983 compared to the 2013 simulation (Figure 4) because of the land subsidence effects, and highly affected areas are located in the high land subsidence rate areas (Figure 2). It is noted that there are no significant changes of flood inundation peak times between the simulations because those peak times are mainly depending on the rainfall pattern and river flood discharges. It was found that the land subsidence in Jakarta contributed by 17.6% to increase of the total flood inundation volume from 1983 to 2013. The total flood inundation volumes in Jakarta in 2013 and 2050 are calculated as 2,779,740 m3 and 3,065,805 m3, respectively. From those results, we concluded that the city’s flood inundation volume would increase by 10.3% by 2050 because of the land subsidence.

It should be noted that the land use/cover change conditions are not taken into account in the above simulations. Moe et al.3) reported that the land use/cover changes were significant in Jakarta from

Fig. 7 Effects of land subsidence and Land use/cover change on the flood inundation area (left) and volume (right).

1990 2000 2010 2020 2030 2040 205045

50

55

60

65

70

75

80

85

Land Use Change Land Subsidence

Year

Inun

datio

n A

rea

[km

2 ]

Land Use Change + Land Subsidence

1983 1990 2000 2010 2020 2030 2040 2050

2

3

[×10+6]Land Use Change Land Subsidence

Year

Inun

datio

n V

olum

e [m

3 ]

Land Use Change + Land Subsidence

1983

I_288

1983 through 2002, and those impacts are also analyzed in this paper. Figure 6 shows the land use/cover maps3) in 1983, 1996 and 2002 provided by PU. The details of the simulations considering the land use/cover change on flooding can be seen in the reference3). As mentioned in the model application section, we assume that there has been no land use/cover change since 2002, so that the 2013 simulation was conducted using the 2002 land use/cover map. This assumption is based on that Jakarta had fully urbanized by 2002 except upstream mountainous regions as shown in Figure 6. Figure 7 shows the simulated flood inundation volumes and areas for the historical and future periods. For the historical period, the simulations were conducted considering only land use change and land subsidence and both of them. It can be seen from Figure 7 that the impacts of the land use/cover change are much greater than those of the land subsidence. This is because the land use/cover change will affect the flood runoff volume itself. However, the land subsidence impacts are also unneglectable, as shown in Figure 5 and 7.

It should be emphasized that the future developments in the upstream of Ciliwung River basin are still expected and possible. Those impacts from land use/cover change on the flood inundation would be more significant than those from the land subsidence. The Jakarta city government should consider countermeasures not only for land subsidence such as ground water extraction regulations but also for land use/cover changes in the upstream region such as land use controls. 6. CONCLUSIONS

In this paper, the land subsidence situations of Jakarta for the historical and future periods were reconstructed and projected in order to evaluate those impacts on the flood inundation by using the calibrated rainfall runoff and flood inundation model.

As the result of analysis, we concluded that the land subsidence in Jakarta contributed by 17.6% to increase of the total flood inundation volume from 1983 to 2013. Also, it was found that the land subsidence situations in 2050 would increase the flood inundation volume by 10.3% compared to those in 2013. However, the impacts from land use/cover changes on the flood inundation were found to be much greater than the impacts from land subsidence. Thus, flood countermeasures should be conducted not only for land subsidence problems in Jakarta, but also land use/cover change problems in the upstream region. A ground water withdrawal regulation in the lowland area and land development regulation in the upstream region should be conducted by the

government as soon as possible. ACKNOWLEDGMENT: This research was supported by the Environment Research and Technology Development Fund (S-14) of the Ministry of the Environment, Japan. REFERENCES 1) Kure S., Farid M., Fukutani Y., Muhari A., Bricker J.D.,

Udo K., and Mano A.: Several Social Factors Contributing to Floods and Characteristics of the January 2013 Flood in Jakarta, Indonesia, Journal of Japan Society of Civil Engineers, Ser. G, Vol. 70, No.5, pp.I_211-I_217, 2014. (in Japanese with English abstract).

2) Farid, M., Mano, A., and Udo, K.: Modeling Flood Runoff Response to Land Cover Change with Rainfall Spatial Distribution in Urbanized Catchment. Annual Journal of Hydraulic Engineering, Vol. 55, pp.19-24, 2011.

3) Moe I.R., S. Kure, M. Farid, K. Udo, S. Kazama and S. Koshimura, Evaluation of Flood Inundation in Jakarta Using Flood Inudation Model Calibrated by Radar Rainfall, Journal of Japan Society of Civil Engineers, Ser. B1 (Hydraulic Engineering), Vol.72, No.4, pp. I_1243-I_1248, 2016.

4) Abidin, H.Z., Y. Fukuda, Y.E. Pohan, T. Deguchi: Land subsidence of Jakarta (Indonesia) and its relation with urban development, Nat Hazards, Vol.59, pp. 1753-1771, 2011.

5) Park. H., Kwon S., and Hadi S.: Land Subsidence Survey and Policy Development in Pantai Mutiara, Jakarta Bay, Indonesia, Journal of Coastal Research, No. 75, pp. 1447-1451, 2016.

6) JAXA: ALOS WORLD 3d – 30 m, Available at: http://www.eorc.jaxa.jp/ALOS/en/aw3d30/index.html accessed April 11, 2016

7) Moe I.R., S. Kure, M. Farid, K. Udo, S. Kazama and S. Koshimura, Numerical Simulation of Flooding in Jakarta and Evaluation of a Countermeasure to Mitigate Flood Damage, Journal of Japan Society of Civil Engineers, Ser. G (Environment), Vol.71, No.5, pp. I_29-I_36, 2015.

8) ATLAS:SAFETY BEACH JAKARTA (Draft), pp.123, 2011, Available at : http://ja.scribd.com/doc/72755643/ATLAS-Jakarta-Coastal-Defense-Strategy

9) JAXA: AVNIR-2 Acquisition Plan, Available at: http://www.eorc.jaxa.jp/ALOS/en/obs/alos_scenario/avnir-2/avnir-2.htm accessed April 11, 2016

10) Murakami S., K. Yasuhara, and N. Mochizuki: An Observational Prediction of Land Subsidence for GIS, Journal of Groundwater Hydrology, Vol. 45, No.4, pp.391-407, 2003. (in Japanese with English abstract).

11) Yasuhara K., S. Murakami and N. Mimura: Inundation Caused by Sea-Level Rise Combined with Land Subsidence, Geotechnical Engineering Journal of the SEAGS & AGSSEA, Vol. 46, No.4, pp.102-109, 2015.

12) Takagi H., M. Esteban, T. Mikami, and D. Fujii: Projection of Coastal Foods in 2050 Jakarta, Urban Climate, Vol.17, pp.135-145, 2016.

13) Kure, S. and Yamada, T.: Nonlinearity of runoff and estimation of effective rainfall in a slope. Proceedings of the 2nd Asia Pacific Association of Hydrology and Water Resources Conference, Vol.2, pp.76-85, 2004.

(Received April 7, 2016)

I_289