Effect of Urbanization on the Groundwater Discharge Into Jakarta Bay

Trends and Sustainability of Groundwater in Highly Stressed Aquifers (Proc. of Symposium JS.2 at 233 the Joint IAHS & IAH Convention, Hyderabad, India, September 2009). IAHS Publ. 329, 2009.

Effect of urbanization on the groundwater discharge into Jakarta Bay

YU UMEZAWA1, SHIN-ICHI ONODERA2, TOMOTOSHI ISHITOBI3, TAKAHIRO HOSONO4, ROBERT DELINOM5, WILLIAM C. BURNETT6 & MAKOTO TANIGUCHI7 1 Faculty of Fisheries, Nagasaki University, 1-14 Bunkyo-machi Nagasaki, 852-8521, Japan [email protected] 2 Faculty of Integrated Arts and Sciences, Hiroshima University, 1-7-1, Kagamiyama, Higashi-Hiroshima 739-8521, Japan 3 Nara City Office, Minami 1-1-1, NijoOhji, Nara, 630-8580, Japan 4 Priority Organization for Innovation and Excellence, Kumamoto University, 2-39-1 Kurokami, Kumamoto 860-8555, Japan 5 Division of Hydrology, Indonesian Institute of Sciences (LIPI), JI. Sangkuriang 21/154B, Bandung 40135, Indonesia 6 Department of Oceanography, Florida State University, Tallahassee, Florida 32306-4320, USA 7 Research Institute for Humanity and Nature, 457-4, Motoyama, Kita-ku, Kyoto 603-8047, Japan

Abstract At Jakarta city in Indonesia, an increase of chemical fertilizer input in suburban areas and over- abstraction of groundwater at unregistered wells have caused many groundwater-related problems such as - NO3 contamination in the shallow aquifer and seawater intrusion in coastal areas. Because groundwater is an essential carrier of land-derived nutrients into adjacent aquatic ecosystems, as well as river water, it is important to understand the effect of urbanization on the submarine groundwater discharges into coastal areas. In this study, we carried out continuous 222Rn and SGD measurements at a fixed location and along the coast of Jakarta Bay in March, 2008. Rn concentrations in groundwater in coastal areas were mostly similar to those in the lower reaches of a nearby river, suggesting that Rn may not be an effective tool as an indicator of groundwater discharge in coastal waters off Jakarta city. However, 222Rn activities were low along the reclaimed land areas (i.e. 0.8–3.0 dpm/L) around the centre of the city, while the values increased up to 6.0 dpm/L along the coast line with natural mangrove vegetation in western suburban areas. 222Rn and conductivity signatures suggested that an increase of Rn around suburban areas can be caused by river water (likely groundwater fed) rather than direct discharge of groundwater. The estimated minor contribution of groundwater to the terrestrial water flux into the ocean corresponds to the observed decline of hydraulic potential caused by urbanization in the Jakarta city area. Key words SGD (Submarine Groundwater Discharge); Jakarta; 222Rn; urbanization

INTRODUCTION Demographic and social trends suggest that past practices leading to coastal nutrient enrichment in developed cities are likely to be repeated in the coming decade in the developing countries of Asia. Total nitrogen load, including natural nitrogen fixation, increased in Asian megacities by a factor of 3.8 between 1961 and 2002, and is predicted to become up to 1.3 or 1.6 times the present load across the entire region by 2020 (Shindo et al., 2006). Jakarta is one of the representative megacities in Asia, and characterized by a high population density (>10 000 km2) and rapid urbanization in these decades (Jago-on et al., 2009). Jakarta Bay is an open bay attached to Jakarta city, and is connected to the Java Sea. In recent years, severe water pollution accompanied by red tides and an oxygen-depleted water mass has been frequently observed in Jakarta Bay, probably due to an increase of land-derived nutrients supply through several major coastal rivers (Susanna & Yanagi, 2002). However, the contribution of groundwater as an agent carrying terrestrial nutrients into the Bay has yet to be investigated in this area, despite the observed importance of submarine groundwater discharge (SGD) in other areas (Burnett et al., 2009). Several methods for measurements of SGD have been employed, including seepage meters (Taniguchi 2002), piezometers (Ullman et al., 2003; Andersen et al., 2007), thermal infrared imagery analysis (Duarte et al., 2006; Johnson et al., 2008), resistivity measurement (Taniguchi et al., 2006) and geochemical/geophysical tracers (Cable et al., 1996; Moore et al., 1996; Burnett et

234 Yu Umezawa et al. al., 2001; Burnett & Dulaiova, 2003). SGD monitoring using a seepage meter, piezometers and resistivity measurements gives us precise information, but can only cover a limited area of the seabed, and a large scale study requires many deployments that can be time consuming and inefficient. On the other hand, a number of methods based on the use of geochemical tracers such as Rn and Ra isotopes provide an assessment of SGD that integrates over a larger area. Therefore, studies combining multiple methods would be useful (Burnett et al., 2006). At Jakarta Bay, a preliminary survey was conducted by our group using automated seepage meters, salinity loggers and resistivity monitoring at a sandy beach (station 3 in Fig. 1) during the dry season, September 2007, but any symptom of SGD was not observed. The objective of the present study was: (1) to check SGD combining the automated seepage meters and 222Rn monitoring during the rainy season at the beach, where SGD was not observed in the dry season, (2) to monitor spatial variation of SGD over wide areas, which cover whole the coastal line of Jakarta city, and finally (3) to understand the relationships between urbanization and SGD characteristics.

STUDY AREA Jakarta, the capital of Indonesia, is located in the lowland on the north coast of western Java (6°7–22′S, 106°40–55′E) (Fig. 1). Jakarta city consists of homogeneous urbanized areas, while a

Fig. 1 Location of the study site and observation wells. Effect of urbanization on the groundwater discharge into Jakarta Bay 235 mixture of residential and agricultural areas spread outside the city (Murakami et al., 2005). The base of the aquifer system in the southern areas and the southern boundary of Jakarta basin consist of an impermeable Tertiary formation, although this formation does not extend beneath Jakarta city, probably due to a fault or erosion. Volcanic fan deposits with relatively high permeability lie over much of this Tertiary base in the southern area. The lower part of the thick basin fills at depths of 50–300 m along the coast, and consists of irregularly alternating layers of permeable volcanic deposits and Pleistocene marine clays of low permeability. These layers are connected and could allow groundwater flow between each other (Lubis et al., 2008). A thin Holocene sequence of marine and flood plain deposits covers these layers. The main river, Ciliwung River, which has about 347 km2 basin area and about 500 × 106 m3 discharge rate to the ocean (Djadjadilaga et al., 2008), flows northward through central Jakarta, and there are many other small streams (e.g. Sunter River, Cipinang River and Krukut River) flowing to Jakarta Bay. River discharge rate to annual rainfall, which is between 1500 mm and 2500 mm, is roughly estimated to be 40 or 50%, depending on the frequency of heavy rainfall and flood conditions. Along the coast line, industrial plants, hotels and an amusement park are located on the reclaimed areas, and there is some natural vegetation in suburban areas (St. 1, Fig.1).

METHODS 222Rn measurements in groundwater and river water To check if 222Rn is applicable as an indicator of groundwater or river water in the Jakarta city area, water samples for 222Rn analysis were collected from seven observation wells (St. A–G, Fig. 1), 1 dug well (St. G, Fig. 1) and in the lower reaches of a major river, Ciliwung River from 26 February to 1 March, 2008. 222Rn measurements were conducted using a commercial Rn-in-air detector (RAD-7, Durridge Co.) which determine the activity of 222Rn by collection and measurement of the α-emitting daughter, 218Po. Briefly, after enough purging with desiccator to dry out the system, 250-ml water samples were aerated in the closed system (RAD-H2O, Durridge Co.) using the pump equipped with RAD-7. Then, radon in air, which was equilibrated with the dissolved radon in the water phase depending on the temperature, was determined by the detector, and converted to the activity in the original water sample.

Survey of SGD flux at a beach located in Jakarta bay Continuous heat-type automated seepage meters (i.e. vented benthic chambers to automatically measure water flow based on the effect of heat convection caused by water flow: Taniguchi & Iwakawa, 2001) were deployed at four stations (20 m, 30 m, 40 m and 50 m from the beach line: Fig. 2) along a transect perpendicular to the beach (Station 3: Fig. 1) from 25 February to 4 March 2008.

Fig. 2 Setting design of instruments along a transect line.

236 Yu Umezawa et al.

222Rn activity in the water column was continuously monitored from 3 March to 4 March at the 40 m location from the beach (Fig. 2), using an automated multi-detector system that consists of a submersible pump, closed air loop, spray chamber (making equilibrium of Rn between air and water), desiccator, and three sets of RAD-7 detectors. Seawater from about 1.0 m below the surface was introduced into the system at approximately 5 L/min, and the three RAD-7 placed in parallel were operated in 20-minute cycles (Lane-Smith et al., 2002; Dulaiova et al., 2005).

222Rn measurements along coast line To survey the spatial distribution of SGD, surface seawater was continuously pumped up to the detection system on the boat, which was run at a slow speed (6–8 km h-1), and Rn activity was monitored using the automated multi-detector system in a similar manner to the time-series monitoring. Therefore, the Rn activities were measured at an average distance of 2.0 or 2.5 km along the coast line. Conductivity and temperature were simultaneously monitored. The surveys were conducted over three overlapping coast lines on 27 February (St.2 to St.3, Fig. 1), 1 March (St.1 to St.2), and 2 March (St.2 to St.4).

RESULTS AND DISCUSSIONS 222Rn in groundwater and river water Rn concentrations in groundwater showed large variations, depending on the well locations (Table 1). Around inland areas, where volcanic debris covers the shallow Tertiary formation, relatively higher values were observed (116–1079 dpm/L at St. C, F, G, and H). On the other hand, the values in groundwater in the coastal flood tidal plain were low except for location A (32–60 dpm/L at St. B, D, and E), and comparable to concentrations measured in river water (33 dpm/L). Although Rn concentrations determined for batch samples using RAD-H2O are not so accurate, especially for the samples having lower Rn concentrations, these results suggest that river waters are likely under the influence of groundwater as observed in other mega-cities in Asia (Dulaiova et al., 2006). At present, the hydraulic potential is below sea level in the coastal aquifers in Jakarta city (Onodera et al., 2008), and as suggested by higher conductivity in the groundwater in the coastal area (i.e. over 10 000 μS/cm), intrusion of seawater into the groundwater looks very severe. Rn concentrations in seawater contaminated into the groundwater can develop along its retention under the ground as well as fresh groundwater, but Rn concentrations in coastal groundwater were low. So such a low Rn activity in coastal groundwater can be explained by low contents of uranium and radium in the flood tidal areas and recent mixing with seawater having lower Rn activity. These results suggest that Rn is not an effective tool for indicating direct groundwater discharge along the coastal areas of Jakarta city.

Table 1 222Rn concentrations and other chemical parameters for groundwater samples collected around Jakarta city. St. 222Rn Temp pH EC DO Well Sampling bottom depth dpm/L °C μS/cm mg/L m m A 262 ±33 31.1 7.3 4210 3.8 59 50 B 32 ±16 31.0 6.3 10780 0.0 150 35 C 116 ±37 30.7 8.8 169 2.1 123 50 D 36 ±15 32.2 7.0 22900 1.2 225 47 E 60 ±25 30.9 6.8 14370 1.7 50 F 356 ±35 28.5 6.7 420 0.8 40 G 1079 ±161 31.1 7.2 302 1.4 253 55 H 362 ±66 28.1 6.2 435 3.1 8 3 river 33 ±12 21.1 7.6 130 8.8 – – Effect of urbanization on the groundwater discharge into Jakarta Bay 237

Fig. 3 Time-series monitoring of SGD, water level, and conductivity (Ex. St. 50 m).

Survey of SGD flux at a beach located in Jakarta bay Specific flow rates obtained every minute at each station by automated seepage meters ranged widely from 1.0 to 200 cm/day (ex., 5 to 80 cm/day at a location 50 m from the beach: Fig. 3). Higher SGD fluxes were observed during low tide at all locations throughout the monitoring period. However, the conductivity trend did not follow the higher SGD fluxes during low tide, suggesting that most SGD was “recirculated seawater” rather than “fresh groundwater”. On the other hand, Rn concentration monitored at a fixed location (40 m from the beach) showed small variation ranging from 0.8 dpm/L to 1.8 dpm/L (Fig. 4). Constantly low Rn values independent of salinity is consistent with the fact that fresh groundwater discharge was minor at this beach. Continuous Rn monitoring along the coast line showed relatively higher Rn concentrations (i.e. over 5.0 dpm/L) off the centre of Jakarta city (Fig. 5). We tried to run near the shore, but seawalls extending offshore often prevented this approach. Therefore, some of lower values (i.e. less than 0.5 dpm/L) were observed at offshore areas due to a higher contribution of seawater. The spatial distribution of Rn activity may be partly attributed to the tidal shift. But the tidal amplitude is less than 0.8 m and any relationship between Rn activity and tidal shift was not observed during the time-series monitoring (Fig. 4). So we feel that higher amounts of freshwater discharge would be usual at some locations.

Fig. 4 Time-series monitoring of Rn concentration performed in surface water. 238 Yu Umezawa et al.

Fig. 5 Distribution of Rn concentration along the coast line.

Fig. 6 Rn vs conductivity variations in freshwater and seawater.

222Rn measurements along the coast To check the source of freshwater, observed Rn concentrations both in freshwater and seawater were plotted as a function of conductivity (Fig. 6). Rn concentrations in seawater regularly increased with a decrease of conductivity, suggesting the mixing of seawater with a freshwater end-member having similar Rn values. The intercept value of the observed regression line was 16.0, suggesting that Rn values in the source freshwater were close to this value. Because observed lower Rn values in coastal groundwater (i.e. 32–60 dpm/L) were likely caused as the result of seawater contamination, Rn signals combined with conductivity (Fig. 6) suggested that higher Rn concentrations in seawater were caused by river water discharge (cf., 33 dpm/L: probably due to groundwater fed in part) rather than direct groundwater discharge. There are two small rivers, Krukut River and Kramat Jati River, around St. 1 and St. 4, respectively, although Rn Effect of urbanization on the groundwater discharge into Jakarta Bay 239 concentrations were not determined in these rivers during our monitoring period. Radon concentrations in river water may be different between city area and suburban areas, because the river water in the city area can be diluted by municipal waste water. So, it is necessary to understand the variation of Rn values in river waters as well as groundwater. Lubis et al. (2008) described that groundwater in the Jakarta city area flows mainly S-SE to N-NW by analysis of the subsurface temperature distribution. So it was first assumed that groundwater discharge could be observed at the coast northwest of Jakarta city. But, as Onodera et al. (2008) reported, this area is governed by seawater intrusion rather than groundwater discharge. Therefore, at present groundwater seems to be a minor agent to directly carry terrestrial nutrients and other contaminants into the ocean compared with river water in this area, where the decline of hydraulic potential due to overpumping of groundwater has yet to recover. Intensive urbanization has affected subsurface environments such as subsurface temperature increase (Taniguchi et al., 2005), land subsidence (Phien-wej et al., 2005) and deterioration of groundwater qualities (Umezawa et al., 2008). We need more examples of Rn and SGD monitoring in areas having different geologies and morphologies to clarify the effect of urbanization on SGD.

Acknowledgements We thank the staff and students at Hiroshima University and Indonesia Institute of Sciences (LIPI) for supporting water sampling and processing. H. Bakti (LIPI), S. Kagabu (Kumamoto University), Y. Shimizu (Hiroshima University) helped us with water sampling and data gathering in Jakarta. F. Lubis (LIPI) gave us the information on river names. We also appreciate the support of D. Lane-Smith (Durridge Co. Ltd.) for technical advice and further discussions. This research was funded by the project “Human Impacts on Urban Subsurface Environment”, Research Institute for Humanity and Nature (RIHN) and partly by Nagasaki University.

REFERENCES Andersen, M. S., Baron, L., Gudbjerg, J., Gregersen, J., Chapellier, D., Jakobsen, R. & Postma, D. (2007) Discharge of nitrate- containing groundwater into a coastal marine environment. J. Hydrol. 336, 98–114. Burnett, W. C. & Dulaiova, H. (2003) Estimating the dynamics of groundwater input into the coastal zone via continuous radon-222 measurements. J. Environmental Radioactivity 69, 21–35. Burnett, W., Kim, G. & Lane-Smith, D. (2001) A continuous monitor for assessment of 222Rn in the coastal ocean. J. Radioanalytical Nuclear Chemistry 249(1), 167–172. Burnett, W. C., Chanyotha, S., Wattayakorn, G., Taniguchi, M., Umezawa, Y. & Ishitobi, T. (2009) Underground sources of nutrient contamination to surface waters in Bangkok, Thailand. Sci. Total Environ. 404. Burnett, W. C., Aggarwal, P. K., Aureli, A., Bokuniewicz, H., Cable, J. E., Charette, M. A., Kontar, E., Krupa, S., Kulkarni, K. M., Loveless, A., Moore, W. S., Oberdorfer, J. A., Oliveira, J., Ozyurt, N., Povinec, P., Privitera, A. M. G., Rajar, R., Ramassur, R. T., Scholten, J., Stieglitz, T., Taniguchi, M. & Turner, J. V. (2006) Quantifying submarine groundwater discharge in the coastal zone via multiple methods. Sci. Total Environ. 367, 498–543. 222 Cable, J. E., Bugna, W. C., Burnett, W. C. & Chanton, J (1996) Application of Rn and CH4 for assessment of groundwater discharge to the coastal ocean. Limnol. Oceanogr. 41, 1347–1353. Djadjadilaga, M., Sigit, H. & Tejalaksana, A. (2008) From Data to Policy – Ciliwung River Water Quality Management – The 3rd WEPA International Forum. Kuala Lumpur, Malaysia. Duarte, T. K., Hemond, H. F., Frankel, D. & Frankel, S. (2006) Assessment of submarine groundwater discharge by handheld aerial infrared imagery: case study of Kaloko fishpond and bay, Hawai’i. Limnol. Oceanogr. Meth. 4, 227–236. Dulaiova, H., Burnett, W. & Lane-Smith, D (2005) A multi-detector continuous monitor for assessment of 222Rn in the coastal ocean. J. Radioanalytical Nuclear Chemistry 263, 361–365. Dulaiova, H., Burnett, W. C., Wattayakorn, G. & Sojisuporn, P. (2006) Are groundwater inputs into river-dominated areas important? The Chao Phraya River-Gulf of Thailand. Limnol. Oceanogr. 51, 2232–2247. Jago-on, K. A. B., Kaneko, S., Fujikura, R., Fujiwara, A., Imai, T., Matsumoto, T., Zhang, J., Tanikawa, H., Tanaka, K. & Taniguchi, M. (2009) Urbanization and subsurface environmental issues: an attempt at DPSIR model application in Asian cities. Sci. Total Environ. (in press). Johnson, A. G., Glenn, C. R., Burnett, W. C., Peterson, R. N. & Lucey, P. G. (2008) Aerial infrared imaging reveals large nutrient-rich groundwater inputs to the ocean. Geophys. Res. Lett. 35, L15606, doi:10.1029/2008GL034574. Lane-Smith, D. R., Burnett, W. C. & Dulaiova, H (2002) Continuous radon-222 measurements in the coastal zone. Sea Technol. 43, 37–45. Lubis, R. F., Sakura, Y. & Delinom, R (2008) Groundwater recharge and discharge processes in the Jakarta groundwater basin, Indonesia. Hydrogeol. J. DOI 10.1007/s10040-008-0278-1. Moore, W. S. & Arnold, R (1996) Measurement of 223Ra and 224Ra in coastal waters using a delayed coincidence counter. J. Geophys. Res. 101, 1321–1329. 240 Yu Umezawa et al.

Murakami, A., Zain Medrial, A., Takeuchi, K., Tsunekawa, A. & Yokota, S. (2005) Trends in urbanization and patterns of land use in the Asian mega-cities Jakarta, Bangkok, and Metro Manila. Landscape Urban Planning 70, 251–259. Onodera, S., Saito, M., Sawano, M., Hosono, T., Taniguchi, M., Shimada, J., Umezawa, Y., Lubis, R. F., Buapeng, S. & Delinom, R. (2008) Effects of intensive urbanization on the intrusion of shallow groundwater into deep groundwater: examples from Bangkok and Jakarta. Sci. Total Environ. 404, 401–410. Phien-wej, N., Giao, P. H. & Nutalaya, P. (2006) Land subsidence in Bangkok, Thailand. Engng Geol. 82, 187–201. Shindo, J., Okamoto, K. & Kawashima, H. (2006) Prediction of the environmental effects of excess nitrogen caused by increasing food demand with rapid economic growth in eastern Asian countries, 1961–2020. Ecol. Modelling 193, 703–720. Susanna, N. & Yanagi, T. (2002) Lower trophic level ecosystem in Jakarta Bay, Indonesia. La Mer 40, 161–170. Taniguchi, M. (2002) Tidal effects on submarine groundwater discharge into the ocean. Geophys. Res. Letters 29, doi:10.1029/2002GL014987. Taniguchi, M. & Iwakawa, H. (2001) Measurements of submarine groundwater discharge rates by a continuous heat-type automated seepage meter in Osaka Bay, Japan. J. Groundwater Hydrol. 43(4), 271–277. Taniguchi, M., Uemura, T. & Sakura, Y. (2005) Effects of urbanization and groundwater flow on subsurface temperature in three megacities in Japan. J. Geophys. Engng 2, 320–325. Taniguchi, M., Ishitobi, T. & Shimada, J. (2006) Dynamics of submarine groundwater discharge and freshwater–seawater interface. J. Geophys. Res. 111, C01008, doi:10.1029/2005JC002924. Ullman, W. J., Chang, B., Millera, D. C. & Madsen, J. A. (2003) Groundwater mixing, nutrient diagenesis, and discharges across a sandy beachface, Cape Henlopen, Delaware (USA). Estuar. Coast Shelf Sci. 57, 539–552. Umezawa, Y., Hosono, T., Onodera, S., Siringan, F. P., Buapeng, S., Delinom, R., Yoshimizu, C., Tayasu, I., Nagata, T. & Taniguchi, M. (2008) Sources of nitrate and ammonium contamination in groundwater under developing Asian megacities. Sci. Total Environ. 404, 361–376.