A quasi-persistent hypoxic water mass in an equatorial coastal sea, Jakarta Bay, Indonesia

Highlights

• Though Jakarta Bay is a shallow open bay, a hypoxic water mass was formed quasi-persistently.

• The hypoxia was generated by large oxygen demand of water and persistent formation of stratification.

• Persistent stratification was generated by weak tidal mixing, high river discharge and lack of continuous convection.

• The recovery of hypoxia was caused by the intrusion of offshore water in the bottom layer induced by wind.

Abstract

Apart from in deep and mainly enclosed systems, the occurrence of hypoxic conditions in coastal seas is generally only a seasonal phenomenon. Despite this, we observed quasi-persistent hypoxia in an equatorial shallow open bay (Jakarta Bay, Indonesia) between 2015 and 2018. There are 3 reasons for the persistence of this hypoxic water mass, namely 1) This bay tends to be stratified due to weak tidal mixing and high river discharge; 2) Due to the lack of continuous vertical convection caused by surface cooling in an equatorial climate, here the water column tends to remain stratified; 3) The oxygen demand in the water is high. Recovery from hypoxia in Sep. 2017 followed the intrusion of offshore water induced by wind. In this study it is suggested that estuaries and ROFIs (Region of Fresh Water Influence) in equatorial regions are susceptible to hypoxia under increasing anthropogenic impacts.

Introduction

Hypoxia generated by anthropogenic eutrophication is a serious problem in many coastal waters (e.g. Diaz, 2001). Hypoxia leads to a decrease in benthic biomass (e.g. Diaz and Rosenberg, 1995), and it alters both the structure and function of benthic communities (e.g. Levin et al., 2009) and the disappearance of fish (e.g. Baden et al., 1990). Therefore, hypoxic water masses are sometimes called dead zones. There are no standard criteria to define hypoxia. Regulatory frameworks tend to define hypoxia where the dissolved oxygen (DO) concentration is below 2–3 mg l⁻¹ in marine and estuarine environments (Farrell and Richards, 2009). Diaz and Rosenberg (1995) defined hypoxia as beginning at 2.0 ml l⁻¹ (2.86 mg l⁻¹). A value of 2 mg l⁻¹ has often been used as a criterion given that DO concentrations below 2 mg l⁻¹ interrupt normal metabolism and behavior of fish and invertebrates and lead to increased mortality (e.g. Diaz, 2001; Kemp et al., 2009), while 3 mg l⁻¹ is commonly used in Japan (e.g. Tokunaga et al., 2009; Nakayama et al., 2010). Ritter and Montagna (1999) showed that the biomass, abundance, and diversity of benthos reduced significantly at < 3 mg l⁻¹ in Corpus Christi Bay, Texas. Here we define hypoxia for a DO concentration lower than 3 mg l⁻¹, considering the high metabolic rate of organisms at high water temperatures in tropical regions.

Research on hypoxia has mainly been conducted in temperate regions, with much less of a focus on tropical regions. Gray et al. (2002) listed the marine areas reported to have been affected by hypoxia, with the majority being located in the temperate or subtropical regions. Altieri et al. (2017) reported that the number of documented dead zones is higher in temperate regions than in the tropics by a factor of 10. One of the reasons for this is that economic development is more concentrated and megacities were developed earlier in temperate regions than in tropical regions. However, the recent economic development in tropical regions is remarkable and megacities have changed from being associated predominantly with developed countries to a phenomenon related to developing and emerging economies (Dsikowitzky et al., 2016). Therefore, the potential threat of hypoxia in tropical regions is now on the increase. One of the other reasons for the lack of research on hypoxia in tropical regions is the climatic characteristics. Annual summertime hypoxia is the most common form of low dissolved oxygen events recorded around the globe (Diaz, 2001). In most temperate estuaries and embayments where hypoxia occurs, it is generated by enhanced stratification with the increase of temperature in summer. On the other hand, the seasonal variation of climate in tropical regions is especially low in equatorial regions, where the water temperature remains constant throughout the year and thermal stratification is weak in coastal areas. These climatic characteristics are the main reason why researchers have not focused on hypoxia in equatorial regions, meaning that seasonal variations in hypoxia in equatorial regions are not well understood.

Jakarta Bay (6.0000° S, 106.8590° E.) is located off the northern coast of Jakarta, Indonesia and surrounded by West Java Province in the east and Banten Province in the west. Its width is about 30 km and its length is about 16 km. It is located in the equatorial region at a latitude of 6° S. Its mean depth is 15 m and there are 13 rivers flowing into Jakarta Bay, of which 10 rivers are small and 3 rivers are of moderate size (Wouthuyzen et al., 2011). The biggest of these is the Citarum River, which is the biggest in West Java. The Citarum River flows into the bay at the northeast end close to the bay mouth. The second biggest river is the Cisadane River flowing into the bay at the northwest end, almost outside the bay. The third biggest river is the Cikarang Bekasi Laut River flowing into the bay at the southwestern coast of the innermost part of the bay. Rather than being a natural river, it is a man-made canal. Active commercial fishing is conducted in Jakarta Bay including many set lift nets (Bagan) and mussel culture facilities. Jakarta City is the biggest city in Indonesia and is located on the coast of the innermost part of the bay. Over the entire Jakarta Metropolitan Area (JMA), the population is about 10.18 million (Badan Pusat Statistik, 2017). As there are no facilities for sewage treatment in Jakarta and its huge hinterland areas, the impact of waste water flow into Jakarta Bay though rivers and canals is considerable (Dsikowitzky et al., 2016).

The seasonal variation in climate in Jakarta is small given its location in an equatorial region. Monthly average air temperature is between 28 and 30 °C with typical changes related to the monsoonal cycle. Two monsoon seasons influence Jakarta Bay, namely the northwest and the southeast monsoon. The northwest monsoon occurs between Dec. and Feb. and the southeast monsoon occurs from Jun. to Aug. (Wyrtki, 1961). The other months are known as the transition season. During the southeast monsoon, the precipitation is low, and this is called the dry season, while the northwest monsoon season is wet, referred to as the rainy season.

With the increase in population of Jakarta City, eutrophication in Jakarta Bay has progressed. Nutrient concentrations in Jakarta Bay have significantly increased between 1975 and 2004 (Arifin, 2004). Eutrophication has been caused by the increase of terrestrial input of nutrients. Compared to other Indonesian coastal areas subject to intense human activities such as the Johor straits between Singapore and Malaysia, nutrient levels in Jakarta Bay are distinctly higher (Damar et al., 2012), and the innermost area of the bay has become hyper-eutrophic (Damar, 2003; Damar et al., 2014). This hyper-eutrophication has been caused by the heavy input of domestic wastewater derived from JMA (Damar, 2003) and its hinterland. Recently, massive fish kills have occurred in Jakarta Bay and have become a problem (e.g. Sidharta, 2004; Thoha et al., 2007). The causes of these fish kills are not well understood. One of the possible causes is the very large blooms of phytoplankton. However, Wouthuyzen et al. (2007) suggested that the forming of oxygen depleted water after the algal blooms may be the key factor that cause the massive fish kills. Suhendar and Heru (2007) reported the occurrence of hypoxia (minimum DO concentration was 0.54 mg l⁻¹) when a massive fish kill occurred, caused by an upwelling of hypoxic water. However, the hypoxia seen in Jakarta Bay is not well understood due to a lack of data. Recently, Ladwig et al. (2016) recorded the distribution of DO concentration in the bottom layer covering the whole of Jakarta Bay, but they did not find hypoxia (minimum concentration DO was 3.2 mg l⁻¹). Huhn et al. (2016) carried out experiments on the stress tolerance of Asian green mussel from Jakarta Bay and from two other natural sites. They found that individuals from the impacted Jakarta Bay performed better under hypoxia than other sites. They mentioned that this was caused by acclimation to hypoxia in Jakarta Bay, suggesting its frequent occurrence. The primary purpose of this study is therefore to report on the distribution of DO concentration in the bottom layer in Jakarta Bay and locate the hypoxic water mass. The second purpose is to clarify any seasonal variations in the hypoxic water mass if it is formed, given that seasonal variations in hypoxia in equatorial regions are not well understood. To these ends, we undertook field surveys of Jakarta Bay 10 times over 27 months and report the results here.