Peter Guest – Heading north from the city centre to the coastline, Jakarta seems to be collapsing in slow motion.
The Indonesian capital sprawls, its black-glass business district giving way to a low-rise hinterland where the bones of the city jut out; long spines of pale concrete pillars bearing kilometres of knotted overpasses and raised highways.
In their shadows are industrial estates in various states of abandonment, stalled construction sites already succumbing to the creep of tropical foliage, sluggish waterways clotted with litter, and thousands upon thousands of houses, from clusters of bare-iron shacks to landed three-storey homes, none the same as its neighbour.
The chaos runs all the way to the seafront, where waterparks, glossy malls and luxury condos jostle for space with container ports and fishing docks crammed so tight with small boats that from above they look like tangles of rusted wire snagged on the shore.
Some of these docks are now hemmed in by giant walls. At Cilincing – a northeastern suburb of the city made up of scattered fishing communities and industrial ports – five-metre-high concrete pillars have been dropped into the shoreline, supporting a sloping buttress that blocks all view of the sea from the land. Less than 50 metres behind it on the landward side is another wall, constructed less than a decade ago, that is now redundant; between them, fishermen use a placid inlet to tie up and maintain their boats.
Twenty kilometres of sea walls have been thrown up around Jakarta Bay in the past three years, along with many more reinforcements along river banks, the first phase of a desperate attempt to fortify the city's waterlogged northern districts.
Jakarta, a megacity of 30 million people, is sinking. In places along the coastline the ground has subsided by four metres over the last few decades, meaning that the concrete barricades are the only thing preventing whole communities from being engulfed by the sea.
Although many coastal cities, from New York to Shanghai, have been forced by the threat of climate change to build high walls to protect themselves, there are few places in the world as vulnerable as Jakarta, where a decades-old problem of land subsidence has intersected with sea level rise caused by global warming, creating an existential threat to the city.
Such is the concern about Jakarta's future that the national government is considering bailing out. In April, president Joko Widodo – himself a former governor of the city – announced a public search for a new capital for Indonesia, in no small part because of its environmental problems.
The city's new walls have bought it some time, but not much, and possibly not enough. Behind them is an alarming case study in how politicking, greed and vested economic interests can lead to a dangerous inertia – a microcosm of the global failure to address climate change. Whether the city saves itself, or whether it becomes the first megacity lost to environmental catastrophe, will depend on a combination of ground-level social change and engineering works of unprecedented scale to hold back the tide.
"If we don't do something, we're doomed," says Oswar Mungkasa, the city's deputy governor. "We will be leaving Jakarta."
Jakarta has always flooded. With so much of the city so close to sea level, the 13 rivers that flow through the metropolitan area take a long time to drain into Jakarta Bay. Even relatively short periods of heavy rain cause water to build up. To manage this, successive governments have built a network of canals, most of them glassy in appearance and slow-flowing, giving the impression that the city is already floating. Pumping stations, weirs and run-off reservoirs sit between malls and offices and in the centre of housing estates.
Over the past decade, this canal infrastructure has come under increasing stress – counterintuitively, because of a lack of water in Jakarta.
As the city grew after the 1970s oil boom – the population of the wider metro region has more than tripled in 50 years – it spread far faster than its supporting infrastructure. Piped water services only reach around 60 per cent of the population and are concentrated in the relatively wealthy areas in south and central Jakarta. The rivers, which should provide a source of fresh water, are largely unusable due to the unregulated dumping of waste, from untreated human excrement to industrial effluent. To get around the lack of affordable water, residents and businesses have sunk boreholes into the aquifers beneath the city. Even some government buildings rely on groundwater pumps.
"There is not sufficient water delivery, so people are pumping out too much groundwater, and because of the rapid urbanisation over the last 30 years, the amount of permeable surface in the city has decreased to a point where you don't have enough recharge in the groundwater," says Kian Goh, assistant professor of urban planning at the University of California, Los Angeles, who has studied Jakarta in depth.
Pumping out the groundwater has quite literally lowered the city's foundations, causing widespread subsidence. Some areas in the north have sunk four metres over the past two decades, putting them so far below the level of the bay that there is nowhere for water to drain out.
Climate change is likely to compound this problem. The Intergovernmental Panel on Climate Change predicts that "business as usual" carbon emissions would drive a one-metre rise in sea levels by 2100, although some studies warn that figure could be far higher. In the more hopeful scenarios envisioned by the 2015 Paris Agreement on climate change – temperature rises of 1.5C or 2C – sea levels are predicted to increase by 0.4m or 0.46m, respectively.
In Jakarta, where so much of the city is already low-lying, the margin for error is nonexistent. "I've seen several studies that say that if the trend of sea level rise continues, by 2030 the north of Jakarta will be flooded, including the international airport," says Arief Wijaya, who heads the climate change programme at the World Resources Institute Indonesia.
Adding to that is the threat of unpredictable and extreme weather. Jakarta is already hit by storm surges and heavy rain from annual cyclones. Global heating is expected to increase the frequency and intensity of those events, making flooding from both the land and the sea more likely.
Successive administrations have been aware of the problem, but have felt able to ignore it, largely because the consequences impacted poorer areas of the city, such as the southern district of Bukit Duri, which straddles the Ciliwung River. There, on one bank, wood-framed houses lean on stilts over the water, which is filthy and clogged with plastic waste; on the other, more solid structures show signs of constant patching and repairs.
Lupus, who has lived in the district for more than 50 years, shows me around the inside of a ruined government building, which was abandoned in the 1970s after being gutted by a flood that reached the rafters. Since then, it has never reopened – although it has been used as a set for several horror movies – and all around it newer buildings carry the scars of water damage. The city government, beset with other challenges, ignored these districts, and residents either adapted to the perennial floods or learned to accept them.
The calculus was changed by two disasters. In 2007, four-metre-deep floods swept through the city. In Bukit Duri, Lupus remembers the water reaching waist height on the second floor. "We had to make tents and sleep on the roof," he says. The 2007 floods were particularly shocking, not only for their scale but because the city was inundated both by rainfall and by seawater coming in from the coast – the land had subsided so far that storm surges carried water inland, engulfing whole neighbourhoods. More than 300,000 people were evacuated and 80 died.
Then, in 2013, several days of sustained rainfall overwhelmed the flood management infrastructure. Canals collapsed and clogged, and the flooding spread beyond the poorer, lower-lying areas of the city and into the central business district. Around 45 people died and thousands of households were evacuated.
That stirred the government into action. The then-governor, Joko Widodo – now Indonesia's president – ordered a large-scale renovation of the city's rivers, reservoirs and flood canals, which had become fatally clogged by decades of an approach to waste management that mirrored the randomised sprawl of construction in the region. Controversially, under an initiative euphemistically called "normalisation" some informal settlements on riverbanks were bulldozed to widen the waterways.
In Muara Baru, a four-metre-high wall was built on one bank of the Ciliwung, protecting the community on that side of the river from most of the smaller floods – although water still spills over in the wet season.
At the same time, the national government began to look at coastal defence in earnest. It launched a new project, the National Capital Integrated Coastal Development, or NCICD, and called in a coalition of international experts, most of them from the Netherlands, which has turned its own centuries-old experience of protecting its low-lying shoreline into a global industry. Among them was Victor Coenen.
Coenen, a tall, genial Dutchman, moved to Jakarta six years ago to head the project with Witteveen+Bos, a Dutch engineering company. His first job was to weigh up the possible scenarios, starting with looking at the feasibility of simply giving up on north Jakarta.
"There are places on the north coast of Indonesia where abandoning is the cheapest way," Coenen says. "Like in the UK, coastal areas are now being given back to the sea because it's simply too expensive to protect them. But that is not the case in north Jakarta." His team calculated that the economic cost alone would be around $200 billion (#158bn), before the human cost of moving between two and three million people was taken into account. "Where would you relocate [the people]?" he says. "So that abandonment scenario was quickly: 'Forget about it. That's not possible.'"
The second option they examined was to build defences onshore. This is not unprecedented. After the Tohoku earthquake and tsunami that hit Japan in 2011, the government ordered the construction of nearly 400km of sea walls on the country's northern coast. Those walls, some more than 12 metres high, cost more than $12bn (#9.5bn).
For Jakarta, Coenen and his team calculated that they would need walls five to seven metres high, stretching the length of the bayfront, with massive pumping stations behind the walls and deep reservoirs that could hold overflows. That would still mean moving large numbers of people and buying up large areas of land.
"The urban impact of such an on-land solution is huge," Coenen says. "Imagine seven-metre-high sea walls on the coastline. It's like a reversed aquarium. The water's outside, the people are inside. It's a scary thought."
An onshore project would have been prohibitively expensive, socially disruptive – and may not even have worked in the long term. "Quickly we said the on-land solution might be the solution for the short term, but it will not be the long-term solution. We have to go offshore," Coenen says.
Construction soon began on a short-term fix, upgrading the existing sea walls, adding 1.5 to 2.5 metres on top of existing structures and filling in gaps in the defences, while more engineers, consultants and government agencies began to work on the shape of the offshore plan.
The island of Pantai Indah Kapuk is linked to the Javan mainland by white half-helix bridges that, uncharacteristically for Jakarta, are clear of traffic and lined with neat pavements. In the glare of the dry season Sun, it looks like a Florida resort village. Multi-storey condos overlook a food court with bright-coloured awnings, while plastic corporate art adorns the grass verges and roundabouts.
The finished buildings are clearly unoccupied, and away from the bridges all work has stopped and the remaining land is fenced off behind high construction hoardings. Idle earth movers are lined up in neat ranks, watched over by security guards, and several sites are going to seed, the bare steel poles of rebar rusting where they juts out of unfinished foundations.
Pantai Indah Kapuk is a vestigial piece of the first – now-cancelled – plan for Jakarta's flood defence, and an emblem of how commercial concerns hijacked the city's attempts to secure its coastline; it is one of 17 islands that were due to be reclaimed from the bay using millions of tonnes of sand and concrete. Those islands were due to stud the inner curve of Jakarta Bay, providing homes for hundreds of thousands of residents, as well as office space, malls and even a new airport. Out in front of them, the plan called for a massive series of sea walls which, viewed from above, would make the shape of the Garuda, a mythical bird that is the emblem of Indonesia. Artists' impressions show a gleaming new city on the sea, a tropical Dubai.
The project, which was unveiled in 2014, was out of control from the very beginning.
"The consultants at that time were given the boundary conditions that the Indonesian government should not invest one penny in this," says Peter Letitre, a groundwater expert from Dutch hydrological institute Deltares and an advisor to the NCICD project. "A typical Jakarta project. The only way to finance it was through land reclamations."
The project, which Letitre calls "megalomaniac", became self-perpetuating: each new part needed more infrastructure and more defences, which meant raising more money, which meant yet more reclamation to make land to sell to developers. As it grew, it began to suck in other projects onshore and offshore. Local media reported that the total cost of the project would be $40 billion (#31.5 billion), and although the engineers and architects disputed that figure, it stuck.
"All of the things that were needed to improve the infrastructure of Jakarta were put under the price tag of the NCICD project. So the amount became higher. And the only way that you could make money back was through more land reclamation," Letitre says. "That was the trap... it became so huge and massive and expensive."
The project was flawed beyond the price tag. To manage the difference in water levels on either side of the sea wall, it called for deep reservoirs on the land side that would have collected the runoff from the rivers, with huge pumps that would have transferred the excess out to sea. However, Jakarta's rivers are some of the most polluted in the world. Without a drastic – and frankly unrealistic – improvement to the city's water quality, the area behind the "Great Garuda" could have become a giant lagoon fouled with the combined effluent of the megacity – an environmental hazard in its own right. One hydrologist, who worked on the project and asked not to be named because he is still working with the national government, says the solution was "a completely insane idea" that would have resulted in polluted water backing up into the city.
Some opponents suspected that the reclamation was the whole point. Since the 1990s, developers had been pushing for the creation of new land; the 17 islands were planned before NCICD was conceived and hastily brought under the umbrella of the project. "It was never about flood defence," says Tubagus Soleh Ahmadi, the executive director of Walhi Jakarta, a local environment campaign group. "This was an economic project."
Opposition to the scheme grew. Coastal communities worried about the destruction of their fishing grounds; inland, discontent lingered over the clearance and eviction of riverside settlements from the last time that the government had "prioritised" flood defence. Work stopped and started throughout 2016. In gubernatorial elections the following year, one candidate, Anies Baswedan, made a moratorium on land reclamation a major part of his platform. He won, and stood by his promise.
The Great Garuda was quietly shelved, although it lingers in the city's imagination, due to the absence of any public announcements about any replacement. Reclamation work was halted. By the time the moratorium took effect, four of the 17 islands – including Pantai Indah Kapuk – had been built. The developments on top of them are now frozen in time.
As the NCICD plan morphed and evolved and stalled, Jakarta kept flooding, and attention shifted to the other side of the equation – stopping the city from sinking any further.
In charge of that rearguard action is Oswar Mungkasa. Tall and professorial, dressed in the military-style khakis of the city's civil servants, Mungkasa is a 25-year veteran of national and regional government and Jakarta's deputy governor in charge of spatial planning and the environment. Since 2017 he has also held the title of "chief resilience officer" – a wide-ranging role that gives him responsibility for fixing everything that threatens the running of the city, from rubbish collection to racial harmony to preparing for a giant earthquake. These challenges grow every year in proportion to the population; 200,000 people move to Jakarta each year, drawn by and adding to the city's huge economic expansion. "The urbanisation process has been too fast. We cannot keep up," Mungkasa says.
His role has hammered home just how fragile the megacity is. "If something happened in Jakarta, our food only lasts maybe one week," he says.
The sinking of the city is one of the most pressing challenges on Mungkasa's slate. It is also one of the most intractable, a complex problem that involves dealing with some of the city's longest-standing environmental issues. It all starts with clean water.
The mechanism by which Jakarta is settling into its foundations has been understood for decades. The soft soil underneath the city is held up by the pressure of water in aquifers and reservoirs deep below the surface. Removing that water lowers the pressure, and the land above it sinks. That issue is exacerbated by building heavy structures on the surface, and by coating it with impermeable materials, like concrete, which prevent water from seeping back down and recharging the subsoil reservoirs.
It is a challenge that has presented itself elsewhere, including in Tokyo and Venice. "They stopped it by simply starting to regulate this enormous overuse of water," says Jorgen Steenfelt, technical director of marine and foundation engineering at consultancy COWI and an expert on urban water issues. "At the root, it's regulation. If you can't regulate it, you have no control over what is happening. If it's the wild west and everybody can do what they please, then there's no stopping it."
Even that is just a palliative measure – groundwater levels would need to be recharged, restoring the pressure deep underground to prevent further slippage. Tokyo, which effectively halted all groundwater extraction by the late 90s, is still sinking by around one centimetre every year.
There are examples where restoring these water levels is being attempted. In Chesapeake Bay, in the US, hundreds of thousands of cubic metres per day of recycled wastewater are being pumped into aquifers to stop the subsidence of farmlands. In the fields around Bangkok, which has also been sinking for decades, the government has put in place regulations requiring farmers to include reservoirs in their properties to store rainwater and allow it to seep back into the soil.
However, there is no precedent of anything of this sort being attempted in a city close to the scale of Jakarta, Steenfelt says. The sheer size of the challenge, and the city's historical failures to meaningfully tackle the problem, mean that he is not hopeful.
"Jakarta is a sad story, and it's very difficult to see how it could be resolved because of the timescale and the number of people," he says. "It's a gigantic task. But you can't just close your eyes."
Mungkasa is well aware of the scale of the problem, and of the rapidly approaching deadlines. By his estimates, the city needs to stop pumping groundwater entirely within five years. It is an enormously ambitious target. "We have to be confident that we can do it," he says.
The alternative is unthinkable. "If this is not a success, if we fail to do it, to reach our target, then there are two [plans]. Plan B and Plan C. Plan B is we need to build another sea wall. The second, of course, if the sea wall still doesn't work, we have to move the people... to move people is really unbelievable."
His first job is to rationalise the ways in which Jakarta has managed, or mismanaged, its water supplies. The city's administration is as tangled and chaotic as the city it serves, with dozens of overlapping ministries, agencies, parastatal companies and private contractors all working on ostensibly similar missions but often not talking to one another. In sanitation alone, Mungkasa says, there are three agencies that share responsibility for managing sewage, which will occasionally duplicate each other's work in some areas while collectively neglecting others.
Even where there are laws governing how water should be treated and recycled, enforcement is poor. An estimated half a million people have no access at all to sanitation, and defecate straight into the water courses. The company responsible for wastewater treatment for the city says it covers around 11 per cent of the total area of the city; Mungkasa insists the reality is closer to five per cent.
"I think this [lack of co-ordination] happens in every city in developing countries," Mungkasa says. "But in Jakarta it is really killing us."
Mungkasa and his small "resilience secretariat" – a group of three young urban experts in a windowless room in an annex of city hall – are putting together a new "master plan" for the water sector, which will be unveiled later this year. It is likely to advise more regulation or taxes on groundwater extraction, and to encourage public sector organisations to lead by example; currently even ministries tasked with fixing the problem of flooding are themselves pumping out groundwater.
Cleaning the rivers will be a huge challenge. As well as human waste, most of the 13 rivers have been contaminated with industrial chemicals, including heavy metals, and it could take years just to stop the ongoing pollution. In the meantime, the city will need to expand its use of rainwater harvesting and other emerging water collection and purification technologies. Property developers are being encouraged to build ponds and reservoirs into their designs in order to store rainwater, and to switch to vertical drainage, pumping used water down into the earth rather than letting it flow back into the rivers.
Mungkasa – who has a politician's gift for making out that intractable problems have simple solutions – says that he has no doubt this master plan will work. "If we provide people with water, we will stop the subsidence," he says.
Few experts would question that, but many are skeptical about whether the necessary measures can be taken in time – if at all. "It could work," says Deltares' Peter Letitre. "It depends on the sense of urgency... we don't think it's impossible."
In June this year, a new plan for Jakarta's offshore sea wall was approved in principle by the ministry of public works, although it could be another 12 months before it gets its final approval and work can start.
NCICD II is modest in scope compared to the Great Garuda, but it is still vast, comprising of a single open dyke that will run 40km across the bay of Jakarta, from the port at Tanjung Priok in the east to the Soekarno-Hatta airport in the west. Lessons have been learned from NCICD I's failure. Gone are the reclaimed islands and the property developments, although a toll road will be built along the dyke in order to help secure financing for the project. The outer dyke is scheduled to begin construction in 2023.
"Reclamation is now off the table, and we're back to flood defence," Coenen says. "We were always against merging the economy and the flood defence in such a way that you depended on economic development for your flood defence. The Great Garuda was a real estate project, basically. But real estate goes up and down. Real estate crashes. You have periods where the market goes up into the sky. What you don't want is that your flood defence depends on that dynamic."
He is also aware, he says, that more work is needed to make sure that the project's architects understand the complex social issues that it brings up. Discussions have often been led by foreign experts and by engineers who are unaware of or unwilling to confront the messy reality on the ground. "You have to commit yourself to the social issues that you are creating," he says. "And don't just scratch the surface, but go very deep."
Coenen estimates that the total cost of the dyke will be around $4bn (#3.2bn), with another $4bn for pumping stations and on-land reservoirs – although that could change. "We are heavily dependent on the sand price. If it goes up one dollar per tonne, we will have to pay Euro 1bn extra, because the volumes are so big. A very small uptick of the price has enormous consequences," he says.
The dyke will act as a huge breakwater to reduce the height of waves entering the bay, and to take the momentum out of storm surges so that they do not wash over the inner sea walls. Crucially, it leaves room for failure. The base level plan assumes that land subsidence will be addressed, but it includes contingencies in case it continues, or if sea level rise occurs faster than anticipated.
"Two scenarios are still on the table: close the bay, or keep it open," Coenen says. "That really depends on what will happen with the land subsidence. If the land subsidence remains as it is or even accelerates, then you have to go to this mega closed system. If the land subsidence is managed then we can still keep the bay of Jakarta open."
That call will be made around 2030 when, at current rates of subsidence, the existing walls will be obsolete. Closing the dyke would be a massive project in its own right, and raise all of the same issues of water pollution that arose during the Great Garuda, but by then there would be very few other options left on the table.
With the timescales for the project so long, there is still a chance that politics could once again get in the way, and Coenen is not entirely certain that everything will be resolved before Jakarta reaches the point of no return.
"I think you can compare it to the problem of climate change – where governments do see the problem, but they postpone very expensive and difficult measures towards the longer term and only focus on the quick wins... That is the nature of this kind of problem and the way politicians solve it," he says. "Indonesia is a country where things can stall, or happen very fast. It doesn't seem like this country has something in between, you know. In an optimistic view, this is a just-in-time society. But in a more realistic view, we're always a little bit too late."