Post-water CliFi — Los Angeles and Riyadh

NB: I wrote these four CliFi scenarios* in 2019 for a paper on life in a “post-water” world, but they had to go, so I am posting them here for your enjoyment (or horror). Please tell me what you think!

Scenario 3: Los Angeles loses its aqueducts

The City of Los Angeles imports about 90 percent of its drinking water via the Los Angeles, California and Colorado River aqueducts, which cross hundreds of kilome- ters of agricultural land on their way to the city (Lin, 2017). Although Los Angeles has been fighting others for decades over water rights, extractions and exports (Got- tlieb & FitzSimmons, 1991; V. Ostrom, 1953; Zetland, 2008), rights to these water sources are relatively secure. The story with conveyance is different because the aqueducts that bring water to Los Angeles could suddenly fracture as stress in- creases due to uneven ground subsidence caused by overdrafting groundwater. Let’s unpack that causal chain.

First, there’s a long history of California farmers using groundwater when surface supplies are absent or reduced due to a drought. In cases of “overdrafting” — using more water than is replaced by natural recharges — ground levels can subside as water is abstracted. In California’s Central Valley, many areas fell by 3–6m (with the extreme of 8.5m) between 1926 and 1970. In the past 50 years, there have been efforts to reduce overdrafting and subsidence (one persistent reason/excuse for importing more surface water to the region is to relieve pressure on aquifers), but normal and drought-response overdrafting continues to lower ground elevations. Between 1995–2010, the ground dropped by 22–60cm in the area through which the Colorado River Aqueduct passes (Sneed, Brandt, & Solt, 2014). In the southern Central Valley (through which the California Aqueduct passes), ground levels dropped by 12cm in most of the area between 2007–2010, with local extremes of 90cm (Faunt, Sneed, Traum, & Brandt, 2016).

Changes in ground elevations wreck havoc with water conveyance infrastructure that must be massive enough to carry large volumes of water but also precisely sloped to maintain flow. Chronic subsidence requires ongoing monitoring and maintenance, but all bets are off if there was an earthquake on the San Andreas Fault, over which all three aqueducts (Lin, 2017) pass. A three-way failure would mean that Angelenos and the other 20 million residents of Southern California would lose 90 percent of their water supply.

Chaos would result. Although 50–70 percent of the the region’s residential drink- ing water is used for grass lawns and landscaping, it is hard to imagine people in the region coping, let alone cooperating. The first response — running the taps to fill bathtubs — would drain local and overwhelm pumps. Water distribution centers need to be numerous to prevent traffic jams that would result if the sprawl converged. These water troubles would be a lucky break. An earthquake is also likely to cause fires, and California is now said to have a “year-round” fire season (Economist, 2019).

Local water management institutions are somewhere between unprepared and lucky. They are unprepared with a system designed importing water, limited local storage capacities, and limited local supplies. Turning to local politics, it would take some time to agree on how to handle a crisis, given ongoing battles over everyday operations.

On the lucky side, people in Los Angeles and the area use vast volumes of water on landscaping, so a drastic cutback wouldn’t be too hard. Second, there is probably enough local storage to keep people going until emergency repairs of aqueducts were made. Finally, the earthquake would be during a rainy winter and not damage too much infrastructure.

In sum, Los Angeles is extremely vulnerable to the damages caused by overdraft- ing farmers whose consumption of a private good (irrigation water) is damaging the commons of the landscape. The city of Los Angeles and others in the region can probably get by if and when an accident happens, but they might be unlucky and pay dearly for ongoing poor management.

Scenario 4: Riyadh is stranded

The Saudi capital, home to at least 6 million people, gets half its water from desali- nation plants located nearly 500km away on the Persian (sometimes called Arabian) Gulf. The rest of the water comes from relatively local wells drilled deep into fossil (non-recharging) aquifers (Ouda et al., 2018). In the summer of 2027, demand for water is extreme due to average daytime temperatures of 36C, a culture of heavy water use, and population of over 8 million people. The Ras Al-Khair desalination plant (RAK) supplies 90 percent of Riyadh’s desalinated water due to the recent retirement of two older plants (Ouda et al., 2018), and SWCC (Saline Water Con- servation Corporation) is running it at full capacity in stifling heat, producing water that is raised 600m on its way to the thirsty distant capital. Groundwater pumps are working as usual, diverting their water into treatment plants whose potable water flows into one of the 50 storage tanks located around the capital (MEED, 2013).

Al Qaeda took these facts into account when planning their strike against the regime of King Mohammed Bin Salman, who has not only maintained diplomatic and military relations with the United States but also weakened the spiritual purity of his people. His reforms are not offensive because they allow women to drive, but because a majority of the population now goes to the movies instead of the mosque on Fridays. Donations have fallen as frivolous spending increases; Al Qaeda’s budget for justice is dropping.

The operatives work in five cells, unknown to each other. They have a date for action: August 15 2027, which is also the Prophet’s Birthday (Peace Be Upon Him). They have trained and prepared. They are ready.

In the early hours of the fifteenth, alarms sound in the RAK control center. Intake pipe #3 (out of 6) seems to be clogged at its mouth, which rests 30m below the gulf, 1.5km offshore. Control operators turn off pumps on #3 and order an inspection crew to go out. Due to the holiday, there are only two maintenance crews on hand, but this routine maintenance is not too difficult, so they leave shore within an hour. Just after 8am, and before that crew can report in, another alarm sounds as a cooling pump 22 shuts down after losing pressure. Perhaps a poor weld has burst due to load or heat. Roughly 15 minutes later, a radio cracks with static — someone from pumping station 4B is trying to get in touch. The control center staff, already distracted by two incidents, have a hard time understanding what the man is screaming, but they hear explosions and then static. Station 4B is 150 km away, on the pipe’s route to Riyadh. Pressure on line B drops to zero as the station goes offline. Only line A continues to operate, pumping water to Riyadh. Omar, the head of operations, suddenly realizes that the situation has exceeded the limits of bad luck. He triggers the emergency plan, which alerts the army, royal palace and SWCC staff that the Kingdom’s water is threatened.

Twenty minutes later, he’s still waiting for a response from the army, which attending upon a royal family pro-occupied with birthday festivities. The palace, likewise, is silent. His SWCC colleagues assure him that local storage is sufficient to meet demand while line B is repaired. Omar turns to report that news when two of his staff run up from opposite directions. “The maintenance crew was attacked by a drone on their way out to #3,” says one. “Their boat is on fire, and they have jumped into the sea.” The other man, barely listening to his colleague, says “we’ve also lost cooling on pumps 8, 17 and 32! Something is wrong with the systems. We’re losing capacity.” Omar flinches. The facility has multiple production units, but the leaks are scattered rather than concentrated. He tells the first one to call the navy. To the second, he says “reduce pressure — we need time to figure out what’s happening. Send out the other crew.”

Just then, the radio cracks to life. His colleague Dasan is calling from the capital. “Omar, we’ve got a real problem: Our storage tanks have been attacked by drones with explosive devices. We’re not sure which ones, yet, but we’re sending out crews.”

Just then, he gets a call from his wife. He doesn’t usually carry his personal phone at work, but his granddaughter is about to give birth any day. He picks up the phone: “Lala — I can’t speak now. We’ve—” She interrupts him: “Omar! Quick, look on Twitter!” He switches to his Twitter app and pales. The screen is filled with panic: #RiyadhDies and #poisonwater are trending in English and Arabic.

The loss of desalinated water supply, combined with mistrust of groundwater quality and ignorance of water safety (Alamri, 2019; Al-Omran, Al-Barakah, Al- tuquq, Aly, & Nadeem, 2015), leads to widespread panic. Tens of thousands flood hospitals and clinics with real or imagined sicknesses. Normal procedures grind to a halt. Over a hundred thousand cars flee the city. Thousands die in car accidents. Thousands drive into the desert to escape traffic jams, but their trust in GPS is defeated by sand washes and hidden cracks. Many stranded families die of thirst, clutching useless cell phones.

In the aftermath, over 50 officers and SWCC staff are arrested on the King’s orders. Five men are found guilty of “failure to protect the nation” and executed. Omar and Dasan are lucky to only be fired, in what the engineers call (under their breath) “Al Qaeda’s Plan B.” The death of 14,500 residents triggers an exodus of families from the capital to cities closer to reliable water supplies. By 2030, the capital’s population has dropped by 1 million, housing prices have dropped by 40 percent, and three-fourths of the staff of international firms have left the country.

These incidents are fictional, but not impossible. Riyadh’s vulnerability to sup-
ply disruptions is well known, as are its excessive demands (around 250 liters/capita/day) and 30 percent water losses (Ouda et al., 2018). Stress on this system will only grow worse as the capital’s population grows and temperatures rise. Riyadh will be able to manage its water with luck, but climate change will test that luck.

* CliFi (or Climate Fiction) draws on science fiction’s long tradition of thinking about possible future by combining human behavior with future technology. In the case of CliFi, the future “technology” is a changing climate, and these examples look into climate-related post-water shocks. I used this speculative method for two volumes of “CliFi” short stories [free to download] that I edited and published a few years ago.

  • Alamri, A. (2019). Water Usage and Human Health: A Preliminary Study in Riyadh, Saudi Arabia (Unpublished master’s thesis). Oregon State University.
  • Al-Omran, A., Al-Barakah, F., Altuquq, A., Aly, A., & Nadeem, M. (2015). Drinking water quality assessment and water quality index of Riyadh, Saudi Arabia. Water Quality Research Journal, 50(3), 287-296. 
  • Economist. (2019). Alaska hotshots. The Economist, 25 July.
  • Faunt, C. C., Sneed, M., Traum, J. A., & Brandt, J. T. (2016). Water availability and land subsidence in the Central Valley, California, USA. Hydrogeology Journal, 24(3), 675-684.
  • Gottlieb, R., & FitzSimmons, M. (1991). Thirst for Growth: Water Agencies as Hidden Government in California. Tucson: University of Arizona Press.
  • Lin, R.-G. I. (2017). California could be hit by an 8.2 mega- earthquake, and it would be catastrophic. Los Angeles Times, 8 Sep.
  • MEED. (2013). Riyadh plans water storage programme. Middle East Business Intelligence, 24 Apr.
  • Ostrom, V. (1953). Water Supply (Vol. VIII). Los Angeles: Haynes Foundation.
  • Ouda, O. K. M., Khalid, Y., Ajbar, A. H., Rehan, M., Shahzad, K., Wazeer, I., & Nizami, A. S. (2018, 02). Long-term desalinated water demand and in- vestment requirements: a case study of Riyadh. Journal of Water Reuse and Desalination, 8(3), 432-446.
  • Sneed, M., Brandt, J. T., & Solt, M. (2014). Land subsidence, groundwater levels, and geology in the Coachella Valley, California, 1993-2010 (Tech. Rep.). Reston, VA: U. S. Geological Survey.
  • Zetland, D. (2008). Conflict and cooperation within an organization: A case study of the Metropolitan Water District of Southern California. Doctoral dissertation, UC Davis (Agricultural and Resource Economics). 

Post-water CliFi — Amsterdam and Jakarta

NB: I wrote these four CliFi scenarios* in 2019 for a paper on life in a “post-water” world, but they had to go, so I am posting them here for your enjoyment (or horror). Please tell me what you think!

Scenario 1: Drought in Amsterdam

Amsterdam is located in Nord Holland, a province in the west of the Netherlands that lies in the Rhine–Meuse–Scheldt delta, and thus at the foot of several major rivers. The local ecology and many methods of managing water take excess water and cooler temperatures for granted, but these gifts are not forever. As I write (July 2019), the Netherlands has just reached its highest recorded temperature (40.4C), and the heat is causing problems for people, farmers and infrastructure. What will happen if these temperatures become common-place? How would Amsterdam cope with the risks from heat and drought?

The good news is that water for humans is unlikely to run out. Amsterdam sits adjacent to the IJsselmeer, the largest freshwater lake in the Netherlands, which has a surface area of 1,133 km2 and average depth of 4.4 meters, meaning a volume of around 5 km2 (Rijkswaterstaat, 2017). That’s over 50 times Amsterdam’s current water use (Waternet, n.d.).

The bad news is that drought would also damage local ecosystems. Falling groundwater would mean dead vegetation and weaker trees, some of which would fall in otherwise “normal” storms. Increased heat will lead to infrastructure failures, such as draw-bridges that lock shut due to metal expansion, or roads, rails and runways that are too hot to bear trucks, trains or planes (Staff, 2019). Individuals faced with “public bads” of hot air and drying ground, will install and use air conditioners and spray drinking water on parched gardens. Poorer citizens and marginal businesses will suffer — unable to afford the costs of equipment or the energy to run it. Some city services will fill the gap, but productivity and happiness will drop (Deryugina & Hsiang, 2014; Kjellstrom, Holmer, & Lemke, 2009).

In terms of institutions for managing water, perhaps the only possible responses to the public bad would be a program of improved public goods to increase local cooling, such as denser tree cover and perhaps restoring water-flows to canals that were filled and converted into roads decades ago. Ignoring sea-level rise and storm surges, Amsterdam should be able to limit the risks from drought and heat.

Scenario 2: Jakarta floods

Jakarta, the capital of Indonesia and home to over ten million people, has been having trouble with sinking land, saltwater intrusion and floods for decades (Pur- nama & Marfai, 2012). These three problems can be attributed to the abstraction, use and discharge of freshwater from local aquifers. This classic case of a tragedy of the commons results from the private use (withdrawal) of freshwater from the common-pool aquifer that everyone can access but also which keeps the land from sinking.

There are two main solutions to these dilemmas: to mitigate or adapt. Mitiga- tion would require building infrastructure to import fresh water and inject treated wastewater under the city, but that’s not happening. Instead, there are efforts to adapt by raising dikes to protect sinking land and building a barrier island to slow down storm surges that risk flooding land (Sherwell, 2016). This “Garuda Project” is complex and controversial, but it is surely cheaper than rebuilding drinking- and wastewater systems to serve ten million, mostly poor residents.[19] Unfortunately, the Garuda project might deplete funds, weaken solidarity, and increase risk. Where’s the post-water element? The people (and leadership) of Jakarta need to live as if they are on an arid island, but they are consuming scarce fresh water as if it’s abundant, which puts them at risk of getting too much salty water.

Let’s assume that the Garuda project is built, and business as usual continues. The ground continues to sink, but the barrier island has created a lagoon on the city’s shore, and flooding has decreased. Barrier island residents live apart from fellow citizens whose houses lie 3m below sea level are protected by higher walls.

Now introduce surprisingly fast climate change based on exceptional methane releases (Weitzman, 2011). Increasing temperatures and greater atmospheric activ- ity means larger typhoons (called cyclones or hurricanes elsewhere in the world). Although Jakarta is not usually struck by typhoons, Typhoon Indra strikes in 2035 with high winds and a storm surge that overwhelms the barrier island (flooding hundreds of expensive cars and cutting power to the whole island) and flows into Jakarta. Thousands die as 8m waves crush down on sunken neighborhoods. When the storm recedes, half of the four districts closest to the sea is gone, replaced by a new shoreline and “beaches” of rubble, crushed cars and bodies. Half a million people are homeless. Fifty thousand are dead or missing.

The overwhelmed local government asks for help. Foreigners bring money and ideas, but no consensus recovery plan. Millions leave the capital for inland regions, hungry and desperately poor. Domestic aid is hard to organize or fund without political leadership, and the rich (such as those on the barrier island) do not feel inclined to pay. They withdraw further into their climate-proof enclaves. The poor cannot grow their own food. Local farmers do not have enough water to grow even their typical crops. Hunger intensifies. Many are sick from drinking polluted water. Aid workers do their best, but a significant minority support a new group that goes by the handle @newgaruda.

This example focusses on three factors: inequality, underinvestment, and climate risk. Inequality makes it hard for people to cooperate, as they “other” neighbors. Underinvestment (in mitigation) falls in to the “penny wise pound foolish” trap of trying to cover a basic problem with a partial (and inadequate) solution. Climate risk is present in all these examples, but this example uses uncertainty — a big storm in an under-prepared location.

* CliFi (or Climate Fiction) draws on science fiction’s long tradition of thinking about possible future by combining human behavior with future technology. In the case of CliFi, the future “technology” is a changing climate, and these examples look into climate-related post-water shocks. I used this speculative method for two volumes of “CliFi” short stories [free to download] that I edited and published a few years ago.

  • Deryugina, T., & Hsiang, S. M. (2014, December). Does the Environment Still Matter? Daily Temperature and Income in the United States. NBER Working Paper, 20750.
  • Kjellstrom, T., Holmer, I., & Lemke, B. (2009, Nov). Workplace heat stress, health and productivity – an increasing challenge for low and middle-income countries during climate change. Glob Health Action, 2.
  • Purnama, S., & Marfai, M. (2012). Saline water intrusion toward groundwater: Issues and its control. Journal of Natural Resources and Development, 2, 25-32.
  • Rijkswaterstaat. (2017, Oct). Natura 2000 Beheerplan IJsselmeergebied 2017–2023 [IJsselmeer region management plan] (Tech. Rep.).
  • Sherwell, P. (2016). $40bn to save Jakarta: the story of the Great Garuda. The Guardian, 22 Nov.
  • Staff. (2019). It has never been hotter since records began: temperature tops 39c. Dutch News, 24 July.
  • Waternet. (n.d.). Ons drinkwater [our drinking water].
  • Weitzman, M. L. (2011). Fat-tailed uncertainty in the economics of catastrophic climate change. Review of Environmental Economics and Policy, 5(2), 275- 292. 

Clean water and American waterworks

FC mentioned that Werner Troesken, who died recently and unexpectedly,  had worked in the same area as me (water services). I went to his Google Scholar to see what he had written and found — amazing! — that I had used his most cited paper (“Population growth in US counties, 1840–1990”) in my PhD dissertation (p84):

A literature review uncovered only one article that discusses the influence of water on urban growth. Beeson et al. (2001) find that precipitation has a significant positive effect on population density in the United States of 1840. By 1990, this significance disappears. In fact, precipitation has a negative correlation with population growth in the 150 years after 1840. These results correspond to what we know about water in the western US: As infrastructure has brought water to arid regions, people have moved from wet, colder areas to dry, warmer areas.

In fact, I just mentioned this result to someone last week.

As I browsed through Troesken’s other papers, I found two with interesting results. In “Municipalizing American Waterworks, 1897–1915” [pdf], Troesken and Geddes (2003) describe how private water companies, fearing seizures by municipalities, would underinvest in infrastructure and thus give municipalities an excuse (under-investment) to take them over! This damned-if-you-do, damned-if-you-don’t result makes sense in the contexts of “bilateral monopoly” (a single seller facing a single buyer) and “stranded assets” (an asset that, once built, cannot be moved or re-used in any way), since in those situations it’s hard to get one side to spend on an investment that the other side might use (or benefit from) without needing to pay. In our paper on the history of the Dutch drinking water sector [pdf], we ran into this situation with English companies underinvesting (or not investing) in Dutch cities that ended up building their own water and wastewater systems.

In a second paper, Ferrie and Troesken (2008) describe “Water and Chicago’s mortality transition, 1850–1925” via the direct reduction in typhoid fever due to access to clean water and a much larger indirect impact of lower mortality among those who survived typhoid but died of another disease. This paper is interesting because the indirect drop in deaths is triple the direct drop. In our drinking water paper, we did not get into the details on the benefits of clean water, but we surely would have cited this paper in support of wide, diffuse benefits.

It’s a pity that Professor Troesken’s life ended prematurely. We need more economic historians like him.

Addendum (1 Nov): I forgot that I had downloaded another paper (“Regime change and corruption: A history of public utility regulation” [pdf]), which I just read. This chapter is interesting for two reason. First, Troesken argues that ownership changes (from public to private ownership of utilities — and vice-versa) is driven by the need for regime change because the existing structure (either private or public) has been compromised by regulatory capture. Second, this paper — and its thesis — fits into my existing idea that the public-private cycle is driven by public underinvestment (to keep prices low) that lead to privatization, which leads to re-municipalization once those investments are made (and prices rise).

The Ostroms on funding municipal services

I’ve tagged this post “from the archives” because I intend to draw attention to old but useful work that is no longer in broad circulation.

Vincent and Elinor Ostrom gave their 1977 chapter the less-than-exciting title of “Public Economy Organization and Service Delivery,” but the work is important for two reasons.

First, it’s the earliest publication (that I’ve found) that describes, defines and compares the four types of goods in the 2×2 matrix* that I use all the time:

In this blog post, I explain how excludable goods are best managed in markets (via economic tools) while non-excludable goods are best managed through political (top down) or community (peer-to-peer) processes where people can be (and must be) jointly obligated to fund public goods (via taxes) or constrained from over-appropriating common pool goods (via regulation). Besides that difference, it’s also important to know that many goods can end up in any of these boxes, depending on governance and the (im)balance between supply and demand. In this paper, we explored how drinking water in the Netherlands cycled among them. 

Second, the chapter’s theoretical discussion is used to set up the case study of interest, i.e., how falling tax revenue may make it difficult for Detroit to offer municipal services (public goods), especially when those goods are “subtractable” (common pooled goods) and insufficient in supply to meet demand. Detroit’s decline — under the twin influences of white flight to the suburbs (where politicians did everything possible to keep tax revenues to themselves) and rising poverty and crime — is well known by now, but this paper’s date suggests that the decline was decades in the making and clearly understood. I wonder how the chapter (or the book that included it) was received by local citizens and policy makers.

It’s nice to learn from the wisdom of the past.

What’s in your drawer that’s worth a read?

  • Their 2×2 is turned on its side, and they are still a bit vague about the difference between public and common pooled goods — as they were in their 1971 paper that doesn’t really mention those latter goods.