User Report

** THIS IS ONLY SAMPLE

1) WHAT is the major environmental or resource ISSUE, HOW is it caused, and WHERE is it occurring?

Issue: Increased temperatures in the city. The Urban Heat Island manifests as warmer nighttime air temperatures in the city. Urban materials like concrete and asphalt store heat energy during the day and release it slowly at night, increasing minimum temperatures (the lowest temperature of the night). The Phoenix Urban Heat Island has higher air temperatures in the city core (downtown Phoenix) as compared to air temperatures in the rural locations (like Wickenburg and Queen Creek). The temperature differences between the city core and rural locations can range from 10-18°F, especially during the warm summer season.

2) WHAT is the main RESOURCE involved and HOW is it being impacted or developed? Describe a consequence

Resource: Energy. In warm climates like Phoenix, not being able to cool effectively at night has negative consequences for human health and comfort. This drives an increase in energy consumption as electric air conditioners are the main tool used to cool buildings and structures. A feedback loop is then created that helps perpetuates the heat island. Air conditioners expel hot air as they operate, which in turn contributes to warmer air temperatures, forcing even more energy usage.

3) WHO are the primary stakeholders or key players (people and/or organizations) driving this issue (or attempting to mitigate the issue) and WHAT are they doing?

Key Player: Phoenix Mayor and Council. The City of Phoenix was named as a finalist in the 2018 Bloomberg Mayors Challenge for their HeatReady proposal. The HeatReady program will help Phoenix prepare their responses to dangerous levels of heat. Additionally, the Phoenix Council is investigating urban forestry as a mitigation technique. Trees assist heat mitigation due to shade and transpiration (evaporating water into the atmosphere, which is a cooling process).

4) HOW is society (or individuals) being impacted by this issue (positively or negatively)? Provide one example.

Society impact: Poor and homeless suffer. The people most at risk from the Urban Heat Island are those that can’t cool effectively at night. In the Phoenix area, the communities most at risk are the poor and homeless. In warm climates like Phoenix, not being able to cool effectively at night has negative consequences for human health. On average, over 100 people die from heat-related illnesses every year.

5) HOW does this issue affect extenuating environmental situations or impact other natural resources? Provide one example.

Other resource: Water. As the city warms, temperatures in urban lakes increase. Warmer lake temperatures deplete dissolved oxygen which affects aquatic organisms in the lake. Warmer air temperatures also drive additional household water usage for pools, lawns, evaporative coolers, and landscaping. Adoption of urban forestry will increase municipal water usage, as well as increase evapotransporation, increase infiltration, and decrease runoff.

6) HOW does this issue relate to a learning concept, topic, theory, or term presented in this class? Provide one example.

Concept: Urban population exceeds rural population. There is a direct relationship between population and the Urban Heat Island. Urban materials like concrete and asphalt store heat energy during the day and release it slowly at night. Globally, as more people relocate and settle in cities, the Urban Heat Island will also increase. Urban sprawl, which is a problem in Phoenix, spreads the heat island over a larger area so that larger numbers of people are impacted.

Sourses:

Q1)

https://www.azcentral.com/story/news/local/arizona-environment/2017/10/19/heres-how-heat-discriminates-what-phoenix-doing-help-those-risk/561116001/

Q2)

https://www.epa.gov/heat-islands/heat-island-impacts

Q3)

https://www.azcentral.com/story/news/local/arizona-environment/2018/05/02/utilitys-tree-planting-program-phoenix-brings-shade-city-need/511319002/

Q4)

https://www.azcentral.com/story/news/local/arizona-environment/2017/10/19/heres-how-heat-discriminates-what-phoenix-doing-help-those-risk/561116001/

Q5)

https://www.epa.gov/heat-islands/heat-island-impacts

Q6)

https://www.epa.gov/heat-islands/heat-island-impacts

IN TOO DEEP The 100-year capitalist experiment that keeps Appalachia poor, sick, and stuck on coal Gwynn Guilford December 30, 2017

Nick Mullins didn't plan on working in the mines. But the industry was nearly impossible to escape. (Roger May for Quartz) The first time Nick Mullins entered Deep Mine 26, a coal mine in southwestern Virginia, the irony hit him hard. Once, his ancestors had owned the coal-seamed cavern that he was now descending into, his trainee miner hard-hat secure. His people had settled the Clintwood and George’s Fork area, along the Appalachian edge of southern Virginia, in the early 17th century. Around the turn of the 1900s, smooth-talking land agents from back east swept through the area, coaxing mountain people into selling the rights to the ground beneath them for cheap. One of Mullins’ ancestors received 12 rifles and 13 hogs—one apiece for each of his children, plus a hog for himself—in exchange for the rights to land that has since produced billions of dollars worth of coal. “I probably ended up mining a lot of that coal,” says Mullins, a broad-shouldered, bearded 38-year-old with an easy smile. There were other ironies to savor too. Mullins was a fifth-generation coal miner. But growing up in the 1990s, his father and uncles—all of them miners—begged him not to get into coal mining. “No one wanted to see you in the mines,” he says. “And they were all union miners too—had it good for a long time.” Those protections were gone by the time Mullins was growing up. The US government’s ongoing assault on organized labor through the 1980s and 1990s meant that the mammoth energy conglomerates that dominated the coal industry were free to open non-union mines with increasing impunity. But mining was still just as rough—replete with injuries, accidents, and black lung deaths. During the coal bust in the 1990s, Mullins’ dad was laid off from Bethlehem Steel’s mines. Mullins recalls living off the green beans his family had diligently canned during the good times, and watching his parents grow desperate. Go to college, they urged him. Mining offered no future. Mullins planned on following their advice. But he, like so many of his friends, family, and neighbors, soon found that the industry that has wreaked havoc on the economy of central Appalachia—composed of southwest Virginia, southern West Virginia, and eastern Kentucky—was also nearly impossible to escape. Passing on the costs Ask most Americans what they know about coal in central Appalachia, and they’ll tell you it’s a dying industry—one that US president Donald Trump famously vowed to revive during the 2016 election. “We’re going to put those miners back to work. We’re going to get those mines open … I see over here a sign, it says ‘Trump digs coal.’ It’s true. I do,” he told a rally in Charleston, West Virginia, in May 2016. “You’re going to be working your asses off.”

“Trump digs coal” signs at a rally in Charleston, West Virginia. (AP Photo/Steve Helber) But the idea that the region’s coal industry is dying is not quite true. For much of the hundred-plus years of its existence, the industry has been on a kind of artificial life support, as state and federal governments have, directly and indirectly, subsidized coal companies to keep the industry afloat. The costs of this subsidy aren’t tallied on corporate or government balance sheets. The destruction of central Appalachia’s economy, environment, social fabric and, ultimately, its people’s health is, in a sense, hidden. But they’re plain enough to see on a map. It could be lung cancer deaths you’re looking at, or diabetes mortality. Or try opioid overdoses. Poverty. Welfare dependency. Chart virtually any measure of human struggle, and there it will be, just right of center on a map of the US—a distinct blotch. This odd cluster is consistently one of America’s worst pockets of affliction. A deeply cynical capitalist experiment has taken place, in which coal companies are kept profitable by passing on the costs they incur to the public. At the root of these problems lies the ironic insight that struck Nick Mullins as he mined coal deep in the earth his family once owned. The extreme imbalance of land ownership in central Appalachia shifted the power over where and how Appalachians lived to corporations. The political and economic impotence of Appalachian residents that resulted has permitted a deeply cynical capitalist experiment to take place, in which coal companies are kept profitable by passing on the costs they incur to the public. The many ways in which politicians and coal barons have kept coal artificially cheap has, over the course of generations, devoured the potential of the area’s residents, and that of their economy. Central Appalachia’s problems stem from its distinctive history. But the pattern of its struggles is not unique. Across America, obscure clusters of misery are growing in number and concentration—as people get sicker, poorer, and more isolated than they were just a few decades ago. Thus untangling the knotty problems of central Appalachia holds lessons for the rest of the country about how imbalances of wealth and power, created generations ago, can trap places and their people in the past. The myth of coal jobs The experiment underway in central Appalachia began with subsidizing coal by suppressing household wealth. To the local politicians who sponsored this strategy, the idea was that making conditions favorable to outside corporations would develop the local economy, creating jobs and enriching residents. How has that played out? Not great, according to the most recent data from the Appalachian Regional Commission (ARC), the economic development partnership between federal, state, and local governments. Between 2011 and 2015, central Appalachia’s median household income was just over $34,000—around 63% of what the typical American household earned. Fewer than 60% of working-age adults had jobs or were looking for work, compared with 77.4% at the national level. At 8.6%, the unemployment rate in central Appalachians is 1.7 percentage points higher that of the US. Nearly a quarter of its residents were in poverty, compared with 15.5% nationwide. Even more remarkably, the region fares much worse on all these measures when compared to Appalachia as a whole, a span of country that reaches from eastern Mississippi to upstate New York. Throughout the last 125 or so years, West Virginia’s economy and state budget has depended far more heavily on coal than Virginia and, to a lesser extent, Kentucky, making it a useful proxy for the socioeconomic trends associated with the region’s coal-reliance. Though coal was first officially spotted in central Appalachia in the 1700s, major commercial production began in earnest only in the 1870s, after railroads finally burrowed through dense, rolling heave of the Cumberland mountains. Output grew exponentially. By the late 1910s, West Virginia was producing around 90 million tons of coal, and employing nearly 100,000 miners. US Steel’s coal operation was West Virginia’s biggest employer. How things have changed: After peaking in 1940, at 140,000—around a third of the state’s workforce—there are now only around 11,600 working miners in West Virginia. That’s the lowest number of coal miners the state has had since 1890, equal to less than 2%

of its workforce. It’s very clear that coal has failed to create jobs. It is, however, still the only way that many West Virginians can earn a decent middle-class wage. Nick Mullins’ story illustrates that paradox—and how decades of subsidizing coal profits over investment in human capital and technology has led to a dearth of opportunities for young central Appalachians. Attending a high school so underfunded that many teachers paid out of pocket to make copies of handouts, Mullins made the National Honors Society. But in eighth grade, an administrator had talked him out of taking the advanced-track classes, telling him his course load looked like too much work for him to handle. Not that he needed much of a push—those classes were filled with the coal-boss kids, who bullied anyone whose dad actually entered a mine. ”I wanted to be a proud, self-sacrificing guy who works two miles underground, where most people won’t go, to provide for my family.” When Mullins asked his guidance counselor how to apply to college as a junior, she told him that, without having taken advanced classes, it was pointless. For the next few years, he bounced from southern Indiana to Knoxville and back to Virginia in search of a decent living, taking whatever work he could find—a Christmas temp gig in Wal-Mart’s layaway department, carpentry jobs, night stockman at a grocery store, a janitor at a chemotherapy complex, clerk at Magic Mart. Nine jobs later, he finally caught a break, landing an $18,000-a-year technician position at a call center just opened by Crutchfield, a catalog and online consumer electronics retailer. After seven years, he was earning nearly $30,000. But with two kids now, and health care costs at $300 a month, he and his wife had nothing left each month to save. As a trainee coal miner, Mullins could earn what he was making at Crutchfield—but with better health benefits and a retirement plan with matching. Despite the ebbing of union clout, coal jobs still offered a level of economic security unheard of in other industries. There really wasn’t much choice to be made; just his family’s fears to push aside. Mullins talked himself into what he calls coal’s “cult of self-worth,” built around following his forbearers into hard, dangerous work. “I wanted to be a proud, self-sacrificing guy who works two miles underground, where most people won’t go, to provide for my family,” he says. Whenever he went into town, he made sure his “Deep Mine 26” ballcap was pulled snug over his head. Coal’s dominion has receded in West Virginia. Why, then, could Mullins find no well-paying work in central Appalachia except mining? Why has the structure of the central Appalachian economy changed so little? Henry Hatfield’s lament Nearly 100 years ago, Henry Hatfield, who became the governor of West Virginia in 1913, predicted this contemporary dilemma. At the time, West Virginia was producing nearly 70 million tons of coal—a roughly 30-fold increase from 30 years prior. Coal was making West Virginia rich. But Hatfield saw trouble ahead. “Rich as we are as West Virginians in our natural resources,” he said in his March 1913 inaugural address, “more than 80% of our fuel and raw material is utilized outside the state.” As a result, he said, low prices fixed “a standard of wages for the miner that is an injustice to him, by reason of the long railroad haul to market.” Indeed, coal operators “extracted more work at less pay from mountain miners, and this substantially lowered their cost of production,” writes historian Ronald Eller, professor emeritus at University of Kentucky, in his book, Miners, Millhands and Mountaineers. The coalmine owners were able to do this because central Appalachian miners were mostly nonunion, unlike their fellow miners in Pennsylvania or the Midwest. “In those days, competition almost universally took the form of competitive wage cutting,” wrote statistician Isador Lubin, the head of the US Bureau of Labor Statistics under president Franklin Roosevelt, in his 1928 book Wages and Cost of Coal. “The operator sold his men rather than his coal.” The resulting boost to profitability “has tended to increase the development of the nonunion fields more rapidly than the country required.” Governor Hatfield dreamed that his state’s endowment of coal would lead to local investment, fuelling factories, and enriching its economy. Instead, its resource wealth chugged out of the state by the train-load, along with the profits it generated. West Virginia produced a lot. But it manufactured almost nothing. And that’s still the case. Whereas US Steel once employed droves of West Virginians to gouge coal from inside mountains, for the last two decades, Wal-Mart has usually claimed the title of the state’s biggest employer.

Coal miners on the steps of a company store in West Virginia, circa 1938. (Library of Congress/Marion Post Wolcott) Since Adam Smith published his magnum opus The Wealth of Nations, capitalist societies have taken as faith that the key to prosperity is specializing in what they’re naturally good at. Central Appalachia’s economy seems to rest firmly on this logic. It has some of the finest-quality coal on the planet, and so it has sought to specialize in selling coal. Why, then, is it so much poorer in both wealth and well-being than the rest of the US? Places with less economic complexity tend to grow more slowly and have much bigger gaps between rich and poor. Perhaps our assumptions about the scale of “specialization” are too simplistic. A new field of research by Cesar A. Hidalgo, a statistical physicist at the Massachusetts Institute of Technology Media Lab, and Ricardo Hausmann, an economist at Harvard University’s Kennedy School of Government, has begun probing that possibility by measuring areas’ “economic complexity”—the mix of products a country makes. It turns out that places with less economic complexity tend to grow more slowly and have much bigger gaps between rich and poor, even after taking into account factors like income and education levels. Why is economic complexity good? For one thing, the complexity and diversity of products made in a region tend to be a proxy for the knowledge and know-how of its workforce. This makes workers more valuable and gives them a broader range of career options, upping their bargaining power. By contrast, countries—or regions like Central Appalachia—that make relatively few things tend to constrain people’s opportunities to learn new skills, whether on the job or at school. The curse of weak institutions Economic complexity illuminates one dimension of central Appalachia’s struggles. Another source of insight comes from research by Daron Acemoglu, an MIT economist. In Why Nations Fail, which he co-wrote with University of Chicago political scientist James A. Robinson, the duo explore why some resource-rich nations are rich while others remain poor. They argue that the fate of a nation is determined in large part by its economic institutions—its financial systems, tax regimes, property rights laws, labor institutions, and markets, among other things. When economic institutions are inclusive, they level the playing field among businesses, create competitive markets, encourage investment in new technologies, and enable people to acquire skills to pursue their talents, according to Acemoglu and Robinson. The institutions are underpinned by a political system that empowers a broad base of citizens to influence political decisions, preventing a single interest group from holding sway. The fate of a nation is determined in large part by its economic institutions—its financial systems, tax regimes, property rights laws, and markets. Extractive institutions, on the other hand, concentrate power among the few. This structure encourages the elite to warp institutions to enrich themselves and their political allies, which tightens their stranglehold on institutions all the more. “Different patterns of institutions today are deeply rooted in the past because once society gets organized in a particular way, this tends to persist,” Acemoglu and Robinson write. While inclusive institutions drive “virtuous circles” of self-perpetuation, extractive ones fuel “vicious circles.” Natural resource wealth can amplify these dynamics. Places with weak institutions—Algeria or Nigeria, for instance—typically succumb to the “resource curse.” Those with strong institutions, like the UK or Norway, tend to distribute that wealth more broadly. The destructiveness of extractive institutions can be hard to discern when times are good. Even though wealth is unequally distributed, the majority of people tend to be buoyed by economic growth. The real trouble arises when growth falters. Instead of letting new sectors emerge to replace the struggling ones, the elite protect the old guard, a key source of their power and wealth. This thwarts creative destruction—the concept pioneered by economist Joseph Schumpeter to describe how the failure of inefficient enterprises frees up resources for more productive firms.

Though central Appalachia isn’t a sovereign state, Acemoglu says that his and Robinson’s framework applies to the region’s lo ng-run problems. He points to research done by economist Robert Tamura, of Clemson University, on the long-lasting effects of slavery on the economic development of the American South, as well as to analysis by Melissa Dell, a Harvard University economist, that shows how areas of Peru and Bolivia subjected to forced mining labor systems in the 16th through 19th centuries currently suffer from lower education levels and investment in public goods and services. The vicious circles cementing central Appalachia’s extractive institutions don’t go back that far. But they do predate the region’s coal boom by a century. Land bonanza Why has the coal industry been permitted so much free rein over central Appalachia, despite the obvious toll it has taken on Appalachian residents? For one thing, the people responsible for devastating the area don’t actually have to live there and experience the consequences of their actions. By the time America won its independence, central Appalachia had already been carved into estates owned from afar. The British crown and colonial authorities granted huge tracts to court favorites, Virginia planters, and eastern capitalists, according to historian Wilma Dunaway’s book, The First American Frontier. From there, land accumulation by distant capitalists snowballed. In the 1790s, Virginia—followed by Kentucky—began selling cheap warrants for frontier land. By 1810, absentee investors owned around 93% of what’s now West Virginia and at least three-quarters of eastern Kentucky, according to Dunaway. Central Appalachia’s extractive institutions allowed the region’s natural wealth to be redistributed to shareholders in faraway metropolises. It wasn’t just far-flung fat cats cashing in. Local merchants, lawyers, judges, and politicians also snapped up land. Largely through control of state governments, this homegrown political and business aristocracy bent policy to enhance that land’s value, says historian Lou Martin, history professor at Chatham University in Pittsburgh. “Central Appalachian states were anxious to get outside investment and took steps to protect capital, making taxes more favorable to large companies,” he says. After the railroads finally bore through central Appalachia’s mountains, in the late 1800s, the coal industry followed a similar development path. At first, smaller coal operators abounded. But by the early 1900s, big, out-of-state companies gobbled up small, locally-owned companies, according to Eller, the University of Kentucky historian. We typically think of “wealth redistribution” as shorthand for taxing corporations and the wealthy to give money to the poor. Central Appalachia’s extractive institutions allowed exactly the reverse—for the region’s natural wealth to be redistributed to shareholders in faraway metropolises, and in a sense, to America’s emerging middle-class consumers who benefited from artificially cheap energy, at the expense of ordinary Appalachians. Typically, tax policy would be used to offset the profiteering of absentee corporations. Not in central Appalachia. In 1902, West Virginia’s state tax commission observed that coal “is being mined from and transported beyond the State, and continuously subtracted from the State’s property values without paying to the State one cent of tribute.” The report’s author warned that “when wholly exhausted, nothing but a comparatively worthless shell will be left behind… the kernel having been removed without the requirement of contribution to the State’s support.” Political bosses often doubled as coal operators, and “company men” were frequently elected to state senate. Had West Virginia moved to change its tax system, it might have used coal companies’ taxes to fund public investments that could have given residents more economic freedom and better quality of life. But supporters of the extractive regime were vastly more powerful. Political bosses often doubled as coal operators, and “company men” were frequently elected to state senate, typically serving as head of the mines and mining committee. Through the vast tentacles of its influence, the coal lobby tied up lawmakers’ efforts to tax coal and pass mine safety laws, despite persistent efforts by Progressive politicians. Only in the 1970s did states pass laws that claimed any share of the earnings on coal extraction on behalf of their citizens. But even with plum tax laws, central Appalachia’s remote coalfields still couldn’t easily compete with the coal mined in Pennsylvania, Indiana, Illinois and elsewhere. To compete—and to reap profit—companies made their coal cheaper not only by suppressing miners’ pay, but also their purchasing power and, therefore, their standard of living. As demand for central Appalachian coal boomed in the late 1800s and early 1900s, displaced farmers and landless sons filled the new ranks of wage laborers entering the mines. To supplement their ranks, companies—sometimes using coercive tactics—recruited poor workers from the American South and even Europe. The region’s population exploded.

With few existing settlements to house this new laboring class, the vast majority was housed in coal camps. A Senate report found that around four-fifths of West Virginia miners lived in company-owned coal camps, and more than two-thirds of Kentucky miners, according to Eller. That compares with 9% of miners in Indiana and Illinois. This wasn’t normal small-town America, or even something resembling other “company towns” around the country. All of these were privately-owned, privately-governed communities designed to give the mining companies maximum control not merely over mineral resources—but over human ones too. Nada White’s story Nada White witnessed the power of the coal camps firsthand. Now in her late 60s, White is a medical assistant who lives with her son, Dustin, outside Charleston, the capital of West Virginia. Her early girlhood was spent with her parents on a farm in the family hollow, tucked up against the belly of Cook Mountain, in southern West Virginia, where ten generations of Cooks had grown up. Those early days were spent swinging on mountain vines, collecting mollymoochee mushrooms in the spring, and picnicking in the glade where all the Cooks before her were buried. Then it came time to go to school, which was two hours away, longer in winter. However, the school bus came right to Wharton, a tiny town a few miles away, where her grandparents lived. So White moved in with them. Back in the 1950s, her granddad did the same thing as all the other men in Wharton: he mined coal for Eastern Gas and Fuel. And like all the others, the Cooks’ lived in one of the 50 or so houses that sat in a double row near the mouth of the hole in th e hillside into which White’s grandpa disappeared each day to gouge coal from the guts of the mountain.

Nada White photographed in Dawes, West Virginia. (Roger May for Quartz) Just as it had powered America’s staggeringly swift industrialization during the previous 50 years, the tar-black ore that came out was now fueling America’s postwar boom. At the edges of Appalachia, mighty factories churned out steel to make Fords and Chevys. The coal needed to smelt that steel came from Wharton and elsewhere in Boone County and throughout the central Appalachian region—as did much of the coal burned to power industry and the new Frigidaires and television sets filling the homes of America’s burgeoning middle class. The people of Wharton and other coal towns weren’t part of this mass-consumption metamorphosis, however. One problem with town planning by plutocracy is that the coal camps suffered the ugliness of industrialization—disease, squalor, poverty. But it did not experience the benefits of a diversified economy and public services that typically accompany industrialization—affordable transportation, easy access to food and markets, job opportunities, newspapers, recreation. Since they were privately owned, the camps enabled mining companies to govern not only production, but services and all means of consumption too: retail, recreation, education, medical care, worship, you name it. Each town was its own mini-monopoly—a phenomenon embodied in the company store. Flouting state laws, mine operators forbade miners from shopping anywhere beside the company store. Flouting state laws, mine operators forbade miners from shopping anywhere beside the company store. White’s grandfather and the others didn’t even earn the same kind of legal tender as factory workers in Pittsburgh and Toledo. He and the other miners were paid in little metal tokens notched in the middle and stamped with the Eastern logo. Scrip, it was called—and it could only be used in one place: the company store. On payday, White’s grandpa filed with everyone else into the Eastern company store to pocket a little mound of scrip, most of which he’d hand right back to pay for pinto beans, potatoes, rent, coal for home-heating, and any replacement mining gear he needed.

(Workers supplied their own.) “Mostly you spent all your money on pinto beans and hoped your fridge didn’t break,” says White. The scrip system worked because there were no other stores. “There was just nowhere else to go.” Up and down the Chesapeake & Ohio rail line that rattled through Wharton two or three times a day were plenty of other small settlements—Robinhood, Twilight, Greenwood, Prenter. Each had its own company store too, but they wouldn’t take Eastern’s scrip. Those took only the scrip of Bethlehem Steel or Armco or whatever other coal company owned the town. And they were all overpriced. “They could charge double or triple the price because they knew you had to come to the company store,” says White.

Coal miners buying groceries in a company store in West Virginia in 1938. Library of Congress/Marion Post Wolcott. “The greatest drain on the miners’ wages was the company store,” explained historian David Alan Corbin in Life, Work, and Rebellion in the Coal Fields. “Coercion, the scrip system, and the physical distance often combined to force the miners to deal at the company store, and through the monopolistic control of food and clothing and tools and powder, the coal companies were able to render wage rates and wage increases meaningless.” “Through the monopolistic control of food and clothing and tools and powder, the coal companies were able to render wage rates and wage increases meaningless.” Ordering from the catalogues of Montgomery Ward or Sears was much cheaper than the company store. That’s where White’s family got new shoes for school. But raising hard cash was tricky. The few folks with family tending farms in the hollows, like White’s, sold chickens or eggs to raise money—or hogs when times got really tough. In the winter, White’s dad also earned cash by driving coal around in his pickup, delivering it to families that needed it for heat (which few coal companies provided). Most people didn’t have pickup trucks or chickens, though. This made it easy to fall into debt, which many of White’s friends’ families did. There were other dangerous dependencies that were simply facts of life in Wharton. White recalls the day the four company men marched into the house next door, where Fred and Bobbie York lived, emptied all their things into the small soot-specked yard, and left. It was threatening to rain, and there sat Bobbie’s piano. White, her grandmother, and Bobbie hauled it up onto the Cook s’ porch, just missing the downpour. The Yorks gathered their damp belongings into a borrowed car and left. A week later, a new miner and his family moved in. White learned later that Fred’s leg had been badly broken in the mines and Eastern was fed up waiting for it to heal. After a while, someone came for Bobbie’s piano. Where the person took it, or what became of the Yorks, White never heard. “You didn’t want to talk bad against them—our whole existence depended on the coal companies.” The company could do that, of course, because Eastern owned the Yorks’ house. They owned everyone’s house, in fact—and everything else in Wharton from the church to the clinic, except the post office. There was no other employer. Those who complained found themselves not only jobless but blackballed, leaving them nowhere to work or live. People kept themselves in line, and quiet, says White. “You didn’t want to talk bad against them—our whole existence depended on the coal companies.” For hundreds of miles in any direction, almost all of the towns were privately owned, privately operated settlements almost all owned by companies like Eastern. There was only the faintest hint of public services—or, even, for that matter, representative government. Mind you, by the 1950s, coal camps were comparatively comfortable compared to what they’d been like before the 1930s, when the passage of the National Labor Relations Act granted central Appalachian miners the right to organize.

Because coal companies owned most of central Appalachia’s towns, “mine guards”—essentially private paramilitary forces— policed the region. In the early decades of the 1900s, because coal companies owned most of central Appalachia’s towns, “mine guards”—essentially private paramilitary forces—policed the region. These forces were broadly supported by the West Virginia government, which declared martial law four times in the early 1900s to put down miner strikes. (Facing a worker revolt that saw thousands of miners join together to oust a particularly brutal sheriff in Logan County, the governor went so far as to dispatch bomber planes.) Thanks to this system, southern West Virginia was home to some of the bloodiest fights in the history of America’s labor movement. To put it most simply, coal camps were economic institutions built to deny people agency. Of course, miners weren’t slaves. But there were parallels. In a 1920s US Senate hearing, a Logan County miner named George Echols who had been fired for heading up the local union explained the combination of chronic underpayment, coercion, and violence prevalent in the coal camps. “I was raised a slave [in antebellum Virginia],” he told the committee. “My master and my mistress called me and I answered, and I know the time when I was a slave and I felt just like we feel now.” Much has changed since the 1950s. As federal labor laws improved the job security and wages of miners, rising electricity demand in the 1960s pulled the coal industry out of a postwar lull. Soaring oil prices resulting from the formation of OPEC a decade later spurred another coal boom. At the same time, a lot hasn’t changed at all. The legacy of coal camps “A manufacturing town may count upon a reasonable degree of permanency. A mining town in a region not suited to other industry is sure to have only a limited life,” a 1923 Congressional report investigating coal-camp life explained. “It is this feature that differentiates the company-owned mining community from the company-owned manufacturing community.” That fatalistic approach to community planning seems to have played out in the coalfields. Through the coal slump of the 1950s and early 1960s, companies razed thousands of coal camp homes to cut their tax burden. Though some companies sold their land holdings, it was often to other big companies—not residents. Others continued charging existing occupants rent, refusing to sell. “I could go up and offer $100,000 for this house and they’d laugh in my face, even if I had it in $100 bills,” a mine safety inspector living in a Logan County company house—whose roof had a gaping hole—told ARC in the late 1970s. “For many Appalachian people, coal camp life is not a bygone era,” said the 1980 ARC report. “Facing no alternative, people remain, often dependent upon the will and wishes of the company landlord. In staying, they face insecurities of tenure, dilapidated housing, and fear of the company’s power.”

Coal tipples at Toms Fork, Cabin Creek, Kanawha County, West Virginia. (Roger May for Quartz) As of 2013, of the 10 largest landowners in West Virginia, none is headquartered in the state, according to a 2013 study by historian Lou Martin and economists at the West Virginia Center on Budget and Policy. In six counties, the top ten landowners control at least half of private land. Five of those six—Mingo, Boone, Logan, Wyoming, and McDowell—are in the central Appalachian coalfields. Jim Justice—who in 2016 ran and won West Virginia’s governorship—is also the state’s sole billionaire, and one of its bigger landlords. The political power of coal hasn’t faded either. As it happens, Jim Justice—who in 2016 ran and won West Virginia’s governorship on the same “bring back coal jobs” promise as Trump—is also the state’s sole billionaire, and one of its bigger landlords. At the time the WVCBP study was conducted, Justice owned 2% of McDowell County, as well as 2% of Greenbrier and 3% of Monroe counties. He also boasts a fortune invested in several Kentucky coal mines, timberlands in West Virginia, and a resort that has catered to coal barons for more than a century (and now fetes the New Orleans Saints).

This pattern of land ownership and extractive institutions, the legacy of policies set centuries ago—set the stage for a new way of passing on the costs of cheap coal to the public. The tragedy of Buffalo Creek Much of the mining in central Appalachia wouldn’t be profitable if companies shouldered the true costs of this environmental blighting. Instead, companies pass those costs on to the public. For one, there are the costs to human safety. Strip-mining and mountaintop removal have been linked with cataclysmic floods that, a few times a decade, destroy tens and even hundreds of millions of dollars in property. The worst floods have killed dozens or even scores of residents. Though few today have heard of it, one of the ghastliest industrial disasters in US history took place in a coal camp-specked hollow in Logan County called Buffalo Creek. One rainy February morning in 1972, a dam holding 157 million gallons of coal waste gave out, sending a 15- to 20-foot tsunami of black gunk ripping through the hollow, according to Everything in Its Path, a study of the disaster by Kai Erikson, a Yale sociologist. The flood killed 125 people, damaging around $50 million in property (about $300 million in current dollars) and leaving four-fifths of residents homeless.

The raging flash flood in Buffalo Creek hollow, West Virginia, February 27, 1972. (AP Photo/Harry Cabluck) Pittston Coal Company—a Virginia-based company that owned the dam—claimed the dam’s sole problem was that it was “incapable of holding the water God poured into it.” But earthlier explanations exist. Raw coal must be bathed of impurities before being burned. Each day, Pittston’s Buffalo Creek mines produced around 4,000 tons of market-ready coal and 1,000 tons of refuse. From the refuse, Pittston built dams to hold the 500,000 gallons of coal-dirtied water it ran through each day—an engineering feat that Appalachian writer Harry Caudill likened to “a pool of gravy in a mound of mashed potatoes.” Though the federal government had in 1969 outlawed this particular type of impoundment, state officials ignored both the law and local residents’ complaints about the dam’s shoddy state. Pittston did eventually foot some of the bill for the Buffalo Creek disaster, settling a class-action lawsuit from residents for a total of $13.5 million (about $13,000 per person). The state sued Pittston for $100 million, half of which was to cover cleanup and damages. Then the extractive institutions swung into action. In 1977, then-governor Arch Moore settled with Pittston, accepting $1 million— less than a fifth of what West Virginia taxpayers spent on cleanup. Moore wasn’t alone in helping keep coal operators profitable in the face of soaring environmental costs—particularly as the easy-to- reach underground seams tapped out, and companies turned to strip-mining. States had passed laws overseeing strip-mining; however, these were so toothless and little enforced, notes Eller, that in both Kentucky and West Virginia, the practice increased after those laws’ passage. In 1977, the federal government finally weighed in. Paradoxically, though, the new law created an above-board way of shifting the astronomical costs of mountaintop removal’s destruction onto taxpayers, with a devastating impact on the lives of central Appalachians. The Surface Mining Control and Reclamation Act was supposed to require that mined land be restored to something approximating its original shape. However, a loophole in the law let coal companies ignore that expensive requirement if the reclaimed land could be put to “a higher and better use.”

Coal companies instantly saw their opportunity, as Ron Eller recounts in his modern history of Appalachia, Uneven Ground. What if the now-flattened mountaintops they were creating could be pitched to government officials as a way to attract outside factories and create more jobs? Higher and better use Almost overnight, strip-mining underwent a hypertrophic mutation, as mining companies turned to blowing up mountain peaks and deploying hulking, 20-story machines to claw open coal seams below, replacing replace verdant ridges with moonscape mesas of lifelessness. The land area of southern West Virginia scarred by active mountaintop removal leapt more than threefold between 1985 and 2005 and has stayed steady since then, according to GIS analysis by SkyTruth, an environmental nonprofit. As of 2015, some 8% of Boone County was being actively mined, compared with 2% in 1985. Today, the biggest mine sites exceed the island of Manhattan in size. It is impossible to restore the land to its former state. And the new law made it so they didn’t have to try. A recent Duke University study found that coal companies had left the area they studied 40% flatter. Mineral rights let coal companies tear off mountaintops—ruining the land permanently—to get at the coal beneath it. The scale of environmental carnage is possible, at least in part, because of the land-grab legacy: corporate control of enormous land parcels has meant residents lack the legal clout to object. It’s also been possible thanks to the signing away of mineral rights that happened in the late 1800s and early 1900s—that saw Nick Mullins’ ancestor part with a hoard of coal wealth for 12 rifles and 13 hogs. This wasn’t merely a matter of swindling farmers out of subsurface riches. Since mineral rights legally trumped surface rights, corporate ownership let coal companies clear cut forests for mine supports, build roads and railroad spurs, pollute and divert streams—all without having to pay taxes on the land they defaced, as Erikson explains. Strip-mining, however, pushed this bait-and-switch to a whole new level. Mineral rights let coal companies tear off mountaintops—ruining the land permanently—to get at the coal beneath it.

The Middle Ridge surface coal mine, which cut off part of Coal River Mountain in Raleigh County, in southern West Virginia. (AP Photo/Michael Virtanen) Then there are the costs that come with trying to find some higher and better use. Desperate to show they’re creating jobs, local politicians have spent many millions in taxpayer funds recruiting out-of-state factories to the “flat land” created by mountaintop removal, according to Eller. Few have panned out. The region’s manufacturing sector was about the same in 1992 as it was in 1967, according to a study commissioned by ARC. It still suffered from low wages, low productivity and over-reliance on branch plants. Meanwhile, a super-max penitentiary built atop a leveled mountain in Virginia struggles with a cracked foundation (the locals dubbed it “Sink-Sink”). Mingo County recently unveiled a new airport on one such dusty tableland; its major industrial park, which sits on a former mine site, remains mostly empty. “Coal creates employment for some, wealth for a few, and sucks oxygen out of the room for the other people.” There are lots more mountaintop removal sites than there ever will be airports or shopping centers,” says Peter Hille, head of Mountain Association for Community Economic Development, or MACED, an economic development group in Kentucky. “Most of these sites are in places where it doesn’t make any sense to build anything. Coal creates economic activity for a period of time but not development. It creates employment for some, wealth for a few, and sucks oxygen out of the room for the other people.” This spending might temporarily boost the area’s GDP. But in the long run, that money isn’t being spent on self -sustaining investments in upping the skills and knowledge of local workers and businesses. Lou Martin, the Chatham University historian, notes, “You can still see state governments worried about losing jobs and trying to help corporations, at times at the expense of employees and residents.”

The false promise of more jobs Despite the state support it’s enjoyed, the shift to mountaintop removal hasn’t created much in the way of jobs. They require far less labor than underground mines. For instance, coal employment in West Virginia fell by three-fifths between 1983 and 2000; 94% of those lost jobs were underground, according to research by Downstream Strategies, an environmental consulting group. In fiscal year 2009, West Virginians paid a net $42.2 million to support coal companies. Nowadays, taxpayers subsidize the coal industry directly too. In fiscal year 2009, West Virginians paid a net $42.2 million to support coal companies, according to a study (pdf) by WVCBP and Downstream Strategies, an environmental consulting firm. The report noted that “this value fails to capture the significant legacy costs resulting from past coal industry activity that have yet to be funded. These costs, which include damages to roads and bridges and funding needs for reclaiming all abandoned mine land and bond forfeiture coal mine sites in the state, amount to nearly $5 billion.” Even those data ignore the human toll, however. As it happens, that is what finally convinced Mullins to leave coal mining behind. In the mid-2000s, a freak fire burned down his home. He and his family lost everything. But they gained a modest homeowners insurance check and some good advice—that maybe they should see the fire as an opportunity instead of a tragedy.

Nick Mullins in Berea, Kentucky. (Roger May for Quartz) “My brother put it to me: ‘Do you want to be a coal miner for the rest of your life?’” says Mullins. “All these things started weighing on me—I had a few run-ins with some of the management that month and realized exactly where we stood as people, as human beings, to the mining company. And I saw how they pitted us against each other and realized I couldn’t stay on that path.” It took him about a month to quit his mining job. A year later, he and his family moved just outside of central Appalachian Kentucky, where he attended Berea College and finally got his degree. He now runs his own public relations firm aimed at bridging political and cultural divides surrounding Appalachian issues. Greater economic opportunity was part of why they left. His biggest concern about staying, however, was their two kids. Eroded by dwindling budgets, local schools were even worse than when Mullins had attended them. He worried about his kids’ health, too. Mountaintop removal sites and underground mines engirdle Clintwood, creating untold volumes of toxic waste buried in unnamed patches of ground. The backyard creek that ran past his house did so increasingly in Technicolor hues. That impetus to leave gained new urgency, however, when a public health scientist named Michael Hendryx began publishing his research. Dark waters Blowing off a mountaintop releases naturally occurring poisons like arsenic, selenium, lead, and manganese. These poisons then seep into streams and groundwater. Meanwhile, the blasting fogs the air with a toxic cocktail of dust that settles on roofs and windows in the valleys below, and cakes the lining of lungs. The displaced soil and vegetation from mountaintop removal is plowed into valleys, creating enormous detritus piles and choking off waterways. Between 1985 and 2001, mining operations buried some 724 miles of rivers and streams, according to the EPA. And since 2002— when the George W. Bush administration redefined “fill material” in the Clean Water Act to include mining waste—the rubble plugging up central Appalachian waterways is laced with toxic muck. For a decade now, peer-reviewed research produced by Hendryx, now a professor of applied health science at Indiana University, exposed a consistent link between mountaintop removal and a broad range of health problems and rising mortality rates. According to his research, since 1990—when amendments to the US Clean Air Act inadvertently stoked the growth of mountaintop

removal (registration required)—parts of central Appalachia with mountaintop removal have had about 1,200 extra deaths per year, adjusting for age, smoking habits, and other factors. Rates of birth defects are 26% higher in areas with mountaintop removal. In addition to all-cause mortality, self-reported rates of some forms of cancer and mortality from diseases of the heart, lung, and kidneys are significantly higher in places with mountaintop removal, compared with places in Appalachia where it doesn’t occur—even when adjusted for age, gender, smoking, work and family history. Rates of birth defects are 26% higher in areas with mountaintop removal than in the non-mining regions of central Appalachia. These findings are ominous because birth defect rates are unusually sensitive to exposure to toxic chemicals, even though the specific mechanisms are little understood. Though it’s hard to put a dollar sign on the human toll, Hendryx estimates that costs associated with the higher mortality of Appalachia’s coal-mining regions between 1979 and 2005 totaled $50 billion a year (in 2005 dollars), compared with the $8 billion contributed by the coal industry in 2005. A 2011 study (registration required) put the health, environment, and other economic costs of coal-fired electricity at $345 billion, and possibly more than $500 billion, doubling—and possibly tripling—the costs of coal- generated electricity. “These and the more difficult to quantify externalities are borne by the general public,” they write. Despite the data, the effect of mountaintop removal on people’s health remains a topic of much debate. The coal companies, unsurprisingly, dismiss Hendryx’s research, as do a long roster of West Virginia politicians. Instead, you often hear that central Appalachians are sick because they’re poor and because they make bad choices. This is the way America commonly understands ill health and early death: that it’s a matter of cigarette drags, cheeseburger bites, and other spasms of feeble character. It’s true that, because towns were developed based on their proximity to coal mines rather than commercial centers, residential communities are often more than an hour’s drive from grocery stores. Gas stop fare of buffalo wings and pepperoni rolls are the quick meal options available to most. But part of coal’s legacy in central Appalachia is that, as with their economic opportunities, individuals often don’t have much control over whether they are healthy or not. Between drugs, poverty, social isolation, and lack of educational opportunities, there is any number of hazards that can derail, or even end, a person’s life. With so many possible pitfalls, it’s difficult to avoid encountering some. The media and lawmakers are increasingly drawing connections between economic stagnation and destructive behaviors like opioid abuse and other “deaths of despair”—a term coined by Monnat and popularized by Nobel laureate economist Angus Deaton, which includes suicide and alcohol-associated liver disease as well as drug overdoses. And economically battered central Appalachia leads the pack. In 1999, rates of these “deaths of despair” among those aged 15-64 were the same inside and outside Appalachia, according to a recent report commissioned by ARC. By 2015, the combined mortality rate from these deaths was nearly 95 per 100,000 in central Appalachia, versus 49 per 100,000 in non-Appalachian US. That’s appalling, of course—but what might be even more disturbing is the larger context of that disparity. It’s not just that central Appalachians are doing worse with opioids than everyone else. They’re dying at higher rates from other causes too. Between 2008 and 2014, heart disease killed nearly 250 people per 100,000 in central Appalachia. Though mortality rates for most major causes of death have improved in the region since the late 1990s, they’re still far higher than the rest of the country. Between 2008 and 2014, heart disease—the leading cause of death in the US—killed nearly 250 people per 100,000 in central Appalachia, 42% higher than the rest of the nation, according to ARC research. Cancer claimed 222 lives per 100,000 in the region, around a third more than the US as a whole. Rates for chronic pulmonary obstructive disease exceeded national rates by more than 80%; diabetes deaths were 40% higher. They’re also dying younger. Between 2011 and 2013, the nation averaged 6,658 “years of potential life lost”—the cumulative number of years a region loses per 100,000 people when its residents die before reaching age of 75, according to ARC. Central Appalachia lost 11,226 years of potential life. But the words typed on a death certificate don’t tell us much about the quality of people’s lives. Central Appalachian residents also have higher rates of disability and chronic disease (pdf, p.86). This may have something to do with the economic precariousness that is a fact of life in central Appalachia. A growing body of scientific evidence links “allostatic load”—basically, chronic wear and tear on the body caused by repeated triggering of stress hormones—with chronic disease and premature death. Many of the maladies strongly associated with allostatic load are problems in central Appalachia, such as diabetes, obesity, and heart disease. Allostatic load is also linked to depression and impaired mental functioning. Some scientists hypothesize that the system’s inability to balance its brain chemistry caused by chronic stress may also encourage addiction.

Central Appalachia’s brutally low living standards certainly bring with them plenty of stress. And the economic realities of the region mean that a sense of community, which might alleviate the effects of stress, is hard to come by. Marlene Spaulding’s story Marlene Spaulding grew up in a Kentucky town called Beauty during the 1970s. “I’m not really sure there’s a whole lot of beautiful there,” she says. The town was once known as Himlerville—the home of the first cooperative coal-mining company in the country. Founded in 1917 by Hungarian immigrants, each employee in Himler Coal Company owned company stock and received a share of its profits, according to Eller’s Miners, Millhands, and Mountaineers. The houses in Himlerville were nice. Each had two fireplaces, gas and electricity, a tub and shower, and a vegetable garden, among other amenities. Miners could choose to buy their home or even b uild their own. Himlerville had a library, an auditorium, a modern school open 10 months out of the year, a local newspaper, as well as commercial developments that included a bank, a hotel and a bake shop. Miners elected by their fellow employees sat on the company’s board of directors. The company and the town prospered throughout most of the 1920s. But in 1928, private capitalists bought its assets. They renamed it Beauty a few years later. By Spaulding’s count, there are only two members of her extended family who have never been to prison. By the time Spaulding was growing up in the town, life in Beauty was hard. Her biological mother was 14 when she gave birth to Spaulding and her twin sister, and a local family adopted the girls. Her adoptive mom came from a family with 18 children, her adoptive dad had 12 siblings. By her count, there are only two members of her extended family who have never been to prison. Drugs were a big problem. Spaulding’s older brother got hooked in high school, and was in prison before long. Back then, they let prisoners out on furlough over the weekends. Her brother would usually run off with his friends. Spaulding recalls awakening one night to her mom calming her as policemen filed into her room to search for her brother under her bed. “And that was normal t o us,” she says.

Marlene Spaulding in front of Able Families in Kermit, West Virginia. (Roger May for Quartz) After a life spent shuttling in and out of prison, her brother died of cancer when he was 48. Six weeks before he died, police busted him for selling morphine “off his deathbed,” as Spaulding puts it, most likely to exchange for something stronger. “One of the hardest parts about him is I really can’t name anything good he did in his life,” she says. “He never carried a job, never had a contributing role in society, was never a productive person.” But her childhood could have been much worse. As an adult, Spaulding tracked down a biological sister who had been born three years after Spaulding but hadn’t been adopted. The two met up at the local Dairy Queen, where she learned that her biological sister had drug problems and was essentially homeless. At that moment, she happened to live eerily close by—beneath a bridge on the Tug River that Spaulding could see from her office window. “That very easily could have been me,” says Spaulding. Long before the rest of the nation noticed, the opioid crisis took root in central Appalachia. By 2015, West Virginia’s rate of opioid- related overdose death for those aged 15-64 was more than three times higher than the non-Appalachian US, while Appalachian Kentucky’s was more than twice the rate. McDowell County, in West Virignia’s southern coalfields, has the highest opioid overdose rate in the country. It’s probably no coincidence that coal-mining towns emerged in the 1990s as the epicenter of the opioid crisis. It’s probably no coincidence that coal-mining towns emerged in the 1990s as the epicenter of the opioid crisis, says Shannon Monnat, sociology professor at Penn State University. Of course, injuries abound and surgery is common, leading to prescriptions for heavy-duty

painkillers. To keep injured miners working—and from claiming disability insurance—some companies had doctors prescribing pain drugs on the job. It’s not just miners who rely on painkillers. Ask central Appalachian residents about addiction, and people will just as casually mention waitresses or hospital workers who dose themselves to keep their jobs. The 2008 ARC study reported that in coal country, women, unemployed people, and those whose households earned less than $35,000 a year were more likely to be admitted for substance abuse treatment in the coalfields than in other parts of Appalachia. Part of the problem may be that high unemployment levels create a ready workforce of traffickers. Central Appalachians themselves will tell you that boredom has a lot to do with it: In a place where movie theaters, restaurants and skating rinks have mostly shuttered, recreational drug use doesn’t have much to compete with. Spaulding escaped opioids, but not teen pregnancy—another problem common in central Appalachia. Kentucky, along with West Virginia, has among the highest rates of teen birth in the country, concentrated in central Appalachian counties. “The day I walked through the [high school] graduation line, I already had a one-year-old,” she says. A few years later, she heard about a family wellbeing charity called Able Families, based in Mingo County, from another young mother. For a while, she participated in its programs, such as playgroups organized for mothers of young children. Then she was asked to join as a home-visit worker. Spaulding is now the director of Able Families, and is completing her masters in social work. In addition to coaching mothers on childcare, Able Families sponsors an after-school program. It used to be that they’d prepare a light snack—until they realized it was often the child’s last meal of the day, and raised money to provide hot meals. On the first day of the new meal program, Spaulding and her staff had planned to let the kids loose on the playground before the meal. But the kids weren’t interested. “The first thing they got off the bus, they were at the fence like, ‘Where’s our food?’” More than a third of central Appalachian children live in poverty, compared with 21.7% nationwide, according to ARC. The even bigger worry is that addicts with needy children aren’t coming forward for help at all. “There’s children starving over the weekend. And we don’t know who they are,” says Sandra Justice (no relation to the West Virginia governor), who works with Spaulding at Able Families. “You go to the suboxone and methadone clinics, and you’ll see pregnant women in line.” The problem could start even earlier, though. “You go to the suboxone and methadone clinics, and you’ll see pregnant women in line,” says Spaulding. The local hospital—where around two drug-dependent babies are born every hour—now has a separate nursery tailored to help infants to undergo the excruciating month-long withdrawal. The infants are treated with methadone, which relieves opioid-addiction symptoms but is itself a powerfully addictive narcotic. It’s not just material needs that local kids are missing out on. When the children put on performances, Able Families pulls out the stops to help parents and grandparents come to watch. It’s hard to get anyone to come, though. Spaulding recalls an episode when, just as the children were about to take to the stage to begin a play they were putting on, she and her staff spied out the window one of the little boy’s mothers driving up. The boy was excited, and the staff stalled while they waited for her. But after a few moments, Spaulding saw the mother pull out again and drive off. She hadn’t been parking to see the pe rformance, but lining up at the drive-through pharmacy that abuts the Able Families lot. Getting painkillers, Spaulding suspects. Graduation events are similarly crushing. At a recent junior high ceremony, besides them, only one mother and one grandmother showed up. “Can you imagine your life if no one ever does that for you?” says Spaulding. “Thinking you could move away to a university—that’s not even in the realm of possibility.” Spaulding’s own daughter was one of only a few kids in her high school graduating class of 106 to leave the area. She graduated from Alice Lloyd College in Kentucky last year and now lives in Nashville, working as a paralegal. (She told her mother she would go anywhere that had at least one stoplight.) Her middle child has wanted to be a chef since early elementary school. She had to explain to him that, with no real restaurants in the area, he would have to leave too. This fall, he started in the culinary program at Fairmont State, West Virginia. “I don’t think there’s anything wrong if they wanted to live here,” she says, “but I want them to know there’s opportunity in other places.” Most kids in Kermit don’t get to glimpse those kinds of opportunities. Driving the ancient white Able Families van along Route 52, Spaulding recounts a recent field trip to Cleveland she took with her students. For nearly all of them, the trip to Hard Rock Café was the first time they’d been to a sit-down restaurant. In fact, many had never before left Kermit. Spaulding was worried that the kids would be bored and disruptive when they toured Case Western Reserve University. However, the campus proved dazzling and exotic—especially their visit to a dorm. “So I could live here?” one girl asked.

“I was thinking, well, probably not—this place is really expensive,” Spaulding says. “But I told her, ‘Yes, you could go to school here and this could be your room.’ And they were all like, ‘To live in?’” she says. “They come from where even a community college is mostly unheard of. Especially thinking you could move away to a university—that’s not even in the realm of possibility.” In this kind of environment, it’s perhaps unsurprising that residents of central Appalachia report feeling mentally unhealthy 25% more frequently than the average American. Nearly 20% of the area’s Medicare beneficiaries suffer from depression, compared with just over 15% nationwide. In May, a metastudy found that social isolation upped a person’s risk of dying by 29%. Perhaps it’s not simply the unhealthy habits and pollution endemic to central Appalachia that are sickening residents and killing them prematurely. Rather, the wear and tear on the body from prolonged social and economic stress could be making people far more vulnerable to their harmful effects. The problem we’re solving for Nada White’s son, Dustin, is now 34. He says that growing up in the 1990s, the most exciting event of the year was the Coal Festival of Boone County. The multi-day coal-themed carnival featured music acts, fireworks and tilt-a-whirls—a much bigger affair than the county fair.

Nada White and Dustin White in Dawes, West Virginia. (Roger May for Quartz) Dustin’s great- great-grandfathers would have looked forward to something very different. Back then, the biggest festival of the year in central Appalachia was election day—a state fair, market day, and political convention rolled into one, as Altina Waller, a University of Connecticut historian, recounts in Feud: Hatfields, McCoys and Social Change in Appalachia, 1860-1900. That disappeared in the coal camp era, as coal operators controlled political machines and, often enough, the votes of their employees. Nowadays, West Virginia frequently has the lowest voter turnout in the US: In 2008 and 2012, rates undershot the average by 11 percentage points. Last year, however, that gap actually narrowed a little—shrinking to nine percentage points—thanks to enthusiasm for Trump. The central Appalachian coalfields were one of his biggest strongholds of support. It’s not hard to guess why. Trump promised to revive central Appalachia’s coal jobs, a feat that would defy market trends. Central Appalachia’s remaining seams are harder to get at, making the coal generated uncompetitive with that mined elsewhere in the US. Environmental regulations have lessened demand from coal plants, though the far bigger factor is the abundance of natural gas, which has undercut coal prices. As a result, the region’s coal industry is in the worst shape it’s ever been. In the last few years, companies have been winding down operations—some by choice, and many by way of bankruptcy. But coal is still the best-paying job around, and in many towns, the only one. Bringing those jobs back, it’s true, would save thousands of families and whole communities. It turns out, Trump’s campaign promises weren’t just talk. In August, the Trump administration halted a comprehensive study, launched by the National Academy of Sciences in early 2017, to assess the health effects of mountaintop removal, citing cost concerns. Meanwhile, Trump appointed as head of the federal Mine Safety and Health Administration a former CEO of a coal company with a record of repeated safety violations, including at a mine where a preventable accident killed a miner in 2011. His energy department is investigating ways to force utility companies to subsidize coal company profits, reported Slate in Oct. 2017. Whether or not this will be enough to postpone the long overdue creative destruction of central Appalachia’s coal industry remains to be seen. But for all the fanfare it’s received, Trump’s appeal to central Appalachians is neither original nor populist. It follows the same values and reasoning championed by a century of Appalachian politicians. Trump, like generations of Appalachian politicians, says that he’s trying to save coal for the sake of the people. But the problem that he’s solving for isn’t how to invest in making people healthier and better educated. Nor is it how to make their children’s lives better or to expand their families’ opportunities.

At a national level, US politicians, corporate chieftains and other civic leaders continue to ignore the flaws riddling their own growth model. Rather, the problem Trump is intent on solving is how best to give vast, distant corporations more opportunities. To revive an old oligopoly, instead of innovating new work. To subsidize producing more coal, instead of spending on making the region’s people more valuable. To stake the region’s future on a finite resource, at the expense of the ultimate renewable one— human potential. This isn’t just a Trump thing. At a national level, US politicians, corporate chieftains and other civic leaders continue to ignore the flaws riddling their own growth model. Like the coal-backed politicians counting on boom to follow bust, the nation’s leaders continue to expect the business cycle to buoy growth, failing to grasp how years of increasing inequality in wealth, income, opportunity, and health have cannibalized the very demand needed to sustain it. While they dither on investing in infrastructure, technology, education and health care, the country’s reliance on welfare continues to climb as labor force participation slides. Taxpayers are subsidizing companies to underpay retail workers, just as they’re paying for coal companies to lop off mountaintops. And though central Appalachia’s health and economic problems are unusually pronounced, the phenomenon certainly isn’t anomalous. According to a recent county-by-county analysis by the Economic Innovation Group, a entrepreneur advocacy and policy group, the number of jobs and new businesses in distressed communities in the US—central Appalachia among them—haven’t grown at all since 2000. Half have suffered net losses of both. Meanwhile, two years ago, national life expectancy in the US fell for the first time in decades. It dropped again last year. It could be an anomaly—two years hardly makes a trend. But it could also mark a shift in its people’s fortunes that is all but unthinkable in a country that has grown to be the world’s richest and most powerful. It’s a blow that hit many central Appalachian counties a decade ago, the grim conclusion of the great cynical experiment played out upon its people. Americans, like central Appalachians, are told that problems of health and poverty stem from the moral failings of individuals, not of society. Meanwhile, their country’s institutions are embracing the same ideology that laid waste to the coalfields—that profit equals prosperity, regardless of its true costs. https://qz.com/1167671/the-100-year-capitalist-experiment-that-keeps-appalachia-poor-sick-and-stuck-on-coal/

One By One Big Hydropower Dams Disrupt Mekong River’s Free Flow August 2, 2017/in Asia, Water & Energy, Water News /by Keith Schneider In unfolding global energy revolution, expensive and ecologically risky dams may not be right choice to generate more electricity.

Plans by Laos to build big hydropower dams on the Mekong River are seen by many villagers, the region’s environmental organizations, and some officials in neighboring countries as making a mockery of the river’s natural beauty and its central value as a source of fish, irrigation, and commerce. Photo © J. Carl Ganter / Circle of Blue By Keith Schneider, Circle of Blue The bend in the Mekong River where Laotian authorities want to build the Pak Beng hydropower dam is described as one of the most beautiful stretches along all of the 4,000-kilometer (2700-mile) river. High bluffs tower above farm fields and tropical forests. Fishing villages lie along the shoreline. The deep and dark waters teem with fish and birds soar above the shallows. Laos’ plan to build the 912-megawatt, $2.3 billion dam is a step in one of the world’s most aggressive hydropower development programs. The tiny landlocked nation of 7.1 million people is setting out to build 350 hydropower projects and add 26,000 megawatts (26 gigawatts) to the 6,300 megawatts of electrical generating capacity that Laos currently produces from its 42 operating hydropower dams. Along with the Pak Beng dam, two are under construction downstream, and three more are planned to cross the Mekong River. “The Department of Energy has the ambitious plan to raise the electrification ratio to 95 percent of families across the country,” Xaypaseuth Phomsoupha, a director general in the Laos Ministry of Energy and Mines, told an international water and hydropower conference last year. “This plan is amongst the priorities of the government to eradicate poverty in the country.” It is also a plan that is seen by many villagers, the region’s environmental organizations, and some officials in neighboring countries as making a mockery of the Mekong’s natural beauty and its central value as a source of fish, irrigation, and businesses that better fit conditions of the 21st century. On the bend of the Mekong designated for the Pak Beng dam and along other big bends upstream in China, and downstream in Laos and Cambodia, one of the world’s great river basins is steadily being hemmed in, and not just by dams.

The Mekong River’s regular cycles of low water during dry seasons and floods during monsoons influence the lives of people in six nations that count on the river for food and their livelihoods. Photo © J. Carl Ganter / Circle of Blue

The Mekong and a good number of the more than 60 million people in four Southeast Asia nations that rely on the river are being pushed around by what they view as obsolete electrical generating priorities. With surprising speed, Southeast Asia has emerged as the latest testing ground for competing theories about the scale and expense of relying on water for electrical energy. The 20th century idea of building immense and expensive hydropower dams is rushing headlong into a wave of less expensive and less disruptive alternative energy technologies and development. Boiled down, the issues are globally momentous. On one side are long-standing practices in project financing, engineering, and economic development theories about the usefulness of big hydropower dams. On the other side are cleaner and less costly generating technologies, civic opposition to big dams, and erratic water supplies caused by climate change that make managing dams much less certain. The outcome will decide how much of the Mekong and the last untouched stretches of the other great rivers in Asia, Africa, and South America will continue to flow freely.

Along big bends upstream in China and downstream in Laos and Cambodia, the Mekong River is steadily being hemmed in by hydropower dams. Photo © J. Carl Ganter / Circle of Blue “The energy revolution is evolving around the world,” said Carl Middleton, an assistant professor of political science at Chulalongkorn University in Bangkok, and a scholar on Mekong River development. “This region is resisting the shift, though, because of the economics of building big new projects. But it seems inevitable that the shift will happen here. It raises questions about continuing to build so many big dams.” A Big and Important River Historically, from its source on the Tibetan plateau, the undammed Mekong gathered water and energy as it flowed through China, Myanmar, Laos, and Thailand before dropping its rich silt and sediment in a rice-growing delta in Vietnam. The Mekong Basin’s water supported one of the world’s largest freshwater fisheries. Its regular cycles of low water during dry seasons and flooding during monsoons influence the lives of people in five nations that count on the river for food and their livelihoods. Historic texts reveal that constructing big Mekong Basin hydropower dams dates to the early 1950s. The Bureau of Reclamation, the water supply manager for the U.S. Department of the Interior, recommended extending its dam building prowess to the region. The idea faded in the conflicts that engulfed Southeast Asia over the following 30 years.

Historically, from its source on the Tibetan plateau, the undammed Mekong gathered water and energy as it flowed 4,000- kilometers (2,700 miles) through China, Myanmar, Laos, Thailand, and Cambodia before dropping its rich silt and sediment in a rice- growing delta in Vietnam. Photo © J. Carl Ganter / Circle of Blue By the mid-1990s China had replaced the United States as the leading dam-building nation, and the Mekong became one of its principal targets. The Mekong’s natural water flows began to change when China built the first of six big hydropower dams in the

river’s upstream reaches. In 1995, recognizing the value of the Mekong’s waters to their wellbeing, Thailand, Cambodia, Laos, and Vietnam formed the Mekong River Commission, a research and planning organization. The four countries formed a pact and a process to consult each other and reach agreement on constructing big dams. There are 286 river basins in the world that encompass boundaries of multiple countries, according to the Institute for Water and Watersheds at Oregon State University. The Mekong River Commission is one of 121 transboundary river organizations in the world. The commission’s research divisions, regarded as highly competent internationally, have published incisive findings about the largely deleterious effects that dam building will have on water flow, fisheries, and farming communities. Moratoriums and Alternatives Last year Cambodia announced a moratorium on new dam construction until at least 2020 based on public opposition and the damage it projected to the Mekong and tributaries in its jurisdiction. Cambodia is nearing completion of the 400-megawatt, $816 million Lower Sesan 2 hydropower dam on the Sesan River, a Mekong tributary. The project unleashed a furor in river communities, which are convinced it will ruin a prime freshwater fishery, a finding supported by commission studies. More than 5,000 people are being forced to move from the 30,000 hectares (74,000 acres) of forest and farmland that will be flooded by the dam’s big backwater lake. Not nearly as effective has been the Mekong River Commission’s prior consultation mandate. Laos has been particularly impassioned about its authority to do what it wants within its borders. Along with the Pak Beng project Laos has started construction on two more Mekong River hydro dams — the $3.8 billion, 1,260-megawatt Xyaburi and the $600 million, 260-megawatt Don Sahong hydropower dams. Both dams lie at the center of an intensifying discussion about energy production, human rights, sovereign authority, and ecological sustainability. Laos authorities assert that its river development program is a sound and safe way to build its economy. It can generate electricity for domestic markets and also sell power to its neighbors.

A good number of the more than 60 million people in four Southeast Asian nations that rely on the Mekong River are being pushed around by what they view as obsolete electrical generating priorities. Photo © J. Carl Ganter / Circle of Blue Laos’ hydro dam development plans are backed by extensive international financing, particularly by Chinese investors. Viraphonh Viravong, the deputy minister of energy and mines and the principal government advocate for hydropower, said he is intent on making his country the “Kuwait of hydropower of Southeast Asia.” Citizens affected by construction plans, though, are reacting with intensifying influence. Mekong Eye, an online news organization that reports on the region, features probing original articles about human hardship in Mekong villages caused by construction and planning. Another operating challenge is climate change. The region’s shifting meteorology is changing the basin’s water supply patterns and causing engineers and power specialists to adjust their power production and revenue estimates. The problem is particularly acute for “run-of-the-river” projects that have no backwater storage reservoirs. Big hydropower dams, say engineers, are not nearly as reliable as they once were for generating power and profits. Some of that message is infiltrating, particularly in Vietnam, which built hydro dams that deliver a third of the nation’s 40 gigawatts of electrical generating capacity, according to the U.S. Energy Information Administration. But Vietnam has developed a bout 80 percent of its hydropower potential and is leery about coal-fired power plants.

Following the lead of China and India, Vietnam is attracting substantial international investment capital to develop renewable energy projects, according to the General Department of Energy. Binh Thuan Province, for instance, has encouraged investors to develop 20 renewable energy projects. The largest is a 200-megawatt solar plant. Solar and wind generating prices are now less than the cost of building coal-fired power plants and very competitive with constructing big hydropower projects. Other Vietnamese provinces are developing wind and solar projects to raise renewable energy to at least 10 percent of the country’s generating capacity by 2030, anticipated to be over 100 gigawatts. http://www.circleofblue.org/2017/world/one-one-big-hydropower-dams-disrupt-mekong-rivers-free-flow/

What It’s Like To Live Through Cape Town’s Massive Water Crisis By ARYN BAKER Photographs by MIKHAEL SUBO TZK Y and JOHNNY M ILLER for TIME Feb 20 18 I knew we were in trouble when I found myself Googling dry composting toilets. That was on Feb. 1, 2018, just after the mayor’s office here in Cape Town announced new water restrictions. We are now limited to using 13 gallons of water per person per day. That’s enough for a 90-second shower, a half-gallon of drinking water, a sinkful to hand-wash dishes or laundry, one cooked meal, two hand washings, two teeth brushings and one toilet flush. I figured I could save an extra couple of gallons by forgoing the daily flush in favor of a dry composting toilet. Hippie friends living off the grid in the country do it. How bad could it be? According to current projections, Cape Town will run out of water in a matter of months. This coastal paradise of 4 million on the southern tip of South Africa is to become the first modern major city in the world to completely run dry. And even though residents aren’t responsible, the burden of making sure it doesn’t happen rests largely on our ability to cut down on water usage. Dramatically.

People queue to buy water at a grocery store in Milnerton, a suburb of Cape Town, on Feb. 3 2018. Mikhael Subotzky—Magnum Photos for TIME Millions of people around the world live without sufficient access to water. But Cape Town is no developing-world urban quagmire. It is a prosperous metropolis, a well-managed global tourist destination responsible for 9.9% of South Africa’s GDP, full of multimillion-dollar beachfront properties, art museums and two of the world’s top 50 restaurants. Cape Town running out of water is like San Diego going dry. Which, if you factor in the looming threat of climate change, may not be that far off. California’s five-year drought, which ended in 2016, had state officials scrambling to enact their own water restrictions. At one point, NASA warned that the state had less than a year’s supply in its reservoirs. As with the California dry spell, climatologists at the University of Cape Town say man-made global warming is a likely factor in the continued drought and that we, like many other cities around the globe, are facing a drier future with increasingly unpredictable rains. What is happening to us in Cape Town might not be an outlier. It could happen to you too. The Cape Town crisis stems from a combination of poor planning, three years of drought and spectacularly bad crisis management. The city’s outdated water infrastructure has long struggled to keep up with the burgeoning population. As dam levels began to decline amid the first two years of drought, the default response by city leadership was a series of vague exhortations to be “water aware.” The water-saving appeals became more urgent in the past year. The rest of us prayed for rain. It was only in September that Capetonians were given a limit of just over 23 gallons of water per day per person. By then the reservoirs were at a third of their capacity. But less than half of city residents met that goal—the tragedy of the commons, in action. As a result, with reservoirs down to the last dregs of accessible water, the 13-gallon limit we are asked to meet is less than the minimum U.N. daily recommendation for domestic water needs. An even grimmer scenario now looms: Day Zero, when the government will turn off the taps for most homes and businesses in the city to conserve the very last supplies. Hospitals and other vital institutions in the city center will still get water, according to officials, but the majority of residents will have to line up at communal water points to collect their daily allotment of 6.6 gallons—about half our current, meager recommended limit—under the gaze of armed guards. This dystopian scenario is no bluff. As of Feb. 5 2018, Day Zero was predicted to fall on May 11 2018, according to an analysis of current usage patterns and dam levels. Although a decline in agricultural use has shifted the date forward since the beginning of the year, there has been no significant reduction in urban consumption, according to the mayor’s office. Day Zero could just as easily be moved closer if city residents don’t continue to conserve. The looming shutdown has prompted chaos, with a run not only on bottled water but also on water tanks and jerricans. Once lush city parks and golf courses have withered, and public restrooms now urge visitors to flush only when absolutely necessary. High-end cafés use paper cups and plates to cut down on dishwashing. Many fear for their livelihoods; analysts estimate that the water crisis will cost some 300,000 jobs in agriculture and tens of thousands

more in the service, hospitality and food sectors. If employees have to take time off from work to wait in line for water, it’s going to have an even greater impact on the economy. Which brings me back to the composting toilet. Come Day Zero, there is no way I am going to flush more than a third of my precious daily water supply down a conventional toilet. Suddenly, spending $250 on a high-tech composting commode seems like a good deal. Until I read the fine print. The contents of the toilet must be allowed to mature for several months on a patch of well-drained soil, several meters away from human habitation. I live in a row house with a concrete terrace the size of two ping-pong tables. Now I’m picturing myself hauling buckets of human waste to the nearest farm. City officials are urging families to stock up with an emergency supply of drinking water. It’s a wise suggestion, except for the fact that bottled water, especially the five-liter (1.1 gallon) version, is impossible to find. They sell out the minute they arrive in stores. My husband now drops off our daughter at school a half hour early so he can hit the supermarkets as soon as they open, hoping to score whatever water may have been delivered overnight. So far, he has had no luck. Even the shopkeepers don’t know when the next shipment will come in. The next best option is to go to one of the city’s natural springs to fill up a jerrican, in an experience that feels like an early rehearsal for what is to come on Day Zero. At the Newlands spring, on the premises of a beer-bottling plant, lines undulate for hours down the street. People arrive long before the 5 a.m. opening, and brewery security guards police the crowd to ensure order. Official plans for Day Zero state that there will be approximately 200 water-collection points around the city, but many questions remain. A quick, back-of-the-envelope calculation suggests that even if only a quarter of residents show up to get their family allotment, each site will see some 5,000 water seekers a day. How will they transport the water? Where will they park? Even if each site has 50 to 200 taps each, as the city promises, the lines are likely to take hours.

Swimming pools dot Kreupelbosch, an upscale Cape Town suburb, amid a worsening water crisis. Mikhael Subotzky and Johnny Miller for TIME Lucky are the neighbors with pools—a built-in bathing option and emergency supply of water combined. For the past six months, it has been illegal to top up pools with municipal water; instead, most people buy well water trucked in by the tanker, sourced from outlying areas unaffected by the restrictions. The newly lucrative business sits in a legal gray area. Technically, it’s illegal to sell water from wells or rivers, but you can charge for labor and transport. In listings on the local version of Craigslist, aquapreneurs are offering to “deliver” water for 4.5 rand, or 36¢, a gallon. If you want drinkable water, it will cost more. Despite lacking a pool, I figured stockpiling that kind of water would help mitigate any Day Zero chaos. All I needed was a big tank. I wasn’t the only person with this idea. All the hardware stores told me there was a six-to-eight-week waiting list for new stock, but a friend knew someone who knew a guy who could get one quicker, if I was willing to pay. I pulled into a sand-dusted parking lot only to be shown a pair of plastic barrels that didn’t look as if they would hold a cup of tea, let alone hundreds of gallons. He was charging $350 each, nearly twice the market price of even the best-quality tanks. I kicked myself for my lack of foresight. Four months ago, I could have filled up an affordable top-of-the-line tank with the spring rainwater gushing from my roof. Instead I let myself be swayed by soothing assurances from Mayor Patricia de Lille that she would “not allow a well-run city to run out of water.” Like the rest of us, the mayor was in denial. It is no small consolation that she has since been removed from her water-crisis-management duties and is facing a vote of no confidence in city government.

Aubrey Witbooi carries water from the Kalk Bay spring on Feb. 1 2018. Mikhael Subotzky—Magnum Photos for TIME

Obaidullah Cader, Muneerah Traut and Saajidah Cader collect water at the Newlands spring on Feb. 3 2018. Mikhael Subotzky— Magnum Photos for TIME But the crisis didn’t start with one deluded mayor. It goes back decades. “City will run out of water ‘in 17 Years,'” declared a Cape Times headline as long ago as April 26, 1990. The story, quoting a Water Research Commission presentation in the country’s Parliament, went on to recommend a sewage-recycling plant to augment supplies from the city’s six rainfall-fed dams. Instead, Cape Town embarked on a laudable conservation effort, but it wasn’t enough. Now the city is playing catch-up, installing expensive desalinization plants to purify seawater and scrambling to tap the underground aquifer. Only two of the seven water-augmentation projects are expected to be up and running by the time Day Zero kicks in. As with my back-ordered water tank, a little extra planning would have gone a long way toward staving off a crisis of unprecedented proportions.

Mwai Halala, with her son, Jaden, holds water from Newlands on Feb. 3 2018. Mikhael Subotzky—Magnum Photos for TIME

John Levi, a water porter at Newlands, on Feb. 3 2018. Mikhael Subotzky—Magnum Photos for TIME Cape Town may be the first major city to run out of water, but it won’t be the last. In Mexico City, residents are already ex periencing cuts to their piped water supply, and officials in Melbourne (another city affected by drought) warn that the city is little more than a decade away from exhausting current water supplies. At one point, before a sustained deluge in 2015, the city of São Paulo was down to less than 20 days’ worth of water, according to the World Resources Institute, a Washington-based research organization that tracks the use of natural resources around the world. It reports that more than a billion people currently live in water-scarce regions and as many as 3.5 billion could experience water scarcity by 2025 if steps are not taken to conserve water now.

On Feb. 3 2018, Theewaterskloof Dam was at 13% of capacity. Mikhael Subotzky and Johnny Miller for TIME As the clock to Day Zero counts down, Capetonians are joining together to share water-saving tips online, and on air through call-in radio shows. Don’t boil food; bake it or cook it on the grill. Use paper plates, and wrap your serving platters in plastic wrap that you can peel off after each meal. Order pizza and eat straight from the box. Shower over a plastic tub and then use the water to wash clothes. Keep a bowl in every sink to collect water from washing hands. Use that and laundry water to flush toilets. The tips, and the attention, have begun to work. On Jan. 30, Mmusi Maimane, head of the Democratic Alliance, a political party that has controlled Cape Town for a decade, announced that residents’ efforts were having an impact. But, he warned, unless consumption fell by at least 25%, Day Zero was still inevitable. The problem is, short of turning off neighborhood taps, there are few effective ways to cut off profligate users. Instead they are fined, in amounts that range from $41 to $248—which in some wealthy neighborhoods is hardly a deterrent.

The future site of a desalination plant on Feb. 1 2018. Mikhael Subotzky and Johnny Miller for TIME The urge to flee before the taps are switched off is strong. Some of our friends are planning to temporarily repatriate thems elves to homes in the U.S. and Europe. Businesses are thinking of relocating to Johannesburg, a two-hour flight away, for the duration. I, for one, plan to stick it out. After all, for the large number of South Africans who don’t live in the mainly white and affluent suburbs that still define Cape Town 25 years after the end of apartheid, Day Zero is just another day in the life. Many township residents already line up at a central tap to get their daily water supply. In a nod to this reality—and out of fear of massive social unrest—city officials have said that townships and informal settlements will be exempted from Day Zero plans to shut off water mains. I’m sure I can handle 6.6 gallons of water a day. I’m already down to two showers a week, and my daughter would go down to once a month if I let her. I’ve sacrificed a corner of my terrace to make way for my back-ordered water tanks. Once they arrive, I will start

praying for rain and interspersing my bucket baths with baby-wipe top-ups. I’ll be cooking my meals on the grill and eating with my fingers. The toilets? Until now, I’ve been using my shower water to flush. But once the taps run dry, we won’t even have that . So I reopen my Internet browser and again type dry composting toilets. With any luck, they won’t have sold out. Aryn Baker is TIME’s Africa Bureau Chief, based in Cape Town. Follow her on Twitter @arynebaker. Mikhael Subotzky is a photographer represented by Magnum Photos. Follow him on Instagram@mikhael_subotzky. Johnny Miller is a photographer and drone pilot based in Cape Town. He is the founder of AfricanDrone, a pan-African drone pilot community. Follow him on Instagram @millefotosa. http://time.com/cape-town-south-africa-water-crisis/

How to Defeat Drought Cape Town is running out of water. Israel offers some lessons on how to avoid that fate. BY SETH M. SIEGEL | APRIL 9, 2018, 8:00 AM

Jewish workers drill for water in Kfar Monash, a farm in the British Mandate of Palestine, in August 1946. (Zoltan Kluger/GPO via Getty Images) Cape Town, a city of some 4 million people, is on the verge of running out of water. Terrifying as it may sound, water scarcity never happens by surprise. At the national, provincial, and municipal levels, South Africa’s leaders ignored Cape Town’s shrinking water supply until it became a crisis. The mistakes were made at every level. Cape Town had no comprehensive, long-term water plan to match water resources with the city’s soaring population. No official program encouraged planting water-efficient crops or the use of water-saving technologies. The city reused only 5 percent of its wastewater for industrial and irrigation purposes. Its water has long been free or heavily subsidized, with no market-based incentive to conserve. And most remarkably, although Cape Town sits on a long seacoast hugging the southern Atlantic Ocean, officials delayed building desalination plants for the city. But solutions to these problems exist. South African officials should have paid attention to a country that long ago figured out how to deal with water shortages before they appear: Israel. From its founding, Israel not only prioritized water conservation but celebrated it. Theodor Herzl, the father of modern Zion ism, wrote in 1902 that water engineers would be the heroes of a future Jewish state. Heeding that call, Israel’s pre-state leaders developed an early ideological commitment to preserving and expanding the water supply. The challenge of keeping a desert country hydrated lured Israel’s most talented minds. (In his 1994 Nobel Peace Prize lecture, Israeli Prime Minister Yitzhak Rabin revealed he, too, once dreamed of becoming a water engineer.) In the early 1960s, Israel began two science fiction-like water initiatives. Recognizing that the amount of water used by showers, dishwashers, and toilets was predictable, Israel turned its national sewage supply into an enormous alternative source of water.

By the early 1980s, Israel had routinized the aggregation and purification of the country’s wastewater and built a parallel water infrastructure system to transport treated water to farms. Today, nearly 90 percent of Israel’s sewage is treated to an ultrapure level for agricultural use. By comparison, Spain — second globally in sewage recycling — reuses only some 17 percent of its wastewater. The United States repurposes far less. Israel’s second grand water initiative — desalination — also began in the 1960s. Around that time, an experimental desalination facility in Israel produced a cubic meter of water (264 gallons) for $14. Israel now produces the same volume for about 50 cents. By filtering seawater through holes that are about 0.5 nanometers wide, the desalination process captures all dissolved material, including salt, while the fresh water passes through. (The resulting water is so pure that minerals need to be added back in.) These two technologies alone could prevent, or at least postpone, another Cape Town-like crisis. Yet both are expensive and sometimes out of reach for countries on the lower end of the economic development scale. But Israel also leads in low-cost efforts to reduce the use of water for agriculture. On average, countries use 70 percent of their fresh water to grow food. Highly inefficient water users — such as Egypt, Ethiopia, and Iran — use as much as 95 percent. Adopting drip irrigation, as Israel has, could reduce those percentages significantly. Drip irrigation uses about half of the water that traditional irrigation does and can also produce larger yields. Liquid fertilizer can be added along with the water droplets, minimizing groundwater contamination. In both India and China, adopting drip irrigation has increased farmers’ yields and limited groundwater contamination from traditional fertilizer. India subsidizes loans for farmers as an incentive for more rapid adoption. Israel also saves water by not losing it: Around the world, water systems lose roughly 30 percent of their water after it has been transported, mainly due to cracked pipes and system power surges. Israel prioritizes keeping leaks to a minimum by fixing pipes or by replacing them before they require emergency repairs. The country’s water loss is now below 10 percent due to the use of algorithms that predict where leaks will occur, underground pipe repair tools, and a sonarlike system that pinpoints the locations of tiny leaks that have the potential to grow larger. What makes Israel such a valuable model for water management isn’t any one thing the country does but the range and integration of its activities. Outside of technology, Israel offers three other replicable elements for all countries. In Israel, professionals, not politicians, determine water policy. Israel’s most senior water official — the water commissioner, who oversees the Israel Water Authority — serves a five-year term after appointment by a cabinet minister. The water authority makes all water-related decisions. The staff comprises highly skilled civil servants who are protected from the type of political meddling common in other countries. Second, the Israeli public bears the cost of sourcing, transporting, and cleaning water; there are no government subsidies for water. People are, therefore, theoretically not only more careful with what they use, but they are also happy to substitute lower-cost technology when it helps save water. This sets off a virtuous cycle of invention leading to greater efficiency. Finally, Israel has built its respect for water conservation into its culture. Beginning in kindergarten, children learn about water efficiency. The United Nations predicts that by 2025 as much as two-thirds of the world’s population could be living underwater- stressed conditions. With climate change, growing populations, and rising affluence, the demand for water is likely to grow substantially in the years ahead. Higher food prices, social unrest, the failure of some fragile states — even mass migrations and a challenge to the world order — could all result. But as Israel has shown, none of those need occur. This article originally appeared in the April 2018 issue of Foreign Policy magazine. Seth M. Siegel is the author of "Let There Be Water: Israel’s Solution for a Water-Starved World." http://foreignpolicy.com/2018/04/09/how-to-defeat-drought-israel-south-africa-cape-town-water/

Submit Your Homework

Let us help you with your homework, we will match you with one of our professional tutors.

My Info

Project Info

Due Date

Describe your Homework Problem

Budget (optional)

By providing your budget we will work on finding the best tutors that can work within it.

Get help from top-rated tutors in any subject.

Efficiently complete your homework and academic assignments by getting help from the experts at homeworkarchive.com