Built in a lake basin, Mexico City experiences destructive flooding, but also struggles with access to clean water. ASU researchers are studying the complex choreography of natural environment, physical infrastructure and human decision-making that contribute to both.
A mural painted by Diego Rivera showing the Aztec city of Tenochtitlan. Photo by Wolfgang Sauber under a CC BY-SA 3.0 license.
By Diane Boudreau
June 2, 2016
Nearly 1,000 years ago, the Aztec people left their ancestral home of Aztlan under orders from Huitzilopochtli, the god of sun and war. He prophesied that the people would find a new home when they saw an eagle perched on a prickly pear cactus, devouring a snake—or so the legend goes.
The people wandered for hundreds of years through what is now Mexico. They settled in several places, but never for very long—their propensity for human sacrifice did not endear them to their neighbors. Finally, in 1325 AD, the Aztecs found their unlikely sign. Unfortunately, the cactus on which the eagle perched grew out of a small island in the middle of Lake Texcoco. Undaunted, the Aztecs built the city of Tenochtitlan, and eventually a flourishing empire, on the swampy site.
“They built this amazing civilization. But of course living in a lake bed imposed a lot of challenges. It was also a huge benefit in that the lake itself provided food for the communities, and it was also a defense,” says Hallie Eakin, a human geographer at Arizona State University.
Eakin studies Mexico City, which was built on the ruins of Tenochtitlan. Today, the city’s metropolitan area is home to 21 million people. It is one of a growing number of megacities—cities with more than 10 million residents—in the world today.
The urbanization of Mexico City has wrought massive changes on the ecosystem it inhabits. For example, Lake Texcoco is almost completely gone. Conversely, the ecosystem has a profound impact on the city and the people who live in it. For instance, the city’s location in a lake basin leads to extreme flooding.
Eakin leads an interdisciplinary research project called MEGADAPT, which explores the challenges of flooding, chronic water scarcity and associated health problems in Mexico City. The project is a collaboration between ASU and the National Autonomous University of Mexico (UNAM), funded by the National Science Foundation.
Unlike some cities that have just begun experiencing major flooding due to climate change, like New York, the people of Mexico City have been adapting to their environment for centuries.
“The city itself appears to be very capable of managing extreme events, in the sense that they don’t cause major collapse,” says Eakin, an associate professor in ASU’s School of Sustainability. “However, that doesn’t mean there aren’t huge costs to that stability. The question for us is whether those forms of adaptation are preventing the system from getting into a more sustainable state, in terms of who bears the risk and then the cost.”
The relationships between a city, the natural environment and human decision-making are extraordinarily complex—particularly at the scale of a megacity. Sometimes solving one problem can inadvertently create others.
For example, even though Mexico City counts more rainy days than London, clean drinking water is often scarce. About 70 percent of residents have no running water for more than half of every day.
One way to get potable water is to pump it from underground. However, this can cause the land to sink down into the emptied aquifer, making the area even more vulnerable to floods.
To understand these complex interactions, the researchers are integrating vast amounts of data about the natural environment, urban infrastructure and human behavior.
“Our aim is to pull this all together and create a platform for decision-making in which we can understand what is really driving the dynamics of risk in this city. How much of it is coming from problems of topography, problems of increased rainfall, and how much is actually due to human decisions?” asks Eakin. Ultimately, the model they produce could be adapted for other megacities around the world.
Building a storm
The team has already made one important discovery. As Mexico City grows, rainfall is increasing. The finding was published in the Journal of Geophysical Research on April 4.
For a city located in a lake basin, more rain can cause problems.
“The city today faces continual problems with chronic flooding, from intersections becoming impassable in the rainy season because of lack of drains, to occasionally severe floods of a meter or so of water over particular neighborhoods. It’s not just the flooding, it’s that this is a combined rainwater and sewage system, as many old cities are. And that water is highly contaminated,” says Eakin.
Matei Georgescu experienced this firsthand when he met with MEGADAPT collaborators in Mexico City.
“I was stuck for about two-and-a-half hours in a bus because of the flooding. When all the main arteries get flooded all the public transportation comes to a complete stop,” he says. “We were finishing up a meeting at UNAM and were supposed to go back to the hotel to meet for dinner. We never made it to dinner. But it was an experience to fully understand what the flooding problems are.”
Georgescu is an assistant professor in the School of Geographical Sciences and Urban Planning. He leads the climate modeling portion of the MEGADAPT project.
“Previous research has shown through observation that rainfall has been increasing over the Mexico City metropolitan area over the last century or so,” he says. “People had assumed that this increase in rainfall was somehow associated with urbanization, or what’s known as the urban heat island. But the link was never made. Correlation doesn’t mean causation.”
His team used satellite data to create a highly detailed map of Mexico City, identifying different land uses such as residential or commercial. They entered that data into the Weather Research and Forecasting (WRF) model and compared their results to the weather that was actually observed in the city.
Valeria Benson-Lira, the lead author on the study, traveled to Mexico City and worked with the Servicio Meteorologico Nacional (the Mexican weather service) to gather the observed weather data. The work was part of her master’s thesis research.
“Because I’m from Mexico, I wanted to do something that will have an impact on my country,” says Benson-Lira, who received her master’s degree in geography in May 2015.
Because of the mass quantity of the information, she had to retrieve it in person. She sifted through data from 16 weather stations that took readings every 10 minutes between 1999 and 2012. That adds up to nearly 11 million data points.
The team compared a subset of this data with the WRF model’s predictions and found that they matched closely. This told them that their adapted model works well.
Next they entered data about the pre-urban land cover of the area, developed at the University of Wisconsin, into the model. They found that Mexico City today experiences more rainfall than in the past, due to the growth of the city.
How does building a city attract more rain? Georgescu says two elements are required for precipitation to occur: moisture in the atmosphere and rising air, which forms condensation as it cools.
Precipitation systems tend to move over Mexico City in the late afternoon and evening, providing the first element—moisture. But they need rising air to unleash a rainstorm.
Urban structures like roads and buildings absorb more heat from the sun during the day than natural environments, creating an “urban heat island” effect. They release that heat in the evening hours. This provides the energy to lift the air, right at the time that precipitation systems move over the city.
“We’re not creating new precipitation systems, we’re just raining more out of what already existed. So essentially rainfall becomes more efficient,” says Georgescu.
The human element
MEGADAPT will take models like Georgescu’s adapted WRF model, integrate them and allow them to “talk” to each other. These include a hydrological watershed model and models of how the city distributes waste and drinking water through its infrastructure.
It’s not enough, however, to understand physical processes. The model also needs to account for human decisions and actions. Eakin interviews many people—from government officials to farmers to urban residents—about what issues they perceive, how they prioritize them and how they respond to them.
For example, Mexican disaster-management officials described major flooding that inundated homes in low-income neighborhoods with sewage-tainted rainwater.
“They basically lose all of their property. They have to evacuate populations. Things that are contaminated all have to be thrown out. Small business owners who are losing all their stock may not have insurance that covers these losses,” she says.
When she visits households, Eakin often finds that people have elevated important appliances like stoves onto bricks. Some will even elevate their thresholds, raising the front door off the ground to prevent water from getting in.
She says these adaptations are helpful, “but on the other hand, they are kind of letting the populations that are the most vulnerable dedicate scarce resources to this type of risk management, which really doesn’t get at the heart of the problem. We’re trying to understand whether there are interventions that can be made in a much more systemic way that would avoid these populations having to bear the burden.”
Every adaptation that people make, whether on an individual level like elevating appliances or on a government level like building infrastructure, is a decision based on certain perceptions, priorities and responsibilities.
For example, a community suffering from water scarcity might petition the government for rainwater capture systems. They see plenty of water washing away during the rainy season, and they get frustrated and angry with the government for not implementing what seems to be an obvious solution.
“And the government may say, ‘It’s all very well to think about rainwater capture but it’s going to be a drop in the bucket when we think about the volume we need to satisfy our population. Therefore it’s not an investment priority for us,’” explains Eakin. “People get frustrated that the solution they think is viable is not being adopted, and the government thinks people just don’t understand the complexity of the problem.”
The MEGADAPT model will allow users to compare the costs and benefits of different solutions, reveal unintended consequences, and offer insights that could lead to potential new solutions, as well. For instance, knowing that urban heat increases rainfall, city officials might consider exploring green infrastructure to reduce temperatures.
The researchers will leverage UNAM’s Decision Theater facility—modeled after the original Decision Theater at ASU—to help stakeholders simulate and visualize the modeling data. The team also plans to make a user-friendly version accessible to the public.
“I would be naïve to say we’ll solve it, but we’re trying to get some better instruments to help people think about these very complex interactions,” Eakin says. “It’s enormously complex.”
The team’s approach in Mexico City could also be applied to other cities.
“The developing world is urbanizing. Their data is not always in the best condition or widely available. There may be plans and protocols for certain things but they may not always be followed to the letter,” says Eakin. “This is the reality of most urban environments in the world. If we can get a hold of how to grapple with these complexities for more effective decision-making in Mexico City, perhaps we can extend this to Dakkar and New Delhi and Guatemala City.”
The MEGADAPT team is led by principal investigator Hallie Eakin and four co-principal investigators:
Charles Redman, founding director and professor, ASU School of Sustainability
David Manuel-Navarrete, assistant professor, School of Sustainability
Enrique Vivoni, professor, School of Earth and Space Exploration
This research is supported by the National Science Foundation under award number 1414052.