Intelligent Climate Adaptation and Resilient Engineering for Urban Sustainability
Auroop R. Ganguly, Northeastern University
Auroop R. Ganguly, Northeastern University, provided an academic perspective of urban sustainability. He said that there is a clear and present need for pathways to urban sustainability, including economic development, social progress, and environmental motivators as key indicators. To highlight issues of risk and resilience quantification, he shared examples of climate risk management frameworks from the Intergovernmental Panel on Climate Change, the emerging resilience paradigm (see Ganguly, Bhatia, and Flynn, 2018), and a recent report1 to support climate action in the United States. While natural climate variability is beyond human control, greenhouse gas emissions and land use change, for instance, are partly within human control. Even hazards such as Hurricane Katrina may not turn into disasters if vulnerability (e.g., the levee that broke) and exposure (e.g., human habitations below sea level) are managed. To do
this, Ganguly suggested that cities integrate consequence management with longer-term adaptation and urban mitigation plans.
Ganguly also discussed the resilience of interdependent systems to stresses and disruptions. Cities need to evaluate how reliably, quickly, and effectively they can recover from natural disasters (i.e., by restoring critical lifelines such as communications, power, transportation, water, and financial services) as well as consider how to maintain overall levels of functionality through graceful degradation. He emphasized that cities should use disruptions as an opportunity to improve their systems rather than just aiming to restore their systems to their original state. He described the value of designing and incentivizing a virtuous cycle of risk and resilience for cities. For example, sea level rise and other indicators of change could be used to assess the value and risk of more systematic recovery approaches (in this case, potentially using a network-topology-based approach). An important part of this is eliminating the cycle of disincentives (e.g., more funds tend to be available for disaster relief than for preparedness efforts) that can stagnate technology development and increase outdated practices (Ganguly et al., 2018). He stressed the need for financial incentives to achieve these goals.
Ganguly suggested that the scope of “urban” should move beyond city boundaries to consider the larger connected ecosystem in which cities exist. Urban impacts can carry across regional, national, and international borders and can include interrelated infrastructures and supply chains. However, few governance structures can look at these mega-regions holistically. Large-scale coupled natural-engineered-human systems can also pose unique challenges. As an example of the interconnected nature of infrastructures, a surge in agricultural water demands in India from a heat wave and delayed monsoons in 2012 led to large groundwater withdrawals, partly as an unintended consequence of price incentives given to farmers. This caused increased stress on the power grid and a massive blackout, which then downed the rail service.
A specific challenge is managing the risks of urban heat on public health. Urban planners could reduce urban heat intensities through decisions about open spaces, bodies of water, and use of green roofs, while the impacts of urban heat islands could be managed through cooling centers, public health facilities, early warning systems, and community education and resilience-building exercises. Ganguly reiterated the value of thinking about consequence management and mitigation when thinking about development pathways (e.g., how to help senior citizens in disadvantaged areas access emergency facilities during a heat wave). Other urban challenges relate to coastal infrastructures and ecosystems as well as the food-energy-water nexus. To maintain urban services—such as buildings, bridges, the power grid, transportation systems, and means
of communication—the gap between the science that is currently understood and the skills or knowledge available to inform engineering and policy needs to be closed, he asserted.
Ganguly presented a selection of novel approaches to connecting science and policy. The Next Generation Digital Earth, which includes a vision for augmented images of city structures and simulations of urban systems, aims to create a digital replica of real systems. Interlinked data systems, physics-guided data science, hybrid physics and data science approaches, and physics-guided uncertainty quantification can all inform decision making. Entrepreneurship and partnerships among private, public, government, and intergovernmental entities can help build communities around these issues. Together, these efforts can enable better understanding of different lifelines, capabilities, kinds of data and ways of processing them, and types of threats, he concluded.
Connecting Research with Civic Action
Jeanne Holm, City of Los Angeles
Jeanne Holm, City of Los Angeles, provided a city government perspective of urban sustainability, emphasizing that connecting academic research and civic action is key to making change. She noted the importance of using varied learning modalities to educate the next generation around data science tools. Cities should be efficient, effective, safe, resilient, and innovative, and data and technologies should be used to create a city ecosystem that benefits its citizens equitably, Holm continued.
Holm explained that the city of Los Angeles is the 20th largest economy in the world—4 million people living in 500 square miles—which brings both great opportunities and challenges. She suggested that technology provides an opportunity to bring people together to decrease the high inequity in Los Angeles County. She emphasized that data-driven approaches could lead to new policies that impact outcomes. A number of innovative projects are under way in the city of Los Angeles, with the support of its tech-focused and data-driven current mayor, Eric Garcetti. In any city, Holm continued, data come from many different sources, including sensors, satellites, and airborne instruments. Smart cities, such as Los Angeles, have sensors on garbage trucks, street sweepers, streetlights, vehicles, and even animals.
Holm and her team are working to understand which of these data could be used to help make better decisions for the city as well as how to make those data interoperable to drive action. She explained that the city’s approach utilizes many technologies, such as artificial intelligence, predictive analytics, personal assistants, machine learning, Internet of
Things, blockchain, gamification, virtual and augmented reality, and open data ecosystems; all of this is possible only with participation from everyone in the city’s ecosystem (e.g., the mayor, businesses, the city council, employees, universities, global partners, and the citizens).
Holm reiterated the value of connecting with academic institutions and research organizations to understand which data will create better actions and outcomes for citizens and to activate citizen scientists. Data interoperability, standardization, provenance, and authenticity are important to enabling these partnerships. She noted that as modeling and simulation are becoming more widespread, it is important to give citizens an opportunity to co-create their future, especially through storytelling. Although futuristic films such as Blade Runner present a dystopian perspective, Holm reminded participants that it is their choice as to whether a dystopian or utopian future is created.
She explained that the city of Los Angeles owes much of its success to formalized partnerships. For example, the Data Science Federation2 involves data science professors and students from 17 universities and colleges in California and Arizona who work with city departments on specific data challenges. This program allows members of the next generation, who are often excited to make a difference in their communities, to understand the power of government service and to explore potential career opportunities. The city of Los Angeles also works with the national Code for America;3 its local organization, Hack for LA,4 connects citizens’ ideas for their communities with academic research and formal data. The LA CyberLab5 partners with more than 500 businesses, the majority of which were started by immigrants, to help them understand cybersecurity vulnerabilities. There is also an effort to encourage citizen scientists to collect data for the city by playing augmented reality games, such as Agents of Discovery, that allow them to explore city parks. Another partnership, SmartAirLA,6 works to address public health issues that arise from air pollution at the Port of Los Angeles: global positioning system smart inhalers are issued to people who live near the port, and they generate heat maps identifying where people had a difficult time
breathing. This helps asthma sufferers avoid high-risk areas and reduce their number of attacks.
City data are also collected from the public via MyLA311,7 an app and call line established for people to report city problems. All of this information is shared on an open data portal and with transportation organizations (e.g., Waze8) so that people are aware of and can avoid unsafe areas. ShakeAlertLA9 is an earthquake early warning system developed in partnership with the U.S. Geological Survey (USGS). Sensors detect when the ground shaking begins and send a signal in 1.8 seconds to the more than 470,000 people who have downloaded the app to let them know how severe the shaking will be at their location. Holm noted that the city invested in federal government work, such as with USGS to build out the earthquake early warning sensor network, because of its value for the citizens of the city of Los Angeles. She added that mayors from more than 400 North American cities have signed a pledge (led by Mayor Garcetti, who chairs Climate Mayors) to take action on climate, and the city of Los Angeles, in particular, has a plan based on the United Nations’ Sustainable Development Goals,10 which it hopes to achieve in time for the 2028 Olympic Games.
The Policy and Planning Perspective
Bill Fulton, Rice University
Bill Fulton, Rice University, provided a policy and planning perspective of urban sustainability and focused on ways that data can be used to improve how municipal governments function. He described a municipal government as a series of operational departments, each with a particular goal and task. It is difficult for these departments to share data in a meaningful way because they have collected data for their own purposes, for
10 The Sustainable Development Goals are as follows: no poverty; zero hunger; good health and well-being; quality education; gender equality; clear water and sanitation; affordable and clean energy; decent work and economic growth; industry, innovation, and infrastructure; reduced inequalities; sustainable cities and communities; responsible production and consumption; climate action; life below water; life on land; peace, justice, and strong institutions; and partnerships for the goals. For more information about these goals, see United Nations, “About the Sustainable Development Goals,” https://www.un.org/sustainabledevelopment/sustainable-development-goals, accessed March 4, 2019.
their own uses, and in their own ways. He added that many municipal governments in the United States are located in small cities and lack analytical capabilities—they are often largely motivated by budget analyses and their impact on operations. Operational data are relatively rare, and although some police departments are sophisticated in the way that they use data, such a practice is fairly atypical, he continued. Municipal government departments are becoming overwhelmed by the abundance of available data. Fulton noted that most city governments are not currently set up to facilitate the use of that amount of data, thus missing opportunities to use data to improve the functioning of the city.
Fulton said that it is crucial to think about how to make data-driven decision making attractive to people who have the power to create political change. First, it is important to pay attention to what can be measured and how, as well as whether the right entities are being measured (i.e., for sustainability as opposed to operations and function). Second, it is necessary for the people who are running municipal governments to understand that investing in data sciences is valuable. Data can then be used in a more efficient, more cost-effective way with the people already on staff. Third, governments need to embed data-driven approaches in their operations so that they become part of the landscape of city management. Fulton explained that change is happening in cities with tech-savvy mayors, but operations should not be dependent on tech-savvy top-level leadership. Fourth, he continued, it is important for municipal governments to understand the value of partnerships, especially with academic institutions—it is impossible for any single municipal government to accomplish such vast change, especially in terms of developing analytic capabilities, without partners. He acknowledged that it can be difficult to initiate partnerships but reiterated that it is essential for progress and is the best way to engage with current data privacy problems.
Katherine Bennett Ensor, Rice University, Moderator
Auroop R. Ganguly, Northeastern University
Jeanne Holm, City of Los Angeles
Bill Fulton, Rice University
Data Sharing and Security
Ulrike Passe, Iowa State University, asked the panelists how to prevent dystopian views of the impact of data, models, prediction, and analysis on society and how to ensure that infrastructure remains secure. Holm responded that this question is essential to understanding the provenance
of data and to ensuring that data remain accessible and available at all times to the people who need them. It is also important to understand the challenges of modeling based on those data, including inferences and biases. Katherine Bennett Ensor, Rice University, acknowledged that data security and access are crucial and said that the Kinder Institute11 is creating the Urban Data Platform,12 which is a library of past and current data about the Houston area. She added that archiving data (and thus having a record of data provenance) is as important as collecting real-time data; such pursuits are opportunities for city–university partnerships. Ganguly wondered if there is a way to expand the archiving and sharing initiatives that are occurring within cities, as Ensor described, across networks of cities so that more data sets are available to more people.
Deborah Goodings, George Mason University, commented on the importance of data repositories. She wondered what type of national organization could host a large data repository into which many cities upload their data to be curated, anonymized, and made available. Fulton said that a university could host such a repository as long as it is tied to the university’s infrastructure and not to an individual professor’s research project. Passe added that land-grant universities should be a part of this conversation because they have the capacity to be in every county in their respective states. Holm explained that some data archiving is available through data.gov/cities, and Ganguly said that national laboratories as well as open access platforms, such as Oasis,13 have an interest in this issue. Ensor championed the value of open data and suggested that published data be held to the same standards as published papers in terms of product ownership and legitimacy. She added that because it is important to understand the data that drive decisions, it is necessary to build infrastructure for transparency and reproducibility.
Jerry Miller, Science for Decisions, referenced the National Academies’ report Pathways to Urban Sustainability: Challenges and Opportunities for the United States (NASEM, 2016) and noted that data access remains a major barrier for small cities. Given the data that are currently available for operations in the urban setting, he asked the panelists how to develop data pipelines that will support decision making decades into the future. Fulton used the example of sensors to show that the presence of data is not synonymous with the value of data. He emphasized the need to be strategic about which data are valuable in helping to achieve sustainability
goals. Holm added that part of the challenge is incentivizing cities to act and to share their data when each city has its own motivation. The city of Los Angeles has a strategic plan for sustainability and a strategic plan for resiliency, both of which are action oriented and connected to the United Nations’ Sustainable Development Goals. She emphasized that all cities have to prepare and would benefit from data-driven planning, even if what they are preparing for is much smaller in scale than the Olympic Games. She encouraged partnerships between data scientists and philanthropic organizations to help actualize research in a city space. In response to a question from Ensor about how such programs are sustained once the philanthropic organization steps out, Holm said that program sustainability should be built into the city budget process. Fulton added that city managers are driven by budget, so it is essential to show a return on investment for any program.
Partnerships for Sustainability
Roland Gamache, George Washington University, asked how a university could act to broker partnerships with cities beyond the jurisdiction of an established city–university partnership. Ganguly described organizations such as the Thriving Earth Exchange of the American Geophysical Union14 that are trying to match stakeholders with smaller, lower-resource cities. He referenced Northeastern University’s Global Resilience Institute15 as well as the Boston Area Research Initiative16 as other innovative programs. Holm added that standards organizations such as the International Organization for Standardization17 and the World Wide Web Consortium18 are working to create new schema around data to improve searches. Ganguly also noted that openly available data could be used to solve issues of importance to these cities and that sensors are becoming more affordable.
Fred Abousleman, Oregon Cascades West Council of Governments, mentioned that developing partnerships can be difficult and wondered how to make information more accessible both nationwide and at the local level. He emphasized that the problem is not a lack of data or resources
16 For more information about the Boston Area Research Initiative, see https://www.northeastern.edu/csshresearch/bostonarearesearchinitiative, accessed March 12, 2019.
but rather a lack of excitement and a fear of high costs. Because cities have so many different approaches, it is difficult to identify a sustainable path forward. Fulton agreed that the presence of these varied approaches is a hindrance right now. However, in time, he believes that a level of standardization will arise, which will make things less expensive and easier for smaller jurisdictions. For now, he suggested leveraging available university and philanthropy capacity as well as any local assets—pooling the cost reduces the burden and increases the return. Holm said that one must always be able to show a cost–benefit for the city; a small up-front investment with a university partner could change the dynamic around a city problem. For example, investing in predictive analytics to try to prevent homelessness through the offering of additional services can provide clear return on investment for a city, Holm explained. Ganguly asserted that there are incentives for those in academia to engage in this type of work.
Chibuzo Okoro, Georgetown University, asked about incentives for communities to create more engagement for data gathering. Holm replied that communities have amazing opportunities for citizen scientists. She referenced the Federal Citizen Science and Crowdsourcing Toolkit,19 which is open to the public and connects citizen science efforts to scientific research (i.e., citizens gather data that can be converted to help scientists and ultimately lead to government action). Ensor said that there is an ongoing conversation about the best ways to integrate this seemingly haphazard information with more measured information, taking into account sampling biases that might be present.
Aniruddha Dasgupta, World Resources Institute, referenced previous commentary from the panelists about the value of city–university partnerships and wondered how to make such partnerships work with nations that are newly industrialized or are in the process of industrializing (referred to as the Global South). He also asked how to help smaller cities make the political point that scientific capabilities are needed for decision making. Fulton explained that U.S. cities are motivated to make the most of their often-limited budgets. This can present impediments for partnerships with academic institutions, especially because incentives do not always exist for tenure-track faculty to engage in these real-world problems. Ganguly said that there are creative ways to get around this lack of incentive structure. In response to Dasgupta’s questions about the Global South, he shared his experiences traveling with U.S. students to India to learn about solving problems from physical infrastructure and natural systems perspectives. Lessons learned include thinking about technological benefits in a sociopolitical context to make improvements
for a city. Holm briefly discussed working with local governments in the Global South to create open data portals and to partner with local universities to inform and drive change in their communities and foster transparency in government. Audience member David Rabinowitz noted that many cities have universities with students who want to do projects in data analysis, but such projects become unsustainable after the students leave. He and Ensor proposed that this could be a great opportunity for industry involvement to sustain the work.
Best Practices and Lessons Learned
Justin Smith, U.S. Census Bureau, asked how to motivate smaller cities (i.e., fewer than 100,000 residents) to invest in data and data science. Fulton responded that small cities with large universities have great opportunities to improve their use of data analytics. Regional planning agencies are another potential vehicle to pool resources, he continued. Holm added that Los Angeles County partnered with the Southern California Association of Governments and now has data science federation projects in a number of the county’s other 88 cities, which prompts cities with resources to share the wealth. She mentioned Code for America, the Metropolitan Information Exchange,20 and the FUSE Corps21 as other programs that work to bridge these resource gaps. She also proposed that federal organizations such as the U.S. Census Bureau, which excels in data anonymization and management, could help cities to become more data literate.
Brian Wee, Neptune and Company, Inc., suggested providing exemplars of how other cities have used data to solve particular problems. Fulton agreed that shared learning is crucial—for example, the MetroLab Network22 presents a series of use cases on its website of how universities and cities worked together to solve problems. He suggested that local government organizations (e.g., the U.S. Conference of Mayors and the National League of Cities) identify successful use cases, publicize them, and create a learning community so that cities know what other cities are doing to succeed.
Sallie Keller, University of Virginia, mentioned that in order to have successful partnerships, research has to unfold at a pace that is often
much faster than a university’s capability. She wondered how to create units within universities that can deliver results to their city partners in a timely fashion. Holm agreed that time is of the essence, so the Data Science Federation projects are completed within 10–12 weeks.
The Future of Urban Sustainability
Ensor said that it is crucial to look toward the future when thinking about sustainability. Ganguly noted that, even with modern modeling and data capabilities, it might not be easy to address resilience for plausible futures, especially given changing conditions and complex uncertainties. Holm commented on the power of predictive analytics: ideas about a future that would be desirable for communities are based on an ability to understand the past and the present. Diverse voices are needed to help visualize and shape cities, she continued. She said that the tools exist; the challenge is in identifying which problems to address. Donald Wuebbles, University of Illinois, noted that people recognize the need to develop new tools to translate science into action.