Meet the Future: Big Data and Sustainability Solutions
Thanks to the digital revolution, a burgeoning amount of big data is available in real-time on a mass scale.
Big data is a collection of larger, more complex data sets, especially from new data sources like social media. It offers scientists unparalleled access to vast amounts of raw information that can be harvested, analyzed and translated into new solutions to the world’s most challenging problems—including sustainability and climate change. Big data allows scientists and other experts to make smarter predictions and decisions than before, and it opens doors for researchers to collaborate across disciplines to better understand people and our planet.
We spoke to four University of Michigan School for Environment and Sustainability (SEAS) professors who are using big data to reimagine sustainability solutions to agriculture, landscape conservation, biological systems and cities.
SEAS Assistant Professor Meha Jain is interested in sustainable agriculture, and big data plays a large role in helping her examine the impacts of environmental change on smallholder farming and ways that farmers may enhance agricultural production in response to that change.
Specifically, Jain’s work focuses on the impacts of weather variability and groundwater depletion on agricultural production in India—the second-largest producer of rice and wheat globally—and whether farmers can alter their cropping practices to mitigate these impacts.
Jain combines remote sensing and geospatial analyses with household-level and census datasets to examine smallholder farmer decision-making and behavior across large spatial and temporal scales. Remote sensing is a low-cost way to obtain crop production statistics for smallholder farms, which are 2 hectares in size or smaller, grow food for consumption and are more vulnerable to the effects of climate change and natural resource degradation.
“We use satellite images that are taken everywhere around the globe—almost daily in some cases—and use remote sensing analysis to create algorithms that can translate those images into agricultural data that we can measure,” said Jain.
“We can map things like crop type, yield, and planting decisions, such as when farmers are sowing or how they’re planting their crops. And by doing this, we can understand how farmers are changing their practices because of climate change, what the impacts of those changes are, and what are some of the interventions that are most beneficial to farmers to increase yields.”
Jain, who was introduced to the field of remote sensing as a PhD student at Columbia University, said big data has proved to be a gamechanger in her work because it creates opportunities for large-scale analysis of agricultural metrics, something that had been difficult to do previously.
When Jain first started her research in 2007, she collected data about farmer management practices and yield outcomes by surveying farmers and collecting measurements on the ground. “Doing that was really cost- and time-intensive,” said Jain, “and there were limitations to that approach, such as not being able to get measurements at very large spatial scales or across time. By using satellite data, we’re able to measure these historically difficult things to map—over decadal time scales, over full countries or even at the global scale, which is exciting.”
Because this type of large-scale data can help to identify which interventions are effective or not effective, as well as which regions are more susceptible to the effects of climate change, Jain said large-scale satellite data is a useful tool for organizations and policymakers to develop real-world solutions for sustainable agriculture.
SEAS Assistant Professor Derek Van Berkel has traveled the world, drawing inspiration from the beautiful cities, landscapes and natural areas he has visited. As a geographer, Van Berkel is motivated by a desire to conserve and restore these places for future generations.
His research uses spatial analysis and geovisualizations of social and environmental data and spatial thinking to develop solutions to environmental challenges. Part of this focuses on discovering the special places that humans value and are drawn to, such as national parks, hiking trails, beaches, or particular neighborhoods and cities.
Specifically, Van Berkel uses social media to mine photographs of the locations that people visit around the globe, which can provide clues about how those locations are being used and how they might be managed to increase their sustainability.
“These photos allow us to see how people are using landscapes and natural areas—where they are walking, if they are encountering wildlife, what they are doing in these locations—which can give us a broader picture of society and how we’re impacting the earth,” Van Berkel said.
By analyzing hundreds of thousands of pictures—“potentially instantaneously”—Van Berkel can glean important information, such as traffic patterns or changes to a landscape over time, which can be used to respond more quickly to issues.
“For instance, if the photos show that people sharing photos on social media are getting too close to wildlife, we might be able to work with park managers to change the design of the park so that we limit disturbances to wildlife,” Van Berkel explained. “Another example is that frequent photos of wildlife encounters on a particular hiking trail can be used to limit how many people hike on particular trails.”
The end goal, said Van Berkel, is to translate that information for policymakers, who can then use it to make informed decisions about keeping these valued places—and society at large—more sustainable and resilient.
“We want to convey that it’s important to conserve special places because of their potential to support the economies of rural communities through the activities of visitors,” Van Berkel said, “but also because of their intrinsic value. People broadly appreciate the beauty of nature, and benefit from these locations throughout the world.”
SEAS Associate Professor Kai Zhu uses ecological modeling and environmental data science in his research on global change biology, which is a field that studies how biological systems are impacted by environmental change at a global scale.
Zhu, whose lab is part of the SEAS Institute for Global Change Biology, studies the interactions between climate, soil microbes, and plants.
Climate change can alter soil microbial communities, Zhu explained, which can impact soil nutrients such as carbon and nitrogen and the plants that rely on them. While changes in soil can have a significant impact on the environment, these changes are not always visible.
Understanding these complex interactions can help anticipate changes in forests and nutrient cycles in the future.
“By studying the interactions between soil and climate change, we can better understand the potential impacts and take action to mitigate them,” Zhu said. “We hope to discover new insights into how soil responds to climate change and how we can protect our ecosystems.”
Another research focus for Zhu involves detecting changes in plant phenology, which refers to the timing of recurring biological events, such as leaf-out and flowering, and how they are related to the environment. Phenology acts like “nature’s calendar,” Zhu said, and is sensitive to climate change.
Warmer temperatures, for instance, can cause earlier spring events and longer growing seasons. Zhu points to the timing of pollen release from wind-pollinated trees and its impact on public health as an example.
“Climate change has caused pollen seasons to start earlier, last longer, and be more intense,” he noted. “By understanding pollen phenology better, we can improve our models and predictions to mitigate the risk of pollen allergies.”
Big data is essential to Zhu’s research, he said, because it fosters new opportunities for understanding biological responses and processes at a grand scale. “It used to take a huge effort to collect a few data points, but now, because of technological innovation, the process has become much easier,” Zhu said. “Thanks to the big data revolution, data are growing in quantity and quality at an unprecedented rate.
“Big data allow us to identify patterns and relationships that would be impossible to see otherwise,” Zhu added. “By analyzing massive amounts of data from multiple sources, we can gain a better understanding of the impacts of climate change and how to solve these problems.”
More than half of the world’s population live in urban areas, said SEAS Professor Josh Newell, so it’s imperative that we develop solutions to ensure that our cities are sustainable, livable and healthy. Big data is a key tool in helping Newell not only identify communities that are prone to climate change vulnerability, but also develop strategies to build resilience.
Through his project, “Mapping Community Vulnerability to Climate Change,” Newell is developing an integrative climate change vulnerability index and corresponding hotspot map that will allow state and local municipalities to identify at-risk communities and develop mitigation strategies to climate change.
Newell is using Twitter data—what he described as “millions of tweets”—to understand the degree of climate change skepticism throughout the country and how it varies at specific county and ZIP code scales.
“Through algorithms, we are able to classify these tweets as either believing in or not believing in climate change,” Newell explained, “which allows us to map out climate change skepticism and belief across the entire United States.”
Climate change acceptance and skepticism are important indicators of how willing or unwilling communities are to develop mitigation and adaptation strategies, Newell said.
Newell is preparing to use other datasets to understand heat and flooding vulnerability in communities across the United States, with the goal of “layering heat vulnerability, flooding vulnerability, and climate change skepticism on top of each other to identify hotspots where particular communities are especially vulnerable to these triple stressors.”
Because not all cities share the same vulnerability to climate change, Newell said big data allows researchers to pinpoint and prioritize those communities that are in most need of adaptation efforts.
“Big data are great because they give us massive amounts of information, and enable us to answer questions we haven’t been able to answer in the past,” Newell said. “Big data also enable us to understand on a very nuanced geographic scale the projected impacts of climate change and other environmental pressures on communities. We can combine big data with surveys, questionnaires, and fieldwork, which provides a rich mixed-methods approach to understanding people and the planet.”