As cities grow and populations move, policy makers want to see the signs as soon as possible in order to plan for the changes and consider what these changes might do to the environment.
A researcher from the University of Iowa thinks changes in vegetation might provide powerful signals for policy makers, and he is developing a tool using data analytics to sound the alert. Xun Zhou has developed an algorithm that detects even minor changes in vegetation caused by human action, which will help policy makers prepare for more sustainable growth.
“Policy makers who want to know what’s happening in cities can use this to see how cities are being affected by things that are being done by human beings,” says Zhou, assistant professor of management sciences in the Tippie College of Business, who has received a $155,000 grant from the National Science Foundation to continue developing the algorithm. “If you observe a change to a local environment, there are probably some human activities or movements that caused it. The algorithm detects changes that can’t be observed easily from an individual’s perspective.”
Zhou’s use of vegetation data makes sense. Few markers are as effective in showing environmental changes resulting from human activities and migration than losses or gains in green space. The research takes advantage of huge datasets that governments and policy-making organizations have collected over the years, including satellite and aerial photos, to detect and identify changes in a community. The analysis will help planners direct appropriate resources to areas in question, which could be something as simple as additional public safety patrols or as complex as laying out new streets or protecting waterways.
In particular, Zhou says the results can help policy makers plan for environmental challenges wrought by the loss of vegetation. By spotting the footprints early, Zhou says, planners will have more tools to encourage sustainable growth and prevent out-of-control sprawl.
At the same time, Zhou says his model can determine if the changes that we think we see are actually happening or are just anomalies. Changes in a community’s greenness, he says, can be the result of many factors, and the trend is not always in one direction or another. In fact, the trend can often be in both directions at the same time.
“This makes it much more challenging to distinguish the real changes from random fluctuations,” he says. “For instance, the temperature of an area may go up and down and go back up again in ways that seem like a pattern but really aren’t.”
Zhou is currently testing his algorithm by comparing its findings in places where environmental changes have been well-tracked for years, such as the Amazon rainforest and farms in the Arabian desert; so far, it has predicted the changes that have occurred.
He is also adapting the algorithm to use shifts in population and urban traffic to capture a wider range of underlying changes related to urban sustainability.
Zhou will use the NSF grant to streamline the algorithm, reducing the time it takes to process the data and making conclusions available to planners more quickly.