Stats4Good: Data for Earth Day 2020

With a PhD in statistical astrophysics, David Corliss leads a data science team at Fiat Chrysler. He serves on the steering committee for the Conference on Statistical Practice and is the founder of Peace-Work, a volunteer cooperative of statisticians and data scientists providing analytic support for charitable groups and applying statistical methods in issue-driven advocacy.

With Earth Day on April 22, this month’s Stats4Good column features the work of Data for Good scientist-activists to support and strengthen environmental action. The opportunities are endless, and D4G can have a huge impact for the benefit of the world around us.

For people just starting out in Data for the Environment, one way to get involved is by working with organizations you are supporting now. Maybe it’s a nature center looking for teachers and other volunteers with an analytic background. Maybe it’s a birding group needing analysis of bird counts to study trends and identify local risk factors. Maybe it’s a community group needing practical, effective ways in which people can reduce their carbon footprint. Working where we are already connected and participating in activities is a great way to introduce the power of statistics for environmental action.

While we cherish and are inspired by global leaders, the truth remains that most actions for preserving, protecting, and sustaining the environment take place in a local setting. The aphorism “Think Globally, Act Locally” is nowhere truer than in Data for Good. Whether supporting sustainable use of energy and other resources; wildlife conservation; human and habitat health; or air, water, and land pollution, the application of statistics and data science to support local projects often makes the biggest difference in turning broad goals into tangible results. One of the most basic questions we can ask is how local organizations participating in good environmental projects can become even better through the application of statistical science. Partnering with local groups—often the same organizations where we are working now—is one of the best ways to get plugged in to a D4G project or start a new one.

Other promising opportunities can be found on the job. Corporations in a variety of industries are increasingly aware of the central role advanced analytics plays in addressing environmental issues. For example, a survey of business leaders undertaken by Intel found 74 percent saw a role for artificial intelligence in addressing these challenges. This trend is especially strong in tech industries: Corporations already using statistical science to guide their business and develop new products—or as their products—will see the value these same capabilities bring to acting as good corporate citizens for the benefit of the community.

The number of hackathons in Data for Good has exploded, and they have become one of the most important avenues for environmental Data for Good. Hackathons offer many advantages. They focus on a single important question; are prepared in advance so participants can focus on developing analytic solutions, networking, and collaborating; and are high in intensity with a specific, short-term time commitment. Local environmental groups and ASA chapters can make perfect partners for hosting a hackathon, so talk with members of your chapter to brainstorm ideas and find the right organization for a partner.

Another great way to use our analytic skills to make a big impact is through educational programs offered by nature centers affiliated with parks, colleges, and other institutions. Many volunteer teachers like to use statistics with M&M’S to show students how math can be used for good. For example, statistics can be used to describe benthic testing of macro-invertebrates—basically bugs that live in stream beds. Different genera have varying affinity for water purity, with some very sensitive to pollution. A few families actually prefer dirty water, and some are found in all water conditions. The equivalent mammal types are chipmunks for very clean environments, rats in adverse locations, and squirrels in almost all conditions.

Benthic testing is a popular and effective means for environmental resource centers to partner with the general public to monitor water quality. For a Statistics for the Environment session, each piece of candy represents one genus found in a water sample—the corresponding bugs only appear in pictures. Blue pieces represent genera requiring very clean water, brown for dirty, green for those associated with algal blooms, and the other colors for genera without a water quality preference. Students can assess the quality of the water in their “sample” and match it to a map of the local watershed marked with colors at places in which the water was tested. Analysis of the patterns shows how pollution travels downstream and identifies locations in which the water quality deteriorates quickly. Classes like this are a fun way to introduce statistical science as a powerful tool for protecting the environment.

These are just a few of the many ways people are getting involved. The list of Data for Good projects to preserve and protect our planet is as colorful and diverse as the Earth itself. Each of us can have a role to play.

Each year in April, Earth Day brings to mind pressing environmental concerns and the different ways people can make a difference. Now is the time to start conversations—with our colleagues, in our communities, in the organizations we support, and in the workplace—about the roles Data for Good can play. From collecting, curating, and publishing environmental data to innovative research, greener manufacturing, and supporting data-driven decision-making and legislation, thousands of statisticians, data scientists, and other researchers are working to make the world a better, cleaner, healthier place. The dedication of countless people in this labor of love is reaching beyond the horizon to create a brighter future for all. In this, as in all our work, the Data for Good movement is changing the way people change with world.