School Notes: School of the Environment
May/June 2019

Report finds urban natural forests require greater management

In cities across the US, natural forest areas are underutilized, and municipalities lack the resources for proper management, according to a landmark report produced by researchers at F&ES and two partner institutions. Yale researchers worked with the New York–based Natural Areas Conservancy (NAC) and the Trust for Public Land to survey 125 organizations, representing 40 states and 111 metropolitan areas with populations of 50,000 or higher. They found that urban natural areas account for 1.7 million acres of land—larger than the state of Delaware—and require “interventions and substantial investments in management to ensure their long-term health.” Urban natural areas are essential to an ecosystem’s resiliency and the health of the local public and are a proven combatant of climate change, regulating temperature and flooding. “We found that many organizations share similar motivations and challenges but don’t have basic types of data to describe who is using the forest, the forest condition, and how it is changing over time,” said coauthor Clara Pregitzer ’18MPhil, a PhD candidate at F&ES and a conservation scientist at NAC. Other coauthors included Mark Bradford, a professor of soils and ecosystem ecology at F&ES, and NAC cofounders Sarah Charlop-Powers ’09MEM and Bram Gunther ’91MEM.

Yale-launched business maps US forests

While he was a student at F&ES, Zach Parisa ’09MFS developed a technology that uses satellite data to determine the size and species of trees in forests. That innovation would eventually become SilviaTerra—cofounded by Parisa and Max Nova, a 2012 graduate of Yale College—which helps improve forest management and improves the return on investment for some of the nation’s largest timber companies. Now, with a grant from Microsoft’s “AI for Good” initiative, the company has created the first high-resolution inventory for all US forests, a tool that will help conservationists, government organizations, and landowners make better forest management decisions. “Forests are made up of complex interconnected systems, and there’s never been a way to measure all of these systems,” says Parisa. “And you can’t manage what you can’t measure.” For this project, SilviaTerra is using Microsoft Azure, high-resolution satellite imagery, and US Forest Service inventory and analysis field data to train machine-learning models to measure forests.