When Zack Parisa decided to become a forester at age 13, one incentive was avoiding a desk job like the one held by his father, a NASA rocket scientist. Back then, Parisa spent all his free hours exploring the Alabama woods. He knew where to find snakes and where the deer slept. He envisioned a career that would let him stay outdoors forever.
These days, Parisa doesn’t stop programming until he gets a nudge from his plott hound, Zoey. Over the past five years, the 29-year-old Yale School of Forestry & Environmental Studies graduate has developed a new method for taking forest inventories. Boston-based SilviaTerra, the business he co-founded in 2010 with fellow Yale grad Max Uhlenhuth, uses Parisa’s method to help foresters tally the size, number, and species of trees they oversee. The detailed inventories help foresters maximize timber yields or locate ideal mushroom habitat.
Traditionally, foresters inventory land by hiking through representative plots and recording each tree. They extrapolate from those samples to get a general sense of the land. The bigger and more diverse the forest, the more plots that must be sampled. Parisa grew frustrated with this method while doing grad school research on biodiversity and development needs in Armenia. “It was obvious that I wasn’t going to get enough information from ground sampling alone,” he says.
Back at Yale, he began crunching numbers. For 18 months he studied satellite data, including spectral images, which reveal overhead views of forests, and radar images, which penetrate the canopy. He correlated that information with inventories he’d gathered the old-fashioned way. The result is an algorithm that helps foresters take stock with far less manpower. Parisa feeds satellite data from clients’ land into his program, which spits out GPS coordinates for the plots that best represent the forest. He then collects on-the-ground data for those few, choice areas. By comparing the satellite and ground data, Parisa calibrates his program to inventory the entire forest.
“It’s an amazingly advanced process,” says Tommy Tadlock, a manager at timber company Plum Creek (PCL), which has been evaluating Parisa’s software. “It cuts down on the amount of fieldwork we have to do, which is the most expensive and time-consuming part of inventory. … And in places where there’s more variablity, we’ll be able to get better data.”
SilviaTerra’s fees depend on the size, location, and complexity of each client’s land. The startup is currently cataloging all of Maine’s forests. While the traditional method would require landowners to sample hundreds of thousands of plots, SilviaTerra will only have to perform on-the-ground sampling for a few thousand.
As for being deskbound, Parisa says he doesn’t mind. “I still believe I could talk with my 13-year-old self, explain what I’m doing, and that I wouldn’t get punched,” he says. “What I do now is just viewing the forest through a different lens.”