Note: This story is part of a special issue of The Warnell Log focused on carbon If you want to quantify carbon, what conversion equations or models do you use, and how do you quantify variability? That’s where Warnell associate professor Joe Dahlen comes in. Dahlen is the director of the Wood Quality Consortium, a cooperative effort with the forest industry and the U.S. Forest Service to better understand wood coming from managed loblolly pine stands. The percent carbon within wood is ultimately attributed to its chemistry. As trees convert sapwood to heartwood, the deposition of extractives in the wood cells further increases the percentage of wood that is carbon. Linking wood chemistry with volume, specific gravity or density helps to calculate the weight of the carbon. Dahlen also measures the annual growth rate of the samples is also measured to track carbon sequestered over time. To collect this, Dahlen has had to get creative, developing new and rapid methods for measuring wood and fiber properties. In many cases, off-the-shelf equipment is modified to suit his needs. In other cases, he’s used circuitry knowledge to design, construct and program his own automated imaging systems. He has two labs, one in Warnell’s Building 3 for measurements and the other at the Pete Phillips Wood Utilization Plant Sciences Building at Whitehall Forest for sample preparation. The work results in large datasets; after collecting the data he develops statistical models that predict these properties. Because of the size of the data and its variability, he recently began experimenting with artificial intelligence and deep learning to analyze and better model the data. Ultimately, the work helps the forest industry better understand the amount of carbon being sequestered within the tree. “Loblolly pine is the most important tree species in the U.S.,” he said. “We should have a good understanding of the amount of carbon we’re sequestering and the sources of variation associated with it.”
