Quantifying forest understory biomass is important for understanding ecological processes, but there are few methods for non-destructive measurement of understory biomass in southeast Alaska. We developed cover-to-biomass equations for common understory species in young-growth Sitka spruce (Picea sitchensis)–western hemlock (Tsuga heterophylla) forests. A sampling method of visually estimating cover and destructively measuring biomass was used at 35 stands aged 10 to 67 years on Prince of Wales Island in southeast Alaska from 2007 to 2018. Linear cover to biomass regressions were fitted for 42 species and other genera. In addition to total biomass, regressions were fitted by part (leaves, twigs, wood) for all woody species. Regressions were also fitted for graminoid, fern, forb, shrub, tree, and conifer functional classes. We demonstrate the utility of these regressions by applying them to the Tongass-wide young-growth studies, a rich dataset with understory cover measurements from treated and un-treated stands in four young-growth age classes. Understory biomass was greater in 0- to 5-year-old even-aged stands than stands greater than 15 years old. Treated stands (thinned, etc.) had a greater understory biomass, annual growth, and carbon than untreated stands older than 15 years. Additionally, biomass composition became less woody with increasing stand age in treated stands. These regressions provide an approach to estimate understory biomass, which can be used for evaluation of forest functions, including understory dynamics, wildlife habitats, and total stand carbon.