Root and stem partitioning of Pinus taeda

We measured root and stem mass at three sites (Piedmont (P), Coastal Plain (C), and Sandhills (S)) in the southeastern United States. Stand density, soil texture and drainage, genetic makeup and environmental conditions varied with site while differences in tree size at each site were induced with fertilizer additions. Across sites, root mass was about one half of stem mass when estimated on a per hectare basis. Stem mass per hectare explained 91% of the variation in root mass per hectare, while mean tree diameter at breast height (D), site, and site by measurement year were significant variables explaining an additional 6% of the variation in root mass per hectare. At the S site, the root:stem ratio decreased from 0.7 to 0.5 when mean tree D increased from 10 to 22 cm. At the P and C sites, where mean root:stem ratios were 0.40 and 0.47, respectively, no significant slope in the root:stem to mean tree D relationship was found over a more narrow range in mean tree D (12–15 and 12–18 cm, respectively). Roots were observed in the deepest layers measured (190, 190, and 290 cm for the P, C, and S sites, respectively); however, the asymptotically decreasing root mass per layer indicated the bulk of roots were measured. Root growth relative to stem growth would need to change with increased mean tree D to explain the results observed here. While these changes in growth rate among plant components may differ across sites, stem mass alone does a good job of estimating root mass across sites.