Tree basal area growth data for fitting allometric equations to 20 species of NE North America

A tree’s basal area and wood volume scale exponentially with tree diameter in species-specific patterns. Recent observed increases in tree growth suggest these allometric relationships are shifting in response to climate change, rising CO2 levels, and/or changes in forest management. We analyzed 9214 cores from nine conifer and 11 broadleaf species grown in managed mixed-species stands in the upper Midwest to quantify how well diameter (DBH) serves to predict basal area (BA) growth and above-ground wood and carbon (C). These samples include many large trees. We fit mixed models to predict BA growth and above-ground biomass/C from diameter, tree height, and the BA of nearby trees while controlling for site effects. Models account for 55-83% of the variance in (log) recent growth, improving predictions over earlier models. Growth-diameter scaling exponents covary with certain leaf and stem (but not wood) functional traits, reflecting growth strategies. Log BA increment scales linearly with log diameter as trees grow bigger in 16/20 species, and growth accelerates in Quercus rubra L. Three other species plateau in growth. Growth only decelerates in red pine, Pinus resinosa Ait. Growth in whole-tree, above-ground biomass, and C accelerates even more strongly with diameter (mean exponent: 2.08 vs. 1.30 for BA growth). Sustained BA growth and accelerating wood/C growth contradict the common assumption that tree growth declines in bigger trees. Yield tables and silvicultural guidelines should be updated to reflect these current relationships. Such revisions will favor delaying harvests in many managed stands to increase wood production and enhance ecosystem values, including C fixation and storage. Further research may resolve the relative roles of thinning, climatic conditions, nitrogen inputs, and rising CO2 levels on changing patterns of tree growth.

Methods
The data are for 20 species / 9,453 trees reflecting a broad range of tree sizes and growth rates in managed public forest (BCPL) stands in N Wisconsin. These stands were managed using thinning to release faster-growing trees. We analyzed growth in 9 conifer and 11 deciduous species. Between 2005 and 2018, the BCPL’s continuous forest inventory team extracted increment cores from the first and third trees sampled within each forest inventory plot (generally >10cm DBH). This ensured sampling trees in proportion to their abundance and avoided sampling adjacent trees whose growth might be correlated. Teams measured DBH to the nearest 2.6mm using a tape and estimated tree height to the nearest foot (31cm) using a clinometer. Tree cores were drilled to a depth of 5-6 cm, reflecting growth from roughly 1985 to 2018. A dissecting microscope (10x) allowed us to count the number of rings spanning the last inch (2.54 cm) of growth (excluding the bark). We estimate DBH at the beginning of the growth period as the final DBH (measured when the core was collected) minus 5.08cm. We estimate basal area increases from differences in cross-sectional area, assuming successive tree disks to be circles of area π r2. Data include plot and stand ID, geographic location, estimated total plot basal area (BA/acre or ha), species, initial DBH, height, crown class, and the number of growth rings in the outer 2.56 cm of wood. We first modeled growth (mean annual basal area increment in the outermost annulus) across all trees as a function of species, diameter (DBH at the start of the growth period), height, crown class, and competition (plot basal area minus focal tree BA). We then fitted species-specific models that included site (Stand ID) as a random factor to adjust the models for different local growing conditions.