Development and evaluation of black spruce (Picea mariana (Miller) B.S.P.) Diameter increment models across silvicultural treatments in northern Minnesota, USA

The black spruce cover type occupies roughly 10% of Minnesota's 7 million hectares of forestland, and is an important species, both ecologically and economically. A clearcut regeneration harvest is the main silvicultural system in black spruce in this region. The effects of managing black spruce with alternative silvicultural methods in the Lake States remains largely understudied. Here, we examine a silviculture study in lowland black spruce to assess the performance of two diameter growth models fit to this data compared to a widely-used model. Six silvicultural treatments (clearcut strips, clearcut patches, thinning, group selection, single-tree selection, and shelterwood) and a control were treated and measured around 1950, with a follow-up measurement occurring 10 years later. Fixed- and mixed-effects growth-models were adapted from the previous work, and fit to 10,231 observations and compared to a recently released diameter growth model. The mixed-effects model using treatment, compartment, and plot as nested random effects outperformed the fixed-effects model, and outperformed a model proposed for use in the Lake States variant of the Forest Vegetation Simulator that was fit to this data. This modeling approach of localized growth models across a wide-range of diameters (9.1-32.1 cm) more accurately predicted the diameter growth in lowland black spruce than the conventional approach of using separate models for large ( > 12.7 cm) and small (≤12.7 cm) diameter trees.