Using heterogeneity indices to adjust basal area – Leaf area index relationship in managed coniferous stands

The structure of contemporary managed forests is complex and deviates from experimental forests which are usually even-aged monocultures and single-storied. To apply theoretical growth and yield functions on managed forests, adjustments are required, especially for leaf area index (LAI) which is a key biophysical variable in process-based growth models. To asses this, the performance of canopy LAI in modelling the basal area (BA) of managed boreal forests dominated by Norway spruce (Picea abies (L). Karst) and Scots pine (Pinus sylvestris L.) was investigated by heterogeneity analysis. The study was based on the assumption that canopy LAI and BA are strongly related and are vital for estimating stand productivity and growth. Managed forests were represented by field data from the 2016 and 2017 Swedish National Forest Inventory (NFI) campaigns. Species-specific LAI conversion parameters were applied on the general plant area index (PAI) values from hemispheric fish-eye photos taken from the permanent and temporary NFI sample plots. The heterogeneity analysis was studied in two parts by: (a) ground-based stand structural heterogeneity (SSH) described by species composition, coefficient of tree diameter variation, tree social status and height-diameter ratio, and (b) spectral heterogeneity (SPH) by vegetation and textural indices developed from Sentinel-2. Species-specific final (with heterogeneity metrics) and base (without heterogeneity metrics) models were fitted for BA-LAI and BA-PAI relationships by nonlinear least squares and generalised additive regression functions, respectively. The performance of models was assessed by the root-mean-squared error (RMSE, m² ha⁻¹) and the relative root-mean-squared error (RelRMSE, %) metrics. For both species, BA-LAI final models (FMs) accounting for heterogeneity resulted in larger explained variance than the base models (BMs). Compared with the BMs, FMs with SSH reduced the variance by 55% in Norway spruce (RMSE = 3.33, RelRMSE = 15.39) and 43% in Scots pine (RMSE = 3.70, RelRMSE = 17.38). The fit between BA-LAI with SPH also showed an improvement for Norway spruce (RMSE = 5.56) and Scots pine (RMSE = 5.66) over the BMs, suggesting the potential use of Sentinel-2 in future growth models. The results of the study suggest that in growth models when extrapolating theoretical growth functions to managed forests, there is a need to calibrate the models with the forest structural heterogeneity. This is important for drawing realistic conclusions from growth and yield modelling of managed stands of Norway spruce and Scots pine.