Estimation of the occurrence, severity, and volume of heartwood rot using airborne laser scanning and optical satellite data

Rot in commercial timber reduces the value of wood substantially and estimating occurrence, severity, volume heartwood rot would be a useful tool decision-making to minimize economic losses. Remotely sensed data has recently been used for mapping on single-tree level, although results have relatively poor, some potential shown. This study applied area-based approaches predict , at an area level. Ground reference were collected from harvester operations 2019–2021. Predictor variables calculated multi-temporal remotely together with environmental variables. Response predictor aggregated grid cells forest stands. Random Forest models built different combinations response subsets, validated both random- spatial cross-validation. The showed that it was not possible estimate occurrence severity modeling procedure (pR2: 0.00–0.16), without spatially close training data. better performance 0.12–0.37) mainly due correlation between volume.