For practical use in model-based growth projections, off-plot
neighbour trees would be independently reconstructed in every
growth simulation run. The competition indices can then be estimated
without bias leading to better growth projections.
Once again the study also highlighted the close relationships of
forest inventory and spatial statistics that have become already
apparent in other work (Pommerening and Stoyan, 2008; Motz
et al., 2010; Nothdurft et al., 2010).
5. Conclusions
In this study, we explored the use of the reconstruction method
for edge-bias mitigation. Stand structure reconstruction has turned
out to be the best and most versatile tool for this purpose and has
great potential in applications of forest monitoring. Our analysis
indicated a statistically good reconstruction of nearest off-plot
neighbours based on a modified simulated annealing algorithm.
This substantially improved the estimations of Hegyi’s competition
index and provided more realistic point patterns compared to
those produced by the reflection method.