A Variance Partitioning Multi-level Model for Forest Inventory Data with a Fixed Plot Design

Abstract Forest inventories are often carried out with a particular design, consisting of multi-level structure observation plots spread over larger domain and fixed plot design exact locations within these plots. Consequently, the resulting data collected intensively equal size but much less intensity at spatial scales. The likely to be spatially correlated both between plots, effects extending two different areas. However, Gaussian process model standard covariance is generally unable capture dependence fine coarse scales variation as well for their interaction. In this paper, we develop computationally feasible that accounts multiple We use data-driven approach determine weight each in partition variability measurements. simulated German small tree inventory evaluate model’s performance.Supplementary material paper provided online.