Skip to main content
Forest Ecosystems
Fig. 4 | Forest Ecosystems

Fig. 4

From: Performance of statistical and machine learning-based methods for predicting biogeographical patterns of fungal productivity in forest ecosystems

Fig. 4

Similarity between the climate conditions of the whole study area and the modeling data. The spatial similarity was based on the number of sampled data with an environment similar to the prediction environment. The per-pixel similarity was obtained by overlapping the pixel position and a density map of the sampled data in a two-dimensional space defined by the first two principal components of a principal component analysis (PCA). The PCA was performed with five and 12 variables according to the machine learning models’ climate predictors (Fig. S3)

Back to article page