Fig. 2

Partial Least Squares Discriminant Analysis (PLS-DA) to classify the objects of study according to the observed or measured values of several variables. The principal coordinates analysis (PCoA) and non-metric multidimensional scaling (NMDS) of all samples using Bray-Curtis distance. a PLS-DA small watersheds; b PLS-DA fruit period; c PCoA small watersheds; d PCoA fruit period; e NMDS small watersheds; f NMDS fruit period. It is generally believed that stress < 0.2 can be represented by 2d NMDS point graph, which has certain explanatory significance. When stress < 0.1, it can be considered a good ranking. When stress < 0.05, it is the most representative