Fig. 4

A dendrogram of clustered plots (n = 120) derived from the agglomerative hierarchical cluster analysis (AHCA) for trees shows five distinct clusters (labeled 1 through 5) in response to treatments (control = C, burned four times = B, mechanical removal two times of stems <10 cm dbh = M, and combination of M and B = MB). Data from 2016 were used in this study (2001 to 2016) of plant community change in response to fuel treatments in Green River Game Land, North Carolina, USA. A Bray-Curtis approach was used as the distance metric and a flexible beta linkage was used as the fusion strategy to determine the appropriate number of clusters. The cluster number was determined from fusion height, a visualization method that shows natural breaks in the data, indicating the highest number of plot similarities. Each line on the dendrogram is denoted by orange (burned four times; B), black (control; C), purple (mechanical treatment two times; M), and blue (mechanical treatment two times plus burned four times; MB) dots that indicate which treatment was applied to that plot. Cluster 1 had 11 plots in B, six in C, two in M, and 15 in MB, categorizing it as having a different response from C (category 2; Cat. 2). Cluster 2 had 13 plots in B, 15 in C, 16 in M, and 14 in MB, categorizing it as having a similar response to C (category 1; Cat. 1). Cluster 3 had six plots in B, one in C, one in M, and 0 in MB, falling under category 2. Cluster 4 had zero plots in B, seven in C, ten in M, and zero in MB, falling under category 2. Cluster 5 had zero plots in B, one in C, one in M, and one in MB, falling under category 1. The horizontal axis at the bottom of the dendrogram represents the distance or dissimilarity between clusters