Fig 1.
Pairwise relationships of the three sampling methods of card counts, pitfall trapping and tuna lures with raw sample-level data (a-c), transect-level data compiled from first pooling sample-level data then applying a log(x+1) transformation (d-f) and transect-level data compiled by pooling sample-level data that had first undergone a log(x+1) transformation (g-i).
Fig 2.
3D scatterplots of the relationship between A. gracilipes abundance determined by the three sampling methods of card counts, pitfall trapping and tuna lures using transect-level data compiled from first pooling sample-level data then applying a log(x+1) transformation (a) and transect-level data compiled by pooling sample-level data that had first undergone a log(x+1) transformation (b).
Fig 3.
Standardised highest yellow crazy ant abundance reported within fourteen studies as calculated by the relationships between the three sampling methods of card counts, pitfall trapping and tuna lures using a) summed log(x+1)-transformed sample data, and b) log(x+1)-transformed pooled data. Studies are: 1) [30]; 2) [31] New Caledonia site; 3) [31] Australia site; 4) [21]; 5) [32]; 6) this study; 7) [33]; 8) [34] 9) [17]; 10) [35]; 11) [36]; 12) [37]; 13) [38]; and 14) [3].
Fig 4.
3D scatterplot of standardised highest yellow crazy ant abundance among fourteen studies as calculated by the relationships between the three sampling methods of card counts, pitfall trapping and tuna lures using sample-based log(x+1)-transformed pooled data.
Studies are: 1) [30]; 2) [31] New Caledonia site; 3) [31] Australia site; 4) [21]; 5) [32]; 6) this study; 7) [33]; 8) [34] 9) [17]; 10) [35]; 11) [36]; 12) [37]; 13) [38]; and 14) [3].