Case study – Cane toads (Bufo marinus)
The distribution of cane toads within NSW appears highly influenced by climatic variables. A classification and regression tree incorporating the BIOCLIM variables minimum temperature in the coolest month, annual precipitation, precipitation in the warmest calendar quarter and precipitation in the wettest month has a sensitivity of 83% and specificity of nearly 100%.
Figure 1 provides an illustration of the recorded density for cane toads and the predicted occurrence from a model (right map). Recorded density is shaded from light to dark, representing low, moderate to high densities. A high density of toads has been recorded in the NE corner of NSW. While the model predicts occurrence in the top corner it does not have the discriminatory capability to pick up the disjunct populations at Yamba.
Recorded cane toad density as reported by land managers (left) and predicted toad occurrence from a model (right). Recorded density is shaded from light to dark reflecting low, moderate and high density. The model corresponds with recorded presence across the Tweed and Brunswick, but not the smaller outlying populations around Yamba. The coarse scale of the pest data is apparent from the ‘blocky’ nature of the display.
More detailed results will be published in peer-reviewed scientific papers.
Page last updated: 26 February 2011