Mean ranking of the key predictor variables (selected by minimising the AICc) across 100 bootstrap models for each Absence��Presence cluster combination for Model 2.

Key to predictor variable names: MIR = Middle Infrared; dLST = daytime Land Surface Temperature; nLST = nighttime Land Surface Temperature; Rain = Rainfall; NDVI = Normalised Difference Vegetation Index; EVI = Enhanced Vegetation Index. Key to Fourier variable names: Mean = average (observed and predicted); amp1, amp2, amp3 = amplitudes of the annual, bi-annual and tri-annual cycles of changes in the respective variables; phase1, phase2, phase3 = phases (or timing) of the annual, bi-annual and tri-annual cycles of changes in the respective variables; min. = minimum of Fourier fitted value (may therefore be negative); max. = maximum of Fourier fitted value; variance = variance of the raw data.