Correlation_variables.csv Columns: Transect=Site name Columns 3-10 (C-J) refer to dawn birds. 11-18 (K-R) refer to nocturnal birds and 19-26 (S-Z) refer to all nocturnal taxa identified. Shannon, Simpson, Pielousevenness, fisher_abund = diversity metrics calculated in Vegan total_reg= Number of 'species detections' (sum of the number species detected in each file, not the number of times each species vocalised) mean_SpR= mean species richness per recording file sr_obs=observed species richness MDS1=axis 1 of an NMDS analysis conducted in vegan with Jaccard index. Index_values.rdat ID=unique ID file=original 2 hour filename Rain=presence of rain determined by hardRain Total.Rain=Total number of 15 s units rain detected in dur=Duration of file analysed date/dat/Time=various time formats Period = categorical time of day (dawn, day, dusk, night) Gradient=categorical levels of forest disturbance. "LB"=Primary forest, logged and burnt. "LU"=Primary forest, logged and unburnt. "SU"=Secondary forest, unburnt. "PB"=Primary forest, unlogged and burnt. "PU"=Undisturbed primary forest. var= Acoustic index and frequency measured. AC=Acoustic Complexity Index. BI=Biacoustic Index. 1= 0.3 kHz to no upper frequency limitation 2= 0.3-4 kHz 3= 0.3-12 kHz 4= 4-12 kHz 5= 12 kHz to no upper limitation ind_value=Acoustic Index value Subsampled_data_SOM.csv Data supplied in same wide format as used for training randomForest models Gradient = as above other variable headers are concatenation of the acoustic index (as above), Period (as above) and the frequency limits, with base being 0.3 to no upper limit. Full_Sensitivity_results.csv Acoustic Index = Acoustic Index (as above) Time Period = Period (as above) Frequency bin = frequency limits (as above) Effect size = effect size of the Kruskal-Wallis test No. of significant differences = The number of forest class pairs shown to be significantly different by a Dunn's test.