Eco-Acoustic Assessment of an Urban Park by Statistical Analysis

We investigated the statistical properties of sounds recorded at a site located in Parco Nord Milan, Italy, characterized by wooded land rich biodiversity and exposed to different sources types anthropogenic disturbances with aim deriving information on its environmental quality terms biophonic presence athropic disturbance. A time series eco-acosutic indices were determined for 616 audio files from 30 April 2019 (5:00 p.m.) 3 May a.m.) 1-min duration followed 5-min pause (10 recordings per hour). In present study, following computed: Acoustic Complexity Index (ACI), Diversity (ADI), Evenness (AEI), Bio-acoustic (BI), Entropy (H), Normalized Difference Soundscape (NSDI) Dynamic Spectral Centroid (DSC). Cluster analysis performed corresponding yielded dimensional reduction seven down three. The results show clear separation eco-acoustic into two clusters, reflecting dynamics diversity behaviour throughout recordings. post-processing aural survey was also performed, aiming determining activities (mainly avian vocalization other animals), characteristics technophonies road traffic noise airplane fly-overs), human (voices steps) geophonies (rain wind). proved be robust tool due good matching obtained outcomes. overall phonic activity found low. Notwithstanding species, highlighted characteristic dawn chorus, both clusters revealed low “scores” NDSI DSC heavily influenced sources. This study represents first step toward realization maps long-term monitoring fragile habitats.