A soundscape assessment of the oak forests in the National Park “Homilshanski Lisy” (Northeastern Ukraine).

Abstract

We investigated the properties of the sounds recorded on the territory of the National Park “Homilshanski Lisy” (Kharkiv region, Ukraine). Record ings were made at five points (in mature, middle-aged, and young oak forests, overgrown clear-cut and aspen forests). Data collection was carried out using on-site positioning of AudioMoth autonomous recorders, located on trees at a height of 1.5 m. The recording was made from April 11 to July 10, 2020, for 3 h in the morning and evening with a 5-min duration followed by a 10-min pause (24 recordings per day). Six acoustic indices (AIs) were calculated: Acoustic complexity index (ACI), acoustic diversity index (ADI), acoustic evenness index (AEI), bioacoustic index (BI), normalized difference soundscape index (NDSI), and acoustic entropy index (H). For the analysis, we used the Friedman test as well as a nonparametric analysis of the variance of the distance matrix and Tukey’s test. The results of the analysis showed the statistical significance of the influence of forest type, date and time of recording, as well as the effect of their pairwise interactions on all six acoustic indices, both in the morning and evening. For three indices – ACI, BI, and NDSI – the highest average values were noted in a mature oak forest and the lowest was in overgrown clear-cuts. We performed a PCA to reduce the number of variables and obtain insight into the variable relevance. The cumulative percentage of variance, explained by the first three principal components, is 84.5%. The first principal component is associated with H, BI, AEI, and ADI. The second and third principal components are associated with NDSI and ACI. The obtained results correspond to the results of quantitative bird counts carried out earlier in this area.