Effect of canal blocking on biodiversity of degraded peatlands: Insight from West Kalimantan

PLOS One | https://doi.org/10.1371/journal.pone.0334014 October 8, 2025 8 / 23 similarities across these indices, contrasting with estate crop, which demonstrated differences in these indices compared to forested habitats (Fig 3). Additionally, disturbed wet shrub displayed discrepancies from forested habitats across sev- eral indices, including Simpson’s Diversity Index, Pielou’s Evenness Index, and Margalef’s Richness Index. According to the results from the GLM test, only 9 out of 12 habitat variables demonstrated linear relationships with the indices, with habitat type significantly influencing most response variables, indicating its substantial impact on determining index values in each habitat (Fig 4). Additionally, variables such as temperature, humidity, month, and understory density also exerted notable impacts on the indices, with nearly all response variables showing a positive relationship with the predictors, except for the Berger Parker Dominance Index, which exhibited a significant negative effect. Furthermore, a comparison test was conducted for habitat parameters linearly related to the indices to assess differences in environmen- tal conditions across habitats. Six variables exhibited statistically significant differences (p > 0.05), including temperature, humidity, seedling density, understory density, canopy cover, and vegetation species richness, whereas ground cover remained consistent across habitats (p < 0.05), suggesting no significant variation between them.

The consistency pattern between forested and non-forested was also detected within acoustics indices. Although it may not be specific to one habitat type, but for ADI and BI the highest value was noticed in forested areas and lowest in non-forested areas (S1 Fig). As for AEI, highest value was in non-forested areas and lower in forested areas. Meanwhile for ACI, less disturbed peat forest holds the lowest value and highest in disturbed wet shrubs. Within month comparison, we identified change from mean value May to August from ADI and AEI where in all habitat types of ADI increased and AEI decreased (S3 Table). Mean BI value increased from May to August in all habitat types except estate crop. Mean ACI increased in disturbed wet shrubs and disturbed peat forest while decreasing in estate crop and less disturbed peat forest. We used LMM to test differences between habitat types. ACI, ADI, AEI, BI were placed as predictors. Habitat was placed as fixed variables, whereas month, recorder position (start and end of the transect), and transect code were set as random variables. Initial Type II ANOVA results from the car package showed significant p-value for ACI, ADI, AEI, and BI, Fig 3. Post-hoc tests using Tukey’s Honestly Significant Difference (HSD) for wildlife attributes. https://doi.org/10.1371/journal.pone.0334014.g003