4.2. Landscape effect on diversity metrics
Diversity metrics were affected differently by landscape composition and configuration in the study area. Opposite trends between diversity metrics were found for various taxa (Sayer et al., 2017; Wong et al., 2020), showing that TD, FD and PD might not respond in the same way along landscape or invasion gradients. Plantation and native forest (montado) cover had a positive effect on TD across each community type, as did the diversity of landscape patches (i.e. SHDI). Nonetheless, agricultural areas (both intensive and extensive) also positively contributed to TD in non-invaded communities, while rice field areas negatively affected TD both in invaded and invaded no alien communities. Indeed, more disturbed areas (extensive and intensive agricultural, urban and rice field areas) had negative effects both on FD and PD. This decrease in FD and PD is usually followed by biotic homogenization of bird communities inhabiting disturbed areas (Liang et al., 2019; Morelli et al., 2016). Therefore, bird communities inhabiting those areas could be more susceptible to future alien species establishment (both for non-invaded and invadedcommunities) as they are likely better at exploiting novel opportunities in disturbed areas compared to native species (the ‘opportunism hypothesis’; Sol et al., 2012). Apart from plantation forest (invaded communities) and landscape diversity (SHDI fornon-invaded communities), FD and PD were not enhanced in less disturbed areas. In the first case, a similar pattern was found in New Zealand where alien forest bird species were associated more with “exotic forest” (Barnagaud et al., 2022). In the latter, higher diversity of the landscape could support the higher niche differentiation of non-invaded communities, since it could be characterized by a more diverse set of resources available.