LAUSR

Abstract A wide range of different biodiversity indicators have been identified in different biomes and types of environments. However, we currently lack a clear understanding under which circumstances proposed biodiversity indicators are valid indicators and when they are not. Here we conducted a large-scale restoration experiment by emulating natural disturbances through prescribed burning and gap-cutting. From this experiment data of resident forest birds was used to, (1) identify biodiversity indicators in the middle and northern boreal zone, (2) test the effect of forest restoration practices on biodiversity indicators and (3) explore the patterns when biodiversity indicator analyses are performed across a wider range of forest types including both restored and unrestored sites. Additionally, to understand the occurrence of species we correlated their occurrence with local habitat structures across different forest types. We found that both the three-toed woodpecker (Picoides tridactylus) and Siberian jay (Perisoreus infaustus) were important biodiversity indicators for the middle/northern boreal zone. The occurrence of these species can generally be used to identify species rich areas across different forest types. However, the three-toed woodpecker, the best indicator for unrestored forest, lost its indicator value after prescribed burning; in these stands the goldcrest (Regulus regulus) was the best biodiversity indicator. Based on habitat associations, we suggest that biodiversity indicators in boreal forest not only indicate high species richness but also structural habitat complexity. Our results draw attention to the fact that one must be careful about using previously identified biodiversity indicators in ecosystems subjected to ecological restoration.