The simplification of forest structural complexity, caused by anthropogenic land-use practices, is one of the main threats to understory specialist birds. We examined the association of both single structural attributes and structural complexity, with the density of four understory bird species in the Global Biodiversity Hotspot “Chilean Winter Rainfall-Valdivian Forests” of South America. Between 2011 and 2013, we surveyed habitat attributes and conducted bird point counts in 505 plots in Andean temperate ecosystems in Chile. In each habitat plot, we measured understory density, volume of coarse woody debris (CWD), number of snags, diameter at breast height (DBH) of trees, and leaf litter depth. With these attributes, we developed an index of stand structural complexity (ISC). On average, old-growth forests had higher values ​​for understory density, CWD volume, DBH, and litter depth than secondary forests and open fields, and thus greater values of ISC. The density of understory birds was positively correlated with the ISC for the Rhinocryptidae Pteroptochos tarnii, Scelorchilus rubecula, and Scytalopus magellanicus. We also found a positive association between understory density and litter depth, with the density of the Furnariidae Sylviorthorhynchus desmursii. However, this latter species showed a negative association with the density of snags. Our results suggest the utility of using an index of structural complexity, rather than single or even additive habitat attributes, for determining the density of understory specialist birds. We recommend that management plans should promote the retention of habitat attributes that contribute to the structural complexity of temperate forests of South America and beyond.

We conducted bird sampling during 2 reproductive seasons (2011-2012 and 2012-2013), between October and January. Point count stations were in the same plots used for habitat sampling, with the center of the plot at the center of the point count station. We sampled 81 sites (80.2%) in 2011-2012 and the remaining 20 (19.8%) in 2012-2013 (Ibarra and Martin 2015a). In each of the 101 sites, we established 5 point counts that were systematically spaced 125 m apart and lasted 6 minutes during which we recorded each individual seen and/or heard within a radius of 50 m (Bibby 2000, Ibarra et al. 2017a).[A1] [A2] 

To increase the detectability of understory specialist birds we performed playbacks. To avoid double counting at nearby stations. we reproduced playback at 3 out of the 5 point count stations. After the period of passive counting (the first 6 minutes), we reproduced the playbacks of the 4 understory specialist species. As each call by 30 seconds was followed by 30 seconds of listening in silence, the counting time for each station with playbacks totaled 15 minutes (Drever et al. 2009, Ibarra et al. 2017a). The distance to all the birds detected was estimated and grouped into one of two intervals for analysis (0-25 m and 26-50 m; Bibby 2000, Ibarra and Martin 2015a). We also registered temperature (°C), humidity (%), and wind speed (m per s) at each point count station (Table 1) using a portable weather monitor (Kestrel 4200, Kestrel-meters, Birmingham, MI), to assess the influence of these covariates on the detectability of each species.