Contracting montane cloud forests: a case study of the Andean alder (Alnus acuminata) and associated fungi in the Yungas

RAES, NIELS; AGUIRRE-GUITERREZ, JESÚS; PACHECO, SILVIA; NOUHRA, EDUARDO; GEML, JÓZSEF; PASTOR, NICOLÁS; WICAKSONO, CHRISTIAN Y.; RAES, NIELS; AGUIRRE-GUITERREZ, JESÚS; PACHECO, SILVIA; NOUHRA, EDUARDO; GEML, JÓZSEF; PASTOR, NICOLÁS; WICAKSONO, CHRISTIAN Y.

BIOTROPICA

WILEY-BLACKWELL PUBLISHING, INC

Año: 2017 vol. 49 p. 141 - 152

0006-3606

Alnus acuminata is a keystone tree species in the Yungas forests and host to a wide range of fungal symbionts. While species distribution models (SDMs) are routinely used for plants and animals to study the effects of climate change on montane forest communities, employing SDMs in fungi has been hindered by the lack of data on their geographic distribution. The well-known host specificity and common biogeographic history of A. acuminata and associated ectomycorrhizal (ECM) fungi provide an exceptional opportunity to model the potential habitat for this symbiotic assemblage and to predict possible climate-driven changes in the future. We (1) modeled the present and future distributions of suitable habitats for A. acuminata; (2) characterized fungal communities in different altitudinal zones of the Yungas using DNA metabarcoding of soil and root samples; and (3) selected fungi that were significant indicators of Alnus. Fungal communities were strongly structured according to altitudinal forest types and the presence of Alnus. Fungal indicators of Alnus, particularly ECM and root endophytic fungi, were also detected in Alnus roots. Current and future (year 2050) habitat models developed for A. acuminata predict a 25?50 percent decrease in suitable area and an upslope shift of the suitable habitat by ca. 184?380 m, depending on the climate change scenario. Although A. acuminata is considered to be an effective disperser, recent studies suggest that Andean grasslands are remarkably resistant to forest invasion, and future range contraction for A. acuminata may be even more pronounced than predicted by our models.