1. Introduction
Global forests are changing rapidly to produce fiber, food and other products (Hansen et al., 2013). Despite their central relevance to global sustainability goals and ecosystem services, including water provision and streamflow regulation, natural forests are being lost in many regions of the world including South America (Creed & van Noordwijk, 2018; Creed et al., 2019; Jones et al., 2017; Paquette & Messier, 2010). Temperate rainforests of south central Chile are a global biodiversity hotspot containing highly threatened endemic species (Olson & Dinerstein, 1998). Since 1974, much of the area of native forests in central Chile has been converted to fast growing plantations of exotic Pinus radiata and Eucalyptus species, or to shrublands, agriculture and pastureland (Aguayo, Pauchard, Azócar, & Parra, 2009; Echeverría et al., 2006; Miranda, Altamirano, Cayuela, Lara, & González, 2016). These changes have been associated with declining annual and summer runoff in small and large catchments (Lara et al., 2009; Little, Lara, McPhee, & Urrutia, 2009; Iroumé & Palacios, 2013) as well as reduced plant diversity (Altamirano, Echeverría, & Lara, 2007) and carbon storage (Hall, van Holt, Daniels, Balthazar, & Lambin, 2012; Heilmayr, Echeverría, & Lambin, 2020).
Ecological restoration aims to increase biodiversity and increase the provision of diverse ecosystem services (Benayas, Newton, Diaz, & Bullock, 2009). Ecological restoration is defined as the process of assisting the recovery of an ecosystem that has been degraded, damaged or destroyed (Gann et al., 2019). Restoration efforts attempt to increase native forest cover, species diversity, ecosystem functionality, and ecosystem services such as water provision and regulation (Little & Lara 2010; Clewell & Aronson, 2013). Functional ecological restoration includes efforts specifically targeted at restoring critical structural ecosystem features such as native vegetation, and ecological processes such as nutrient dynamics (Palmer, Hondula, & Koch, 2014).
Ecological restoration is distinct from afforestation, but these two contrasting concepts are sometimes confused in the hydrological literature. Afforestation with fast-growing species has been linked to adverse hydrologic effects such as reduced streamflow (Farley, Jobbágy, & Jackson, 2005; Filoso, Bezerra, Weiss, & Palmer, 2017) and groundwater (Lu, Zhao, Shi, & Cao, 2018). In contrast, the restoration of native vegetation is intended to restore forest hydrology. For example, native forest restoration was associated with increased soil moisture and water table levels and increased forest biodiversity, including plants, fungi, and lichens (Mazziotta et al., 2016).
This study addresses a gap in knowledge about hydrologic response to forest restoration. We describe a 14-year (2006 to 2019) catchment study to determine how streamflow responded to the first nine years of a 130 to 180-year forest restoration experiment (Lara, Little, González, & Lobos, 2013). After five years of pre-treatment streamflow measurements,Eucalyptus globulus plantations were clear-cut in 2011. Restoration was initiated by this clearcut, followed by planting seedlings of native Nothofagus dombeyi tree species and fostering regeneration of other native species. The study is located in the Valdivian Coastal Reserve in south central Chile, a Nature Conservancy (TNC) reserve on former private industrial forest land. This study is one of the longest catchment experiments in South America, and to our knowledge, the only one involving a long-term effort to restore native forests on land formerly in short-rotation intensively managed forest plantations.
This study addressed the following questions:
  1. How did streamflow respond to restoration of native forest in formerEucalyptus plantations?
  2. How has native vegetation changed in catchments under restoration?
  3. What factors explain variation in streamflow response?