“…We propose the hypothesis that the presence of forests in large basins is determinant for the occurrence of a given partitioning pattern, mainly because forests have a strong potential to modify the 10 components of the surface water balance. There is a variety of mechanisms through which forests can exert strong effects on the components of long-term water balance in river basins These factors include, but are not limited to: accumulation and redistribution of soil moisture by root systems (Nadezhdina et al, 2010;Nepstad et al, 1994;Lee et al, 2005;Bond et al, 2002), strong capacity for stomatal regulation related to the large cumulative surface area of leaves (Berry et al, 2010;Costa and Foley, 1997;Katul et al, 2012), physiological adaptations for water and light use efficiency (Nadezhdina et al, 2010), 15 landscape-scale energy balance effects and overall dynamics of E (Villegas et al, 2014), land-atmosphere interactions that enhance the capacity of river basins to store water as a natural "reservoir" (Salazar et al, 2017), activation of shallow convection through transpiration (Wright et al, 2017), below-canopy stability that restricts direct soil evaporation (Henao et al, Submitted), and soil moisture control via canopy effects on hydrological partitioning (Fleischbein et al, 2005). Collectively, these mechanisms define biophysical relations between the presence of forests and the dynamics of P , E and R, and imply a 20 strong potential of forests for modifying the long-term patterns of water balance partitioning.…”