Hydraulic architecture explains species moisture dependency but not mortality rates across a tropical rainfall gradient

Intensified droughts are affecting tropical forests across the globe. However, the underlying mechanisms of tree drought response and mortality are poorly understood. Hydraulic traits and especially hydraulic safety margins (HSMs), that is, the extent to which plants buffer themselves from thresholds of water stress, provide insights into species-specific drought vulnerability. We investigated hydraulic traits during an intense drought triggered by the 2015–2016 El Niño on 27 canopy tree species across three tropical forest sites with differing precipitation. We capitalized on the drought event as a time when plant water status might approach or exceed thresholds of water stress. We investigated the degree to which these traits varied across the rainfall gradient, as well as relationships among hydraulic traits and species-specific optimal moisture and mortality rates. There were no differences among sites for any measured trait. There was strong coordination among traits, with a network analysis revealing two major groups of coordinated traits. In one group, there were water potentials, turgor loss point, sapwood capacitance and density, HSMs, and mortality rate. In the second group, there was leaf mass per area, leaf dry matter content, hydraulic architecture (leaf area to sapwood area ratio), and species-specific optimal moisture. These results demonstrated that while species with greater safety from turgor loss had lower mortality rates, hydraulic architecture was the only trait that explained species’ moisture dependency. Species with a greater leaf area to sapwood area ratio were associated with drier sites and reduced their transpirational demand during the dry season via deciduousness.

Alexandria L. Pivovaroff, Brett T. Wolfe, Nate McDowell, Bradley Christoffersen, Stuart Davies, L. Turin Dickman, Charlotte Grossiord, Riley T. Leff, Alistair Rogers, Shawn P. Serbin, S. Joseph Wright, Jin Wu, Chonggang Xu, & Jeffrey Q. Chambers