Predicting mangrove forest dynamics across a soil salinity gradient using an individual-based vegetation model linked with plant hydraulics

Abstract

Abstract. In mangrove forests, soil salinity is one of the most significant environmental factors determining forest distribution and productivity as it limits plant water uptake and carbon gain. However, salinity control on mangrove productivity through plant hydraulics has not been investigated by existing mangrove models. Here we present a new individual-based model linked with plant hydraulics to incorporate physiological characterization of mangrove growth under salt stress. Plant hydraulics was associated with mangroves' nutrient uptake and biomass allocation apart from water flux and carbon gain. The developed model was performed for two coexisting species – Rhizophora stylosa and Bruguiera gymnorrhiza – in a subtropical mangrove forest in Japan. The model predicted that the productivity of both species was affected by soil salinity through downregulation of stomatal conductance. Under low-soil-salinity conditions (< 28 ‰), B. gymnorrhiza trees grew faster and suppressed the growth of R. stylosa trees by shading that resulted in a B. gymnorrhiza-dominated forest. As soil salinity increased, the productivity of B. gymnorrhiza was significantly reduced compared to R. stylosa, which led to an increase in biomass of R. stylosa despite the enhanced salt stress (> 30 ‰). These predicted patterns in forest structures across the soil salinity gradient remarkably agreed with field data, highlighting the control of salinity on productivity and tree competition as factors that shape the mangrove forest structures. The model reproducibility of forest structures was also supported by the predicted self-thinning processes, which likewise agreed with field data. Aside from soil salinity, seasonal dynamics in atmospheric variables (solar radiation and temperature) were highlighted as factors that influence mangrove productivity in a subtropical region. This physiological principle-based improved model has the potential to be extended to other mangrove forests in various environmental settings, thus contributing to a better understanding of mangrove dynamics under future global climate change.