Winners and losers in dryland reforestation: species survival, growth, and recruitment along a 33-year planting chronosequence

Reforestation in the Lower Rio Grande Valley (LRGV) of Texas began in the 1960s and to date over 6,475 ha of land has been reforested. However, there has been minimal assessment to understand differential species success, compositional trends, and the aboveground carbon sequestration potential of these developing forests. We coupled quantitative planting information of \u3e50 native woody tree and shrub species with surveys of 5,223 stems of 4,606 individuals in a chronosequence of restored forests ranging in age from 1 to 33 years to estimate species-specific mortality rates, biomass accumulation and recruitment, as well as compositional trends in the herbaceous understory. We show that 7–15 years are required for mortality rates of the transplanted cohort to stabilize to background levels observed in other dry forests. A small number of species, mostly N-fixing trees with a deep rooting habit, persisted on the landscape beyond 15 years. Even so, aboveground biomass (corrected for differences in initial planting density) accumulated at an average rate of 1.41 Mg ha−1 yr−1 compared to 0.35 Mg ha−1 yr−1 for a fallow old-field. Species biomass growth rates increased with decreasing mortality, as did the abundance of recruits, suggesting a degree of reproduction by initial planted cohorts. However, a suite of highly competitive exotic grasses increases in density over a 25-year period, which we link to suppressed seedling recruitment. This poses a serious challenge to the long-term sustainability of planted forests in the LRGV. We highlight potential avenues of research and modification to restoration methodologies