Patterns and determinants of tree mortality under assisted natural regeneration in Kibale National Park, Uganda

Abstract

Tree mortality is significantly increasing in tropical forests which could jeopardize the efforts to restoration for sustainable forest management. However, there is limited knowledge of dynamics of tree mortality in restored tropical forests, particularly forests under assisted natural regeneration. This study examined the intrinsic drivers of tree mortality under assisted natural regeneration in Kibale National Park. Specifically, the study assessed successional dynamics of mortality rate and the influence of species characteristics and stand characteristics on tree mortality. Trees were measured for diameter, height and crown width in 63 plots (2000 m2 each) in a regenerated forest (16–22-year old) and 5 plots in an old‐growth forest in 2011, 2014, 2017 and 2020, and their status (dead or alive) was also recorded over the same period. Species characteristics (successional guild, wood density, specific leaf area) were obtained from literature. Data were analyzed using regression models in R. The study revealed that tree mortality increases with the age of the regenerating forest. The assisted naturally regenerated forest attained the same rate of mortality as old growth and it significantly increased with forest age. On average, Mean diameter, height, crown width, wood density of dead trees among different species ranged from 5.2-80 cm, 2.40-19.06 m, 2.48-12.36 m, 0.23 -0.98 g/cm3, respectively. Mortality increased with diameter and wood density but declined with height, while crown width, specific leaf area and successional guild did not affect mortality. Forest dependent species showed lower mortality rates compared to open habitat species and forest nondependent species. Absolute mortality and exponential mortality declined with tree species richness and increased with basal area while exponential mortality and height heterogeneity were not significantly influenced by basal area in the assisted natural regenerated forest. This study has demonstrated that forest age and some, but not all, species and stand characteristics can influence tree mortality under assisted natural regeneration. It is important to strategically integrate these factors during assisted natural regeneration interventions in order to optimize the health and resilience of tropical forests. This study monitored an assisted naturally regenerated forest up to 25 years old. Future studies need to monitor tree mortality in older restoration sites to ascertain whether younger and older tropical forests show consistent mortality patterns.