Mon, Aug 02, 2021:On Demand
Background/Question/Methods
Insect outbreaks in low-diversity tropical forests are seldom reported but can offer useful perspectives about outbreaks in other systems. We investigated the outbreak of the endemic koa moth (Scotorythra paludicola), which defoliated endemic koa trees (Acacia koa) over a third of their range on Hawai‘i Island during 2013–2014. At Hakalau Forest NWR, we observed outbreak dynamics and its impacts on host trees, nutrient cycling, and insectivorous consumers in reforestation stands of densely planted koa and natural stands of mixed koa and ‘ōhi‘a (Metrosideros polymorpha), the most widespread and abundant native species. This allowed us to evaluate resource concentration hypothesis (RCH) predictions about herbivore behavior in response to variation in host tree dominance and density. We tracked caterpillar abundance and biomass from branch clipping; determined rates of development, frass production, and parasitism in the lab; and estimated moth abundance with malaise trapping. Marked koa were assessed for defoliation severity, refoliation rate, and survival. Nitrogen and phosphorous redistribution from frass and litterfall were assayed in soil samples and leaves of understory plants. We analyzed bird diet change using fecal samples, and measured changes in bird mass and habitat use in the field. Bat foraging behavior was assessed with acoustic recorders.
Results/Conclusions Contrary to RCH, caterpillar biomass and defoliation severity were high in mixed koa-‘ōhi‘a stands. Mortality was high for heavily defoliated small koa. Refoliation occurred quickly for lightly defoliated small koa, but the opposite occurred for large trees. Small koa initially refoliated with true leaves, the foliage type generally avoided by caterpillars compared to phyllodes. Nitrogen from frass was redistributed to the soil but was taken up by alien grasses instead of native shrubs. ‘Ōhi‘a flower abundance and koa defoliation severity affected bird habitat use, diets switched to outbreaking caterpillar prey, and mass increased during the outbreak. Bat foraging times decreased due to rapid satiation each night. Parasitoid wasps increased with caterpillar abundance but had little influence on outbreak dynamics. Defoliated stands eventually recovered, suggesting resilience in both planted and natural forests, but tree density and diversity affected outbreak dynamics in sometimes unexpected ways. Unique among outbreaks worldwide, and in contrast to the single defoliation-refoliation cycle at Hakalau, were three, rapid cycles of defoliation-refoliation observed at warmer elevations below Hakalau, potentially due to faster rates of refoliation and caterpillar development and sparser grass cover. Our results provide new insights about outbreaks generally and in low-diversity tropical forests.
Results/Conclusions Contrary to RCH, caterpillar biomass and defoliation severity were high in mixed koa-‘ōhi‘a stands. Mortality was high for heavily defoliated small koa. Refoliation occurred quickly for lightly defoliated small koa, but the opposite occurred for large trees. Small koa initially refoliated with true leaves, the foliage type generally avoided by caterpillars compared to phyllodes. Nitrogen from frass was redistributed to the soil but was taken up by alien grasses instead of native shrubs. ‘Ōhi‘a flower abundance and koa defoliation severity affected bird habitat use, diets switched to outbreaking caterpillar prey, and mass increased during the outbreak. Bat foraging times decreased due to rapid satiation each night. Parasitoid wasps increased with caterpillar abundance but had little influence on outbreak dynamics. Defoliated stands eventually recovered, suggesting resilience in both planted and natural forests, but tree density and diversity affected outbreak dynamics in sometimes unexpected ways. Unique among outbreaks worldwide, and in contrast to the single defoliation-refoliation cycle at Hakalau, were three, rapid cycles of defoliation-refoliation observed at warmer elevations below Hakalau, potentially due to faster rates of refoliation and caterpillar development and sparser grass cover. Our results provide new insights about outbreaks generally and in low-diversity tropical forests.