Biodiv Sci ›› 2022, Vol. 30 ›› Issue (11): 22069.

• Original Papers:Plant Diversity •

### Effect of plant species loss on grassland caterpillar in alpine meadows

Yongqingcuomu , Xinqiang Xi, Kechang Niu()

1. School of Life Sciences, Nanjing University, Nanjing 210023
• Received:2022-02-09 Accepted:2022-04-24 Online:2022-11-20 Published:2022-06-23
• Contact: Kechang Niu

Abstract:

Aims: Plant diversity plays a crucial role in regulating food web dynamics and ecosystem functioning. Existing studies have documented significant impact on large herbivores and ecosystem functioning from loss in plant species diversity. Yet, little is known about effect of plant diversity loss on small herbivores and its functional consequence on food web dynamics, ecosystem functioning and ecosystem service, due to difficulty in distinguishing the role of plant diversity from larger fluctuations of small herbivore populations under the confounding effect of multiple ecological factors and demographic stochasticity.

Methods: Based on our long-term experiment of plant species removal in an alpine meadow on the Tibetan Plateau, we repeatedly measured the number of grassland caterpillar (Gynaephora alpherakiif) in each plot (i.e. density) in July and August over five years (2016-2020). We examined the effect of plant species removal on caterpillar population by associating changes in caterpillar counts with a shift in plant community attributes following plant removal and using a set of generalized linear mixed models.

Results: Plant removal induced a difference in the plant community, which significantly and consistently influenced the caterpillar counts over years and months independent of the observed variation in caterpillar counts accounted for by the difference in years and observational times within each month. Comparing the caterpillar counts in plots without removing plant species, (i) the caterpillar counts decreased in plots subjected to removal of dominant species (Kobresia capillifolia) in both July and August across five years, and increased in plots subjected to removal of forbs species in August of 2016 and 2019, and (ii) the caterpillar counts tended to increase in plots subjected to removal of legume species but only significantly in few cases, and did not significantly differ with plots subjected to removal of partial species (accounting for 20% aboveground biomass) of each functional groups. Further analysis showed that neither total aboveground biomass nor plant species richness and total individual numbers was significantly correlated with the caterpillar counts. The variation in caterpillar counts was associated positively with aboveground biomass of sedges and grasses as well as grass species richness, but negatively with the amount of forbs and legumes. The structural equation model further revealed that the removal of Kobresiaand grass species affected caterpillar counts through decreasing aboveground biomass of graminoids, but increasing graminoids following removal of forb and legume species promoted the density of caterpillars.

Conclusion: Based on field observations on the grassland caterpillar populations in plots subjected to different removal protocols, our study provides robust evidence that the decline in relative abundance of graminoids significantly reduced the fitness of caterpillars, and thus potentially impacts grassland functioning and ecosystem service. This study also suggests that future research on biodiversity effects and biological control of pests needs to pay more attention to the effects of plant functional diversity and demographic stochasticity on small herbivores.