Contributions of deforestation, climate change, pollination, and seed dispersal to the population dynamics of a tropical herb

Background/Question/Methods

Deforestation can affect the diversity, abundance, and movement behavior of animals, which could indirectly affect plant populations via pollination and seed dispersal. In addition, plant growth, survival, and reproduction may be directly affected by deforestation and climate change. Using a demographic analysis, we compared these effects in the understory herb Heliconia tortuosa. First, we assessed if the abundance and richness of seed dispersers was affected by deforestation and compared our results with previous studies on the community of pollinators. Second, using telemetry data from one of the main seed dispersers of H. tortuosa, Turdus grayi, we developed a movement behavior model to simulate seed dispersal in the fragmented landscape. Lastly, we collected demographic information of H. tortuosa populations during 2013-2015 and 2015-2017 and developed a demographic model that included estimates of pollination and seed dispersal rates. We performed an elasticity analysis to understand the importance of different parts of the life cycle for H. tortuosa population growth rate (λ). We performed a Life Table Response Experiment (LTRE) to understand which parts of the life cycle of H. tortuosa were important for explaining the observed differences between sampling periods, and we related these differences to meteorological data for the study area.

Results/Conclusions

Contrary to pollinators, the abundance and diversity of the community of seed dispersers of H. tortuosa was little affected by deforestation. Seed dispersal simulations showed that most of the seeds stayed within 30 m of the maternal plant (63.30%) or landed in the same forest patch (16.10%), 5.21% dispersed to a different forest patch, and 15.4% were lost to the non-forested landscape matrix.. The elasticity analysis showed that growth, survival, and seedling establishment had the highest contribution to λ, while seed dispersal and pollination had a low contribution. The LTREs showed that survival was responsible for the observed differences in λ between populations sampled in 2013-2015 and 2015-2017. The period with the highest mortality coincided with an unusually dry wet season in 2015 for the study area, but not with the very strong el Niño event of 2015/16. Our results suggest that climatic differences between years, and future climate change, may have a greater impact on H. tortuosa population dynamics than the direct or indirect effects of forest loss and fragmentation.