2022 ESA Annual Meeting (August 14 - 19)

COS 33-2 Long-term impacts of foundation species loss are not influenced by compensatory dynamics: results from an experimental manipulation of foundation species in a temperate northeastern deciduous forest.

3:45 PM-4:00 PM
516E
Sarah Bruner, Columbia University;Mukund P. Rao,University Corporation for Atmospheric Reserach;Jan Eitel,University of Idaho;William Schuster,Black Rock Forest Consortium;Katie Terlizzi,Black Rock Forest Consortium;Kevin L. Griffin,Columbia University;Shahid Naeem,Columbia University;
Background/Question/Methods

Temperate forests are often dominated by single species or conspecific species which, through their physical architecture and ecophysiology, characterize the structure and function of the ecosystem. Loss of these “foundation species” from introduced pests or pathogens drastically shifts the local environment, leaving a forest that is significantly changed in species composition, form, and ecosystem properties. Whether this loss causes a permanent shift or if altered forests can recover their lost function through compensatory growth – and how long this might take – remains unclear. Using selective girdling of mature trees to manipulate foundation species density at Black Rock Forest, NY, we tested whether compensatory dynamics, were reflected in spectral indices ten years after the manipulation. We established four treatments: (1) where oak trees were killed by girdling, (2) were reduced to half their density by the same process, (3) all non-oak species were girdled, and (4) a control. We censused treatments yearly, starting in 2008, and matched these data with above-canopy spectral indices, specifically normalized difference vegetation index (NDVI) and green chromatic coordinate (GCC), captured by an unmanned aerial vehicle in 2017-2018.

Results/Conclusions

Results revealed that the temporal patterns in NDVI and GCC were significantly different among treatments over the course of the year. During the growing season, the standard deviation of NDVI varied significantly by treatment, with the oak-girdled plots having the highest value (0.05) along with the lowest total aboveground biomass (µ= 5,601 kg). Standard deviation of NDVI was negatively correlated with aboveground biomass, stand density, evenness, but had a positive relationship with disturbance (R2=0.45, 0.20, 0.26, and 0.52, respectively). Taken together, these initial findings suggest that, since spatial homogeneity of the spectral indices are related to the size, quantity, and evenness of trees, compensation for the loss of the oak species has not occurred and the impacts of foundation species loss are long-lasting within this system.