Our research explores the spatial and temporal patterns of forest succession following landslide disturbances in the Aotea and Coromandel regions using remote sensing and field-based approaches.
Common globally in mountainous regions and often driven by rainfall or seismic events, landslides trigger the process of primary succession by removing existing vegetation and topsoil. While early landslide succession research exclusively used field studies to monitor forest changes, modern approaches have increasingly integrated remote sensing and predictive modelling methods.
In New Zealand, however, research on post-landslide forest recovery has been limited and exclusively involved the use of vegetation surveys. Our research aims to address this gap while providing data processing pipelines for continued recovery monitoring. We aim to first build a multi-temporal landslide database for the Aotea and Coromandel regions using a semi-automatic approach with freely available satellite and aerial imagery from the 1940s to present. Canopy recovery will be monitored annually on all detected landslides using spectral identification models.
Field surveys and seed bank analyses will be used to complement the findings of all remote sensing work, providing insight into the community level changes within the landslide and across the broader landscape.
By surveying in a select chronosequence of landslides and the surrounding undisturbed forests, she aims to assess the relative changes in species diversity, functional traits, and invasive species prevalence over time. By integrating remote sensing and field-based methods, this research aims to provide an enhanced understanding of local landslide-recovery processes and insight into potential community shifts that further landslide events could cause.
With a projected increase in landslide frequency due to climate change and land use intensification, the community effects of this disturbance method will continue to be of importance for local conservation and ecosystem management efforts.
