The primary goal of this project is to assess how well photogrammetry (also known as Structure-from-Motion) can be used to improve our understanding of coral reef demographics and vital rates from colony to habitat-scale. This project builds out the innovative software pipeline for generating 3D dense point clouds and extracting reef metrics. Using imagery collected across a broad range of habitats, coral communities, and environments, in the Main Hawaiian Islands, we aim to extract coral demographic data, benthic cover and habitat complexity data from a subset of 100 sites collected in summer 2019. In FY20, we are conducting a statistical analysis to determine whether there are measurable biases between in situ demographic data and SfM-generated data, and if so, whether we can model or adjust for these biases. Through this analysis we will identify preliminary limitations of SfM for generating demographic data. We are also testing whether SfM can be used to track individual colonies through time to measure growth, mortality and recruitment information (vital rates). As part of this project we are collaborating with the Hawaii Institute of Marine Biology to develop integral projection models using vital rate data to identify which demographic processes are driving coral resilience. Through this project we have the capacity to eventually reduce our operational complexity in the field and answer fundamental questions about the status of reefs across the Pacific and underlying demographic processes governing the resilience of reefs to local and global threats. In addition, the processing framework developed during this project is likely transferable to other benthic habitat and marine ecosystem types. We also continue to develop and foster partnerships with colleagues across the federal, academic and private sectors to develop cyber infrastructure that leverages machine learning and cloud processing, leading to more efficient 3D model generation and extraction of benthic data from imagery. More specifically we are collaborating with Scripps Institution of Oceanography, University of California San Diego, University of Hawaii at Hilo and NASA to develop the SfM processing pipeline and artificial intelligence tools needed to more efficiently extract benthic data from photomosaics. We are continuing to build capacity with local partners during in person SfM training workshops with Hawaii Division of Aquatic Resources, Fish and Wildlife Service, University of Hawaii, The Nature Conservancy, The National Park Service and the Coral Reef Alliance.
This project will result in coral demographic data (colony density, size, % mortality and health status) from 100 stratified random sites, as well as growth rate, mortality rate and reproduction from 10 fixed sites in the Main Hawaiian Islands. The belt data will be used for a statistical analysis of method comparability between in situ and SfM-generated demographic data. These results will be included in a technical report that will be submitted to PIFSC’s editorial board by 9/30/2020.
Jursdiction Priority Sites:
- Ka‘anapali-Kahekili (Maui)
- Pelekane Bay-Puako-Anaeho‘omalu Bay (Hawai‘i)