Dronescape: UAV Observations of TERN Ecosystem Surveillance Monitoring Plots, Data Collection 

Dronescape Dataset Description

Dronescape is a research infrastructure initiative within TERN that integrates uncrewed aerial vehicle (UAV) remote sensing techniques into Australia's terrestrial ecosystem monitoring framework. The project addresses critical spatio-temporal gaps between field-based measurements and satellite observations by deploying UAV systems carrying RGB, 10-band multispectral, and LiDAR sensors across TERN's surveillance monitoring plot network. Through standardized protocols and automated processing workflows, Dronescape generates analysis-ready products including orthomosaics, reflectance maps, vegetation indices, and 3D point clouds.

As of 2025, the project has surveyed over 120 plots across multiple bioregions, with ongoing expansion to additional monitoring sites. Each plot covers approximately 90,000 m², contributing to a continuously growing dataset that enables long-term monitoring of Australia's diverse vegetation communities.

Project Scope

• Coverage: More than 120 plots across multiple bioregions (ongoing expansion)
• Plot Size: Approximately 90,000 m² per monitoring site
• Temporal Scale: 2–10 year monitoring cycles
• Data Types: RGB, 10-band multispectral, and LiDAR

 

We at TERN acknowledge the Traditional Owners and Custodians throughout Australia, New Zealand and all nations. We honour their profound connections to land, water, biodiversity and culture and pay our respects to their Elders past, present and emerging. We thank the TERN Surveillance team, particularly Sally O'Neill, Carla Franson and Troy Bektas for logistics and field support. We also acknowledge the Terraluma lab members from the University of Tasmania, in particular Poornima Sivanandam, Darren Turner, Alice Robbins and Caitlyn Smart for their support in protocol development and data collection. We further acknowledge the support of the TERN Data Services team. This project was funded by TERN Landscapes & Surveillance. 

The purpose of dronescape is to support TERN's multi-scale monitoring framework by providing high-quality, F.A.I.R. datasets that bridge the gap between field-based measurements and satellite imagery. The project enables detailed studies of vegetation community composition, structure, and spatio-temporal dynamics, facilitating research into species distribution, biodiversity, canopy architecture, phenology, and ecosystem responses to environmental changes. Additionally, dronescape serves as a platform for the remote sensing research community, providing resources for calibration/validation, protocol development, and processing algorithm innovation. By integrating UAV technology into standard ecosystem monitoring protocols, dronescape supports TERN's mission of providing comprehensive environmental research infrastructure at national and continental scales, while establishing standardized workflows for data collection, processing, and delivery through TERN's open-access infrastructure. 

Data Collection

The TERN Dronescape UAV observations follow the standard TERN Ecosystem Surveilance Protocols.

Technical Specifications Platform
• UAV: DJI Matrice 350 RTK
• Sensors:
• DJI Zenmuse P1 (RGB)
• Micasense RedEdge-P Dual (10-band multispectral)
• DJI Zenmuse L2 (LiDAR)
• Georeferencing: DJI D-RTK2 base station and onboard GNSS RTK
Multispectral Bands
• Band 1: Coastal, Center 444 nm, FWHM 28 nm
• Band 2: Blue, Center 475 nm, FWHM 32 nm
• Band 3: Green, Center 531 nm, FWHM 14 nm
• Band 4: Green, Center 560 nm, FWHM 27 nm
• Band 5: Red, Center 650 nm, FWHM 16 nm
• Band 6: Red, Center 668 nm, FWHM 14 nm
• Band 7: RedEdge, Center 705 nm, FWHM 10 nm
• Band 8: RedEdge, Center 717 nm, FWHM 12 nm
• Band 9: Near-Infrared (NIR), Center 740 nm, FWHM 18 nm
• Band 10: Near-Infrared (NIR), Center 842 nm, FWHM 57 nm
• FWHM: Full Width at Half Maximum
LiDAR sensor
• Penta-return @ 240 khz
• Non-repetitive
• Variable LiDAR overlap based on canopy height (52% to 75%)
• RGB frontal overlap (70%)
Data Processing
• RGB and Multispectral: Structure from Motion (SfM) using Agisoft Metashape v2.1.1
• LiDAR: Processed using DJI Terra 4.0.10
• Resolution: 1–5 cm/pixel for orthomosaics
• Co-registration: RGB and multispectral data co-registered during post-processing
Data Processing Levels
Level 0: Raw Sensor Data
• RGB: Original JPEG imagery with navigation files (.nav, .mrk, .obs, .bin)
• Multispectral: Raw 10-band TIF imagery with positioning data
• LiDAR: Raw point cloud data (LDR) with calibration and navigation files (CLC, CLI, RTK, RTL, RTS, RTB, IMU)
Level 1: Analysis-Ready Products
• RGB: Cloud-optimized GeoTIFF orthomosaic
• Multispectral: 10-band cloud-optimized GeoTIFF orthomosaic
• LiDAR: Processed point clouds (LAS, PLY) and digital terrain models (COG)


Details on the Data Directory Structure are provided in the Readme text file.