Research Institute for the Environment and Livelihoods

Remote sensing and Earth observation

Research group

false colour image of trees made up of points, red at the top, with colours changing through yellow and green to blue at the bottom

This research group is focused on the application of Earth observation data and artificial intelligence (AI) for landscape-scale monitoring of tropical savannas, rangelands, and agriculture in northern Australia.

We develop new methods and improve existing methods of monitoring agricultural systems and natural resources in northern Australia. We use remotely sensed imagery to understand the climate and anthropogenic impacts on the environment and support pathways to environmental stewardship through robust remote sensing-based carbon accounting tools. Our research prioritises collaboration with First Nations communities and organisations through research, development, and adoption of projects.

We are a multicultural group of students and postdoctoral researchers with a broad range of backgrounds, interests, and skill sets. Our group includes First Nations and non-First Nations Australians as well as members from across the world.

Meet the team

Group leader

Dr Richard Crabbe

Research fellows

Glen Shennan (Spatial Analyst - Ecological Remote Sensing)

Dr Rafael Bohn Reckziegel (Research Fellow, Remote sensing riparian vegetation)

CDU affiliates

Prof Lindsay Hutley
Prof Brett Murphy
Prof Sarah Legge
Prof Stephen Xu
Dr Andrew Edwards
Dr Dylan Irvine

Honours

James Pike – Mapping groundwater-dependent vegetation using Landsat 8 and Random Forest

External collaborators

Dr Malcolm Lindsay (Environs Kimberley)
Dr Surantha Salgadoe (Wayamba University of Sri Lanka)

Former students

Michael Ogungbuyi (PhD, University of Tasmania) – Improving real-time pasture management using remote sensing, machine learning, and big data analytics
Cristina Rodriguez (master degree, CDU) – Detecting land clearing and regrowth in northern Australia using Sentinel-2
Jacques Theron (honours, Charles Sturt University) – Using Sentinel-1 synthetic aperture radar to characterize planophile and erectophile crop architecture and biomass
Sean Fisher (honours, Charles Sturt University) – Monitoring bushfires in Australia using NovaSAR-1 and Sentinel-1

black drone with 4 rotors standing on short green grass

Our research focus

Carbon accounting
Biodiversity and natural resource management
Agricultural productivity
Detection of invasive and rare native plants
Groundwater-dependent vegetation
Land condition assessment
Field-based spectral measurements

Importance of this research

Monitoring tropical savanna and rangeland ecosystems at local and global scales is central to understanding the impacts of climate change and management practices but requires efficient tools and methods for improved success.

Our research provides and evaluates efficient tools and remote sensing methodologies that are scalable to support national and international programs related to carbon accounting, biodiversity assessment, and sustainable agriculture.

Key projects

Innovative biodiversity monitoring

This work measures the habitat condition of desert and savanna vegetation integration on ground biodiversity survey data, drone and satellite remote sensing.

We are monitoring vegetation structure and habitat condition using methods at different scales, establishing a relationship between on-ground biodiversity surveys (i.e., vegetation and trapping surveys), drone light detection and ranging (LiDAR) and photogrammetry, and various synthetic aperture radar data.

Sites occur in two habitats in the Kimberley region: the spinifex desert on Karajarri Country (Great Sandy Desert) and the savanna on the Dampier Peninsula on Yawuru Country. The project includes training Aboriginal Rangers in driving satellite remote sensing data. The project finishes in March 2026.

Mapping invasive plants

This research focuses on remote weed detection using AI and satellite and uncrewed aerial vehicle (UAV) imagery.

This is a collaborative project, including Charles Sturt University, exploring AI algorithms and drone and satellite imagery to detect and map hawkweed (Pilosella officinarum), bitou bush (Chrysanthemoides monilifera subsp. rotundata), and African lovegrass (Eragrostis curvula). The project finishes in June 2024.

satellite with four solar panels on both sides, above cloudy blue atmosphere, against sky ranging in colour from black through light blue to white near the horizon, with partial moon visible in background

Potential PhD projects

Biomass and carbon modelling
Habitat condition assessment
Synthetic aperture radar and fire regimes
Detecting weeds of national significance
Modelling crop and pasture yields

Example publications

Shennan, G., Crabbe, R., 2024. A Review of Spaceborne Synthetic Aperture Radar for Invasive Alien Plant Research. Remote Sensing Applications: Society and Environment 101358. https://doi.org/10.1016/j.rsase.2024.101358

Crabbe, R.A., Lamb, D.W., Edwards, C., 2021. Investigating the potential of Sentinel-1 to detect varying spatial heterogeneity in pasture cover in grasslands. International Journal of Remote Sensing 42, 274–285. https://doi.org/10.1080/01431161.2020.1812129

Crabbe, R.A., Lamb, D., Edwards, C., 2020. Discrimination of species composition types of a grazed pasture landscape using Sentinel-1 and Sentinel-2 data. International Journal of Applied Earth Observation and Geoinformation 84, 101978. https://doi.org/10.1016/j.jag.2019.101978

Crabbe, R.A., Janouš, D., Dařenová, E., Pavelka, M., 2019. Exploring the potential of LANDSAT-8 for estimation of forest soil CO2 efflux. International Journal of Applied Earth Observation and Geoinformation 77, 42–52. https://doi.org/10.1016/j.jag.2018.12.007