Spatial sciences
Research leaders in spatial intelligence
The Spatial Sciences Group conducts research to improve our understanding, management and monitoring of the environment at landscape scales.
We use remote sensing, geographic information systems, ecological modelling and multi-objective decision support systems to understand spatial variability, temporal dynamics, change and interrelationships in the environment.
We work in a diversity of natural and managed ecosystems, terrestrial, aquatic and marine environments.
Our capability is nationally and internationally unique because of a co-location of expertise in all scales of spatial imaging from unmanned aircraft to satellites, spatial big-data analysis, and environmental management.
Major themes of our research are:
- Spatial variation in landscape – what occurs where and why?
- Biodiversity and landscape composition
- Improving environmental and resource mapping
- Assessing land, habitat, vegetation, soil and water condition
- Monitoring environmental change over time
- Natural resource and wildlife management planning and decision support
Research strengths
-
Ecosystems, habitats and biodiversity
Geospatial information and analysis provide powerful new tools for characterising ecosystems, assessing the distribution and condition of habitats and monitoring native and invasive species.
Key projects
-
Wildlife
Wildlife management is about decision making in space. Our research uses spatial monitoring and analysis tools at the interface of natural and agricultural environments, addressing questions related to cost-efficient management of spaces for species conservation.
Key projects
Spatial technologies supporting management of the Southern Hairy-nosed Wombat
ARC Linkage Project LP160100937
-
Fire
We use remotely sensed images from aerial surveys, drones, Earth observation satellites and geographic information systems (GIS) for applications relevant to many aspects of bushfire assessment, monitoring and prediction.
These include immediate response, medium term monitoring of impacts and recovery, through to long term assessment of past fire frequencies to inform infrastructure and environmental management and planning.
Key projects
A new remote sensing method for mapping and monitoring burnt areas in arid South Australia
Image by Peripitus (CC-BY-SA by 2.5)
-
Arid lands
Our research uses remote sensing and geospatial data to provide objective, spatially explicit and comprehensive assessments of soil and vegetation cover and land condition over extensive arid landscapes.
We analyse time series of images to understand short and long term trends in condition, and the influences of climate and land management.
Key projects
CARMS - Condition Assessment and Risk Management System
A new automated tool to assist land condition assessment across South Australian Pastoral Leases using satellite remote sensing, geospatial data analysis and machine learning. This is a collaboration with PIRSA and AIML.
-
Wetlands and inland waters
We apply a wide range of satellite and airborne imagery and geospatial analyses to develop new tools for monitoring inundation in inland waters and the condition of their associated wetlands.
Key projects
More objective and repeatable monitoring of artesian springs influenced by mining
Collaboration with BHP and Department of Environment and Water
-
Coasts and marine environments
We use satellite image-derived indicators of ocean water quality to understand regional patterns and drivers in the marine environment and advanced airborne assessment of seagrass and near-shore benthic habitats.
Key projects
Monitoring seagrass on Metropolitan Adelaide coast
A collaboration with SA Water, SA Environment Protection Agency, Department of Environment & Water, Airborne Research Australia and Flinders University
-
Agriculture and viticulture
Our research informs agriculture and viticulture at a range of scales from plantations and properties to regional and national evaluations of production.
Remote sensing and geospatial analyses offer new technologies for assessing soils, water availability, nutrients and disease, and understanding the distribution of constraints and variability in plant production.
Key projects
-
Socio-economic impacts of climate change
Geospatial data and analysis provides valuable insights into the distribution and severity of impacts of weather and climate change.
Key collaborations
We collaborate with a wide range of scientists and environmental managers to develop and apply new approaches and tools for their assessments and monitoring, including:
- Environment Institute
- Unmanned Research Aircraft Facility
- Australian Institute for Machine Learning
- Goyder Institute for Water Research
- Terrestrial Ecosystems Research Network - we are the SA node of TERN Landscapes platform
- The Plant Accelerator, Australian Plant Phenomics Facility
- Surveying and Spatial Sciences Institute - We are an Education Sustaining Partner
- Erasmus Mundus International Masters in Applied Ecology
- Bureau of Meteorology
- Primary Industries and Regions SA
- SA Department of Environment and Water
- SA Water
Our people
- Ingrid Ahmer
- Matthew Farrell
- Hannah Auricht
- Hasan Musana
- John Nairn
- Angus Retallack
- Yeniu (Mickey) Wang
- Jennie Weinheimer
- Lauren Werner
News
Spatial Points blog
5
Dec
Using deep learning to detect an indicator arid shrub in ultra-high-resolution UAV imagery
Article link: Using deep learning to detect an indicator arid shrub in ultra-high-resolution UAV imagery My PhD research is in collaboration with Bush Heritage Australia. The second chapter of my thesis concerns the use of UAV imagery for gathering species-composition information, a component of biodiversity where measurement has typically been restricted to on-ground methods. I …
28
Nov
21
Nov
HOW TO SEE SEAGRASS? WITH DRONES!
Seagrass forms an incredibly important ecosystem worldwide, providing a wide range of functions. However, as with many other lifeforms, their health has been threatened by human advancements. Significant declines of seagrass meadows extent (up to a third of meadows along the Adelaide Metropolitan Coastline) leading up to the early 2000s, has been due to poor …
13
Sep
Conference Poster – Using Deep Learning to Detect an Arid Shrub Species in UAV Imagery
This poster was presented at the Advancing Earth Observation Forum in Brisbane in August 2022, and shows the results of the second chapter of my thesis. The full resolution PDF of the poster can be viewed here. Additional information about the research can be seen in this handout. The research paper detailing the research presented …
20
Apr
Deep Learning in ArcGIS Pro using your GPU
ArcGIS Pro includes built in tools that allow end-to-end deep learning, all within the Arc interface – from training sample labelling, through model training and final image classification / object detection. This removes the need for coding and installing the correct versions of the required libraries for machine / deep learning in Python. However, training …