Top of page

Student/Staff Portal
Global Site Navigation

School of Science

Local Section Navigation
You are here: Main Content

Related Contents

An Artificial Intelligence based multimodal approach for monitoring peatlands

Project Outline

The peatlands that occur throughout southwestern Australia, including those in the Walpole Wilderness Area, are unique ecosystems at threat from climate change and in particular, fire. Soil moisture levels within peats are integral to maintaining the hydrological integrity of these systems and their resilience to fire. However, the hydrology of these systems is poorly understood. Significant improvements in remote sensing technologies mean that we now can estimate temporal fluctuations in soil moisture over large spatial scales by combining various remote sensing products. Recent advances in Internet of Things (IoT) and Artificial Intelligence (AI) especially the transformer-based deep learning models also allow the obtaining and processing of a large volume of images, texts and other types of data relevant to the vegetation and fire-severity classifications. This project will develop an efficient and robust machine-learning approach to the fusion of multimodal data that can be used for reliable and automated monitoring of different biodiversity and geo-diversity aspects of the peatlands.

Desired skills

Prospective candidates should have Masters or Honours degree (first class or equivalent). Good interpersonal communication and the ability to work as part of a multidisciplinary team will be expected from the candidate. The ability for independent, organised work and advanced communication skills in English (oral and written) are also essential.

Project Area: Computer Science

Supervisor(s): Dr Shams Islam (ECU), Dr David Blake (ECU), Dr Fabian Boesl and Dr Mohammad Awrangjeb

Project level: PhD

Funding: Ian Potter Foundation

Start date: Any

Skip to top of page