On November 5 2015 the Fundao iron ore tailings dam in Brazil failed. This lead to 60 million cubic meters of water and waste flowing into the Doce River. The escaped water destroyed the town of the Bento Rodrigues and caused 17 deaths.
In March 2016, co-owners of the mines, BHP Billiton and Vale, reached a US$2.4 billion settlement. Four weeks later the share price of BHP Billiton fell 25% and the company suffered a US$8.3 Billion loss for 2016.
The Funando dam failure was not an isolated event. In the past decade, an average of two or three tailings dam failures have occurred worldwide each year. Part of the problem is incredibly simple: The tailings dams are simply holding too much water.
Remote sensing offers a cost-effective mechanism to automatically monitor the volume of water in tailings dams, regardless of the operational context. Moreover, the monitoring could be extensive, performed at a national or global scale, and continuous, occurring up to a weekly or fortnightly basis.
We are looking to analyse satellite images to see if we automatically identify how much water is in tailings dams either nationally or overseas.
As part of your research project, you will be involved in:
- reviewing literature about existing methods
- developing an algorithm
- downloading suitable test cases
- testing the algorithm.
The project will develop an algorithm to detect water levels in mine tailing dams.
Skills and experience
To be considered for this project, you will need to have experience with remote sensing (such as ArcGIS and QGIS) and with coding with lanuages such as Python, R and C++.
You may be able to apply for a research scholarship in our annual scholarship round.
Contact the supervisor for more information.