Stochastic Spatial Random Forest for Detecting Remotely Sensed Forest Change Despite Missing Data

Forest cover is an indicator of species habitat and biodiversity that can be monitored effectively using satellite images. The benefits of using satellite images for large scale forest monitoring are that they are freely available globally and frequently updated, which reduces the need for extensive field data collection. Field data collection to monitor forest change can be prohibitively costly in many places around the world. A challenge of working with these images is missing data due to clouds, particularly in tropical regions where forest monitoring is essential. Existing methods for interpolating missing data based on only past observations, such as compositing, are effective for stable land cover but inaccurate for dynamic and substantially changing landscapes. In this talk I present joint work with Dr Kate Helmstedt and Distinguished Professor Kerrie Mengersen: our new machine learning method Spatial Stochastic Random Forest (SS-RF). Our method accurately interpolates missing forest and land cover under simulated forest clearing scenarios by taking spatial relationships in the landscape and past and current data into account to produce probabilities of land cover classifications. This is necessary because monitoring changing landscapes and modelling missing data are highly uncertain problems.

We found our SS-RF method detected different land clearing scenarios accurately, and importantly offers more accurate and robust estimates with associated uncertainty measurements not possible with traditional compositing approaches. This method has promise for use for other remotely sensed environmental monitoring cases, and I will discuss future plans to explore use of our method in the Antarctic environment.

Please join the event.

About Jacinta Holloway-Brown

Dr Jacinta Holloway-Brown is a lecturer in the School of Mathematical Sciences at University of Adelaide, South Australia. Her research focuses on developing hybrid approaches of spatial and Bayesian statistics and machine learning methods to model biodiversity and environmental change over time using remote sensing data.

Previously, Jacinta worked as a postdoctoral fellow in the Queensland University of Technology (QUT) Centre for Data Science, and a research associate in the ARC Centre of Excellence for Mathematical and Statistical Frontiers (ACEMS) at QUT, where she also completed a PhD. She has developed and taught hands on workshops on machine learning methods for analysing satellite imagery data for the United Nations, and run these workshops in Bogota, Colombia and Bangkok, Thailand. She also worked for the Australian Bureau of Statistics for years, more recently in methodology and tourism statistics roles.

Jacinta has degrees in statistics, journalism and economics, and enjoys working collaboratively to use data science to better understand, monitor and manage the environment.