Background

Dr. Jigme Thinley is a geospatial scientists and systems engineer specializing in developing intelligent Artificial Intelligence (AI) and Machine Learning (ML) models to answer complex environmental, climate, and developmental questions using geospatial and Earth Observation (EO) data. Dr. Thinley has more than a decade of multi-disciplinary research and development in Australia, Europe (Switzerland and Austria), Asia (Bhutan, India and Nepal) and Africa (Kenya). The particular research areas where he has been involved include biodiversity monitoring, urban forest biomass mapping, Glacier monitoring, Landslide susceptibility mapping, flood risk mapping, bush fire monitoring and design of national level digital monitoring systems. Through his research, he has been able to publish more than 7 peer reviewed articles both as a lead author and also as an effective contributor. Further has been involved in developing project reports and training manuals for use by different institutions in Bhutan. Beyond research and development, Dr. Thinley has been involved in capacity building and training both formally within the university as a lecturer in Geoinformatics and IT, and also through multi-stakeholder capacity building platforms. With his strong foundation in IT and software development, Dr. Thinley has also been involved in developing novel and intelligent software solutions to support in proper water resource and water utilities management, an example of his contribution is in the eMita digital solution for water utilities management that is currently being piloted by different resource constrained water utility companies in Kenya.

Download CV

Education

2022-2025

Griffith University

PhD in Geospatial Science (Forest Ecology, AI & Remote Sensing) | Griffith University, Australia | 2025

Thesis: Multi-Scale Characterization of Urban Forests Using Machine Learning and Hyperspectral Remote Sensing

The thesis focused on developing machine learning models (multilinear regression models, random forest classifiers and convolution neural networks CNN) for quantifying aboveground biomass in planted and remnant urban forests using remote sensing data from various sources including airborne sensors, multispectral drone imagery, LiDAR point clouds, and passive and active satellite-based imagery from Landsat, Sentinel-1 Synthetic Aperture Radar (SAR), Sentinel-2 and MODIS. The case study locations were in Logan and Nathan in Queensland, Australia. The result showed that combining multisource remote sensing data with field-sampled information using intelligent and self-learning models is a viable, non-intrusive, and accurate method for estimating above ground biomass even for small (less than 10 hectare) urban forests. The outcome of the work is particularly relevant in supporting in documenting and reporting on the contribution of nature-based solutions to climate mitigating efforts in busy urban environments. Apart from the thesis, five (5) high quality publications were developed from the research.

2011-2013

University of Salzburg

M.Sc in Applied Geoinformatics | University of Salzburg, Austria | 2013

Thesis: “Land Use and Land Cover Changes Using Object-Based Image Analysis in Thimphu Bhutan”

This research applied novel object-based segmentation and feature modelling techniques to map land cover changes around Thimphu City in Bhutan. As opposed to pixel-based land cover maps, the methods applied resulted to semantically relevant classes of urban surface, cropland, forest area, and landslide prone bare lands within the city boundary. The outcome of the work was relevant for updating the land cover maps of the city. Apart from the thesis work, the program provided novel foundational knowledge and methods for geospatial data acquisition, modelling and visualization. In addition, course units in geospatial software engineering provided handy knowledge and technical skills for full cycle development of digital solutions for geospatial data workflows.