Wei Ji Leong
Phone: 04 463 5233 x8373
Office: CO 522
BSc VUW (2014); BSc(Hons) VUW (2015)
Deep Learning in the Cryosphere: Using deep neural networks to investigate subglacial water
Project aim and description
Water distributed underneath the Antarctic ice sheet affects the rate at which glaciers flow into the ocean. These subglacial water systems are dynamic with change observed over short timescales. This study uses a novel deep learning approach to handle the large and varied remote sensing datasets available to find subglacial water features. Convolutional Neural Networks will be used to increase the spatial resolution of cryospheric datasets and classify subglacial lakes in Antarctica. The final product is a revised inventory of subglacial lakes, channels and aquifers, with information on their evolution over time and the implications for Antarctic ice flow.