Extreme weather events such as droughts are expected to increase tree mortality rates, and large trees are particularly vulnerable. However, we cannot accurately measure large tree mortality rates with current methods. Field data does not sample enough large trees, while remote sensing methods currently only measure the total canopy area disturbed. In this project, I will apply novel machine learning methods to detect and segment large trees in remote sensing data. This enables me to track individual large trees over time and therefore quantify tree mortality from remote sensing. This dramatically increases the sample size, and I will use this increased sample size to explore the drivers of large tree mortality and how these may be affected by climate change.