2x PhD Positions as part of the SNSF Project “From Alps to Arctic: Satellite-based Assessment of Forest Canopy Height across Decades” 80 %

The EcoVision Lab, led by Prof. Jan Dirk Wegner, in the Department of Mathematical Modeling and Machine Learning (DM3L) at University of Zurich (UZH) is seeking applications for two Doctoral Candidates with a strong background in computer vision and machine learning for developing novel deep learning methods for estimating forest parameters across the Arctic and the Alps over the last decades from historical remote sensing imagery as part of a joint project collaboration with the Remote Sensing of Environmental Change (RSE) group at UZH led by Prof. Livia Piermattei (see open RSE jobs here).


We offer an exciting and stimulating environment to study and work in. The University of Zurich has several internationally recognized research groups dedicated to data science, machine learning, remote sensing, historical aerial image interpretation, forest analysis, and more broadly ecology. We also collaborate with several other institutions and companies in the fields of computer vision, machine learning and earth observation, in Switzerland and abroad. The EcoVision Lab is a member of UZH.ai, the ETH AI Center, the UZH Digital Society Initiative, the UN-ETH partnership, and the ETH for Development Center (ETH4D).

We are looking for two highly motivated PhD candidates to join a Swiss project consortium of 1x PostDoc and 3x PhDs across two research groups at UZH aimed at transforming how forest analysis across the Alps and Arctic is accomplished for the last decades. The project's goal is ambitious: using historical satellite imagery to build a harmonized time-series of high spatial resolution of forest parameters like canopy height back to the 1970s. The objective is to gain a better understanding of forest change across the Alps and the Arctic to eventually inform policymakers and support environmental decision-making internationally.

Why This Project Matters

The project will develop a harmonized, open, and scalable workflow for generating high-resolution, multitemporal spatially continuous canopy height maps (CHMs) at regional-to-biome scales. It will deliver 5 m ground sampling distance (GSD) CHMs for 2000-2025 at five-year intervals, with annual products to capture rapid forest dynamics and extend CHM predictions back to the 1970s (at 10 m GSD) by integrating Landsat with declassified stereo satellite imagery. This will establish the first multi-decadal baseline of canopy height and more forest parameters, enabling to disentangle natural variability from anthropogenic trends and detect changes in forest growth, productivity, degradation, and regeneration. Implementation of the workflow will begin in the Alps, leveraging Switzerland's exceptional validation resources (LiDAR, aerial imagery, field data), before scaling to boreal forests worldwide (~12 million km2), which is an essential region to the global carbon cycle and is undergoing rapid climate-driven transformation.