Skip to main content
Bayi Glacier in Qilian Mountain, China (Credit: Xiaoming Wang, distributed via

Job advertisement PhD at Aarhus University: Assessing future changes in Greenland runoff using machine learning and climate models

EGU logo

European Geosciences Union

PhD at Aarhus University: Assessing future changes in Greenland runoff using machine learning and climate models

PhD at Aarhus University: Assessing future changes in Greenland runoff using machine learning and climate models


Aarhus University


Roskilde, Denmark


Relevant divisions
Climate: Past, Present & Future (CL)
Cryospheric Sciences (CR)
Nonlinear Processes in Geosciences (NP)

Full time

Entry level


Required education

Application deadline
5 October 2023

15 September 2023

Job description

We seek a candidate for a fully-funded PhD project investigating the range of present and future runoff from Greenland under different climate scenarios. Output from high-resolution regional climate models and in-situ observations of runoff will be used to develop machine learning tools that can provide estimates of future runoff, with a combination of in-situ and satellite data to evaluate the approach. The project is a collaboration between Aarhus University, the Geological Survey of Denmark and Greenland (GEUS), and the National Centre for Climate Research (NCKF) at the Danish Meteorological Institute (DMI).

Runoff from Greenland includes components from the ice sheet, marginal glaciers and non-glaciated land surfaces. Temporal and spatial runoff variability can impact infrastructure (buildings, bridges and roads), security of supply (energy, water), industry/production, as well as fisheries and marine ecosystems. The potential impacts are emphasized with ongoing and future warming in the Arctic and knowledge of the runoff is therefore critical both for climate adaptation planning and mitigation efforts in and outside the Arctic.

State-of-the-art model simulations show a wide variation in melt and runoff estimates in the near future. While runoff calculations from high-resolution climate models are important to determine future potentials, there is still a limited number of simulations available, and these do not sample the full range of outcomes. At the same time, there is only limited observational data to assess these model simulations.

This project will therefore develop machine learning-based tools to be applied to different global and regional climate model outputs to produce an ensemble of medium- to long-term projections (2050 to 2150) of runoff under a range of future scenarios, allowing a statistical assessment of the likely range in future runoff.

This project offers the opportunity to work with research that will contribute to climate adaptation in Greenland and Denmark. International and domestic collaboration is essential and some fieldwork in Greenland can be expected. We seek a motivated and independent student with well-developed numerical skills and ideally experience in the field of machine learning. The candidate should expect to be situated both at DMI in Copenhagen and Aarhus University in Roskilde.

Qualifications and specific competences:

  • MSc education in physics, maths, computer science, geophysics, engineering, glaciology, environmental science or a related subject.
  • Well-developed numerical skills and ideally experience in developing and using machine learning models.
  • Experience in programming, data analysis and visualisation in Python, R, Julia, Matlab or similar.
  • Experience with climate model output or large observational datasets of climatic variables as well as knowledge of the Arctic climate system and climate change are further advantages.