PhD position in climate related statistical hydrological modelling
University of Bergen
The University of Bergen (UoB) is located in Bergen, Norway and serves approximately 17,000 students and employ 3 600 staff. PhD candidates are paid employees of staff, making the doctoral degree at UoB particularly attractive for rising talent. About one in three graduating doctors are from outside Norway. UoB has three strategic areas: Marine research, Climate and energy transitions and Global challenges. Within these areas, UoB contribute to society with excellent research, education, interdisciplinary cooperation and dissemination of knowledge and innovation.
Climate: Past, Present & Future (CL)
Hydrological Sciences (HS)
There is a vacancy for a PhD position at the Geophysical Institute, University of Bergen, Norway within climate related statistical hydrological modelling. The position is for a fixed-term period of 4 years and is and is funded by the University of Bergen through the Bjerknes Centre for Climate Research.
About the project/work tasks
Traditionally simulation of historical and future changes in streamflow is often done by hydrological models that require high-resolution quality meteorological input as well as streamflow observations for calibration of model parameters. Current estimates of flood recurrence intervals are often based on statistical extreme value techniques with instrumental time series of streamflow as input. In Norway, repeated what has been calculated to be 200-year floods have occurred over the last decades. This hints to the need for longer time series in order to calculate more realistic recurrence intervals in a changing climate. The use of information based on reconstructions of paleo floods and long climate simulations may therefore prove very beneficial for such calculations. With relatively new machine learning techniques that perform advanced statistical classification and prediction, and the emergence of temporal high resolution (daily) climate model output over several centuries it is possible to rethink the traditional approaches to hydrological modelling and flood estimation within climate.
The PhD project aim to develop and apply a hydrological modelling tool which link century scale atmospheric reanalysis, century to millennia atmospheric climate modelling data and paleo flood archives in order to estimate regional flood probability changes back in time as well as into the future for different flood types in order to better quantify and understand changes in flood frequency in a changing climate. For such a tool to be usable on historical, paleo and future climate simulations, it needs to be forced with parameters that are regularly stored by groups doing these types of simulations and be able to incorporate multi-decadal to seasonal and short-term signals. As a result of these constraints, we envision the development of a data-driven system using computational statistics/machine learning techniques providing links between regional flood occurrences of different types of floods and large scale atmospheric information from reanalysis, long model simulations and paleo flood records. The system will be trained on instrumental streamflow records.
The applicant must hold a master or an equivalent degree in computational statistics, machine learning, hydrology, atmospheric sciences or earth science. Master students can apply provided they complete their final master exam before 01.09.2018. It is a condition of employment that the master’s degree has been awarded.
- We seek candidates with knowledge in spatio-temporal statistical methods relevant for climate science and downscaling.
- Prior knowledge of computational statistics/machine learning techniques is preferable.
- Experience with evaluating the performance of observations or simulations within hydrology, weather or climate using statistical analysis, or experience with frequency analyses of time series is an advantage.
- Practical experience with programming (Fortran, C++, matlab, python, R or similar) is required.
- Demonstrated ability to work independently and structured, and good collaborative skills.
- Proficiency in English, both written and oral.