Master thesis topics

List of possible master thesis topics

Publisert: (Updated: )

Early 20th century warming in the Arctic

Supervisors: Lea Svendsen (UiB) and Ingo Bethke (UiB)

Background

The Arctic is now warming more than 2-3 times faster than the global average due to human greenhouse gas emissions. However, the Arctic also had a warming period in the early 20th century (1910-1940) that cannot be explained fully by greenhouse gas emissions.

Tasks

The student will analyze data from large ensemble experiments with the Norwegian Climate Prediction Model

  1. Identify simulations which reproduce the observed early 20th century Arctic warming
  2. Investigate the Atlantic and Pacific Oceans in these selected simulations
  3. Calculate heat transport to the Arctic (investigate mechanisms) 

Relevant literature: 

  • Bokuchava, D. D., and V. A. Semenov. "Mechanisms of the early 20th century warming in the Arctic." Earth-Science Reviews 222 (2021): 103820.
  • Svendsen, Lea, et al. "Pacific contribution to the early twentieth-century warming in the Arctic." Nature Climate Change 8.9 (2018): 793-797. 

 

 A prediction model for spring rain in Ethiopia

Supervisors: Lea Svendsen (UiB) and Rondrotiana Barimalala (NORCE)

Background

The greater horn of Africa, and especially Ethiopia has experienced prolonged drought over the last years. Ethiopia has several rainy seasons throughout the year, but the spring rainy season from March to May is especially important in the southern part of Ethiopia. However, the spring rain has proven to be especially difficult to predict. 

Tasks may include: 

  1. Build and evaluate a prediction model for spring rain in Ethiopia based on climate variables already identified using observations.
  2. Use data from the Norwegian climate prediction model to improve the prediction model
  3. Perform future predictions for spring rain in Ethiopia

Relevant literature: 

Carina Knudsen (2022) Factors influencing interannual variability of Belg rain in Ethiopia. Master thesis. https://hdl.handle.net/11250/3059081 (ekstern lenke)

 

 

Master’s projects in Climate Futures: developing climate services in collaboration with user partners

Contact: Lea Svendsen (UiB)

Background

Climate Futures is a center for research-based innovation developing climate predictions for handling climate risk. The researchers in Climate Futures work together with nearly 40 partners from industry and public organizations across sectors and disciplines to tackle one of the most urgent challenges of our time: changing nature of weather and climate. In Climate Futures we have several master’s projects where the student will collaborate with one or more of these partners from industry and/or public organizations, where the project will directly target the needs of that partner related to climate risk. These partners cover areas such as renewable energy, insurance, food production and shipping, and the master’s projects are related to extreme weather, seasonal forecasts, and climate predictions. 

 

 Drivers of decadal variability in the Atlantic: role of wind stress

Supervisors: Lea Svendsen (UiB), Helene Asbjørnsen (UiB), Marius Årthun (UiB) 

Background

Climate predictions in the North Atlantic as especially promising on decadal timescales. However, what the major drivers of decadal variations in the North Atlantic is still uncertain. The atmosphere forces variability in the Atlantic, but is it only the heat fluxes or can the wind stress itself change the temperature and circulation in the North Atlantic Ocean on decadal timescales? 

Tasks: 

Analyze the Atlantic Ocean circulation in an experiment with the Norwegian earth system model (NorESM) where observed wind stress is prescribed to the ocean surface over the Atlantic, and compare the circulation with a fully coupled historical simulation with the same model. 

Relevant literature: 

Zhang, Rong, et al. "A review of the role of the Atlantic meridional overturning circulation in Atlantic multidecadal variability and associated climate impacts." Reviews of Geophysics 57.2 (2019): 316-375.

 

 The role of the North Atlantic Ocean in climate prediction of the North Pacific Ocean

Supervisors: Lea Svendsen, Helene R. Langehaug (NERSC)
 

Background:

While climate change studies focus on our climate at the end of this century (year 2100), it may be more useful (for f.ex. politicians, energy sector, municipal planning) to know our climate the next 10 years. In Bergen, we are developing a new state-of-the-art climate prediction model, the Norwegian Climate Prediction Model (NorCPM) building on the Norwegian Earth System Model, to improve seasonal-to-decadal climate predictions so they can be beneficial and useful for society. 

The North Atlantic Ocean is a region which is found to have high climate predictability, and climate prediction using climate models show promising result when predicting upper ocean temperature several years ahead. North Pacific Ocean temperatures, on the other hand, have so far been difficult to predict.

North Pacific Subtropical Mode Water is a relatively warm water mass occupying large areas and volumes of the upper ocean in the western North Pacific subtropical gyre. This water mass has an important role in distributing heat and nutrients from the surface to the subsurface. A recent study found that the decadal variations in the temperature of this water mass is controlled by the Atlantic Multidecadal Variability (AMV). AMV is the slow changes in the sea surface temperature in the North Atlantic Ocean. 


Tasks:The aim of this thesis is to investigate the relationship between North Pacific Subtropical Mode Water and AMV, and evaluate the possibility of using information about the North Atlantic to improve climate predictions in the North Pacific Ocean. The student will analyze output from NorCPM over the time period 1960-2010. The tasks involve identifying the North Pacific Subtropical Mode Water in the model, calculating the AMV mode, and investigating the forecast skills of both North Pacific Subtropical Mode Water and AMV mode and their relationship. 

Relevant literature: 

 

 The Indian summer monsoon in the Norwegian climate prediction model 

Supervisors: Lea Svendsen (UiB), Noel Keenlyside (UiB)

Background
Since India receives about 80 % of their annual rainfall during the summer monsoon season, the strength of the monsoon has a large impact on agriculture, economy and livelihood in this region. The strength of the Indian summer monsoon is given by large-scale dynamics based on for instance wind. But the large-scale circulation does not always match the rainfall, and climate models have difficulty reproducing the geographical pattern and magnitude of rain. 

Tasks: 

The student will analyze the large-scale atmospheric circulation related to Indian summer monsoon rainfall in the Norwegian climate model and compare with observations. The aim of the master thesis is to first evaluate the performance of the Norwegian climate model in simulating the monsoon, and then to use the model to study factors controlling variability of the monsoon. The student could also test and compare different definitions of the monsoon.

Relevant literature: 

  • Webster, P. J., Magana, V. O., Palmer, T. N., Shukla, J., Tomas, R. A., Yanai, M. U., & Yasunari, T. (1998). Monsoons: Processes, predictability, and the prospects for prediction. Journal of Geophysical Research: Oceans, 103(C7), 14451-14510.
  • Cherchi, A., & Navarra, A. (2007). Sensitivity of the Asian summer monsoon to the horizontal resolution: differences between AMIP-type and coupled model experiments. Climate dynamics,28, 273-290.

 

 Exploring climate services in collaboration with user partners and forecast providers

Supervisors: Noel Keenlyside, Ingo Bethke, Francois Counillon, Erik Kolstad

Background

Climate services is a rapidly evolving field that aims to deliver climate information in form useful to users for making decisions related to climate from sub-seasonal to multi-annual, and longer timescales. We are leading activities at the Bjerknes Centre in the development of climate predictions and climate services. We can offer projects related to development, evaluation, and provision of climate prediction and services. Through the recently funded Centre for Innovation Based Research – Climate Futures, there will be opportunities to develop projects in conjunction with user partners in the fields of renewable energy, sustainable food production, resilient societies, and smart shipping.

More information about these projects are available at the following websites:

https://bcpu.w.uib.no/ (ekstern lenke)

https://www.climatefutures.no (ekstern lenke)

 

 Why is the ITCZ North of the Equator?

Supervisor: Noel Keenlyside

Background

The intertropical convergence zone (ITCZ) is a rain band extending across the tropics that is essentially confined to the region just north of the equator. Early studies argued that this was because of local ocean-atmosphere interaction and the shape of continental coastlines in the tropics. However, recently the interhemispheric energy transport was suggested to be the main reason that the ITCZ is in the northern hemisphere.

Tasks
The aim here is analyze observations and idealized experiments to understand the role of oceanic circulation and continental distribution in positioning the ITCZ.

Relevant literature:

  • Philander, S. G. H., D. Gu, G. Lambert, T. Li, D. Halpern, N. C. Lau, and R. C. Pacanowski, 1996: Why the ITCZ Is Mostly North of the Equator. Journal of Climate, 9, 2958-2972.
  • Schneider, T., T. Bischoff, and G. H. Haug, 2014: Migrations and dynamics of the intertropical convergence zone. Nature, 513, 45-53.

 

 Developing climate services

Supervisor: Noel Keenlyside

Background

Climate services is the provision of climate information in a form useful to users. These include products based on outlooks of precipitation, wind speed, and temperature that can be useful for agriculture, renewable energy, health sectors, aquaculture, and fisheries. At the Bjerknes Centre we are developing the ability to predict climate from sub-seasonal to multi-annual timescales and at regional scales. We are also working towards making the predictions useful to a variety of stakeholders within the SFI Climate Futures.

Tasks
Thesis projects are available in a variety of topics, for different regions, timescales, and sectors. All projects will be defined with the involvement of a user partner, who will also follow thesis. Tasks will include the analysis of climate reanalysis, historical observations, and climate prediction data. Standard and modern (machine-learning) approaches to analyze data will be employed. Projects may involve performing and analyzing new climate model experiments. Outcome of the project will be information and product useful to the stakeholder partner.

Relevant links

 

 Variability and predictability of the Atlantic Niño

Supervisors: Noel Keenlyside, Shunya Koseki

Background

The Atlantic Niño affects marine ecosystems, African and South American climate, and also El Niño Southern Oscillation. The mechanisms for this variability remain uncertain. Particularly debated are the contribution of thermodynamic and dynamic ocean-atmosphere interaction, and the role of extra-tropical atmospheric forcing. 

Tasks
Analysis of observations and targeted simulations with the Norwegian Earth System Model will be used to better understand the mechanisms for the Atlantic Niño, its predictability, and how it will be impacted by climate change.

Relevant literature:

  • Lübbecke, J. F., C. W. Böning, N. S. Keenlyside, and S.-P. Xie, 2010: On the connection between Benguela and equatorial Atlantic Niños and the role of the South Atlantic Anticyclone. J. Geophys. Res., 115, C09015.
  • Nnamchi, H. C., J. Li, F. Kucharski, I.-S. Kang, N. S. Keenlyside, P. Chang, and R. Farneti, 2015: Thermodynamic controls of the Atlantic Nino. Nat Commun, 6:8895.

 

 Atlantic decadal variability and prediction

Supervisors: Noel Keenlyside, Francois Counillon

Background

North Atlantic ocean exhibits pronounced shifts on the timescale of decades. These affect marine ecosystems, climate of surrounding continents, and the Arctic. The variations are so strong that they can temporarily offset the impacts of global warming. Climate models can partly simulate and predict these variations, but large uncertainties exist in the underlying processes. Recently even the role of ocean dynamics has been questioned. 

Tasks

Here we will address the contribution of ocean dynamics versus thermodynamics by careful budget analysis of ocean observations and reanalysis, and simulations and predictions with the Norwegian Earth System Model.

Relevant literature:

  • Keenlyside, N. S., J. Ba, J. Mecking, N.-O. Omrani, M. Latif, R. Zhang, and R. Msadek, 2015: North Atlantic multi-decadal variability - mechanisms and predictability. Climate Change: Multidecadal and Beyond, C.-P. Chang, M. Ghil, M. Latif, and M. Wallace, Eds., World Scientific Publishing Company, Singapore, n/a. ISBN 978-9814579926.
  • Clement, A., K. Bellomo, L. N. Murphy, M. A. Cane, T. Mauritsen, G. Rädel, and B. Stevens, 2015: The Atlantic Multidecadal Oscillation without a role for ocean circulation. Science, 350, 320.

 

 Does ENSO matter for European climate?

Supervisors: Camille Li, Noel Keenlyside, Martin King

Background: 

The El Niño Southern Oscillation (ENSO) is the dominant mode of climate variability on Earth and is characterized by warming (cooling) of tropical Pacific sea surface temperatures during its El Niño (La Niña) phase. ENSO creates teleconnections that affect weather across large parts of the globe, for example, setting risks of floods and droughts in Australia, and shifting where good ski conditions are found along the west coast of North America. In the Euro-Atlantic sector, however, the more regional North Atlantic Oscillation (NAO) dominates, leading to open questions about whether and how the two modes interact (Mezzina et al. 2020). A better understanding of the relationship between ENSO and the NAO is important for many reasons, from advancing knowledge of forced and internal climate variability to seasonal predictions of European climate to evaluating (hopefully improving) climate models.

Tasks: 

The NAO dominates Euro-Atlantic variability in the real world. A filtered composite analysis suggests that part of this signal (~30% of the total SLP variability) originates from ENSO (see figure). This project aims to investigate the ENSO teleconnection to the Euro-Atlantic sector in climate models by repeating the filtered composite analysis on model simulations, starting with the Norwegian Earth System Model and Norwegian Climate Prediction Model, and expanding to other CMIP models and/or seasonal hindcasts. The differences in model behaviour will guide an exploration of when ENSO affects the NAO and when it doesn’t, evaluated against the considerable envelope of internal variability in the North Atlantic (King et al. 2021). This should also help provide insight into model performance in capturing the mean state and variability of European climate, with implications for near-term predictions and future projections of climate change.

This project is suitable for students who have mastered the topics taught in GEOF210, GEOF212, GEOF213, GEOF348 (preferably also GEOF352) and are enthusiastic about gaining physical understanding from quantitative analysis of the climate system. Experience with coding (Fortran, C, Matlab, python or similar) and some familiarity with the unix/linux environment are essential.


Relevant literature: 

 

 Constraining European climate impacts from a large-scale perspective

Supervisors: Camille Li, Stefan Sobolowski

Background: 

The North Atlantic sector exhibits some of the most stubborn climate model biases (errors) in key features of the large-scale atmospheric circulation. These errors include a weak stationary wave, a jet stream that extends too far into Europe (and often too far south), and a storm track that is too weak (Woollings 2010, Chang et al. 2012).  These biases affect our ability to accurately simulate downstream weather and climate over Europe, which has implications for near-term predictions and future projections of local climate and extremes.  Fixing the errors would of course be the best solution, but this has proven to be extremely difficult! In the meantime, we can make progress by better understanding how large-scale circulation biases affect local/regional European conditions.

Tasks: 

There are two possible approaches to tackling this problem starting from known Euro-Atlantic biases that have persisted over several generations of global climate models used in IPCC Assessment Reports (Simpson et al. 2020). The first is to use a “storyline” approach (Zappa and Shepherd, 2017), matching model biases to European climate using a tool called Empirical Statistical Downscaling to constrain impacts over specific regions. The second is to evaluate the performance of the Seasonal Forecast Engine (https://klimavarsling.no/ (ekstern lenke) and https://www.climatefutures.no/en/home-en/ (ekstern lenke)) against the ensemble of prediction models that it aims to improve upon. We can use this information to obtain insight into mechanisms and drivers of persistent biases and the conditions under which they are exacerbated or mitigated. Depending on the skills and interest of the student, this Masters project could combine the two approaches, or focus on just one. 

This project is suitable for students who have mastered the topics taught in GEOF210, GEOF212 and GEOF313, and are enthusiastic about gaining physical understanding from quantitative analysis of the climate system. Experience with coding (Fortran, C, Matlab, python or similar) and some familiarity with the unix/linux environment are essential. 

Relevant literature: 

  • Chang, E. K. M., Guo, Y., & Xia, X. (2012). CMIP5 multimodel ensemble projection of storm track change under global warming. JGR, 117, D23118.
  • Simpson, I. R., Bacmeister, J., et al. (2020). An evaluation of the large‐scale atmospheric circulation and its variability in CESM2 and other CMIP models. JGR 125, e2020JD032835.
  • Woollings, T. (2010). Dynamical influences on European climate: An uncertain future. Philosophical Transactions of the Royal Society of London, 368, 3733–3756.
  • Zappa, G., & Shepherd, T. G. (2017). Storylines of Atmospheric Circulation Change for European Regional Climate Impact Assessment, Journal of Climate, 30, 6561-6577

 

 Arctic sea ice, climate and predictability 

Supervisors: Tor Eldevik, Marius Årthun

Motivating papers and popular info:

 

 An AI based climate emulator of precipitation in Norway

Supervisor: Asgeir Sorteberg 

Background:

 Climate emulators based on machine learning (ML) are tools designed to approximate the output of complex regional climate models (RCMs) much faster and with lower computational cost. New results have shown great promise for the use of such models.

Task: 

The student will have to gain knowledge on machine learning to use and further develop an existing climate emulator for Norway. Key research topics will be to what extent the emulator is able to mimic the results of a regional climate model, investigation into ways to improve the emulator and using various methods  to understand the results (explainable AI). 

Relevant literature: 

Doury et. al. (2024). On the suitability of a convolutional neural network based RCM-emulator for fine spatio-temporal precipitation. https://link.springer.com/article/10.1007/s00382-024-07350-8 (ekstern lenke)

Doury et. al. (2023) Regional climate model emulator based on deep learning: concept and first evaluation of a novel hybrid downscaling approach. https://link.springer.com/article/10.1007/s00382-022-06343-9 (ekstern lenke)

 

 Precipitation and insured damage

Supervisor: Asgeir Sorteberg

Background
Extreme precipitation and related damages is a major factor in the Norwegian society’s overall weather and climate vulnerability. Flood damage makes up around 40% of the insurance payments. Planning for and mitigating extreme precipitation impacts requires detailed knowledge of the historical severity, location and frequency of extreme precipitation and related flooding as well as  their relation to observed damage. Furthermore, understanding how extreme precipitation may change in the future is crucial in assessing future damage and risk .

Tasks
The student will analyze observations and the new  high-resolution hourly NORA3 atmospheric hindcast dataset from the Meteorological office (1980-present) to calculate return values for extreme events in Norway. Through the Norwegian Natural Perils Pool (Naturskadefondet), Norway is in the unique situation of having high quality nationwide daily statistics on municipality level on insured losses going back to 1980 (Finance Norway, 2018). The figure on the right shows an example of insurance payments per capita for the year 2011. The new hindcast dataset provides a great opportunity to link precipitation severity to damage. 
Relevant literature:

  • Cortès et. al (2018). The relationship between precipitation and insurance data for floods in a Mediterranean region (northeast Spain). Nat. Hazards Earth Syst. Sci., 18, 857–868, 201

 

 The sensitivity of short-term extreme precipitation to climate change in Norway - the case of Hans

Supervisor: Asgeir Sorteberg

Background
The relationship between atmospheric temperature and precipitation intensity  has been extensively investigated, mainly as a basis for predicting the effect of climate change on extreme rainfall. Findings are usually compared to the Clausius-Clapeyron (CC) relation, which gives an about 7% per °C increase of the water holding capacity of the atmosphere with temperature. Since precipitation feeds on atmospheric moisture, it is often argued that rainfall intensity should increase at about the Clausius Clapeyron rate, but it may be possibly faster or slower due to the additional effect of temperature on vertical fluxes and the effect of increased latent heat release. 

While daily and longer terms precipitation extremes have been extensively investigated in the literature, shorter term (hourly) changes have received less attention. In Norway short term extreme precipitation is mainly a summer phenomenon related to fronts with embedded convective activity and the intensity of such events has been observed to increase.  High-resolution so-called “pseudo global warming simulations” where a warming or cooling has been imposed on historic extreme events has previously been used to investigate the effect of global warming on individual weather events. 

Tasks
The student will run and analyze high-resolution simulations of the extreme weather Hans (2023) using the WRF model in order to investigate the effect of a +-2°C and +-4°C warming on the intensity of the event. The main aim is to estimate and understand the simulated precipitation changes through investigation into how the warming changes the vertical velocity, convective available potential, amount and types of hydrometeors, the rainout efficiency etc.
Relevant literature: 

 

 Temperature and convective activity dependence of observed rainfall extremes globally

Supervisor: Asgeir Sorteberg

Background

Several European studies have found evidence of increasing rainfall intensity with temperature (via increased atmospheric water content), but global analysis show that this is not a universal result. Recently investigations into the behavior of observed rainfall extremes in the US indicate that enhanced atmospheric convection has to be accounted for and not only temperature.  

Tasks
The student will analyze global precipitation observations and use atmospheric reanalysis to investigate to what degree the lack of the increasing rainfall intensity with temperature in certain regions of the world can be explained by changes in convective activity. The main aim is to better understand the complexity of extreme precipitation responses to global warming.

Relevant literature: 

 

 How will the partitioning between snowfall and rainfall  change globally  in the future?

Supervisor: Asgeir Sorteberg

Background

From a hazard risk standpoint, it makes a big difference if the precipitation comes as rainfall or snowfall. Change in global precipitation as emission of greenhouse gasses increase has been extensively studied, however, surprisingly little attention has been paid to the change in rainfall versus snowfall. 

Task
The student will analyze changes in snow and rainfall using an ensemble of global climate models. This cannot be done directly as these models suffer from large temperature biases that makes it difficult to estimate if the precipitation is snow or rain. They therefore have to be bias corrected against observations and paired with high resolution topographical data.  The main aim is to provide worldwide estimates of changes in snowfall and rainfall for different future emission scenarios. 

Relevant literature: 

 

 Precipitation interpolation using machine learning 

Supervisor: Asgeir Sorteberg

Background

The interpolation of point data to a gridded spatial field can be done in numerous ways. These gridded field are the backbone of weather and climate impact studies as they are often taken as “reality” by impact modelers (hydrological modelling, agricultural modelling etc.). In Norway the Se Norge (http://www.senorge.no/ (ekstern lenke)) temperature and precipitation data from the Norwegian Met. Institute form the input to hydrological modelling and bias corrections of climate models.  Any systematic errors in these gridded fields are translated into errors in the impact modelling. New machine learning techniques are rapidly making their way into environmental sciences and gives opportunities to do spatial interpolation in new ways.  Machine learning is a subset of methods within artificial intelligence that uses statistical techniques to give computers the ability to "learn” These are methods that up to now have not been applied much to spatial interpolation of climatic variables, but have been demonstrated to be powerful when large amounts of secondary variables that show correlation with the primary variable are available. The aim of the study is to use available Norwegian precipitation observations and do spatial interpolation based on machine learning techniques and compare the result to the conventional Se Norge data.

Task

Learn basic machine learning techniques. Perform machine learning based spatial interpolation of Norwegian daily precipitation data using additional topographic and atmospheric information to enhance the interpolation. Analyze the quality and compare it against existing gridded precipitation products like the Se Norge data.

Relevant literature: 

 

 The effect of climate change on alpine skiing

Supervisor: Asgeir Sorteberg

Background

Little is known about how future climate change will affect alpine skiing and winter tourism in Norway.  Contrary to cross country skiing, alpine sites often have the ability to produce artificial snow in order to extend and maintain their skiing season.  However, the technology for producing snow is costly and needs a certain threshold wet bulb temperature in order to work. Increased temperatures will reduce the season favorable for snow production and increase melting which will increase the demand for artificially produced snow in order to maintain skiing conditions. On the other hand, increased precipitation may lead to increased snowfall and reduce the need for snow production in high elevations. For alpine resorts to be profitable the increased cost of snow production has to be paid by the customers. Thus, analysis of future changes in snow production is of great interest for alpine resorts to optimize their future investments. 

Task

Integrate snow making procedures into a snowpack model. Model the change in snow production season as climate warms. Model the additional snow production needed to sustain skiing conditions and estimate the cost related to the increased snow production.

Relevant literature: 

 

 Analysis and interpretation of variations of ocean temperature around the Greenland coast during the last 120 years

Supervisor: Helge Drange

Background

Major changes in North Atlantic temperature and circulation strength/patterns have been observed since the instrumental measurements began more than 100 years ago. There are reasons to believe that these changes have significantly affected marine life in the ocean (Hatun et al. 2009), as well as contributing to the melting of the many glaciers flowing into the sea along the Greenland coasts.

Observed heating after 1990 has received a lot of attention recently. This period shows rapidly rising sea surface temperatures across large parts of the North Atlantic. Increasing temperatures have been particularly pronounced along the west coast of Greenland, as shown in Fig. 1 (right). The high sea temperatures, is likely an important component of the rapid, recent melting of the Greenland ice sheet.

It is also well known that there was an anomalously warm period in the North Atlantic region between 1920-1950 (see Fig. 1, left). During this period, there were particularly high catchments of the (warm water) Atlantic cod along the west coast of Greenland. There are reasons to believe that high sea temperatures at that time also affected Greenland's ice cap.

Task

This project will be focused on analyzing observed and modelled North Atlantic ocean climate for the period between around year 1900 until today; to quantify the marine warming along the west and east coasts of Greenland, with a dynamic assessment of how circulation and heat changes in the North Atlantic can affect the transport of warm and saline water towards (and from) the coasts of Greenland.

The candidate must familiarize her-/himself with existing literature related to the topic and must master/be willing to be engaged in the use of matlab, python or other analysis tools for analysing observations and model fields.

Relevant literature:

  • Holland D M., Thomas R.H., Younn B .d, Ribergaard M. H., Lyberth, B. (2008). "Acceleration of Jakobshavn Isbrae triggered by warm ocean waters". Nature Geoscience (ekstern lenke) 1 (10): 659–664. [fil (ekstern lenke)]
  • Hátún, H., Mark Payne, Gregory Beaugrand, Philip Chris Reid, Annebritt Sandø, Helge Drange, Bogi Hansen, Jan Arge Jacobsenand Dorete Bloch (2009, Large biogeographical shifts in the north-eastern Atlantic Ocean: From thesubpolar gyre, via plankton, to blue whiting andpilot whales, Prog. Oceanogr., doi:10.1016/j.pocean.2009.03.001 [file (ekstern lenke)]
  • He, Y.-C., H. Drange, Y. Gao and M. Bentsen (2016), Simulated Atlantic Meridional Overturning Circulation in the 20th century with ocean model forced by reanalysis-based atmospheric dataset, Ocean Modelling, doi:10.1016/j.ocemod.2015.12.011 [file (ekstern lenke)]
  • Lohmann, K., H. Drange, and M. Bentsen (2009), A possible mechanism for the strong weakening of the North Atlantic subpolar gyre in the mid-1990s, Geophys. Res. Lett., 36, L15602, doi:10.1029/2009GL039166. [file (ekstern lenke)]
  • Lohmann, K., H. Drange, and M. Bentsen (2008): Response of the North Atlantic subpolar gyre to persistent North Atlantic oscillation like forcing, Clim. Dyn., 10.1007/s00382-008-0467-6 [file (ekstern lenke)]
  • Hátún, H., A. B. Sandø, H. Drange, B. Hansen, and H. Valdimarsson (2005), Influence of the Atlantic Subpolar Gyre on the Thermohaline Circulation, Science, 309, 1841-1844 [article (ekstern lenke)]
  • Årthun, M., Eldevik, T., Viste, E., Drange, H., Furevik, T., Johnson, H. L., and Keenlyside, N. S. (2017), Skillful prediction of northern climate provided by the ocean, Nature Comm., doi:10.1038/ncomms15875 [file (ekstern lenke) and supplementary material (ekstern lenke)]

 

Extreme European rainfall - constraining future changes by combining dynamics and machine learning

Camille Li, Josh Dorrington, Robin Guillaume-Castel

Background:
Dynamical precursors describe the large-scale atmospheric circulation patterns that produce extreme European rain events. The framework was originally developed using ERA5 reanalysis data (Dorrington et al. 2023) and has since been applied to simulations from the Coupled Model Intercomparison Project phase 6 (CMIP6) generation of climate models to study how extreme rainfall and their drivers change in the future. The precursor analysis automatically partitions model biases into different source terms, allowing us to classify models into groups that must be used differently when assessing future changes in rainfall extremes.

Tasks:
This project will adapt and apply the precursor analysis for all available model simulations from CMIP7 to see how this new generation of climate models performs in terms of their ability to produce extreme rainfall. We will explore the benefits of including machine-learning tools in the framework, and whether these data-driven components can “learn” dynamics. 

Skills and knowledge: 

atmospheric dynamics, statistical analysis of climate data, python programming, basic shell scripting, working with climate model output

References:

Dorrington et al. 2023 : https://rmets.onlinelibrary.wiley.com/doi/full/10.1002/qj.4622 (ekstern lenke)