Lead Supervisor: Buwen Dong, Department of Meteorology, University of Reading and National Centre for Atmospheric Science

Email: b.dong@reading.ac.uk

Co-supervisors: Robin Clark, Met Office; Ed Hawkins, Department of Meteorology, University of Reading and National Centre for Atmospheric Science; Paul-Arthur Monerie, Department of Meteorology, University of Reading and National Centre for Atmospheric Science

Determining the time of emergence (ToE) of signals arising from anthropogenic climate change is a key issue in the subject of climate change attribution [1]. ToE is defined as the date for which the effect of climate change is projected to exceed the effects of the natural variations of the climate system (Figure 1). For regions highly vulnerable to climate variability, such as the tropics, this is especially relevant for decision makers in their development of adaptation and mitigation strategies for dealing with the changes coming in the next few decades.

Graph showing precipitation in the Shale between 1960 and 1999

Figure 1: Changes in central Sahel precipitation (10-20°N; 0-20°E) (mm/d) for each year and relative to the period 1960-1999. The effect of the anthropogenic activity (red line) is projected to become stronger than the effect of internal climate variability (grey shading) in the late 2050s (the ToE). Results are obtained with one climate model, with 40 ensemble members, and with a high emission scenario. Figure adapted from [11]. 

Up to now, studies focussing on ToE have typically used small ensembles, often of as few as 5 climate simulations [2, 3]. For robust results, however much larger ensembles are required. These are now available from international projects and perturbed physics ensembles such as those at the Met Office. We thus feel that the time is right for a student to undertake a project to produce some new and robust ToE conclusions using the most up-to-date simulations now available. A focus will be made on extreme events, on precipitation and temperature extremes, which have strong societal impacts.

The objectives of the project are as follows:

O1 – Quantifying the ToE and model uncertainty for extreme rainfall and temperature

O2 – Assessing the effects of differences in models’ physics in simulated ToEs

O3 – Addressing the role of the modes of sea surface temperature variability on ToEs

For the project we propose using single-model initial-condition large ensembles (SMILEs) that can provide robust estimates of both the effects of climate change and internal climate variability [4]. ToEs of changes in extreme rainfall and temperature will be assessed with the SMILEs provided by the 5th and 6th phases of the Coupled climate model intercomparison project (CMIP5 [5] and CMIP6 [6]) (~12 climate models), assessing O1.

Differences of ToE results from the SMILEs would likely be due to differences in models’ physics. Where possible, these will thus be compared to simulations from the Met Office’s HadGEM3 perturbed parameter ensemble (PPE) [7] using statistical methods developed in earlier work [2], as part of O2. The results of the PPE will allow developing an history-matching approach [8], that is a storyline based on different physics of the model, and that will help providing information for improving climate models and understanding future changes in rare extreme events.

Decadal to multi decadal modes of sea surface temperatures have strong effects on precipitation and temperature variability. The project would aim to include assessments of how these modes of variability can affect changes in extreme rainfall and temperature for a near-term projection (2021-2040, relative to 1995-2014), hereby affecting ToE of changes in precipitation and temperature. Since climate models underestimate the decadal variability in precipitation and temperature, we will correct statistically effects of the surface temperature on climate using calibration methods [9, 10] based on observation and define what the ToE could be in the real world. This will address O3.

The student will also test hypotheses by running sensitivity experiments at Reading, testing for instance, different patterns in sea surface temperature as obtained during the project (addressing O3) and in an AMIP-like mode, using the Met Office atmospheric model GA7.

Training Opportunities

The student will be offered courses in data analysis and computer programming in the computer science department and NCAS courses in atmospheric science. The student will also have the opportunity to perform simulations with a climate model, and to follow a training course on scientific modelling. The project offers an opportunity for the student to study at the Met Office, where they will experience a highly interdisciplinary, professional research environment. The SCENARIO provides the student with opportunities to develop their presentation skills, and to network at conferences, with further opportunities for discussion of their work provided by the multi-institutional and multi-disciplinary supervisory team.

Student Profile

This project would be suitable for students with a degree in Meteorology or a closely related environmental or physical science. The student must have strong analytical skills. During the project the student will be expected to develop the necessary computer programming and climate data analysis skills. Prior experience with programming software (e.g., Python) and the Unix/Linux environment would be highly beneficial.

Funding Particulars

This project has CASE support from the Met Office.

References

  1. Kirtman, B. et al. Climate change 2013: the physical science basis. Contrib. Work. Gr. I to fifth Assess. Rep. Intergov. panel Clim. Chang. 953–1028 (2013).
  2. Hawkins, E. & Sutton, R. Time of emergence of climate signals. Geophys. Res. Lett. 39, (2012).
  3. Lyu, K., Zhang, X., Church, J. A., Slangen, A. B. A. & Hu, J. Time of emergence for regional sea-level change. Nat. Clim. Chang. 4, 1006–1010 (2014).
  4. Deser, C., Phillips, A., Bourdette, V. & Teng, H. Uncertainty in climate change projections: the role of internal variability. Clim. Dyn. 38, 527–546 (2012).
  5. Taylor, K. E., Stouffer, R. J. & Meehl, G. A. An overview of CMIP5 and the experiment design. Bulletin of the American Meteorological Society vol. 93 485–498 (2012).
  6. Eyring, V. et al. Overview of the Coupled Model Intercomparison Project Phase 6 (CMIP6) experimental design and organization. Geosci. Model Dev. 9, 1937–1958 (2016).
  7. Yamazaki, K. et al. A perturbed parameter ensemble of HadGEM3-GC3.05 coupled model projections: part 2: global performance and future changes. Clim. Dyn. 56, 3437–3471 (2021).
  8. Hourdin, F. et al. Process-Based Climate Model Development Harnessing Machine Learning: II. Model Calibration From Single Column to Global. J. Adv. Model. Earth Syst. 13, e2020MS002225 (2021).
  9. Huang, X. et al. The Recent Decline and Recovery of Indian Summer Monsoon Rainfall: Relative Roles of External Forcing and Internal Variability. J. Clim. 33, 5035–5060 (2020).
  10. O’Reilly, C. H., Befort, D. J. & Weisheimer, A. Calibrating large-ensemble European climate projections using observational data. Earth Syst. Dyn. 11, 1033–1049 (2020).
  11. Monerie, P. A., Sanchez-Gomez, E., Pohl, B., Robson, J., & Dong, B. (2017). Impact of internal variability on projections of Sahel precipitation change. Environmental Research Letters, 12(11), 114003.