Working with climate model ensembles

1. Working with climate model ensembles PRECIS workshop, Reading, UK, 13th-17th May 2013

2. Explain how different types of climate models ensembles are used to explore different types of uncertainty Demonstrate how the ‘QUMP’ ensemble can be used with PRECIS to generate an ensemble of downscaled projections Demonstrate sub-selection of ensemble members Aims of this session

3. Working with climate model ensembles Table of Contents • What is an ‘ensemble’? • 3 types of ensemble • Generating ensembles of high-resolution simulations with PRECIS • Sub-selecting ensemble members for downscaling with PRECIS – an example from Vietnam

4. What is an ‘Ensemble’ Ensemble = “all the parts of a thing taken together, so that each part is considered only in relation to the whole.” Or, (in climate-modeling context)… “ the results from several models, so that each single model is considered only in context of the results of all of the models” 3 main types: • Initial conditions ensemble • Multi-model ensemble • Perturbed-physics ensemble

5. Revisiting the sources of uncertainties

6. What types of ensemble can we downscale with PRECIS? • Initial conditions ensemble • 3 - member HadAM3P ensemble • Useful for capturing climate change signal in regions/variables with variability on inter-annual or decadal timescales (‘noisy’ variables) • E.g. precipitation extremes. • Perturbed-Physics ensemble • 17-member ensemble of HadCM3 (HadCM3 Q0-16) ‘QUMP’ • We can select a sub-set of the 17 models in order to run a computationally ‘affordable’ experiment. • In 2013 (ish)…Multi-model Ensemble (CMIP5) • A carefully selected sub-set of CMIP5 models

7. 3 types of ensemble

8. Run 1, winter Run 1, summer Run 2, summer Run 2, winter Run 3, winter Run 3, summer -80 -40 -20 -10 -5 5 10 20 40 80 Change (%) 1.) Initial Conditions ensembles • Internal or natural variability • 1 model run more than once gives slightly different responses • Which changes are ‘signal’, and which changes are ‘noise’..? • I.e. which changes are reliable? • Internal variability greatest issue when we are looking for: • (a) Spatial or temporal details (e.g. extremes) • (b) Variables with strong multi-decadal variability

9. Mean, winter Mean, summer Top 5%, summer Top 5%, winter Top 1%, winter Top 1%, summer -10 -5 -2 -1 0 1 2 5 10 Initial conditions Ensembles • Coloured areas where signal is discernable from noise, and changes are ‘reliable’. • White areas where signal is not clear, and changes are ‘unreliable’. • Can discern significant changes over much of Europe in winter and parts of Europe in summer, but signal is still unclear in many areas, particularly in extremes. Kendon, E. J., D. P. Rowell, R. G. Jones, and E. Buonomo, 2008: Robustness of future changes in local precipitation extremes. J. Climate, doi: 10.1175/2008JCLI2082.1.

10. 2.) Multi-model ensembles • IPCC CMIP3 ensemble: modelling centres submitted results from equivalent simulations to allow inter-model comparisons for fourth assessment report • Models from different modelling centres around the world use different structural choices in model formulation → different future climate projections: • ‘Structural uncertainties’ • ‘Disagreements’ between models can be large i.e. between an overall increase or decrease in rainfall in a region

11. Consensus in CMIP3 models on rainfall change White regions= Models do not agree on increases, decreases or ‘no change’. McSweeney and Jones, In Press

12. Rainfall change: CMIP3 • Combination of pattern and some sign differences lead to lack of consensus

13. 3.) Perturbed-Physics Ensembles • An alternative route to exploring GCM uncertainty • Many processes in GCMs are ‘parameterised’ • Parameterisations represent sub-gridscale processes • Values of parameters are unobservable and uncertain • Explore model uncertainty by varying the values of the parameters in one model

14. The QUMP 17-member perturbed physics ensemble • 17 members (HadCM3 Q0..Q16) • ‘parameter space’ = range/combinations of plausible values of parameters • Range of plausible values of parameters determined according to expert opinion • 17 models sample ‘parameter space’ systematically Quantifying Uncertainty in Model Projections

15. Structural and parameter uncertainties Structural Uncertainty (e.g. IPCC CMIP3 multi-model ensemble) Parameter Uncertainty (e.g. QUMP perturbed physics ensemble)

16. Rainfall change: Hadley QUMP • Again significant range of different projected changes • Similar range and behaviour to IPCC models?

17. Generating ensembles of high-resolution simulations with PRECIS

18. Ensemble regional prediction: Issues and approaches • The major uncertainties in the simulated broad-scale climate changes come from GCMs • Global climate models provide information which is often too coarse for applications thus downscaling is required • To provide the best possible information currently available we can: • Consider a range of simulated climate changes from the global climate models • Downscale these to provide information relevant to applications which accounts for this range of possible future climate changes • Also provides a set of internally-consistent inputs for impacts models

19. PRECIS+QUMP • PRECIS version 1.8.2 and later • 17 member GCM driving data is ready • Users should use a sub-set of these as it is inefficient to downscale all 17!

20. Process for requesting ‘QUMP’ boundary data from the Hadley Centre 1.) Email us to let us know that you are interested in using the QUMP ensemble with PRECIS, and let us know what you are planning to use the data for. 2.) Download the mean GCM fields from BADC http://badc.nerc.ac.uk/browse/badc/hadcm3/data/PRECIS/ Use these fields to choose a model subset which validates well, and spans a wide range of future outcomes 3.) Email us a 1-2 page summary of your analysis of the GCM fields, and your selected ensemble members 4.) If we agree that your selection is based on good criteria, we’ll then send your boundary data, and you can begin your runs

21. Sub-selecting ensemble members for downscaling with PRECIS – an example from Vietnam

22. Selecting a subset of the ‘QUMP’ ensemble Choose members that: • do well in validation, and • sample the range of future outcomes • Both can be analysed using the GCM monthly mean fields available from BADC • Finally… (c)How does the spread of futures results indicated by QUMP compare with CMIP3 models…? • i.e. does the range of outcomes from our small ensemble capture those indicated by CMIP3 too?

23. Example of Model sub-selection: Vietnam Criteria for selection • Validation • Selected models should represent Asian summer monsoon (position, timing, magnitude), and associated rainfall well, as this is key process • Future • Magnitude of response: greatest/least regional/local warming, greatest/least magnitude of change in precipitation • Characteristics of response • Direction of change in wet-season precipitation (increases and decreases) • Spatial patterns of precipitation response over south-east Asia • Response of the monsoon circulation

24. Validation:Monsoon Onset • Monsoon flow has some systematic error – a little too high, but timing (and position) of features is very good. • All do a reasonably good job at simulating rainfall in the region • Those that best represent the characteristics of the monsoonal flow don’t necessarily also best represent the local rainfall… • No reason to eliminate any models on grounds of validation

25. Range of Future changes

26. Spatial patterns of future changes (precip) ←Typical Atypical →

27. Recommended QUMP members for this region • HadCM3Q0 – The standard model • HadCM3Q3 – A model with low sensitivity (smaller temperature changes) • HadCM3Q13 – A model with high sensitivity (larger temperature changes) • HadCM3Q10 – A model that gives the driest projections • HadCM3Q11 – A model that gives the wettest projections • Including Q10 and Q13 means that we also cover models which characterise the different spatial patterns of rainfall change, and different monsoon responses.

28. QUMP vs CMIP3 responses

29. Summary • Using an ensemble of models gives us an idea of how confident we can be in modelling outcomes • Warning: model consensus does not necessarily mean certainty! • Need to have confidence in the underlying mechanisms. • Can use ensemble(s), or sub-sets of ensembles, to try to capture as wide a range of plausible outcomes as we can. • Sub-selection strategies will become increasingly important in coming years as modelling centres make boundary data from CMIP5 models available

30. Questions and answers