Climatic Research Unit : Research and Projects : CRANIUM

View of a Skull (Leonardo da Vinci) View of a Skull (Leonardo da Vinci)

CRANIUM

Climate change Risk Assessment: New Impact and Uncertainty Methods


Examples of the CRANIUM probabilistic climate scenarios for UK point locations

UEA Project leader: Clare Goodess, Climatic Research Unit, School of Environmental Sciences, University of East Anglia (UEA), Norwich, NR4 7TJ c.goodess@uea.ac.uk

UEA team:  Colin Harpham, Phil Jones, Mike Salmon, Craig Wallace

CRANIUM is one of the research projects in the EPSRC/UKCIP Building Knowledge for a Changing Climate (BKCC) programme.

CRANIUM has developed  new methodologies for analysing uncertainy and making robust risk-based decisions for infrastructure design and management in the face of climate change.  New methods for analysing uncertainties in key climate variables, for example rainfall and temperature, have been developed and applied. CRANIUM has demonstrated how these climate changes impact upon the performance of infrastructure systems, including railways and hydro-electric power. It has developed methods to enable this type of analysis to be carried out faster, more robustly and comprehensively.  Finally, CRANIUM has investigated with stakeholders how, in the light of these insights, decision making about operation of, or investment in, infrastructure systems can be managed or modified to reflect potential climate change impacts and, in particular, the uncertainties surrounding them.

CRANIUM is managed by Professor Jim Hall at the University of Newcastle, and also involves teams from the University of East Anglia (UEA) and University of Leeds.  The project as a whole finishes in April 2007, while the UEA work formally finished in May 2006.

This web page describes the contribution of the Climatic Research Unit (CRU) at UEA to CRANIUM and provides public access to probabilistic scenarios for 10 UK locations.  The CRU work focused on CRANIUM tasks 1a and 1c:

Task 1a achievements

Task 1c achievements

BKCC is a portfolio of research projects looking at how climate change will affect different aspects of the built environment. It addresses some of the research needs of decision-makers dealing with buildings, transport and utilities infrastructure in responding to the impacts of a changing climate in the UK. It is promoted jointly by the Engineering and Physical Sciences Research Council (EPSRC) and the UK Climate Impacts Programme (UKCIP).

The construction, communication and use of probabilistic climate change scenarios: summary of the CRU contribution to CRANIUM

The first daily probabilistic scenarios of weather extremes for point locations in the UK have been developed.    These scenarios are based on state-of-the-art European regional climate model (RCM) output from the PRUDENCE project and the daily weather generator developed by UEA during the BETWIXT project – thus demonstrating that a weather generator can be successfully linked with RCM output in a probabilistic framework.

The CRANIUM climate change scenarios for 10 UK locations (selected for their relevance to the BKCC programme) are publicly available here.  They are provided for the 2080s and the UKCIP Medium-high emissions scenario (equivalent to the IPCC A2 scenario). They are presented in a number of different formats (histograms, probability and cumulative density functions, percentiles, class probabilities and event thresholds) to suit the needs of a range of different users and to illustrate the range of output formats under consideration for the next national UK climate change scenarios (UKCIPnext). The event thresholds that have been used relate to ‘memorable’ extreme years. It is shown, for example, that there is a 98% probability that average summer day-time temperatures at Gatwick will be as high or higher than experienced in 1976 by the 2080s - and a 55% probability that summers will be as dry or dryer than 1976. 

While probabilistic scenarios are the best way of making use of new information emerging from climate modelling, dealing with probabilistic climate projections will involve more work, with a steep learning curve for both developers and users.  A series of key questions has been identified by UEA from discussions with CRANIUM researchers and stakeholders. These questions will form the basis for ongoing discussions and further research in the EPSRC Sustaining Knowledge for a Changing Climate (SKCC) project.

Dissemination and communication of CRU CRANIUM results

This web site provides the main point of dissemination for our results

Three journal articles are in preparation:

The CRANIUM probabilistic scenarios for Coltishall, Norfolk were used in a short play and poster presentation as part of the 'Beer today champagne tomorrow' event held in Norwich during the 2006 BA Festival of Science

CRANIUM results will be presented during the SKCC workshop on the use of probabilistic climate scenarios in impacts assessment and adaptation studies to be held at UEA in November 2006.  They will also be reported at the SKCC dissemination conference (to be held at the Institute of Civil Engineers, London in January 2007) and the related conference publication

Meeting presentations

Overview of UEA CRANIUM results: presented by Clare Goodess at the final CRANIUM project meeting, 3 May 2006, Norwich

Working with probabilistic climate change scenarios: the challenges for the users and the developers: presented by Clare Goodess at the BKCC Integrating Framework (IF) meeting, 17 November 2005, Newcastle and the BKCC Stakeholder Forum, 21 November 2005, London.
Briefing note and key questions to accompany this presentation
Questions and discussion from the IF meeting

Exploring inter-model uncertainties in scenarios of UK climate extremes using a daily weather generator: presented by Craig Wallace at the Royal Meteorological Society Conference 2005, 11-16 September 2005, Exeter

An introduction to climate scenario uncertainties

Uncertainties in climate scenarios are related to:

These sources of uncertainty were recognised a number of years ago and are sometimes referred to as the cascade or explosion of uncertainty.  Techniques for handling them have also been identified and recommended.  Inter-scenario variability, for example, can be represented by using multiple emissions scenarios and the pattern scaling technique.  Intra-model variability can be explored using intra-model ensembles, in which the same model is run a number of times, with a different starting point or parameter values each time. 

Despite the recognition of these uncertainties and ways of representing them, in practice these techniques have not yet been widely or comprehensively used, particularly with respect to regional climate scenarios. The UKCIP02 scenarios, for example, are based on four emissions scenarios and a single suite of climate models, hence they only reflect a small part of the uncertainty range.

The growing availability of large climate model ensembles permits a more comprehensive approach to the assessment of uncertainty.  The size of these ensembles reflects the complexity and computing time required to run the different types of climate model.  Multi-model ensembles encompassing dozens of regional climate model runs were performed during the European Union-funded PRUDENCE project for example.  Larger ensembles are possible for coarser scale global climate models - encompassing hundred of runs in the case of the Hadley Centre ‘perturbed physics’ QUMP (Quantifying Uncertainties in Model Predictions) simulations, and thousands of runs in the case of the climateprediction.net initiative.

These large ensembles permit the construction of probabilistic climate scenarios, which are now acknowledged by the climate modelling and scenario community as the way forward.  The UKCIPnext scenarios will, for example, be probabilistic.  This approach is also being pursued in the European Union-funded ENSEMBLES project.  ENSEMBLES work on probabilistic regional scenarios includes, for example, assessment of weighting, pattern-scaling and ensemble-averaging techniques.

Exploring inter-model uncertainties in scenarios of UK weather extremes - the CRANIUM probabilistic scenarios

As part of the CRANIUM project, the UEA team has explored inter-model scenario uncertainties using output from 10 different European regional climate models (RCMs).  The RCM runs were undertaken as part of the PRUDENCE project. Most of the RCMs were forced by the Hadley Centre global climate model (GCM) - the model used in UKCIP02, but three of the RCMs were also forced by a second global model  (ECHAM4) - giving a total of 13 RCM runs, all for the IPCC SRES A2 emissions scenario (equivalent to the UKCIP02 medium-high scenario).  Changes in mean temperature and precipitation, together with changes in their variability, were taken from each RCM run for the grid square nearest to each UK location of interest, and used to perturb the parameters of the CRU weather generator developed for the BETWIXT project (see BETWIXT Technical Briefing Notes 1 and 4).  For each of the 13 RCM runs, the weather generator was run 100 times.  Ten seasonal indices of mean and extreme temperature and rainfall  (see table below) were then constructed from the daily time series.

The scenario results are presented below, using a number of different formats to suit a range of different users.  These formats reflect the choices offered to potential users as part of the online consultation for the UKCIPnext scenarios.

It is important to note the following features of the CRANIUM probabilistic scenarios:


The CRANIUM probabilistic scenarios: figures and tables

The following acronyms are used to identify the ten seasonal indices of mean and extreme temperature and rainfall in the figures and tables below.  All scenario  results are for the 2080s (2071-2100).

Index Description User-friendly name
txav Mean Tmax Average maximum temperature
tnav Mean Tmin Average minimum temperature
tav Mean Tmean Average temperature
txhw90 Heat Wave Duration (days) Longest heatwave
txf90 % days Tmax > 90th percentile Number of hot days
tnf10 % days Tmin < 10th percentile Number of cold nights
tnf90 % days Tmin > 90th percentile Number of warm nights
pav Mean climatological precipitation (mm/day) Average daily rainfall
pxcdd Max number of consecutive dry days Longest dry period
pfl90 fraction of total rainfall from events > long-term 90th percentile Heavy rainfall proportion

The percentile threshold values calculated from observed data which were used to calculate these extremes are available here (Excel spreadsheet).

Figures

Three sets of figures for 10 stations and 10 indices/variables are provided, as described below.  Each figure is available as a gif for immediate viewing, and as a pdf and eps for higher quality.

Histograms

In this set of figures, the weather generator output is plotted as histograms, with the number or magnitude of events on the horizontal axis and the probability of the event occurring in any one season on the vertical axis. Each individual histogram is constructed using 3000 values (i.e., 30 years x 100 weather generator runs).    

The heavy black line in both left- and right-hand panels shows the weather generator results based on observations for the reference period 1961-1990.  In the left-hand panel, each of the coloured lines represents weather generator results for the 2080s based on a different RCM run.

In the right-hand panel, the red line shows the ensemble average, i.e., the average of the coloured lines from the left-hand side.

For the A2 emissions scenario, 13 different RCM runs were used – hence there are 13 coloured lines on the left-hand side.  Histograms are also shown for the B2 emissions scenario – for which 11 runs were available (note that histograms only are provided for the B2 scenario – no further analysis of these scenario results has been undertaken).

Average temperatureTemperature extremesAverage
rainfall
Rainfall extremes
txav tnav tav txhw90 txf90 tnf10 tnf90 pav pxcdd pfl90
Bradford
53.82N 1.77W
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
Coltishall
52.77N 1.35E
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
Elmdon
52.45N 1.73W
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
Eskdalemuir
55.32N 3.2W
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
Gatwick
51.15N 0.18W
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
Heathrow
51.48N 0.45W
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
Hemsby
52.68N 1.68E
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
Paisley
55.86N 4.43W
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
Ringway
53.35N 2.28W
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
Yeovilton
51.00N 2.63W
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript
A2:GIF previewPortable Document FormatEncapsulated PostScript
B2:GIF previewPortable Document FormatEncapsulated PostScript

Interpretation of the histograms

As an illustrative example, we consider results for the number of hot days (txf90 – 4th column above) at Elmdon (Birmingham - 3rd row above), for the A2 scenario. 

For all RCM runs, the histograms shift to the right – indicating an increased probability of more frequent hot days in the 2080s for the A2 emissions scenario.  This shift to the right is greater in summer than other seasons.  The left-hand panel indicates the spread across results based on individual RCM runs.  Two RCM runs give larger changes than the others – these are forced with a different global climate model to the others (i.e., ECHAM4 rather than HadDAM3). This indicates the importance of considering inter-model uncertainty in the forcing model as well as the downscaling model.

As well as being shifted to the right, the future histograms are broader, indicating the range of uncertainty in the projections.  If we just consider the mode or centre point of the distribution, however, the ensemble average indicates an increase in summer from around 5 hot days (i.e., days with Tmax > 24.3°C in this case) in the reference period to about 40 days in the 2080s.  This represents a major increase in the risk of occurrence of hot days.  For comparison, the results for Elmdon produced as part of the BETWIXT project (based on the Hadley Centre RCM – a slightly different version of which is used here), indicate an average of 45.6 summer hot days. 

Probability Density Functions (PDFs) - A2 scenario changes

In this set of figures, changes for the A2 emissions scenario calculated from weather generator output are  plotted as Probability Density Functions (PDFs).  The magnitude of the change is shown on the horizontal axis and the density (frequency) of the change occurring in any one season on the vertical axis. 

In the left-hand panel, each of the coloured lines represents scenario changes for the 2080s based on a different RCM run.  Each individual PDF is constructed using 3000 paired changes (i.e., 30 years x 100 weather generator runs).  The changes are calculated as the difference between each of the 3000 values for the 2080s and the equivalent control-period value.      

In the right-hand panel, the red line shows results for all 39,000 paired changes (i.e., 30 years x 100 weather generator runs x 13 RCM runs).  Thus it represents the ensemble average.

In order to estimate probabilities of change from these plots, it is necessary to integrate the area under the part of the curve of interest.  However, these PDFs do provide a clear visual picture of the most likely change (indicated by the peak or mode of the distribution) and the uncertainty (indicated by the shape and spread of the distribution). 

Average temperatureTemperature extremesAverage
rainfall
Rainfall extremes
txav tnav tav txhw90 txf90 tnf10 tnf90 pav pxcdd pfl90
Bradford
53.82N 1.77W
GIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScript
Coltishall
52.77N 1.35E
GIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScript
Elmdon
52.45N 1.73W
GIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScript
Eskdalemuir
55.32N 3.2W
GIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScript
Gatwick
51.15N 0.18W
GIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScript
Heathrow
51.48N 0.45W
GIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScript
Hemsby
52.68N 1.68E
GIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScript
Paisley
55.86N 4.43W
GIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScript
Ringway
53.35N 2.28W
GIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScript
Yeovilton
51.00N 2.63W
GIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScript

Interpretation of the PDFs

As an illustrative example, we consider results for average daily rainfall (pav - 7th column above) at Gatwick (5th row above).

The importance of driving GCM uncertainty is evident – particularly in spring, when the two ECHAM4-driven RCMs give more negative changes (i.e., less rainfall), while the RCMs driven by HadAM3 indicate more positive changes (i.e., more rainfall).

It is also evident that none of the ensemble-average PDFs for Gatwick rainfall are normally distributed - but are skewed, with very long tails to the right-hand side.

In summer, the ensemble-average PDF is mainly located on the negative part of the horizontal axis – increasing confidence that the rainfall change will be negative.  In spring, however, the PDF is centred close to zero encompassing negative changes, but extending further to the right-hand side (thus indicating a higher probability that the change will be positive rather than negative).

In contrast, the average temperature changes for Gatwick (tav, third column above) are more normally distributed – although the spring PDF again appears stretched to the right – again because of uncertainties due to the driving GCM.  The possible range of change (i.e., the spread of the PDF) is, however, somewhat narrower in spring than in winter.  

Cumulative Density Functions (CDFs) - A2 scenario changes

This set of figures is exactly the same as for the PDFs, except that here, cumulative density functions (CDFs) are plotted.

This allows the cumulative probability of particular changes to be easily read.   Where the changes are positive, the cumulative probability for a change greater than any particular value on the horizontal axis is read as 1 minus the value on the vertical scale.

Average temperatureTemperature extremesAverage
rainfall
Rainfall extremes
txav tnav tav txhw90 txf90 tnf10 tnf90 pav pxcdd pfl90
Bradford
53.82N 1.77W
GIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScript
Coltishall
52.77N 1.35E
GIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScript
Elmdon
52.45N 1.73W
GIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScript
Eskdalemuir
55.32N 3.2W
GIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScript
Gatwick
51.15N 0.18W
GIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScript
Heathrow
51.48N 0.45W
GIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScript
Hemsby
52.68N 1.68E
GIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScript
Paisley
55.86N 4.43W
GIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScript
Ringway
53.35N 2.28W
GIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScript
Yeovilton
51.00N 2.63W
GIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScriptGIF previewPortable Document FormatEncapsulated PostScript

Interpretation of the CDFs

Taking again some of the illustrative examples from above, the CDFs indicate:

Tables

In the two tables below, probability values derived from the figures above are shown.

The CDF percentiles give lower (10th percentile), mean (50th percentile) and upper (90th percentile) estimates of changes derived from the PDFs/CDFs shown above.

The class probabilities give estimates of the probability of changes lying within fixed classes or bins. About 10 classes are used for each index, with the same ranges or bin sizes used for all stations, depending on the variable considered:

CDF
percentiles
Class
probabilities
BradfordComma-Separated ValuesComma-Separated Values
ColtishallComma-Separated ValuesComma-Separated Values
ElmdonComma-Separated ValuesComma-Separated Values
EskdalemuirComma-Separated ValuesComma-Separated Values
GatwickComma-Separated ValuesComma-Separated Values
HeathrowComma-Separated ValuesComma-Separated Values
HemsbyComma-Separated ValuesComma-Separated Values
PaisleyComma-Separated ValuesComma-Separated Values
RingwayComma-Separated ValuesComma-Separated Values
YeoviltonComma-Separated ValuesComma-Separated Values

Interpretation of the percentile/class probability tables

If the 10th, 50th and 90th percentile changes for a particular station/season/variable have the same sign (positive - an increase, or negative - a decrease), we can have more confidence in the projected direction of change.

For example, the 50th percentile value for the change in number of summer hot days (txf90) at Gatwick is +39 days, while the 10th and 90th percentile values are +25 and +56 days respectively.  Thus it is reasonable to conclude that we have high confidence that the frequency of hot days in summer will increase at Gatwick.

In the case of average daily summer rainfall (pav) at Gatwick, the 50th percentile change is -54%.  The 10th percentile change is -79% (25% greater than the modal change, while the 90th percentile change is only -5% (49% less than the modal change) – reflecting the skewed nature of the PDFs and CDFs.  For spring, the 50th percentile change is +7%, while the 10th and 90th percentile changes are -40% and +91% respectively – indicating uncertainty in both the sign and magnitude of change.

The class probabilities provide perhaps the easiest way of quantifying scenario changes.  Taking a new example of Coltishall, they indicate, for example:

Probabilities of threshold events

Finally, probabilities associated with selected 'memorable' extreme years are presented here.

The upper half of the table shows the probability of a change as large or larger than the selected event occurring in the 2080s - while the bottom half of the table shows the seasonal anomaly (calculated from observed data) for the selected event (i.e., the magnitude of the threshold event).

Four temperature events defined by tav anomalies are considered:

Two rainfall events defined by pav anomalies are considered:

The most memorable of these events is the summer of 1976, for which the table indicates:

Further reading on uncertainty and probabilistic climate change scenarios

Links

BETWIXT http://www.cru.uea.ac.uk/cru/projects/betwixt/
BKCC http://www.ukcip.org.uk/resources/publications/pub_dets.asp?ID=74
http://www.ukcip.org.uk/resources/sector/projectsdets.asp?sector=1&project_ref=5
Climateprediction.net http://www.climateprediction.net/
CRANIUM http://www.ncl.ac.uk/cranium/
ENSEMBLES http://www.ensembles-eu.org/
EPSRC http://www.epsrc.ac.uk/default.htm
PRUDENCE http://prudence.dmi.dk/
SKCC http://www.k4cc.org/
UKCIP  http://www.ukcip.org.uk/
UKCIPnext http://www.ukcip.org.uk/scenarios/ukcipnext/default.asp


Last updated: January 2007 - Clare Goodess, Colin Harpham and Mike Salmon