P1254325874WBDtF

1

• CRU CRANIUM work
• Techniques for the construction of
  probabilistic high-resolution climate scenarios
• Discussion on the use of information on
  probabilities and extremes in climate impacts
  analysis

http//www.cru.uea.ac.uk/projects/cranium
2

• First point location daily probabilistic
  scenarios of extremes for UK
• Based on state-of-art European RCMs
• Demonstration that weather generator can be
  linked with RCM output in a probabilistic
  framework

3
1 Incorporating RCM uncertainty the CRANIUM
methodology
Change in T P 2071-2100
n runs
Weather generator
RCM 1
Weather generator
RCM 2
The CRU weather generator

• developed for BETWIXT
• mean and s.d. constrained by RCM changes

4
1 Incorporating RCM uncertainty the CRANIUM
methodology
PRUDENCE simulations
Control 1961-1990 Future Scen A2 2071-2100 Future Scen B2 2071-2100
HIRHAM (x2) HadRM3P CHRM CLM REMO RCAO (x2) PROMES RegCM RACMO Arpege (x2) HIRHAM (x2) HadRM3P CHRM CLM REMO RCAO (x2) PROMES RegCM RACMO Arpege (x2) HIRHAM HadRM3P RCAO (x2) PROMES RegCM Arpege (x2)
5
1 Incorporating RCM uncertainty the CRANIUM
methodology
Station Lat Lon
1 Paisley 2 Eskdalemuir 3 Ringway 4 Bradford 5 Coltishall 6 Hemsby 7 Elmdon 8 Heathrow 9 Gatwick 10 Yeovilton 55.85 55.32 53.85 53.82 52.77 52.68 52.45 51.48 51.15 51.00 -4.43 -3.20 -2.28 -1.77 1.35 1.68 -1.73 -0.45 -0.18 -2.63
Which climate extremes are we interested in? Heatwave duration Number of hot days Number of cold days Longest dry spell length Fraction of rainfall from intense events..
6
Building Knowledge for a Changing Climate
Probabilistic scenarios for Ringway (Manchester)
2080s, Medium-high emissions scenario
Hot days
Fraction intense events
Longest dry period
Black baseline, Coloured lines RCMs, Red
ensemble average
7
http//www.cru.uea.ac.uk/projects/cranium

• Tables of figures
• Non-technical background information, including
  assumptions etc.
• Descriptive summary of results
• Briefing note and questions
• Powerpoint presentations
• Alternative forms of presentation
• PDFs, CDFs, percentiles, classes, thresholds

8
Probabilistic scenarios on stage!
9
New probabilistic scenarios for Coltishall in the
2080s assuming a medium-high emissions scenario
hotter summers
fewer frosts
longer droughts
????
10
New probabilistic scenarios for Coltishall in the
2080s assuming a medium-high emissions scenario

• 52 chance of 5-15 fewer cold nights in spring
• 84 chance of 2-5C rise in maximum summer
  temperature
• 87 chance of up to 15 day longer summer dry
  period
• 50 chance autumn rainfall increasing by up to
  100
• 90 chance of average summer day-time
  temperatures as high or higher than 1976 -
  and a 52 chance that summers will be as dry or
  dryer than 1976

hotter summers
fewer frosts
longer droughts
????
11
Downscaling in ENSEMBLES

• RT2B is working on construction of probabilistic
  regional scenarios
• Using dynamical and statistical downscaling
• Co-ordinated by Clare Goodess and Daniela Jacob

http//www.ensembles-eu.org
12
Timescales Climate change (ACC) Seasonal-to-decad
al (s2d)
Spatial scales Global climate models Regional
climate models WP2B.1 Statistical downscaling
WP2B.2
Forcing Emissions scenarios (SRES) Reanalysis Per
turbed physics
Construction of probabilistic scenarios WP2B.2
WP2B.3 Monte Carlo, Bayesian, REA, weighting,
scaling, etc. etc.
Probabilistic regional climate scenarios and tools
13
Downscaling issues being addressed in ENSEMBLES
RT2B

• Shift of emphasis to tools
• Adapting methods for probabilistic scenario
  construction (weighting, scaling, ensemble
  averaging etc)
• Integration of climate change and
  seasonal-forecasting timescales
• Synergistic use of dynamical and statistical
  downscaling

See RT2B website at http//www.ensembles-eu.org