Møde xx – CPH 2017 Hydrological Modelling for Assessing Climate Change Impacts at Different Scales - www.hyacints.dk Climate models: From projections to predictions of climate change Jens Hesselbjerg Christensen Danish Meteorological Institute Seminar on Climate Change Impacts on Water Resources: 23-9-2008 - Geocenter
26
Embed
Climate models: From projections to predictions of climate ...hyacints.dk/xpdf/christensen-water-resources.pdfMøde xx – CPH 2017 Hydrological Modelling for Assessing Climate Change
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Møde xx – CPH 2017Hydrological Modelling for Assessing Climate Change Impacts at Different Scales - www.hyacints.dk
Climate models: From projections to predictions of climate change
Jens Hesselbjerg ChristensenDanish Meteorological Institute
Seminar on Climate Change Impacts onWater Resources: 23-9-2008 - Geocenter
Møde xx – CPH 2017Hydrological Modelling for Assessing Climate Change Impacts at Different Scales - www.hyacints.dk
Acknowledgements
PRUNDENCE and ENSEMBLES partners– EU Framework Programmes 5 & 6– http://prudence.dmi.dk 2001-2004– http://ensembles-eu.metoffice.com 2004-2009
Ole Bøssing Christensen, Martin Drews, Philippe Lucas-Picher, Fredrik Boberg - DMI
Møde xx – CPH 2017Hydrological Modelling for Assessing Climate Change Impacts at Different Scales - www.hyacints.dk
Outline
Uncertainties related to regional climate changeState-of-the-artUsing high resolution informationThe ENSEMBLES regional climate modeling product
Møde xx – CPH 2017Hydrological Modelling for Assessing Climate Change Impacts at Different Scales - www.hyacints.dk
Uncertainties in climate change projections
Uncertainty due to observational limitations
• use multiple means of validation
Uncertainty in future emissions
• use a range of SRES emissions scenarios
Natural variability
• use a number of different initial conditions (ensembles)
Uncertainty in the response of the climate system
• use a range of climate modelling systems
• AND/OR assess confidence in climate projections (better models
Need for a large-scale coordinated effort
Møde xx – CPH 2017Hydrological Modelling for Assessing Climate Change Impacts at Different Scales - www.hyacints.dk
Source: Hadley Centre
Global temperature rise
Signal
Noise
Møde xx – CPH 2017Hydrological Modelling for Assessing Climate Change Impacts at Different Scales - www.hyacints.dk
Møde xx – CPH 2017Hydrological Modelling for Assessing Climate Change Impacts at Different Scales - www.hyacints.dk
Møde xx – CPH 2017Hydrological Modelling for Assessing Climate Change Impacts at Different Scales - www.hyacints.dk
3B2+ARPEGE3
11B2+ECHAM4
1111B2+HadAM3H
1A2+ARPEGE3
11A2+ECHAM4
1111/113113A2+HadAM3H
UCMSMHIMPIKNMI/met.noICTPHCGKSSETHDMICNRM
50km time slice
PRUDENCE RCM
Møde xx – CPH 2017Hydrological Modelling for Assessing Climate Change Impacts at Different Scales - www.hyacints.dk
Schär et al. (2004)
Heat wave summer 2003
Møde xx – CPH 2017Hydrological Modelling for Assessing Climate Change Impacts at Different Scales - www.hyacints.dk
Schär et al. (2004)
Møde xx – CPH 2017Hydrological Modelling for Assessing Climate Change Impacts at Different Scales - www.hyacints.dk
xxxxxx….
xxxxxx….
xxxxxx….
xxxxxxxxx….
xxxxxxxxxGCM2
xxxxxxxxxGCM1
....….….….RCM3RCM2RCM125kmTrans.
ENSEMBLES idea
Møde xx – CPH 2017Hydrological Modelling for Assessing Climate Change Impacts at Different Scales - www.hyacints.dk
ENSEMBLES GCM-RCM Matrix
1
1950-2050*
CGCM3
1OURANOS**
18 (20)23264 (6)Total (1950-2050)
11950-2050*CHMI**
11950-2050*Met.No**
11950-2050*GKSS**
11950-2050C4I
11950-2050UCLM
21950-2050*1950-2050SMHI
11950-2050ICTP
11950-2050KNMI
11950-2050ETH
21950-2050*1950-2100DMI
21950-2050CNRM
21950-2050*1950-2100MPIMET
2 (4)1950-21001950-2100£METO-HC
Total numberNERSCCNRMIPSLMPIMETMETO-HCGlobal modelRegional model
Møde xx – CPH 2017Hydrological Modelling for Assessing Climate Change Impacts at Different Scales - www.hyacints.dk
Rockel domains
Møde xx – CPH 2017Hydrological Modelling for Assessing Climate Change Impacts at Different Scales - www.hyacints.dk
ENSEMBLES GCM-RCM Matrix
1
1950-2050*
CGCM3
1OURANOS**
18 (20)23264 (6)1960-2002
11950-2050*CHMI**
11950-2050*Met.No**
11950-2050*GKSS**
11950-2050C4I
11950-2050UCLM
21950-2050*1950-2050SMHI
11950-2050ICTP
11950-2050KNMI
11950-2050ETH
21950-2050*1950-2100DMI
21950-2050CNRM
21950-2050*1950-2100MPIMET
2 (4)1950-21001950-2100£METO-HC
Total numberNERSCCNRMIPSLMPIMETMETO-HCERA40
Møde xx – CPH 2017Hydrological Modelling for Assessing Climate Change Impacts at Different Scales - www.hyacints.dk
Christensen et al. GRL 2008
Møde xx – CPH 2017Hydrological Modelling for Assessing Climate Change Impacts at Different Scales - www.hyacints.dk
Christensen et al. GRL 2008
Møde xx – CPH 2017Hydrological Modelling for Assessing Climate Change Impacts at Different Scales - www.hyacints.dk
ENSEMBLES GCM-RCM Matrix
1
1950-2050*
CGCM3
1OURANOS**
18 (20)23264 (6)1960-2002
11950-2050*CHMI**
11950-2050*Met.No**
11950-2050*GKSS**
11950-2050C4I
11950-2050UCLM
21950-2050*1950-2050SMHI
11950-2050ICTP
11950-2050KNMI
11950-2050ETH
21950-2050*1950-2100DMI
21950-2050CNRM
21950-2050*1950-2100MPIMET
2 (4)1950-21001950-2100£METO-HC
Total numberNERSCCNRMIPSLMPIMETMETO-HCERA40
Møde xx – CPH 2017Hydrological Modelling for Assessing Climate Change Impacts at Different Scales - www.hyacints.dk
ENSEMBLES GCM-RCM Matrix
1
1950-2050*
CGCM3
1OURANOS**
18 (20)23264 (6)1960-2002
11950-2050*CHMI**
11950-2050*Met.No**
11950-2050*GKSS**
11950-2050C4I
11950-2050UCLM
21950-2050*1950-2050SMHI
11950-2050ICTP
11950-2050KNMI
11950-2050ETH
21950-2050*1950-2100DMI
21950-2050CNRM
21950-2050*1950-2100MPIMET
2 (4)1950-21001950-2100£METO-HC
Total numberNERSCCNRMIPSLMPIMETMETO-HCERA40
Møde xx – CPH 2017Hydrological Modelling for Assessing Climate Change Impacts at Different Scales - www.hyacints.dk
Making sense of it all?
Chris Ferro (UniReading) tought us in PRUDENCE: ANOVA (Analysis of variance)
Producing weights for each RCM/GCM pair
Møde xx – CPH 2017Hydrological Modelling for Assessing Climate Change Impacts at Different Scales - www.hyacints.dk
The ENSEMBLES RCM weights are based on the following specific metrics, deduced from the ERA40 driven simulations at 25 km
f1: large scale circulation based on a weather regime classification (PI Meteo France)f2: meso-scale signal based on seasonal temperature and precipitation analysis (PI ICTP)f3: probability density distribution match of daily and monthly temperature and precipitation analysis (PI DMI/SMHI)f4: extremes in terms of re-occurrence periods for temperature and precipitation (PI KNMI/Hadley)f5: trend analysis for temperature (PI MPI)f6: representation of the annual cycle in temperature and precipitation (PI CUNI)
Møde xx – CPH 2017Hydrological Modelling for Assessing Climate Change Impacts at Different Scales - www.hyacints.dk
Full RCM weight
∏=
=6
1i
n
iRCM fw i
Møde xx – CPH 2017Hydrological Modelling for Assessing Climate Change Impacts at Different Scales - www.hyacints.dk
Møde xx – CPH 2017Hydrological Modelling for Assessing Climate Change Impacts at Different Scales - www.hyacints.dk
GCM weights
The RCM weights introduced are basically independent on GCM performance as they are all based on the ERA40 experiments. The final weights in the RCM/GCM matrix, therefore needs to include element of assessing skill of the GCM as well.
• The quality of the GCMs used as driving models for the regional climate change production runs will also have a characteristic assigned weight,
• These precise weights are still to be deduced
Møde xx – CPH 2017Hydrological Modelling for Assessing Climate Change Impacts at Different Scales - www.hyacints.dk
Final system
Convert discrete data set into a continuous PDFs of climate change variables. • This will be done using a Gaussian Kernel algorithm
applied to the discrete dataset with the aim to take into account also the GCM/RCM model specific weights
Møde xx – CPH 2017Hydrological Modelling for Assessing Climate Change Impacts at Different Scales - www.hyacints.dk
Møde xx – CPH 2017Hydrological Modelling for Assessing Climate Change Impacts at Different Scales - www.hyacints.dk
Conclusions
Regional information provided by high resolution modeling adds credibility in projectionsCombining multiple simulations allows to assess uncertaintyElaborated model quality checks will reduce model spread – not necessarily uncertainty!