we get from climate models with higher resolution Gustav Strandberg Paula Gonzalez Galia Guentchev and Gerard van der Schrier Swedish Meteorological and Hydrological Institute ID: 1034395
Download Presentation The PPT/PDF document "What new information do" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
1. What new information do we get from climate models with higher resolution?Gustav Strandberg*, Paula Gonzalez, Galia Guentchev and Gerard van der Schrier *Swedish Meteorological and Hydrological Institute
2. Why is model resolution important?Topography in different models (meter above sea level)Primavera low res. ~100 km Primavera high res. ~50 km CORDEX high res. ~12 km
3. High resolution is expensive50 km
4. High resolution is expensive25 km50 km
5. High resolution is expensive12.5 km25 km50 kmDoubling the resolution (e.g. 50 to 25 km) means 8 times the computer power
6. Some important abbreviationsModel families:CMIP5 = Global climate models used in the last IPCC assessment PRIMAVERA = Global models used in the PRIMAVERA projectCORDEX = Regional climate models of higher resolution
7. By the looks of it …Low res.High res.MPI-ESM-1-2 CNRM-CM6-1 HadGEM3-GC31 EC-Earth3 ECMWF-IFSEOBSPrmean (mm/day)Precipitation benefits from higher resolution
8. By the looks of it …Low res.High res.MPI-ESM-1-2 CNRM-CM6-1 HadGEM3-GC31 EC-Earth3 ECMWF-IFSEOBSEspecially extreme precipitation benefits from higher resolutionrx1day (mm)
9. Individual models over a specific regionBRI = British IslesNo systematic differences between high and low res. for temperatureTemperature
10. Individual models over a specific regionBRI = British IslesNo systematic differences between high and low res. for temperatureSome tendency for more precipitation in high res. Higher maximum precip.TemperaturePrecipitation
11. PRIMAVERA in perspective of CMIP5 and CORDEXHigh res.Low res.ScandinaviaPrecip.
12. PRIMAVERA in perspective of CMIP5 and CORDEXIncreasing precipitation with higher resolution is a robust feature over different ensemblesHigh res.Low res.High res.Low res.ScandinaviaBritish IslesPrecip.Precip.
13. Precipitation intensitiesMedian99th percentileX Low res.X High res.
14. Precipitation intensitiesMedianMedian99th percentile99th percentileX Low res.X High res.X High res. – low res.Higher resolution mainly changes precipitation in the high end of the distribution (more extreme precipitation).
15. Precipitation intensitiesThe Alps
16. Precipitation intensitiesHigher resolution mainly changes precipitation in the high end of the distribution (more extreme precipitation).The AlpsMid Europe
17. Precipitation intensitiesHigher resolution mainly changes precipitation in the high end of the distribution (more extreme precipitation).The AlpsMid Europe
18. Let’s look at extra-tropical cyclones (ETCs)
19. ETC densitiesChange in storm track bias (as compared to ERA Interim) between CMIP5 and PRIMAVERA models.Mainly PRIMAVERA models are characterised by smaller biases overall compared to the CMIP5 models CMCCCNRMEC-EarthRed colours = reduction in bias for PRIMAVERAMPIMOHC
20. Frequency Anomalies of maximum Vorticity of ETCSBiases are largely reduced in the PRIMAVERA models especially regarding the underestimation of the stronger storms with higher vorticityCMIP5 vs ERAMaximum vorticityPRIMAVERA vs ERAMaximum vorticity
21. Let’s have a closer look at the upper Danube basinUpper Danube basin and its topography.
22. Hydrological impact over the upper Danube basinCMIP5CORDEXPRIMAVERANormalized overall error as compared with CRU TS3.25 at the basin scale (1950-2005)Mean temperatureTemperature variability
23. Hydrological impact over the upper Danube basinCMIP5CORDEXPRIMAVERANormalized overall error as compared with CRU TS3.25 at the basin scale (1950-2005)Mean precipitationPrecipitation variability
24. summaryThe general improvement of having higher resolution can be difficult to quantify – is it enough that it looks better?Improvements for specific processes are easier to quantify:Extreme precipitation is better reproduced by high resolution models. Extra Tropical Cyclones are better represented in the PRIMAVERA models compared to CMIP5 A study over the Danube basin shows that the PRIMAVERA models even outperform the CORDEX RCM for a selection of climate variables
25. Thanks for listening!Are there questions?