Transcript Il CMCC e la sfida della Scienza del Clima

Performance evaluation on Sub-Saharian Africa
region from high resolution simulations with
COSMO-CLM
Paola Mercogliano, CIRA and CMCC
Edoardo Bucchignani, CIRA and CMCC
Myriam Montesarchio, CMCC
Enrico Scoccimarro, INGV and CMCC
Cava dé Tirreni,
30 August -2 September 2011
1. Outlook
•The CLUVA project
•The domains simulated
•List of simulations and status
•NIGERIA- World Bank: validation over Nigeria
•The West domain: validation
•The Lower East domain: validation
•Conclusions
2. CLUVA – Climate change and urban vulnerability in Africa
Project Co-ordinator: AMRA, Center of Competence in the field of Analysis and
Monitoring of Environmental Risk, Italy
The project objective is to develop methods and knowledge to be applied to African cities, to
manage climate risks, to reduce vulnerabilities and to improve their coping capacity and
resilience towards climate changes. The project will explore the issues of climate change
vulnerability, resilience, risk management and adaptation in selected African cities with local
partners.
The aim is to set up methods and work out probabilistic scenarios of climate change
affected hazards having a resolution that fits for regional and urban systems (for the 5
selected cities) and related uncertainties. More detailed aims are:
•
To produce downscaled regional climate scenarios (IPCC scenarios: RCP4.5 and RCP8.5)
for selected African areas surrounding the African cities of interest, at high resolution
(about 8 km). Boundary from CMCC-MED.
•
To produce very high resolution projection (about 1-2 km) for the climate of some African
cities using specific and accurate statistical techniques
3. Areas of interest for CLUVA 1950-2050
Spatial Resolution: 8 km
EAST Domains:
WEST Domain:
(18 W -15.17 E; 3.3 – 16.8 N)
465 x 190 grid points
N (34.4 – 42.9 E; 6.1N – 12.5N)
120 x 90 grid points
S (34.5 – 41.3 E; 11.8S – 2.1S)
95 x 135 grid points
St.Louis
(16.5 W, 16.03 N)
Ougadougou (1.55 W, 12.37 N)
Douala
(9.71 E, 4.045 N)
Addis Abeba (38.75 E, 9.02 N)
Dar es Salaam (39.27 E, 6.82 S)
4. West Domain : orography of the area
Area of interest:
(18W-15.17 E ; 3.3–16.8 N)
Extension: 3491 x 1489 km
Computational grid: 465x190 grid points;
40 levels
5. Lower east domain
Area of interest:
(34.54 - 41.26E ; 11.8 – 2.2 S)
Extension: 747 x 1064 km
Computational grid: 95x135 grid points;
40 levels
Orography of the area
6. Upper east domain
Area of interest:
(34.45 - 43.03E ; 6.16 – 12.59 N)
Extension: 952 x 714 km
Computational grid :120x90 grid points;
40 levels
Orography of the area
7. Details of the Numerical simulations
8 km resolution
•
•
•
•
•
Supercomputer used:
Cluster of 30 IBM P575 nodes (32 cores per node) (peak power 18 TF)
Driving data: CMCC-MED 80 km resolution
Time step: 40 sec.
Numerical scheme: Runge-Kutta 2-time level HE-VI integration
Simulated periods and domains :
1970-2065 (A1B scenario, only for west domain)
1950-2050 (RCP4.5 and RCP8.5 scenarios,
for west domain and east domains)
8. List of simulations
 1970- 2065 (completed)
•
Scenario A1B West domain
•
•
Scenario RCP 4.5 Lower East domain  1950-2050 (completed)
Scenario RCP 4.5 Upper East domain  1950-2050 (completed)
•
•
•
•
Scenario RCP 4.5 West domain  to be completed within November 2011
Scenario RCP 8.5 Lower East domain  to be completed within November 2011
Scenario RCP 8.5 Upper East domain  to be completed within January 2012
Scenario RCP 8.5 West domain  to be completed within January 2012
9. NIGERIA- World Bank
Validation over Nigeria region: T2M (1976-2005)
T2m yearly averaged value history
(from Climate Risk Analysis in Nigeria, Interim Report, June 2011)
10. NIGERIA- World Bank
Validation over Nigeria region: Total Precipitation
Total precipitation yearly averaged
value history
(from Climate Risk Analysis in Nigeria, Interim Report, June 2011)
11. NIGERIA- World Bank
Climate projection over Nigeria region: T2M (up to 2065)
(from Climate Risk Analysis in Nigeria, Interim Report, June 2011)
12. NIGERIA- World Bank
Climate projection over Nigeria region: Total Precipitation (up to 2065)
(from Climate Risk Analysis in Nigeria, Interim Report, June 2011)
13. Mean temperature bias with CRU
DJF
Not well defined bias
map.
Bias is between -3 and
+3 degrees
JJA
Hot bias of 3-4 degrees
in the north part.
In the south part, the
bias is between -1 and
+1 degrees.
(COSMO-CRU)
14. Mean precipitation (mm/month) bias with CRU
DJF
Underestimation in the
south coastal area.
Quite good agreement
in the other parts.
JJA
A strong bias is
registered.
15. Temperature variation :
future (2021-2050) minus past (1971-2000)
DJF
JJA
16. Precipitation variation (mm/month):
future (2021-2050) minus past (1971-2000)
DJF
There is a big
difference between
winter and summer.
In winter, there is a
slight decrease of
precipitation, while in
summer there is a
general increase with
some exceptions.
JJA
17. Mean temperature bias with CRU (east lower domain)
DJF
JJA
18. Temperature variation:
future (2021-2050) minus past (1971-2000)
DJF
JJA
19. Conclusions
• Good agreement with observed data over Nigeria
•An increase of temperature is expected in the XXI century in Nigeria, while
precipitations are substantially unchanged.
•Numerical results related to the simulation of the climate of the west and the
east lower domain at high resolution have been shown.
•for the west domain: in winter, temperature bias is not well defined and
the precipitation is underestimated in the south coastal area with a quite
good agreement in the other parts, while in summer there is a strong bias;
•for the east lower domain: the temperature is underestimated in winter
and overestimated in summer;
•for the west domain in the future, the temperature will increase, especially
in winter; the precipitation will decrease slightly in winter and will greatly
increase in summer;
•for the east lower domain in the future, the temperature will increase,
especially in winter.
Thanks