Sensitivity analysis of City Delta calculations for several Spanish cities

Transcription

1 33 rd meeting of the Task Force on Integrated Assessment Modelling 33 rd meeting of the Task Force on Integrated Assessment Modelling Sensitivity analysis of CD for Spain Prague. 3 May, 2007 National Integrated Assessment activities: Sensitivity analysis of City Delta calculations for several Spanish cities J. Lumbreras, R. Borge, E. Rodríguez jlumbreras@etsii.upm.es UNIVERSIDAD POLITÉCNICA DE MADRID Department of Chemical and Environmental Engineering

2 INTRODUCTION SENSITIVITY ANALYSIS CONCLUSIONS

3 Introduction National Emission Inventory SEP methodology Directives, agreements and protocols Possible future emission levels Policy Policy assessment assessment and and decision-making decision-making support support Modelling system? Legislation, standards Possible future air quality levels AOT40 [ppm.hours]m 2010 Evaluation of City Delta results in some Spanish cities using national estimations

4 City Delta calculations (IIASA Interim Report IR ) Where: CD concentration increment computed with the 3 models α, β regression coefficients D city diameter U wind speed Q change in emission fluxes d number of winter days with low wind speed Ac Urban shape area A E EMEP grid cell area

5 INTRODUCTION SENSITIVITY ANALYSIS CONCLUSIONS

6 33rd meeting of the Task Force on Integrated Assessment Modelling Sensitivity analysis of CD for Spain Sensitivity analysis for the two largest cities in Spain (Madrid & Barcelona)

7 Sensitivity parameters analysed Area: Calculation method Surface considered Population Emissions Meteorological data: Wind speed Number of low wind speed days during winter

8 Areas (Madrid) Urban shape = km 2 Urban shape diameter = 39.8 km Calculated area = km 2 City area = km 2 Region area = 8022 km 2 Urban fabric Industrial, commercial and transport units CORINE Land Cover 2000 E-W N-S

9 CD (µgm -3 )

10 Areas (Barcelona) Urban shape = km 2 Urban shape diameter = 23.3 km Calculated area = km 2 City area = km 2 Region area = 7728 km 2 Urban fabric Industrial, commercial and transport units CORINE Land Cover 2000 Max Min 20 10

11 CD (µgm -3 )

12 Population IIASA value: Official city population (2004): Official region population (2004): CD (µgm -3 )

13 Population IIASA value: Official city population (2004): Official region population (2004): CD (µgm -3 )

14 Emissions CD (µgm -3 ) CD (µgm -3 )

15 Wind speed MADRID Alcob Alcalá Alcor Cosl Fuen Get Leg Móst Tor r Bar Average BASE Average # days < # winter days < CD (µgm -3 ) CD (µgm -3 )

16 BARCELONA El Prat CMT Areyns Fabra Average BASE Average # days < # winter days < Changes in wind speed for Barcelona City City Delta ((µgm g/m 3 ) -3 ) Base # low wind days=0.365 # low wind days=1 # low wind days=58 Changes in number of low wind speed days for Barcelona City Delta (µgm -3 ) City Delta ( g/m 3 ) Base # low wind days=0.365 # low wind days=1 # low wind days=58

17 Increment intervals for Madrid MADRID Urban shape Population Diameter Area City emissions Wind speed # low wind speed days City Delta Base 4,575, Highest CD 4,575, Intermediate CD 4,575, Lowest CD 4,575, City Delta (µgm -3 )

18 Increment intervals for Barcelona Urban shape BARCELONA Population Diameter Area City emissions Wind speed # low wind speed days City Delta Base 2,859, Highest CD 2,859, Lowest CD 2,859, City Delta (µgm -3 )

19 INTRODUCTION SENSITIVITY ANALYSIS CONCLUSIONS

20 Comparison to Mar-2007 NECPI document (NEC Scenario Analysis Report Nr. 3) Reported for (sub)urban background stations Reported for traffic background stations Reported for regional background stations

21 Comparison to lowest CD Reported for (sub)urban background stations Reported for traffic background stations Reported for regional background stations

22 Comparison to other CD > 100 Reported for (sub)urban background stations Reported for traffic background stations Reported for regional background stations

23 Conclusions (1) It was necessary IIASA support to reproduce calculations Current computed urban concentrations do not match with PM 2.5 measurements for Madrid and Barcelona CD computation is: very sensitive to geometric and meteorological parameters little sensitive to emissions unsensitive to population Sensitivity is city-dependent: relatively robust to input changes in Madrid (CD ranging from 0.6 to 9.3 mg/m 3 ) highly sensitive to input changes in Barcelona (CD varies from mg/m 3 )

24 Conclusions (2) It is necessary to obtain improved values from countries, especially regarding: Wind data (both wind speed and number of low ws days) Urban shape area Since there is large variability between monitoring stations, it appears critical to have comparable criteria among countries to select the monitoring stations: for meteorological input data for PM 2.5 concentration comparison