1. 4 th Training Course on WMO SDS-WAS products (satellite and ground observation and modeling of atmospheric dust) Casablanca-Morocco, November 17-20, 2014 Sergio Rodríguez (srodriguezg@aemet.es) & Emilio Cuevas ( ecuevasa@aemet.es ) Ground observations of mineral dust AEMET, Spain

2. AEMET, Agencia Estatal de Meteorología 2 Session of practical work (Ground-based observation) WMO SDS-WAS Sergio Rodríguez (srodriguezg@aemet.es)srodriguezg@aemet.es Casablanca-Morocco, 17-20 November 2014 Long term 1987 - 2012 trends of sulfate, nitrate and dust mixing in the Saharan Air Layer

3. Izaña, Tenerife Island 2400 m.a.s.l., free troposphere (night) Long term aerosols: Total number concentration 2006 Size distribution 0.5 – 20 µm APS 2007 Scattering total and back 3 TSI neph 2008 Absorption 1 MAAP 2007 Aethalometer 7 2012 chemical composition 1987 Size distribution 10 – 400 nm SMPS 2008

4. aerosol chemical composition at Izaña (since 1987): dust (Al, Fe,...), SO 4 =, NO 3 -, NH 4 +, Na, and Cl - sample collection on filter 1987-1999 30 m3/h Dust: ash method (normalized Al/dust – 8%) SO 4 =, NO 3 -, NH 4 +, Cl - : ion chromatography Al, Na, Fe: INAA cellulose PM T : total particulate matter 2002- up to the date 30 m3/h Dust: elemental composition IPC- AES, ICP-MS (normalized Al/dust – 8%) SO 4 =, NO 3 -,, Cl -: : ion chromatography NH 4 + : capillary electrophoresis, specific electrode OC, EC: TOR quartz microfibber filter PM 10 : particulate matter diameter ≤ 10 µm PM 2.5 : particulate matter diameter ≤ 2.5 µm PM T : total particulate matter 1987-2014 27 years aerosol chemistry in the free troposphere samples collected at night free troposphere The two data sets were jointed for the firs time

5. PM T 0.9 elemental carbon 0.2% none ammonium-sulfate dust (Al, Fe, Ca, Ti..) Al = 8% dust 91% 2.2% 1.2% 0.4% 1.9% 3.8% ammonium-sulfate ammonium nitrate organic matter 47.3 µg/m 3 42.6 1.0 0.5 0.07 0.2 1.8 PM 2.5 0.2 elemental carbon 0.4% none ammonium-sulfate dust 85% 3.0% 2.7% 1.0% 1.1% 5.8% ammonium-sulfate ammonium nitrate organic matter 18.5 µg/m 3 15.8 0.6 0.5 0.07 0.2 1.1 PM 10 0.6 elemental carbon 0.2% none ammonium-sulfate dust 91% 2.2% 1.2% 0.4% 1.3% 3.4% ammonium-sulfate ammonium nitrate organic matter 42.0 µg/m 3 38.3 0.9 0.5 0.07 0.2 1.4

6. PM T 0.9 0.2% 91% 2.2% 1.2% 0.4% 1.9% 3.8% 47.3 µg/m 3 42.6 1.0 0.5 0.07 0.2 1.8 Satellite (Earth Probe, Nimbus 7, Aura): Total Ozone Monitor Spectrometer (1987-2001) Ozone Monitor Instrument (2005-2012) Saharan Air Layer MDFA: Major Dust Frequency Activity number days UV Absorbing Aerosol Index > 1 total number of days in the month MDFA = Izaña UV Absorbing Aerosol Index = sensitive to iron oxides in dust

7. aerosol chemical composition (since 1987): dust (Al, Fe,...), SO 4 =, NO 3 -, NH 4 +, Na, and Cl - part-1: long term evolution of dust part-2: long term evolution of sulfate

8. aerosol chemical composition (since 1987): dust (Al, Fe,...), SO 4 =, NO 3 -, NH 4 +, Na, and Cl - part-1: long term evolution of dust part-2: long term evolution of sulfate

10. part-1: long term evolution of dust Sahara Sahel summer winter Winter: dust export modulated by the North Atlantic Oscillation (Ginoux et al., 2004) Sahel: Monsoon rains (Prospero and Lamb, 2003) winter

11. part-1: long term evolution of dust Sahara Sahel summer winter Which are the large scale processes that influence on long term inter-annual variability in Saharan dust export in summer? We have focused in summer scientific question Why? Is the season when maximum dust emissions occurs in North Africa due to the activation of subtropical Saharan sources Processes that modulated inter-annual variability in dust export are still unknown Sahel: Monsoon rains (Prospero and Lamb, 2003) Winter: dust export modulated by the North Atlantic Oscillation (Ginoux et al., 2004)

12. dust T, µg/m 3 0 120 80 40 19881989199019921987199319941996199119971998199920002002199520042005200720012003200820092010201220062011 part-1: long term evolution of dust summer dust view from Izaña: subsidence free-troposphere Saharan Air Layer

13. dust T, µg/m 3 0 120 80 40 19881989199019921987199319941996199119971998199920002002199520042005200720012003200820092010201220062011 part-1: long term evolution of dust summer dust Methods for data analysis: 1.Meteorological re-analysis data 2.Satellite data

14. dust T, µg/m 3 0 120 80 40 19881989199019921987199319941996199119971998199920002002199520042005200720012003200820092010201220062011 part-1: long term evolution of dust summer dust Methods for data analysis: 1.Meteorological re-analysis data National Center for Environmental Prediction (NCEP) National Center for Atmospheric Research (NCAR) 2.Satellite data

15. Bamako – Mali Morocco North AFrican Dipole Intensity dust T, µg/m 3 0 120 80 40 19881989199019921987199319941996199119971998199920002002199520042005200720012003200820092010201220062011 part-1: long term evolution of dust summer dust 700 hPa: relevant level for dust export dust export occurs at altitudes between 1 to 5km above sea level: 700hPa geopotential level North AFrican Dipole Intensity: Difference of the anomalies of the geopotential between sub-tropic (Morocco) and the tropic (Bamako- Malí)

16. Bamako – Mali Morocco North AFrican Dipole Intensity dust T, µg/m 3 0 120 80 40 19881989199019921987199319941996199119971998199920002002199520042005200720012003200820092010201220062011 -2 +3 -3 +2 0 +1 NAFDI 700 hPa: relevant level for dust export part-1: long term evolution of dust Pearson correlation between NAFDI and the dust at Izaña = +0.75 R1: Variability in NAFDI influence on dust export geostrophic wind influence on dust export

17. 0 120 80 40 19881989199019921987199319941996199119971998199920002002199520042005200720012003200820092010201220062011 -2 +3 -3 +2 0 +1 NAFDI Correlation coefficient (1987- 2012) between NAFDI and precipitation rates zonal wind (925hPa) MDAF dust T, µg/m 3 part-1: long term evolution of dust SSS: Subtropical Saharan Stripe

18. 0 120 80 40 19881989199019921987199319941996199119971998199920002002199520042005200720012003200820092010201220062011 -2 +3 -3 +2 0 +1 NAFDI Correlation coefficient between NAFDI and precipitation rates zonal wind MDAF dust T, µg/m 3 part-1: long term evolution of dust 1987-2012 back trajectories 10 50 400 20 30 40 1987-2013 frequency R1: Variability in NAFDI influence on trade winds at the north of the ITCZ (surface-925hPa Harmattan)  dust export (1-5 km)

19. dust T, µg/m 3 0 120 80 40 19881989199019921987199319941996199119971998199920002002199520042005200720012003200820092010201220062011 part-1: long term evolution of dust summer dust Methods for data analysis: 1.Meteorological re-analysis data 2.Satellite data Satellite (Earth Probe, Nimbus 7, Aura): Total Ozone Monitor Spectrometer (1987-2001) Ozone Monitor Instrument (2005-2012) MDFA: Major Dust Frequency Activity number days UV Absorbing Aerosol Index > 1 total number of days in the month MDFA = Izaña UV Absorbing Aerosol Index = sensitive to iron oxides in dust ….201219871988……1990 Location of the Saharan Air layer for every summer:

20. 0 120 80 40 19881989199019921987199319941996199119971998199920002002199520042005200720012003200820092010201220062011 -2 +3 -3 +2 0 +1 NAFDI Correlation coefficient between NAFDI and precipitation rates zonal wind MDAF dust T, µg/m 3 part-1: long term evolution of dust satellite product Correlation coefficient between NAFDI and the MDAF (location of the Saharan Air Layer) 1. More dust activity at the north of the ITCZ 2. Less dust activity at the south of the ICTZ reason ?

21. 0 120 80 40 19881989199019921987199319941996199119971998199920002002199520042005200720012003200820092010201220062011 -2 +3 -3 +2 0 +1 NAFDI Correlation coefficient between NAFDI and precipitation rates zonal wind MDAF dust T, µg/m 3 part-1: long term evolution of dust Correlation coefficient between NAFDI and The MDAF (location of the Saharan Air Layer) NAFDI also connected with monsoon rainfalls (Sahel) R: Variability in NAFDI influence on: 1. trade winds at the north of the ITCZ (Harmattan)  dust export 2. monsoon rainfalls, including the Sahel 3. latitudinal shifts of the Saharan Air Layer 1. More dust activity at the north of the ITCZ 2. Less dust activity at the south of the ICTZ reason ?

22. 700hPa: related to dust export SSS: Subtropical Saharan Stripe 925hPa: related to potential dust emissions Northern shift in the monsoon rains

23. Low NAFDI Low Wind Coarse particles (65%) Enhanced effects of dry deposition 0 120 80 40 19881989199019921987199319941996199119971998199920002002199520042005200720012003200820092010201220062011 -2 +3 -3 +2 0 +1 NAFDI dust T, µg/m 3 Dust particles size ? 2002-2012 High NAFDI Windy Coarse particles (80%) More dry deposition

24. part-1: long term evolution of dust 5. Connection global to climate (?) Variability in NAFDI influence on Saharan dust export during 25-y 1. trade winds at the north of the ITCZ (Harmattan)  dust export 2. monsoon rainfalls, including on the Sahel 3. latitudinal shifts of the Saharan Air Layer 4. Dust particle size NAFDI and ENSO connected to variability in subtropical and tropical climate La NiñaEl Niño La Niña Low dust years tend to be associated with El Niño – low (negative) MEI values Multivariate ENSO (El Niño Southern Oscillation) Index (MEI), – calculated with sea level pressure, zonal and meridional components of the surface wind, sea surface temperature, surface air temperature and total cloudiness fraction of the sky over the 20 tropical Pacific Ocean

25. Less monsoon rainfalls Sahara Air Layer: Dust impacts: shifted to northshifted to south More impacts in Morocco Less impacts in Morocco Less impacts of dust+biomass burning from Central Africa Tropical Africa: Benin, Togo, Ghana, Malí, Burkina Faso, Côte d’Ivore, Liberia, Sierre Leone, Guiné Bissau, Senegal More impacts of dust+biomass burning from Central Africa More monsoon rainfalls

26. aerosol chemical composition (since 1987): dust (Al, Fe,...), SO 4 =, NO 3 -, NH 4 +, Na, and Cl - part-1: long term evolution of dust part-2: long term evolution of sulfate

27. trend of sulfate in the Saharan Air Layer concentrations of ammonium-sulfate in the Saharan Air Layer decreased through the 1990s nitrate during dust condition in Izaña view from Izaña: subsidence free-troposphere Saharan Air Layer sulfate present as ammonium sulfate (a-SO 4 = ) NO 3 -, µg/m 3 total <10 µm <2.5 µm a-SO 4 =, µg/m 3

28. trend of sulfate in the Saharan Air Layer < 10 µm total < 2.5 µm a-SO 4 =, µg/m 3 y = -0.0545x + 110.29 SO 4 = as ammonium-sulfate August 1987-2012, dust days TgS no significant change 55% decrease (1)air laden in Saharan dust has previously passed over the Mediterranean and Europe (2)North African emissions of SO 2 did not changed significantly during the 1990s SO 2 emissions (3) Decrease in a-sulfate in the SAL is correlated with the decrease in European SO2 emissions

29. trend of sulfate in the Saharan Air Layer a-SO 4 =, µg/m 3 European Union emissions, TgS a-SO 4 =, µg/m 3 European Union SO 2 emissions, TgS SAL decrease in a-sulfate in the Saharan Air Layer is correlated with the decrease in European SO 2 emissions

30. a-SO 4 =, µg/m 3 SAL European Union SO 2 emissions, TgS a-SO 4 = (SAL), µg/m 3 European Union emissions, TgS a-SO 4 =, µg/m 3 SAL other sources contributing ship emissions in the Mediterranean (?) north African emissions (?) Rodríguez et al., 2011 Cobett et al., 2007

31. a-SO 4 =, µg/m 3 SAL Europe North Africa Ships Mediterranean Europe  Need to develop regional emission inventories

32. dust belt Tanaka and Chiba (2006)

33. 4 th Training Course on WMO SDS-WAS products (satellite and ground observation and modeling of atmospheric dust) Casablanca-Morocco, November 17-20, 2014 Sergio Rodríguez (srodriguezg@aemet.es) & Emilio Cuevas ( ecuevasa@aemet.es ) Ground observations of mineral dust AEMET, Spain