Tropospheric halocompounds and nitrous oxide monitored at a remote site in the Mediterranean

Atmospheric Environment 44 (2010) 4944e4953

Contents lists available at ScienceDirect

Atmospheric Environment journal homepage: www.elsevier.com/locate/atmosenv

Tropospheric halocompounds and nitrous oxide monitored at a remote site in the Mediterranean F. Artuso a, *, P. Chamard a,1, S. Chiavarini a, A. di Sarra a, D. Meloni a, S. Piacentino b, M.D. Sferlazzo c a

ENEA, Casaccia, Via Anguillarese 301, 00123 S. Maria di Galeria, Rome, Italy ENEA, Via Catania 2, 90141 Palermo, Italy c ENEA, Capo Grecale, Lampedusa, Italy b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 8 January 2010 Received in revised form 20 July 2010 Accepted 9 August 2010

Analysis of time series and trends of nitrous oxide (N2O) and halocompounds weekly monitored at the Mediterranean island of Lampedusa are discussed. Atmospheric N2O levels showed a linear upward growth rate of 0.78 ppb yr
1 and mixing ratios comparable with Northern Hemisphere global stations. CFC-11 and CFC-12 time series displayed a decline consistent with their phase-out. Chlorofluorocarbons (CFCs) replacing compounds and SF6 exhibited an increasing temporal behaviour. The most rapid growth rate was recorded for HFC-134a with a value of 9.6% yr
1. The industrial solvents CCl4 and CH3CCl3, banned by the Montreal Protocol, showed opposite trends. While CH3CCl3 reported an expected decay of
1.8 ppt yr
1, an increasing rate of 5.7 ppt yr
1 was recorded for CCl4 and it is probably related to its relatively long lifetime and persisting emissions. Chlorinated halomethanes showed seasonality with a maximum in early April and a minimum at the end of September. Halon-1301 and Halon-1211 displayed a decreasing trend consistent with industry emission estimates. An interspecies correlation analysis gave positive high correlations between HCFC-22 and HFC-134a (þ0.84) highlighting the common extensive employment as refrigerants. Sharing sources inferred the high coupling between CH3Cl and CH3Br (þ0.73) and between CHCl3 and CH2Cl2 (þ0.77). A singular strong relationship (þ0.55) between HFC-134a and CH3I suggested the influence of an unknown anthropogenic source of CH3I. Constraining of source and sink distribution was carried out by transport studies. Results were compared with the European Environment Agency (EEA) emission database. In contrast with the emission database results, our back trajectory analysis highlighted the release of large amounts of HFC134a and SF6 from Eastern Europe. Observations also showed that African SF6 emissions may be considerable. Leakages from SF6 insulated electrical equipments located in the industrialized Northern African areas justify our observations. Ó 2010 Elsevier Ltd. All rights reserved.

Keywords: Nitrous oxide Halocompounds Greenhouse gases Air mass transport

1. Introduction The recent increasing impact of N2O and tropospheric halogenated hydrocarbons (halocarbons) to the energy balance of the Earth-atmosphere system was assessed by the last IPCC report (IPCC, 2007) which indicated halocarbons and N2O to be among the four principal long-lived greenhouse gases (GHGs) emitted by human activities. N2O is a powerful GHG produced both naturally and anthropogenically. Natural sources include microbial processes of

* Corresponding author. Tel.: þ39 0694005584. E-mail address: fl[email protected] (F. Artuso). 1 Retired. 1352-2310/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.atmosenv.2010.08.019

nitrification and denitrification in soils and water. Use of nitrate and ammonium fertilizers noteworthy increased the emission of N2O from soils. Other anthropogenic sources of N2O are fuel combustion, adipic and nitric acid industrial production and waste management activities (Cicerone, 1989). Its mixing ratio in the atmosphere has considerably risen during the past two centuries (Wolff and Spahni, 2007). N2O has an atmospheric lifetime of about 114 years and its global warming potential (GWP) is 298 times higher than CO2. Human-made halocarbons have low mixing ratios but are very effective GHGs because of their high GWPs and long lifetimes in the atmosphere (IPCC, 2007). Chlorine- and bromine-containing halocarbons play also a significant role on stratospheric ozone depletion (Molina and Rowland, 1974). The increase of the halocarbons emissions in the last years, together with the discovery of the

F. Artuso et al. / Atmospheric Environment 44 (2010) 4944e4953

Antarctic ozone hole (Farman et al., 1985), lead to the formulation of the Montreal Protocol (WMO, 1988) which regulates the production of the ozone-depleting substances (ODSs) and called for the elimination of their production by 1996. The principal ODSs include chlorofluorocarbons (CFCs), which are responsible of the 95% of the halocarbons radiative forcing (RF) (IPCC, 2007). The most important ones are CFC-11, CFC-12 and CFC113, used mainly as foam blowing agents, refrigerants or solvents in electronic industry. Ground based measurements of troposphere mixing ratios showed that most of the Montreal Protocol species, with the exception of CFC-12, began to decay since the mid-1990s in response to reduced global emissions (Clerbaux and Cunnold, 2007). CFC-12 concentration peaked in 2003 and is now levelling off because of its long lifetime (100 years). Halons (Halon-1301, Halon-1211) are bromine-containing haloalkanes primarily used as fire extinguishers and also banned by the Montreal Protocol (Montzka et al., 1999). The species designed as replacements of CFCs and halons are SF6 and the partially halogenated hydrochlorofluorocarbons (HCFCs) and hydrofluorocarbons (HFCs). Both classes of compounds have low or null ozone depleting potentials (ODPs) and shorter lifetimes than CFCs but may contribute significantly to the greenhouse effect because of their long-wave radiation absorption. SF6 is a stable trace gas used predominantly in gas-insulated electrical equipment, in degassing of molten reactive metals and also as tracer in atmospheric transport and oceanography studies (Maiss and Brenninkmeijer, 1998). Its physical and chemical stability combined with its strong absorption in the infrared region make SF6 a very long-lived and the most efficient GHG in the atmosphere (IPCC, 2007). Due to its unperturbed constant emissions, almost uniquely anthropogenic, SF6 mixing ratio is increasing over the past decades. HCFCs were widely used in refrigeration and foam blowing applications. These compounds are GHGs and because of their ODP their production and usage are only temporary (UNEP, 2001). Surface observations of HCFC-141b and HCFC-142b and HCFC-22 remarked their recent rapid growth in the troposphere (Derwent et al., 2007; O’Doherty et al., 2004). For this reason the European Union accelerated the curtailment of HCFC production with a complete ban within 2010 (European Communities, 2000). HFCs are the second generation alternative to CFCs and are included in the Kyoto Protocol because of their positive radiative forcing (Forster et al., 2005). HFC-134a is the most rapidly increasing HFC because of its extensive use as replacement of CFC refrigerants. Halocarbons used as solvents (CCl4, CH2Cl2, CH3CCl3) are of particular concern in climate and air quality studies (Simmonds et al., 1998; Simmonds et al., 2006). CCl4 and CH3CCl3 global mixing ratios are rapidly declining as a result of the compliance to the Montreal Protocol. CH2Cl2 is not included among the ODSs because its high reactivity with OH results in a short persistence in the troposphere but it is a toxic air pollutant (Cox et al., 2003). Ozone depleting halomethanes such as CH3I, CH3Cl, CH3Br, and CHCl3 are primarily emitted by biogenic processes from oceans and soils (Smythe-Wright et al., 2006; Dimmer et al., 2001). The natural contribution to their fluxes is difficult to quantify and, consequently, their global budget is still uncertain. Long-term observations of atmospheric concentration of halocarbons, SF6 and N2O in the troposphere are crucial issues for the investigation of global climate changes. Knowledge of their radiative properties, lifetime and atmospheric balance are essential data for the prediction of their future trends. For this reason a number of global monitoring studies were carried out during the last years for determining trends and distribution of these compounds in the troposphere (Prinn et al., 1990; O’Doherty et al., 2004).

4945

In this work we present and analyse for the first time halocompounds and N2O weekly records measured at the remote site of Lampedusa Island, located in the middle of the Mediterranean basin. The measurements are part of the ENEA (Italian National Agency for New technologies, Energy and Sustainable Economic Development) GHG monitoring program (Chamard et al., 2003; Artuso et al., 2007, 2009). For its favourable geographical characteristics Lampedusa is a unique observing site in the heart of the Mediterranean and is part of the major global monitoring programs such as the Global Atmosphere Watch (GAW), established by the World Meteorological Organization (WMO, 2005), and the National Oceanic and Atmospheric Administration (NOAA) networks (Conway et al., 1994). Measurements of N2O, CFC-11 and CFC-12 atmospheric mixing ratios started in 1996 and since December 2003 several CFC replacement compounds (HCFC-22, HFC-134a, HCFC-141b, HCFC142b and SF6), were also detected with a weekly frequency. Other anthropogenic halocarbons, such as chlorinated solvents, halomethanes and halons were regularly monitored since September 2007. We analysed time series and growth rates of the above mentioned species and made a comparison with data obtained in other globally distributed stations in order to estimate the representativeness of Lampedusa Station on global and regional scale. Interspecies correlations were considered for determining common sources. Analysis of backward airmass trajectories was also reported for some halogenated compounds and the results were compared with data reported by emission inventories in order to evaluate possible source and sink areas. 2. Experimental 2.1. Sampling program and analysis Lampedusa is a small Mediterranean island with scarce vegetation and almost unperturbed by significant anthropogenic pollution sources. A detailed description of its geographical characteristics and meteorological conditions may be found elsewhere (Artuso et al., 2007, 2009). For CFCs and N2O analysis, air samples were collected weekly in 2 L glass flasks, firstly evacuated and flushed many times with ambient air before sampling. Air was dried by a sodium perchlorate trap and then pumped into the flasks at a pressure of 3 atm. Sampling inlet was located 3 m above the ground level at an altitude of about 45 m a.s.l. Since October 2005 on-line in situ measurements were carried out in parallel with the fixed event ones, but in this work only weekly data were presented. The technique used for analysis was conventional electron-capture-gas chromatography (GC-ECD). The GC (6890, Agilent Technologies) was equipped with a 25 ml loop and an Altech Porasil B packed column. The carrier gas was a mixture of argon/methane (5%) and the column temperature was maintained at 50  C during analysis. Before each injection the loop was flushed for about 2 minutes at 50 cc min
1. Since December 2003 several halocompounds (HCFC-22, HFC-134a, HCFC-141b, HCFC-142b and SF6) were also weekly monitored and from September 2007 halocarbons measurement program was extended to a number of halogenated trace species (CFC-113, CH2Cl2, CHCl3, CH3Cl, CH3I, CH3Br, CH2Br2, CCl4, CH3CCl3, Halon-1301, Halon-1211). Samples were collected on a weekly basis in 6 L stainless steel canisters (Entech Instruments Inc.), internally passivated by a silcosteel layer, and then shipped to the ENEA Casaccia Laboratory (Rome), where they were analysed by GC with mass spectrometric detection (GC-MS, 6890-5873N Agilent Technologies). Trace species were dried by a nafion tube and then preconcentrated by an adsorption trap system (Markes Air Server/

F. Artuso et al. / Atmospheric Environment 44 (2010) 4944e4953

330 325 320

a

Mace Head Lampedusa Cape Grim

2

N O (ppb)

Unity thermal Desorber). The trap was maintained at
15  C during the adsorption step (t ¼ 60 min; flow rate ¼ 40 cc min
1) and heated at 300  C for 1 min during the thermal desorption. All compounds, except the most volatile SF6, were quantitatively trapped with this method. Once released by the trap, the analytes were injected into the column (Gaspro 0.32 mm I.D., 60 m; carrier: He) through a capillary transfer line (0.01 mm I.D.) and separated with the following temperature profile: 40  C for 3 min followed by a programmed 8  C min
1 temperature rise up to 120  C, followed by a 1 min hold time and finally by a 6  C min
1 ramp up to 250  C. The detector was operating in a selected ion monitoring (SIM) mode. All data measured at Lampedusa were regularly submitted to the World Data Centre for GHGs (WDCGG) established by the GAW Programme.

315 310

b

350 CFC-11 (ppt)

4946

300 250 200 150

2.2. Calibration

c

CFC-12 (ppt)

650 600 550 500 450 400 1996

1998

2000

2002 2004 time (year)

2006

85

a

80 75 70

4

CCl (ppt)

110

Calibration scale

CFC-113 HCFC-22 HCFC-141b HCFC-142b HFC-134a SF6 CH3Cl CH3Br CH2Cl2 CCl4 CH3CCl3 Halon-1211 Halon-1301 CH2Br2 CH3I CHCl3

76.0 205.3 21.1 18.9 57.6 6.49 551.3 8.75 47.9 89.9 11.9 4.27 3.11 1.3 0.67 10.9

1.1 3.6 0.5 0.4 1.1 0.35 11.8 0.34 2.0 1.8 0.4 0.15 0.07 0.1 0.06 0.3

2002 2006 1994 1994 1995 2006 2003 2003 1992 2002 2003 2006 2006 2003 2004 1992

b

90 80 70 2

y = 3578.7 - 1.7759x R = 0.71353 16

3

Uncertainty (ppt)

y = -11306 + 5.6742x R = 0.27006

100

14

3

Mean mixing ratio (ppt)

CH CCl (ppt)

Compound

2010

aimed to establish a quality assurance system for global halocarbons measurements. The results obtained during the intercomparison experiment were all within the standard deviation given by the reference laboratory.

2

Table 1 2008 mean mixing ratios, uncertainty and calibration scales of halocarbons measured by GC-MS.

2008

Fig. 1. (a) Monthly means N2O time series at Lampedusa, Mace Head and Cape Grim. Weekly time series of (b) CFC-11 and (c) CFC-12.Ticks on x axis mark beginning of each year.

CFC-113 (ppt)

N2O and CFCs data recorded before 2000 were referred to standards provided by a national manufacturer (SIAD, Milan). Then two standard mixtures provided by the Climate Monitoring and Diagnostic Laboratory/National Oceanic and Atmospheric Administration, (CMDL/NOAA), were used and previous data were rescaled and referred to the CMDL83 scale. The analysis sequence alternated measurements of the two standards with aliquots of the air sample. Ten measurements were performed on each flask sample and standard deviations and arithmetic mean molar mixing ratios were calculated on the ten records. Standard deviations of N2O, CFC-11 and CFC-12 were typically 0.2, 0.4 and 0.1% respectively. Halocompounds concentrations measured by GC/MS were calibrated using a primary standard prepared by gravimetric methods at the NOAA Global Monitoring Division. Certified mixing ratios (dry air mole fraction) of the species of interest were all close to the ones occurring in actual air samples. Calibration scales for each compound were named according to the year they were adopted and were listed in Table 1. The same table summarized the 2008 mean mixing ratios and uncertainties of all halocompounds measured by GC-MS. The analysis sequence consisted in a series of four sample measurements bracketed by the primary standard (one standard every two sample aliquots). Standard deviations, calculated on the four sample data set, were typically between 2 and 8%. The accuracy of the analytical method was tested during the International HALocarbons in Air Comparison Experiment (IHALACE) (Hall et al., 2008). The inter-laboratory round robin test was managed by the GAW program and was

12

c

10 8 2007.6

2008

2008.4 time (year)

2008.8

2009.2

Fig. 2. Records of (a) CFC-113, (b) CCl4 and (c) CH3CCl3 mixing ratios.

F. Artuso et al. / Atmospheric Environment 44 (2010) 4944e4953

3. Results and discussion 3.1. Analysis of data series and trends 3.1.1. Nitrous oxide Monitoring of the N2O atmospheric concentration in Lampedusa started in 1996. Outliers were rejected in order to prevent distortion of the fitting curve used for background data selection. Records deviating from the trend line more than 3s, where s indicates the mean standard deviation, were rejected. Only 76% of all available data were taken into account in the period 1996e2008. Monthly mean data were calculated from weekly records and the linear trend line was used to evaluate the annual growth rate. Fig. 1a reports N2O Lampedusa time series together with data from two GAW observation sites located in the northern (NH) and southern (SH) hemispheres, Mace Head (Ireland, 53.33 N, 9.90 W) and Cape Grim (Tasmania, 40.68 S, 144.68 E). Both stations are part of the AGAGE network (Prinn et al., 1990). In agreement with data

9 2

y = -518.87 + 0.26152x R = 0.60484

a

7

6

SF (ppt)

8

6 5

HCFC-22 (ppt)

240

2

y = -12327 + 6.2392x R = 0.6607

b

220 200 180

4947

reported by the IPCC, 2007, the mean N2O value recorded at Lampedusa in this period was 319.6  0.6 ppb. This finding remarks the representativeness of Lampedusa on global and regional scale. Atmospheric N2O levels have linearly risen at a growth rate of about 0.78 ppb yr
1 and reached in 2008 a mean mixing ratio of 322.5 ppb. Annual trends at Cape Grim and Mace Head are both around 0.73 ppb yr
1(data source: http://gaw.kishou.go.jp/ wdcgg/). Our data mirror the N2O pattern recorded at Mace Head and show a similar trend, but higher concentrations, respect to values obtained at Cape Grim station. Such behaviour reflects the N2O latitudinal gradient due to the high concentration of N2O sources distributed in the 30 Ne90 N semi hemisphere. Fossil fuel combustion and the enhanced microbial production caused by the expansion of fertilized agricultural lands are thought to be the main causes of the observed global temporal N2O increase. 3.1.2. Chlorofluorocarbons: CFC-11, CFC-12, CFC-113 Measurements of CFC-11 and CFC-12 atmospheric concentrations started in 1996, but in this paper 1996 records were neglected because they were not qualitatively valid. Data from January 1997 to December 2008 were taken into account for calculations. As reported in literature the concentration of CFC-11 reached its maximum around 1993 and then started to decrease (Clerbaux and Cunnold, 2007). The resulting mixing ratio time series observed at Lampedusa since January 1997 are shown in Fig. 1b. CFC-11 has a mean value of 256.2  1.2 ppt, a declining rate of 1.4 ppt yr
1 and is now at a level of 245.9 ppt (mean 2008 value). In contrast, the mixing ratio of the longer lived CFC-12 slowly increased until 2002 and then declined in the last 7 years at a rate of 2.0 ppt yr
1(see Fig. 1c). Its mean level for the overall period of analysis is 536.9  0.7 ppt. The 2008 mixing ratio mean value is 529.5 ppt. These findings are in tune with global CFC trends. CFC-113 was widely used as industrial solvent in the past and was successfully substituted after the application of the international regulations. Fig. 2a reports the CFC-113 time series recorded

160 y = -540.07 + 0.27931x R = 0.22004

22

3

20 18

y = -2216.7 + 1.1137x R = 0.84091

b CHCl (ppt)

20

16 14 2

y = -9549 + 4.7827x R = 0.74636

10

e

60

c 75

2

55

2

50 45 40 35 2005

15

3

18

65

600

d

CH Cl (ppt)

HCFC-142b (ppt)

22

a

675

525 2

HFC-134a (ppt)

750

c CH Cl (ppt)

HCFC-141b (ppt)

2

24

50 25

2006

2007 time (year)

2008

2009

Fig. 3. Records of (a) SF6, (b) HCFC-22, (c) HCFC-141b, (d) HCFC-142b and (f) HFC-134a mixing ratios, error bars and curve fit.

0

Jan 2008

May 2008 time (month_year)

Oct 2008

Fig. 4. Records of (a) CH3Cl and (b) CHCl3 and CH2Cl2(c) mixing ratios and error bars.

4948

F. Artuso et al. / Atmospheric Environment 44 (2010) 4944e4953

at Lampedusa in the period September 2007eDecember 2008. The short observation time does not allow deriving a decreasing trend. The mixing ratio seems to be constant around the mean value of 76.1 1.2 ppt in accordance with global measurements (http://gaw. kishou.go.jp/wdcgg/). 3.1.3. CFCs and halons replacement compounds: SF6, HCFC-22, HCFC-141b, HCFC142b, HFC-134a The 2004 records of ODSs alternative species measured at Lampedusa are not reported being affected by large errors due to the non-optimisation of the chromatographic method. In contrast to the CFC decline, their mixing ratios in the troposphere exhibit a linearly increasing pattern. Records of CFCs replacement compounds and SF6 measured at Lampedusa are reported in Fig. 3. SF6 is the most potent greenhouse gas with a GWP of 22800 over a 100 year period and its atmospheric lifetime is estimated to be 3200 years (Ravishankara et al., 1993). Its long residence time in the atmosphere is not counterbalanced by significant sinks and the accumulation in the atmosphere causes a rapid increase in its concentration. SF6 data set recorded at Lampedusa in the period

January 2005eDecember 2008 is shown in Fig. 3a. A rapid linear growth of 0.26 ppt yr
1 (4.3% yr
1), is recorded at Lampedusa. The actual 2008 mean value is 6.49 ppt (see Table 1). HCFC-22 (CHClF2) is the most widely used refrigerant in the world and for this reason it is the most abundant CFC replacement compound in the atmosphere. HCFC-22 measurements performed at Lampedusa during the 4-year period 2005e2008 (see Fig. 3b) show a growth rate of 6.2 ppt yr
1 (3.2% yr
1) and a mean value of 195.5  2.9 ppt. HCFC-142b (CH3CClF2) and HCFC-141b (CH3CCl2F) are mainly employed as refrigerants and foam-blowing agents in substitution of CFC-11. HCFC-141b is also used as a solvent in cleaning applications as alternative for CFC-113. The growth rates recorded at Lampedusa for HCFC-141b and HCFC-142b during the 4-year period 2005e2008 are, 0.3 and 1.1 ppt yr
1 respectively (see Fig. 3c and d). HFC-134a (CH2FCF3) is considered as an alternative for refrigeration and air conditioning, medical dose delivery systems, and as a foam blowing agent (O’Doherty et al., 2004). Due to its low impact on ozone depletion its production is increasing. The four years data set recorded at Lampedusa ENEA station is shown in Fig. 3e. A linear

Fig. 5. Hysplit backward trajectory ending at Lampedusa on 11th of April 2008 at 7 UTC.

F. Artuso et al. / Atmospheric Environment 44 (2010) 4944e4953

increasing trend of 4.8 ppt yr
1 is displayed, passing from the 2005 mean value of 43.3  0.3 ppt to 57.6  1.1 ppt in 2008.

2

2

CH Br (ppt)

a

3.0 2.0 1.0

3

CH Br (ppt)

b 10.5 9.0 7.5

3

CH I (ppt)

c 1.5 1.0 0.5 0.0 2007.6

2008.0

2008.4 time (year)

2008.8

2009.2

Fig. 6. Records of (a) Halon-1211 and (b) Halon-1301 mixing ratios and error bars.

dibromomethane (CH2Br2) and methyl iodide (CH3I) (Dimmer et al., 2001; Smythe-Wright et al., 2006). Natural sources of bromocarbons include primarily coastal marine macroalgae and open ocean phytoplankton (Carpenter and Liss, 2000). Antarctic firn air measurements have shown a rapid increase in global CH3Br concentration starting from 1960s, due to anthropic emissions, and correlation with CH2Br2 and CH3I mixing ratios due to a natural common source (Butler et al., 1999). CH3I is the main iodine organic compound in the troposphere and it is almost exclusively produced by marine sources. Lampedusa time series of the three species, reported in Fig. 6, show a seasonal pattern but no mutual correlations. 3.1.6. Halons: Halon-1301, Halon-1211 Halons are bromofluorocarbons of exclusively man-made origin and were extensively used as fire extinguishers. The two halons

6.0 Halon-1211 (ppt)

3.1.5. Halomethanes: CH2Cl2, CH3Cl, CHCl3, CH3Br, CH2Br2, CH3I Halomethanes are short-living species contributing to the stratospheric ozone loss. Dichloromethane (CH2Cl2), methylchloride (CH3Cl) and chloroform (CHCl3) account for about 15% of stratospheric chlorine (Kurylo et al., 1999), while methyl bromide (CH3Br) and dibromomethane (CH2Br2) are among the major bromine carriers. CH2Cl2 emissions are largely anthropogenic due to its employment as solvent while natural CH2Cl2 sources are still uncertain in magnitude and include mainly ocean and biomass burning (Keene et al.,1999). In contrast with CH2Cl2, sources of CH3Cl and CHCl3 are predominantly natural. CH3Cl global natural emissions are related to tropical biomass burning events, ocean processes, tropical plants (Yokouchi et al., 2000) while anthropogenic processes include combustion of fossil fuels and waste. CHCl3 global sources are dominated by the oceanic emissions that represent about 60% of the total budget (Keene et al., 1999). High frequency observations performed within the AGAGE network at Mace Head and Cape Grim showed that CH2Cl2, CH3Cl and CHCl3 concentrations exhibit seasonal cycles with maxima in late winter/early spring and minima in late summer/early autumn (Simmonds et al., 2006). CH2Cl2, CHCl3 and CH3Cl weekly observations at Lampedusa show the same seasonality (see Fig. 4), but a longer period of observations and higher frequency measurements are needed to confirm the seasonal pattern. All those three species present a correlated high mixing ratio event on the 11th of April indicating a common source. To check the origin of the pollution event the corresponding backward airmass trajectory has been analysed. Fig. 5 shows the backward trajectory ending at Lampedusa the 11th of April 2008 at 7 UTC. The 72 h back-trajectory was computed using the Hybrid SingleParticle Lagrangian Integrated Trajectories (HYSPLIT-4) modelling system (Draxler and Rolph, 2003). Airmass origin is localized in North Africa. It starts in Algeria, then passes near the southern border of Tunisia and crosses over the north-western part of Lybia. The trajectory was compared with the oil and gas pipeline map available on the following web site: http://www.theodora.com/pipelines/ world_oil_gas_and_products_pipelines.html. The comparison suggests that large emissions of chlorinated solvents may be originated from Lybia in the north-western part of the country, where numerous industries and gas-petroleum pipelines are located. Among the halomethanes of biogenic origin affecting the Earth’s radiative budget particular concern was recently dedicated to the bromine and iodine carriers, such as methyl bromide (CH3Br),

4.0

Halon-1301 (ppt)

3.1.4. Montreal protocol chlorinated solvents: CH3CCl3, CCl4 Chlorine-containing hydrocarbons, such as CCl4, methylchloroform (CH3CCl3), were extensively employed as solvents in commercial and industrial applications. CH3CCl3 global mixing ratios were growing up at the end of the seventies and reached a maximum in 1992 (Prinn et al., 2005). Results from combined studies of surface observations and transport modelling (Reimann et al., 2005) demonstrated that emissions derived from industrial databases are underestimated. After the 2003 levelling off, the atmospheric CH3CCl3 levels started to decline in consequence to the Montreal Protocol ban. As shown in Fig. 2c, observations at Lampedusa confirm the declining CH3CCl3 trend recording a descending rate of 1.8 ppt yr
1 and a mean value of 12.0  0.4 ppt. In contrast CCl4 exhibits a rising pattern (Fig. 2b) even if its emissions are globally rapidly decreasing. An increasing rate of 5.7 ppt yr
1 and a mean level of 89.7  1.9 ppt are recorded at Lampedusa. Such unexpected behaviour may be justified by the CCl4 relatively long lifetime of 26 years (Clerbaux and Cunnold, 2007) and by its persisting emissions in the North Hemisphere.

4949

a

5.0

4.0

b 3.5 3.0 2.5 2.0 2007.6

2008.0

2008.4 time (year)

2008.8

Fig. 7. Dendrogram resulting from cluster analysis.

2009.2

4950

F. Artuso et al. / Atmospheric Environment 44 (2010) 4944e4953

Table 2 Pearson’s index correlation matrix. Table reports only values >0.50 and with significance at 0.01 level. CFC-11 CFC-12 CFC-11 SF6 HCFC-22 HFC-134a CH3Cl CH3Br HCFC-142b CH3I HCFC-141b CHCl3 CH3CCl3

HCFC-22

SF6

HFC-134a

CH3Br

HCFC-142b

CH3I

CH2Cl2

HCFC-141b

CH3CCl3

0.59 0.62 0.57

0.52 0.78 0.67

0.84

0.60 0.66 0.57

0.55

0.73 0.57 0.57 0.77

1301 and 1211 are of particular concern in climate investigation because they are the major sources of BrOx radicals in the stratosphere that strongly affect the ozone layer. Their decomposition by UV photolysis reduces their lifetimes in the atmosphere to 65 and 16 years for Halon-1301 and 1211 respectively (IPCC, 2007). The short time series of Halons 1301 and 1211 are displayed in Fig. 7. A slightly decreasing trend results from the short-period of analysis of both species. This result is consistent with industry emission estimates that report a decrease in their production due to the compliance to international treaties designed to protect the ozone layer (Butler et al., 1992). Their production was banned by 1994 in most of the developed countries but its release is still sustained by the halon “banks” contained in existing equipment. 3.2. Inter-species correlation analysis To evaluate the characteristics of the N2O and halocompounds common sources, an interspecies correlation analysis was carried out on all 2008 time series. For this purpose Pearson’s indices and cluster analysis were used as investigation means. In Table 2 we report the correlation matrix of all species of interest. Data series

used to calculate the interspecies correlations were previously screened for outliers which can cause misleading results. From our calculation no statistically significant negative correlations were found. Among the positive correlations the ones with Pearson coefficient above 0.5 are all significant at the 0.01 level. As expected CFC-11 is only correlated with CFC-12 and viceversa (þ0.59). The highest correlation was found between HCFC-22 and HFC-134a (þ0.84), remarking the common extensive use as refrigerants and their interchangeability due to their common chemical and physical properties. HCFC-22 also shows an excellent significant relationship with HCFC-142b (þ0.78) and HCFC-141b (þ0.66), confirming that they are originated from the same sources. Good correlation coefficients among SF6 and HFC-134a with HCFCs imply their common production and industrial employment in substitution of ODS. Surprisingly HFC-134a displays a good correlation with CH3I too (þ0.55 with significance at 0.01 level). However the CH3I mixing ratio is not well related with other methyl halides. Such uncorrelated behaviour has also been observed at Cape Grim, Tasmania, within the AGAGE experiment (Cohan et al., 2003). In contrast CH3Cl and CH3Br are strongly interrelated (þ0.73) at Lampedusa because of their common emission sources such as

Fig. 8. Hysplit backward trajectory ending at Lampedusa on 11th of April 2008 at 7 UTC.

F. Artuso et al. / Atmospheric Environment 44 (2010) 4944e4953

ocean, biomass burning and fossil fuel combustion. CFC-11 and CH3CCl3 are probably correlated (þ0.62) because of their concomitant phase out due to the Montreal Protocol ban and their consequent common decline in the atmosphere. A very high correlation (þ0.77) with a significance at 0.01 level exists between CHCl3 and CH2Cl2 because these species are both used as industrial solvents and are naturally emitted from ocean (Cox et al., 2003). To further understand the potential sources of the species analyzed in this work and to compare their chemical characteristics a hierarchical cluster analysis was also performed. The objective of cluster analysis is to sort samples into groups of similar characteristics. There are several ways to define the similarity between samples and there are also different strategies to group objects. In this work squared Euclidean distance was used as clustering method to sort the all analyzed species. The result is shown in Fig. 8, where dendrogram is displayed and the major clusters of samples were identified. Using a cutting distance of 10, only four small groups can be found. The first cluster, with the smallest distance, results from the combination of CH2Cl2 and CHCl3, which are correlated by their common use as solvents. Cluster 2 includes CFC-11, CFC-12 and CH3CCl3. All those species have been banned by the Montreal Protocol and during the last years display a similar descending trend in the atmosphere. The cluster analysis confirms also the relation between CH3Cl and CH3Br (cluster 3), as previously highlighted by the Pearson’s coefficient analysis, due to their natural emissions from combustion processes and from ocean. The fourth group contains HCFC-22, HFC-134a and HCFC- 142b, all CFC substitutes, and CH3I. The presence of CH3I in this cluster and the previously found high correlation between CH3I and HFC-134a suggest that CH3I trend recorded at Lampedusa is not only influenced by natural emissions but also by an unknown anthropogenic source. 3.3. Trajectory analysis An airmass back-trajectory analysis combined with the halocompounds (SF6, HCFC-22, HFC-134a, HCFC-141b, HCFC-142) time series in the period 2005e2008 was performed with the aim to identify the possible source regions leading to elevated concentrations at Lampedusa. Back-trajectories, computed using the HYSPLIT-4 model, were classified on the basis of their residing time in specific geographical regions. Trajectories are driven by meteorological input from the National Centre for Environmental Prediction/National Centre for Atmospheric Research (NCEP/NCAR) database. Deviations of the weekly mixing ratios from the trend line, defined as d, were associated with the corresponding 5-day air mass back trajectories. The trajectories were classified according to the residing time (50% permanence criterion) of the air parcel over each of the following 4 regions: Western Europe (sector EUW); Eastern Europe and Russia (sector EUE); Africa (sector AFR); a sea region (sector SEA), which includes the Mediterranean Sea and the Atlantic Ocean. Origin regions were defined in relation to the distribution of known sources and are the same individuated for CO2 in Artuso et al. (2009). Results are reported in Table 3 and in Fig. 9. Combined analysis of backward airmass trajectories and weekly time series shows that anomalous high mixing ratios correspond to airmasses originating from North-continental area. Higher emissions from Western Europe are detectable for HCFC-142b, while HCFC-22 and SF6 seem to have pronounced sources in Eastern Europe. As expected, the contribution from Northern Africa is negligible, except for SF6. Back trajectory results were compared with the emission data for SF6 and HFCs inferred from the Annual European Community greenhouse gas inventory 1990e2007 and inventory report 2009 (http://www. eea.europa.eu/publications/european-community-greenhouse-gasinventory-2009/).

4951

Table 3 Average value of d, and standard deviation of average, sd, satisfying 50% permanence criterion for each geographical sector. EUW

EUE

AFR

SEA

HFC-134a

d sd

1.45 0.64

1.47 0.83


1.83 0.60


0.27 0.28

0.13 0.13


0.01 0.24


0.10 0.13


0.01 0.06

0.17 0.11

0.14 0.17


0.04 0.13

0.07 0.09

0.52 1.11

2.40 1.50


1.88 0.85

0.41 0.45

0.07 0.05

0.22 0.15

0.07 0.06

0.01 0.02

HCFC-141b

d sd HCFC-142b

d sd HCFC-22

d sd SF6

d sd

According to the emission database, HFCs emissions should prevalently originate from EUW, in particular from France, Germany and UK (26%, 20% and 17% of total EUW emissions respectively during 2007). The back trajectory analysis remarked that large amounts of HFC-134a are released also from the Eastern sector while African contribution was negligible. Conversely SF6 showed a singular behaviour. Despite the difficulty to evaluate the SF6 African contribution because of lack of information from this area, our analysis highlighted the presence of considerable African SF6 emissions. Possible SF6 sources located in North Africa may be linked to leakages from SF6 insulated electrical installations in the industrialized areas. Database records reported that the SF6 major

Fig. 9. Backward trajectories ending at Lampedusa at 7 UT. Map reports 4 geographical sectors defined in the text. Trajectory colour indicates HFC-134a measured value of d. Trajectories associated with
2ppt < d < þ 2ppt were not reported (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.).

4952

F. Artuso et al. / Atmospheric Environment 44 (2010) 4944e4953

source country in EUW is Germany (61%) and in EUE is Turkey (60%). Transport analysis, combined with Lampedusa observations, remarked the presence of an unknown SF6 source located in Eastern Europe. Estimates of HCFC emissions were not available on EEA web site but our study suggested a differentiation of HCFC production in both North continental areas. We supposed that in Western Europe HCFC-141b and HCFC-142b are widely used, while in Eastern Europe prevails the employment of HCFC-22.

Acknowledgments We gratefully acknowledge Dr. Francesco Monteleone from ENEA for the logistic management at Lampedusa Station and Dr. Daniela Romano from Istituto Superiore per la Protezione e la Ricerca Ambientale (ISPRA) for the information provided on GHGs inventories. Dr.Spezzano is acknowledged for his precious suggestions. We also thank the AGAGE team for making available the N2O data recorded at Mace Head and Cape Grim.

4. Conclusions Time series of N2O and halocompounds measured at the Lampedusa ENEA Station in the Mediterranean were analysed and presented in this work. Atmospheric N2O levels showed a linear upward growth rate of about 0.78 ppb yr
1 and mixing ratios in line with observations recorded at Mace Head in the NH. CFC-11 and CFC-12 time series showed a decline due to their phase-out while CFC-113 short records did not allow deriving a significant trend. All CFCs and halons replacing compounds exhibited an increasing trend and the highest growth rates were recorded for SF6 and HFC-134a with values of 0.26 ppt yr
1 (4.3% yr
1) and 4.8 ppt yr
1 (9.6% yr
1) respectively. Industrial chlorinated solvents such as CCl4 and CH3CCl3 showed an opposite behaviour. While a decay was detected for CH3CCl3, an unexpected increasing CCl4 rate of 5.7 ppt yr
1 was recorded due to its relatively long lifetime and persisting emissions. CH2Cl2, CH3Cl, and CHCl3 time series showed seasonality with a maximum in early April and a minimum at the end of September. A correlated high mixing ratio event on the 11th of April 2008 was detected by our observations. Through a backward trajectory analysis the airmass origin corresponding to the pollution event was localized in Northern Africa, suggesting that large emissions of chlorinated solvents may be originated from the Lybian areas in the North-Western part of the country, where industries and gaspetroleum pipelines are located. CH3I, CH3Br and CH2Br2 showed a seasonal pattern but no correlations with each other. From Pearson’s index calculation, positive high correlation was found between HCFC-22 and HFC-134a (þ0.84), highlighting the common extensive employment as refrigerants. Sharing sources such as ocean, biomass burning and fossil fuel combustion probably inferred the high Pearson’s index (þ0.73) between CH3Cl and CH3Br. CHCl3 and CH2Cl2 were also tightly interrelated (þ0.73) because of their usage as industrial solvents and common emissions from ocean. A singular strong relationship (þ0.55) was detected between HFC-134a and CH3I, as confirmed by the cluster analysis, suggesting the influence of an unknown anthropogenic source of CH3I. Combined analysis of airmass back trajectories and observed values, aimed to identify possible source regions, were compared with the EEA emission database. According to HFC emission database the most impacting emission areas is North Europe. However our analysis remarked the release of large amounts of HFC-134a also from the EUE sector. Our observations and airmass transport analysis showed that African SF6 emissions may be considerable and comparable with the European ones. Leakages from SF6 insulated electrical equipments located in the industrialized Northern African areas may justify our observations. Our study prompted the presence of HCFCs persisting European sources with a prevalence of HCFC-141b and HCFC-142b in EUW and HCFC-22 in EUE. We demonstrated that the remote site of Lampedusa may be considered representative of halocompound tropospheric features in the Mediterranean area. The prosecution of the monitoring program and the eventual start up of in-situ continuous observations may therefore offer in the future useful information for constraining atmospheric models in order to derive better source/sink scenarios in this area.

References Artuso, F., Chamard, P., Piacentino, S., di Sarra, A., Meloni, D., Monteleone, F., Sferlazzo, D., Thiery, F., 2007. Atmospheric methane in the Mediterranean: analysis of measurements at the island of Lampedusa during 1995e2005. Atmospheric Environment 41, 3877e3888. Artuso, F., Chamard, P., Piacentino, S., Sferlazzo, D., De Silvestri, L., di Sarra, A., Meloni, D., Monteleone, F., 2009. Influence of transport and trends in atmospheric CO2 at Lampedusa. Atmospheric Environment 43 (19), 3044e3051. Butler, J.H., Elkins, J.W., Hall, B.D., Cummings, S.O., Montzka, S.A., 1992. A decrease in the growth rates of atmospheric halon concentrations. Nature 359, 403e405. Butler, J.H., Battle, M., Bender, M.L., Montzka, S.A., Clarke, A.D., Saltzman, E.S., Sucher, C.M., Severinghaus, J.P., Elkins, J.W., 1999. A record of atmospheric halocarbons during the twentieth century from polar air. Nature 399, 749e755. Carpenter, L.J., Liss, P.S., 2000. On temperate sources of bromoform and other reactive organic bromine gases. Journal of Geophysical Research 105, 20,539e20,547. Chamard, P., Thiery, F., di Sarra, A., Ciattaglia, L., De Silvestri, L., Grigioni, P., Monteleone, F., Piacentino, S., 2003. Interannual variability of atmospheric CO2 in the Mediterranean: measurements at the island of Lampedusa. Tellus 55B, 83e93. Cohan, D.S., Sturrock, G.A., Biazar, A.P., Fraser, P.J., 2003. Atmospheric methyl iodide at Cape Grim, Tasmania, from AGAGE observations. Journal of Atmospheric Chemistry 44 (2), 131e150. Conway, T.J., Tans, P.P., Waterman, L.S., Thoning, K.W., 1994. Evidence for the interannual variability of the carbon cycle from the National Oceanic and Atmospheric Administration/Climate Monitoring and Diagnostics Laboratory Global Air Sampling Network. Journal of Geophysical Research 99 (D11), 22,831e22,855. Cicerone, R.J., 1989. Analysis of Sources and Sinks of Atmospheric Nitrous Oxide (N2O). Journal of Geophysical Research 94 (D15), 18,265e18,271. Clerbaux, C., Cunnold, D.M., 2007. Long-lived compounds. In: Ajavon, A.-L.N., Albritton, D.L., Watson, R.T. (Eds.), Scientific Assessment of Ozone Depletion: 2006. World Meteorological Organization, Geneva (Chapter 1). Cox, M.L., Sturrock, G.A., Fraser, P.J., Siem, S.T., Krummel, P.B., O’Doherty, S., 2003. Regional sources of methyl chloride, chloroform a dichloromethane identified from AGAGE observations at Cape Grim, Tasmania, 1998e2000. Journal of Atmospheric Chemistry 45 (1), 79e99. Derwent, R.G., Simmonds, P.G., Greally, B.R., O’Doherty, S., McCulloch, A., Manning, A., Reimann, S., Folini, D., Vollmer, M.K., 2007. The phase-in and phase-out of European emissions of HCFC-141b and HCFC-142b under the Montreal Protocol: evidence from observations at Mace Head, Ireland and Jungfraujoch, Switzerland from 1994 to 2004. Atmospheric Environment 41, 757e767. Dimmer, C.H., Simmonds, P.G., Nickless, G., Bassford, M.R., 2001. Biogenic fluxes of halomethanes from Irish peatland ecosystems. Atmospheric Environment 35 (2), 321e330. Draxler, R.R., Rolph, G.D., 2003. HYSPLIT (HYbrid Single-Particle Lagrangian Integrated Trajectory) Model access via NOAA ARL READY Website. NOAA Air Resources Laboratory, Silver Spring, MD. http://www.arl.noaa.gov/ready/ hysplit4.html. European Communities, 2000. Regulation (EC) No 2037/2000 of the European Parliament and of the Council of 29 June 2000 on substances that deplete the ozone layer. Official Journal of the European Communities L 244, 1e24. Farman, J.C., Gardiner, B.G., Shanklin, J.D., 1985. Large losses of total ozone in Antarctica reveal seasonal ClOx/NOx interaction. Nature 315 (6016), 207e210. Forster, P.M.D., Burkholder, J.B., Clerbaux, C., Coheur, P.F., Dutta, M., Gohar, L.K., Hurley, M.D., Myhre, G., Portmann, R.W., Shine, K.P., Wallington, T.J., Wuebbles, D., 2005. Resolution of the uncertainties in the radiative forcing of HFC-134a. Journal of Quantitative Spectroscopy and Radiative Transfer 93 (4), 447e460. Hall B., Engel A., Weiss R.F., Elkins J.W., Muhle J., Harth C., Salameh P.K., Rhoderick J., Miller W., Happel J., Atlas E., O’Doherty S.J., Simmonds P.J., Greally B.R., Saltzman E., Aydin M., Blake D., Vollmer M., Montzka S., Maione M., Yochuchi Y., Fraser P., Krummel P., Scheel H.E., Artuso F., Worthy D., Brunke E.G., Levin I., Loewenstein M., 2008. The International Halocarbons in Air Comparison Experiment: first results. Proceedings of the AGU fall meeting, San Francisco. Intergovernmental Panel on Climate Change (IPCC), Contribution of Working Group I to the Fourth Assessment Report 2007. Forster, P., Ramaswamy, V., Artaxo, P., Berntsen, T., Betts, R., Fahey, D.W., Haywood, J., Lean, J., Lowe, D.C., Myhre, G., Jnganga, Prinn, R., Raga, G., Schulz, M., Van Dorland, R., 2007. Changes in

F. Artuso et al. / Atmospheric Environment 44 (2010) 4944e4953 atmospheric constituents and in radiative forcing. In: Solomon, S., et al. (Eds.), Climate Change 2007: The Physical Science Basis. Cambridge Univ. Press, Cambridge, UK. Keene, W.C., Khalil, M.A.K., Erickson, D.J., McCulloch, A., Graedel, T.E., Lobert, J.M., Aucott, M.L., Gong, S.L., Harper, D.B., Kleiman, G., Midgley, P., Moore, R.M., Seuzaret, C., Sturges, W.T., Benkovitz, C.M., Koropalov, V., Barrie, L.A., Li, Y.F., 1999. Composite global emissions of reactive chlorine from anthropogenic and natural sources, reactive chlorine emissions inventory. Journal of Geophysical Research 104, 8429e8440. Kurylo, M.J., Rodriguez, J.M., Andreae, M.O., Atlas, E.L., Blake, D.R., Butler, J.H.S., Lary, D.J., Midgeley, P.M., Montzka, S.A., Novelli, P.C., Reeves, C.E., Simmonds, P.G., Steele, P.L., Sturges, W.T., Weiss, R.F., Yokouchi, Y., 1999. Short-Lived Ozone-Related Compounds, Chap. 2 in Scientific Assessment of Ozone Depletion: 1998, WMO Global Ozone Research and Monitoring Project e Report No. 44, 2.1e2.56. Maiss, M., Brenninkmeijer, C.A.M., 1998. Atmospheric SF6: trends, sources, and prospects. Environmental Science and Technology 32 (20), 3077e3086. Molina, M.J., Rowland, F.S., 1974. Stratospheric sink for chlorofluoromethanes: chlorine atom catalyzed destruction of ozone. Nature 249, 810. Montzka, S.A., et al., 1999. Present and future trends in the atmospheric burden of ozone-depleting halogens. Nature 398, 690e694. O’Doherty, S., Cunnold, D.M., Manning, A., Miller, B.R., Wang, R.H.J., Krummel, P.B., Fraser, P.J., Simmonds, P.G., McCulloch, A., Weiss, R.F., Salameh, P., Porter, L.W., Prinn, R.G., Huang, J., Sturrock, G., Ryall, D., Derwent, R.G., Montzka, 2004. Rapid growth of hydrofluorocarbon 134a and hydrochlorofluorocarbons 141b, 142b, and 22 from Advanced Global Atmospheric Gases Experiment (AGAGE) observations at Cape Grim, Tasmania, and Mace Head, Ireland. Journal of Geophysical Research Atmospheres 109 (D6) Art. No. D06310. Prinn, R.G., Cunnold, D.M., Rasmussen, R., Simmonds, P.G., Alyea, F.N., Crawford, A., Fraser, P.J., Rosen, R., 1990. Atmospheric emissions and trends of nitrous oxide deduced from ten years of ALE-GAGE data. Journal of Geophysical Research 95, 18369e18385. Prinn, R.G., Huang, J., Weiss, R.F., Cunnold, D.M., Fraser, P.J., Simmonds, P.G., McCulloch, A., Harth, C., Reimann, S., Salameh, P., O’Doherty, S., Wang, R.H.J., Porter, L., Miller, B.R., Krummel, P.B., 2005. Evidence for variability of atmospheric hydroxyl radicals over the past quarter century. Geophysical Research Letters 32 (7), L07809.

4953

Ravishankara, A.R., Solomon, S., Turnipseed, A.A., Warren, R.F., 1993. Atmospheric lifetimes of long-lived halogenated species. Science 259, 194e199. Reimann, S., Manning, A.J., Simmonds, P.G., Cunnold, D.M., Wang, R.H.J., Li, J., McCulloch, A., Prinn, R.G., Huang, J., Weiss, R.F., Fraser, P.J., O’Doherty, S., Greally, B.R., Stemmler, K., Hill, M., Folin, D., 2005. Low methyl chloroform emissions inferred from long-term atmospheric measurements. Nature 433, 506e508. doi:10.1038/nature03220. Simmonds, P.G., Cunnold, D.M., Weiss, R.F., Prinn, R.G., Fraser, P.J., McCulloch, A., Alyea, F.N., O’Doherty, S., 1998. Global trends and emission estimates of CCl4 from in situ background observations from July 1978 to June 1996. Journal of Geophysical Research 103, 16017e16027. Simmonds, P.G., Manning, A.J., Cunnold, D.M., McCulloch, A., O’Doherty, S., Derwent, R.G., Krummel, P.B., Fraser, P.J., Dunse, B., Porter, L.W., Wang, R.H.J., Greally, B.R., Miller, B.R., Salameh, P., Weiss, R.F., Prinn, R.G., 2006. Global trends, seasonal cycles, and European emissions of dichloromethane, trichloroethene, and tetrachloroethene from the AGAGE observations at Mace Head, Ireland, and Cape Grim, Tasmania. Journal of Geophysical Research-Atmospheres 111 (D18), D18304. Smythe-Wright, D., Boswell, S.M., Breithaupt, P., Davidson, R.D., Dimmer, C.H., Eiras Diaz, L.B., 2006. Methyl iodide production in the ocean: implications for climate change. Global Biogeochemical Cycles 20, GB3003. doi:10.1029/2005GB002642. UNEP, 2001. Backgrounder. Basic Facts and Data on the Science and Politics of Ozone Protection. United Nations Environment Programme, Ozone Secretariat, P.O. Box 30552, Nairobi, Kenya. Yokouchi, Y., Noijiri, Y., Barrie, L.A., Toom-Sauntry, D., Machida, T., Inuzuka, Y., Akimoto, H., Li, H.-J., Fuginuma, Y., Aoki, S., 2000. A strong source of methyl chloride to the atmosphere from tropical coastal land. Nature 403, 295e298. Wolff, E., Spahni, R., 2007. CH4 and N2O in the ice core record. Philosophical Transactions of the Royal Society-A Mathematical Physical and Engineering Sciences 365 (1856), 1775e1792. World Meteorological Organization (WMO), 1988. The Montreal protocol on substances that deplete the ozone layer. WMO Bulletin 37, 94e97. World Meteorological Organization Global Atmosphere Watch No. 168, 2005. Report of the 13th WMO/IAEA Meeting of Experts on CO2 concentration and relatedtracers measurement techniques. Boulder, Colorado, USA, 19e22 September.