history of chemically and radiatively important atmospheric gases ... - mit global … · 2019. 1....

The MIT Joint Program on the Science and Policy of Global Change combines cutting-edge scientific research with independent policy analysis to provide a solid foundation for the public and private decisions needed to mitigate and adapt to unavoidable global environmental changes. Being data-driven, the Joint Program uses extensive Earth system and economic data and models to produce quantitative analysis and predictions of the risks of climate change and the challenges of limiting human influence on the environment—essential knowledge for the international dialogue toward a global response to climate change.

To this end, the Joint Program brings together an interdisciplinary group from two established MIT research centers: the Center for Global Change Science (CGCS) and the Center for Energy and Environmental Policy Research (CEEPR). These two centers—along with collaborators from the Marine Biology Laboratory (MBL) at

Woods Hole and short- and long-term visitors—provide the united vision needed to solve global challenges.

At the heart of much of the program’s work lies MIT’s Integrated Global System Model. Through this integrated model, the program seeks to discover new interactions among natural and human climate system components; objectively assess uncertainty in economic and climate projections; critically and quantitatively analyze environmental management and policy proposals; understand complex connections among the many forces that will shape our future; and improve methods to model, monitor and verify greenhouse gas emissions and climatic impacts.

This reprint is intended to communicate research results and improve public understanding of global environment and energy challenges, thereby contributing to informed debate about climate change and the economic and social implications of policy alternatives.

—Ronald G. Prinn and John M. Reilly, Joint Program Co-Directors

MIT Joint Program on the Science and Policy of Global Change

Massachusetts Institute of Technology 77 Massachusetts Ave., E19-411 Cambridge MA 02139-4307 (USA)

T (617) 253-7492 F (617) 253-9845 [email protected] http://globalchange.mit.edu

Reprint 2018-16

Reprinted with permission from Earth System Science Data, 10: 985–1018. © 2018 the authors

History of chemically and radiatively important atmospheric gases from the Advanced Global Atmospheric Gases Experiment (AGAGE)R.G. Prinn, R.F. Weiss, J. Arduini, T. Arnold, H.L. DeWitt, P.J. Fraser, A.L. Ganesan, J. Gasore, C.M. Harth, O. Hermansen, J. Kim, P.B. Krummel, S. Li, Z.M. Loh, C.R. Lunder, M. Maione, A.J. Manning, B.R. Miller, B. Mitrevski, J. Mühle, S. O’Doherty, S. Park, St. Reimann, M. Rigby, T. Saito, P.K. Salameh, R. Schmidt, P.G. Simmonds, L.P. Steele, M.K. Vollmer, R.H. Wang, B. Yao, Y. Yokouchi, D. Young, and L. Zhou H. Lee, X. Liu, Z. Lu, S. Albani and C. Wang

mailto:globalchange%40mit.edu?subject=http://globalchange.mit.edu

Earth Syst. Sci. Data, 10, 985–1018, 2018https://doi.org/10.5194/essd-10-985-2018© Author(s) 2018. This work is distributed underthe Creative Commons Attribution 4.0 License.

History of chemically and radiatively importantatmospheric gases from the Advanced Global

Atmospheric Gases Experiment (AGAGE)

Ronald G. Prinn1, Ray F. Weiss2, Jgor Arduini3, Tim Arnold4, H. Langley DeWitt1, Paul J. Fraser5,Anita L. Ganesan6, Jimmy Gasore7, Christina M. Harth2, Ove Hermansen8, Jooil Kim2,

Paul B. Krummel5, Shanlan Li9, Zoë M. Loh5, Chris R. Lunder8, Michela Maione3,Alistair J. Manning10,11, Ben R. Miller12, Blagoj Mitrevski5, Jens Mühle2, Simon O’Doherty11,

Sunyoung Park9, Stefan Reimann13, Matt Rigby11, Takuya Saito14, Peter K. Salameh2,Roland Schmidt2, Peter G. Simmonds6, L. Paul Steele5, Martin K. Vollmer13, Ray H. Wang15, Bo Yao16,

Yoko Yokouchi14, Dickon Young11, and Lingxi Zhou161Center for Global Change Science, Massachusetts Institute of Technology, Cambridge, MA, USA

2Scripps Institution of Oceanography, University of California San Diego, La Jolla, CA, USA3Department of Pure and Applied Sciences, University of Urbino, Urbino, Italy

4National Physical Laboratory, Teddington, Middlesex, UK andSchool of GeoSciences, University of Edinburgh, Edinburgh, UK

5Climate Science Centre, Oceans and Atmosphere, Commonwealth Scientificand Industrial Research Organization (CSIRO), Aspendale, Victoria, Australia

6School of Geographical Sciences, University of Bristol, Bristol, UK7Rwanda Climate Observatory Secretariat, Ministry of Education of Rwanda, Kigali, Rwanda

8Norwegian Institute for Air Research (NILU), Kjeller, Norway9Department of Oceanography, Kyungpook National University, Daegu, Republic of Korea

10Hadley Centre, The Met Office, Exeter, UK11School of Chemistry, University of Bristol, Bristol, UK

12National Oceanic and Atmospheric Administration (NOAA), Earth System Research Laboratory,Boulder, CO, USA

13Laboratory for Air Pollution and Environmental Technology (Empa),Swiss Federal Laboratories for Materials Science and Technology, Dübendorf, Switzerland

14National Institute for Environmental Studies (NIES), Tsukuba, Japan15Georgia Institute of Technology, Atlanta, GA, USA

16China Meteorological Administration (CMA), Beijing, China

Correspondence: Ronald G. Prinn ([email protected])

Received: 5 December 2017 – Discussion started: 4 January 2018Revised: 15 April 2018 – Accepted: 27 April 2018 – Published: 6 June 2018

Abstract. We present the organization, instrumentation, datasets, data interpretation, modeling, and accom-plishments of the multinational global atmospheric measurement program AGAGE (Advanced Global Atmo-spheric Gases Experiment). AGAGE is distinguished by its capability to measure globally, at high frequency,and at multiple sites all the important species in the Montreal Protocol and all the important non-carbon-dioxide(non-CO2) gases assessed by the Intergovernmental Panel on Climate Change (CO2 is also measured at severalsites). The scientific objectives of AGAGE are important in furthering our understanding of global chemical andclimatic phenomena. They are the following: (1) to accurately measure the temporal and spatial distributions ofanthropogenic gases that contribute the majority of reactive halogen to the stratosphere and/or are strong infrared

Published by Copernicus Publications.

986 R. G. Prinn et al.: History of environmentally important atmospheric gases

absorbers (chlorocarbons, chlorofluorocarbons – CFCs, bromocarbons, hydrochlorofluorocarbons – HCFCs, hy-drofluorocarbons – HFCs and polyfluorinated compounds (perfluorocarbons – PFCs), nitrogen trifluoride – NF3,sulfuryl fluoride – SO2F2, and sulfur hexafluoride – SF6) and use these measurements to determine the globalrates of their emission and/or destruction (i.e., lifetimes); (2) to accurately measure the global distributions andtemporal behaviors and determine the sources and sinks of non-CO2 biogenic–anthropogenic gases important toclimate change and/or ozone depletion (methane – CH4, nitrous oxide – N2O, carbon monoxide – CO, molecularhydrogen – H2, methyl chloride – CH3Cl, and methyl bromide – CH3Br); (3) to identify new long-lived green-house and ozone-depleting gases (e.g., SO2F2, NF3, heavy PFCs (C4F10, C5F12, C6F14, C7F16, and C8F18) andhydrofluoroolefins (HFOs; e.g., CH2=CFCF3) have been identified in AGAGE), initiate the real-time monitor-ing of these new gases, and reconstruct their past histories from AGAGE, air archive, and firn air measurements;(4) to determine the average concentrations and trends of tropospheric hydroxyl radicals (OH) from the ratesof destruction of atmospheric trichloroethane (CH3CCl3), HFCs, and HCFCs and estimates of their emissions;(5) to determine from atmospheric observations and estimates of their destruction rates the magnitudes and distri-butions by region of surface sources and sinks of all measured gases; (6) to provide accurate data on the global ac-cumulation of many of these trace gases that are used to test the synoptic-, regional-, and global-scale circulationspredicted by three-dimensional models; and (7) to provide global and regional measurements of methane, car-bon monoxide, and molecular hydrogen and estimates of hydroxyl levels to test primary atmospheric oxidationpathways at midlatitudes and the tropics. Network Information and Data Repository: http://agage.mit.edu/dataor http://cdiac.ess-dive.lbl.gov/ndps/alegage.html (https://doi.org/10.3334/CDIAC/atg.db1001).

1 Introduction

The Advanced Global Atmospheric Gases Experiment(AGAGE: 1993–present) and its predecessors (AtmosphericLifetime Experiment, ALE: 1978–1981; Global Atmo-spheric Gases Experiment, GAGE: 1982–1992) have mea-sured the greenhouse gas and ozone-depleting gas composi-tion of the global atmosphere continuously since 1978. TheALE program was instigated to measure the then five majorozone-depleting gases (CFC-11 (CFCl3), CFC-12 (CCl2F2),CCl4, CH3CCl3, N2O) in the atmosphere four times per dayusing automated gas chromatographs with electron-capturedetectors (GC-ECDs) at four stations around the globe andto determine the atmospheric lifetimes of the purely anthro-pogenic of these gases from their measurements and indus-try data on their emissions (Prinn et al., 1983a). The GAGEproject broadened the global coverage to five stations, thenumber of gases being measured to eight (adding CFC-113 (CCl2FCClF2), CHCl3, and CH4 to the ALE list), andthe frequency to 12 per day by improving the GC-ECDsand adding gas chromatographs with flame-ionization de-tectors (GC-FIDs; Prinn et al., 2000). The AGAGE pro-gram then significantly improved upon the GAGE instru-ments by increasing their measurement precision and fre-quency (to 36 per day) and adding gas chromatographs withmercuric oxide reduction detectors, to measure 10 biogenicand/or anthropogenic gases overall (adding H2 and CO tothe GAGE list). AGAGE also introduced powerful new gaschromatographs with mass spectrometric detection and cryo-genic pre-concentration measuring over 50 trace gases 20times per day. In this overview paper, while we address theentire 1978–present database and its public availability, we

focus more on the evolution of the network after 2000; de-tails of the period before that are addressed in the previouscomprehensive overviews provided by Prinn et al. (2000) andPrinn et al. (1983a). The case for high-frequency measure-ment networks like AGAGE with data available to operatorsin real time is strong, and the observations and their interpre-tation are important inputs to the scientific understanding ofozone depletion and climate change. AGAGE is character-ized by its capability to measure globally the trends at highfrequency and estimate emissions from these trends for allof the important species in the Montreal Protocol on Sub-stances that Deplete the Ozone Layer, and all of the impor-tant non-carbon-dioxide (non-CO2) trace gases assessed bythe Intergovernmental Panel on Climate Change. More re-cently, AGAGE has also been measuring CO2 using high-frequency optical spectroscopy (focusing on sites where suchmeasurements are not made by other groups; Sect. 2.3 and2.4). The scientific objectives of AGAGE (summarized inthe Abstract) are of considerable significance in furtheringour understanding of important global chemical and climaticphenomena. The remainder of this Introduction is devotedto describing the network of stations (Sect. 1.1), the mea-surements (Sect. 1.2), and the place of AGAGE in the globalobserving system (Sect. 1.3). Then Sect. 2 addresses the in-strumentation, calibration, and station infrastructure, Sect. 3the data analysis and modeling, Sect. 4 the scientific accom-plishments, and Sect. 5 the AGAGE data availability.

1.1 A Global network of stations

The ALE/GAGE/AGAGE stations are coastal or mountainsites around the world, chosen primarily to provide accurate

Earth Syst. Sci. Data, 10, 985–1018, 2018 www.earth-syst-sci-data.net/10/985/2018/

http://agage.mit.edu/datahttp://cdiac.ess-dive.lbl.gov/ndps/alegage.htmlhttps://doi.org/10.3334/CDIAC/atg.db1001

R. G. Prinn et al.: History of environmentally important atmospheric gases 987

Ny-Alesund(Norway)

Ny-Alesund

(Svalbard)

Monte Cimone

(Italy)

Ragged Point

(Barbados)

Hateruma

(Japan) Cape Matatula

(American Samoa)

Cape Grim

(Tasmania)

Mace Head

(Ireland)

Trinidad Head

(California)

Jungfraujoch

(Switzerland)

AGAGE measurement stations collaborative measurement stations

Gosan

(Korea)

Shangdianzi

(China)

Mount Mugogo(Rwanda)

Cape Ochiishi(Japan)

AGAGEMedusastationsAGAGEaffiliatestations

o

Figure 1. Locations of the 10 current AGAGE primary stations (red highlighted stations) that have Medusa gas chromatograph–mass spec-trometer (GC-MS) instruments and the 3 current AGAGE affiliate stations (green highlighted stations) that have alternative pre-concentrationGC-MS instruments. AGAGE and the other major global air sampling network, NOAA-ESRL-GMD, are independent but closely cooperat-ing, including frequent data intercomparisons, especially at the American Samoa shared site.

measurements of trace gases whose lifetimes are long com-pared to global atmospheric circulation times (Fig. 1). The10 “primary” AGAGE stations that all share common cal-ibrations and gas chromatographic–mass spectrometric in-strumentation (see Sect. 1.2) are the following: (a) on Ire-land’s west coast, first at Adrigole (52◦ N, 10◦W; 50 m (inletheight a.s.l. here and for all other stations), 1978–1983), thenat Mace Head (53◦ N, 10◦W; 25 m 1987 to present); (b) onthe US west coast, first at Cape Meares, Oregon (45◦ N,124◦W; 30 m, 1979–1989), then at Trinidad Head, Califor-nia (41◦ N, 124◦W; 140 m, 1995 to present); (c) at RaggedPoint, Barbados (13◦ N, 59◦W; 42 m, 1978 to present); (d) atCape Matatula, American Samoa (14◦ S, 171◦W; 77 m, 1978to present); (e) at Cape Grim, Tasmania, Australia (41◦ S,145◦ E; 164 m, 169 m, 1978 to present); (f) on the Jungfrau-joch, Switzerland (47◦ N, 8◦ E; 3580 m, 2000 to present);(g) on Zeppelin Mountain, Ny-Ålesund, Svalbard, Norway(79◦ N, 12◦ E; 489 m, 2001 to present); (h) at Gosan, JejuIsland, Korea (33◦ N, 126◦ E; 89 m, 2007 to present); (i) atShangdianzi, China (41◦ N, 117◦ E; 383 m, 2010 to presentwith gap) and (j) Mt. Mugogo, Rwanda (1.6◦ S, 29.6◦ E;2640 m, 2015 to present). The AGAGE network also includesthree AGAGE-compatible (but not identical) instruments inthe following locations: (k) Hateruma Island, Japan (24◦ N,123.8◦ E; 47 m, 2004 to present); (l) Cape Ochiishi, Japan(43◦ N, 145.5◦ E; 100 m, 2006 to present), and (m) Monte Ci-mone, Italy (44◦ N, 10◦ E; 2165 m, 2004 to present). Theseare called AGAGE “affiliate” stations in Fig. 1. There arealso “secondary”, usually continental and some urban, sta-

tions that are linked to and complement the primary and af-filiate stations (discussed below).

1.2 Measurements

At its primary stations, AGAGE uses in situ gas chromatog-raphy with mass spectrometry (GC-MS) in the “Medusa”system (Miller et al., 2008; Arnold et al., 2012) to measureover 50 largely synthetic gases including hydrochlorofluoro-carbons (e.g., HCFC-22; CHClF2) and hydrofluorocarbons(e.g., HFC-134a; CH2FCF3), which are interim or long-termalternatives to chlorofluorocarbons (CFCs) now restricted bythe Montreal Protocol, other hydrohalocarbons (e.g., methylchloride; CH3Cl), halons (e.g., Halon-1211; CBrClF2), per-fluorocarbons (e.g., PFC-14; CF4), and trace chlorofluoro-carbons, all of which, except CH3Cl, are involved in theMontreal or Kyoto Protocol. Affiliate stations use similar butnot identical cryogenic pre-concentration GC-MS systems(Maione et al., 2013; Yokouchi et al., 2006).

At its Mace Head, Trinidad Head, Ragged Point, CapeMatatula, and Cape Grim primary stations, AGAGE also usesin situ gas chromatographs (GC) with electron-capture de-tection (ECD), flame-ionization detection (FID), mercuricoxide reduction detection (MRD, at Mace Head and CapeGrim only), and pulsed discharge detection (PDD, at CapeGrim only) to measure five biogenic–anthropogenic gases(methane – CH4, nitrous oxide – N2O, and chloroform –CHCl3 at all sites; carbon monoxide – CO and hydrogen– H2 at Mace Head and Cape Grim only) and five anthro-pogenic gases at all five sites: CFC-11 (CCl3F), CFC-12(CCl2F2), and CFC-113 (CCl2FCClF2), methyl chloroform

www.earth-syst-sci-data.net/10/985/2018/ Earth Syst. Sci. Data, 10, 985–1018, 2018


(CH3CCl3), and carbon tetrachloride (CCl4) 36 times per day(Prinn et al., 2000). The list of gases measured with these gaschromatography “multidetector” (GC-MD) systems includesthe three major chlorofluorocarbons (CFCs) restricted by theMontreal Protocol and the four major long-lived non-CO2greenhouse gases (GHGs). Table 1 lists all the major gasesbeing measured in AGAGE using the Medusa GC-MS andGC-MD instruments, their 2016 global average mole frac-tions, and their typical measurement precisions.

The precisions for each species are determined from theinterspersed measurements of the on-site station calibrationtanks and are reported along with the mole fractions of theinterspersed atmospheric measurements in the AGAGE dataarchives. In general the precisions in Table 1 are highest(< 0.1 %) for the species with the highest absolute mole frac-tions and lowest (∼ 10 %) for those with the lowest molefractions; there are also more subtle differences dependingon the species behavior in the trapping (Medusa), separa-tion (GC), and detection (MS, MD; ECD, FID, MRD) stages.The accuracy of the measurements is determined by calibra-tion scale and tertiary tank accuracies that are discussed inSect. 2.6.

Recent developments have enabled precise analyses ofCH4, CO2, CO, and N2O by laser spectroscopic detectionto begin in AGAGE. These optical instruments are now ex-panding the measurement capabilities within AGAGE, andthere are advantages in switching from the GC-MD approachfor measuring CH4, N2O, and CO to these less operationallydemanding optical spectroscopy methods resulting in near-continuous measurements of comparable or better precision.As discussed in Sect. 2.3 and 2.4, this transition is happeningalready at several AGAGE stations. The GC-MD and opticalspectroscopy instruments will follow the AGAGE protocolused for all cases in which a new improved instrument re-places an earlier one; namely, the two instruments are runtogether for at least several months (and years for gases cur-rently measured on both the GC-MD and Medusa GC-MS)to ensure data comparability and verify improvements.

Each instrument system is automated and under computercontrol. All chromatograms, instrumental data, and operatorlogs are transmitted via the internet to the data processingsites. AGAGE includes timely public archiving and publica-tion of all data, regular intercomparisons of AGAGE mea-surements, absolute calibrations with other networks (e.g.,NOAA’s Global Monitoring Division, GMD), and contribu-tions to national and international assessments of ozone de-pletion and climate change. The data are calibrated againston-site air standards, which are calibrated relative to off-siteparent standards before and after use at each station. AGAGEdepends upon well-defined absolute gravimetric calibrationprocedures that are repeated periodically to ensure the accu-racy of the long-term measured trends (Prinn et al., 2000).

To emphasize the need for very frequent real-time mea-surements we show data for several trace gases (Fig. 2a–d)for the years 2004 and 2016. These GC-MD and GC-MS

data demonstrate the existence of regional pollution-inducedor local sink-induced (e.g., for H2; shown in red) and large-scale transport-induced (shown in black) variability, whichare not captured with weekly flask measurements typicallydesigned to avoid local pollution. Our approach for identi-fying these pollution events is discussed in Sect. 3.1. Notealso the evolution of the sizes of these pollution events be-tween 2004 and 2016 associated with the decreases in theemissions of regulated gases and the growth of emissions ofunregulated ones. This high-frequency sampling enables thepollution events in particular to be used to estimate emis-sions from nearby source regions (e.g., Cape Grim stationfor SE Australian emissions; e.g., Dunse et al., 2005; Stohlet al., 2009; O’Doherty et al., 2009; Fraser et al., 2014; Luntet al., 2015), Trinidad Head for the west coast US emis-sions (e.g., Li et al., 2005; O’Doherty et al., 2009; Lunt etal., 2015; Fortems-Cheiney et al., 2015), Mace Head and theother European stations for European and in some cases east-ern USA emissions (e.g., O’Doherty et al., 2009; Stohl et al.,2009; Keller et al., 2012; Simmonds et al., 2015; Lunt et al.,2015; Fortems-Cheiney et al., 2015; Graziosi et al., 2017),and Hateruma, Shangdianzi, and Gosan for East Asian emis-sions (e.g., Stohl et al., 2009, 2010; Kim et al., 2010; Li etal., 2011; Yao et al., 2012a, b; Saito et al., 2015; Fang et al.,2015; Lunt et al., 2015; Fortems-Cheiney et al., 2015). Thesources of many anthropogenic and natural trace gases mea-sured in AGAGE are often colocated so that measurementof a wide range of gases enhances the ability to accuratelyestimate their sources and sinks. The AGAGE data in graph-ical and digital forms are available for most stations at theAGAGE website: http://agage.mit.edu (last access: 21 May2018) (Sect. 3.2).

1.3 Integral element of the global observing system

AGAGE is part of a powerful complementary observing sys-tem that is measuring various aspects of the evolving com-position of Earth’s atmosphere and providing the fundamen-tal understanding needed to preserve this vital sphere of lifeon our planet. Sharing the AGAGE surface-based perspec-tive are, for example, the remote-sensing Network for Detec-tion of Atmospheric Composition Change (NDACC; see DeMazière et al., 2018) supported by NASA and other agenciesand nations (AGAGE is an NDACC Cooperating Network)and the NOAA-ESRL Global Monitoring Division in situand flask networks. AGAGE contributes to the World Meteo-rological Organization’s Global Atmosphere Watch (WMO-GAW) and regularly provides its data to the WMO-GAW’sWorld Data Center for Greenhouse Gases (WDCGG) web-site (see Sect. 5). AGAGE European stations provide data tothe Integrated Carbon Observation System (ICOS)that coor-dinates pan-European observations of GHGs, and Monte Ci-mone, Jungfraujoch, and Ny Ålesund are now formally join-ing. Also measuring atmospheric composition (as columnprofiles or abundances) are instruments onboard the NASA


http://agage.mit.edu


Table 1. Primary AGAGE measured species using Medusa GC-MS and GC-MD systems. Gases measured with Medusa GC-MS and GC-MD only are in black regular font; those measured with both systems are in italic font. Calibrations are on AGAGE SIO gravimetric scales(Sect. 2.6) unless otherwise noted.

Compound Global mean 2016 Typical Compound Global mean 2016 Typicalconc. (pptc) precision (%) conc. (pptc) precision (%)

PFC-14 82.7 0.15 CFC-114h 16.3 0.3PFC-116 4.56 1 CFC-115 8.48 0.7PFC-218 0.63 3 Halon-1211 3.59 0.4PFC-c318 1.56 1.5 Halon-1301 3.37 1.7PFC-5-1-14 0.31 3 Halon-2402 0.41 2SF6 8.88 0.6 CH3Cl 552 0.2SF5CF3 0.17 7 CH3Br 6.96 0.6SO2F2 2.26 2 CH3Ie 0.58 2NF3 1.44 1 CH2Cl2 31.1 0.5HFC-23 28.9 0.7 CH2Br2e 1.08 1.5HFC-32 12.6 3 CHCl3 8.78 0.4HFC-134a 89.3 0.5 CHBr3e 1.84 0.6HFC-152a 6.71 1.4 CCl4 79.9 1HFC-125 20.8 0.7 CH3CCl3 2.61 0.7HFC-143a 19.3 1 CHCl=CCl2 ∼ 0.11 3HFC-227ea 1.24 2.2 CCl2=CCl2e 1.07 0.5HFC-236fa 0.15 10 COSe 543 0.5HFC-245fa 2.42 3 C2H6d 586 0.3HFC-365mfc 1.00 5 C3H8f 9.04 0.6HFC-43-10mee 0.27 3 C6H6d 17.9 0.3HCFC-22 237 0.3 C7H8d 4.19 0.6HCFC-141b 24.5 0.5HCFC-142b 22.6 0.4HCFC-124d 1.11 2 GC-MD only (ppbc)CFC-11 230 0.2 CH4 1842 0.2CFC-12 516 0.1 N2O 329.3 0.05CFC-13g 3.28 2 COa 54 to 115 0.2CFC-113 71.4 0.2 H2a 515 to 550 0.6 (0.08)b

a CO and H2 measured at Mace Head and Cape Grim only (range for annual means of these two stations given). b GC-PDD system at Cape Grim.c ppt: parts per trillion and ppb: parts per billion. d Preliminary (AGAGE) scale (Sect. 2.6), e preliminary (transfer of NOAA) scale (Sect. 2.6),f preliminary (Empa) scale (Sect. 2.6), g METAS-2017 (Empa) scale (Sect. 2.6), h quasi-linear sum of CFC-114 and CFC-114a.

TERRA and AURA satellites and the ESA ENVISAT satel-lite. Aircraft- and balloon-borne instruments provide vitalin situ measurements in the middle troposphere and lowerstratosphere. The combination of all of these complemen-tary data with state-of-the-art global chemistry and circula-tion models is providing major advances in our understand-ing of the global sources, chemistry, transport, and sinks ofatmospheric trace substances and allows for the determina-tion of atmospheric composition and air quality, the radia-tive forcing of climate change, and impacts on stratosphericozone.

2 Instruments, calibration, and infrastructure

The AGAGE program has placed a strong emphasis on in-strumental innovation and the gravimetric preparation of pri-mary standards to obtain high-frequency and high-precision

automated trace gas measurements at all the AGAGE mea-surement sites. In this section, the first four subsectionsdiscuss the AGAGE GC-MD (Sect. 2.1), Medusa GC-MS(Sect. 2.2), optical spectroscopy (Sect. 2.3), and isotopic(Sect. 2.4) instruments. Then we address data acquisition andprocessing (Sect. 2.5), instrumental calibration (Sect. 2.6),primary and affiliate station facilities and infrastructure(Sect. 2.7), secondary stations (Sect. 2.8), and stored airarchives (Sect. 2.9).

In the early 1990s the GC-MD instruments were devel-oped and deployed at the Mace Head, Trinidad Head, RaggedPoint, Cape Matatula, and Cape Grim stations and at theScripps Institution of Oceanography (SIO) calibration lab-oratory (Prinn et al., 2000). In the late 1990s, AGAGE pio-neered the deployment of automated GC-MS instruments atour stations in Mace Head and Cape Grim and at the Uni-versity of Bristol. These instruments featured an adsorption–desorption system (ADS) with cryogenic (−50 ◦C) pre-



Figure 2. A total of 7 months of data for gases measured at Mace Head, Ireland: (1) with the GC-MD in (a) 2004 and (b) 2016 (units: molefractions; ppb for N2O, CH4, H2, and CO; ppt for all others) and (2) with the Medusa GC-MS for selected gases in (c) 2004 and (d) 2016(units: mole fractions in ppt for all gases). In all four panels, measurements in polluted air originating from Europe (also in air affected bylocal sinks; see text) are shown in red, while those in clean air off the Atlantic Ocean are shown in black. Note that pollution events aredefined separately for each gas due to their often differing sources.

concentration of analytes from 2 L air samples (Simmonds etal., 1995). The technological developments incorporated intothese instruments, the methods of data collection, transmis-sion, and processing, the primary and secondary calibrationstandards produced at the SIO calibration laboratory, and theon-site tertiary (from SIO) and quaternary (calibrated on-site

from the tertiary) standards necessary to sustain the AGAGEnetwork are partly described in the first AGAGE overview(Prinn et al., 2000), but updated here in Sect. 2.6 and 2.7.

Beginning in the early 2000s, the AGAGE team recog-nized that modern refrigeration technology made it possi-ble to make major improvements to the ADS concept and



Table 2. GC–multidetector instruments at current AGAGE primary and secondary stations. Detectors: ECD for CFC-11, CFC-12, CFC-113,CH3CCl3, CCl4, N2O, and CHCl3; FID for CH4; MRD for CO and H2; and PDD for H2.

GC-ECD-FID GC-ECD-FID-MRD GC-ECD-FID-MRD-PDD

Trinidad Head, CA, USA Mace Head, Ireland Cape Grim, TasmaniaRagged Point, Barbados Tacolnestona, UKCape Matatula, Samoa Aspendaleb, AustraliaLa Jolla, CA, USARidge Hilla, UKBilsdalea, UKHeathfielda, UK

a Modified version of the GC-MD without FID channel. b Uses three individual GC systems with ECD,FID, and MRD detectors.

to greatly extend the range of compounds that could be mea-sured by enhanced cryogenic pre-concentration at −165 ◦C.As a result, the AGAGE GC-MS effort was redirected to thedevelopment of the new Medusa instrument (Miller et al.,2008; Arnold et al., 2012).

2.1 GC–multidetector instruments

The current AGAGE GC-MD instruments replaced the ear-lier GAGE GC-MD instruments in 1993–1996 (Table 2).These Agilent© GC instruments employ two electron-capture detector (ECD) channels and one flame-ionizationdetection (FID) channel to measure the principal chlorine-bearing anthropogenic ozone-depleting compounds nowbanned by the Montreal Protocol (CFC-11, CFC-12, CFC-113, CCl4, and CH3CCl3), as well as the both natural andanthropogenic compounds N2O, CH4, and CHCl3 (see Ta-ble 1). The GC-MDs at Mace Head and Cape Grim in-clude an extra channel for the measurement of CO and H2by a mercuric oxide reduction detector (MRD; Prinn et al.,2000). In early 2015, the GC-MD system at Cape Grim alsoadded a further extra channel for the measurement of H2 bypulsed discharge detector (PDD), bringing a more than 10-fold improvement in precision. The GC-MD measurementsare made on dried whole-air samples automatically injectedby a computer-controlled sampling module. Each analysiscycle takes 20 min.

Compared to its ALE and GAGE predecessors, theAGAGE GC-MD provides greatly enhanced precision andmeasurement frequency, custom software (GCWerks©, http://www.gcwerks.com, last access: 21 May 2018) for instru-ment control and digital acquisition of all chromatogramsand measurement parameters, and use of the internet fordata transmission and remote diagnosis and control (Prinnet al., 2000, Sect. 2.5). These instruments can also carryout pressure-programmed injections to assess their own non-linearities and use flexible custom algorithms for the post-analysis quantitative interpretation of chromatograms. Theperformance and reliability of these instruments have beenand continue to be exceptional, leading to important ad-

vances in scientific interpretation, as discussed below. Forsome of the species that the GC-MDs measure, AGAGE isnow also beginning to deploy new technologies includingGC-MS, cavity ring-down spectroscopy (CRDS), and quan-tum cascade laser (QCL; optical) methods that offer im-proved sensitivity as discussed in the following sections. TheGC-MD instruments will continue to be operated until suchtime as they can be phased out after careful overlap in thefield using these newer technologies.

2.2 Medusa GC-MS instruments

The AGAGE Medusa GC-MS instruments have become themajor instruments of the AGAGE network and collaboratingmeasurement laboratories. Instrument development work be-yond that described by Miller et al. (2008) continues, withenhanced operational parameters, upgrades, and new speciesbeing added over time. For example, subsequent importantchanges were made in the Medusa flow scheme and col-umn configuration that add the powerful greenhouse gas NF3emitted by the electronics industry to its measurement capa-bility without sacrificing any of its other capabilities (Arnoldet al., 2012). The reader is directed to these two papers for afull description of the current Medusa configuration – only abrief overview is given here.

A complement of 19 AGAGE Medusas has now been de-ployed (Table 3), with one at each of the 10 primary sta-tions (red labels in Fig. 1), two at the SIO calibration andinstrument development laboratory, and seven more at othersecondary stations or laboratories in the UK (Tacolneston& Bristol), Switzerland (Dübendorf), Australia (two at As-pendale), Norway (Kjeller), and China (Beijing).

At the heart of the Medusa is a Polycold© “Cryotiger”cold end that maintains a temperature of about −175 ◦Cwithin the Medusa’s vacuum chamber, even with a substan-tial heat load, using a simple single-stage compressor witha proprietary mixed-gas refrigerant. This cold end conduc-tively cools dual micro-traps to about −165 ◦C. By usingstandoffs of limited thermal conductivity to connect the trapsto the cold head, each trap can independently be heated re-


http://www.gcwerks.comhttp://www.gcwerks.com


Table 3. GC-MS instruments at AGAGE primary, affiliate, and secondary monitoring stations and at laboratories.

Primary or affiliate station (by latitude) Instrument Secondary station or laboratory (by country) Instrument

Ny-Ålesund Medusa La Jolla, USA (laboratorya and secondary) Two MedusasMace Head Medusa Tacolneston, UK MedusaJungfraujoch Medusa Bristol, UK (laboratory) MedusaMonte Cimone Affiliate Dübendorf, Switzerland (laboratory) MedusaCape Ochiishi Affiliate Aspendale, Australia (laboratory and secondary) Two MedusasShangdianzi Medusa Kjeller, Norway (laboratory) MedusaTrinidad Head Medusa Beijing, China (laboratory) MedusaGosan MedusaHateruma AffiliateRagged Point MedusaMount Mugogo Medusab

Cape Matatula MedusaCape Grim Medusa

a Central AGAGE Calibration Laboratory. b Installed in spring 2018.

sistively to any temperature from −165 to +100 ◦C or more,while the cold end remains cold. The use of two traps withextraordinarily wide programmable temperature ranges, cou-pled with the development of appropriate trap adsorbentsand the use of separating columns between traps, permitsthe desired analytes from 2 L air samples to be effectivelyseparated from more abundant gases that would otherwiseinterfere with chromatographic separation or mass spectro-metric detection, such as nitrogen (N2), oxygen (O2), argon(Ar), water vapor (H2O), CO2, CH4, krypton (Kr), and xenon(Xe). Importantly, the dual micro-trap and revised columnconfiguration also permit the analytes to be purified of in-terfering compounds from the larger first-stage trap (T1) byfractional distillation, chromatographic separation, and refo-cusing onto a smaller trap (T2) at very low temperatures sothat the resulting injections to the main chromatographic col-umn in the Agilent© 5975C quadrapole GC-MS are sharpand reproducible. By trapping and eluting analytes at verylow temperatures, the range of compounds that can be mea-sured is greatly extended to include a number of importantvolatile compounds, and problems with the reaction of an-alytes on the traps at higher temperatures are avoided. TheMedusa system uses high-precision integrating mass-flowcontrollers for the measurement of sample volumes. In ad-dition, significant advances have been made in the software(GCWerks) to control and acquire data from the Medusa andthe GC-MS itself so that the entire system has programma-bility, versatility, and ease of operation comparable to thatof the AGAGE GC-MD instruments. The original Agilent5973 mass-selective detectors (MSDs) used in the six earlyMedusas have been replaced with newer and more sensi-tive Agilent 5975C MSDs. As a result, sensitivities on theMedusas with the new MSDs increased 1.5- to 2-fold overthose with the old MSDs, which has especially benefittedmeasurements of the lowest-abundance species.

As noted above, instrument development work on theMedusas continues. The species routinely measured atMedusa field stations are listed in Table 1. Compounds addedonly recently to routine Medusa measurements (and there-fore not yet in Table 1) are HCFC-133a and CF3CFOCF2,while the light hydrocarbons C2H2 and C2H4, although stillmeasured, are also not included in Table 1 because co-elutioncompromises their measurement as the GC column ages. TheAGAGE Medusas were the first instruments monitoring insitu the global distributions and trends of the high-GWP in-dustrial gases CF4, NF3, and SO2F2 (Mühle et al., 2009,2010; Weiss et al., 2008; Arnold et al., 2013). In additionto the compounds listed in Table 1, additional species (e.g.,CFC-112) are in various stages of being added to the stationmeasurements. Recently, the “fourth-generation” halocar-bons HFC-1234yf, HFC-1234ze(E), and HCFC-1233zd(E),as well as HCFC-31 and four inhalation anesthetics havebeen measured in the atmosphere using the Medusa sys-tem (Vollmer et al., 2015a, b; Schoenenberger et al., 2015).The development work on the Medusa utilizes the two in-struments in this central laboratory. These instruments al-low a wide range of development work to be undertakenwhile maintaining the important functions of primary andsecondary calibration of the global AGAGE network and alsocontinuing “urban” AGAGE ambient measurements of airpumped from the SIO pier at La Jolla. At CSIRO Aspendale,one Medusa instrument is deployed in an urban air monitor-ing mode and the other is generally deployed for flask sam-ple measurements, in particular analyses of the Cape Grimair archive. The Medusas at the other five secondary stationslisted in Table 3 are deployed either for monitoring or labo-ratory functions.

The Medusa technology continues to evolve in responseto the needs of AGAGE researchers to measure new com-pounds, improvements in software, including data process-ing, diagnostics and alarms, and improvements in available



technology. Most notably, the Polycold Cryotiger cold-endtechnology that was so revolutionary at the outset of theMedusa program is nearing the end of its useful life, butvery fortunately Stirling cooling technology has advancedconsiderably with improved performance and reliability andreduced cost during the same time period. One Medusa atthe SIO laboratory has been retrofitted to Stirling cooling(Sunpower CryoTel-GT) and is performing extremely well,as well as offering increased flexibility in trapping parame-ters. At the Empa and SIO laboratories, efforts are also un-derway to upgrade current Medusa technology to time-of-flight mass spectrometry (TOF-MS) in place of quadrupolemass spectrometric detection. This offers the advantage ofvery high mass resolution (∼ 4000) that is capable of sepa-rating gases with the same integer masses but different actualmasses that interfere with each other in the chromatogramsusing quadrupole technology (e.g., Obersteiner et al., 2016).

There are also three AGAGE-affiliated stations that usesimilar but not identical automated GC-MS measurementswith cryogenic pre-concentration (stations denoted “affil-iate” in Table 3), but are tied to AGAGE standards, atHateruma Island and Cape Ochiishi, Japan (NIES) and atMonte Cimone, Italy (University of Urbino). Monte Cimoneuses a GC-MS (Agilent 6850 and 5975, respectively) withan autosampling and pre-concentration device (Markes In-ternational©, UNITY2-Air Server2©) to enrich the halocar-bons on a focusing adsorbent trap (Maione et al., 2013) andAGAGE-derived calibrations. Hateruma and Ochiishi bothuse a GC-MS (Agilent 6890 and 5973, respectively) with aunique cryogenic pre-concentration module (Yokouchi et al.,2006, 2012) and independently produced gravimetric stan-dards that are intercompared with AGAGE standards to pro-vide intercalibration factors.

2.3 Optical spectroscopic instruments

Recent advances in wavelength-scanned cavity ring-downspectroscopy (CRDS) have enabled precise analyses of CH4,CO2, CO, N2O, and H2O without chromatographic separa-tion to begin in AGAGE. The analyzed air sample needs tobe dried or, if not dried, corrections applied using the an-cillary H2O measurement. The Nafion sample drying andgas sampling approach used in AGAGE has been adaptedto a sampling module with an MKS Instruments© inlet pres-sure controller for CRDS instruments that has been designedby SIO and built by Earth Networks© (Welp et al., 2013).These optical instruments are now expanding the measure-ment capabilities within AGAGE. There are several advan-tages in switching from the GC-FID approach for measur-ing CH4, the GC-ECD approach for N2O, and the GC-MRDapproach for CO in AGAGE to these optical spectroscopymethods: no chromatography (so no carrier gases needed),essentially continuous, reduced costs including ongoing in-strument maintenance, and improved linearity of response

(for N2O, CO). This transition is happening already at sev-eral AGAGE stations (see Table 4).

The CSIRO Picarro© G2301 for CO2, CH4, and H2O atCape Grim (which is being operated at present without dry-ing the sample gas) has been compared with the AGAGEGC-MD CH4 data at Cape Grim and the agreement is verygood, with a mean offset of only∼ 0.26 ppb (∼ 0.02 %) whenreported on the same calibration scale. The AGAGE groupat SIO, in collaboration with the laboratory of R. F. Keel-ing, the company Earth Networks©, and the California AirResources Board (CARB), has been evaluating the perfor-mance of various CRDS instruments, including calibrationoptimization, using Allan variance analyses (Allan, 1966;Werle et al., 1993). This has included the Picarro G2301,the Picarro G2401 for CO2, CO, CH4, and H2O, the PicarroG5205 (prototype) and G5310 mid-IR for N2O and H2O, andthe Los Gatos Research (LGR©) high-precision mid-IR in-strument for N2O, CO, and H2O. For CO, the LGR mid-IRinstrument is an order of magnitude more precise than the Pi-carro G2401, but to take full advantage of the LGR’s preci-sion requires frequent calibration (hourly or less) that is im-practical for long-term atmospheric monitoring. With onlydaily calibration this difference is reduced to about a fac-tor of 2. The precisions of the G5310 (and G5205) and toa lesser extent of the G2401 are improved by drying the airsample to minimize the H2O correction using the aforemen-tioned sampling modules built by Earth Networks, and thesemodules have been adopted at the Ragged Point, Mt. Mu-gogo, and Cape Matatula stations. Finally, CSIRO is oper-ating high-precision Aerodyne Research© quantum cascadelaser (QCL) spectroscopy systems for CO and N2O at As-pendale, Australia.

2.4 Isotopomer–isotopologue instruments

For GHGs that have natural, anthropogenic, industrial, andbiogenic sources, such as CO2, CH4, and N2O, measure-ments of atmospheric abundances alone are often inadequateto precisely differentiate among these different sources.High-frequency in situ measurements of not just the totalmole fractions of these gases, but also their stable isotopiccompositions (12C, 13C, 14N, 15N, 16O, 18O, H, D) are a newfrontier in global monitoring and hold the promise of revo-lutionizing our understanding of the global cycles of thesegases (e.g., Rigby et al., 2012). High-frequency in situ iso-topic measurements are now feasible using optical (laser) de-tection.

MIT and Aerodyne Research have codeveloped and de-ployed (2015–2017) at the Mace Head station an automatedhigh-frequency instrument for the analysis of the isotopiccomposition of N2O using tunable infrared laser differen-tial absorption spectroscopy (TILDAS) with mid-infraredquantum cascade lasers (Harris et al., 2013). This instru-ment is fully automated and can be accessed and con-trolled via the internet. The new instrument monitors the



Table4.C

RD

Sspectroscopic

instruments

atAG

AG

Eprim

arystations

andsecondary

stations(including

theU

KD

erivingE

missions

relatedto

Clim

ateC

hange(D

EC

C)

network

andU

KN

ationalPhysicalLaboratory

(NPL

)stations).Instruments

with

Earth

Netw

orks(E

N)driers

lowerthe

sample

watervaporm

olefractions

todecrease

H2 O

interferences.

Instrument

Gases

Stations

PicarroG

1301C

H4 ,C

O2 ,H

2 OJungfraujoch

(G2401

after2011)M

aceH

eadPicarro

G2301

CH

4 ,CO

2 ,H2 O

La

Jolla(+

EN

drier),TrinidadH

ead(+

EN

drier)C

apeG

rimM

aceH

eadB

ristol,Tacolneston

(+E

Ndrier),

Ridge

Hill(U

KD

EC

C)

Aspendale

PicarroG

2401C

H4 ,C

O2 ,C

O,H

2 OR

aggedPoint(+

EN

drier)C

apeM

atatula(+

EN

drier)M

t.Mugogo

(+E

Ndrier)

Heathfield

(UK

NPL

),B

ilsdale(U

KD

EC

C)

Ny-Å

lesund

PicarroG

5205orG

5310N

2 O,H

2 OM

t.Mugogo

(+E

Ndrier)

Ny-Å

lesund(G

5310)L

GR

highperform

anceN

2 O,C

O,H

2 OL

aJolla

(+E

Ndrier)

TacolnestonH

igh-precisionA

erodyneQ

CL

CO

,N2 O

Aspendale,A

ustralia

four major isotopologues and isotopomers of nitrous oxide(15N14N16O, 14N15N16O, 14N14N18O, and 14N14N16O) witha precision of at least 0.3 per mil (‰) for individual mea-surements spanning 28 min. For at least 0.1 per mil (‰)precision, we need to average 3–11 such measurements de-pending on the isotope (Harris et al., 2013). The neededpre-concentration was achieved through the development ofa new high-efficiency cryo-focusing trap and sample trans-fer module (called Stheno) using concepts from the AGAGEMedusa module (Potter et al., 2013).

Similar automated N2O isotope instrumentation has beendeveloped at Empa (Wächter et al., 2008; Heil et al., 2014)and has been used for analyzing flask samples from Jungfrau-joch. Also, a similar pre-concentration system has been de-veloped by Mohn et al. (2010) and their pre-concentrationTILDAS system has shown excellent compatibility with iso-tope ratio MS in an interlaboratory comparison campaign(Mohn et al., 2014). The pre-concentration technique hasbeen further developed at Empa by implementing a morepowerful Stirling cooler and a moveable trap design for quan-titative CH4 adsorption (Eyer et al., 2016). Also, CSIRO op-erates an Aerodyne Research quantum cascade laser systemfor the three stable isotopologues of CO2 (12CO2, 13CO2,and 18O12C16O) at Cape Grim.

Further developments in these instruments will facilitatetheir future deployment at AGAGE stations for continuoushigh-frequency in situ isotopic composition measurements ofCO2, CH4, and N2O.

2.5 Data acquisition and processing

The custom data acquisition and processing software (GCW-erks) used in AGAGE for both the GC-MD and Medusa GC-MS instruments and run under the Linux operating system isdescribed in moderate detail by Miller et al. (2008) and Prinnet al. (2000). There are many benefits to using this customsoftware approach, including complete source-code controlover all instrument operation software, integration and dataprocessing algorithms, and the ability to improve the soft-ware interactively. All AGAGE stations (except Haterumaand Ochiishi) and laboratories are linked via the internet sothat functions such as instrument control and software up-dating can be done remotely. The strength of this approachis illustrated by the fact that, in addition to being used for allMedusa instruments in the AGAGE network, portions of theGCWerks software have been adopted by other leading lab-oratories engaged in non-AGAGE atmospheric and oceanictrace gas measurements, including NOAA/ESRL, CSIRO,the University of Bristol, and Empa.

Chromatograms are acquired and displayed in real timeand are stored in a highly compressed format. Electronic stripcharts record critical instrument parameters and a multitudeof log files are generated as well, which contain parameterscritical for data quality control. The GCWerks software al-lows operators and data processors to quickly review and



batch-integrate chromatograms and produce time series anddiagnostic plots of integration results to assess instrumentalperformance. The AGAGE data processing system relies onhaving identical software and databases at the field stationsand at the data processing sites. This allows the station opera-tors and investigators to review identical chromatograms andinstrumental data in a timely manner and fosters constructiveexchanges among the AGAGE investigators. The SIO servermaintains a complete database for all stations and producesfinal results for all sites once the periodic data reviews havebeen completed. Data are routinely reviewed at regular inter-vals, and a final review is done approximately every 6 monthsprior to and at each AGAGE team meeting, with all the dataprocessing sites involved concurrently.

New software (GCCompare, http://www.gcwerks.com,last access: 21 May 2018) continues to be developed for dataprocessing, quality control, and visualization. This softwarehas greatly streamlined the review and editing of AGAGEdata that takes place over the internet and at AGAGE meet-ings twice a year. This software is highly interactive and hasfeatures such as being able to click on individual measure-ments and display back trajectories from the UK Met Of-fice’s NAME model (Jones et al., 2007) to help diagnoseobserved departures from background values. Recent stationsoftware developments continue, including enhancements ofautomated alarms to improve the oversight of day-to-dayfield operations and, importantly, to protect the instrumen-tation from damage when key components fail. Software forthe correction of occasional drifts in more reactive gases inthe on-site tertiary and quaternary calibration standards con-tinues to be improved and implemented. Working in collab-oration with NOAA/GMD, the software has also been mod-ified to remove the need to divide the acquisition of peakdata into time “windows”. This had caused problems in opti-mizing dwell times on certain masses and in following smalldrifts in retention times of peaks located near transitions be-tween windows. This change also allows for a reduction, tosome degree, in the numbers of ions acquired at a given time,thereby improving precisions and detection limits, especiallyfor the less abundant emerging compounds. GCWerks alsokeeps all of the raw data, including the chromatograms, thusenabling the routine reprocessing of the entire record for eachspecies at each station whenever needed (e.g., when calibra-tion scales are updated (see Sect. 2.6) or when peak integra-tion methods are improved).

Finally, this GCWerks software is becoming an increas-ingly important “spin-off” from the AGAGE project. In par-ticular, considerable progress has been made in adaptingAGAGE data acquisition, visualization, and quality-controlsoftware for discrete sample GC and GC-MS instruments toapplications involving continuous optical instruments suchas the cavity ring-down spectrometer (CRDS) instruments ofPicarro and Los Gatos Research (LGR) and the quantum cas-cade laser (QCL) instruments of Aerodyne Research.

2.6 Calibration

One of the strengths of AGAGE is its dependence uponwell-defined internal absolute gravimetric calibration proce-dures that can be repeated periodically to ensure the accu-racy of the long-term measured trends. During the periodof AGAGE there have been seven absolute primary cali-bration efforts, SIO-93, SIO-98, SIO-05, SIO-07, SIO-12,SIO-14, and SIO-16, named after the SIO laboratory andthe year in which the scale was completed. The “bootstrap”methods used to prepare primary gravimetric standards atppt levels and the way in which these standards are in-tegrated to define a calibration scale are described in theAGAGE “history paper” (Prinn et al., 2000). The methodsused to propagate these scales to the species measured bythe Medusa GC-MS are discussed by Miller et al. (2008).At present, ambient-level SIO primary calibration scaleshave been prepared for 42 AGAGE species: N2O, PFC-14 (CF4), PFC-116 (C2F6), PFC-218 (C3F8), PFC-318 (c-C4F8), PFC-3-1-10 (C4F10), PFC-4-1-12 (C5F12), PFC-5-1-14 (C6F14), PFC-6-1-16 (C7F16), PFC-7-1-18 (C8F18), SF6,SF5CF3, SO2F2, NF3, HFC-23, HFC-32, HFC-125, HFC-134a, HFC-143a, HFC-152a, HFC-227ea, HFC-236fa, HFC-245fa, HFC-356mfc, HFC-43-10mee, HCFC-22, HCFC-141b, HCFC-142b, CFC-11, CFC-12, CFC-113, CFC-114,CFC-115, Halon-1211, Halon-1301, Halon-2402, CH3Br,CH3Cl, CH2Cl2, CHCl3, CH3CCl3, and CCl4. Among them,NF3, C4F10, C5F12, C6F14, C7F16, and C8F18 were calibratedby the method of internal additions, which is by spiking realair with gravimetrically determined amounts of the analyte(Arnold et al., 2012), while the remaining gases were cali-brated by the conventional AGAGE method of adding gravi-metrically determined amounts of the analytes to analyte-free artificial “zero air”. For CF4, the primary calibrationshave been made both ways with excellent agreement. For thevolatile gases like CF4 and NF3, the use of the internal addi-tions method is particularly valuable to avoid biases in theirseparation or detection due to interferences from the pres-ence of krypton and other inert gases in real air but not inartificial zero air. The precisions of these calibration scales,based on the internal consistency among the individual pri-mary standards, range from about 2 % for the least abundantcompounds to < 0.1 % for the more abundant compounds.The absolute accuracies of these scales, based on estimatesof maximum systematic uncertainties, including the puritiesof the reagents used in their preparation and possible system-atic analytical interferences, are between 0.3 and 2 % greaterthan the statistical uncertainties depending on the compoundand its atmospheric abundance.

The evolution of GC-MS techniques in AGAGE hasgreatly increased the number of species that are measured inthe program and has thus exceeded, at least temporarily, ourcapacity to prepare and maintain gravimetric primary cali-bration scales. To bridge this gap and, very importantly, todecouple the long-term measurement program for the evolv-


http://www.gcwerks.com


ing and independent primary calibration process, AGAGEhas adopted a relative calibration scale for all Medusa andGC-MD measurements. This scale, designated R1, is definedby regular intercomparisons of trace gas concentrations ina suite of whole-air secondary (“gold”) tanks maintained atthe SIO laboratory. These tanks are compared against eachother to assess possible drift and against primary standardsfor those species for which we have primary standard cal-ibrations. Every year, this suite of secondary tanks is ex-tended with at least one new tank filled under clean air con-ditions in winter or spring and the intercomparison is re-peated. Other tanks filled at the same time are calibratedagainst this suite of tanks and sent to each station as cal-ibration “tertiary” standards, where they are either directlymeasured (GC-MD) or used to calibrate working “quater-nary” standards (Medusa) at each measurement site. As pri-mary calibration scales evolve at SIO, NOAA/ESRL, Bris-tol, Empa, NCAR, NIES, or any other laboratory, the rela-tionships of their scales to the R1 scale can be measured toobtain a set of factors by which our R1 values can be multi-plied to report Medusa data on any of these calibration scales.The R1 scale is flexible to designate tanks other than R1as a reference tank for individual compounds, which werenot present at sufficient concentrations or were not measuredin the original R1 tank. Looking to the future, this enablesus to keep pace with the changing atmospheric concentra-tions of many species and to incorporate corrections for pos-sible nonlinearities in the calibration process and for possi-ble drifts in standard mixtures. This technique has been usedto provide calibrations for species not on an SIO scale suchas CFC-13 (METAS-2017), CHBr3 (NOAA-2009P), PCE(NOAA-2003B), and HCFC-133a (Empa-2013; Vollmer etal., 2015c).

AGAGE gravimetric calibration activities are independentfrom those in other laboratories (except for the CO2 cali-brations used in the bootstrap method that come from theKeeling laboratory at SIO), but there are also strong syner-gies, especially with NOAA/ESRL. For example, the SIO-14 calibrations showed excellent agreement with NOAA forHalon-2402 (Vollmer et al., 2016), while AGAGE atmo-spheric CH2Cl2 mole fractions based on the SIO-14 scale aresignificantly higher than those reported by NOAA (Carpen-ter et al., 2014). This subject of intercalibration is discussedfurther in Sect. 3.2.

Whole-air and synthetic mixture calibration standardsused in AGAGE are stored in 34 L high-pressure (60 bar)electropolished stainless steel canisters designed at SIO andmanufactured by Essex Industries© that are legal for interna-tional shipment. Although the adoption of a single primarycalibration scale from a central calibration facility for eachmeasured species has been advocated by some researchers,AGAGE does not favor this approach. The existence of morethan one independent high-precision traceable calibrationscale for each measured species, with frequent intercom-parisons among independently calibrated field measurements

(see Table 5, Sect. 3.2) and with direct intercomparison of thecalibration standards themselves (Hall et al., 2014), reducesvulnerability to systematic errors and long-term calibrationdrifts for all participating primary calibration and measure-ment programs.

2.7 Primary and affiliate station facilities andinfrastructure

While the individual station size and infrastructure varies de-pending on their location and the presence of other com-plementary gas and aerosol measurement programs, all sta-tions consist of permanent buildings (wood, concrete, steel,fiberglass) with air samples drawn using non-contaminatingpumps through lines with inlets located on adjacent high tow-ers. The details about the general air sampling setup for eachinstrument are provided in Miller et al. (2008) and Prinn etal. (2000). The sampling lines are either stainless steel orlayered polyethylene–aluminum–Mylar (Dekabon© or Syn-flex©). For more information on individual stations, we referthe reader to the AGAGE website (http://agage.mit.edu (lastaccess: 21 May 2018). All stations (except Hateruma andCape Ochiishi) periodically exchange stainless steel on-siteEssex calibration tanks (tertiary standards) calibrated at SIOlinking the measurements to the AGAGE SIO primary andsecondary standards. Some stations also use modified RIX©oil-free air compressors and the tertiary standards to preparequaternary standards either on-site, in their home laborato-ries, or supplied by SIO to extend the lifetime of the tertiarystandards. At Cape Grim and Ny-Ålesund, the quaternarystandards are prepared by a cryogenic collection of wholeair with subsequent ejection of condensed water.

2.8 Secondary stations

In addition to the primary and affiliate stations in AGAGE,there are complementary secondary stations, usually at eithermore polluted urban locations or at more remote sites thatshare some or all of the AGAGE technology and calibrations.

SIO carries out continuous measurements of all AGAGEgases in La Jolla in conjunction with its extensive calibration(Sect. 2.6) and instrument development operations.

The University of Bristol runs the UK DECC (DerivingEmissions related to Climate Change) network of tall tow-ers at Ridge Hill, Angus (now decommissioned), Tacolneston(in collaboration with the University of East Anglia), Heath-field (UK National Physical Laboratory), and Bilsdale in theUK measuring CO2, CO, CH4, N2O, and SF6 and linked tothe AGAGE Mace Head station and to AGAGE calibrationsand some technologies. Tacolneston also includes measure-ments of H2 and CO via MRD and a Medusa GCMS.

CSIRO is operating two Medusa GCMSs at Aspendale,and Picarro CRDS CH4 and CO2 (and CO at one sta-tion) instruments at Burncluith (26◦ S, G2401), Ironbark(27◦ S, G2301), Aspendale (38◦ S, G2301), Macquarie Is-




Tabl

e5.

Scal

eco

nver

sion

fact

ors

betw

een

NO

AA

and

AG

AG

E(S

IO)

expr

esse

das

aN

OA

A/

AG

AG

Era

tioba

sed

ona

com

pari

son

ofN

OA

A/E

SRL

/GM

Dfla

skda

tato

AG

AG

Ein

situ

data

atco

mm

onsi

tes.

For

CH

4,N

2O,a

ndSF

6,N

OA

Afla

skda

tafr

omth

eca

rbon

cycl

ean

dgr

eenh

ouse

gase

s(C

CG

G)

grou

pha

vebe

enus

ed;f

oral

loth

ersp

ecie

sN

OA

Afla

skda

tafr

omth

eha

loca

rbon

san

dot

hera

tmos

pher

ictr

ace

spec

ies

(HA

TS)

grou

par

eus

ed.T

here

spec

tive

scal

esus

edin

each

netw

ork

are

indi

cate

din

the

tabl

eal

ong

with

the

inst

rum

enta

lm

etho

dus

edfo

rthe

anal

ysis

.The

site

sus

edin

the

com

pari

sons

are

liste

din

colu

mn

five,

follo

wed

byth

ele

ngth

ofth

eco

mpa

riso

npe

riod

.Las

tly,c

omm

ents

onth

eco

nsis

tenc

yof

the

com

pari

sons

fore

ach

spec

ies

are

give

n.

Spec

ies

Rat

io(N

OA

A/

AG

AG

E)

NO

AA

scal

em

etho

dA

GA

GE

(SIO

)sc

ale

met

hod

Site

sTi

me

peri

odC

omm

ent

CH

41.

0001±

0.00

07N

OA

A-2

004A

GC

-FID

Toho

kuU

nive

rsity

GC

-FID

(GC

-MD

)Fi

vesi

tes

(CG

O,S

MO

,RPB

,TH

D,M

HD

)19

93–2

017

0.1

%co

nsis

tenc

yov

ertim

e

N2O

0.99

83±

0.00

05N

OA

A-2

006A

GC

-EC

DSI

O-1

6G

C-E

CD

(GC

-MD

)Fi

vesi

tes

(CG

O,S

MO

,RPB

,TH

D,M

HD

)19

97–2

017

0.1–

0.2

%co

nsis

tenc

yov

ertim

e,sl

ight

incr

eas-

ing

tren

dof

0.08

%pe

rdec

ade

SF6

1.00

49±

0.00

29N

OA

A-2

014

GC

-EC

DSI

O-0

5G

C-M

SM

edus

aSi

xsi

tes

(CG

O,S

MO

,RPB

,TH

D,M

HD

,ZE

P)20

04–2

017

Smal

lste

pin

2010

,0.5

%co

nsis

tenc

yov

ertim

e

CFC

-11

0.99

93±

0.00

09N

OA

A-2

016

GC

-EC

DSI

O-0

5G

C-E

CD

(GC

-MD

)Fo

ursi

tes

(CG

O,S

MO

,TH

D,M

HD

)19

93–2

017

∼1

%co

nsis

tenc

yov

ertim

e

CFC

-12

0.99

62±

0.00

10N

OA

A-2

008

GC

-EC

DSI

O-0

5G

C-E

CD

(GC

-MD

)Fo

ursi

tes

(CG

O,S

MO

,TH

D,M

HD

)19

93–2

017

0.5

%co

nsis

tenc

yov

ertim

e

CFC

-113

1.00

03±

0.00

23N

OA

A-2

003M

SG

C-M

SSI

O-0

5G

C-E

CD

–GC

-MS

Med

Four

site

s(C

GO

,SM

O,T

HD

,MH

D)

1993

–201

7∼

1%

cons

iste

ncy

over

time

CC

l 41.

015–

1.03

8(n

otco

n-st

ant,

see

com

men

ts)

NO

AA

-200

8G

C-E

CD

SIO

-05

GC

-EC

D(G

C-M

D)

Four

site

s(C

GO

,SM

O,T

HD

,MH

D)

1995

–201

7Tr

end:

3.5–

4.0

%di

ffer

ence

in19

95–2

000,

toap

prox

imat

ely

1.5

%di

ffer

ence

in20

13–2

017

CH

3CC

l 31.

0055±

0.01

09N

OA

A-2

003

GC

-MS

SIO

-05

GC

-EC

D–G

C-M

SM

edFo

ursi

tes

(CG

O,S

MO

,TH

D,M

HD

)19

93–2

017

Initi

altr

end

duri

ng19

93–2

000,

from

3%

dow

nto

0.5

%di

ffer

ence

,the

ngo

odag

reem

entw

ithin

1%

HC

FC-2

20.

9971±

0.00

27N

OA

A-2

006

GC

-MS

SIO

-05

GC

-MS-

AD

SM

edFo

ursi

tes

(CG

O,S

MO

,TH

D,M

HD

)19

98–2

017

1–2

%co

nsis

tenc

yov

ertim

e

HC

FC-1

41b

0.99

41±

0.00

49N

OA

A-1

994

GC

-MS

SIO

-05

GC

-MS-

AD

SM

edFo

ursi

tes

(CG

O,S

MO

,TH

D,M

HD

)19

98–2

017

∼2

%co

nsis

tenc

yov

ertim

e

HC

FC-1

42b

0.97

43±

0.00

52N

OA

A-1

994

GC

-MS

SIO

-05

GC

-MS-

AD

SM

edFo

ursi

tes

(CG

O,S

MO

,TH

D,M

HD

)19

98–2

017

∼2

%co

nsis

tenc

yov

ertim

e

HFC

-134

a1.

0015±

0.00

48N

OA

A-1

995

GC

-MS

SIO

-05

GC

-MS-

AD

SM

edFo

ursi

tes

(CG

O,S

MO

,TH

D,M

HD

)19

98–2

017

∼2

%co

nsis

tenc

y,be

tterr

ecen

tly

HFC

-152

a0.

9976±

0.02

27N

OA

A-2

004

GC

-MS

SIO

-05

GC

-MS-

AD

SM

edFo

ursi

tes

(CG

O,S

MO

,TH

D,M

HD

)19

98–2

017

2–3

%co

nsis

tenc

yov

ertim

e

H-1

211

0.97

99±

0.00

50N

OA

A-2

006

GC

-MS

SIO

-05

GC

-MS-

AD

SM

edFo

ursi

tes

(CG

O,S

MO

,TH

D,M

HD

)19

98–2

017

∼2

%co

nsis

tenc

yov

ertim

e

H-1

301

0.97

66±

0.00

98N

OA

A-2

006

GC

-MS

SIO

-05

GC

-MS

Med

usa

Thr

eesi

tes

(CG

O,S

MO

,TH

D)

2004

–201

5∼

2%

cons

iste

ncy

over

time

H-2

402

1.02

08±

0.01

00N

OA

A-1

992

GC

-MS

SIO

-14

GC

-MS

Med

usa

Four

site

s(C

GO

,SM

O,T

HD

,MH

D)

2004

–201

7Sm

all

step

chan

ge20

08–2

009,

3–4

%co

nsis

-te

ncy

over

time

CH

3Cl

1.00

74±

0.00

73N

OA

A-2

003

GC

-MS

SIO

-05

GC

-MS-

AD

SM

edFo

ursi

tes

(CG

O,S

MO

,TH

D,M

HD

)19

98–2

017

2%

cons

iste

ncy

over

time

Tabl

eno

tes:

com

pari

sons

betw

een

NO

AA

HA

TS

data

and

AG

AG

Ein

situ

wer

epe

rfor

med

base

don

the

NO

AA

data

post

edon

the

ftp

site

:ftp

://ft

p.cm

dl.n

oaa.

gov/

hats

/(la

stac

cess

:21

May

2018

).G

C-M

S-A

DS

Med

indi

cate

sda

tafr

omth

eA

DS

inst

rum

ents

atC

ape

Gri

man

dM

ace

Hea

dus

edfr

om19

98–2

003,

with

Med

usa

data

used

from

2004

onw

ards

atth

esi

tes

indi

cate

d.G

C-E

CD

–GC

-MS

Med

indi

cate

sa

com

bine

dda

tare

cord

from

the

GC

-EC

D(G

C-M

D)i

nstr

umen

tsw

ithth

eG

C-M

SM

edus

ada

taus

edfo

rthe

latte

rpar

toft

here

cord

.Si

tes:

CG

O–

Cap

eG

rim

,Aus

tral

ia;S

MO

–C

ape

Mat

atul

a,Sa

moa

;RPB

–R

agge

dPo

int,

Bar

bado

s;T

HD

–Tr

inid

adH

ead,

USA

;MH

D–

Mac

eH

ead,

Irel

and;

ZE

P–

Zep

pelin

Mou

ntai

n,N

y-Å

lesu

nd,N

orw

ay.

Som

esp

ecie

sar

em

easu

red

bym

ultip

lein

stru

men

tsan

d/or

flask

sam

ples

;sel

ecte

dre

sults

show

nhe

re.


ftp://ftp.cmdl.noaa.gov/hats/


land (55◦ S, G2301), Casey Station, Antarctica (66◦ S, orig-inally a G1301 now replaced by a G2301), and onboard thenew CSIRO research vessel the RV Investigator (G2301).Picarro CRDS CH4 and CO2 instruments were also previ-ously operated at Gunn Point, northern tropical Australia(11◦ S, G1301, 2010–2017, currently suspended), Arcturus(22◦ S, G1301 replaced by G2301, 2010–2014), and Ot-way (38◦ S, ESP1000, 2009–2012). CSIRO is also operatinghigh-precision Aerodyne Research QCL systems for CO andN2O and another for the stable isotopes of CO2 at Aspendale.All of these instruments are configured to run with AGAGE–GCWerks software (see Sect. 3.3).

2.9 Air archives

CSIRO has been collecting and archiving pressurized 34 Lelectropolished canisters of cryo-trapped air collected dur-ing clean air conditions at Cape Grim since the mid-1970s,and plans to continue into the future (Fraser et al., 2017).This “Southern Hemisphere air archive” has proven to be aninvaluable resource to the international atmospheric chem-istry community, including AGAGE, because a wide rangeof species that could not be measured at the time of collec-tion can be measured retrospectively in the archive as long asthose species are conserved in these canisters. Until 2013 atarget of four Cape Grim air archive samples were collectedeach year, while from 2014 onwards six air archive tanks arecollected each year. Measurements from this Southern Hemi-sphere archive have made significant contributions to severalrecent AGAGE papers by addressing the following: HFC-23(Miller et al., 2010); PFCs (Mühle et al., 2010; Trudinger etal., 2016); SF6 (Rigby et al., 2010); CFC-13, CFC-114, andCFC-115 (Vollmer et al., 2018); Halon-1211, Halon-1301,and Halon-2402 (Vollmer et al., 2016); and HFC-365mfc,HFC-245fa, HFC-227ea, and HFC-236fa (Vollmer et al.,2011). There was a parallel “Northern Hemisphere archive”collected by Rei Rasmussen at Cape Meares, Oregon dur-ing the ALE and GAGE programs, but these samples are nolonger accessible to this program and are mostly used up.The SIO AGAGE group has been storing a Northern Hemi-sphere archive of air compressed at Trinidad Head and LaJolla since the mid-1990s and has collected a series of North-ern Hemisphere air samples from various sources (e.g., SIOlaboratories of Charles D. Keeling and Ray F. Weiss, NOAA-GMD, and NILU) and of varying integrity for trace gas mea-surements that extends this record back to the early 1970s.Measurements from this Northern Hemisphere archive havemade significant contributions to several recent AGAGE pa-pers, especially for more inert species such as the PFCs, NF3,and SF6 (e.g., Mühle et al., 2009, 2010; Rigby et al., 2010;Weiss et al., 2008; Arnold et al., 2013).

Additional air archive samples used in AGAGE studieswere derived from firn air collections in Greenland andAntarctica obtained by international consortia. The AGAGEanalyses of firn air used Medusa GC-MS instruments and

substantially extended mole fraction data back in time alongwith emission estimates derived from the data, specificallyfor Halons (Vollmer et al., 2016), PFCs (Trudinger et al.,2016), and minor CFCs (Vollmer et al., 2018).

3 Data analysis and modeling

In this section, the seven subsections address the following:meteorological interpretation of data (Sect. 3.1), data inter-comparisons (Sect. 3.2), flux estimation using data and mod-els (Sect. 3.3), and flux estimation using 3-D Eulerian models(Sect. 3.4), 3-D Lagrangian models (Sect. 3.5), merged 3-DEulerian and Lagrangian models (Sect. 3.6), and simplified(2-D) models (Sect. 3.7).

3.1 Meteorological interpretation

As part of processing the AGAGE data, we place an identi-fication flag on each measured value in an attempt to sepa-rate regional and/or local pollution events from backgroundmeasurements. The current, objective (statistically based) al-gorithm has been successfully implemented and uniformlyapplied to the entire ALE/GAGE/AGAGE time series includ-ing data from all AGAGE primary and affiliate stations (ex-cept Hateruma and Cape Ochiishi) and all instruments (GC-MS, GC-MD, Picarro). Moreover, the algorithm has been de-signed to be easily reapplied to the entire dataset in the eventof (minor) modifications to the algorithm. The concept ofthe algorithm is to examine the statistical distributions of 4-month bins of measurements (approximately 4320 GC-MDor 1440 Medusa GC-MS values) of any species at a speci-fied site and centered on one day at a time after removingthe trend over the period (O’Doherty et al., 2001; Cunnold etal., 2002). The algorithm can be applied to the results from3-D models to separate the background and polluted values(Ryall et al., 2001; Simmonds et al., 2005). We also use a 3-DLagrangian back-trajectory model driven by reanalyzed me-teorology, specifically the UK Met Office’s Numerical Atmo-spheric dispersion Modelling Environment (NAME; Ryall etal., 1998; Jones et al., 2007), to further evaluate the statis-tical pollution algorithm (O’Doherty et al., 2001; Cunnoldet al., 2002) and include this evaluation as part of the pollu-tion and background identification flag associated with eachmeasurement. NAME is Lagrangian (Sect. 3.5). In NAME,large numbers of particles at the station are effectively ad-vected backwards in time by 3-D reanalysis meteorologi-cal fields, with turbulent dispersion represented by a ran-dom walk technique. Particles first encountering the surfaceor surface boundary layer in known trace-gas-emitting re-gions are then flagged as polluted. An observation is alsoconsidered potentially polluted if the atmosphere at the sta-tion is stable with very low winds and known nearby tracegas sources. NAME back trajectories are automatically com-puted for every AGAGE measurement and used extensivelyin the semiannual AGAGE data reviews.



3.2 Data intercomparisons

AGAGE cooperates with other groups carrying out flasksampling and/or in situ real-time tropospheric measurementsin order to produce harmonized global datasets for use inmodeling. Toward this end, AGAGE routinely collaborateswith NOAA/ESRL/GMD to develop best estimates of thedifferences in absolute calibrations and field site calibra-tions between them and the AGAGE–SIO scales (see Elkinset al., 2015, and the NOAA/ESRL/GMD website for theNOAA/ESRL/GMD database). This is undertaken in severalways: comparisons involving exchanges of tanks (checkingabsolute calibration); comparisons of hemispheric and globalmean trends estimated by the two networks; examination ofdifferences between the AGAGE and GMD in situ instru-ments at our common in situ site, Cape Matatula (checkingthe propagation of standards to remote sites); and ongoingextensive comparisons between AGAGE in situ GC-MD andGC-MS data and GMD flask data at the six AGAGE siteswhere GMD flasks are filled (Zeppelin, Mace Head, TrinidadHead, Ragged Point, Cape Matatula, and Cape Grim), withthe results reported at the semiannual AGAGE meetings. Tohelp ensure progress on this and other cooperative endeavors,leaders and members of the relevant NOAA/GMD group reg-ularly attend the semiannual AGAGE meetings; other jointmeetings with GMD personnel are held from time to time.Examples of the scale conversion factors determined fromthe comparison of AGAGE in situ data to NOAA flask resultsare given in Table 5. There is generally good consistency withtime for these with some exceptions, most notably CCl4. TheCCl4 comparison shows a trend with time from around 3.5–4.0 % in 1995–2000 to approximately 1.5 % in 2013–2017.Because these factors are updated when additional intercom-parisons occur, we advise data users to consult the AGAGEwebsite (http://agage.mit.edu, last access: 21 May 2018) forpossible updates.

Also, comparisons between AGAGE in situ GC-MD andGC-MS data at Cape Grim and flask data from other groups(CSIRO, NIES, U. East Anglia, SIO, U. Heidelberg, MaxPlanck Inst. Mainz) have been and continue to be made. Ex-changes of tanks between the collaborating NIES group andAGAGE–SIO are also performed to compare absolute cali-brations. Also, there are routine data intercomparisons car-ried out within AGAGE for those gases measured on boththe AGAGE Medusa GC-MS and AGAGE GC-MD instru-ments. Finally, three AGAGE sites (SIO, Mace Head, andCape Grim) participated in the WMO-organized IHALACE(International HALocarbon in Air Comparison Experiment),round robin intercomparisons (Hall et al., 2014).

3.3 Flux estimation using measurements and models

A major goal of AGAGE is to estimate surface fluxes and/oratmospheric sinks (lifetimes) of trace gases by merging mea-surements and models using advanced statistical methods

(Prinn et al., 2000; Weiss and Prinn, 2011). Specifically, weuse a range of Bayesian methods, in which a priori estimatesof atmospheric sinks and surface fluxes (or uncertain param-eters in flux models) are adjusted to improve agreement withthe trace gas observations within estimated uncertainties, andit is important to ensure that the problems are well posed,that the ill-conditioning inherent in our emission estimationsis minimized, and that model and measurement imperfec-tions are accounted for properly (e.g., Prinn, 2000; Taran-tola, 2005). A basic requirement for all our inverse schemesis an accurate and realistic atmospheric chemical transportmodel (CTM). Even small transport errors can lead to signif-icant errors in estimated sources or sinks (Hartley and Prinn,1993; Mahowald et al., 1997; Mulquiney et al., 1998). Weuse a range of CTMs to estimate trace gas budgets at differentspatial scales: two-dimensional “box” models provide globalsource and sink estimates using baseline observations, globalthree-dimensional Eulerian models are used for estimatingfluxes at national to continental scales, and high-resolutionregional Lagrangian models provide fine-scale source esti-mations close to AGAGE monitoring sites.

Here, and in Sect. 3.4–3.7, we summarize the methodsand models actually used by AGAGE scientists to interpretAGAGE measurements. There are alternative methods andmodels that may give differences in estimated emissions, es-pecially at regional scales. The AGAGE publications gener-ally address the issue of differences, if any, between the esti-mated emissions and those reported in prior studies by othernon-AGAGE scientists. Some of these alternative methodsare addressed in Sect. 4.6 and 4.8, but it is beyond the scopeof this paper to review all the alternatives. Instead we referthe reader to two comprehensive books that provide in-depthsummaries of most of the major models and methods usedto estimate sources and sinks from measurements (Enting,2002; Kasibhatla et al., 2000).

We relate the vector of measured atmospheric mole frac-tions (y) to emissions or initial conditions in a “parametersvector” (x) using the “measurement” equation yobs =Hx+e.Here H is a matrix of sensitivities, or partial derivatives, ofsimulated measurements in y (=Hx) to each element in xand is derived using the CTMs; e describes the random com-ponent of the error due to errors in the measurements and inthe CTM. These errors form the error covariance matrix R.A prior estimate of x (xprior) is generally needed, with uncer-tainties contained in the error covariance matrix Pprior. For-mally, since the chemical lifetime for a reactive trace gas candepend on emissions of that gas, then H=H(x) so the equa-tion y =Hx is nonlinear in x. For many ozone-depleting andgreenhouse gases this nonlinearity is negligibly weak andis ignored. Exceptions exist, as discussed later. There are anumber of statistical approaches that have been developedand implemented to make these estimations (e.g., Kasibhatlaet al., 2000; Prinn, 2000; Rigby et al., 2011; Ganesan et al.,2014).




A common Bayesian statistical approach is “optimal es-timation” (e.g., Kasibhatla et al., 2000) in which one mini-mizes a “cost” function (J ) that is the sum of two quadraticforms: (yobs−y)

TR−1(yobs−y) that minimizes the weighteddifference between measured and modeled mole fractionsand (x−xprior)T P−1prior(x−xprior) that minimizes the weighteddifference between the estimated parameters and their prior.This minimization yields analytical solutions to x = xprior+G(yobs− y), P= (I−GH)Pprior, and the “gain” matrix G=PpriorHT (HPpriorHT +R)−1. Examples of this approach us-ing global 3-D Eulerian models are provided by Chen andPrinn (2005, 2006) for CH4, Xiao et al. (2010a) for CH3Cl,Xiao et al. (2010b) for CCl4, Rigby et al. (2010, 2011) forSF6, Saikawa et al. (2012, 2014b) for HCFC-22, and Huanget al. (2008) and Saikawa et al. (2014a) for N2O. Weak non-linearities may occur when lifetimes vary with emissions(e.g., OH depends on CO and CH4 emissions). This problemcan be addressed by recalculating the time-dependent partialderivative (sensitivity) H matrix after inversion of all the dataand then repeating the inversion with the new H matrix to en-sure convergence (Prinn, 2000).

Random measurement imperfections are associated within situ instrument precision, satellite retrieval errors, andinadequate sampling in space and time. If known, ran-dom model errors can also be incorporated into the model–measurement error covariance matrix (R). It is also impor-tant to recognize that correlated model–measurement errors,which comprise R, and errors in the prior contained in Ppriorare often poorly known quantities. Ganesan et al. (2014) ex-plicitly allow such uncertainties to be derived in the inver-sion to minimize the effect of subjective assumptions onderived fluxes. This hierarchical Bayesian method (Gane-san et al., 2014) incorporates “hyper parameters” that de-scribe the model–measurement and/or prior uncertainty co-variance matrices (R and P) in the inversion. This approachleads to solutions that are less sensitive to the (often sub-jective) assumptions that are required about uncertainties intraditional Bayesian approaches. The hierarchical inversionscheme cannot, in general, be solved analytically, and there-fore Markov chain Monte Carlo (MCMC) methods must beapplied to that sample from the posterior distribution usinga large number (∼ 104–105) of realizations of the parame-ter space (e.g., Rigby et al., 2011). Recently, this MCMCapproach has been extended to include problems in whichthe dimension of the parameter space is itself considered un-known using a so-called “reversible jump” MCMC algorithm(Lunt et al., 2016). This method has been applied to high-resolution regional inversions using a Lagrangian model tosample from a range of possible basis function decompo-sitions of the flux space, objectively determining the levelof decomposition that is appropriate to effectively minimize“aggregation error” (i.e., an inflexibility in the space thatcould lead to errors in the prior distribution unduly influ-encing the outcome of the inversion; Kaminski et al., 2001),

while maintaining an acceptable level of uncertainty reduc-tion.

We also address model structural errors and random andsystematic transport errors (i.e., errors in H) through the uti-lization of multiple model versions (Locatelli et al., 2013)and Monte Carlo methods (Prinn et al., 2001, 2005; Huanget al., 2008). The Monte Carlo methods also include system-atic errors in measurement calibration.

For the determination of the regional sources of tracegases, beginning with Chen and Prinn (2006) we nowfrequently merge measurements from the AGAGE andNOAA/ESRL/GMD stations and also aircraft and satelliteswhenever appropriate (e.g., Ganesan et al., 2017). Becausesource and

history of chemically and radiatively important atmospheric gases ... - mit global … · 2019. 1....

Documents