Skip navigation.
Home

Validation

Printer-friendly versionSend to friend

The validation requirements

Validation studies and resulting documentation within the Ozone CCI will address the following targets:

  • Time series of ECV total ozone data and of the main measurement and retrieval parameters with potential impact on the data quality (AMF, cloud properties, SZA, etc) should be visualised, at least in selected latitude zones and at a few representative ground-based stations. Any obvious quality issue like the frequent occurrence of outliers and unrealistic values should be detected, documented and filtered out appropriately before performing quantitative comparisons.
  • The error bar on ECV total ozone data (δTOC) shall be assessed and expressed as the percent relative difference with respect to correlative measurements of reference.
  • Statistical estimators of the difference like the bias and the spread shall be calculated over different time periods and over different ranges of relevant parameters as listed below. In case of frequent occurrence of outliers, median and interpercentile values shall be preferred over mean and standard deviation values as they reduce the influence of outliers. Calculation of mean values and associated standard deviation is nevertheless encouraged. In case of doubt, histograms of the relative difference might be helpful in determining the validity of statistical estimators.
  • In the treatment of statistics, care will be given to decouple as far as possible the different sources of ECV uncertainty and avoid misleading cancellation of mutually compensating errors.
  • Decadal stability of the bias and spread shall be assessed and expressed in %/decade.
  • Based on at least bi-weekly sampling of the time series over at least five years, shorter term stability of the bias and spread shall be assessed, including annual cycle, interannual variability and shorter term variability of the bias.
  • The dependence of the ECV data quality on main measurement and retrieval parameters like the solar zenith angle, ozone column amount, latitude, and cloud parameters (fractional cloud cover, cloud top height and albedo, etc. as appropriate) shall be investigated.
  • Studies shall be carried out at least in three geographical zones, in both hemispheres: tropics, middle latitudes and polar areas. Higher meridian and regional sampling is encouraged where possible.

The round robin protocol is contained in the product validation plan.

View an example: Validation of GDP5 total ozone

 

Stations

SAOZ stations

Stations/Latitude (North >0)/Longitude (West <0)/Altitude (m)  

  • Ny Alesund 78.908 11.883   0
  • Scoresby Sund 70.485 -21.952 67
  • Sodankyla 67.368 26.633 170
  • Zhigansk 66.793 123.351 200
  • Salekhard 66.531 66.676 137 
  • OHP 43.935 5.7116 683
  • Reunion -20.901 55.484 110
  • Bauru -22.347 -49.027 640
  • Kerguelen -49.352 70.256 36
  • Rio Gallegos -51.601 -69.319 15
  • Dumont d'Urville -66.666 140.017 45
  • Dome C -75.10 123.35 3250

SAOZ stations

Dobson Brewer Stations

The Ground-based measurements

Archived total ozone column measurements from the World Meterological Organization (WMO) - Global Atmosphere Watch (GAW) network routinely deposited at the World Ozone and Ultraviolet Radiation Data Centre (WOUDC) in Toronto, Canada (http://www.woudc.org) are utilized for the ground-based measurements reference. The WOUDC archive contains total ozone column data mainly from Dobson and Brewer UV spectrophotometers as well as from M-124 UV filter radiometers from the early fifties onwards. In general, spatial and temporal coincidences offered by the Dobson and Brewer networks are sufficient to cover a wide geographical extent for the validation of a satellite sensor, however, with better coverage over land with respect to sea and over the Northern Hemisphere compared to the Southern Hemisphere. Total ozone column data from a large number of stations have already been used extensively both for trend studies [e.g. WMO 1998, 2002, 2006] as well as for validation of satellite total ozone data [e.g. Lambert et al., 1999; Fioletov et al., 1999; Lambert et al., 2000; Bramstedt et al., 2003; Labow et al., 2004, Weber et al., 2005, Balis et al., 2007a].

Van Roozendael et al., [1998] have shown that Dobson and Brewer data can agree within 1% when the major sources of discrepancy are properly accounted for. Dobson measurements suffer from a temperature dependence of the ozone absorption coefficients used in the retrievals which might account for a seasonal variation in the error of ±0.9% in the middle latitudes and ±1.7% in the Arctic, and for systematic errors of up to 4% [Bernhard et al., 2005]. The error of individual total ozone measurements for a well maintained Brewer instrument is about 1% (e.g. Kerr, 1988). Despite the similar performance between the Brewer and Dobson stations, small differences within ±0.6% are introduced due to the use of different wavelengths and different temperature dependence for the ozone absorption coefficients [Staehelin et al., 2003]. Dobson and Brewer instruments might also suffer from long-term drift associated with calibration changes. Additional problems arise at solar elevations lower than 15°, for which diffuse and direct radiation contributions can be of the same order of magnitude.

To prepare the ground-based data set for trend analysis, we investigated the quality of the total ozone values of each station and instrument that deposited data at WOUDC for any time period after 1995. For detailed discussion on the selection process and the exclusion procedures please refer to Balis et al., 2007b. Using the methodology described in great detail in that publication, 32 Brewer and 47 Dobson stations were considered as potentials for the comparisons with GOME, Sciamachy and GOME-2 total ozone column data. These stations are sorted with increasing latitude and listed in Table 1 for each of the two types of ground-based instrument.

 

Table 1: List of ground-based stations used for the comparisons

NUMBER

INSTR/NT

NAME

LATITUDE

LONGITUDE

ELEVATION

21

brewer

EDMONTON

53.57

53.57

668

24

brewer

RESOLUTE

74.72

74.72

64

35

brewer

AROSA

46.77

46.77

1860

53

brewer

UCCLE

50.8

50.8

100

65

brewer

TORONTO

43.78

43.78

198

76

brewer

GOOSE

53.32

53.32

44

77

brewer

CHURCHILL

58.75

58.75

35

96

brewer

HRADEC-KRALOVE

50.18

50.18

285

99

brewer

HOHENPEISSENBERG

47.8

47.8

975

100

brewer

BUDAPEST

47.43

47.43

140

123

brewer

YAKUTSK

62.08

62.08

98

174

brewer

LINDENBERG

52.22

52.22

98

213

brewer

EL-ARENOSILLO

37.1

37.1

41

261

brewer

THESSALONIKI

40.52

40.52

4

262

brewer

SODANKYLA

67.37

67.37

179

267

brewer

SONDRESTROM

67

67

150

279

brewer

NORKOPING

58.58

58.58

0

284

brewer

VINDELN

64.25

64.25

0

290

brewer

SATURNA

48.78

48.78

0

295

brewer

MT.WALIGUAN

36.17

36.17

3816

301

brewer

ISPRA

45.8

45.8

0

305

brewer

ROME-UNIVERSITY

41.9

41.9

0

308

brewer

MADRID

40.45

40.45

0

314

brewer

BELGRANO

-77.87

-77.87

255

315

brewer

EUREKA

79.89

79.89

10

316

brewer

DEBILT

52

52

0

322

brewer

PETALING-JAYA

3.1

3.1

46

326

brewer

LONGFENSHAN

44.75

44.75

0

331

brewer

POPRAD-GANOVCE

49.03

49.03

0

332

brewer

POHANG

36.03

36.03

0

338

brewer

REGINA

50.21

50.21

0

346

brewer

MURCIA

38

38

69

2

dobson

TAMANRASSET

22.8

22.8

1395

7

dobson

KAGOSHIMA

31.63

31.63

283

11

dobson

QUETTA

30.18

30.18

1799

12

dobson

SAPPORO

43.05

43.05

19

14

dobson

TATENO

36.05

36.05

31

19

dobson

BISMARCK

46.77

46.77

511

20

dobson

CARIBOU

46.87

46.87

192

27

dobson

BRISBANE

-27.47

-27.47

5

29

dobson

MACQUARIE-ISLAND

-54.48

-54.48

6

31

dobson

MAUNA-LOA

19.53

19.53

3397

35

dobson

AROSA

46.77

46.77

1860

40

dobson

HAUTE-PROVINCE

43.92

43.92

580

43

dobson

LERWICK

60.15

60.15

90

53

dobson

UCCLE

50.8

50.8

100

57

dobson

HALLEY-BAY

-75.52

-75.52

31

67

dobson

BOULDER

40.02

40.02

1634

68

dobson

BELSK

51.83

51.83

180

84

dobson

DARWIN

-12.47

-12.47

0

91

dobson

BUENOS-AIRES

-34.58

-34.58

25

96

dobson

HRADEC-KRALOVE

50.18

50.18

285

99

dobson

HOHENPEISSENBERG

47.8

47.8

975

101

dobson

SYOWA

-69

-69

21

105

dobson

FAIRBANKS

64.8

64.8

138

106

dobson

NASHVILLE

36.25

36.25

182

107

dobson

WALLOPS-ISLAND

37.87

37.87

4

111

dobson

AMUNDSEN-SCOTT

-89.98

-89.98

2835

152

dobson

CAIRO

30.08

30.08

35

159

dobson

PERTH

-31.95

-31.95

2

175

dobson

NAIROBI

-1.27

-1.27

1710

190

dobson

NAHA

26.2

26.2

29

191

dobson

SAMOA

-14.25

-14.25

82

199

dobson

BARROW

71.32

71.32

11

200

dobson

CACHOEIRA-PAULISTA

-22.68

-22.68

573

208

dobson

SHIANGHER

39.77

39.77

13

209

dobson

KUNMING

25.02

25.02

1917

213

dobson

EL-ARENOSILLO

37.1

37.1

41

214

dobson

SINGAPORE

1.33

1.33

14

216

dobson

BANGKOK

13.73

13.73

2

219

dobson

NATAL

-5.83

-5.83

32

226

dobson

BUCHAREST

44.48

44.48

92

232

dobson

VERNADSKY-FARADAY

-65.25

-65.25

7

245

dobson

ASWAN

23.97

23.97

193

252

dobson

SEOUL

37.57

37.57

84

253

dobson

MELBOURNE

-37.48

-37.48

125

256

dobson

LAUDER

-45.03

-45.03

3701

265

dobson

IRENE

-25.25

-25.25

1524

268

dobson

ARRIVAL-HEIGHTS

-77.83

-77.83

250

 

The monthly mean averages and trend analysis input files necessary for the purposes of this study were then created using the daily comparison measurements with two different sets of only coincident datasets being considered: the monthly mean and associated standard deviation of the ground-based measurements and the equivalent one for the satellite measurements. Due to the differences in time-span and spatial resolution between the three satellite datasets considered, not all of the above shown ground-based stations were utilized in the comparisons of all three satellite instruments. The locations of the two types of stations around the globe are shown in the two graphs below.

 

 

Figure 1a. The locations of the Brewer network of stations on a global scale.

  

Figure 1b.The locations of the Dobson network of stations on a global scale.

Ozone Lidar Sondes 

 

DATA ON THE VERTICAL DISTRIBUTION OF OZONE 

The vertical distribution of ozone is measured complementarily by ground-based stratospheric ozone lidars and by balloon-borne electrochemical ozonesondes. Those ground-based instruments are operated in networks which contribute to WMO’s Global Atmosphere Watch programme (GAW) [WMO TD No. 1384]. Data sets suitable for the correlative analysis of CCI products are collected from the Data Host Facility (DHF) of the Network for the Detection of Atmospheric Composition Change (NDACC) and from the World Ozone and Ultraviolet Radiation Data Centre (WOUDC). The geographical distribution of ozone profile instruments having archived data to the WOUDC and NDACC DHF in the Envisat era is displayed in Figure 2, on top of an illustrative global field of total ozone. Further details regarding the availability of data suitable for the validation of CCI products are given hereafter. 

 

Figure 2. Geographical distribution of ground-based lidar and ozonesonde stations having archived regularly ozone profile data to the NDACC DHF and/or the WOUDC in the Envisat era.

STRATOSPHERIC OZONE LIDARS (NDACC)

Originating system

Lidar spectroscopy is an analytical technique with a long history in environmental science and chemistry. This method has been used widely in atmospheric chemistry and has a heritage in ground-based as well as space-based instruments. The NDACC has accepted lidar measurement techniques as valid methods for measuring and monitoring stratospheric ozone, temperature, and aerosols, tropospheric ozone, and tropospheric and lower stratospheric water vapor. Atmospheric ozone profiles from around 12-18 km up to 45–50 km can be measured using the DIfferential Absorption Lidar (DIAL) technique [Mégie et al., 1977]. In NDACC there are thirteen DIAL systems dedicated to the measurement of stratospheric ozone at fixed locations, and one mobile DIAL system.

Data class

Ground-based remote sensing data

Sensor type and key technical characteristics

The basic principle of the DIAL technique is to transmit two short laser pulses vertically into the atmosphere, one having a wavelength in an absorption band of ozone (typically 308 nm) and the other not absorbed by ozone – or not so strongly absorbed (typically 355 nm). The light is scattered by the atmospheric molecules and particles, and a fraction is collected back on the ground with a telescope. Knowing the speed of light, the distance to a scattering molecule or particle is deduced from the travel time of the photons on their way upward and then back to the lidar. The light collected by the lidar telescope is geometrically and spectrally separated (e.g., with optical filters and beam splitters) and detected with photosensitive devices (photomultipliers, abbreviated PMTs) where it is converted to electro-photon counts, the so-called “lidar signals”. The signals are sampled in time (i.e., distance) and after various corrections are proportional to the product of the number of photons emitted by the number of backscattering molecules. This proportionality is expressed by the so-called “lidar equation”. This equation is the starting point for the retrieval of many atmospheric properties. The ozone number density can be retrieved from the difference in slope between the absorbed and non-absorbed (reference) backscattered laser signals. Following major volcanic eruptions, it is necessary to avoid corruption of the backscattered signal caused by enhanced aerosols. Nitrogen Raman scattering can be used with the DIAL principle to derive ozone in the lower stratosphere under these conditions.

The lidars involved in the NDACC differ by the size of their receiver telescope and their laser power, i.e., power–aperture product. However, these differences do not significantly affect the derivation methodology and their main effect is on the level of the counting noise that only restricts the altitude range of the measurement or the integration time for a given accuracy. In addition, many variations in the actual lidar implementations can be noted, which can explain the differences observed between the various lidars involved in the NDACC. The review by Keckhut et al. [2004] describes briefly the different stratospheric ozone instruments accepted for NDACC, gives appropriate references, and summaries the results of validation exercises carried out in the framework of NDACC.

Data availability & coverage

The geographical distribution of stratospheric ozone lidars having archived regularly data to the NDACC DHF in the Envisat era is displayed in Figure 2. The availability of lidar data records at DHF in the 1991-2010 era is displayed in Table 1. NDACC stratospheric ozone lidars operate only by clear-sky nights. Some of them perform measurements systematically, weather permitting, while others operate only on campaign basis or if resources permit.

Source data product name & reference to product technical specification documents

This documentation is available on http://ndacc.org

Data quantity

One file of the order of 50-100 kB is produced by night of measurement. NDACC stratospheric ozone lidars operate only by clear-sky nights. Some of them perform measurements systematically, weather permitting, while others operate only on campaign basis or if resources permit.

Data quality and reliability

To ensure quality and consistency of the NDACC lidars operation and products, a number of protocols have been formulated covering such topics as validation, measurements and instruments intercomparisons, and theory and analysis. The members of the NDACC Lidar Working Group (LWG) are committed to follow the principles of these protocols, and the LWG meets every two years to review and coordinate the activities necessary to the valuable contribution of the lidars to NDACC. Lidar Investigators must provide the following information:

  • A document describing the instrument and data acquisition procedures.
  • A document describing the algorithm to be used, including the forward and retrieval models, the method of error analysis, and the ancillary data (spectroscopic data, atmospheric parameters) used for the inversion.
  • The validation record of the instrument.

In addition, NDACC lidar Instrument Investigators are required to participate in ongoing validation exercises such as algorithm intercomparisons and satellite data long-term analysis.

On average, the ozone measurement bias achieved by NDACC lidars is around 5-10% below 20 km for instruments without Raman channels and 5% for instruments with Raman channels, around 2% at altitudes within 20-35 km, and around 5-10% at altitudes above 40 km [Keckhut et al., 2004]. The ozone measurement precision achieved by NDACC lidars is around 1% up to 30 km, 2–5% at 40 km and 5–25% at 50 km. The network of NDACC lidars can, in principle, be considered as homogeneous within ±2% between 20–35 km.

Ordering and delivery mechanism

The NDACC Data Host Facility provides public access to the data.

 

Table 1. Availability of lidar

ozone profile data in the NDACC DHF

[from NDACC website

BALLOON-BORNE ELECTRO-CHEMICAL OZONESONDES (WOUDC and NDACC)

 

Originating system

Ozonesondes were introduced into atmospheric science in the 1960s [Brewer and Milford, 1960; Komhyr, 1964, 1967, 1969] and have had a long development history. The ozonesonde is balloon-borne instrument of light weight that is coupled to a meteorological radiosonde. It measures the vertical profile of pressure, temperature and humidity (PTU) as the balloon ascends through the atmosphere. During the ascent the ozonesonde/radiosonde package telemeters ozone and PTU data – and also sometimes wind direction and speed – to a ground receiving station through the radiosonde transmitter. These in situ instruments using balloon platforms are unique in providing ozone profiles at vertical resolution of about 150 m in the troposphere and lower stratosphere, with maximum altitudes at balloon burst, usually near 30 km. In the framework of WMO’s Global Atmosphere Watch [WMO TD No.1384], there is a network of global ozone sounding stations which partially overlaps the NDACC network.

Data class

In situ measurement from lightweight meteorological balloons

Sensor type and key technical characteristics

The heart of the ozonesonde is an electrochemical concentration cell that senses ozone as is reacts with a dilute solution of potassium iodide to produce a weak electrical current proportional to the ozone concentration of the sampled air according to the following redox reaction:

Ambient air is continuously forced into the sensing cell by a battery driven sampling pump. An electrical current is generated proportional to the mass flow rate of ozone through the cell. By knowing the volume flow rate and temperature, the electrical current can be converted to an ozone concentration under the assumption that the ozone reaction with potassium iodide is quantitatively known.

Long term monitoring networks of ozone sounding stations as well as project dedicated networks have developed optimal practices over the years. Within these networks three different types of ozonesondes are still employed: electrochemical concentration cell (ECC), Brewer Mast (BM), and the Japanese KC sonde.

The ECC ozonesonde was developed by Komhyr [1969, 1971]. It consists of two half cells, made of Teflon, which serve as cathode and anode chamber, respectively. Both half cells contain a platinum mesh serving as electrodes. They are immersed in potassium iodide solution of different concentrations. The two chambers are linked together by an ion bridge in order to provide an ion pathway and to prevent mixing of the cathode and anode electrolytes. In contrast to the Brewer-Milford type of electrochemical ozone sensors (see below), the ECC does not require an external electrical potential. ECC ozonesondes are now the most widely used ozonesonde type. Two companies produce ECC sondes, Science Pump Corporation (SPC) and ENSCI Corporation. The two manufacturers recommend their own procedures, which differ slightly. These along with expertise gained in the operational ozonesonde networks, such as NDACC, and comparisons organized by the World Meteorological Organization (WMO) have been used to improve these recommendations. These improvements have evolved to a stage where the publication of SOPs for ECC ozonesondes will be made available soon by the WMO.

The Brewer-Mast sonde evolved from the Oxford-Kew ozonesonde developed by Brewer and Milford (1960). The Brewer-Milford type ozone sensor consisted of a single electrochemical cell with a silver anode and platinum cathode immersed in an alkaline potassium iodide solution. The Brewer-Mast sondes were manufactured by the Mast Keystone Corporation (Reno, Nevada, USA) and its predecessor the Mast Development Corporation. Since 1976, a document defining the SOPs for the Brewer-Mast ozonesonde has been available. It defines the different steps to complete proper and reproducible ozone profiles with BM sondes. Brewer-Mast sondes have constituted several long-term data records of interest. Presently only one station (Hohenpeissenberg) is still using Brewer-Mast ozonesondes operationally.

The third still active instrument is the Japanese sonde KC92 [Kobayashi et al., 1966; Fujimoto et al., 2004]. This ozonesonde type is based on a modified version of the carbon-iodine ozone sensor [Komhyr, 1969]. The ozone sensor is an electrochemical cell containing a platinum gauze as cathode and an activated carbon anode immersed in an aqueous neutral potassium iodide/potassium bromide solution. Successive versions (KC-68, RSII-KC79, KC92…) of this instrument have been used by the Japanese Meteorological Agency and produced long-term data records of interest at several sounding stations, however, KC92 sondes are now being replaced by ECC sondes. No other stations have used KC sondes.

Data availability & coverage

The geographical distribution of ozonesondes archiving regularly data to the WOUDC and NDACC DHF is displayed in Figure 2. Ozonesonde data records with clear interest for this CCI-Ozone project are listed in Table 2.

Source data product name & reference to product technical specification documents

Ozonesonde data format was discussed extensively at the Ozonesonde WG meeting in February 2009 in Jülich, Germany. A working group was formed at that time to complete the work done there and to provide a document to describe the format and provide example files. The format is based on the NASA/AMES 2160 format and efforts are taken to standardize this format amongst all stations to avoid the need for a multiplicity of readers to access NDACC ozonesonde data. The relevant documents are posted on the ozonesonde working group web site. NDACC investigators will be encouraged to submit all new data with the revised format, and, although not required, to consider resubmitting all their previous data in the new format.

This documentation and other relevant information are available on http://woudc.org and http://ndacc.org

Data quantity

One file of the order of 50 kB is produced by ozonesonde flight. Some stations launch an ozonesonde twice a week, others only once a month, others only during special events (e.g. ozone hole season) or campaigns.

Data quality and reliability

Many intercomparisons between different ozonesonde types and reference instruments have been conducted over the last 40 years [Attmannspacher and Dütsch, 1970; 1981; Barnes et al., 1985; Hilsenrath et al. 1986; Kerr et al., 1994; Beekmann et al.,1994; 1995; Komhyr et al., 1995a; 1995b; Reid et al., 1996; Boyd et al., 1998; Johnson et al., 2002; Fioletov et al. 2006; Terao and Logan, 2007; Smit et al., 2007; Deshler et al., 2008; Stübi et al. 2008]. Therefore, this is a proven technique which doesn't require further justification to be accepted as a reference instrument and as a validation source.

The peculiarity of ozonesondes is that every instrument is usually new and flown only once. Therefore, the notion of a reference/standard instrument has to be interpreted differently than for other types of instruments. In the case of ozonesondes, the main emphasis is on the standard operating procedures (SOPs) for preparation of the instruments for flight, and on the data processing. The WMO has attributed the role of the world calibration center for ozonesondes (WCCOS) to the Research Center in Jülich. The primary goals of the WCCOS are to promote understanding of the instrument, to establish well documented SOPs, and to assess differences in instrument manufacturers and in variations of SOPs in use. The WCCOS along with NDACC investigators were instrumental in establishing the guidelines behind the presently recommended SOPs which should be available on the WMO web site soon. At that time this document will be cross linked from the NDACC ozonesonde web site. WCCOS continues to periodically test the quality of ECC ozonesondes provided by the two manufacturers. The role of the WCCOS is endorsed by the NDACC ozonesonde working group and there is a good collaboration between NDACC and WCCOS.

From recent laboratory [Smit et al., 2007, WMO TD No. 1218 and 1225] and field [Deshler et al., 2008] experiments, it can be concluded that, if the SOPs are strictly followed, the variability (precision) between sondes is estimated to be ± 0.1 mPa in the troposphere and ± 0.2 mPa (± 2%), that is, very reproducible and consistent results.

Ordering and delivery mechanism

The NDACC Data Host Facility provides public access to the data.

Table 2. Availability of ozonesonde data records in the WOUDC in the ERS-2/Envisat timeframe: station, country code, latitude, longitude, instrument type (ECC/Brewer or Mast), instrument model, first and last date of measurement as of 31.12.2009 [from information retrieved from the WOUDC]

 

 

REFERENCES

Attmannspacher, W. and H. Dütsch, International Ozone Sonde Intercomparison at the Observatory of Hohenpeissenberg, Berichte des Deutschen Wetterdienstes, 120, 1970.

Attmannspacher, W. and H. Dütsch, 2nd International Ozone Sonde Intercomparison at the Observatory of Hohenpeissenberg, Berichte des Deutschen Wetterdienstes, 157, 1981.

Barnes, R.A., A.R. Bandy, and A.L. Torres, Electrochemical concentration cell ozonesonde accuracy and precision, J. Geophys. Res., 90, 7881-7887, 1985.

Beekmann, M., G. Ancellet, G. Megie, H.G.J. Smit, and D. Kley, Intercomparison campaign for vertical ozone profiles including electrochemical sondes of ECC and Brewer-Mast type and a ground based UV-differential absorption lidar, J. Atmos. Chem., 19, 259-288, 1994.

Beekmann, M., G. Ancellet, D. Martin, C. Abonnel, G. Duverneuil, F. Eidelimen, P. Bessemoulin, N. Fritz, and E. Gizard, Intercomparison of tropospheric ozone profiles obtained by electrochemical sondes, a ground based lidar and airborne UV-photometer, Atm. Env., 29, 1027- 1042, 1995

Boyd, I.S., G.E. Bodeker, B.J. Connor, D.P.J. Swart, E.J. Brinksma , An assessment of ECC ozonesondes operated using 1% and 0.5% KI cathode solutions at Lauder, New Zealand, Geophys.Res. Lett., 25, 2409-2412, 1998

Brewer, A.W., and Milford, J.R., The Oxford-Kew ozone sonde. Proc. Roy. Soc. London, 256, 470-95, 1960.

Deshler, T., J. M. Mercer, H. G.J. Smit, R. Stubi, G. Levrat, B. J. Johnson, S. J. Oltmans, R. Kivi, A. M. Thompson, J. Witte, J. Davies, F. J. Schmidlin, G. Brothers, T. Sasaki, Atmospheric comparison of electrochemical cell ozonesondes from different manufacturers, and with different cathode solution strengths: The Balloon Experiment on Standards for Ozonesondes, J. Geophys. Res., 113, D04307, doi:10.1029/2007JD008975, 2008

Fioletov, V. E., D. W. Tarasick, and I. Petropavlovskikh, Estimating ozone variability and instrument uncertainties from SBUV(/2), ozonesonde, Umkehr, and SAGE II measurements: Shortterm variations, J. Geophys. Res., 111, D02305, doi:10.1029/2005JD006340, 2006

Fujimoto Toshifumi, Takahiro Sato, Katsue Nagai, Tatsumi Nakano, Masanori Shitamichi, Yoshihiro Kamata, Seiji Miyauchi, Kazuaki Akagi, Toru Sasaki, Further evaluation and improvements of Japanese KC-Ozonesonde through JOSIE-2000, Quadrennial Ozone Symposium 2004, extended abstracts, 224, 2004

Hilsenrath, E., W. Attmannspacher, A. Bass, W. Evans, R. Hagemeyer, R.A. Barnes, W. Komhyr, K. Mauersberger, J. Mentall, M. Proffitt, D. Robbins, S. Taylor, A. Torres, and E. Weinstock, Results from the balloon ozone intercomparison campaign (BOIC), J. Geophys. Res., 91, 13137-13152, 1986

Johnson, B.J., S.J. Oltmans, H. Voemel, H.G.J. Smit, T. Deshler, and C. Kroeger, ECC Ozonesonde pump efficiency measurements and tests on the sensitivity to ozone of buffered and unbuffered ECC sensor cathode solutions, J. Geophys. Res., 107, D19 doi: 10.1029/2001JD000557, 2002

Keckhut, P., S. McDermid, D. Swart,T. McGee, S. Godin-Beekmann, A. Adriani, J. Barnes,J-L. Baray, H. Bencherif,H. Claude,A. G. di Sarra,i G. Fiocco,G. Hansen, A. Hauchecorne,T. Leblanc, Choo Hie Lee, S. Pal, G. Megie, H. Nakane, R. Neuber, W. Steinbrechth, and J. Thayer, Review of ozone and temperature lidar validations performed within the framework of the Network for the Detection of Stratospheric Change, J . Environ. Monit. , 6 , 721 – 733, 2004.

Kerr, J.B., H.Fast, C.T. McElroy, S.J. Oltmans, J.A. Lathrop, E. Kyro, A. Paukkunen, H. Claude, U. Köhler, C.R. Sreedharan, T. Takao, and Y.,Tsukagoshi, The 1991 WMO International OzoneSonde Intercomparison at Vanscoy, Canada, Atm. Ocean, 32, 685-716, 1994

Kobayashi, J., and Y. Toyama., On various methods of measuring the vertical distribution of atmospheric ozone (III) - Carbon iodide type chemical ozonesonde. Pap. Met. Geophys., 17, 113-126, 1966

Komhyr, W.D., A carbon-iodide ozone sensor for atmospheric sounding, Proceedings of the Ozone Symposium, Alburquerque, N.M., Edited by H.V. Dutsch, Secretariat of the World Meteorological Organization, Geneva, Switzerland, p. 26, 1964.

Komhyr, W.D., Nonreactive gas sampling pump, Rev. Sci. Inst., 38, 981-983, 1967.

Komhyr, W.D., Electrochemical concentration cells for gas analysis, Ann.Geoph., 25, 203-210, 1969.

Komhyr, W. D., R. A. Barnes, G. B. Brothers, J. A. Lathrop, D. P. Opperman, Electrochemical concentration cell ozonesonde performance evaluation during STOIC 1989, J. Geophys. Res., 100, 9231-9244, 1995a,

Komhyr, W.D., B.J. Connor, I.S. Mcdermid, T.J. McGee, A.D. Parrish, J.J. Margitan, Comparison of STOIC 1989 ground-based lidar, microwave spectrometer, and Dobson spectrophotometer Umkehr ozone profiles with ozone profiles from balloon-borne electrochemical concentration cell ozonesondes, J. Geophys. Res., 100, 9273-9282, 1995b

Reid S.J., G. Vaughan, A.R. Marsh, and H.G.J. Smit, Intercomparison of ozone measurements by ECC sondes and BENDIX chemiluminescent analyser, J. Atm. Chem., 25, 215-226, 1996

Smit, H. G.J, W. Straeter, B. J. Johnson, S. J. Oltmans, J. Davies, D. W. Tarasick, B. Hoegger, R. Stubi, F. J. Schmidlin, T. Northam, A. M. Thompson, J. Witte, I. Boyd, F. Posny, Assessment of the performance of ECC-ozonesondes under quasi-flight conditions in the environmental simulation chamber: Insights from the Jülich Ozone Sonde Intercomparison Experiment (JOSIE), J. Geophys. Res., 112, D19306, doi:10.1029/2006JD007308, 2007

Stübi R., Levrat G., Hoegger B., Viatte P., Staehelin J., Schmidlin F. J., In-flight comparison of Brewer-Mast and electrochemical concentration cell ozonesondes, J. Geophys. Res., 113, D13302, doi:10.1029/2007JD009091, 2008

Terao, Y., and J. A. Logan, Consistency of time series and trends of stratospheric ozone as seen by ozonesonde, SAGE II, HALOE, and SBUV(/2), J. Geophys. Res., 112, doi:10.1029/2006JD007667, 2007

World Meteorological Organization, JOSIE-1998: Performance of ECC Ozone Sondes of SPC-6A and ENSCI-Z Type, GAW Report No. 157 / WMO TD No. 1218

World Meteorological Organization, JOSIE-2000: Jülich Ozone Sonde Intercomparison Experiment 2000, GAW Report No. 158 / WMO TD No. 1225.

World Meteorological Organization, WMO Global Atmosphere Watch (GAW) Strategic Plan: 2008-2015, GAW Report No. 172 / WMO TD No. 1384.