Appendix II - UV/Vis Instruments

P. V. Johnston, J-P. Pommereau and H. K. Roscoe

Revision - November 1998 (finalised August 1999)

Introduction

This document describes the validation process for new zenith viewing UV/Vis instruments, and the criteria for maintaining data quality from existing instruments. It is written to cover measurements of the species NO₂ and ozone, which are the ones most accurately measured using the zenith sky viewing technique at this time. However, much of this document will also apply to measurements of the species OClO and BrO. NO₂ and ozone were the primary species measured at the most recent NDACC UV/Vis Instrument Intercomparison which was held at the Observatoire de Haute-Provence (OHP), France, in June 1996, and this revison of the appendix includes new procedures advanced after discussions with the UV/Vis community at OHP. The SCUVS-3 BrO working group also held a BrO measurement intercomparison at OHP in 1996 following the formal NDACC intercomparison, and the resulting report is expected to be a reference for future NDACC BrO measurement intercomparisons. The reports on the UV/Vis Data Analysis Intercomparison in 1991, the Lauder Instrument Intercomparison in 1992 and the OHP Instrument Intercomparison, should be read with this document (see references).

Because UV/Vis spectroscopic measurements are used to answer a variety of scientific questions, this appendix will attempt to define the certification of groups and their instruments for two applications: (Further ones should be defined in the future.)

(1) NDACC trend determination and global studies;

(2) NDACC stratospheric chemistry process studies and satellite validation.

Trend measurements and global studies require a long term dedicated approach to the maintenance of the quality of the measurements and the archiving of data. To determine long term trends requires stable and well calibrated instruments operated by groups that have a thorough understanding of the measurement technique and who have demonstrated competence in the management of long term programmes. Global studies include, for example, identifying possible interhemispheric asymmetries of species such as NO₂, which requires the maintenance of good inter-instrument calibration. Type (1) measurements of the highest possible accuracy at solar zenith angles up to 95° in the visible are the most suitable for NO₂ vertical profiling (Preston et al., in preparation). Because type (1) studies would probably involve instruments permanently located at primary or secondary sites, instrument portability and deployment flexibility are not important criteria.

In contrast, process studies and satellite validation campaigns would be better facilitated by a denser network of measurements at locations (such as polar regions) being studied to answer specific questions. Except for intercomparisons, accuracy and long term stability would be less important in this case, whereas deployment flexibility plus operational reliability in hostile environments would be more important. Typical of the issues that can and have been investigated in such campaigns are: Antarctic total ozone measurements, NO₂ seasonal cycles, SAGE / TOMS / GOME / ENVISAT etc., satellite validations and PSC statistics. Some satellite validations may require type (1) measurements after the initial debug stage however.

Ideally a network of a few type (1) measurements and many type (2) measurements in polar regions would best serve vortex and collar region chemistry studies. This multiple specification approach will be extended to cover further scientific needs as they are identified. Such an approach is more realistic with UV/Vis instruments because their cost is lower than IR or microwave instruments.

The accuracy of vertical column derived from zenith measurements is determined by several factors: measurement accuracy (random and systematic), airmass factor (AMF) calculations and the errors introduced by multiple scattering in clouds containing pollution (mainly in the continental northern hemisphere). The limiting accuracy of the most accurate instruments operating at clean sites is determined by the accuracy of the AMF calculations and the accuracy of the calculation of the NO₂/O₃ amount in the reference spectrum. Work to address AMF and cloud effect modeling limitations is a planned future UV/Vis activity.

To summarise, the principal acceptance criteria for each type are: type (1) certification; highest possible accuracy and long term commitment: type (2) certification; the ability to deploy instruments that can operate reliably in a remote environment and that have an accuracy adequate to answer the scientific questions posed.

Before a formal intercomparison with a certified instrument is planned, the group whose instrument is being assessed may be asked to supply the following to the NDACC steering committee’s UV/Vis Instrument Group or designated representative.

• A detailed technical description of the instrument including sensitivity limits and general operating parameters.

• An outline of the spectral analysis technique used with particular details of the number and source of the cross sections used. If the cross sections are not published or those commonly used by the UV/Vis community, a comparison between the cross section normally used and a generally accepted one is required. Sufficient information is required so that it can be seen that the laboratory cross sections used are of sufficient resolution that their convolution with the measured instrument (slit) function will produce sufficiently accurate analysis cross sections.

• An example of a raw measured spectrum, a ratio spectrum of a twilight spectrum (near 90 degrees SZA) to a midday spectrum, and a spectrum demonstrating the quality of the cross sections fit to the ratio spectrum.

• Spectra or data showing the instrument resolution (slit function), an estimate of the stray light levels and the instrument polarisation characteristics, in the wavelength interval expected to be used for the intercomparison. This would normally consist of:

(1) one or more spectra containing lines from spectral lamps (e.g., low pressure mercury, neon, argon, krypton or xenon) or a laser;

(2) clear sky spectra measured using fixed detector gain, with and without suitable Schott glass short λ cut filters;

(3) two or three spectra showing the relative transmission of the instrument for different polarisation axes that has been measured using a white light source filling the field of view together with a suitable film polariser. For instruments not using an input fibre optic cable, the polariser is positioned at the entrance aperture and two measurements are made with the polariser orientated parallel and normal to the grating ruling. For instruments using a fibre optic cable, normal and parallel to the grating ruling has no meaning at the entrance to the fibre where the film polariser must be placed for the polarisation measurement. This is because many fibre optics cables partially randomise the polarisation, a characteristic that is often used to reduce the sensitivity of zenith viewing spectrometers to sky polarisation effects. In this case, the polariser is positioned at the entrance of the fibre optic cable and three measurements are made with the polariser rotated 45 degrees between them.

• Examples of any existing measurement data and the results of any previous intercomparisons (including conditions and references).

Discussions and data exchange between the PI and the Instrument Group may be required, as the Instrument Group must be satisfied with this part of the evaluation before proceeding to an intercomparison.

The instrument intercomparison must follow the “blindness” rules detailed below and adhere to the general philosophy of the NDACC Instrument Intercomparison Protocol. The instrument specific requirements of this protocol follow. In the general case, one or more new instruments are to be evaluated by comparison with an already certified instrument or instruments (called the reference instrument(s) in the following) under the supervision of an impartial referee. The certified instrument(s) to be used will be designated by the Instrument Group.

• The intercomparison should be conducted at a site that is reasonably free from tropospheric NO₂ contamination. Rapidly changing contamination due to a close source is unacceptable, making it essential that the site chosen for the intercomparison is remote from emissions such as those from cities and industrial plants. Some tropospheric contamination may be acceptable if it is only present for part of the day, such as in the late afternoon when it only affects only the evening measurements. Clean morning measurements can be sufficient for some assessments. The site should also be one that would normally experience both clear (>30% expectation) and cloudy days over the intercomparison period. Measurements during both clear and cloudy conditions are important to ensure that different skies do not introduce errors in the results.

• The intercomparison should be conducted for a period of not less than 10 days with all instruments operating correctly.

• Measurements, taken by the one or more instruments being evaluated and the reference instrument(s), should be made over the whole day if possible, and not less than the SZA range 70 to 95 degrees (or as dark as acceptable signal to noise dictates) together with a period near midday, for both morning and evening twilight periods, each day of the intercomparison. The measurement integration period should be less than the time taken for a 1 degree change of solar zenith angle at twilight (5 minutes at midlatitudes) or a maximum of 5 minutes.

• Measurements taken by the instrument(s) being evaluated and the reference instrument(s) should be coincident in time. If this is not possible the compared data sets must be interpolated for comparison using a formula agreed to by the Instrument Group. Also, the effects that changing light intensity and changing column amount have on the results must be properly understood. The resulting error in the compared column amounts due to the use of different integration methods must be demonstrated to be small. For the 1992 Lauder Intercomparison such errors were calculated to be less than 1%.

• The wavelength interval used should be the same for the instrument(s) being evaluated and the reference instrument(s). This requirement may be difficult, but without it close comparison accuracy (better than 5%) may not be possible. This could create unnecessary doubts about the instrument being evaluated.

• The cross sections used in the analysis must be from the same laboratory measured source, or have been previously intercompared at a similar resolution.

• Periodically, an NO₂ calibration cell should be measured by the instrument(s) being evaluated and the reference instrument(s). This checks both the cross sections and analysis methods used during the intercomparison. The cell should be temperature stabilised or be one that has sufficiently low NO₂ concentration that dimmer formation is not a significant source of error for the temperature variations expected. Because of unexpected variations seen in some cells after long periods in the dark, or after exposure to sunlight, it is necessary to characterise and monitor the cell by measuring it often with one instrument. Managing the use of the cell is the responsibility of the referee.

• The polarisation characteristics of the instrument(s) being evaluated must be determined. Ideally, all instruments used for zenith sky measurements should be either unpolarised or polarised (tracking the sun). Because sky and grating polarisation effects are intertwined, the “Ring” cross section is likely to be misfitted otherwise.

If the polarisation characteristics are not supplied by the PI of an instrument being evaluated then they must be measured during the intercomparison (see previous section Evaluation of Instrument Design and Data Analysis, bullet point 4, for the details). Differences between the measured polarised light spectra indicates that a misfit of the “Ring” cross section is likely.

If white light measurements are not available or cannot be made during the intercomparison, then sky measurements can be substituted provided the measurements are made during cloudy conditions. To ensure maximum depolarisation use dense clouds and a high sun, and interlace the measurements at opposite polarisations so variations in cloud density can be detected and the effects of them minimised.

• Analysis should provide two sets of results: (1) analysis using a daily selected midday reference spectrum, to be submitted to the referee normally within 1 day of the measurements; (2) analysis using a single midday reference spectrum for the whole intercomparison data set to be submitted to the referee within 1 month of the end of the intercomparison. Final “polished” results will also be submitted then. The choice of the daily reference spectrum and the single campaign reference spectrum will be made by the referee in consultation with the instrument groups.

• Blindness rules are important for formal NDACC intercomparisons. The goal is to provide NDACC data users with evaluations that are clearly a true picture of an instruments performance. Preliminary data submitted by participants can be displayed by the referee during the campaign providing it is in a form that does not enable participants to identify individual instruments. While total “blindness” would achieve this, it limits the opportunity for groups to learn and was therefore rejected at the 1996 OHP intercomparison. Displays of data that don’t identify groups, but enable participants to see the general form of the others measurements, were found to not compromise the integrity of the OHP intercomparison and so this practice is now permitted within the NDACC UV/Vis intercomparison specification. (It was not at the 1992 Lauder intercomparison.)

Individual results must not be exchanged between any participants being evaluated until final results are submitted by all instrument groups. This includes the group(s) with the reference instrument(s) .

• Auxiliary data should include cloud information so that differences resulting from the handling of instrument polarisation effects, “Ring Effect”, aerosol scattering, and absorption by O₄ and H₂O, can be identified. The cloud data should include cloud type and altitude, and the frequency of observations should match the rate at which cloud change is occurring. In addition, if significant tropospheric NO₂ contamination is expected, an in-situ measurement of it should be made. Note the above requirement for choosing a clean site.

The Instrument Group or its designated representative(s) will examine the results of the Intercomparison and make a recommendation to the NDACC Steering Committee. While a number of factors may enter the considerations, the principal criteria for the two types of certification currently used are as follows.

Because no absolute calibration is possible, accuracy is determined by quantifying the accuracy of each instrument being considered for certification, relative to the designated reference instrument(s) and, as appropriate, other instruments that are assessed as producing high quality results.

Spectral measurements made during the intercomparison period will be analysed by all participants using agreed criteria (wavelength interval, cross sections, etc.) to obtain their intercomparison results. A reliable method to determine which instruments meet this certification type is a regression analysis (Roscoe et al., 1998). This intercompares, using linear regression, all combinations of the twilight sets of measurements. Each set must include the results for a range of measurement SZAs, typically 70 to 93 degrees. Matrices of residual error, slope and intercept are generated in order to identify the instruments that agree most closely. The results from these instruments can then be used as the reference results for comparing the results of the other participating instruments against. In 1996 at the OHP intercomparison, the NO₂ results from three instruments, (one was the designated reference instrument) agreed to within 1.00 (+0.00 -0.02) in slope and 0.00 (+0.02 -0.07) cm-2 in intercept in one analysis (same wavelength interval and NO₂ cross sections), with similar or better ozone results. This close agreement is the basis for choosing the following figures as reasonably achievable at this time:

Groups with intercomparison results that meet these accuracy criteria together with the general acceptance criteria (below), and who are able to meet the quality control and long term commitment requirements, should be certified for type (1), global studies and trend measurements. These specifications are not to taken as a rigid criterion for certification, but rather as a goal which can now be meet. Some groups may have instruments that produce results that are close to these figures over a more limited SZA range and these should be considered on their merits. Limited certification can be used to recognise the potential of such instruments and the groups given assistance by the fully certified groups to improve their measurement accuracy.

The conversion of measured slant columns to vertical columns, using existing air mass factor (AMF) calculations, results in AMF errors that are greater than the measurement errors being sought in the above specifications, suggesting that these specifications are too tight. However, data of higher quality must be sought because, in the long term it can reasonably be expected that AMF calculations (including those at high solar zenith angles) will improve. Ancillary measurements, such as accurate meteorological parameters, aerosol profiles and cloud data, which are now available at many NDACC sites, will play an important role in these improvements. Other reasons for seeking the highest possible accuracy are the detection of trends in slant column amounts and the need for high accuracy measurements when making NO₂ height retrievals, particularly at high SZAs, (Preston et al., in preparation).

The following provides guidelines for the assessment. While some criteria must be met others can be applied as required to meet the overall goals of the assessment.

• The general instrument tests to measure resolution (slit function), polarisation characteristics, and stray light levels must be acceptable. See the discussion in the following section Instrument Tests for further information.

• Good result self consistency: this can be assessed by examining the “smoothness” of the twilight data series, especially when varying cloud conditions occur (as long as the troposphere is clean).

• At campaigns where the NO₂ or ozone amounts measured at each SZA don’t vary noticeably over several days (as occurred at Lauder in 1992 - see Hofmann, 1995), overlaying the results plotted against SZA can quantify an instrument’s uncertainty over a wide range of values .

• Low midday result variations during variable cloud periods and between midday values on clear and cloudy days. At clean sites these are naturally small, < 1x10¹⁵ cm^-2 for NO₂ and < 1x10¹⁸ cm^-2 for ozone slant columns, so large variations can indicate instrument or analysis problems.

• Acceptable signal to noise at high (near 95 degrees for spectra in the visible) and small (near 70 degrees) SZAs. This can be estimated by examining residual spectra or the “smoothness” of the result series (above). The reference instrument(s) errors at these solar zenith angles can be used as guide for acceptance. An important reason for considering high/low solar zenith angle results is that at midlatitude intercomparison sites, where there is normally plenty of NO₂ , they provide a good estimate of performance at high latitudes where NO₂ is lesser.

• Good consistency between the results obtained using the daily reference spectrum and the results obtained using the campaign single midday reference spectrum. This helps identify problems caused by long period (10 day) drifts in the instrument function or spectral wavelength repeatability. The reference instrument errors can be used as a guide for acceptance.

• Low (< 2x10¹⁵ cm-2 NO₂ , < 1x10¹⁸ cm^-2 ozone, slant columns) systematic errors. These are difficult to quantify, but examination of residual spectra, and comparisons of results obtained by using different wavelength subsets in the spectra, can be carried out.

Other methods to analyse intercomparison results can also be used to determine certification. One method that is being developed is to use measurements taken over the more limited range of SZAs, 85 to 91 degrees. These are converted to one vertical column result for each twilight measurement set by averaging the 85 to 91 degrees measurements after scaling each one by the air mass factor for the SZA of the measurement. Most instruments produce their most accurate measurements in this SZA range. Such an intercomparison method is a variation on the “fractional difference” technique used in the 1996 OHP intercomparison (Roscoe et al., 1998). It is expected that this technique is more suitable for certifying type (2) instruments. Using airmass factor weighted values also allows results obtained using different wavelength intervals to be compared (subject to the errors arising from the airmass calculations).

Ratios of the derived (airmass factor weighted) vertical columns are used to measure the agreement between the reference and compared instruments. For a 10 day intercomparison, 20 twilight periods, and therefore 20 such ratios, provide sufficient data from which a mean ratio and a fractional standard deviation of this mean ratio can be estimated. These statistics are used to assess the agreement of both NO₂ and ozone measurements.

Because both AM and PM results are available from the NO₂ measurements intercomparison, an offset value can be estimated by using a linear regression on the campaigns (>= 10 days) results. Because AM mid-latitude NO2 values are typically 0.6 times the PM values, the error on such an estimate is expected to be acceptable in determining the offset tabled below for NO₂.

A suggested accuracy for certification is:

Groups with an instrument that meets these accuracy criteria in a formal intercomparison and who are able to participate in process study campaigns with reliable instruments, should be certified for type (2), process studies and satellite validation measurements.

Most of the General Acceptance Criteria for type (1) certification also apply to type (2) certification with some relaxation of limits. The working group will define these as required.

The slit function full width half max. (FWHM) should be appropriate for the specie being measured. This is typically <1.5 nm for NO₂.

Using a Schott glass filter that cuts off light below a wavelength that is just above the highest wavelength used for measurements, gives some estimate of stray light levels. The limitation of using this technique is that the filter cuts a fraction (between a factor of 2 and 3 from model calculations for a GG475 filter) of the light that would otherwise contribute to the stray light signal. When the measured stray light is small (<0.2% of the light measured without the filter) this error is not important. However, with some array detector spectrographs the stray light levels are > 0.5% so this error is a problem. Another way to estimate stray light levels is to rotate the grating to below the atmospheric cutoff wavelength (∼300 nm) and measure the signal. This is often not an option however.

Stray light produces errors that are approximately proportional to the fraction of stray light in a measured spectrum. For a 1% stray light level, a 1% error in NO₂ and O₃ measurements occurs. The ratio technique does not cancel this error.

Another spectral artifact that can affect UV/Vis measurements is due to re-entrant light which can produce unwelcome structures in measured spectra. Generally this only occurs when Zerner-Turner spectrometers with grating rulings < 600 groves/mm are used in the UV region. The occurrence of re-entrant light can be predicted in the spectrometer design.

Polarisation

Two assessments of the polarisation tests are required to determine if an instrument is capable of making high quality zenith sky measurements.

The first is the variation with wavelength of Tp/Ts, where Tp is the relative transmission with the polariser parallel to the entrance slit and Ts the relative transmission with the polariser normal to the slit. Ring effect, which is primarily due to rotational Raman scattering, will vary with wavelength when clear sky twilight measurements are made with a instrument that does not have Tp = constant x Ts. Misfit of the Ring cross section can result, with consequent errors in the measured absorbers due to correlation of the fitted cross sections with the residual of the Ring fit. A variation of 20% across the 435 - 450 nm wavelength region has been found experimentally to be acceptable (< ± 0.1 x 10¹⁶ molec. cm^-2 offset) in tests at NIWA, Lauder. All the campaign instruments met this criterion.

The second assessment required is to estimate the effect that differences between Tp and Ts have on NO₂ measurement errors (%) due to the rotation Raman light in the spectra, i.e., Ring effect. A good reference on this question is Fish and Jones, 1995, who estimate that approximately 7% of the light in a twilight spectrum is due to rotational Raman scattering. Because the spectral shape of NO₂ is largely lost in the Raman scattered light, due to the smoothing effect of the stokes and antistokes lines, the error in NO₂ measurements due to this light can be estimated by treating it as a simple offset added to the Rayleigh + Mie scattered light. A 7% Ring amount will therefore produce an NO₂ error of approximately 7%. In actual measurements, the Ring amount depends on the polarisation characteristics of the instrument and the presence of cloud, which can partially depolarise the normally polarised clear sky Rayleigh scattered light. In measurements at NIWA Lauder, the amount ranges between 2 and 7% for both polarised and unpolarised instruments over the twilight period. This can be approximately corrected for in the data analysis. The question of how a Tp/Ts ratio that is different to 1.0 affects the Ring amount can be estimated for the worse case, that is, when the instrument is oriented so that the highest transmission axis is aligned parallel to the sun zenith plane and when the twilight sky is very clean and clear. This is the orientation that produces a minimum signal. For a Tp/Ts = 2 or 0.5, the Ring amount would increase to approximately 9%, while for a Tp/Ts = 4 or 0.25, it would increase to nearly 12%. For Ts or Tp = 0, it increases to over 20%, a figure that is clearly too high for good NO₂ measurements. A criterion of 0.5 < Tp/Ts < 2.0 has been chosen as a reasonable value for NDACC instruments.

A further source of error that arises when Tp is very different to Ts, is the change in airmass that results. One test by NIWA using a polarised system showed a 5% difference in ozone airmass between measurements taken using opposite polarisations.

• The group will be responsible for maintaining data quality: this should include routine procedures to check on instrument performance (e.g. NO₂ cell checks, stray light monitoring, instrument resolution tests, error figure monitoring and residual spectra checks).

• The group must maintain suitable instrument operation and maintenance records. Repairs and changes to equipment must be carefully logged and calibrations made afterwards to identify any changes in accuracy.

• Where available, groups should use data from other instruments at the measurement site to compare with their UV/Vis measurements. For example: Dobson, Sonde and Lidar data for ozone total column comparisons. The use of more than one UV/Vis instrument and the comparison of results also offers a higher level of confidence in the data.

• The group must maintain a routine data archiving procedure. The maximum time between measurement and data submission should not exceed 12 months. This is a vital commitment.

• (Primary sites only) When equipment problems compromise data continuity or quality, the group should promptly discuss the situation with the NDACC UV/Vis Working Group. This requirement is intended to encourage the open exchange of information from which all groups can learn. Also, when instrument failure seems likely to result in an extended loss of data, other groups may be able to help (e.g. with the temporary deployment of an available instrument to fill the gap).

• The group must be willing to participate in blind Intercomparisons. However, because of the enormous effort and high cost of these, the timing of them should be determined primarily by need, with a maximum gap between intercomparisons of 5 years (tentative figure at this time). For example, when improved or new measurement or analysis techniques are proposed, the Instrument Group, after favorable evaluation of them, could choose to conduct a suitable intercomparison inside the 5 year window. Analysis intercomparisons are easier to hold more often because they can be conducted by correspondence.

A possibly more cost effective, but not necessarily preferred (see following), way of maintaining standards within the Primary and Complementary Instrument community, is the occasional use of a traveling intercalibration instrument. This could be circulated on (say) a 3 year cycle initially, and where no significant change is detected after one cycle, the cycle could be raised to (say) 5 or more years. In addition, the use of a calibration cell (following section) provides a useful measure of instrument stability at the 10% level. It does not however necessarily detect changing vulnerability to some types of instrument drift (e.g. stray light and detector changes), and therefore the use of a traveling instrument, and blind intercomparisons when appropriate, needs to be maintained.

Intercomparisons offer advantages that the use of a traveling standard does not -

- they result in most of the community working together for extended periods of time during which workshops and discussions between groups results in a considerable sharing of knowledge and experience. Future directions and new techniques are also popular topics.

- they help break down cultural differences between groups from different countries enabling better cooperation and understanding within the global UV/Vis community.

- they provide a forum to identify and sometimes solve measurement and analysis issues.

- they enable new groups seeking to learn about more UV/Vis techniques to participate informally.

The consensus opinion of those who attended the 1996 OHP instrument intercomparison in France was that the interaction of the community resulted in a sharing of knowledge that was valuable to all groups. The question of how often these should be held has not been agreed by the community, but the figure of 5 yearly is judged as reasonable at present. Obviously some combination of formal intercomparisons and traveling instruments can be used to maintain NDACC UV/Vis standards and test new instruments for NDACC certification (e.g., the use of the NIWA traveling instrument at the Zvenigorod, Russia, intercomparison in September 1997). The final decisions will be made by the NDACC UV/Vis working group and be based on current circumstances.

• An NDACC NO₂ calibration cell will be circulated to all NDACC instrument groups every year. The group should follow the instructions accompanying the cell and promptly return it to the person or organisation designated by the Instrument Group to manage the cell calibration program. It is expected that the cell will be checked on one of the certified instruments between each field calibration in order to track any cell degradation. The repeatability of cell calibrations can depend on cell history, hence the need to maintain regular tie point checks using one instrument. The overall repeatability however is expected to be better than 10%.

• Instrument and analysis improvements that enhance scientific output or data quality are encouraged. The group should ensure however that data continuity and quality is maintained. Where possible an improved instrument should be operated in parallel with the existing instrument for a period of at least 6 months, and the data carefully compared. When an instrument or analysis technique improvement results in a change in the measurement results, this must be fully reported and recorded in the archive.

• Approximately every 2-3 years, the NDACC UV/Vis Instrument group and community should hold a workshop to discuss ways of improving measurement and analysis quality, and solving the remaining problems (e.g. scattering models, polarisation and Ring effect, standard cross sections, signal to noise, etc.). These workshops have been very effective in the past (Boulder 91, 92; OHP 96, Paris 98) and must be continued. Groups experience staff changes occasionally and the training and the free exchange of experience and knowledge within the UV/Vis community that workshops provide is an excellent way of providing training for new people. Workshops also foster continuing collaboration between the many groups that will help to ensure the quality of network UV/Vis measurements into the future.

NDACC UV/Visible Data Analysis Intercomparison: Workshop Report compiled by Susan Solomon from material presented at the Data Analysis Workshop held at the NOAA Aeronomy Laboratory, Boulder, Colorado, 23-25 July, 1991

International Intercomparison of Stratospheric NO₂ UV/Vis Measuring Instruments: Referee’s Report prepared for the Instrument Intercomparison Workshop held at Boulder, Colorado, 13-15 October, 1992, by David Hofmann (formal referee).

UV/Vis Intercomparison II; Retrievals of NO2, O3, OClO from Synthetic Spectra: Report prepared and presented by Aaron Goldman (Data Analysis Intercomparison Referee) at the Instrument Intercomparison Workshop held at Boulder, Colorado, 13-15 October, 1992.

Hofmann, D., P. Bonasoni, M. De Maziere, F. Evangelisti, G. Giovanelli, A. Goldman, F. Goutail, J. Harder, R. Jakoubek, P. Johnston, J. Kerr, W. Matthews, T. McElroy, R. McKenzie, G. Mount, U. Platt, JP. Pommereau, A. Sarkissian, P. Simon, S. Solomon, J. Stutz, A. Thomas, M. Van Roozendael, and E. Wu, Intercomparison of UV/visible spectrometers for measurements of stratospheric NO₂ for the network for the detection of stratospheric change, J. Geophys. Res., 100, 16,765, 1995.

Roscoe, H.K.; Johnston, P.V.; Van Roozendael, M.; Richter, A.; Roscoe, J.; Preston, K.E.; Lambert, J.-C.; Hermans, C.; DeCuyper, W.; Dzienus, S.; Winterrath, T.; Burrows, J.; Sarkissian, A.; Goutail, F.; Pommereau, J.-P.; D'Almeida, E.; Hottier, J.; Coureul, C.; Didier, R.; Pundt, I.; Barlett, L.M.; McElroy, C.T.; Kerr, J.E.; Elokhov, A.; Giovanelli, G.; Ravegnani, F.; Premuda, M.; Kostadinov, I.; Erle, F.; Wagner, T.; Pfeilsticker, K.; Kenntner, M.; Marquard, L.C.; Gil, M.; Puentedura, O.; Arlander, W.; Kastad Hoiskar, B.A.; Tellefsen, C.W.; Heese, B.; Jones, R.L.; Aliwell, S.R.; Freshwater, R.A. (Accepted, October 1998). Slant column measurements of O₃ and NO₂ during the NDACC intercomparison of zenith-sky UV-visible spectrometers in June 1996. Journal of Atmospheric Chemistry, 32: 281 - 314, 1999.

Certification Assessment for NDACC UV/Vis Instrument Intercomparison, Observatoire Haute Provence, France 12 - 21 June, 1996: report compiled by Paul Johnston from material in Roscoe et al., and additional regression analyses not otherwise published. Presented to NDACC Steering Committee, December 1998.