Operation of Brewer Instruments at Mauna Loa Observatory
J.B. Kerr
Atmospheric Environment Service, Environment Canada, Downsview, Ontario Canada M3H 5T4
Brewer instrument No. 009 was installed at the Mauna Loa Observatory (MLO) on March 24, 1997, by the Atmospheric Environment Service (AES) of Environment Canada (EC). The instrument measures total ozone, the vertical profile of ozone using the Umkehr method, and UV-B radiation. An all-sky camera that records digital images of the sky every 10 minutes is mounted in the top of the instrument. These images provide information about sky conditions to assist in the data analysis, particularly the UV-B radiation.
The instrument was initially located on a concrete pad to the southeast of the present site of the Network for Detection of Stratospheric Change (NDSC) building. Supporting computer equipment for the instrument was inside the dome to the east of the NDSC building. Software operations for the instrument, such as programming and data retrieval, are carried out remotely by telephone and the Internet from AES in Toronto. On-site operations, such as hardware maintenance and routine checks, are carried out by the staff at MLO.
The instrument operated nearly flawlessly until August 15, 1997, when the observatory took a direct hit from lightning. A surge from the lightning damaged the control computer, probably through the telephone port, but left the Brewer instrument undamaged. Observations resumed on November 18, 1997, when the computer was replaced and the instrument was relocated to the deck of the NDSC building. Double monochromator Brewer instrument (No. 119) was then installed adjacent to No. 009 on the NDSC deck. Since then both instruments have been operating satisfactorily.
Ozone Measurements
Between 20 and 30 direct sun total ozone measurements are attempted each day depending on the time of year, and a total of 5045 attempts were made during the 188 days that Brewer instrument No. 009 operated in 1997. Of these attempts, 3905 (77.4%) were of good quality and the remaining ones were usually influenced by clouds. There were only 2 days of the 188 when no good quality direct sun measurements were obtained.
Extraterrestrial calibrations for the Brewer (and Dobson) reference instruments are regularly carried out at MLO. These calibrations require several days (about 1 week) of direct sun measurements made on clear days (or half days) with airmass values between 1 and 3. The extraterrestrial calibration is the extrapolation of the daily data sets to zero airmass [see Kerr et al., 1988]. The quality control on data from an instrument operating permanently at MLO is enhanced by the extraterrestrial calibration which is measured every day, so the stability of the instrument is readily monitored and the calibration can be adjusted if necessary. Figure 1 shows the record of daily mean total ozone values measured during 1997.
Fig. 1. Record of direct sun total ozone measurements made with Brewer instrument No. 009 at MLO during 1997.
UV-B Measurements
Spectral measurements of UV-B global irradiance are made throughout each day two or three times per hour [Kerr and McElroy, 1993]. A total of 4151 spectra were made with instrument No. 009 in 1997. The data have been processed with the standard World Ozone and Ultraviolet radiation Data Centre (WOUDC) processing and quality assurance procedures [Wardle et al., 1996]. The measurements are generally of high quality and about 5% of the scans have been flagged to show variability during the course of the scan, mostly due to changing cloud conditions.
Examples of analyzed UV data are shown in Figures 2 and 3 which summarize results for the entire year as functions of time of day and time of year. Figure 3 shows irradiance data in the UV-A at 324 nm, a wavelength that is negligibly absorbed by ozone. Here the irradiance values are given as the percentage of clear sky values derived from a statistical analysis of data from seven sites at lower altitude [Fioletov et al., 1997]. The red and purple colors indicate clear sky conditions and the banded darker colors are for days when clouds were present. Note that most mornings are clear and clouds occur more often later in the day. Another feature seen in Figure 2 is that the sky often becomes brighter (increasing from ~100% to ~110%) near 1100 LST. This may be the result of the buildup of convective clouds below the MLO altitude causing an increase in upward scattered radiation (similar to the enhancements caused by increased surface albedo reported in Wardle et al., 1997).
Fig. 2. Summary of UV global irradiance measurements made at 324 nm as functions of time of day and time of year for 1997. Intensity measurements are given as the percentage of clear sky values. Radiation at this wavelength is negligibly affected by ozone.
Figure 3 summarizes the UV Index values as functions of time of day and time of year. At noon during the summer when the sun is directly overhead, UV Index values are typically 15 or 16. These values are significantly higher than typical summertime values of about 10 seen at midlatitudes and are due to high sun angle, clear skies, high altitude, and relatively low ozone values.
Fig. 3. Summary of measured UV Index as functions of time of day and time of year for 1997. Maximum values at noon in summer are near 16. This value is significantly more than typical values seen at mid-latitudes in summer (about 10).
Sky Images
Figure 4 shows examples of images made by the all-sky camera mounted on Brewer instrument No. 009. The dark band from the center to the right of the images is a shade to prevent the suns bright image from saturating and degrading the camera sensor. Since the Brewer instrument (and camera) follow the sun in azimuth, the shade band is always aligned with the sun. The sun rises at the eastern horizon at the right of the image behind the band. Until solar noon the sun moves behind the band toward the center of the image. After noon the sun moves back toward the right of the image until is sets on the western horizon. Throughout the day features on the horizon rotate clockwise around the perimeter of image.
The top image in Figure 1 was taken at 1147 (local standard time (LST) on March 30, 1998, when the sky was cloudless. The tower to the west of the observatory is seen on the horizon in the upper left of the image. The top of Brewer No. 119 is seen at the lower left of the image. The bottom image in Figure 1 was taken at 1325 LST on March 20, 1998, showing convective up-slope clouds that developed in the early afternoon. At this tower is now in the upper right of the image. All images for 1997 have been archived on CD disks and are available from AES on request.
Fig. 4. Example of all sky images made with a video camera mounted on the Brewer instrument showing clear (top) and partly cloudy (bottom) situations. The Brewer instrument is normally aligned so that the shade band protects the camera from the sun and features on the horizon move clockwise around the perimeter of the image during the day. The record of sky conditions are useful for interpreting ozone and UV-B data.
References
Fioletov, V.E., J.B. Kerr, and D.I. Wardle, The relationship between total ozone and spectral UV irradiance from Brewer observations and its use for derivation of total ozone from UV measurements, Geophys. Res. Lett., 24, 2997-3000, 1997.
Kerr, J.B., I.A. Asbridge, and W.F.J. Evans, Intercomparison of total ozone measured by the Brewer and Dobson spectrophotometers at Toronto, J. Geophys. Res., 93, 11,129-11,140, 1988.
Kerr, J.B., and C.T. McElroy, Evidence for large upward trends of ultraviolet-B radiation linked to ozone depletion, Science, 262, 1032-1034, 1993.
Wardle, D.I., E.W. Hare, D.V. Barton, and C.T. McElroy, The World Ozone and Ultraviolet radiation Data Centre - content and submission, Proc. 18th Quad. Ozone Symp., LAquila, Italy, 1996.
Wardle, D.I., J.B. Kerr, C.T. McElroy, and D.R. Francis (Eds.), Ozone science: A Canadian perspective on the changing ozone layer, Rep. CARD 97-3, 19 pp, Environ. Canada Int., Downsview, Ontario, 1997.