7.3.5 Calibration for surface ozone observation

   JMA was using a KI-method based calibration system prescribed by JIS B7957 until March 2002, because there was no concentration standard for surface ozone observation designated by WMO at that time. This method makes no comparison between the ozone monitor and ozone calibrator, but rather determines the absolute concentration through the calibration procedure itself. This method also required additional measurement work, such as preparation of a KI solution and accurate dilution of the solution, as well as special skills in analytical chemistry. WMO adopted the standard reference photometer, developed by the National Institute of Standards and Technology (NIST) and utilizing optical method with good stability and small error, as the standard for surface ozone observation in the GAW programme. Subsequently, NIST was designated as the Central Calibration Laboratory for maintaining the standard for GAW surface ozone observation in the GAW Strategic Plan (WMO, 2001).

   In March 2002, JMA upgraded its calibration system for surface ozone to comply with the WMO standards, and started calibration utilizing an optical method by comparing the operational ozone monitor with the standard reference. The new calibration system (Figure 7.3.5.1) consists of an ozone calibrator-based concentration unit, a zero air gas generation unit and a computer-based control unit.

   The new calibration system uses a MODEL 49C-PS ozone calibrator (Nippon Thermo) with a calibration precision of 1 ppb, linearity of not exceeding ±0.1% (full scale), and response speed of 20 seconds (up to 95%). The ozone calibrator is sent to NIST on a regular basis to compare with the WMO standards (Figure 7.3.5.2). The surface ozone calibration system maintains traceability to the WMO standard scale through this process.

   Since the standard scales differ between the new and old calibration systems, the calibration coefficients were adjusted to be consistent with each other and correct the observation data obtained at the three JMA observation stations to secure continuity of data (JMA, 2004).

   The ozone monitor to be used with the new calibration system is calibrated in the following manner. First, ozone-free gas is produced from the zero air gas generator, followd by standard gases of 150 ppb, 100 ppb and 50 ppb in this order from the ozone calibrator. The concentrations are measured for duration of 15 minutes after allowing 5 minutes for replacement. This calibration cycle is repeated 10 times for the above four concentrations. Concentration is measured every 10 seconds, mean concentrations and standard deviations for the calibrator and the ozone monitor to be calibrated are calculated for each of the prescribed concentrations, and then calibration coefficients are obtained for each cycle. After all the calibrations are completed, the 10 calibration coefficients are averaged to obtain the calibration coefficient for the ozone monitor.

   The ozone monitors are calibrated at the JMA headquarters and then sent to each station. After about 6 months of use for observation at the station, the ozone monitors being sent back to the JMA headquarters. The returned ozone monitors are calibrated again to check if there is any change in the calibration coefficient. After this, consumable parts of the ozone monitor such as mercury lamps are replaced, maintenance overhauls are carried out, and then the monitor is put on standby until the next observation. At the stations, two ozone monitors are operated in parallel for about a month for the confirmation of proper observation when ozone monitors are replaced.