7.2.2 Atmospheric CH₄ observation

   Atmospheric CH₄ is observed using an NDIR, which utilizes the infrared absorption characteristics of CH₄, at Ryori and Minamitorishima; and by a gaschromatograph (GC) equipped with a flame ionization detector (FID) at Yonagunijima from January 2008. The system models used at the stations are listed in Tables 7.1.1.2, 7.1.1.3, and 7.1.1.4, and they are depicted in Figure 7.2.2.1. The stability of the NDIR observation, expressed as a 16-minute mean standard deviation, does not exceed 5 ppb at Ryori and 4 ppb at Minamitorishima; and the stability of the observation system at Yonagunijima, expressed as a standard deviation of repeated 6 times of measurements, does not exceed 2 ppb. The repeatability, which is expressed by the difference between the 2 runs of the above stability test, does not exceed ±5 ppb, ±1 ppb, and ±2 ppb at Ryori, Minamitorishima, and Yonagunijima, respectively.

   Air samples are introduced from the same intake used for CO₂ observations, and then is diverted. In the analysis by NDIR, hydrocarbons other than CH₄ are removed by combustion using a catalyst so as not to interfere with the subsequent infrared absorption of CH₄. Two cylinders (4 cylinders at Yonagunijima) of working standard gases of natural air base in different concentrations are used in observations. The samples are supplied to the analyzer at a flow rate of 3.5 L per minute (250 mL per minute at Yonagunijima). The concentrations of these working standard gases are determined and their drifts are evaluated by calibrating against JMA's reference gases using a GC-based CH₄ calibration system before and after use.

   The CH₄ concentrations of each of the 2 cylinders (4 cylinders at Yonagunijima) of CH₄ working standard gases are measured for 7 minutes each (every 10 minutes at Yonagunijima), and then the CH₄ concentration in an air sample are measured for a prescribed duration. This observation cycle is repeated every 12 hours (every 8 hours at Yonagunijima). For NDIR, the concentrations of the working standard gases and air samples are determined by calculating the average of measured analyzer output voltages every second over a period of 30 seconds. For GC, the concentrations of the working standard gases and air samples are measured by analyzing the peaks from the analyzer every 10 minutes to obtain the outputs from the system at the station, which are later converted into concentrations at the JMA headquarters.

   To convert into concentrations, a linear calibration curve (or quadratic curve for Yonagunijima) expressing the relationship between the concentration and output for each observation cycle is determined from the known concentration of the 2 cylinders (or 4 cylinders at Yonagunijima) of working standard gases and outputs. The output of the air sample is converted into the concentration using these calibration curves. To minimize errors caused by variations in the response time of the analyzer, 2 concentrations are calculated for every measurement of air sample using 2 different calibration curves, one immediately before and one after the sample measurement. These 2 concentrations are then time-interpolated to obtain 1 concentration every 30 seconds (or every 10 minutes for Yonagunijima). If a drift of the concentration for a working standard gas is detected by JMA's calibration system as a result of the comparison before and after the use, these observations are corrected by interpolation.

   The selection procedure for BG data to be used to calculate hourly, daily, and monthly mean values is the same as that for CO₂, with the threshold values shown in Table 7.2.2.1.