Lne

20:00 22:00

00:00 02:00 04:00 Time

FIGURE 11.47 Nighttime measurements of OH at Mauna Loa Observatory, Hawaii, in May 1992 made by the mass spectrometry derivatization technique (adapted from Tanner and Eisele, 1995).

20:00 22:00

00:00 02:00 04:00 Time

06:00

FIGURE 11.47 Nighttime measurements of OH at Mauna Loa Observatory, Hawaii, in May 1992 made by the mass spectrometry derivatization technique (adapted from Tanner and Eisele, 1995).

UV Light

UV Light

FIGURE 11.48 Schematic diagram of radiochemical OH measurement apparatus (adapted from Felton et al., 1990).

Figure 11.48 is a schematic diagram of this apparatus (Felton et al., 1990). Air is introduced into a sampling manifold consisting of a quartz tube where it is mixed with the 14CO. The air is collected downstream after a measured reaction time and analyzed for l4C02.

There are several assumptions inherent in this method (e.g., see Felton et al., 1990, 1992). For example, the concentration of l4C02 in ambient air must be negligible compared to that formed in the reaction and the OH concentration in air is assumed to be unperturbed either by the addition of l4CO or by the sampling system itself, e.g., by loss on the walls. While straightforward in principle, as discussed by Felton et al. (1990, 1992), it is experimentally challenging. For example, accurately measuring the small concentrations of 14C02 formed is difficult, imposing stringent requirements on the purity of the l4CO tracer and on the purification techniques used for the product l4C02.

Intercomparisons. A number of intercomparison studies have been carried out for the different OH measurement techniques (e.g., see Beck et al., 1987; Mount and Eisele, 1992; Eisele et al., 1994; Campbell et al., 1995; Brauers et al., 1996; Mount et al., 1997a, 1997b; and Hofzumahaus et al., 1998). Overall, given the extreme difficulty in sampling and measuring this highly reactive free radical at the sub-ppt concentrations found in air, the agreement is generally quite good.

Figure 11.49, for example, shows measurements of the diurnal variation of OH made using LIF and UV absorption, respectively, on two different days in a rural area in Germany (Hofzumahaus et al., 1998). The agreement is, in most cases, excellent. These data also illustrate a typical diurnal variation of OH, being below the detection limits of the instruments at night (5 X fO5 radicals cm-3 for LIF and 1.5 X 106 radicals cm-3 for

DOAS) and rising to a peak of ~107 radicals cm~3 at noon when photolysis of its precursors peaks. Similar diurnal behavior has been observed in remote areas such as the Mauna Loa Observatory (e.g., Eisele et al., 1996) and in more polluted areas as well (e.g., Felton et al., 1990; Hard et al., 1995; Mount et al., 1997b). Typical peak OH concentrations are usually in the range of ~(2-f0) X 106 radicals cm~3.

Figure 11.50 shows for this particular intercomparison study a plot of OH measured by DOAS against those obtained simultaneously by LIF. The correlation coefficient is r = 0.85. Disagreement was greatest when the wind was from a particular direction, which gave higher DOAS readings. The reason for this is not clear, but Hofzumahaus and co-workers propose that it may

Was this article helpful?

0 0

Post a comment