Effect of diffusion and mixing on e

One factor that may cause the stratospheric fractionation constant to deviate from the measured laboratory values is the effect of eddy diffusion. Rahn et al. (1998) calculate

Fig. 14.8. Measurements of stratospheric S 15Nbulk indicate that fractionation increases from the lower to the upper stratosphere. When the data are plotted as in the figure, the slope is an estimate of the fractionation constant. The transition in the slope occurs near 24 km. (From Toyoda et al., 2001. Reproduced by permission of American Geophysical Union.)

Fig. 14.8. Measurements of stratospheric S 15Nbulk indicate that fractionation increases from the lower to the upper stratosphere. When the data are plotted as in the figure, the slope is an estimate of the fractionation constant. The transition in the slope occurs near 24 km. (From Toyoda et al., 2001. Reproduced by permission of American Geophysical Union.)

that in a diffusion-limited one-dimensional (1D) stratosphere, the apparent fractionation constant (estrat) is expected to be about half the true value (£lab). This result is promising since this is approximately the relation between fractionation constants measured in the lower stratosphere and in the laboratory. However, for diffusion and photolysis rates typical of stratospheric conditions, photolysis in the lower stratosphere is not diffusion-limited, but rather rate-limited, in which case estrat would be similar to £lab (Toyoda et al., 2004). In addition, it is likely that photolysis becomes diffusion-limited in the upper stratosphere, which would cause £strat to decrease with altitude. This is opposite to the trend observed. Changes in the diffusion constant may however produce small seasonal and interannual variations in £strat in the more sensitive mid-to-upper stratosphere (Toyoda et al., 2004).

It appears that mixing can explain some if not most of the departure of £strat from £lab and also the increase of £strat with altitude. We expect measurements of isotopic enrichments in the stratosphere to follow Rayleigh fractionation only if they track the chemical evolution of a single air mass. If multiple air masses are mixed, the combined enrichment will reflect contributions from multiple chemical histories. This tends to dilute the overall enrichment and lower the fractionation constant.

Dynamical mixing is common throughout the lower stratosphere. Although most tro-pospheric air enters the stratosphere through vertical advection in the tropics, there is considerable stratosphere-troposphere exchange at extra-tropical latitudes along isentropic surfaces and during tropopause folding events. Incursion of non-enriched tropospheric air dilutes the photolysed stratospheric air mass, thus reducing £strat. Although the dynamics of the stratospheric circulation is complex, observations indicate that N2O-depleted air is brought down from the upper stratosphere to the lower and middle stratospheres through the polar vortex at high latitudes (Park et al., 2004). The mixing of these air masses with the N2O-enriched air of the lower stratosphere creates a lighter isotopic composition and also reduces £strat. Since subsidence is great est at high latitudes, mixing effects produce a gradient in the latitudinal variation of £strat. Air in the subtropics undergoes less mixing than in the high latitudes, which causes £strat to decrease with latitude as mixing becomes more pronounced (Park et al., 2004).

The impact of mixing processes on £strat has been simulated by a number of atmospheric chemical-tracer models (McLinden et al., 2003; Morgan et al., 2004). These models successfully reproduce the observed increases in the N2O fractionation constants in the middle stratosphere. Both models find that the enrichment of isotopologues in the stratosphere can be described by two fractionation constants: one for the lower stratosphere stretching from the tropopause to ~25 km, and another for the stratosphere above 25 km. This boundary is consistent with the transition altitude measured by Toyoda et al. (2004).

Was this article helpful?

0 0

Post a comment