FIGURE 12.36 Deviation of (a) particle surface area at 16-20 km and (b) monthly mean ozone at 15.3-19.8 km measured in Germany from 1991 to 1994 (adapted from Ansmann et al., 1996).

Arctic stratosphere due to chemistry that is qualitatively similar to that in the Antarctic, an analogous "ozone hole" is not formed. The major reason for this difference is the different meteorology and dynamics (Schoeberl et al., 1992; Manney and Zurek, 1993; Man-ney et al., 1996).

First, the temperatures found in the Arctic stratosphere are warmer by about f 0 K compared to those in the Antarctic. The Arctic stratospheric temperatures do not drop below 195 K as frequently (e.g., see Paw-son et al., 1995), so that PSCs, particularly Type II, which consists primarily of ice and requires temperatures of ~ 188 K, do not form as readily nor persist for the lengths of time that they do in the Antarctic polar vortex. In addition, as discussed in Section C.5b, the uptake of HC1, a key species in the heterogeneous chemistry, into liquid solutions found in the stratosphere is highly temperature dependent, with the Henry's law constant increasing as the temperature decreases (Table 12.4). The heterogeneous reaction probabilities also depend on temperature. However, it appears that mountain-induced gravity waves cause local reductions of up to 10-15 K in the temperature of the stratosphere, which can increase PSC formation and hence increased halogen activation in the Arctic stratosphere (Carslaw et al., 1998a).

Second, the northern polar vortex is much less stable and hence less isolated from mixing with external air masses compared to the Antarctic case; events in January and February in which there was substantial mixing of air from midlatitudes into the vortex have been reported (e.g., see Browell et al., 1993; Plumb et al., 1994). This makes it particularly important to make both measurements and model predictions with sufficient resolution (Edouard et al., 1996). In addition, the Arctic polar vortex tends to break up earlier than the Southern Hemisphere polar vortex; since ozone destruction is determined to a large degree by the extent of exposure to sunlight, the earlier breakup and mixing with air external to the vortex cuts the ozone loss short.

Finally, the dynamics are quite different, with ozone concentrations in the Arctic stratosphere usually increasing in December and into the early spring due to the normal large-scale transport of air containing higher ozone concentrations from the tropics at higher altitudes, followed by downward transport (see Section A.l). Any decreases due to the chemical destruction processes already described are superimposed on these normal increases. Hence the chemically induced losses of total column ozone can be at least in part masked by natural variations (e.g., see Proffitt et al., 1990, 1993; Waters et al., 1993; Manney et al., 1994a, 1994b; Hen-riksen et al., 1994; Santee et al., 1995; Solomon et al., 1996; and Zhao et al., 1996).

For example, while the vortex-averaged 03 concentration at one altitude in the Arctic in the spring of 1994 was measured to decrease by ~10%, the net chemical loss was estimated at ~20% but this was partially compensated by an increase due to transport of air containing higher ozone concentrations from higher altitudes (Manney et al., 1995). Similar amounts of chemical ozone loss in the Arctic polar vortex have been calculated based on measurements of CIO, BrO, and 03 (e.g., Brune et al., 1991; Salawitch et al., 1993).

Despite these differences, it is clear that ozone destruction due to CFCs and halons also occurs in the the lower stratosphere in the Arctic. For example, total 03 losses of the order of 50-f00 DU have been deduced in the Arctic polar vortex during the 1991-1995 winters (e.g., see Larsen et al., 1994; and Miiller et al., 1996). Again, the increasing importance of heterogeneous chemistry at lower temperatures is evident. For example, Fig. 12.37 shows the measured concentrations of CIO as a function of the minimum temperatures experienced by the air masses obtained using back trajectories; also shown is the deficit in HC1, defined as the difference between the measured HC1 concentrations and those expected in these air masses based on the concentrations of N20, which can be used as a tracer (Toohey et al., 1993; Webster et al., 1993a). Clearly, at the lower temperatures, below 196 + 4 K, where heterogenous chemistry is expected to convert HC1 and C10N02 to active forms of chlorine, there is a greater HC1 deficit than otherwise expected and much higher levels of CIO. Furthermore, these studies showed that the amounts of active chlorine in the form of CIO and its dimer, C1202, were equivalent to twice the observed HC1 deficit, consistent with the heterogeneous reaction of HC1 with C10N02 (Webster et al., 1993a).

Similarly, Fig. 12.38 shows some typical measurements of CIO and the HC1 deficit at latitudes both outside and inside the Arctic polar vortex (Webster et al., 1993b). As expected based on the known chemistry, there is a significant HC1 deficit inside the vortex, accompanied by increased CIO concentrations (the more gradual increase in CIO at the edge of the vortex is attributed to air that has undergone PSC chemistry in the past but is now partially recovered; Webster et al., 1993b). Indeed, almost complete conversion to active forms of chlorine has been measured in the winter Arctic vortex. Figure 12.39, for example, shows one estimate of the partitioning of total inorganic chlorine, C\y (Cly = CI + CIO + 2C12Oz + HC1 + C10N02 + HOC1; OCIO, BrCl, and 2C12 are also included in this if present), both outside and inside the Arctic polar vortex in January and February 1989 based on measurements of NO, NO , CIO, N20, and total organic

0 0

Post a comment