155 1008 a yor Fig, 122.--9 Maysurface weather map, 1200 CST the lake breeze had already passed the three lake front stations. All stations still had low SOs readings. The 1400 and 1800 maps showthe penetration of the lake breeze many kilometers inland. The first high SO. values were observed at station 6 at 1800, An hour later (Figure 127) the lake breeze front had trated inland as far as Argonne by 1900. High pollution levels appeared first at 1800 well behind the lake breeze front in easterly winds at TAM-6§. The pollution moved to the northwest, north, and then east; it was detected next at station 8, then 7, and then almost of the wind to southerly and a jump in SQ» at station simultaneously at 2,3, and 4. The fact that the pollution first appeared at station 6 in easterly winds suggests a source between TAM-6 and the lake shore. We traced the trajectorv of TAM-2 pollution backwards from 2200 and arrived at the samearea. Theoretically an onshore wind from a cool lake breeze developed between 1000 and 1100 and pene- mation of an unstable boundary layer that deepens with distance from the shore. When this unstable layer intersects plumes aloft, contaminants will mix downwardto the surface in high concentrations. Circumstances in this case suggest that SOQ. from the South Chicago-Northwest Indiana industrial complex may have been brought to the surface in this way passed Argonne. Winds behind the front were southeasterly, and station 7 also had an increase in SOs. The reading at station 6 had dropped from 0.38 to 0.27 ppm. The next map (2000) shows continued veering 8. Two hours later (2200), winds were southwest and SOs concentrations were 0.38 ppm or more atall stations except 5 and 6. At 0200 on the 10th, winds were light and variable and SOs concentrations were lower at all stations. The highest levels appeared in a band along the lake shore. This case may be summarized as follows: A lake should behave as shown in Figure 128, with the for-