RAPPs (continued) and they made another estimate. However, they stuck to the expanding cylinder with the homogeneous distribution & they did not consider a vertical change of the cylinder, Cur latest model, “ark iI ‘‘odel, allows for a different type of expansion of the cloud. <“e have considered a clohd to expand according to a fickian law of diffusion and we have chosed a diffusion constant that is as big as we can get it without tearing the cloud apart. ‘¢ want to try keep this cloud intact and expanding so we can't allow f too big a diffusion coefficient, this would tear the clou Also, because we don't want to tear the cloud apart, we haven't considered the shear. But this now gives you a picture of what our cloud looks like at the end of six hours. te start out with an elipsoid of hozogeneous con- centration 3 kilometers thick and S kiloneters in diase and we assune that the center wont to a height of10 moters. As the diffusion spread this cloud, the peak co tration started to fall and these Lines here represent sh where the concentration has fallen to 1% of the original centration, 0.1% to 0.01% of the original concentration. You must try to imacine that these shells initially, wien the cloud is homogeneous, are all together, we cre upat 1 Then as it starts to spread these low concentrations spre. - out rapidly, then dilution begins to take offect so that Department of Evcvw! \ Historion’: 77 yaomrae e :