between the approach velocities determined for the primary photo-boundary and the outer smooth boundaryindicates a relatively even surge outline in the néignborhood of the station; converseiy, large discrepancies suggest lobes or irregularities. The approach velocity for the airborne radioactive material is determined from the gamma dose rate record by the rate of rise to the first major peak, a process fully described in Section 3.3.4. The value obtained depends on the surge model used (Section 1.3.2). Only the range of possible rate-of-rise velocities is tabulated in Table 3.11. The rough general agreement between the several velocity determinations sug- gests that the visible surge and the radioactive aerosol are moved by the same mechanical forces but does not necessarily imply that they are the same body of airborne material. For both shots, a somewhat better comparison results for rate-of-rise velocities determined with surge models os greater thickness. Since all these derived velocities are affected by a large numberof arbi- trary assumptions necessary for their determination, this distinction may indeed be fictitious. A more consistent difference between the Wahoo and Umbrella records becomes apparent when the shortest distance to the outer primary photo-boundary P, at the time of the major dose rate peak is conSidered. These distances obtained from the boundary plots are given in Table 3.11. In accordance with the sign conventions previously described, a negative value indicates that the outer primary boundary has not yet reached the station; a zero value indicates _ that its arrival is coincident with the time of the first major peak; and a positive value indicates that it has already passed the station. The major dose rate peak may be assumed to correspond to a position of optimum detector geometry relative to the airborne radioactive material or to a region of maximum radioactive concentration within the visible base surge (such regions of in- creased radioactivity were previously suggested in Section 3.2 by variations in instrument response). Regardless of its actual cause, this point is referred to as the source center. Although no particular significance is placed on the numerical values because of the stated limitations on the accuracy of ali photo-boundaries, the fact that the values for Wahoo are preponderantly positive suggest that the source center lags approximately 1,000 feet behind the outer visible boundary somewhere near the inner edge of the primary base surge. This suggestion is also supported by the observation in Section 3.2 that a source center approximately 1,000 feet behind the photo-boundary is required in correcting the differences between ASEL- GITR and std-GITR responses at Wahoo Stations CL 4.6 and CR 4.1 (Figures 3.7 and 3.9). For Umbrella the preponderantly negative values suggest a source center closely associated with the outer primary photo-boundary or possibly somewhat in advance of that boundary. At the distant stations, such differences might be ascribed to errors in assumed surface winds, but _ at the closer stations, which constitute approximately 80 percent of all records, the position of the surge boundary is a matter of photographic record. Use of the NOL radii places the source center even farther behind the visible surge boundary for Wahoo, whereas for Umbrella their use moves the source center to an apparently more reasonable 500 feet behind the outer primary photo-boundary. The distinction that the source center for Wahoo lags far behind that for Umbrella, however, remains essentially unchanged. AS already indicated, existence of an invisible radioactive material in advance of the primary photo-boundary might be the result of an overexposure in the photographic printing process. In- deed, the anomalous behavior shown by Station DRR 12.8 during Wahoo is probably due to such photographic disappearance. If the section of the primary surge boundary that finally intersects this station is assumed to expand from its 2-minute position in a manner exactly Similar to the remainder of the surge, the source center lies behind the primary photo-boundary at a position similar to that observed at other Wahoo stations. Nevertheless, an exactly similar disappearance could result from evaporation at the outer surge boundaries. Under the appropriate ambient conditions, the base surge droplets could evaporate leaving a more or Jess invisible radioactive aerosol. Thus, these differences in source center position relative to visible boundaries suggest that, aithough the airborne radioactive material is often closely associated with the visible material, sach association cannot be tacitly assumed. In the preceding general discussion the Wahoo base surge is described as a double toroid having a number of diffuse clouds at its center whereas the Umbrella base surge is described 130