CONFEIDENTFHATL limiting factor to carrying out a mass evacuation within a given period of time. The Washington area was used as a model. As described in Fig. 2, the numberof traffic lanes diminishes from 149 at 4 miles from the city’s center to 101 at 16 miles from the city’s center. The modelis designed to set an upper limit on evacuation feasibility, and assumes that immediately on warning of attack all vehicles will start moving radially outward at a rate of 1000 vehicles per lane per hour.* It is further assumed that all vehicles will be used to optimum advantage (the population of each ring distributed evenly throughout the numberof vehicles in the ring), and that each vehicle will contain five passengers. Table 5 shows how the population is distributed after 1, 2, 3, and 4 hr of evacuation. At the end of 4.5 hr the target is empty. Figure 10 shows the same data in terms of ring density after 1,2, and 3 hr of evacuation. For comparison this figure assumes that evacuees might be held in the 16- to 20-mile ring; it will be noted that under these conditions density is optimally uniform after about 3 hr of evacuation. TABLE 5 POPULATION DISTRIBUTION OF WASHINGTON FOR Various EvacuaTION TIMES FCDA damage zone A B C D Outside ; Zone radius, miles 0-4 4-8 8-12 12-16 16-20 Traffic lanes leavin 4 city 149 120 112 101 —_— Population, thous No 1-hr 2-hr 3-hr 4-hr movement evac evac evac evac 647 748 218 64 _— 0 996 239 100 342 0 577 8286 136 678 0 19 86409 236 1013 0 0 0 200 1477 The approximate times required to empty the six remaining targets under consideration (based on the numberof traffic lanes leaving the target area) are as follows: Boston, 8.0 hr; Dayton, 3.3 hr; Milwaukee, 6.7 hr; St. Louis, 6.0 hr; and San Francisco, 11+ hr. A comparison of the times required to evacuate totally the six urban targets assuming the probable warning times expected for them through 1959 (Table 4) reveals that only one, Dayton, can complete mass evacuation within the expected period of warning. Against attack with ballistic missiles, with perhaps 15-min warnings, such a tactic is completely impossible. It should be further emphasized that these figures represent the minimum times required. The figures given would have to be degraded by many factors, including the fol- lowing. (a) There is some loss in starting time, e.g., losses in making the warning public (probably 12 min), 11 or more min to emptylarge office buildings under normalwell-practiced circumstances (recent air-raid drills), and 20 min and upward to empty parking lots, depending on their size. The time required to reunite fathers, mothers, and school-age children for evacuation purposes would also degrade this figure by some large factor, and additional starting losses would be created as prospective evacuees loaded their automobiles with needed supplies. *Lane capacities as high as 2300 vehicles per lane per hour have been measured on somelimited-entryegress highways; other highways have capacities as low as 600 vehicles per lane per hour. A capacity of 1000 was chosen as an all-road, all-season, day-night average. TAn sri study" of three California floods indicated that 70 percent of the population had left their homes within 30 min of the warning. After the first 30 min, the remaining population responded slowly; at 4 hr from the warning only an additional 15 percent had evacuated. 29 ORO-R-17 (App B)