ML20040B127
| ML20040B127 | |
| Person / Time | |
|---|---|
| Site: | Diablo Canyon  |
| Issue date: | 01/19/1982 |
| From: | Jerrica Johnson CALIFORNIA, UNIV. OF, LOS ANGELES, CA, JOINT INTERVENORS - DIABLO CANYON |
| To: | |
| Shared Package | |
| ML20040B117 | List: |
| References | |
| NUDOCS 8201250188 | |
| Download: ML20040B127 (30) | |
Text
-.
t y. : > -
1 TESTIMONY OF 2
DR. JAMES 11. JOIINSON, JR.
'82 JAN 17 P7:19
\\
3 ON BEIIALF OF JOINT INTERVENORS La p~
4 JANUARY 19, 1982 H M 'T 5
CONTENTION 1 0
7 My name is James !!. Johnson, Jr.
I am an Assistant 8
Professor of Geography at UCLA.
I received a B.S. degree in 9
geography from North Carolina Central University in 1975, a 10 M.S.
in 1977 from the University of Wisconsin at Madison, and 11 a Ph.D. in 1980 from Michigan State University.
I specialize 12 in urban-social geography, survey design, and sampling 13 techniques.
My primary research interests are social and 11 spatial disparities within cities, migration, and environ-15 mental conflict.
16 Much of my research in the area of environmental conflict 17 has focused on issues pertaining to nuclear power.
In 1978, I l
IS completed a study of the locational conflict surrounding the 19 proposed siting of a nuclear waste disposal facility in 20 northeastern Michigan.
In 1979, I co-authored (with 21 Professors Stanley D.
Brunn and Donald J.
Zeigler) a report on 22 a social survey of Three Mile Island (TMI) area residents, 23 which was conducted within one month of the nuclear accident.
21 Along with Professors Brunn and Zeigler, I also co-authored an 25 article entitled " Evacuation From a Nuclear Technological 26 Disaster," vhich describes both the TMI study and its findings 27 regarding evacuation.
(That ar ticle, which appeared in the 28
///
j 8201250188 920111 PDR ADOCK 05000275 PDR T
lt
i 1
1 January, 1981 issue of The Geographical Review, is attached to 2
this testimony.)
We conducted a follow-up survey of TMI area 3
residents in 1980, but we have yet to complete the analyses on 4
that data.
A copy of my curriculum vitae is attached.
5 I have been asked to summarize briefly the results of 6
that part of our TMI report which dealt with evacuation and to 7
comment in light of our findings at TMI on certain of the 8
evacuation plans developed for the Diablo Canyon Nuclear Power 9
Plant.
10 The accident at TMI near Harrisburg, Pennsylvania, on I
11 March 28, 1979, provided the first opportunity for an 12 empirical examination of the evacuation process in the 13 aftermath of an unexpected and unprecedented nuclear accident l
14 and near disaster.
As geographers, the primary objectives of 15 our study were to (1) identify the spatial and temporal 16 dimension of evacuation behavior among TMI area residents; and 17 (2) develop a conceptual model of evacuation-decision making 18 in response to a nuclear accident.
Because our study offered 19 an accurate and comprehensive assessment of evacuation l
20 behavior in a case study context -- the first such assessment 21 for a nuclear power plant accident -- our findings should be l
22 useful in the development of evacuation plans for nuclear 23 power plants throughout the United States.
24 The significant findings of our study can be summarized L
25 as follows:
26 (1)
At TMI, the official evacuation advisory caused 27 departure from a much larger area than was originally 28 intended or called for by the advisory.
If only the _
t 1
persons advised to evacuate had left the area, the number 2
of evacuees would have been limited to the estimated 3
2,500 pre-school children and pregnant women within a 4
five-mile radius of the plant.
But instead an estimated 5
144,000 persons, or 39 percent of the total population, 6
evacuated their homes in the area as far as fifteen miles 7
from the plant.
We describe this tendency to evacuate by 8
people outside the designated zone.of evacuation as the 0
evacuation shadow phenomenon.
10 (2)
The evacuation from TMI is the longest on 11 record.
Our study revealed that the evacuees fled a 12 median distance of 85 miles.
Most chose destinations 13 north and west of the crippled reactor.
This strong 14 directional bias appears to reflect a preference for a 15 site upwind from the plant, a psychological attraction to 16 the mountains in time of danger, and a reluctance to 17 select a destination in the more densely populated JS metropolitan areas to the east.
19 (3)
The majority (81 percent) of the evacuees stayed 20 with relatives and friends.
Few people used the 21 l evacuation shelter at Hershey, Pennsylvania, in all 22 probability because of the perceived social stress of
'3 living in mass quarters and of the perceived locational 2
24 stress o0 residing at a site so close (ten miles) to the 25 threatening reactor.
The maximum number of persons who 26 used the shelter in one day was estimated at only 180.
27
///
28
/// l
i r
I i
What are the implications of these findings for 2
evacuation planning at Diablo Canyon?
In an attempt to answer 3
this question, I reviewed the following documents:
1 (1) Evacuation Times Assessment for Diablo Canyon Nuclear 5
Power Plant, Phase I and II, prepared for Pacific Gas and i
6 Electric Company, by Alan M. Voorhees and Associates, i
7 Berkeley, California, April and September, 1980.
8 (2) San Luis Obispo County Nuclear Power Plant Emergency 9
Response Plan, Draft Plan Revision B, prepared by the San Luis 10 Obispo County Office of Emergency Services, October, 1981.
11 (3) " Evacuation Times Estimates," Chapter 4 in _ Earthquake 12 Emergency Planning at Diablo Canyon, prepared by the TERA 13 Corporation, September, 1981.
14 I am of the opinion that the evacuation times estimates 15 derived in these studies are likely to be grossly 16 underestimated, in large part because each fails to consider t
l i
17 and allow for any of the findinga of our TMI study discussed 18 above, including the directional bias of the evacuees, the 19 distances traveled by them, and, most importantly, the 20 evacuation shadow phenomenon.
Although the evacuation shadow 21 may not be a major consideration in evacuation planning for 22 natural hazards, it is especially critical in planning for 23 future nuclear accidents precisely because the geographical 21 scope of an invisible danger such as ionizing radiation is 25 difficult for public officials and private citizens to 26 determine.
Studies by Paul Slovic and others of Decisions 27 Research in Eugene, Oregon have shown that the risks 28 associated with nuclear power evoke greater feelings of dread i
i 1
than any other activities except terrorism and warfare.
In 2
planning for evacuation from a nuclear disaster, it can, 3
therefore, be projected that any order to evacuate will cause 1
the departure of residents not only from the designated zone, 5
but also from its peripheries.
The planning process should 6
accommodate responses from the two areas.
No such 7
accommodation or consideration appears in any of the documents 8
which I have reviewed regarding Diablo Canyon.
For example, 9
the concept of staged or partial evacuation to reduce 10 congestion during evacuation seems questionable in light of 11 the evacuation shadow phenomenon observed at TMI.
12 The plans also assume the willingness of the evacuees to 13 follow orders concerning the direction of evacuation, the 11 evacuation routes, and the evacuation destinations.
As 15 described above, however, this assumption is not borne out by 16 our study at TMI, which showed, for example, that the evacuees 17 stayed at the homes of relatives and friends rather than 18 utilizing designated shelters and that their evacuation 19 destinations reflect the influence of other subjective 20 i factors, such as a strong directional bias, a psychological 1
21 attraction to the mountains, and a reluctance to select a i
l 22 l destination in the more densely populated metropolitan areas.
23 The failure to consider these kinds of factors in the Diablo l
21 Canyon plans raises doubt in my mind about the accuracy of 25 their evacuation times estimates.
26 To accommodate the evacuation shadow phenomenon and 27 related factors in the evacuation plans for the Diablo Canyon 28 Nuclear Power Plant, a survey should be designed and a profile 1
c 1
of the population residing within a fifty-mile radius of the 3
plant should be taken.
Specifically, the survey instrument 3
should be designed to gather the following types of data or 1
information about the population:
5 (1) socioeconomic and demographic characteristics 6
(e.g.,
race or ethnicity, age and sex structure, family 7
size, occupation, education, automobile ownership, etc.);
g (2) daily activity patterns (neighboring and other 9
interaction patterns, including journeys to work, school, 10 recreation, shopping, medical care, and church);
11 (3)
Images, attitudes, and beliefs about (a) nuclear 12 power, (b) disasters, and (c) the agencies involved in 13 the implementation of evacuation plans; and 11 (4) decision-making in response to hypothetical 15 nuclear disasters of various magnitudes.
1G Previous research on both natural and man-made disasters, 17 including our own study of the TMI accident, indicates that 18 this kind of information may influence evacuation decision-19 making and, therefore, that it should be considered and go utilized in developing evacuation plans.
21 The results of such a survey, combined with data on the 1
i institutional population and the physical characteristics of 33 l 33 the area, should then be used as a basis for intelligent 3.;
decision-making.
Specifically, the data could be utilized to 35 (a) describe the population and settlement geographies of the 36 potentially affected area; (b) determine the transportation 37 routes that would be most suitable for evacuation; (c) 28 establish optimal locations for evacuation shelters; (d).
i
i I
1 l
1 predict the movement patterns of evacuees in order to regulate 2
the mass evacuation of an area; (e) plan for the delivery of 3
emergency services and supplies in the evacuation field; (f)
,1 create networks for the communication of disaster information 5
and for the delivery of services in the zone of evacuation; 6
and (g) identify the locationa that would be most difficult to 7
evacuate because of physical constraints, personal immobility, g
or attitudinal resistance.
9 In conclusion, I believe it is unlikely that the current I
evacuation plans for Diablo Canyon will work, primarily 10 lj because they fail to incorporate our findings regarding 13 population evacuation behavior during the TMI accident.
13 Principally, the existence of the evacuation shadow phenomenon 11 could increase congestion along evacuation routes and 15 significantly increase the time it takes to evacuate a 16 designated area.
The alternative approach outlined above, 17 which incorporates in the planning process a broad range of 18 social, economic, demographic, attitudinal, and physical 19 factors unique to the area and its population, is likely to 30 provide a basis both for more accurate evacuation times 21 estimates and more effective evacuation plans to be utilized 33 in efforts to protect the public in the event of a serious 33 accident at the Diablo Canyon Nuclear Power Plant.
21 25 !
26 l
27 l
28 i
n Stade m tjmteJ hrates of Amern a Reponted from Gr o<,p sem< u Rr s tr w Vol. 71. No. I, J.muar y IWI Cop >nght 9 IWI t'y the Amencan Geographical Soucty of Nt
- York k
w;.m. - q q EVACUATION FROM A NUCLEAR TECIINOLOGICAL DISASTER
- DON AI.D J. Zl.lGIJ.R, STANI.I'Y D. IIRUNN, and J AM1.S I1. JOIINSON J R.
unique peacetime technological disaster occurred in northeastern United States in the spring of 1979; an accident in a nuclear-generating piant.
Because of the proliferation of nuclear power plants throughout the wodd, the possibihty of comparable disasters elsewhere increases. We chose to examine one aspect of the 1979 American disaster: evacuation of the affected population. 'Ibe nuclear accident at 'Ihree Mile Island ('IMI) near liarrisburg, Pennsylvania, on Wednesday, March 28, 1979, dramatically emphasized the need to broaden the range of evacuation studies to include technological dis-asters, particularly ones of nuclear origin The crisis at 'lhree Mile Island pro-vided the first opportunity for an empirical examination of the evacuation pro-cess in the after math of an unexpected and unprecedented nuclear disaster. We used it as a case study in order to seek a foundation for geographical research in the nascent field of evacuation behavior and planning in response to tech-nological disasters. Our objectives are to identify the spatial and temporal di-mensions of evacuation behavior among 'IMI residents, to offer a conceptual model of evacuation-decision making in response to a nuclear disaster, and to suggest the role for geographers in evacuation planning. Because of the uaique-
..,m ness of the case study, we offer generalizations and models to explain the decision making process for nuclear evacuation not as definitive conclusions but rati cr as by,,otheses for future studies.
Joseph llans and 'lhomas Sell compiled a list of more than 500 natural and tet knological disasters that rettoh ed evacuation during the period 1960 to 1973.'
'iheir figures indicated that an average of almost 90.000 persons per year were lon ed to evacuate their homes because of hurricanes, lloods, train derailments
' N lhank ( Wula t'aurr. thret for of the Cattography I al> oratory. Unnemty et Kentusky, for t on-struttmg the graphis s.
' jowph Nt llans Jr. anil 'Ibomas C. s ll, I. vat uation Had s---An i valuanon 0 as Vegn. Nev.- llS.
L o
I nnronmental Protet tron Ac,ctn y, I'd 'il. pp.101 -l M O t)H 7t h.t i n is an assotant prof essor of geography at Old Deminion University,
+
Noifolk, Vuninta 2.UuS. Dn. lheNN ts a professor of geography at the University of Kentut ky, I cunnton, kentut ky 4uh.1)u loussos a an assistant professor of geog-taphy at the Unn ersity of California. I os <\\ngeles. Cahtornia **th)2 4.
n lasI ha t y hwn n t;cecar cal mutu cf ww w t
e Orvndt e
- 'g
i t
t r
i l
,I L
i 2
Tnt GloGR APinCAL RF. VIEW l
i I
involving toxic chemicals, and other types of disaster. Evacuations from tech-nological disasters accounted for only one-fourth of the incidents listed by Ilans and Sell in terms of affected area and population each technological i
f incident lacked the large scale effect that characterized natural disasters.8 Part-(
ly because of this limited scope of impact, study of evacuation from techno-i logical disaster has been neglected. Such evacuation has traditionally been
}
} N@W 13f-M viewed as a mechanistic problem, merely a question of logistics, in this article we hope to advance understanding of individual behavioral patterns during evacuation from a technological accident. 't his understanding will allow public officials and planners to base emergency-evacuation designs on documented
[
j behavioral responses rather than on assumptions derived from the experience of evacuations from natural disasters.
'lhe data for this study, hereafter referred to as the Michigan State Univer-sity (MSU) study, were obtained from a survey questionnaire mailed to a strat-ified sandem sample of 300 imuseholds in south central Pennsylvania approx-imately one month after the accident at TMI. The sample included 178 l
households w thin fif teen miles of the plant and 122 households in Carlisle, j
7 Duncannon, and I.ancaster, three communities beyond the fifteen-mile radius that we chose to include in the sample. Of the 267 questionnaires that reached their destinations, 150 were completed and returned, a response rate of 56 percent. A detailed description of the survey design and a copy of the ques-tionnaire appear in the final repmt on the TMI incident that we published elsewhere.' In this article we make reference to two other surveys of TMI area r
residents, although the final results of each survey are yet to be published, i
One survey was conducted by Mountain West Research for the Nuclear Reg-ulatmy Conunission (NRC), and the second was done by a group of geogra-r
)
phers at Rutgers University.'
l
't hese three 'IMI studies provide the basis for examining the emergency-plamung prmess in general and evacuation planning in particular. Methods of
..-4
' flarry I stdl N1me and otimrs, lictore the Wind-A Study of the ReTimse w lfurricane Carla, Dr Nrr hivNo 19 (Washmgton, D. Cc Natmnal Academy of Soi.wevNaimn 1 Rescan h Coun-
- 01. l%h, I homa L. Drab L. sooal harv es in Dnaster: l'amdy lxa. uaten. Sormi ProMcrro. Vet 4
in, tw). pp m m, IL kl lirsk, Communicatmn m Cn* !aptaming Tsamt...o Symt.t.hcahy, Ceanruorvatwo Ecuan h. Vol. 2,1975, pp & 49,1 arl 1. llan er, Prrdit tmg Revonse to llusu<anc 4
Wa: rungs A Re.tnalym of Data f rori inar 5tdes. Atsw be,rryenan, Yu!. 4, PC9, pp. %2 O and i
RonaM W. Proy. I vatuation Drouon-Mak mg m Natmal Dnasters, Alm burgenoo. Vol 4. PCu.
j pp 5 Ut.
- staa'cy D Hrunn lar* cs ll Johnson jr, and Donald j. ZrigIrt. I gnal Rrport on a Smjal Survey J
of ihn-c stdr Island Arca Hr uh-nts ti ast I ansing. high.. hinhgan
- tate 81nnetuty, Department of f.mgraphy. Peq. pp. le l
- Mountam West Riwan h, Inc., 't hree Al& Island lotcphone Survey. hehnunary Repo't on Pro-I
< edures and I m.hngs tiy ( )nttua It f ly nn, perparta for the Nurtear Regulatory Commiwton (Wnhmgton. D (.. US. Gmegnment Punhng Othic 19?C, Mountam Wea.: Resean h. Inc., with sm sal Impact Revun h. Inc. lhe Sm ul and liononne i flects of the Arodent at Three Mdc bland
[
by (antina Il lly nn and J.unes A ( halmen. prtparrd for the Nuitrar Regulatory Cumnuwon j
(Wnhmgton, D. C. US Cosernment Pontmg (Hhte,19s0L and kent llarnes. lames ltrosius, f
s sm Cultor. and James Ma(hril. Revonws of Impasted Populatiorn to the ihree Mile bland f
u i
Nmicar Reanor Aiodent An Imtui Awrwment. Doomwn Papre Nm 11 (New tirunswnk. N la i
Rutgrrs Unnersity. Orpartment of (,cography Pr79 the NRC study was sondusted by teltT one h
in July and August of 19N and the Rutgers study was based on,e questmnnaire maded in Aprd
[
of P69 In renrral the results of the MsU. NRC. and Rutgers studies are mutuath supportne; L
maior differomes are in the mn cptuahialmn and the yatial analym of evatuation behavior and druuon mA mg. topu s that are most tully developed m the MsU report.
1 i
l
't s
F t
,--.,_.,-n_.,-
.,, -. -,. - -. ~ ~.
. - ~ -,., _.. -,,
= = - _ _ _ _ - - _ _ -
4 9
y i
i j
EVACUATION WoM NUCLEAR DISASTER 3
e i
j coping with the consequences of nuclear disasters are certain to attract consid-l erably more interest than they have to date for several reasons. The TMI acci-i I
dent demonstrated that "societics using nuc! car power today must accept major j
{
accidents not only as a theoretical possibility of no practical consequence, but j
i as a risk to include in actual plannin;;."$ The results of state and nationwide opinion polls conducted since the accident at 'IMI indicate that supporters of i
{
nuclear power, though now insisting on higher safety standards, still outnum-l ber persons who oppose it? Nuclear-generating facilities in the shoa run, at i
i c'.
least, will probably continue to operate and to proliferate. Of the existing and
)
..;.u h w w u c d planned reactors in the United States,85 percent are sited within sixty miles i
i
~
of a metropolitan core and thus cast a nuclear threat over a large proportica of
{
f the population in the country.'
~
.,1 EVACUAT!oN-DLCistoN Max!NG l
5 Our study of the Three Mile Island evacuation was one of the first attempts l
to document the pracess of evacuation under the threat of a severe technological j
f-disaster. We designed the questionnaire to ascertain whether the respondent i
evacuated and to identify the factors that influenced the decision. The results I
i of the surv 'y indicate that 53 percent of the population within twelve miles
[
of TMI evacuated, while only 9 percent beyond this limit left their homes.
We propose a tentative model of the evacuation decision-making process and I
the spatial outcome of those decisions (Fig.1). The first question posed was
]
]~
this question. The remainder considered evacuation, but only 31 percent of the
[
whether even to consider evacuation; 21 percent of the sample never considered b
'W
~ 0.
sample decided to evacuate. Several external constraints on the flow of deci-i sions existed. Some potential evacuees were undoubtedly dissuaded from leav-l ing by temporal (when), spatial (where), and operational (how) constraints.
l 1he relationships in the diagram should thus be interpreted to present a system j
of interlockmg decisions rather than a series of unrelated options. Further re-1 search on evacuations from nuclear and other technological disasters may sug-l gest revision and refinement of the decision-making model and, perhaps even more importantly, may help to identify the critical factors that influence the decision making process.
o -
Two spatially distinct population groups were identified on the basis of their reaction to the 'l*di incident. One group, composed of individuals who remamed in their usual place of residence during the crisis, may be called the l
residual population; the other group, comprising the individuals who de-parted, is the redistnhuted population or evacuees. The MSU study found no l
stat.stically significant differences between these two groups in terms of oc-cupation, irnome, age of household head, length of residence in south central I
Pennsylvania, and political ideology. Similarly the NRC samly found that dif-ferences in irdome, education, and occupation had no significant bearing on an individual's decision to evacuate."
i 5 fient Sorenson. Nudcar Powc t he Ansacr 't hat Hecanw a Qucstnin, Ambs. Vol. H PU9, p.17.
j
- Rot crt C. Mitdidl. l'utiler Opuuon ani! Nudcar Power ticfore and After lhree Mdc 1 tand.
[
Rcwuren. Resoun cs for the l otuse, lanuary-Apnl 19NL pp 5-7.
i
' Pohey Rocarch Awniates, mioccunonuc Impxh Nudcar Power Niing P> tate College, Pi.:
Pois y Rescan h Awoates for the Nui tcar Regulatory Comnussion, PC"), p. II.
,I
- Mount.un West Researt h, Inc. woh Soual knput IIc carth. Inc., footnote 4 abtnc, p 24.
i I
t l
- . r: l %.* *..
lu j~.&) 'r;5;):4,
"%>\\',
d
,h;
,2, 7.s y..p.;s., p. * * ').~..
- a,
.y i
.w
t y
_,s j
4 1HE crocnArtiscat utvitw i
EVACUATION DECISION M AKING
]
Decisions to be Made SpatialOutcome of the Decision-Making Proceu Should n e (po)
Consider I v acua tion 8
.e a.
.'. :-Wid),4 M2.newi.n4 l
<,os i
Remain Should We at Home l
I vac uatei (Res; dual I
a t7 Populat.on) r k
p Shou d W hat Part l
t ntire I amily (no of I amily i
iva< vater Should E vacuate7 ttt it fy e,
=,
....;, f -
li'
.il 4
A E
1 When L...
Should We Cor l
fI 4i 4
T ra s el to a nd
^
S
~
Wtere Arrival dt i
Should We Gos Destination i
m <. i,.. ; c.
.c.,,%i,:
- J 6I II **'"##'I f
H <wv Should We C o) 1 L..
]
Whe e
/*
Should We R eturni N.
h
{
l'n. 1 -trat uation det ision snakmg l
lleginning with the study of the population affected by llurricane Carla, l
studies of natural.tisasters have repeatedly confirmed the hypothesis that a fanuly is the unit making the decision about evacuation."livacuees from sud-I den natural rate.trophes typically leave the hazard zone in family groups and remain together during the crisis. 'lhe results of our survey suggest that while
[
the majority of evacuees left in complete family units, the proportion of partial
{
' Moore and others, focinore 2 atu r. p.. ".
1 I
4:..
..~.7*.
j
1 l
[
l I
i t
I EVACUATION rRoNf NUCLLAR DIsASIER 5
l famdies fleeing the disaster was larger than would be expected from the con-l 4
t clusions of natural hazard research. In charting the basic patterns of evacuation j
behavior and in planning many phases of the evacuation process, the concept i
I of an evacuatic.n unit, as opposed to the individual or the family, may best serve as the fundamental analytical entity. We define an evacuation unit as a single individual acting alone or a group of individuals acting in unison during
- e, hher;6J the evacuation process. Because the members of a household may be unable to agree on a decision or on procedures, a single household may generate several evacuation units. In the MSU survey, partial families composed one-
[
third of all evacuation units, but in the sample communities beyond fifteen
-..s-miles from the plant, evacuation units were more likely to be partial families j
than complete families. Within six miles of the plant, complete families out-numbered partial families by more than three to one.1he high percentage of j
partial families evacuating the 1MI area may be accounted for by the high degree of uncertainty surrounding the accident itself and by the inability of either individuals or public officials to gauge accurately the magnitude of the j
malfunction at the plant.
i In his study of the sudden and unexpected impact of the Denver flood of l
1%5,1homas E. Drabek demonstrated that evacuation is not always the result of a simple scenario in which families receive a warning, seek to confirm the 4-4 l
. c danger, and decide to evacuate."Instead he proposed four separate evacuation processes: evacuation by default, evacuation by invitation, evacuation by com-L promise, and evacuation by decision. Although Drabek's classification scheme was specific to forced evacuation in response to a natural disaster, we propose a similar, but somewhat modifiod classification system specific to voluntary
'~
evacuation in response to a technological disaster. In terms of the response of families in the Three Mile Island area, evacuation seemingly resulted from three f
dif ferent processes: evacuation by division, in which some members of a family decided to (ave while ether members decided to remain; evacuation by con-
- n. ~ n sensus, in which the whole family decided that evacuation was the best course
{
of protective action to follow, and evacuation by compromise, in which a dead-
{
imk was resolved by a dominant family member in favor of evacuation. Dra-l i
bek's model of evacuation by default would be applicable only in the event of l
a forced evacuation.
t t he principal factor motivating TMI residents to evacuate was concern about li personal safety: 94 percent of the evacuees gave this reason (Table 1). Conflict-ing repotts from governmental and utility-company officials were another crit-ical factor. One-fifth of the evacuees indicated that the news media played a role in their decision. The NRC study ab o cited th> perception of danger and l
the volume of confusing information as the major reasons for evacuation."
t he reasons given by members of the residual population for not evacuatmg were varied (Table II) 1he most freiluently given explanation was that no order i
to evacuate was issued. '!he NRC study also found this response to be the most heiluent argument for staying."'Ihe existence of many conflicting reports was 4
" Dralvk, footnote 2 atmc. pp. 315-14A
" Ntountain West Hewanh. Inc. with Social impxt Researth. Inc.. footnote 4 abow, p.1R j
" Ntountain West Rewanh. Inc. with %:al tmpa(t Rewan h. inc.. Iootnote 4 abow, p. 21.
i
?$
l i
s A
6 1stL ci OGRArtticAI. <tvit.w T Aen t I-M ASohs top l'V AC U A1lN(,
- v... _, = u. :
.=:==:.=======.===z===--.==
I'I k C l N I AG E Or i V At U AllON pt ASON%
UNt!$
Centerned about safety 91 Conth< ting reports from government anit utshty.(ompany ofhcials 44
. wgyampi.vud A%w*4 Confhiting scports f rom unhtyaompany offaiah 26 Confht tmg irportr. f roin government of f u sals 28 News mrdia 20 I'Vrfj orH' WJ% CYaf uating 2
Orderrd to evatuate 4
5muc e: ;.15U Suncy, test hmtnote 3.
I Altil II-I(f A%ONS 8014 NOr I.VA(UAllNG
. = =. = = = = _ _ = =. = = = = _ = = =. = = _ = = = = = =
f4 K E NT ALF of kl A4oN%
P ONI V AL't,'t I.t.
No osJer to e vac uate was issued 62 3
Too many confhoing reports 42 No apparent tra on to evacuate 3M 42 llome was.: *,af e datante from plant 31 I rar of h.otmg 28 No clubhrn involved 23 Could na,a trave job or business 21 No one che in aree evac uated 16 Nerded to tA e < arc of f utn hvestos k 6
No pla< e to go 5
' loo old to evacuate 3
t hndiopera 2
Smoc c: MsU Suivry, tcat 6minote 3 the ~ctond-most frequently (ited reason for not evacuating. Paradoxically this ew wf..s;ac-a w,,. a,o the,n onJ-mW widely cited n'ason among th persons who chose to evacuate. Conflicting infonnation was thus used by some residents to justify a decision to 1" ave and by others to justify a decision to stay.
Tin. Gi ocnArmy or livAroA rloN s itoM 'IMI On the basis of the redisteilmtion of population in the immediate af termath of the lluce Mile Island disaster, we we re able to delineate two distinct but overlapping segions: the zone of evacuation and the evacuation field. The first zone comprises the ateas ! cit by the evacuees, and the second was the area to whit h the evacuees fled.
/oNI of LVACUAlloN 1he pattern of volontary evacuation from 't hree Mile Island clearly reveals a distance-decay eclationship that illustrates both the effect of governmental directives and the evacuation shadow phenomenon. The distance decay func-tion shows a sharp discontinuity approximately twelve miles from the plant (Table !!!). Within a twelve mile radms of the disabled reattor,53 percent of l
the sample reported that at least part of the household evacuated. Ileyond twelve miles, only 9 percent of the sample reported evacuation. The sharp
?-
. r.1
.. _ ~
l i
f i
i I
l EVACUAT!oN FHoM NUrLEAR DISASTER 7
i
~.
1 ABLE t!!-DMTANCE AND bVACUAMON RFSPON5E
)
~
TE RCE NT AGE OF PESPONDING DISTANCE 70NE HOU5tHot D% FROM WHICH IROM TMt SOME MLMBE R5 EVACUATED 1 to 3 miles 55 j
4 to 6 miles 56 I
v' 7 to 9 miles 53 10 to 12 miles 47 13 to 15 miles 13 More thar. 5 miles 9
Total sample 31 i
MhNd:p, rA.W.j Source: MSU Survey, text footnote 3.
discontinuity in the vicinity of twelve miles reveals the impact of two directives y
issued by the office of the governor of Pennsylvania on Friday, March 30. In V
the first, everyone within a ten-mile radius was advised to remain indoors, an action knmvn as sheltenng. In the second, all pregnant women and preschool children within a five-mile radius of the plant were advised to evacuate. The first directive seemed to establish the critical evacuation boundary in the minds of area residents. lleyond the ten-mile limit the proportion of respondents who evacuated declined sharply.
The evacuation-shadow phenomenon is the term used to describe the ten-dency of an official evacuation advisory to cause departure from a much larger area than was originally intended. The evacuation shadow cast by the public announcement of a very limited evacuation order extended wcil beyond the zone to which the order applied. It only the persons advised to evacuate had i
left the area, the number of evacuees would have been limited to approximately 2,500 preschool childien and pregnant women. Instead an estimated 14i,000 persons or 39 percent of the population, evacuated their homes in the area as j
far as fifteen miles from the plant." Although the evacuation-shadow phenom-enon may be a minor Consideration in evacuation planning for natural hazards, the impact of the phenomenon needs to be emphasized in planning for future nuclear accidents precisely because delineation of the geographical scope of an
.,w. %,.....e a.i.
invisible danger such as ionizing radiation is difficult for public officials and private citizens to determine. In planning for an evacuation from a nuclear disaster, it can therefore be projected that any order to evacuate will c..use the departure of iesidents not only from a designated zone but also from its periph-cries. The planning process should accommodate responses from the two areas.
LvACUATloN FILI.D In order to analyze the spatial patterns of evacuation behavior, we asked each evacuation unit to indicate its destination. Taken together, these desti-nations constitute the evacuation field of the survey respondents (Fig. 2). The i
spatial pattern, as inferred from the locations of these sites, suggests a calm l
and orderly movement rather than a hysterical flight. Evacuees fled a median distance of eighty-five miles from 'Ihree Mile Island. In the NRC study the median distance was found to be one hundred miles from the plant." In com-
" MountaE west Rewanh. inc., with wial Imrait Rescanh, Inc., footnote 4 aliose, p. 22.
" Mountain West Newan h. Inc., with Nomal tmpatt Researth, Inc., botnote 4 alu e, p.17
?
g4 ",,
,yg, gg
} ]
g.
e a
,,-T.
, O. [- $ *..
- f
, I.
., si 4 4
%=
,s
- - =. _.. -. - _.- - - ~.. -
. ~ - _ - _ = _ _ ~ -
_=_-.. _... _.
i
(
I e
L I
l
~
f I
7
'3 Till. Gl oGP Al'flicAL Hl VII.W e
i n
\\
f DESTINATIONS OF TMI EVACUEES
.s i
1 I
t-..-..__..._.._...._._..._..__..._..-.._.-..\\..
/
l'l r4NSf LVANIA I
i t
\\
.G.v.e w..., a.Oq
'\\
\\
l N
r n!O o:
W f
fl l
(
%egl, l
e
,/
\\.
- )
k
--n
~
N x
I, s
u
- I
%m Nt*W_ N; r
-cm. N i.
N g
a e !
l
, is ri,
)
j
(
N/
w
}
g*'
Wf ST vihGI N6A l. MAF(YLAf D./.d..% *
- [~~~~~
9, N
j
.s.
.. d l
r
!,/
\\
N
\\
a
~
.s 1
[,'N, / )
h
\\r-
'/
Vil(>lNIA
,n.
,.; y
.c l'../
j
\\n
(
N-N i sc. 2-Destiruhorn tif 'I MI cvacuers.
i
.aa.j, -.oo parison with ' e list of evacuations compiled by llans and Sell, the median liight of ev.ico..on from Three Mile Island is the longest on record. 'lhe longest median distan(e given in that study was eighty miles in response to !!urricane i
Carla in lW.I.
flaff of the evacuation units in the MSU survey chose (iestinations between f
forty five and ninety miles from the plant. We define the area in these limits as the /one of perceived safety t'ar from the nuclear site. 'ihe inner boundary this /one, shown on !!ig. 2 by a dashed line, suggests that the evacuees f
I o
sought destinations far enough from the plant to put a territori.d bufier between themselves arul the soorte of po'stble danger. 'lhe outer hmit of the /one i.eems to imply a telu(tance on the part of most evacuees to venture any farther than necessary from home. 'lhe tone of perceived safety represents the spatial out.
o.me of the tension between centrifugal forces generated by the perception of danger and centripetal forces renerated by the attachnwnt to home.
4 l
A strong duettional bias, si.nilar to that identified in studies of the intra-urban mobihty process, appears to have inthienced the configuration of the
[
e n >n,,a s,m. s.,, e, i..ime. n, m.o q
l
.j
- I ', e%,,,., g.;
l
i i
l
)
i I
LVACUAT!oN FROM NUCLEAR DI$AsTLR 9
cvacuation field. Although only one of every ten evacuation units chose des-tinations in the quadrant southeast of TA11, almost half chose destinations in the quadrant northwest of the crippled reactor. The directional bias was the consequence of severalinterrelated factors. The most important considerations seem to have been a preference for a site upwind from the plant, a psychological attraction to the mountains in time of danger, and a reluctance to select a
-f.vmeo..p yy destination in the more densely populated metropolitan areas to the east. These factors, and possibly others, require further investigation before behavioral
~~ '
j models of the evacuation site-selection process can be constructed.
In addition to sketching the configuration of the actual evacuation field, we j
j attempted to delineate a potential evacuation field for all respondents. Everyone j
was asked to supply a choice of destination, if a presidential order had required j
a full evacuation of the area. The map of potential sites displays a galaxy of destinanons to the north and the west of Three Stile Island and an evacuation
[
hollow, an area shunned by evacuees, around the reactor (Fig. 3). The maps of actual and potential fields are similar in many respects, although the map of potential destinations has a less clearly defined zone of perceived safety. The potential destinations were also more widely dispersed, and some were not shown on the map because they were as far away as California, Arizona, and Florida.
LVACUATION QUARTLRs The homes of relatives and friends proved to be the preferred evacuation quarters among both the actual and the potential evacuees. The $151.s survey found that 81 percent of the evacuees stayed with relatives and friends. The i
comparable figures were 78 percent in ihe NRC study and 74 percent in the Rutgers study."' ~lhese proportions exceed those characteristic of evacuations from natural disasters." Despite the ubiquity of hotels and motels in the evac-f nation held, their use by evacuees from Three hiile Island was limited, in all w.e g w..
likelihood, by the financial strain that such accommodations would have im-j posed on family budgets. The use of the designated evacuation shelter in lier-shey, ten miles from the plant, might have been limited by the perceived social 3
stress of life in mass quarters and by the perceived locational stress that evac-j uees would have experienced in a site so close to the threatening reactor. No respondent in either the htSU or the Rutgers survey repmted utilization of the public shelter in !!ershey, and only one of the 1,500 households surveyed in the NRC study used the evacuation shelter."'The maximum number of persons who u: co the shelter in one day was estimated at only 180, a situation that seems to confirm the finding that " shelter centers are used only if nothing else is available or if one cannot Imancially care for himself."" A reasonable con.
l clusion is that the low utilization of the shelter at IIershey was partially the i
" Mountain West Reseanh, Inc.. with Sosial Impatt Reseanh, Inc., footnote 4 above, p.17; and i
Harnes and others, footnote 4 above, p.17
" Moore and others, footnote 2 above. p. H; and '!homas E Drabrk and Kruh S. Ihygs. Famdies in DN ter: Reactions and Reiain es, le:arol of Af arnagr a,nf flic rana.'w. Vol. 30, lore, pp. 4 4b451.
i
" Da nes and others, footnot< 4 above, p.17; and Mountam West Research,Inc., with SodalImpact 1
Newrth, Inc, footnote 4 abme, p. 25 "llans and Sell, footnote I alm e, p. 52, and Mountam West Research, Inc., with Social Impact Resear(h, Inc., footnote 4 abo.c. p. 25.
M*,5. d y
d 10 un crocnaruscat utvnw
/
POTENTIAL DESTINATIONS IN THE EVENT OF AN ORDERED EVACUATION e-..~.._..-..-..-.._......_..-....-.--..
[
Fif4N5YLVANIA
..\\
l t
,\\
..,W/y.uw e gj l
7 4 0 ;l
-/ kDdO 0:
I 9
4 j
9 4'
.I
/
e.
=
+
/
t.
j
\\,
.h *.,
- t
\\
f I
. y()
, a me.
i J
~"f.
- y west VU<Ca fMA f ~"
,g,. /. %+ %,
MARYLAND i
s
!/
N,
\\. D
- n K
./
A
'/
/
Im
's
/
A
/ VIRGINIA
\\
c
..o 1.
n.. n..
l gg e
/.
-'l.
.c.v
/
%n N';
gi
/
ss~
i n. 3 -l'ot. niial destmations of ~l N11 area residenk in event of an ordered evatuatmn.
result of its location in the ten-mile /one from the reactor. Within ten to twelve miles of 'IMI,17 pert ent of the survey tespondents evacuated.
't he pattern of evacuation was innuenced by both spatial and temporal piocesses (Fig ;r in this diagram each prism of the cube represents the average behavior of evacuees originating in each of the six distance /ones used in this analysis. 't he distante of the horae from Three Mile Island was found to be thre[tly related to the destination chosen by an evacuee. In general, persons living tartber from the plant fled to more distant locations than did individuals living close to the plant. 'lhe same tendency was observed in the NRC study. '"
'Ilus linding adds a new dimension to evacuation behavior that has not been previously observed or pieditted, and several explanations may tentatively be of fered. First, persons living closest to the plant were likely to be the most concerned about the safety of their homes and property. They were therefore inclined to remain as close as possible to home. Second, only in the closest distance /ones were residents with high personal evacuation threshohls suf-ficiently motivated to abandon their homes. If these evacuees lived a few miles I
> stou$n west Researth. Inc., with sooat Impa ites.. ire b. Inc. footnote 4 above p.17.
". i
1 i
1 l
j ll TER
- vacUADON FROM NUCU AR DISAS SPATIAL AND TEMPORAL DIMENSIONS OF EVACUATION o../,,,,,,,,,,,,,,,,
N M 8"
~
0 6:- +-
>r~;.-,...,,.,.,
j.
O D#}Es o.
""" - i ( Il i
/
1, s s
{ l l l ( t~tyi.,,,
f I
u-I 3o,
o i
h I
g s..
M
+
th I
o f
C n
- f,
- f 18 O
5 c
o2
-~z~-
<m ms, t.
j 15-1\\ 16tIIe.*_
su, 8,.
I I l 41 1
lI i2I l
l F
AI4 f "i l
1g4 gig 6g?f l1g)t 9 g ; f's; Log ti Dates of Departure and Return Iw. 4-spatiat.in<t temporal dimensions of evacuanon.
l J
i further hom 'IMI, they probably would not have evacuated at all. T dents who live far from the disabled reactor would be expected to s little or no improve-uition sites in the closest zones because they would offer d that evacuees ment over the conditions of the home site. It can be hypothesize ld the seg-originating at great distances from 'Ihree Mih' lsland would inc u e hat would ment of the population with low personal evacuation thresholds t Fourth, because evac-consequently be likely to seek more distant destmations.
d i ntly women uation units residmg more than fifteen miles away were pre om na d.
and children, inany constraints on evacuation might have lessene
~
A
l.
i I
1
/
i 4
i i
12 Tur ca.ocuArmcat. nr.vlow
}
The temporal dimension of evacuation is along the third axis of the evac-l uation cube and represents the date and the duration of evacuation (Fig. 4).
The length of the hinges on which the prisms hang denotes the average du-1 ration of evacuation, while the position of the hinges denotes the average date I
of departure and return.1he length of stay away from home among respon-4 j
dents ranged from one to sixteen days, but 51 percent of all evacuation units i
returned home two to four days after departure. In the three-mile zone closest
! qqhq..y to the reactor,58 percent (none of whom had preschool children) stayed away l
three to six days, and 42 percent (all of whom had preschool children) were absent nine to thirteen days. As distance between home and plant increased j'
y up to the fifteen-mile radius, the duration of stay away frc m home decreased.,
i An increase in the duration of evacuation was observed in respondents from the three sample communities that were outside the fifteen-mile limit.
The majority of evacuation units (M percent) left on Friday, N1 arch 30, two days after the accident and the beginning of what was termed the crisis-re-l sponse period." An identical percentage was cited by the Rutgers study, and the NRC study reported 55 percent." The departure of so many persons that j
day can probably be attributed to a combination of factc.a. Fini, the ;;overnor's sheltering and evacuation directives were issued on Friday when serious con-i sideration of a full evacuation first became public. Second, two major con-straints on evacuation were lifted because Friday is the end of both the work-I week and the school week. Evacuees living close to the plant were likely to j
leave earlier than those living in the outlying communities. Whereas 77 percent of the evacuation units living within six miles of the plant left on or before Friday, only one-third of the evacuation units living ten or more miles away l
evacuated on Fuday. ill of the evacuees who reported leaving on hionday
]
lived ten or nwre miles from the plant.
A CONCI PIUAI. hloDr.l. or SUM ss-INDUcru EvAcurtion t
i The decision to evacuate from the Three h1ile Island area may be concep-j tua19ed as a behavioral adjustment to the stressful environmental conditions i
caused by the sudden nmlear accident. Evacuation in anticipation of disaster therefore becomes a stress management technique whereby an evacuee moves from one location to another in an effort to reduce the strain imposed by the pctception of danger."'Ihe stress-inducing factors during the 'Ih11 crisis were ihe knowledge that radioactivity hed leaked into the environment and, more importantly, the fear of an even !.uger catastrophe, that is, a core meltdown,
" Russell R. D> nes and others Repon of the limergency Preparednew and Response lask I orte, i
l staff Report to the President % Comnusuon on the Anident at 1hree Mde bland tWa-.hmgton.
l D C th Government Urbiting Ouice,1979), p. 43 Dynes divided the time af ter the auident in the emergent y re ponse period from Wednesday, Marsh 28 to Friday morning, Mars h 30. and the I
aius responw penod, begmmng on Friday mornmg. Man h 30.
i
' Harnes and others, biotnote I above, p.17; and Mount.nn West Rewart h Inc,, with Sodal impat t l
Ecse.nch. Inc., tootnote I at ove, p 24
" Stantev D. Hrunn Spatul Causes and Consequentes of Psp hosocial C.trew. in 1he Geer.caphy of flealth and Ibease (cihted by John M. Ilunter; Chapel Hdl. N. Ca Univeruty of North Cato'ma.
Dep.utment of Geography. l'C4). pp 138-lM; W A V. Clark and Martm Cadwallader, t o-(ational Atrew and Resniential Mobihty, f ner onmmt.mJ <armr. Vel s. PC3, pp. 24-41; Harold D. I oster. T he Geography of strew, Area. Vol 11, PC4. pp.10-10% and Juhan Wolpert Migration as an Ad ustment to I nvironmental Strew. /nmal et huf h3nes Vol. 22.1% pp **2-102.
i l
l Oq
. m
=,.. _,. _
1,vACUATION f FoN1 NUCII Alt DISAsTLR 13 GENERALIZED PERSON AL STRESS CURVES Do.qudt,num
.s,..#issen
- es.a.4%,#' d si k
5 E
F
~
.,.h'] ET $'.2", N
/
/
N~%'
~__/
~~~_
i qud bnum i
scy
- s.:cco
- n+wd incun i i e< u swn ',oen ' wxn ' r.no, 'a+r.cni reanI
. T.,';,.,
I su 5--Genenhn 3 personal trew curves.
at the plant, Only permanent rclocation is a more radical adjustment to per-(cived risk than evacuation. Less radical adaptations surfaced among TMl area residents, for esample, modifications of the daily personal routine such as remaining indoors and (onstant tuning to local and regional news.
3 vACU ATioN-10 Si oNst' lilRI.Silot Ds
,;a, w:vm.nc n n At any given distance from lhree Mile Island, the propensity of a household to evacuate depends on the evacuation-response thresholds of individual fam-ily members and on the availability and the desirability of evacuation quarters at varying distances from the source of danger. The evacuation-response threshold is that point along an individual's personal-stress continuum when the decision to eva(uate is made. Individuals with low thresholds wih tend to evacuate even if they live far from the source of danger, whde persons with high tluesholds will evacuate only if they live very near that source. As distance fmm the plant mcreases, the proportion of the evac uating population decreases, and the evacuating population includes an increased number of individuals with low evatuation-response thresholds. 'Ihe tendency of persons with low thresholds to move farther from the stricken plant than persons with high j
thresholds helps to explain the pattern of evacuation site selection with respect to the two distance variables presented in the evacuation cube (1:ig. 4).
I Generali/ed postaccident personal stress curves otfer another temporal mea-soie of responses to the disaster (l'ig. 5). ~lhe increased perception of stress on 1:riday is a;, parent on bo:h stress curves, but only the curve for the evacuees rises above the evacuation-response threshohl ~Ibe precipitous drop in the evacuees' level of perceived stress on Saturday was the result of departing for
o j
s i
i T110 GloGRAPlHCAL REVILW
{
~
1-1 SELECTION OF EVACUATION QUARTERS
' ' ' ~. <
t a
M Ys, dl
.cb s
"f b
C @ s a n,Q im pm
- MJes f rom TM) -
v. __p 7 ry_.yu w
7s n
m n_ 4 y
en_
(
I n
/
a sn o,id Home c
d l
~_
f emely
[
I tr w h l,wm.]
-) =
t f
i w
Morel shelter I
I J
omo.
l j
j t
t O r. u.a u_no, %,,n, m.m.. a l'u;. 6-Solntion of evacuation quarters.
sun, ue,num Consequently was considered a safe distance from the plant.
a destination that was reestablished.
h ed to alle-I;vacuation may also be motivated by reasons other than t e ne equilibrium k below individual evacua-viate stress. Some evacuees' stress curves may peahe persons may have acquiesce l
. ~. a tion-response thresholds, an indication that t though individually they l
in the decision by their family to evacuate, even by governmental author-i would not have taken the action. Forced evacuat on ld be other examples of ities and previously arranged plans to be absent wouidual may also rise abov d
such a phenomenon. The stress curve of an in ivti n is not an automatic res ceacuation-response threshold, but evacua o l
to 3;o, may have been con-example, some individuals may have had no p ace d on their mobility might consider by a job or other mnunitment.'Ihe possib!c evacuation sites th can be entered in a matrix of the search for evacud stress curves may be projected tifies available options (Fig. 6). Personali/c f the map of actual evacuation along each axis of the matrn. On the basis odestina fi and ninety miles from the disabled reactor. The locationa f perceived safety. A pears to dip in this range and to demarcate a zone o h total social and finan-personal or nonlocational stress curve, representing t e t
Or, h,
o e-f ene,., e, *..e a, s,.i 3 *, o
=,
I
O LvAcUATION IRoM NUCLEAR DisAsTrk 15 cial strain perceived to be associated with vanous types of evacuation quarters, y
was drawn to conform to the preferences expressed by survey respondents.
T he curve peaks at public shelters and dimimshes through motels, friends, and L
relatives. Although the use of second homes and campgrounds would be lim-g ited by personal circumstances, they are included as potential destinations.
LVACUAllON SPACE-SL ARCil M A1 RIX In the space-search matrix the most desirable evacuation sites can be iden-sa....me. e.4 tified by projecting the " lowest" segment of each stress curve into the matrix.
The area, delineated in Fig. 6 by a shaded border, has three potential sites that wouhi be open to this particular evacuation unit. The final choice under such circumstances would be made on the basis of nonlocational factors that enter the selectmn process. While locational factors would prevail to discourage the selection of evacuation sites either very near or very far from the nuclear plant, nonlocational factors would influence the selection of a specific site in tha geographical zone of perceived safety. Each individual would perceive the stress associated with location and types of evacuation quarters differently.
' Ibis personalized decision-making schema is only a first attempt to analyze the thinking and the planning by which individuals and families search in the N
surroundmg territory for an acceptable evacuation destination.
Tut Gl oGRAPur R'S 140! E IN EMLRGLNCY-liLSPoNSE P! ANNING Untd the ac cident at Three Mile Island, emergency-response and evacuation planning reteived surprisingly little attention from either the Nuclear Regu-lato y Conunission 0: government officials. Prior to the TMI accident, NRC had required nuclear plant operators to develop emergency plans only for the fa-cihty itself and the surrounding low-population zone. The zone around 1MI*
cxtended only 2.2 miles from the facihty. At the time of the accident, no evac-nation plans existed for the local jurisdictions in the area. Although the three closest counties had five-mile emergency-response plans on file, only one plan incorporated a fully developed course of action. Two emergency plans were develogd for the state at the time of the accident, but neither one had been
-,.i appmved by NRCM Alter noting the low priority that the Nuclear Regulatory Commission had accorded emergency-response planning, the President's Commission on the Anident at 1hice Mile Island dcommended that emergency plans, including evacuation, be designed for existing and proposed nuclear power plants on the basis of alternative disaster scenarios for any given plant. Scenarios would specify appropriate responses from state and utility-company officials on the ba es of both the magnitude of the disaster and the distance of residents from the genciatmg station. 'lhe commission considered a single evacuation plan based on a fivd set of distances and a fixed set of responses to be inadequate?
i
" A th4 Hough a flinjut of the plaris an ef fect on Marc h 2 8, 1!87'8 and of the ad Itu pl.itinin.*, dont-monts ihat twtved m response to the nuticar emerr,rnq n in Dvnes and others, lootnote 21 abose, pp 101 198
' the l* resident's Conunwon on the Au sdent at lhete Mile island. Ib Nent for Chanc.e T he I egan of IMI (Waslungton, it C.: U.S. (.overnment Pnntmg Othte.109. pp. 7t -77
' t w y-i gg W
b
~
16 THE croGRAPHICAL REVIEW In view of the minimal attention to emergency-response planning and the
~
recommendation of the presidential commission to identify appropriate re-
/
sponses for a range of conditions, there seems to be ample opportunity for geographers to contribute to the design and the implementation of emergency-response plans for nuclear emergencies. De role of the geographer in emer-gericy p.eparedness is considered most essential in the design of plans for
-t w e s.M. o y m evacuation and for delivery of emergency services. Expertise in spatial and
~
locational matters is especially critical in response to nuclear accidents and other technological disasters. Specific contributions of the geographer include the identification of the areas to be evacuated on the bases of distance and direction from the disaster site, the description of the population and settle-ment geographics of the potentially affected areas as a basis for intelligent decision making, the determination of the transportation routes that would be most suitable for an evacuation, and the establishment of the locations for evacuation shelters. Additional important contributions that geographers may make are the prediction of the movement patterns of evacuees in order to regulate the mass evacuation of an area and to plan for the delivery of emer-gency services and supplies in the evacuation field, the creation of the networks for the communication of disaster information and for the delivery of emer-gency services in the zone of evacuation, and the identification of the locations that would be most difficult to evacuate because of physical constraints, per-sonal immobility, or attitudinal resistances.
In addition to the magnitude of the accident, other factors may require the formulation of contingency plans to cope with the invisible danger and de-struction associated with a nuclear emergency. Evacuation, particularly if it begins as a voluntary process, will vary according to the season of the year, the day of the week, the specific weather conditions, and the availability of gasoline supplies.1: actors unique to the affected area will also need to be con-M-a.,-;m.....a sidered in anticipating the public response to an evacuation order, particularly rural-urban population mix, automobile ownership, ownership of campers, vans, and second homes, available public transportation, proportion of the population confined to institutions, location of friends and relatives, obstruc-tions in the transportation network, and extent of cooperation among local governments. A clear understanding of responsibilities and prior planning of appropriate emergency responses will help to facilitate evacuation from nuclear j
and other technological disasters.
l l
e e
d h
.~u.nG.K.,
Yf
rt:RRICULl'M V1'I AE Jam"n !!. Johnson, Jr.
Mailin: Address lione Address Department of Geography 11811 Venice ulvd. //218 University of Calif ornia Los Angeles, Calif orn ia 90066 Lo s Angel es, California 90024 (213) 397-4093 (213) 82 5-1415 or 825-1071 liirt h Place and Dat e.
Greenville, : orth Carolina: June 4, 1954 EDUCATION:
II.E., Geography, fiort h Ca rol ina Cent ral University,197 5.
M.S., Geography, Un iversity of Winconsin-Mad ison, 1977.
Thesis: lunovatlonn in Flue-crued Tobacco l'arning and Their Inplicationn for Energy Une.
(C.W. Olnstead, Advisor)
Ph.D., Geography, Michigaa Stat e University, June, 1980.
Dissertation:
I ncum ben t. Upgrading and Centrification in the Inner Cit :r: A Case Study of IIcighborhood Revitalization Activitica in Eacta w, Grand Papido. (C.D. Brunn, Advisor)
FIELDS Oi' S P ECI Al.l Z AT IO:: Teaching and Hencarch S y s t em.it ic : energy policy anc glann ing, urban, social geography of the inner city
!!e t hod s research des ign, theory and methodology in geography SPI:Cl A1. '!ONnRS :
Ih c i p i en t of t he 'Ibeodore R. Speigner Award for the hi;; hest academic average, Departnent of Geography, !! ort h Carol ina Cen t ral Un ivers it y, 1975.
Suma Cum Laude Grr.duat e, Department of Geography, Nort h Carolina Central Universit y, 1975.
Selected as an outstandine, Younn Man in America by t he Jayceet. in 1976.
Graduat e Fellow, Depar t ment of Geography Ur Iversity of Uinconnin-Madinon, 1975-77.
Itec i p i en t of ti c lirst place award ($100) for the bent graduate student paper presented at the East Lat es Division of the Asnociation of American Georraphern Annual :!eet in:,n, Michigan State Un iversit y, Septenher 15-16, 1978.
10B Expl:! iENCI:
Job Title Eng inee r ' ' Ann i.st ant Location:
U.S. Environmental Prot ect ion Agency, Research Tr iangle Park, Do t ham, '!o rt h Carol ina Put ie:,
Run Clien i. cal Ana l:, ;i r for double alkali scrohher syntens Yearn 19 7.!-7 5 (St ay in School P ro:; ram)
Job TitIe:
Junfor Computer Operalor Location:
Duke University Computat Ion Center, Durham, North Carolina Do t ie:.
Command IDM/%0 Comput er Years February 1975 to Augunt 1975.
.\\
James !!. Johnson, Jr.
Page 2 Job
Title:
Field Laboratory Technician 1.oca t ion :
Air Pollut ion !!onitoring Field Proj ec t, Tulsa, Oklahoma Research Triangle Institute, Research Trian;;1e Park, Durham,
!; ort h Carol ina Duties Operated and maintained two air pollution monitoring field research
""Y*
st at ions; annalyzed data f ro:a 03, 2, and M2 """
Job Tit le:
Teaching Assistant (lialf Time)
Locatlon:
Department of Geography,'!Ichigan State l'n iv e r s it y Du t ies.
Range f rom direct ing d iscussion sect ions to lecturing in the followint; courses:
Future Worlds -- Fall 1977; Spring 1978 Geography of Culture -- Winter 1978 World Regional Geography -- Winter 1978 Anm. ican Cit y and Region -- Winter 1979 Quantitative Methods in Geography -- Spring 1978; Fall 1978; Spring 1979 Advanced Quant it at ive !!et hods -- Wint er 1979 Years Fall 1977 to Spring 1979 Job Title Research Assistant (Quart er-t ime)
Location; Center for I?rhan Affairs, College of Urban Development, Michigan State l'n ive r s i t y Datles Computer analynis of data f or various proj ect s Years Summer 1978 to Spring 1980 Job Title Research Asuistant (Half-time)
Location:
Departnentu of Geograph-j and Agricultural Economics, Center for Environmental Quality, Michigan State liniversity Duties-Research and writ Ing of report s and presentations to legislative and consumer groups rea.arding Electric Ut ilit y Rate Reform in ?!ichigan (under contract # 71 - %93 )
Years Sumner 1979 to Spring 1930 Job Tit le Research Annistant Location:
- lic hi;;an Depa r t men t of Commerce, Energy Administration, Lan s ing Dutie:
Review grant proposals from schoolt and hospitals for energy conservation assist ance Re:earch on vanpooling,.
Years-June 1980 to August 1980 Pi;OPESSit G AL AND CIVIC AFFILI ATIO'S :
Church:
%mber, St. Joha :li: sionary Baptist Church Falkland, North Carolina Ot her Organi::at ions Gamma Pi Chapter, Gamma Thet a Upuilon Asaociation of American Geographers Trianyle Geegraphers Association Mi:,o i: ' ippi Counc il of Geographic Educat ion Rura l Soc iological Soc iet y of Anerica World Fut u re Society Key Club International
.\\
James I!. Johnson, Jr.
o Page 3 RESCAHeH 1" PROCPES3:
A Research Agenda."
Submitted to "Enert;y Policy and Public Wel f are Prof e:... f ona l Gemrapljer (co-authored).
"A 1.ocatlon-Allocation Model of Vanpool Park and Ride Points The Lansing, Michigan Exanple."
Suhait t ed to Geograph ical Pe r sp e c t im s (co-authored).
"Tecl.nology As llana rd. " AAG Resour., Paper for College Geography, to be publ ish-ed in early 1931.
"The Rol e of Coa: nun it y Act ion in *;e ir,hborhood Rev ital izat ion." Submit ted to
.U r ba n_ Ceg,; r_ a pi g.
"Toward:, a Geography of Technological llazards " Submit ted to Area (co-authored).
" Ext erna l it y Ef f ec t:. of an Abandoned liazardous Waste Dunp in Fullerton, California."
" Community Reactionu Touards If.iza r d ou r Uante Dinposal Fac i l it ies in West Covina, California."
" :a pp i ng the Perceived Inpact of the T:ll Nuclear Acc ident by Area Pesidents."
" Implications of Electricity Utility Rate Heform for Low-Income llouseholds in Oakland and Livingston Count ies, Michigan."
Submit t ed to the Journal of the Connon i t y Developnent Societv PU BI,I C AT I O::S 1917
" Black Migration ar a Henponse to Social Psychological Stress A Note on Migrant lettern, 1916-1918,"
Proceedinet, "ew En" land-St. Iaurence Valley Geo r,r a p h ic a l Society, Volone 6, pp. 42-46 (with Walter C Farrell, Jr.).
1978 "Blac k M ip,ra t ion a: a Hesponse to Social-Psycholop,feal Stresa: A Note on M i r,r a n t 1.e t t e r :
1916-1913,"
Tlu. Geographical Survey, Volu w 7, pp. 22-27 (wit h Ual t er C.
Parrell, Jr.); a reviacd version.
1973 "Blaci Philadelphians, A Factorial Ecology,"
Proceeding, Pennsylvania Academy of Sciences, Volume 52, pp. 91-95 (uith Walter C.
Farrell, Jr. and John D. Oliver).
1974 "The Qualit y of Af ro-American 1.if e in llouston, Texan A Ceographical Pern-pectize,"
in J.
P i l.:e r, U.
Pinder and M Proctor, P e r s p,e c t i v e;., on t !a.
U r b.m Sout h:
';el ec t ed Paper-t ro: the Fourt h Annual Conference, pp. 20/-221.
Nor f o l k, V irr,i n i.i : Nortoll. State College and i 'l Dominion Universit y,
- 1979, (w i t h '1a l t er C. Farrell, Jr. and Pat ricia Johnson).
1919 "Pecent Met hodolog ical Devel opment s and the Geography of Plack America," The Penn. fivan ia Geo7rapher, Volume 17, pp. 19-30
(> it h Wal t er C Farrell,
.I r. ).
\\
O Jiir:en 11. Jol nson, J r.
Pa f'y 4 PI:n L I CAT I O::S (cont. ) :
1979
" P h e n o: enoleg/ i n Geoy,r ap!cz," lhe @fy.ryhical Su m y, Volume 8, pp. 3-9 (w i t h
'.la l t e r C. Farrell, Jr.).
19/9
" Educational Concern of Inner City Plack Parents," FPIC Docum n.t (Sp r in;; 1979),
- to be anninned (w ith Ua l ter C.
Farrell, Jr.).
1919 Pre l in i na ry Repo r t _on._.a Socia! Survey of Three Mile in. land Area Penidepts.
Department of Geography, flichigan Sta t e University, May J')/9 (ulth Stanley D. Brunn and liona ld
.1.
Ze i r, l e r ).
- L ytjaj Survev of ]hree ; file Inl:ntd Area Hruidf.:nt s,
l W il Ei n a 1J',i;p o r t en S
1)e pa r t t.en t of Geogtaphy, Michinan St at e Cnivernit y, Augu:,t 1979 (uith
'itanIcy D.
Br unn an I Donahi J.
Zeigler).
1979 P t.e l ini na ry Rev i.
and Anal.nis of Electrie Ut f lityJat e 1:e f orm:
Implications t e.r o.@ land and I.!v ine :t on Count ica, Michigan.
Report for :lichigan Comuni ty Ac t ion Agency lu.nocia t ion and Oak land-Livin;' ;t on liuran Service: Agency, under centract # / l -- M 9 3 Departments of Geog ra phy and Ay,r ic u l tu ra l Ec onmn i c: and Center for Env i ron: ental Quality, Michigan State Un iver s it y, September 24, 1979 hii t h Brad ley T.
Cullen).
1979
"',pa t i a l Pat t erne of Alcohol tintlet: in the Washington, D.C. Black Conn. unity,"
Preccedinc., Pennsylvania Acadcov of Sciences, Vol. 51, pp. 89-97 (with
'arvin P.
1)a..'k i n s an ! Ual t er C.
Farrell, Jr.).
1930 "ises i d en t f a l P re t e r enc e Patternn of Afro-Ancrican College St uth nt : in 'lhree Dif f erent Staten," liie Profes;ional C r oy,r anlj e r,,
Vol. 32, pp. 3/-42.
19cm "cpatial and Behavioral A:.pec t : of the Count ernt rean i!igration of Blacb to the :;ou t h," in, Brunn, St an l ey I). and J;rne:, O.
Wheeler, i. d.., The Aeerican tie t ropo l i t,in Syn t c:.
Precept and rot nr.
New Yore
- John Uile' and Sonn, cripta Se r i e:.
in Geography, V ic t or Uin:,t on Publ i nher, in pre:,
j 19Hu "1.oc a t i on.i l Conflict and Puhlit At t i t ude: Rep,arding the Durial of Nuclear l
U.c. t e The Alpena, Michi; on ! < per ience," F,. ; t La b e < Gen 'ragh r, Vol. l '),
t pp. 2 ', - 4 0 (wi t h St an le-I;runn m.1 Mrlao 'cGirr).
l 19 ;',1 "E"a cua t l on l i o-the '.uc l ea r 'l echno liin ic a l D I:,an t e r at Three ' tile Inland,"
l
.i n prer ;, lhe f :coc carla.i c a l _ lh zif'<,
.lanuary (co-authored).
I j
.g3 h
d a
.b p
3 J r i ', t ' a j :
l'> b'.
[!. ! I ' d o 'd y lIs
- 1 a S I E Un C-
'A t '/ b'iLOC 5 > t t 5 0n Cea:' ' '.. i t ' a f
.c..!
th Dyie-1.
i 'o u a t iGn, J.um iey, l'Ut p;i.
~%
o.
James 11. Johnson, Jr.
Page 5 PRESENTATIONS:
1976 Prairie Vlev A!J1 University, Department of Economics, Geography, and Social Scienc e and School of Agriculture, "The Effects of "echanization on the Production of Flue-cured Tobacco in the Coastal Plain of North Carolina,"
Prarie View, Texas.
1976 Annual : lect ing, N et, England-St. 1.awrence Valley Geo;;raphical Soceity, " Black liigration as a Response to Psycho-Social Stress A Note on !!i ; rant Letters,"
l Salem, !!as. ;achuset t n.
1976 Annual !!eetin;;, Association for the Study of Afro-American Life and llistory, Con.nentator for ses: ion " Considerations in Culture," Chicago, Illinois.
1977 Annual :lecting, Popular Culture Associzition, " Folk Reasons for Early 20th Centur: Black :ligrat ion," Ba lt i mo re, 5taryland (with Ualter C. Farrell, Jr.
and Phillip Kitchart).
1977 Cheyney State College, Urban Studies Departnent, " Data Gathering Techniques in the Social Sciences," Cheyney, Pennsjlvania.
1978 Annual leeting, Norfolk State College and Old Dominion Universi.ty Conference on the Urban South, "The Qua lity of Af ro-American Life in Houston, Texas A
Geographic Perspective," Norfolk, Virginia (with Walter C.
Farrell, Jr.).
19/3 Annual 'teetinn, Michigan Acadcay of Sciences, " Factors influencing the Decline of White and Non-Wh i t e Operated Farms in the Central Coastal Plain of :: orth Carolina, 1945-1969," Ypsilanti, !!ichigan (with Walter C.
- Farrell, Jr.).
1973 Annual "eet i nn, Pennsy lvan ia Acadeny of Science, " Black Philadelphians-A Fa c t o lal l:co l ogy," Charp l on, Pennsylvania (with Ualter C.
Farrell, Jr.).
19/8
!!ont h ly : tert inn, Woodbr idge Neighborhood Citizen Counci:1, "The Proces-of Neighborhoo41 nevitalization:
A Case Study of the Woodbridge Community - A Research Proposal," Detroit, Bl i c h i p,a n.
1973 Annual :teet inn, East I.a k ea Divinien, Assoc iat ion of American Geographers, "Sp.i t i a l V.ir i a t i ons in At t itudes Toward Nuclear Waste Disposal in Alpena, Michigan," East lanning, !!ichigan (with Brian J. StcGirr).
1979 Annual 3'cet inn, Wr f olk '; t.i t e Colls,;e emd Old Dom i ni on Un iversi t :, Conference on t he Urban Sout h, " Black :ligration to the South:
A l'oeur on Durham, North Carolina," Norfolk, Virginia (with Walter C.
Farrell, Jr.).
1979
'ii:;t h Annua l ':ational Conference on the Black Family, "Scarching for Elbow Roon:
A Perspective on Southern Black 'ti:; rat ion," 1.ou i sv i l l e, Kentucky (wi t h W.il t er C Pa r r e l l, J r. ).
\\
Janen 11. Johnnon, Jr.
Page 6 P!:F.Si::; TAT IO::S (cont. ) :
1979 Annual Meet inn, "ichigan Academy of Sc iences, "I:rhan I!onent endin:; and liintoric Preservation:
Problen and Prospects," Mt. Pleasant, Michigan (with Ualter C.
Fa r r e l l, J r. ).
1979 Annual Meeting, Penunylvania Acad e n:/ of Science, "Sp:s t i a l Patterns of Alcohol Outletn in the 1:anhington, D.C. lil ac k Com:.un i ty," Moun t Pocono, Pennsylvania (with Marvin P.
Daukins and Walter C.
Pa r re l l, J r. ).
1979 Annual Meet ing, East Lakes Division, Annociation of Ancrican Geographers,
" Elect ric it y Ut ility Ra t e 1:cforn or Maint enance of the Statun Quo?" Akron, Ohio (ui t h lirad ley T. Cullen).
1979 Quirterly "eeting, Michigan Comuni t y Act ion Agency Associat ion, " Preliminary lhvies and Anal:, si: of 1:lectric litility Rate Reform:
Implications for Oakland and Livin;p;t ee Count j en, Michigan, Lansing, Michigan.
1979 Mit hl y "ect ing, Oakland-Liv ingst on Human Services Agency, "P,ene f it s and Cos t s of IIfeline Electric Ut ility Rate Reforn," Pontiac, Michigan.
1919 Annual "erting, Southeast Divinion, Association of American Geographera, "Locational Conflict and Puhile At t itude: Megarding the Burial of Nuclear Uante The Alpena, Michigan Experience," Manhville, Tennerque (with Stanley D. Brunn.nul Drian J. McGirr).
198I l'acu l ty Sen inar Ser ics, UCLA Cent er f or Af ro-/aerican Stud f es, " Spatial Perspect ive: on Counter-st rean !!)ach Migration to the South."
198i Annu i1 ':ect in;, Uentern Soclai Sefence Assoclation, " Electric IIti1ity Hate Ref o r:.
'l h e '; i gn i f i canc e of the Spatial Factor," San Diego, California.
1931 Annuai M"eting, Southuentern 1.conomica Annociation, "Combinational Progran-ning Solut ion to a Park anti Ride Man, 'I r a n s i t Problen," Dallas, Te::an (with Bradley T. Cullen).
1931 Annual "eet inn, Annoe ia t ion of American Geographern, "1.oca t i on-A l loc a t ion Male l of Vanpool Park and Rido '; i t e:,
The 1.a n n i n g, Mi cit i gan F.::a m p l e, "
, gle :, Calidornia (with Bradi,y T.
Cullen and Lawrence M. So:r 1 rn).
! x, MEDIA 1:; M r /I E' S AND PlmLICAT ro!",.
191/
" loc ent ive to Po o r " response to editorial "No Free Abortion," Tulsa WonLd
( Au p,u s t-26, l9/7),
- p. 5.
197H Wi: A!' (rad io), Eant 1
- ing, Michigan.
" Nuclear Waste Disposal in North-ca, tern 'lichigan."
"tatr_Je n (October 9, 1979) with 19/9 "Liteline liill Will Not Aid Poor," The nrad l e:
T.
Cullen.
- \\
\\'
Janes H. Johnson, Jr.
Pa;,e 7 MI:DIA I N i i:M'/ ! !!,.:. A::D Pl:BI.I CAT IOS'. - (c on t. ) :
1979 WFA!< ( r;u! 1 o ), l'a s t I.a n s.i n c,,
iii t h i p,a n.
" Social Survey of 'i hree :lJ 1e Island
/i r e.i Re::id en t :." ('.li t h ': t a n l ey D. Brunn and Donald I. Ze i;;l er) 1979 UE!!!, Channe l 11, 1:a s t I.a n :, i n g,, !!! c h i ga n.
"Prelininary Report on a Social Survey of Three Mile island Area Res ident n."
(Uith Stanely D.
Brunn and Donald J. Ze il;] e r )
3 ARTICI.ES ';HITTEN IW OllIEi"; OS VARIOU'; TOPICS:
1918 Fran Murray, "Alpena Re:,ident s S t il l Oppose Nuclear Waste Site," p r e s t.
release, Departnent of Inf ornat ion Services, Mich igan S ta te l'niversi ty, E.i n t I.a ns i ng, :li c h i r,a n, September 21, 1978.
1978 United Pres, international.
"A! pena Count y St il l Av.a inst a Uaste Dump Site,"
Det roi t ::eu, (September 24, 1973).
A number of other !!! chip,an cit ies '
upaper:. carried account s from the o r ip, i n a l um release.
nc-19/9 F r.in ::u r ra y, "i:e s i d en t.< of 'lbree !!Lle Island Area l' ear Inpact of Disaster,"
pres, release, Depa r t r.en' of In f orm t ion Services, Mich ip,:m ';t a t e University, I:a s t 1.a n s i n;;, :!i c h i r,a n,. fay 18, 1919.
19/9 l'ran "urray, "li Survey of 'Ihrec !!1 l e Island issue," MSl! News Bulletin (Oc t ober 18, 1979), p.
2.
19/9 1:l l i r, ro s e, "In the Al t ert:a t h of Three !!il e Island," Det ro i t Fren Press (oc t oher 31, 19/9), p.
10/t.
19/9 E.iren Zura.e ki, 'T.u. t o.cn Ce uunity Ao:ociation S t u,!'.
Moven"n t s o f Peop le,"
Phuto
.norter (octoher I,
1979), p.
1.
1
}
1 l
)