EFFECTIVENESS OF BOMBING IN NORTH VIETNAM

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP78S02149R000200060006-7
• Release Decision: RIPPUB
• Original Classification: S
• Document Page Count: 16
• Document Creation Date: December 15, 2016
• Document Release Date: December 31, 2003
• Sequence Number: 6
• Content Type: REPORT

Approved For Release 2 04/02/02 : CIA-RDP78S02149 000200060006-7 25X1
1. The cumulative effects of the air raids on
North Vietnam continue to limit the capability of
the North Vietnamese forces for overt aggression,
but they have not reduced the ability to support
military activity in South Vietnam and Laos, either
at present or increased levels of combat and manpower.
2. The bombings appear to be having some dele-
terious effect on public morale. We have reports of
consumer shortages, and the extensive evacuation of
civilian personnel has caused some economic difficul-
ties. There is no good evidence, however, that the
leadership's resolve to continue the war has been
weakened.
3. The bombings have forced the diversion of
major resources and work forces to repair damage.
North Vietnam has been denied the use of larger
coastal carriers, and is severely limited in the use
of established port facilities. Measurable losses
to the economy and military establishment are esti-
mated at $184 million.
4. On the other hand, POL supplies have not
dropped below 60-day levels since September 1, and
attacks on rail, highway and waterway transportation
have reduced efficiency but have not stopped move-
ment. Energetic construction of replacement and al-
ternate bridges have rail capabilities at their high-
est level since the bombing stand-down a year ago.
25X1
Approved For Release 20)4/02/02 : CIA-RDP78S02149 000200060006-7
 Approved For Release
Tab B
CASUALTIES RESULTING FROM THE BOMBING OF
NORTH VIETNAM
I. Total Casualties
1. The cumulative killed and injured attributable
to the bombing of North Vietnam, estimated
through September 1966, total about 29,000,
of which 18,000 are believed to be logistics
workers and other civilian personnel. Some
13,200 of the total casualties occurred in
1965, of which 6,000 were in the "civilian"
category. During the first nine months of
1966 we estimate total casualties of 15,700
of which about 11,900 were civilians. (See
Table I). On the basis of sample data through
September 1966 the total casualties in 1966
are estimated to have been about 40 percent
killed and 60 percent injured. This division
is based on averages derived from the sample
cases for which we have hard evidence.
2. The composition of the casualties resulting
from the 1966 air campaign differs notably
from that in 1965. Over 75 percent of the
casualties in 1966 were logistics workers
and other civilians, compared with 45 percent
in 1965. Armed reconnaissance against mili-
tary supply routes, which accounted for only
52 percent of the casualties in 1965 accounted
for an estimated 93 percent of the total in
1966.
3. The increase in casualties during 1966 is a
function of the growing scale of the air
campaign, the greater emphasis on armed re-
connaissance against lines of communications
and the selection of a few fixed targets
located in more heavily populated areas.
Despite the increase in casualties, however,
the number per mission has remained about
stable.
25X1
25X1
Approved For Release 2004/02/02 : CIA-RDP78S02149R0p0200060006-7
 Approved For Release
4. The preponderance of civilian casualties
resulting from the acceleration of armed
reconnaissance has in large measure in-
volved those civilians working at military
targets or directly engaged in the mainte-
nance and operation of the logistics system
which moves supplies and personnel into Laos
and South Vietnam. Hence, it is heavily
weighted with transport repair crews, truck
drivers, and personnel operating logistic
craft on the rivers. The bombing campaign
has generally been quite selective. The
few official North Vietnamese statements
protesting the extent of civilian casualties,
while vigorously presented, actually cite
very small numbers of personnel. The May
1966 North Vietnamese letter to the Interna-
tional Red Cross claimed only 239 civilian casu-
alties since 31 January 1966, although implying
many more. The Education Ministry in October
1966 stated that 300 students and 30 teachers
had been killed since the bombings began.
The well-publicized propaganda statement on
the attacks against Nam Dinh, where heavy
casualties would be expected, indicated that
37 percent of total casualties were innocent
victims of the war--women, children and the
aged. One would expect this propaganda
statement to make the maximum claim possible.
Even if the specific casualty claims for Nam
Dinh were accurate, this percentage would not
seem unduly high. Nam Dinh is a heavily de-
fended population center subjected to heavy
attacks and with many of its military targets
located in densely populated areas. It is
hardly a typical case, and this percentage
should probably be regarded as the upper limit
for casualties sustained by so-called innocent
victims.
25X1
25X1
Approved For Release 2004/02/02 : CIA-RDP78S02149R0p0200060006-7
 Approved For Release
II. Methodology for Estimating Casualties
5. Although we cannot claim absolute precision
for our estimates of the number of civilian
casualties in North Vietnam, a variety of
available sources and analytical techniques
makes us reasonably confident that these es-
timates are of the right order of magnitude.
25X1
25X1
6. With the information provided by this wide
variety of sources a number of analytic tech-
niques are used to arrive at our estimates of
casualties.
Approved For ReleaseI2004/02/02 : CIA-RDP78SO2149ROgO200060006-7
25X1
25X1
 -Approved For Release
a. Fixed Targets
The basic method of determining casualty
estimates for attacks on individual
fixed targets rests on a casualty pre-
diction methodology developed by US
intelligence analysts. This method
is essentially an application of the pre-
dictable lethal effects of given types of
ordnance to known target environments,
particularly their population density.
These factors are applied in conjunction with
roof cover bomb damage assessments based
on photographic analysis, and operations
reports from specific attack missions.
The number*of casualties is determined by
multiplying the. average lethal area of
the bombs dropped in target areas by the
population density of the area and the
number of structures in the area. These
estimates are expressed in a broad range
of minimum and maximum casualties.* The
maximum figures assume that the popula-
tion in the target area was unwarned; the
minimum figures assume warning and full
resort to civil defense and protective
measures.
The casualty estimates derived by US
intelligence specialists are weighted
by CIA on the basis of sample evidence of
actual casualties obtained from Prisoner
of War (POW) reports
We have found fairly consistently that the
initial attacks on JCS targets have in-
flicted casualties that are four times
greater than the minimum military esti-
mates. Consequently, we have adjusted the
*The maximum figure would generally be 10 times
greater than the minimum.
25X1
25X1
25X1
Approved For Release 2004/02/02 : CIA-RDP78S02149RO90200060006-7
 Approved For Relea
military estimates of minimum number of
casualties,by a factor of four for all
initial strikes on fixed targets, and ac-
cepted the minimum estimate for all tar-
gets undergoing restrikes. [This assumes,
and we believe the assumption to be reason-
able, that once the target has been struck,
warning is heeded and passive defense
measures are adhered to.]
b. Armed Reconnaissance
i. Civilian Casualties
The CIA estimates of casualties re-
sulting from armed reconnaissance
strikes are expressed through a casu-
alty/sortie rate derived from exten-
sive analysis
_This analysis yielded a rate of .23
casualties for each armed reconnaissance
sortie.
25X1
25X1
ii. Military Casualties
The information base for estimating mil-
itary casualties resulting from armed
reconnaissance missions is extremely
oor
25X1
25X1
25X1
Approved For Release 2004/02/02 : CIA-RDP78S02149IR000200060006-7
 Approved For Relea
25X1
The estimate of military casualties
is, therefore, based on pilot reports
of the'number of targets destroyed or
damaged, weighted by Defense Intelligence
Agency'(DIA) factors for the probable
number of casualties to be expected for
each type of target destroyed or damaged.
In two specific categories--trucks and
watercraft--the DIA casualty weights
were decreased to reflect the informa-
tion gained from POW reports 25X1
25X1 In view of
the poor data base available the esti-
mates of military casualties resulting
from armed reconnaissance are probably
the most unreliable of all our casualty
estimates.
25X1
Approved For Release X2004/02/02 : CIA-RDP78S02149R00g200060006-7
 25X1
Approved For Releas
III. Applicability of the Nam Dinh Claims
The propaganda statement* distributed by the
DRV Ambassador to foreign correspondents in
Moscow early in November 1966 claims that
from early 1965 to September 20, 1966, air
attacks against Nam Dinh killed 89 persons
among them 23 children and 39 women. Wounded
numbered 405, including 41 children and 81
women.
The methodology used in calculating casual-
ties in urban areas in the CIA report An Eval-
uation of Allied (US and GVN) Air Attacks
Against North Vietnam
8 November 1965, was based on the city of
Nam Dinh, which was used as a case study.
The casualties at Nam Dinh in November 1965
were computed using composite estimates based
on ?e-strike estimates prepared for the JCS;
a DIA estimate, assuming warning and based on
examination of post-strike photography; and
an estimate drawn from World War II experi-
ence. The latter was based on a review of
casualties in bombed German cities during
1943, before blockbusters were used and fire
storms were created, It was concluded that
casualties (killed and wounded) for each air
strike in urban areas would range. from a min-
imum of 1 per 18,000 in population exposed to a
probable of 1 per 12,000 to a maximum of 1 per
8,333. Applying these. factors to the popula-
tion of Nam Dinh (about 90,000)** and to the
six air strikes that had been launched against
this city in November 1965, casualties were
estimated as follows:
*Report on US War Crimes in Nam Dinh, by Committee
Investigation of US Imperialists War Crimes
in Viet Nam of Nam Ha Province.
**Based on a city with a population of 90,000 casu-
alties per air strike range from a minimum of 5 and
a probable of 705 to a maximum of 10.8?
Approved For Release) 2004/02/02 : CIA-RDP78SO2149R0Q0200060006-7
25X1
25X1
 Approved For Rele~
The propaganda statement mentioned above
claims that from early 1965 to 20 Septem-
ber 1966 a total of 47 day and night air
strikes were completed against Nam Dinh,
killing and wounding a total of 494 per-
sons. Applying the estimating methodology
used in the November 1965 CIA report to
the number of air strikes claimed by the
Communists produces casualty estimates as
follows:
In a press conference on 29 December 1966,
a Pentagon spokesman stated that military
targets in Nam Dinh had been struck 64
times. Our methodology, described above,
based on this number of strikes would pro-
duce casualty estimates' as follows:
On the basis of these various methods of
measuring probable casualties, we conclude
that official North Vietnamese claims on
Nam Dinh may be accurate as to total num-
bers.
8. Casualties Among Women and Children
Armed reconnaissance sorties accounted for
more than 99 percent of the total attack
sorties flown in th first nine months of
1966, compared with about 75 percent of
those flown in 1965. The emphasis on
armed reconnaissance rather than attacks
on ftxed targets during 1966 resulted in
armed reconnaissance accounting for an
estimated 93 percent of total casualties,
compared with 52 percent in 1965. The
25X1
25X1
Approved For Rellase 2004/02/02 : CIA-RDP78S02149R00g200060006-7
 Approved For Releo
swing away from attacks on JCS-designated
fixed targets had several notable results
in terms of casualties in North Vietnam.
In 1965 a greater emphasis on attacks on
fixed targets, which are predominantly
military, resulted in military personnel
accounting for about 55 percent of total
casualties. In 1966, however, military
personnel accounted for only about 24
percent of total casualties.
The preponderance of civilian casualties
resulting from the acceleration of armed
reconnaissance has in large measure in-
volved those civilians who are most di-
rectly engaged in maintenance and opera-
tion of the logistic system moving sup-
plies and personnel into Laos and South
Vietnam. Through September 1966, over
20,000 units of transport equipment
and miscellaneous transport facilities
have been destroyed or damaged by armed
reconnaissance missions.
Armed reconnaissance missions also have
destroyed or damaged over 10,000 miscel-
laneous military facilities such as bar-
racks, supply warehouses, antiaircraft
sites, radar and communications sites.
Through 30 September 1966 only 175 JCS
fixed targets of all types had been at-
tacked, of which 93 were direct military
targets and 54 were railroad and highway
bridges. There were less than 30 JCS
fixed targets struck in the built-up areas
where one might reasonably expect to find
some concentration of women and children.
This is not to say that there have been
no casualties among women and children
as a result of substantial damage to
military and transport facilities and
equipment, including railroad and highway
bridges. But by the nature of the func-
tions these targets perform and their rel-
ative isolation from civilian activities
Approved For Releo
25X1
25X1
 Approved For Rel
the probability of there being women and
children in the target areas should not be
high. This probability must be substan-
tially less where men's work is being done,
even by Vietnamese standards, than it would
be in attacks on the more important economic
targets in built-up areas. In the case of
Nam Dinh the Communists claim that over one-
third of the casualties to 20 September 1966
were women and children. There is no way of
confirming or denying this statement. Never-
theless, it would appear to establish an up-
per limit to the percentage of this type of
casualties in urban attacks. The attacks
against Nam Dinh are not typical of the
preponderance of the air strikes against
North Vietnam. It would be logical to char-
acterize the attacks against Nam Dinh as one
extreme, and the bulk of the armed reconnais-
sance attacks against transport and military
targets as the other extreme. In the aggre-
gate of casualties, therefore, women and child-
ren would not be a large element.
Casualties among women and children would come
from those too imprudent to stay away from re-
pair activity, or those so unfo unate as to be
present when our pilots have had'to dump
their bomb loads in order to avoid operational
disasters.
25X1
Approved For Wlease 2004/02/02 : CIA-RDP78S04149R000200060006-7
 25X1
? Approved For Rel
Casualties Resulting from the Bombing of North Vietnam
January-
1965 September 1966 Total
7,-6-00 ,9 0 17,900
Attacks on fixed targets 2,000 700
Attacks on armed recon-
naissance missions 4,000 11,200
2,700
15,200
Military 7,200 3,800 11,000
Attacks on fixed targets 4,300 400 4,700
Attacks on armed recon-
naissance missions 2,900 3,400 6,300
25X1
Approved For Release 2004/02/02 : CIA-RDP78SO2149RO90200060006-7
 Approve
EVIDENCE OF THE GENERAL ACCURACY OF US AIR STRIKES ON
NORTH VIETNAM
1. Post-strike photography provides good evi-
dence that US air strikes have generally been ac-
curate. There have been some occasions when ord-
nance was off target as a ,result of aircraft being
hit and misfiring or because of some other accident,
but the evidence indicates that such mishaps are in-
frequent. Eyewitness reports on the accuracy of US
air strikes on North Vietnam are hard to come by
2. Despite these limitations, there have been
25X1 have provided
evidence of the generally precise nature of the air
strikes as well as information indicating that ci-
vilian casualties from these air strikes have been
low.
3. The most recent eyewitness testimony of the
fact that US pilots make every effort to expend
their ordnance on target was provided by British
journalist Norman Barrymaine, who just completed a
three-week stay in the Haiphong area. He has stated
that US pilots "were very courageous" in pressing
home their attacks and in "not unloading their bombs
on civilians by design."
4. Other Westerners who have been in North Viet-
nam have volunteered similar observations. French
Press Agency (AFP) correspondent Jean Raffaelli, who
spent six months in North Vietnam last year, reported
25X1
25X1
25X1
25X1
Approved For Release 2004/02/02 : CIA-RDP78S02149R00020006g006-7
 Approved
on 14 October 1966 that "Haiphong and its suburbs
have been attacked a dozen times since June, but
there is no sign of damage in the city center."
25X1
5. There was general agreement among a variety
of sources that the US air strikes in late June 1966
against petroleum storage facilites in the Hanoi and
Haiphong areas were extremely accurate.]
25X1
25X1
25X1
Approved F
 Approved
8. Even the North Vietnamese on occasion have
admitted that civilian casualties from the air strikes
are limited. The North Vietnamese press representa-
tive in Phnom Penh, for example, told
25X1 that "the bombed areas o
Hanoi and Haiphong have been almost completely evac-
uated and consequently the number of victims has been
relatively low in relation to property damage."
25X1
25X1
25X1
Approved Igor Release 2004/02/02 : CIA-RDP78S02149R000200060g06-7
 25X1 Approved For Release 2004/02/02 : CIA-RDP78SO2149R000200060006-7
Next 1 Page(s) In Document Exempt
Approved For Release 2004/02/02 : CIA-RDP78SO2149R000200060006-7