time of the year, is to be expected more frequently during winter
months with January, February, and March probably being the months
of greatest expectation,

The upper-level cyclone situation appears

to have little, if any, dependence on season,

Because of its per=

sistence, up to ten days, it must always be taken into account in

planning operations in the northern Marshalls,
PART III
Meteorology in Relation to Atomic Tests,
It is probably safe to assume in planning future atomic tests

in the Marshall Islands that each of the three categories of weather
situation described in Part II will occur at least once during any
period longer than one month.

As already stated, there will be a

tendency for low-level cyclones to be more frequent in summer and

fall while the trade situation should be expected to occur with
highest frequency in summer and spring.

In the past, there has

been a tendency to assume in planning more complicated operations
that the wind systems aloft are associated only at random with ~
cloud and weather.

Prior to Operation CASTLE, for exemple, there

had been operational requirements that the winds over the Marshalls
‘up to 60,000 feet be, in the period following a detonation, from
the southeast or south,

At the same time, corollary air operations

had been predicated on the assumption that trade cumulus without middle or upper cloud would prevail over the entire Marshall Islands

area,

Experience indicates that this is stipulating an incompati-

ble distribution of weather elements.

L7L

It is true that such condi-

Select target paragraph3