UNCLASSIFIED
The average failing rate for a group of irregular-shaped particles
of a given size will be given by the equations. However, individual particles of the group may deviate from this average.
.
2.1.5
Marshali Isiands Atmosshere
Marsha’ !s'ands atmospheric conditions determined the values
for the density and viscosity parameters used in computing particle
" falling rates, A-ailabie data on the temperature, pressure, censity and
viscosity as functions of ait:tude for the atmosphere common to the
Marshaii island area in che spring months follow.
It was not possidle to use a “standard atmosphere" in this
problem because such use introduced a large error in the particle falling
rate at higs altitudes. This error originates primarily because an isothermai layer is assumed above the tropopause in the standard atmosphere —an unveal'stic assumption.
Temperature Dis
bution.
Therefore the profile cf the vertica1 temperature gradient
(Fig. 4) was based on measured dita to 67,000 #1 ard extrapolated to
120,000 ft on the basi¢ of scpporting climatological data and temperatu
re
measurements made at argh altitudes with rockets.
Pressere Distribution.
Published high alsitade measarements
- of the pressure distrivbatia: weze obtained on two occasions at Operation
CASTLE. These measurements,° made at Bikini on 7 April 1954 and
on 26 April 1954, were no: taker: above 65,000 ft. Abows thas altitade
the pressure sas extrapolated as a straight line on semi-log paper to
120,006 ft. Agreement with published rocket data from White Sands,
Mew Mexico was good to $9,000 ft (Fig. 5).
.
2 March 1954 0690 M Bik.ni
27 March 1954 0600 M Bikini
7 April 1954 0620 M Bikin:
26 April 1954 9619 M Bixini
No da.a were available above 67,000 ft. Fortunately two of these runs
pen-trated the tropopause whick waz located at approx:mately 55,000 ft,
To extend the measured data beyond 67,000 ft, climatological averages?
for latitude 12°N were emplovei. Agreement with measured data was
satisfactory except for the range from 50,000 to 25,000 it where the
climatologicai data indicated a well-defired isothermal layer.
The
most significant iinc.ng from the measured data was the complete lack
of an isothermal! Jayer above the tropopause. Instead, a distiact and
rapid inversion was observed which, when extrapolated as 2 straight
line, agreed with the cl.matologica_ data above 70,000 ft.
P
_ where the gas constant was tzken for dry air. The @Ssemztion
of no
moisture in the mixture introdccesan error of several percent in the
lower layers of the atmosphere where the relative homadity 1s high.
However, this assumption can de safely neglected. Also, the lazest
theories om the composition of the aimospaere indicate it to be constant
to altitudrs abare 150,060 {t which justified the asscmption of a noavarying gas constant.
Viscosity Distribzties., The variation of absoinze Viscosity with
altitade was compcted {rom the observed temperatcre
cistribatioa asirg
Sutherland's formula,*
.
= po
p= 0.01709
Since the
atmosphere was to be defined to 120,900 ft, further extrapolation was
Temperature data a-ailabte at these igher altitudes were
taken by rockets” over White Sands, New Mexico, A plot of three points
from the rocket data justified to some extent a continued extrapolation
of the curve to 129,000 ft. ,
The density distribation of the atmosphere
f° * RE
From the weather data published
by Task Force Weather Certral at Cperation CASTLE, four published
radiosonde runs obtained temperature measurements to high aititides:
necessary.
Density Distribztion.
(Fig, 6} was calculated from the perfect gas law using the aoove pressere
and temperature distrib:tions,
.
.:
To + 124
T
Te ble
To
2
(387.17\ foi
+ 116 }
.
afz
273.17
where tj = temperature in degrees Kelvin and p is viscosity ia centipoises. These data are plotted in Fig. i,
ntenen ene
-* Sq. 7.0. -13 Opecaiioa be mo No. 34. MS Agel 1868,
Fra ets os
-6-
ee
nee ee ee ee