CODES FOR CLOUD FORMS
AND STATES OF THE SKY

ACCORDING TO THE
INTERNATIONAL SYSTEM OF
CLASSIFICATION

PREAH ILY SLVS RST i ITO DB TANT ONT I ee a ee te ee on as PELLET
VAL A Hole ‘ At by i aco
: ‘i / m WIRE WO Fale tena e Ped ale Te a a it hast Rone ie =“
b OWWOOUS VIC Qceanozranhic SHON: |

OA Te De Pa Rd ONS
ATLAS GAZETTEER COLLECI iOf

RACER ELE HEPA AENEID AS Aad SSNS Oct Aa I see Sader oin arka ae esooewen

U. S. DEPARTMENT OF AGRICULTURE
WEATHER BUREAU
1938

Given in Loving Memory of

Raymond Braislin Montgomery
Scientist, R/V Atlantis maiden voyage
2 July - 26 August, 1931
KK KK KK
Woods Hole Oceanographic Institution
Physical Oceanographer
1940-1949
Non-Resident Statf
1950-1960
Visiting Committee
1962-1963
Corporation Member
1970-1980
KEK KKK
Faculty, New York University

1940-1944 :
Faculty, Brown University
1949-1954
Faculty, Johns Hopkins University
1954-1961
Professor of Oceanography,
Johns Hopkins University
1961-1975

Cireular S
W.B. No. 1249

i eS

INO

_ MBL/WHOI

CO

O 0301 00597448 j

U. 8. DEPARTMENT OF AGRICULTURE
WEATHER BUREAU

CODES FOR CLOUD FORMS

AND STATES OF THE SkY

ACCORDING TO THE
INTERNATIONAL SYSTEM OF
CLASSIFICATION

Prepared by the
WEATHER BUREAU COMMITTEE ON CLOUDS AND CLOUD FORMS:
BENJAMIN C. KADEL
JOSEPH B. KINCER
RICHMOND T. ZOCH

IvAN R. TANNEHILL, Chairman
HORACE R. BYERS
EDGAR B. CALVERT

UNITED STATES
GOVERNMENT PRINTING OFFICE
WASHINGTON: 19388

ee ———
Price 15 cents

For sale by the Superintendent of Documents, Washington, D. C.

CONTENTS

Page

Imbtrocwiebionee Rs fo 2C MSS Ree Se EE Gohl hes ee 5801 pee ee es III
iINamestand per giits: ofc] oul see ay se ee ee ee il
Classification: 2s. *S- Cae 8 ee ee Al ee eee 1
Definitions and descriptions of the forms of clouds____________________- 3
Girrset = 32 ts ol FE a ee ee Si Eee 9 3
Cirrocumulisuc. 25472 tae os Pees of ee ee ee 4
@irrostratuiseses. 2 2-2 gee be ee ee ee 4
Altocumulus. = 2 22) ee ew es Be 5
AITOStTAUUS A= es Soe So oe ee SE ee ee a
Stravocumawluis ess 52. 8 Ses ee ee ee ee 8
DtratUse.=S-= 2. Un Pee ee GU ee ee ee 10
INimbostratus=. “22. S452 sa. 2 ees Se re 10
@umuluss 2 25 SE See ea ke ee iil
Cumulonimbus 2s = 27. a a ee ee ee 172
Varietiesand details 246 . oS42 5 3. eek) oe eee 13
Hmnitrygo feeloma sromynre Orci Or Ses ee ee en ne 15
Abbreviations: 22452 2-6 aoe ai a ee 2 ee 15
Amount ana Gdinechiony Ofanol ome sa = 5m 16
Observation’ of cloudsitor codedimessages!—- 22 ss 222 2 7
Codertables==s2 22 Ss 55 so ee ee oe ee 19
The state of the sky and the evolution of the clouds___________________ 21
Codeispecifications 22 222 sa eee een eae le ee ee 22
Wowercloudsi==s2 St ee et ee oe ees en oe a ee 22
Middie*cloudsss 3-2 24 Ses oo Se See ee 25
Wipper clouds= 3422.42 47S eae See ee es 29
Keyatodentitication of cloudtcodertypess= ee = saa ae ee ee 32
Skyzandycloudspimyarstyjorc alaGis Gusto am Cees eee ey ee ee 35
Photosraphs wathed escripe G1 om sass se ee ee ee 36
Neri alevdews) OlGloudse === a= ae Cpe a apt ee 225) 90

INTRODUCTION

At Copenhagen, Denmark, in 1929, the International Meteorologi-
eal Organization adopted code specifications for clouds including
lower, middle, upper, and predominating clouds. The third edition
of the International Atlas of Clouds and of the States of the Sky which
was published in 1932 and the abridged edition which was published
in 1930, contain detailed instructions for observing and coding cloud
forms in accordance with the specifications adopted at Copenhagen.
The cloud codes form a part of the International figure code for syn-
optic messages.

Definitions, descriptions, and_ codes in the following pages are
based on the third edition of the International Atlas and are intended
for the use of observers who report clouds in synoptic messages in
the International Code. This publication is not designed for public
distribution. The cloud codes are of little interest to the general
public. The Weather Bureau has prepared a separate publication
entitled “Cloud Forms,’’ which is especially designed for general
use in the identification and study of the forms of clouds.

Observers who have access to the International Atlas should also
study the illustrations, diagrams, and comments contained therein.

The observer will note the following details in which International
practices differ from those heretofore in effect in the Weather Bureau:

1. The definition of cirrocumulus clouds has been restricted more
than in previous editions of the International Atlas. Real cirrocu-
mulus is now considered to be uncommon; it must have an evident
connection with cirrus or cirrostratus; it must result from a change
in cirrus or cirrostratus; and it must show some characteristics of
ice crystal clouds.

2. The distinction between cumulus and cumulonimbus has been
modified. By the new definitions, masses of cumulus clouds, however
heavy they may be, and however creat their vertical development, are
not classed as cumulonimbus unless the whole or a part of their tops is
transformed or is in process of transformation Into a cirrus mass.
In the species cumulonimbus calvus, however, no cirriform parts can
be detected at first but nevertheless the freezing of the upper parts
has already begun and the tops are beginning to lose their rounded
outlines and clear-cut contours; the hard and ‘cauliflower’ swellings
soon become confused and melt away so that nothing can be seen
in the white mass but more or less vertical fibers. Hence, the cri-
terion for distinction in cases difficult to identify is the fibrous summit
that must be present if the cloud is to be called cumulonimbus with-
out having the clearly defined cirriform top. Precipitation may fall
from cumulus clouds, though only in relatively small quantities.

3. Altocumulus clouds are now distinguished from stratocumulus
by the specific criterion that the well-defined, regularly arranged
elements which are observed in the layer are not altocumulus if they
are greater than 10 solar diameters in their smallest diameters, 1. e.,
the apparent width of three fingers when the arm is held extended.

(III)

4. Dense sheets of middle-level nonfibrous clouds, heretofore called
altostratus, have been changed to altocumulus opacus. This change
restricts the dense altostratus forms to the dense fibrous sheets of
rain (sometimes snow) that usually succeed cirrostratus by lowering,
a development that is of much importance to the forecasters. If a
dense sheet of middle-level clouds shows definite relief on the lower
surface it is classed as altocumulus and not altostratus.

5. The name ‘nimbostratus” has been introduced and the name
‘nimbus” has been eliminated. The observer should read the
definitions carefully and note particularly that the name ‘‘nimbo-
stratus”? should not be applied to the lower parts of a cumulonimbus
cloud. When precipitation falls from nimbostratus clouds it is con-
tinuous in character. Nimbostratus is evolved by lowering of a
sheet of altostratus or in exceptional cases from a sheet of strato-
cumulus but not from cumulonimbus.

6. The hyphens in cloud names have been eliminated, e. g., cirro-
stratus has been changed to cirrostratus.

Vi

NAMES AND HEIGHTS OF CLOUDS

Clouds are classified according to form and appearance, but con-
sideration is also given to the physical processes producing them.
There is a general relation between the forms of clouds and their
heights as shown by actual measurements. The temperature,
moisture, and movements of the air differ characteristically at the
various levels above the earth’s surface and hence there are significant
differences in the formation of clouds as well as the apparent effects
of distance or height above the observer.

In the International system there are 4 families and 10 genera,
and in addition certain species, varieties, and special features. The
10 genera and their heights, as given in the International Atlas, are
shown in the following table, where each word preceded by a number
is the name of a cloud form which constitutes a distinct genus. The
mean heights are for temperate latitudes and refer not to sea level
but to the general level of land in the region. There is nearly always
some variation from the mean height, which serves merely as a guide
to the general elevation. In certain cases there may be large de-
partures from the mean height.

Cirrus clouds sometimes may be observed as low as 3,000 meters
in temperate regions and at lower levels in higher latitudes. In the
polar regions, and in extremely cold weather elsewhere, ice spicules
are occasionally observed in the air at and near the surface.
When ice crystals or spicules are seen floating in the air at the surface,
the phenomenon is recorded as ice fog or ice crystals, depending on
visibility and other criteria for identifying hydrometeors.

Clouds composed at least partially of water droplets can exist at
very low temperatures; sometimes low clouds are composed of a
mixture of water droplets and ice crystals or snow. The observer
must be guided by the appearance of the clouds. If they have the
usual appearance of cirrus forms, they should be so recorded and
reported in regular synoptic messages. Clouds at very low levels
which have the appearance of ordinary stratus clouds should be
recorded and reported as such, even when upper air soundings show
that they are made up of snow crystals, ice spicules, or a mixture of
either or both with water droplets. In special cases an explanatory
note should be entered in the record to show the composition of the
clouds as revealed by airplane soundings. In this connection the
observer is referred to the section, ‘‘Aerial views of clouds.”’

CLASSIFICATION '
Family A: High Clouds

(Mean lower level, 6,000 meters, 20,000 feet)

1. Cirrus.
2. Cirrocumulus.
3. Cirrostratus.

1 Species, varieties, and special features are described in the sections which follow.

1

Family B: Middle Clouds

(Mean upper level, 6,000 meters, 20,000 feet; mean lower level, 2,000 meters,
6,500 feet)

4. Altocumulus.
5. Altostratus.

Family C: Low Clouds
(Mean upper level, 2,000 meters, 6,500 feet; mean lower level, close to the surface)

6. Stratocumulus.
7. Stratus.
8. Nimbostratus.

Family D: Clouds With Vertical Development
(Mean upper level, that of cirrus; mean lower level, 500 meters, 1,600 feet)

9. Cumulus.
10. Cumulonimbus.

DEFINITIONS AND DESCRIPTIONS OF THE
FORMS OF CLOUDS

CIRRUS
Gil, 42) 43) 44,45, 46, 51, 52)

A. Definition

Detached clouds of delicate and fibrous appearance, without shad-
ing,” generally white in color, often of a silky appearance.

Cirrus appears in the most varied forms, such as isolated tufts,
lines drawn across a blue sky, branching feather-like plumes, curved
lines ending in tufts, etc.; - they are often arranged in bands which
cross the sky like meridian lines, and which, owing to the effect of
perspective, converge to a point on the horizon, or to two opposite
points (cirrostratus and cirrocumulus often take part in the formation
of these bands).

B. Explanatory Remarks

Cirrus clouds are always composed of ice crystals, and their trans-
parent character depends upon the degree of separation of the
crystals.

As a rule when these clouds cross the sun’s disk they hardly di-
minish its brightness. But when they are exceptionally thick they
may veil its light and obliterate its contour. This would also be the
case with patches of altostratus, but cirrus is distinguished by the
dazzling and silky whiteness of its edges.

Halos are rather rare in cirrus.

Sometimes isolated wisps of snow are seen against the blue sky,
and resemble cirrus; they are of a less pure white and less silky than
cirrus; wisps of rain are definitely gray, and a rainbow, should one be
visible, shows their nature at once, for this cannot be produced in
cirrus.

Before sunrise and after sunset, cirrus is often colored bright yellow
orred. These clouds are lit up long before other clouds and fade out
much later; some time after sunset they become gray. At all hours
of the day cirrus near the horizon is often of a yellowish color; this is
gue a distance and to the great thickness of air traversed by the rays
of hght.

Cirrus, being in general more or less inclined to the horizontal,
tends less than other clouds to become parallel to the horizon, under
the effect. of perspective, as the horizon is approached; often on the
contrary it seems to converge to a point on the horizon.

1 Numbers inserted in parentheses in connection with a cloud name refer to the pictures in which that
cloud is illustrated.
2 Dense cirrus derived from an anvil (cirrus nothus) may have shading.

3

C. Species

Among the principal forms one may note:

Cirrus filosus (42).—More or less straight or irregularly curved
filaments (neither tufts nor little hooks and without any of the parts
being fused together).

Cirrus uncinus (45, 52).—Cirrus in the shape of a comma, the
upper part either ends in a little tuft or is pointed.

Cirrus densus (43).—Cirrus clouds with such thickness that without
care an observer might mistake them for middle or low clouds.

Cirrus nothus (44).—Cirrus proceeding from a cumulonimbus and
composed of the debris of the upper frozen parts of these clouds.

D. Varieties

Ordinary cirrus may appear in many very different forms. One
may particularly note the forms jfloccus and vertebratus which are
really aspects of the varieties cumuliformis and undulatus radiatus,
respectively.

CIRROCUMULUS

(49, 51, 52)
A. Definition

A cirriform layer or patch composed of small white flakes or of very
small globular masses, usually without shadows, which are arranged in
eroups or lines, or more often in ripples resembling those of the sand
on the seashore.

B. Explanatory Remarks

In general cirrocumulus represents a degraded state of cirrus and
cirrostratus both of which may change into it. In this case the
changing patches often retain some fibrous structure in places.

Real cirrocumulus is uncommon. It must not be confused with
small altocumulus on the edges of altocumulus sheets. There are in
fact all states of transition between cirrocumulus and altocumulus
proper; this is only to be expected as the process of formation is the
same. In the absence of any other criterion the term cirrocumulus
should only be used when—

1. There is evident connection with cirrus or cirrostratus.

2. The cloud observed results from a change in cirrus or cirro-
stratus.

3. The cloud observed shows some of the characteristics of ice
crystal clouds which will be found enumerated under cirrus.

Clear rifts are often seen in a sheet of cirrocumulus.

CIRROSTRATUS
(46, 47, 48, 49, 50)
A. Definition

A thin whitish veil which does not blur the outlines of the sun or
moon, but usually gives rise to halos. Sometimes it is quite diffuse

+

and merely gives the sky a milky look; sometimes it more or less dis-
tinctly shows a fibrous structure with disordered filaments.

B. Explanatory Remarks

A sheet of cirrostratus which is very extensive, though in places
it may be interrupted by rifts, nearly always ends by covering the
whole sky. The border of the sheet may be straight edged and clear-
cut but more often it is ragged or cut up.

During the day, when the sun is sufficiently high above the horizon,
the sheet is never thick enough to prevent shadows of objects on the
ground.

A milky veil of fog (or thin stratus) is distinguished from a veil of
cirrostratus of a similar appearance by the halo phenomena which
the sun or the moon nearly always produces in a layer of cirrostratus.

The following are the principal halo phenomena: A circle of 22°
radius round the sun or moon; this is roughly the angle subtended by
the hand placed at right angles to the arm when the latter is extended;
this halo is sometimes, but rarely, accompanied by one of 46° radius.
Parhelia, paraselenae (mock suns or mock moons), luminous patches,
often showing prismatic colors, a little over 22° from the sun or moon
and at the same elevation. A luminous column, e. g., sun pillar,
extending vertically above and below the luminary.

Often only small fragments of these appearances are visible but
they are none the less characteristic of high clouds.

What has been said above of the transparent character and colors
of cirrus is true to a great extent of cirrostratus.

C. Species

Cirrostratus has two principal aspects which correspond to the
two following species:

Cirrostratus nebulosus (48, 50).—A very uniform nebulous veil,
sometimes very thin and hardly visible, sometimes relatively dense,
but always without definite details and usuaily with halo phenomena.

Cirrostratus filosus (47)—A white fibrous veil, where the strands
are more or less definite, often resembling a sheet of cirrus densus
from which indeed it may originate.

ALTOCUMULUS
(28 to 38, inclusive)
A. Definition

A layer (or patches) composed of laminae or rather flattened
globular masses, the smallest elements of the regularly arranged
layer being fairly small and thin, with or without shading.? T hese
elements are arranged in groups, in lines, or waves, following one or
two directions, and are sometimes so close together that their edges
join.

The thin and translucent edges of the elements often show irisations
which are rather characteristic of this class of cloud.

From the definition it follows that altocumulus comprises the sub-
genera:

3 See p. IV for distinction between dense sheets of altocumulus and altostratus.

5

Altocumulus translucidus (28, 29, 32)—Altocumulus formed of
elements whose color—from dazzling white to dark gray—and whose
thickness vary much from one example to another, or even in the
same layer; the elements are more or less regularly arranged and
distinct. In the definition of the elements it is the variation in the
transparency of the layer, variable from one point to another, that
plays the essential part. "There appears in the interstices either the
blue of the sky, or at least a marked lightening of the layer of cloud
due to a thinning out.

Altocumulus opacus (35, 36).—An altocumulus sheet which is
continuous, at least over the ereater part of the layer, and consisting
of dark and more or less ir regular elements, in the definition of which
transparency does not play a great part, owing to the thickness and
density of the layer; but the elements show in real relief on the lower
surface of the cloud sheet.

B. Explanatory Remarks

The limits within which altocumulus is met are very wide.

At the greatest heights, altocumulus made up of small elements
resembles cirrocumulus; altocumulus, however, is distinguished by
not possessing any of the following characters of cirrocumulus:

1. Connection with cirrus or cirrostratus.

2. An evolution from cirrus or cirrostratus.

3. Properties due to physical structure (ice crystals) enumerated
under cirrus.

At lower levels, where altocumulus may be derived from a spreading
out of the tops of cumulus clouds, it may easily be mistaken for
stratocumulus; the convention is that the cloud is altocumulus if the
smallest, well defined, and regularly arranged elements which are
observed in the layer (leaving out the detached elements which are
generally seen on the edges) are not greater than 10 solar diameters
in their smallest diameters, 1. e., approximately the width of three
fingers when the arm is held extended.

When the edge or a thin semitransparent patch of altocumulus
passes in front of the sun or moon a corona appears close up to (within
a few degrees of) them; this is a colored ring with red outside and blue
inside; the colors may be repeated more than once. ‘This phenomenon
is infrequent in the case of cirrocumulus and only the higher forms of
stratocumulus show it.

Irisation, mentioned above, is a phenomenon of the same type as
the corona; it is characteristic of altocumulus as distinguished from
cirrocumulus or stratocumulus.

Altocumulus clouds often appear at different levels at one and the
same time. Often, too, they are associated with other types of cloud.

The atmosphere is often hazy just below altocumulus clouds.

When the elements of a sheet of altocumulus fuse together and
make a continuous layer altostratus or nimbostratus is the result.
On the other hand a sheet of altostratus can change into altocumulus.
It may happen that these two aspects of a cloud sheet may alternate
with each other during the whole course of a day. It is also not rare
to have a layer of altocumulus coexisting with a veil resembling alto-
stratus at a height very little less than the altocumulus.

It is interesting to note that one may often observe filiform descend-
ing trails to which the name virga has been given.

6

C. Species

Among the principal kinds one may note:

Altocumulus cumulogenitus (33).—This is an altocumulus cloud
formed by the spreading out of the tops of cumulus, the lower parts
of the cumulus clouds having melted away; the layer in the first stages
of its growth has the appearance of altocumulus opacus.

D. Varieties

An important variety of altocumulus should be noted, namely,
altocumulus cumuliformis, which has two different aspects:

Altocumulus floccus (88).—Tufts resembling small cumulus clouds
without a base and more or less ragged.

Altocumulus castellatus (37)—Cumuliform masses with more or
less vertical development, arranged in a line, and resting on
a common horizontal base, which gives the cloud a crenellated
appearance.

The caps or hoods which form above a cumulus by the uplift of a
damp layer, and which may be pierced by the tops of the cumulus
are considered as a detail of cumulus, and denoted by the term pileus
attached to the name cumulus; but in reality they are aberrant forms
of altocumulus translucidus._ Moreover, similar clouds, independent
of cumulus, can be formed by the same process by the effect of a
rising current caused by a mountain or any obstacle. They are then
named altocumulus, and they are classed, on account of their form,
with the variety lenticularis (30).

ALTOSTRATUS
(UE, A, AG, Bel, 3%)
A. Definition

Striated or fibrous veil, more or less gray or bluish in color. This
cloud is like thick cirrostratus but without halo phenomena; the sun
or moon shows vaguely, with a faint gleam, as though through ground
glass. Sometimes the sheet is thin, with forms intermediate with
cirrostratus. Sometimes it is very thick and dark, sometimes even
completely hiding the sun or moon. In this case differences of thick-
ness may cause relatively ight patches between very dark parts; but
the surface never shows real relief,* and the striated or fibrous struc-
ture is always seen in places in the body of the cloud.

Every form is observed between high altostratus and cirrostratus
on the one hand, and low altostratus and nimbostratus on the other.

Rain or snow may fall from altostratus (altostratus precipitans), but
when the rain is heavy the cloud layer will have grown thicker and
lower, becoming nimbostratus; but heavy snow may fall from a layer
that is definitely altostratus.

From the definition of altostratus it follows that there are three
subgenera:

1. Altostratus translucidus (15, 25)—A sheet of altostratus re-
sembling thick cirrostratus; the sun and the moon show as through
eround glass.

4 See p. IV for distinction between dense sheets of altocumulus and altostratus.

G

Altostratus opacus (26, 27).—An opaque layer of altostratus of
waite thickness which may entirely hide the sun, at any rate, in
parts, but showing a fibrous structure in some parts.

3. Altostratus precipitans—A layer of opaque altostratus which
has not yet lost its fibrous character, and from which there are light
falls of rain or snow, either continuous or intermittent. This precipi-
tation may not reach the ground in which case it forms virga.

B. Explanatory Remarks

The limits between which altostratus may be met with are fairly
wide (about 5,000 to 2,000 meters).

A sheet of high altostratus is distinguished from a rather similar
sheet of cirrostratus by the convention that halo phenomena are not
seen in altostratus, nor are the shadows of objects on the ground
visible.

A sheet of low altostratus may be distinguished from a somewhat
similar sheet of nimbostratus by the following characters: Nimbo-
stratus is of a much darker and more uniform gray, and shows nowhere
any whitish gleam or fibrous structure; one cannot definitely see the
limit of its undersurface which has a wet look, due to the rain (or snow),
which may not reach the ground.

The convention is also made that nimbostratus always hides the
sun and moon in every part of it, while altostratus only hides them in
places behind its darker portions, but they reappear through the lighter
parts.

Careful observation may often detect virga hanging from altostratus,
and these may even reach the ground causing slight precipitation. If
the sheet still has the character of altostratus it will then be called
altostratus precipitans, but if not it has become nimbostratus.

A sheet of altostratus, even if it has rifts in places, has a general
fibrous (or ground glass) character. A cloud layer, even a continuous
one, which has no fibrous structure, and in which rounded cloud
masses may be seen is classed as altocumulus or stratocumulus accord-
ing to circumstances.

Altostratus may result from a transformation of a sheet of alto-
cumulus, and on the other hand altostratus may often break up into
altocumulus.

STRATOCUMULUS
(G10, 11: 12; 18,.19;-20,-211,.98)
A. Definitions

A layer (or patches) composed of laminae, globular masses or rolls;
the smallest of the regularly arranged elements are fairly large; they
are soft and gray, with darker parts.°

These elements are arranged in groups, in lines, or in waves, alined
in one or in two directions. Very often the rolls are so close that their
edges join; when they cover the whole sky they have a wavy appearance,
z From the definition it follows that stratocumulus comprises two
<inds:

5 Also, a low, continuous sheet, thick or thin, with distinct irregularities of large size. See explanatory
remarks under ‘‘Stratus.”’

Stratocumulus translucidus (12).—A not very thick layer; in the
Pie sie: between its elements either the blue sky appears, or at any
rate there are much lighter parts of the cloud sheet, which here is
thinned out on its upper surface.

2. Stratocumulus opacus (21).—A very thick layer made up of a con-
tinuous sheet of large dark rolls or rounded masses; their shape is
seen not by a difference in transparency, but they stand out in real
relief from the under surface of the cloud layer.

There are transitional forms between stratocumulus and _ alto-
cumulus on the one hand and between stratocumulus and stratus
on the other.

B. Explanatory Remarks

The difference between stratocumulus and altocumulus is given
under the latter.

It should also be noted that the cloud sheet called altocumulus by
an observer at a small height might appear as stratocumulus to an
observer at a greater height (12).

It often happens that stratocumulus is not associated with any
clouds of the second or third families; but it fairly often coexists with
clouds of the fourth family.

The elements of thick Lied omaailis (stratocumulus opacus) often
tend to fuse together completely, and the layer can, in certain cases,
change into nimbostratus. The cloud is called nimbostratus when
the cloud elements of stratocumulus have completely disappeared
and when, owing to the trails of falling precipitation, the lower surface
has no longer a clear cut boundary.

Stratecumulus can change into stratus, and vice versa. The
stratus being lower, the elements appear very large and very soft,
so that the structure of regularly arranged globular masses and waves
disappears as far as the observer can see. The cloud will be called
stratocumulus as long as the structure remains visible.

C. Species

Among the principal species may be mentioned:

1. Stratocumulus vesperalis (9)—This name is given to flat elon-
gated clouds which are often seen to form about sunset as the final
product of the diurnal changes of cumulus.

2. Stratocumulus cumulogenitus (10).—Stratocumulus formed by
the spreading out of the tops of cumulus clouds, which latter have
disappeared; the layer in the early stages of its formation looks like
stratocwmulus opacus.

D. Varieties

The cloud called roll cumulus m England and Germany is desig-
nated stratocumulus undulatus (11); its wave system is in one direction
only. It must not be confused with flat cumulus clouds ranged in
line. Stratocumulus often has a mammatus (festooned) character;
that is to say there is a high relief on the lower surface where pendant
rounded masses or cor rugations are observed, and at times these look
as though they would become detached from the cloud. Care must
be taken not confuse this cloud with some kinds of altostratus opacus
whose under surface may appear to be slightly corrugated or mam-
millated ; a/tostratus opacus is distinguished by its fibrous structure.

9

STRATUS
(13, 14)
A. Definition

A low uniform layer of cloud, resembling fog, but not resting on the
eround.

Fractostratus.—When this very low layer is broken up into irregular
shreds it is designated fractostratus (13, 25).

B. Explanatory Remarks

A veil of true stratus generally gives the sky a hazy appearance
which is very characteristic, but which in certain cases may cause
confusion with nimbostratus. When there is precipitation the dif-
ference is manifest: nimbostratus gives continuous rain (sometimes
snow), precipitation composed of drops which may be small and sparse,
or else large (at least some of them) and close together, while stratus
only gives a drizzle, that is to say, small drops very close together.

When there is no precipitation a dark and uniform layer of stratus
can easily be mistaken for nimbostratus. ‘The lower surface of nimbo-
stratus how ever has always a wet appearance (widespread trailing
precipitation, ‘‘virga’’); it is quite uniform and it is not possible to
make out definite detail; stratus on the other hand has a drier appear-
ance, and however uniform it may be it shows some contrasts and
some lighter transparent parts, that is, places less dark where the cloud
is thinner, corresponding to the interstices between the rolls and
globular masses of stratocumulus, but considerably larger, while
nimbostratus seems only to be feebly illuminated, as though lt up
from within.

Stratus is often a local cloud, and when it breaks up the blue sky
is seen.

Fractostratus sometimes originates from the breaking up of a layer
of stratus, sometimes it forms independently and develops till it forms
a layer below altostratus or nimbostratus, which latter may be seen
in the interstices.

A layer of fractostratus may be distinguished from nimbostratus
by its darker appearance, and by being broken up into cloud elements.
If these elements have a cumuliform appearance in places the cloud
layer is called fractocumulus and not fractostratus.

NIMBOSTRATUS
(iG. LA)
A. Definition

A low, amorphous, and rainy layer, of a dark gray color, usually nearly
uniform : feebly illuminated seemingly from inside. When it gives
precipitation it is in the form of continuous rain or snow.

But precipitation alone is not a sufficient criterion to distinguish
the cloud which should be called nimbostratus even when no rain or
snow falls from it.

There is often precipitation which does not reach the ground;
in this case the base of the cloud is usually diffuse and looks wet on

10

account of the general trailing precipitation, virga, so that it is not
possible to determine the limit of its lower surface.

B. Explanatory Remarks

The usual evolution is as follows: A layer of altostratus grows thicker
and lower until it becomes a layer of nimbostratus. Beneath the latter
there is generally a progressive development of very low ragged clouds,
isolated at first, then fusing together into an almost continuous layer,
in the interstices of which, however, the nimbostratus can generally
be seen. These very low clouds are called fractocumulus or fracto-
stratus according as to whether they appear more or less cumuliform
or stratiform.

Generally the rain only falls after the formation of these very low
clouds, which are then hidden by the precipitation or may even melt
away under its action. The vertical visibility then becomes very bad.

In certain cases the precipitation may precede the formation of
fractocumulus or fractostratus, or it may happen that these clouds do
not form at all.

Rather rarely a sheet of nimbostratus may form by an evolution
from a stratocumulus.

CUMULUS
Cl, 2, B; 4 18, 19, 20, 2B, 3B, Br, 24)
A. Definition

Dense clouds with vertical development; the upper surface is dome
shaped and exhibits rounded protuberances, while the base is nearly
horizontal.

When the cloud is opposite the sun the surfaces normal to the
observer are brighter than the edges of the protuberances. When the
light comes from the side, the clouds exhibit strong contrasts of light
and shade; against the sun, on the other hand, they look dark with a
bright edge.

Fractocumulus—True cumulus is definitely limited above and
below, its surface often appears hard and clear cut. But one may also
observe a cloud resembling ragged cumulus in which the different
parts show constant change. This cloud is designated fractocumulus
(®).

B. Explanatory Remarks

Typical cumulus, over land areas, develops on days of clear skies,
and is due to the currents of diurnal convection; it appears in the
morning, grows, and, then more or less dissolves again toward the
evening.

Cumulus, whose base is generally of a gray color, has a uniform
structure, that is to say it is composed of rounded parts right up to its
summit, with no fibrous structure. Even when highly developed,
cumulus can only produce light precipitation.

Cumulus, when it reaches the altocumulus level, is sometimes capped
with a light, diffuse, and white veil of more or less lenticular shape,
with a delicate striated or flaky structure on its edges; it is generally
shaped like a bow which may cover several domes of the cumulus, and

itt

finally be pierced by them. This cloud which does not constitute a
species is given the name of pileus, a cap or hood.
The clouds which form below altostratus or nimbostratus and which
can develop into a complete layer, through whose interstices the alto-
stratus or nimbostratus is oenerally seen, are usually fractostratus; but
if they have a cumulform appearance they should be classed as
fractocumulus. They rarely have this appearance during or soon
after rain; on the other hand it is frequent at the beginning of the
formation of the low cloud, and alien it breaks up.

C. Species

Among the principal species one may note:

1. Cumulus humilis (1)—Cumulus with little vertical development,
and seemingly flattened. These clouds are generally seen in fine
weather.

2. Cumulus congestus (3, 4).—Very distended and_ sprouting
cumulus, whose domes have a cauliflower appearance.

CUMULONIMBUS
(5, 6, 7, 8, 21, 22, 24)
A. Definition

Heavy masses of cloud, with great vertical development, whose
cumuliform summits rise in the form of mountains or towers, the
upper parts having a fibrous texture and often spreading out in the
shape of an anvil.

The base resembles nimbostratus, and one generally notices virga.
This base has often a layer of very low ragged clouds below it (frac-
tostratus, fractocumulus).

Cumulonimbus clouds generally produce showers of rain or snow
and sometimes of hail, and often thunderstorms as well.

If the whole of the cloud cannot be seen the fall of a real shower is
enough to characterize the cloud as a cumulonimbus.

B. Explanatory Remarks

Even if a cumulonimbus were not distinguished by its shape from a
strongly developed cumulus its essential character is evident in the
difference of structure of its upper parts, when these are visible (fibrous
structure and cumuliform structure). Masses of cumulus, however
heavy they may be, and however great their vertical development,
should never be classed as cumulonimbus unless the whole or a part of
their tops is transformed or 1s in process of transformation into a cirrus
mass.

Although the upper cirriform parts of a cumulonimbus may take on
very VE ried shapes, yet in certain cases they spread out into the form
ofananvil. To this interesting feature the name incus is given.

In certain types of cumulonimbus, which are especially common in
spring in moderately high latitudes, the fibrous structure extends to
nearly the whole cloud mass, so that the cumuliform parts almost
wholly disappear; the cloud is reduced to a mass of cirrus and of virga.

The veil cloud pileus is seen with cumulonimbus clouds as with
cumulus.

When a cumulonimbus covers nearly all the sky the base alone is
visible, and resembles nimbostratus, with or without fractostratus or
fractocumulus below. The difference between the base of a cumu-
lonimbus and a nimbostratus is often rather difficult to make out. If
the cloud mass does not cover all the sky, and if even small portions
of the upper parts of the cumulonimbus appear, the difference is
evident. If not it can only be made out if the preceding evolution of
the clouds has been followed, or if precipitation occurs; its character
is violent and intermittent (showers) i in the case of cumulonimbus, as
opposed to the relatively gentle and continuous precipitation of a
nimbostratus.

The front of a thunder cloud of great extent is sometimes accom-
panied by a roll cloud of a dark color in the shape of an arch, of a
frayed out appearance, and circumscribing a part of the sky of a
lighter gray. This cloud is named arcus and is nothing more or less
than a particular case of fractocumulus or fractostratus.

Fairly often a mammatus structure appears in cumulonimbus,
ence at the base, or on the lower surface of the lateral parts of the
anvi

When a layer of menacing cloud covers the sky and virga and mam-
matus structure are both seen it is a sure sign “that the cloud is the
base of a cumulonimbus, even in the absence ‘of all other signs.

Cumulonimbus is a real factory of clouds: it is responsible in great
measure for the clouds in the rear of disturbances. By the spreading
out of the more or less high parts and the melting away of the under-
lying parts, cumulonimbus can produce more or less thick sheets of
altocumulus or stratocumulus (spreading out of the cumuliform
parts) and dense cirrus (spreading out of the cirriform parts).

C. Species

Among the principal species may be noted:

1. Cumulonimbus calvus (5, 6)—Cumulonimbus characterized by
the thunderstorm or the shower that it causes, or by virga, but in
which no cirriform parts can be made out. Nevertheless the freezing
of the upper parts has already begun; the tops are beginning to lose
their cumulus structure, that is to say, their rounded outlines and
clear cut contours; the hard and “cauliflower” swellings soon become
confused and melt: away so that nothing can be seen in the white mass
but more or less vertical fibers. The freezing, accompanied by the
change into a fibrous structure, often goes on very rapidly.

2. Cumulonimbus capillatus (7)—Cumulonimbus which displays
distinct cirriform parts, having sometimes, but not always, the shape

of an anvil.
PRINCIPAL VARIETIES

The chief varieties common to different genera are as follows:

1. Fumulus.—At all levels, from cirrus “to stratus, a very thin veil
may form, so delicate that it may be almost invisible. These veils
seem to be most frequent on hot days, and in low latitudes. Occa-
sionally they may be observed to thicken rapidly, forming clouds
easily visible, especially cirrus and cumulus. The clouds thus pro-
duced seem unstable however, and usually melt away soon after their
formation.

98863°—38 13

Cirrus fumulus must not be confused with cirrostratus nebulosus.
The latter is much more stable and does not show the phenomenon of
the formation and subsequent rapid disappearance of cirrus clouds.

2. Lenticularis (30).—Clouds of an ovoid shape, with clean-cut
edges, and sometimes irisations, especially common on days of strong,
dry winds in rough country. This form exists at all levels from cir-
rostratus to stratus.

3. Oumuliformis——The rounded form resembling cumulus which
the upper parts of other clouds may sometimes assume. This may be
seen at all levels from cirrus to stratus.

4. Mammatus.—This description is given to all clouds whose lower
surfaces form pouches or breasts. This form is found especially in
stratocumulus and in cumulonimbus, either at the base, or even more
often on the lower surface of anvil projections. It is also found,
though rarely, in cirrus clouds, probably when they have originated
in the anvil of a dispersing cumulonimbus.

5. Undulatus.—This term is applied to clouds composed of elongated
and parallel elements, like waves of the sea. There is sometimes an
appearance of two distinct systems, as when the cloud is divided into
rounded masses by undulations in two directions.

6. Radiatus.—This term is applied to clouds in parallel bands (polar
bands), which owing to perspective seem to converge to a point on the
horizon, or to two opposite points if the bands cross the whole sky. The
point is called the radiant point, or vanishing point.

CHIEF CASUAL DETAILS

The chief casual varieties are the following:

1. Virga: Wisps or falling trails of precipitation; applied princi-
pally to altocumulus and altostratus.

2. Pileus: A cap or hood; applied principally to cumulus or cumu-
lonimbus.

3. Incus: Anvil; upper part of cumulonimbus.

4. Arcus: Arch cloud; usually associated with cumulonimbus.

14

ENTRY OF CLOUDS ON RECORD FORMS

The proper method of entering cloud observations on record forms,
including abbreviations, are described in the instructions accompany-
ing the forms.

Abbreviations.—Abbreviations in the International Atlas are of the
form shown by the following examples: Cist., Acu., Stcu., and Cunb.,
etc. These merely constituted a simplification of the abbreviations
previously used, for example, Ci.-St., A.-Cu., St.-Cu., and Cu.-Nb.,
etc. The International Meteorological Organization in a meeting at
Warsaw in 1935 adopted shorter abbreviations which supersede those
shown in the International Atlas. The Warsaw abbreviations are as
follows:

(Giinsrsi1s ea ise foo odse (Cil || Starhrocunmultig. 2 Se
Cirrocumuluss2 222252 20s ee (OG Smee Re OL RS Rees eee ese aes St
@irrostratuss 22-2. joe ee CsieNimibostratise= == ae eer Ns
ANooumnmllig 2 oe Aci Cumuluseeen = tee ee ee eee Cu
Altostratus: 2-22-25. 22. 23050 22 AS) |: Cumulonimbuste2==252 5520 sae Cb

The Warsaw abbreviations and, in addition, the following four
special abbreviations have been adopted by the Weather Bureau,
effective January 1, 1939:

OrACTOSULA TUS mee se ee Fs | Cumulonimbus mammatus (Mam-
Fractocumulus__________________ Fe TNA GON CUMAUL US) eee Cm
Altocumulus castellatus____.____- Acc |

15

AMOUNT AND DIRECTION OF MOTION

An observation of the state of the sky should also contain an esti-
mate of the cloudiness; both the total cloud amount, that is the
amount of the sky in tenth parts covered by all the clouds, and the
partial cloud amounts for each genus of cloud.

In estimating the cloud amount it is best to ignore the part of the
sky close to the horizon, for there the clouds seem to be packed
together owing to perspective, so that the cloud amount seems greater
than it really is. When the amount of cloud is large it is easier to
estimate the amount of the sky that is free from cloud, and from this,
by subtraction, to obtain the cloud amount.

In entering cloud araounts in record forms the observer will be
guided by instructions printed on the forms or special instructions
accompanying them.

The direction of motion of clouds may often be easily observed by
sighting on a steeple, tower, pole, or other structure in an open place.
If the motion of the cloud be very slow, the head should be steadied
by using a rest. This method is readily applied only to clouds that
are well above the horizon, preferably near the zenith. When a
nephoscope is available it should be used at or near the time of obser-
vation and if feasible, at some period prior to the observation when
plenty of time is available. Whether with or without a nephoscope,
the observer should take note of the kind of clouds and their direction
of movement during a period of 2 or 3 hours prior to the observation
whenever practicable so that the information obtained will be of
assistance at the time of observation.

16

OBSERVATION OF CLOUDS FOR CODED MESSAGES

A careful examination of this publication, “Cloud Codes” will show
the observer that the International plan of reporting clouds in synoptic
messages is radically different from that employed by the Weather
Bureau for many years in the word code. The principal difference is
that the Weather Bureau has heretofore based its coding on the kinds
or forms of clouds and their movements, whereas the International
Cloud Code is arranged to give information concerning states of the
sky which include cloud forms, indications of structure, and the
general evolution and organization of the clouds.

In the International synoptic message one figure, 0-9, is included
for each of the three principal cloud levels. In each case, ‘‘0”’ signifies
“no clouds.”’ Thus there are 27 cloud code classifications, 9 for each
level—lower, middle, and upper—as shown in the tables on pages 19
and 20.

One form of cloud may be found in several different code classifi-
cations. For example, in the table for lower clouds, cumulus is men-
tioned in the specification opposite the code figures 1, 2, 4, 7, 8, and 9.
These represent different stages in the evolution of the clouds or differ-
ent arrangements in combination with other clouds in the sky. The
same is true of clouds at the other levels—middle and upper—as will
be seen in an examination of the appropriate tables.

Even if the observer is attempting to determine only the form of
cloud, a more or less continuous watch of the sky is helpful because
cloud forms are more easily identified when the evolution of the clouds
is observed. In using the International Code, it is important that
the evolution of the clouds and the states of the sky be care-
fully watched whenever possible because they are the basis of the
classifications.

The states of the sky are classified with respect to the position of the
observer relative to the center of a disturbance. Thus the cloud ob-
servations, if carefully made, should give valuable supplementary in-
formation relative to the structure and evolution of the disturbance.
In order that the observer may code the cloud forms and states of the
sky correctly he must study the code specifications, cloud descriptions,
and illustrations so that he will have them clearly in mind when watch-
ing the evolution of clouds and making observations for coded messages.
The time allotted to the observation is insufficient to permit the ob-
server to search specifications and descriptions; he must have all of this
in mind and whenever possible should also be fortified with a knowledge
of the development and arrangement of clouds during an adequate
interval (2 or 3 hours if practicable) preceding the observation.

In the issue of instructions for observing clouds, it is fully realized
that many observations are taken in darkness without any preceding
period of daylight to permit a preparatory period of observation; also,
that conditions may change materially just prior to an observation

AL

taken in darkness so that the preceding notes taken in daylight will
not be appropriate for use in coding clouds for the synoptic message.
However, it is intended that the observer will take every advantage of
periods of daylight or visibility of the moon and stars that is practica-
ble without undue hardship in extending his hours of duty or depriving
him of needed rest. The cooperation of other employees in the office
should be secured in making cloud notes during periods when the
observer cannot be on watch.

It is also important that doubtful cloud observations be omitted
from the coded messages. When there is serious doubt about the
direction of movement, it should be coded as ‘‘unknown.’’? When
there is real doubt as to the form of scattered clouds because they are
seen only near the horizon, or for other reasons the clouds cannot be
accurately observed at or prior to the observation, no cloud word
should be sent. However, doubt should exist only after every effort
has been made to ascertain the facts. For example, a definite cloud
bank on the horizon should be reported and if it is observed to be ad-
vancing slowly its direction of movement should be given. It is im-
portant to note that a bank of what appears to be cirrostratus on
the horizon, definitely advancing, is coded H; when the high clouds
do not extend 45° above the horizon. This may be the first indication
in the sky of the approach of a disturbance.

It is especially desirable that care be exercised in observations of
high and intermediate clouds, particularly altostratus and altocumulus
clouds, because accurate data concerning clouds at these levels are
of very great value to the forecasters. Preconceived ideas as to the
probable direction of upper and intermediate clouds based on average
conditions or past experience should not influence the observer.
Intermediate and high clouds may be visible at one station when the
sky is completely covered by low clouds at surrounding stations. In
such cases, isolated observations of intermediate and high clouds
become of critical importance.

Special instructions are necessary for observing and coding clouds
at mountain stations where they may exist at levels below the ob-
server’s station.

18

cover the largest portion of the sky at the time of observation.

CODE TABLES

The form of predominating cloud is the genus which is observed to

All

species or special forms of each cloud genus will be counted in deter-
mining the total amount of sky covered by that genus.

FORM OF PREDOMINATING CLOUD

ae Form of cloud ee Form of cloud
1 | Cirrus. 6 | Stratocumulus.
2 | Cirrostratus. 7 | Nimbostratus.
3 | Cirrocumulus. 8 | Cumulus or fractocumulus.
4 | Altocumulus. 9 | Cumulonimbus.
5 | Altostratus. 0 | Stratus or fractostratus.

NE Si 2

CO NR oT BANWHS

[SS S)

Codes for the forms of clouds and states of the sky, lower, middle,
and upper levels, are as follows:

I. LOWER CLOUDS C,

Cumulus heavy and swelling, without anvil top_____________________
Cima Orig Us eyes eee Oa ES rr ee oe enn ee eS
Stratocumulus formed by the flattening of cumulus clouds___________
May erromstravusionstracocumuUlus=: a0 e wae eee Sen aes
Howabrokenjupicloudsiof badiweathers= ss s22" 5 5s 5s 4e eee
Cumulus of fine weather and stratocumulus_______________________-
Heavy or swelling cumulus, or cumulonimbus, and stratocumulus_____
Heavy or swelling cumulus (or cumulonimbus) and low ragged clouds

GLE |OBKC! CBRNE SES ARN Os le See ee ee ae ee a erie en Wee ccd

INommiddlerclouds 355 so) oe Se ey ae Oe Ne Se Se ee

vl micalvaltostratus thine: feo wee ees a Ee Se ee

Typical altostratus, thick (or nimbostratus)________________________

. Altocumulus, or high stratocumulus, sheet at one level only__________
. Altocumulus in small isolated patches; individual clouds often show

signs of evaporation and are more or less lenticular in shape________
Altocumulus arranged in more or less parallel bands, or an ordered
la ViCTLAC VAN CIN GVOVeTatMeis kaysette J soe sn) ny
Altocumulus formed by a spreading out of the tops of cumulus_-______
Altocumulus associated with altostratus or altostratus with a partially
SOc umauluste la Rac be Te ee eae mye eee rote opie ta een ated aa one

. Altocumulus castellatus, or scattered cumuliform tufts___-____________
. Altocumulus in several sheets at different levels, generally associated

with thick fibrous veils of cloud and a choatic appearance of the sky__
JUOL, (NPR, (Civ ONONDIS; Ce,

Be NOMUP DCTACL OG sie mess ee tesa eaten a mn Eic hee a eaee eet Hau biec” Bile ee ee Ch dee
. Cirrus, delicate, not increasing, scattered and isolated masses_________

19

Cirrus, delicate, not increasing, abundant but not forming a continuous
Teer 2 Sees eee eae re re ee a
Cirnustofanvailycloudsssustall lyn ems coe ee ee
Cirrus, increasing, generally in the form of hooks ending in a point or
invassmalligutte 8 see oe ee ee eee
Cirrus (often in polar bands) or cirrostratus advancing over the sky
bud notaumore than! 45> above the horizons === = a=
Cirrus (often in polar bands) or cirrostratus advancing over the sky and
TRIVOINE) Hovain, AGI A ovonie ielaves lous o = = = ee See =
Veil of cirrostratus covering the whole sky__.-.___---____-=---===-=
Cirrostratus not increasing and not covering the whole sky__-___-_----
Cirrocumulus predominating, associated with a small quantity of cirrus_

30
30

30
30
31
31

31
31

THE STATE OF THE SKY AND THE EVOLUTION
OF THE CLOUDS

The necessity of noting the state of the sky as a whole.—It is evident
from the specifications of the cloud code that to describe the sky at the
station at a particular time logically and completely it is not enough to
know the genera or even the species of the clouds present; for example
altocumulus appears in seven specifications of the code, and cirrus in
nine. In reality each specification of the code, as the explanations
show, is not so much a dry enumeration of the genera or species of
clouds in the sky as a general indication of the structure, the organiza-
tion, and the evolution of the cloud complex which makes up the state
of the sky. Some specifications only refer to the general structure;
for example Cy=9 is a thundery sky; everyone knows that in thundery
conditions degenerate cloud forms are met with which are very difficult
to classify, while the thundery look of the sky is apparent immediately
and without any doubt.

Each specification of the code corresponds to a state of the sky—
lower, middle, or high.!| The observer should have at his ‘fingers’
ends”? the commentaries accompanying the definitions; he should
consider as a whole the lower, middle, or high clouds there described,
and try to make a considered judgment of the observed sky as a whole,
so that he can directly apply the proper numbers of the code.

The detailed analysis of the individual clouds should follow and not
precede this recognition of the state of the sky as a whole. If the observer
becomes accustomed to this course he will find in a short time that the
different states of the sky—lower, middle, and high—corresponding
with the code, will seem just as ‘‘live”’ as the typical cloud forms, and it
will be just as easy to identify a state of the sky as the form of a cloud.

Necessity of following the evolution of the sky.—The aspect of the sky
is continually changing, and many transitional forms exist between
the different types of cloud. Itis relatively rare for the observer to see
typical clouds of one genus which float past, or persist in the sky for
any considerable time; in most cases he will find that he has difficulties
at the time of observation if he has not taken the trouble to watch the
sky since the last observation. If, however, he has taken this precau-
tion he will often be able to refer a confusing state of the sky, or a
particular cloud, to a previous state which was typical and easy to
identify. Moreover, most of the specifications of the cloud code take
into account the evolution of the clouds. A single isolated observation
is insufficient.

As regards evolution, the recognition of the state of the sky, recom-
mended in the previous paragraph, is better than the identification of
clouds considered by themselves, for, as a matter of fact, the evolution
of the state of the sky can be followed indefinitely at one station, while
the evolution of a cloud, if, as is usual, it is a “migrant’’ can only be
observed during the relatively short time that it takes to cross the sky.

1 In the International Atlas, the term ‘‘state of the sky”’ has two distinct meanings. (1) The form and ar-
rangement of the clouds in the sky at any given level—lower, middle, or high—is a ‘“‘state of the sky”’ at that
level; this is the meaning of ‘‘state of the sky”’ in the code specifications. (2) The combination of all clouds
in the sky is also described as a ‘‘state of thesky,’’ for example, a ‘‘front zone sky,’’ an “‘emissary sky,’’ ete.

In the latter sense it is used by the forecaster as an interpretation of the combination of code figures in the
synoptic message.
21

CODE SPECIFICATIONS
LOWER CLOUDS C,

0. No Lower Clouds.

1. Cumulus of Fine Weather (1).'

This is observed under different aspects:

A. In astate of formation, in general in the morning.

B. Completely formed in general in the middle of the day, with
definite horizontal bases, the air being more or less calm; these are
(a) with rounded tops but without ‘‘cauliflower’ heads, or (6) flat
and ‘“‘deflated.”

C. Completely formed but broken up by the wind; in this case they
remain separate and are white in color (2).

The cumulus clouds of fine weather usually have a marked diurnal
period on the continent, growing until the middle of the afternoon,
and decreasing later, both as to amount of cloud in the sky and in
their vertical development.

These cumulus clouds are found only at a distance trom disturb-
ances. When the veil of cirrostratus, which fringes the front of a
disturbance, begins to cover the sky, the cumulus of fine weather
changes from type Ba to type Bb and finally disappears entirely.

The fractocumulus of fine weather noted above and coded as C,=1,
must not be confused with the fractocumulus of bad weather which
is coded C,=6, or C,=9. The former are (C,=1) detached white
clouds usually in a blue sky, and remain detached; the latter are
found in the central part of a disturbance or in its rear; in the first
case (C,=6) they form? under a gray sheet of altostratus or of
nimbostratus; in the second case (C,=9), in a sky crowded with
clouds at all altitudes, they form under the bases of cumulonimbus or
very large cumulus clouds or in the spaces between these. In both
cases they are dark, receiving little light, and generally become very
numerous, while the fractocumulus clouds of fine weather usually
show white on a blue sky, and remain detached.

2. Cumulus, Heavy and Swelling, Without Anvil Top (3, 4).

There are two types of these:

A. In calm air and especially on hot days with a thundery tendency
they form heavy masses with horizontal bases, and very great vertical
development. They are sometimes in the form of towers, sometimes
of complex heaps with ‘cauliflower’ formation. They often have
caps or hoods (pileus).

B. In strong winds in the rear of disturbances they also form tower-
ing masses, with great vertical development but tossed about and
broken up.

These cumulus clouds, especially those of the second class, are often
associated with thick cirrus (C,—3) and with extensions of strato-

1 Numbers in parentheses after code classifications refer to cloud pictures which illustrate the codes.

2 Fractocumulus clouds of bad weather (C.=6) develop under a sheet of altostratus or nimbostratus and
should not be confused with fractocumulus which persist from a previous disturbance or exist independently.

22

cumulus and altocumulus (C,,—6). None of these cumulus clouds
should show ice crystal clouds at their tops; this would mean that
they had reached the cumulonimbus stage which would entail code
3. Cumulonimbus (5, 6, 7, 8).

Cumulus clouds of great vertical development, with the tops com-
posed of ice crystal clouds. Sometimes the nascent ice crystal cloud
is merely mingled with the “‘cauliflower” tops, where a fibrous struc-
ture appears and the clear-cut outlines fray out; sometimes the
completely formed ice crystal clouds crown the cumulus with a definite
plume of cirrus of a shape more or less like that of an anvil. Some-
times, especially in the spring and in high latitudes, the ice crystal
formation involves nearly the whole cloud even to the base. At the
end of the growth of a cumulonimbus the lower cumuliform part of
the cloud often tends to disappear leaving only the upper or cirrus
part.

When the anvil, reaching nearly to the zenith, begins to overshadow
the observer, a mammatus structure will often be seen on the lower
surface of the anvil projection.

Like the heavy and swelling cumulus, cumulonimbus is formed either
in calms, especially on hot thundery days, or in a strong wind in the
rear of disturbances.

Cumulonimbus is a regular factory of clouds. By extension at
various levels it often produces either cirrus masses by an extension
of the ice crystal parts, or masses of altocumulus or stratocumulus by
an extension of the cumuliform parts, and these may end by becoming
detached from the parent cloud. Thus cumulonimbus, C,=3, may
coexist with cloud sheets that should be coded C,=3 or C Ge

At the end of the evolution of cumulonimbus, C,=3 should only
be coded when the cumuliform parts are still visible. When the
cumuliform clouds have degenerated, the anvil top or the clouds
derived therefrom should be ‘coded C8.

When cumulonimbus nears the zenith, and its base, with low dark
clouds under it, often in the form of a roller or an arch, has covered all
or nearly all the sky, code C,=3 should be replaced by C,=9.

4. Stratocumulus Formed by the Flattening of Cumulus Clouds (9, 10).

Cumulus tops may settle down and the bases may spread out; this is
a frequent end to the progressive changes of the cumulus of fine
weather. On the other hand, the bases may melt away and the tops
may spread out; this is a common phenomenon in the rear of a dis-
turbance after squalls or showers. Very opaque sheets or a layer of
stratocumulus may be formed in this way, often showing a festooned
formation (mammatus) in places. At the end of the process the clouds
thus formed may thin out.

The first case is stratocumulus vesperalis, the second stratocumulus
cumulogenitus.

5. Layer of Stratus or Stratocumulus (11, 12, 13, 14).

Clouds usually forming a single layer, fale, cullen and not very
dark or menacing; they have a certain stability. "The stratocumulus
has often semitransparent parts, or even clear spaces between the
elements of the cloud.

23

These cloud formations are common on the continent, especially in
winter, and are found outside the regions of disturbances, or on their
extreme lateral edges.

The layer of stratocumulus may often be broken up; code C,=5 is
used only for those sheets of stratocumulus that are not formed from
cumulus; otherwise they are coded C,=4. The observer may be in
doubt between C,=5 and (,,=—3; the code C,=5 is used only when
the stratocumulus is fairly low and rather like stratus (large and rather
diffuse tesselations or waves); if it is clearly high up and related to
altocumulus it is coded C,,=3.

6. Low Broken Up Clouds of Bad Weather (15, 16, 17).

The following is the ordinary course of formation of these clouds:
When a veil of altostratus becomes lower, and tends to turn into
nimbostratus, it usually has below it a gradually increasing layer of
fractocumulus or fractostratus; these clouds are isolated at first.
They ultimately fuse together into a continuous sheet; but through
interstices the veil of relatively light higher cloud may be seen (16).
The continuous rain does not usually occur until after the formation
of the fractostratus or fractocumulus, which is then hidden by the
precipitation, or may even disappear under its influence.

This type is found in the middle of a typical disturbance.

For the difference between the fractocumulus of bad weather and
that of fine weather see the explanation under C,=2.

7. Cumulus of Fine Weather and Stratocumulus (18, 19).

Cumulus clouds may form below an already existing sheet of strato-
cumulus, which they do not penetrate.

Before the formation of the cumulus, the layer of stratocumulus
would have been coded C,=5 or C,,=3 according to its height.

If the cumulus penetrates the stratocumulus layer it must be
coded C,=8.

8. Heavy or Swelling Cumulus, or Cumulonimbus, and Stratocumulus
QUa2h):

Heavy or swelling cumulus, or cumulonimbus may form below an
already existing sheet of stratocumulus, and some of the cumulus or
cumulonimbus may penetrate the layer.

This is a formation analagous to C,=7 except that the vertical
development of the convection clouds is more marked, so that in the
absence of the upper layer they would be coded C,=2 or C,=3.

9. Heavy or Swelling Cumulus (or Cumulonimbus) and Low Ragged
Clouds of Bad Weather (22, 23, 24).

When a heavy or swelling cumulus or a cumulonimbus nears the
zenith it can hide all, or nearly all the sky with its base; the latter may
somewhat resemble nimbostratus, but is distinguished from it by its
previous history, or by the broken character of the rain, which, in the
case in question, falls to the ground in showers, or, if it does not reach
the ground, is seen falling from the lower surface of the cloud in wisps
of rain or snow called virga.

Generally below the base of such a great cloud there is a greater or
smaller amount of lower cloud broken up in the manner of fracto-
cumulus or fractostratus, these low clouds may sometimes be in the
form of a roller, or an arch.

It may also happen that low dark clouds of the nature of fracto-
cumulus or fractostratus may grow in amount in a sky covered with
heavy or swelling cumulus or with cumulonimbus and may fill the
spaces between the main cumulus clouds.

Only in these two cases should C,=9 be coded:

A. When a cumulus cloud approaches the zenith and its upper parts
are still visible it should be coded C,=2 or C,=3 according as to
whether it is a heavy and swelling cumuius or a cumulonimbus.

B. In damp climates it may happen that dark fractocumulus or
fractostratus forms a layer completely closing up the spaces between
the masses of heavy cumulus. In such cases one cannot distinguish
individual cumulonimbus clouds under their typical aspect; but their
passage overhead is manifest by a temporary darkening of the sky
and by showers. Their presence, thus made known, allows the
code C,=9 to be used and not merely C,=6.

MIDDLE CLOUDS C,,

0. No Middle Clouds.
1. Typical Altostratus, Thin (15, 25).

A sheet of this cloud resembles thick cirrostratus, from which it is
often derived without any break; but halo phenomena, sun pillar, etc.,
are not seen in altostratus, and the sun appears as though shining
through ground glass and does not cast shadows.

This cloud is found in the central part of a typical disturbance.

If there were halo phenomena or if the sun cast shadows one would
code C,—5, C,—6, C,=7, or C,=8 according to circumstances. If
the sun were hidden, or lable to be completely hidden by a thick part
of the sheet, the code C,,=2 should be used.

2. Typical Altostratus, Thick (er nimbostratus) (26, 27).

The sun and moon are completely hidden, at any rate by some parts
of the cloud sheet. Typical thick altostratus can be formed either by
a thickening of typical thin altostratus, C,=1 or by the fusing to-
gether of the cloudlets in a sheet of altocumulus, O,,=7.

Nimbostratus is derived either by a change from typical thick
altostratus, or by the fusing together of the cloudlets in a sheet of
stratocumulus (C,=5).

This type is found in the central parts of a typical disturbance.

In the case of a transition from altocumulus into altostratus, if it
shows either in whole or part the structure of altocumulus (ripples,
waves, or tesselations) it is coded C,=7. In the case of a change
from stratocumulus to nimbostratus, it will also be coded C,,=7 if the
lower surface shows a real relief (waves or tesselations) instead of
showing a smooth under surface. Thick typical altostratus and nim-
bostratus are often accompanied by underlying and very low clouds,
which are ragged and dark (fractocumulus or fractostratus); in the
gaps one can generally see the altostratus or the nimbostratus of a
lighter gray; this case will be coded as C,=6, and C,,=2. If the lowest
clouds form a continuous sheet the observer will not assume what is

above, but will code C,=6.

3. Altocumulus, or High Stratocumulus, Sheet at One Level Only
(28, 29).

A cloud generally forming a single layer; it is fairly regular, and of
uniform thickness, the cloudlets, tesselations or waves, being always

25

separated by clear spaces or lighter gaps; the cloudlets are neither very
large nor very dark. This layer is generally fairly stable, that is to
say it changes but slowly. In tropical and subtropical regions such
a layer often forms at the end of the night in calm weather, even at a
high altitude, in the absence of any disturbance.

This layer of cloud is sometimes broken up but it is only coded as
O,,—=3 if the layer does not proceed from the extensions of the tops of
the cumulus clouds; in the latter case it is coded as C,=6. For the
difference between C,,=3 and C,=5 see the explanatory remarks on
C,=5. The layer of altocumutus that should be coded Cy=3 is
distinguished from that coded C,,=5 by its stability with no tendency
to increase, and by a greater regularity and uniformity.

4. Altocumulus in Small Isolated Patches, Individual Clouds Often
Showing Signs of Evaporation and Being More or Less Lenticular
in Shape (30, 31).

As regards the smallness of the cloudlets these little patches of
altocumulus resemble cirrocumulus, but do not show the characters
of clouds formed of ice crystals. Lenticular altocumulus clouds
show the most beautiful irisations; when this is the case they are lens
shaped, fairly thick, but with little or no shadows and of a pure dazzl-
ing white; they are slightly wavy at the edges.

Generally they are scattered over the sky quite irregularly and are
often at different levels; they are mostly in constant change so that
if one looks away from the sky for only a few minutes it 1s difficult
afterward to identify the clouds previously seen. Individually they
are often in process of dissolution, but the amount of cloud over the
whole sky does not in general become greater or less.

This type of cloud is common on the extreme lateral edge of dis-
turbances and also in mountainous districts under the influence of the
foehn wind.

The characteristics mentioned above are very typical. Neverthe-
less in cases where the observer is doubtful between C,,=4 and C,,=6
or between C,,—4 and C,,=5 they can be discriminated in the follow-
ing way: The sheets of altocumulus of C,,—=4 are higher and more
delicate than those of C,,=6 and they do not have the same regular
structure as those of C,,=5. Wisps of rain or snow (virga) may some-
times be seen falling from the underside of sheets of altocumulus of
class C,,=4.

5. Altocumulus Arranged in More or Less Parallel Bands, or an
Ordered Layer Advancing Over the Sky (82).

The essential feature of this type is that the sky becomes more and
more covered; the process often begins with altocumulus in large
parallel bands, which often have a roughly lenticular shape. On
their edges they may show signs of evaporating, but on the whole the
amount of cloud and the thickness of the sheets increases.

Sometimes altocumulus appears from the beginning as a sheet,
often under the form of a vast pavement with more or less rectilinear
joints, and the semitransparent parts allow the blue of the sky to be
seen. This layer very soon thickens in places, or has another layer,
lower and darker, forming beneath it.

This cloud is met with on the lateral side of a typical disturbance
or on the front of a weak disturbance.

26

This type of altocumulus is distinguished from that coded as
C,,=3 by the progressive deterioration of the sky, and by the irregular
thickness of the layer. Again the proportions of these sheets and
their regular structure differentiate them from those coded as C,,—4.

6. Altocumulus Formed by a Spreading Out of the Tops of Cumulus
(33).

Cumulus clouds of sufficiently great vertical development may
undergo an extension of their summits while their bases may gradu-
ally melt away. These sheets of altocumulus are generally fairly
thick and opaque at first with rather large elements, dark and soft;
later they may thin out and finally have rifts in them, or at any rate
semitransparent interstices.

The phenomenon of the extension of cumuliform masses is common
in the rear of disturbances, after squalls and showers.

These extensions from the summits of cumulus clouds must not be
confused with the ice crystal extensions of a cumulonimbus from
which the anvil and hybrid cirrus is formed. The anvil, and even the
cirrus, detached from the cumulus which produced it, may sometimes
have a festooned appearance on the lower surfaces, and thus may
have a certain likeness to a sheet of altocumulus. But the alto-
cumulus does not have the whiteness, the silky appearance, nor the
fibrous structure of a cirrus anvil.

To discriminate between the altocumulus coded C,,=6 on the one
hand, and the altocumulus coded C,=3, C,=4, and C,,=5 on the
other, see the explanations of C,,=3, Cy,=4, and O,,=5.

The observer may in certain cases be in doubt between C,,=6 and
C,=5. At the end of the evclution of sheets of altocumulus cumulo-
gemitus when they may have thinned out and become semitransparent
one would not of course think of coding C,=5; but at the beginning
of the change when the cloud elements are rather large, dark, and soft
there are all the transitions between stratocumulus and altocumulus.
It is a question of the apparent size of the elements; the convention
is that the cloud in question is altocumulus when the smallest well-
defined cloud elements observed over the sheet as a whole (and
ignoring the separated portions which may exist beyond the edges
of the sheet) are not greater than 10 diameters of the sun, i. e., approxi-
mately the width of three fingers when the arm is held extended;
in case of doubt it is best to code C,,=6, which allows one to note
(C,=2 or C,=3) cumulus clouds which are visible at the same time.

7. Altocumulus Associated With Altostratus, or Altostratus With a
Partially Altocumulus Character (34, 35, 36).

Different clouds are comprised in this section:

A. Typical altostratus can lie above sheets of altocumulus which
are at a definitely lower level. This type including typical alto-
stratus can only exist in the central part of a disturbance.

B. A more or less continuous layer of altocumulus can have below
it a gray veil of cloud, often hardly visible, lying at a level very little
lower; in places, and for short times it hides the cloudlets of the alto-
cumulus sheet sufficiently to give it in places the appearance of alto-
stratus. This type is found in the central part of a weak disturbance,
or on the lateral edge of the central part of a typical disturbance.

C. In a layer of altocumulus which is growing thicker ((,,=5)
the cloudlets may fuse together and the layer may become alto-

27

stratus; the cloud loses its tessellations, and takes on a fibrous struc-
ture; this transitional state is coded C,,=7, as is also the transition
from stratocumulus to nimbostratus. It 1s observed in the same
situation as the preceding cloud.

D. Again, altostratus, and even nimbostratus can change pro-
gressively into altocumulus, and in this case also the state of transition
is coded C,,=7. This phenomenon is common after continuous rain,
when the altostratus breaks up, that is to say at the passing away of
the end of the central part of a typical disturbance. It is also found
in the actual center when the disturbance is filling up.

E. The opaque cloud sheets with a more or less irregular corru-
gated structure, too dense and thick for the transparency of the ripples
to afford any criterion for their classification, should be coded as
O,,=7. The wave structure, though not visible as lighter parts, is
nevertheless apparent, and gives a corrugated appearance to the under
surface. These sheets were sometimes called ‘‘wrinkled”’ altostratus
in the old classification, but they should now be ealled altocwmulus
opacus. They are met with in the same situation as classes B and
C above.

These different types have this in common, that they denote a
weakening of a disturbance, because they are related either to one
which is weakened as a whole, or to a weakened, lateral part of a typical
disturbance.

As regards the usually ephemeral change of altostratus, or nim-
bostratus to altocumulus or stratocumulus, or vice versa (class
C and D above) and the differences between codes C,=2 and C,=7,
see the notes on C,,=2.

Corrugated altocumulus opacus (class E above) has two origins;
either it forms (a) like the cloud sheets of class C above, by the
uniting of the cloudlets and the general thickening of a layer of
altocumulus coded as C,,=5 (or more rarely C,,=3), or (6) as in class
D above, it forms from thick altostratus or from nimbostratus (C,,=2)
which is in process of dissolution; but it represents a definitely more
stable type than C or D; in case of these latter there is a quick and
almost direct change of the altocumulus translucidus into altostratus
or vice versa; in the case of class 5 the intermediate stage of alto-
cumulus opacus can be maintained for some time. The layer of
corrugated altocumulus opacus (C,,=7) is distinguished from sheets
of altocumulus coded C,,=3 or C,,—5 by its greater thickness which
gives it a darker color and a menacing appearance, and by the very
marked irregularity of the relief of the under surface which gives it a
characteristic pendant appearance (festoon clouds, pendant cor-
rugations).

8. Altocumulus Castellatus, or Scattered Cumuliform Tufts (87, 38).

The character common to these types of altocumulus is a domed
shape, but they have very different aspects.

A. Altocumulus castellatus is composed of a series of small cumuli-
form masses with more or less vertical development, arranged in
line and resting on a common horizontal base (reduced sometimes to
plain gray wisps); this gives the cloud a crenellated appearance.

B. The second kind is observed in scattered tufts, white or gray,
but without definite shadow, with the rounded parts very slightly
domed; they resemble very small, more or less broken up cumulus
clouds without a base.

28

These cloud forms are the precursors of thunderstorms, but the
first named variety may appear a long way in advance.

9. Altocumulus in Several Sheets at Different Levels, Generally Asso-
ciated With Thick Fibrous Veils of Cloud, and a Chaotic Appear-
ance of the Sky (39, 40).

This type is very difficult to analyse; it is generally complex, and
patches of middle cloud, more or less fragmentary, are seen super-
posed; they are often badly defined and with soft outlines; there are
all the transitional forms between low altocumulus and the fibrous
veil. The sky moreover is covered with clouds of different layers,
but as in general there is not a continuous sheet, blue patches may
be seen. Luckily, though these middle clouds are very difficult to
classify individually, the state of the whole sky is very typical; it has
a disordered, heavy, and stagnant appearance (calm or light wind).

These appearances are characteristic of the central parts of thun-
dery disturbances.

The clouds of other levels which almost always exist in a thundery
sky are principally cirrus densus (Cy=3) and large cumulus (C,=2)
or cumulonimbus (C,=3).

It would be too complicated, and hardly worth while, to attempt
to analyse in detail the individual characters of the altocumulus
sheets in a thundery sky. It should be carefully noted that such a
sky can be identified by its general appearance as a whole, which is
quite typical. There is only one other type which in this respect

may be confused with a thundery sky, because it also has a more or
less disordered appearance; this is the sky in the rear of a disturbance,
but this latter is fairly easily distinguished by its ‘“‘tossed about”
and “‘windswept”’ appearance.

UPPER CLOUDS C,

0. No Upper Clouds.

1. Cirrus, Delicate, Not Increasing, Scattered and Isolated Masses
(41).

This type of isolated cirrus is widely scattered; its amount does
not noticeably increase either in time or in any par ticular direction.
It does not collect into sheets or bands, and there is no tendency for
the elements to fuse together into masses of cirrostratus. The cirrus
clouds whose strands end in an upturned hook or tuft must not be
included in this class.

Cirrus clouds of this type are indications of a very distant dis-
turbance, and are found either on its front or on its borders. The
place where these clouds are observed is generally not influenced by
the disturbance, at any rate for some time.

The chief characteristic of the type C,=1 is the sparseness of the
cirrus. It is only distinguished from the type C,=2 by this fact of
the cirrus being more sparse.

This type of cirrus is distinguished: From that of class C,=3 in
that it is delicate and does not originate in anvil clouds; from class
C,,=4 in that it does not increase either in time or in any particular
direction of the sky, and that it does not belong to the uncinus type
(strands with upturned ends); from class C, —5 and class Cz Grim
that it does not increase in time or in any particular direction, that

98863°—38

29

it does not form into sheets or bands, and that it does not show, in
any part of the sky, any tendency to pass into cirrostratus.

2. Cirrus, Delicate, not Increasing, Abundant but not Forming a
Continuous Layer (42).

The definition of this type is precisely the same as the preceding
with the exception that in this type the cirrus is more abundant
over the whole sky, but without any tendency to increase in any
particular direction.

Cirrus of this type is the forerunner of a disturbance either on its
front or on its borders.

3. Cirrus of Anvil Clouds, Usually Dense (43, 44).

In this type the cirrus either actually proceeds from the anvil of
a cumulonimbus, in which case one sees cirriform masses attached to
the remains of cumulus, or it has probably so originated, being still
anvil-like in shape, or else remarkable for its density and frayed-out
appearance, generally showing virga in places.

Originating in cumulonimbus, this type of cirrus is met with either
in the rear of typical disturbances or else around thunderstorms.

In the first case the cloud is hybrid cirrus; in the second it is dense
cirrus, in which case the cumulonimbus stage may have ended some
time previously.

4, Cirrus, Increasing, Generally in the Form of Hooks Ending in a
Point or in a Small Tuft (45).

This type of cirrus which is often in the form of streaks ending in a
little upturned point or in a small tuft, increases in amount both in
time and in a certain direction. In this direction it reaches the
horizon where there is a tendency for the cloud elements to fuse
together, but the clouds do not pass into cirrostratus.

This type often occurs on the front of a typical disturbance.

If the cirrus increases in time and in a certain direction tending
also to pass into cirrostratus in this direction it should be coded as C,=6
or C,,—5 according as to whether the front of the cirrus sheet, formed
out of the scattered cirrus is at a greater or lesser altitude than 45°
above the horizon.

5. Cirrus (Often in Polar Bands), or Cirrostratus Advancing Over the
Sky, but not More Than 45° Above the Horizon (46).

A sheet of fibrous cirrus partly uniting into cirrostratus, especially
toward the horizon in the direction where the cirrus strands tend to
fuse together; the cirrus is often in a herring-bone formation, or is
in great bands converging more or less to a point on the horizon. In
this class is also included a sheet of cirrostratus without any cirrus.
In either case the front of the sheet is not more than 45° above the
horizon.

This type occurs in the front of a typical disturbance.

The varieties included in this definition are ‘‘herring-bone cirrus”
and ‘‘polar bands.’”’ What is termed the front of the sheet means
the front either of the cirrostratus sheet, if this cloud exists without
cirrus, or of that part of the sky where the cirrus fibers are close enough
to each other to appear welded together, instead of being discrete
masses like the cirrus which precedes this sheet.

30

6. Cirrus (Often in Polar Bands) or Cirrostratus Advancing Over the
Sky, and More Than 45° Above the Horizon (47).

The definition of this type is exactly the same as the previous one,
with the sole exception that the edge of the sheet is more than 45°
above the horizon.

This type is found in the front part of a typical disturbance, a little
nearer the center than the former type.

The same remarks on the limits of the cloud sheet apply as for

Cp=&,
7. Veil of Cirrostratus Covering the Whole Sky (48).

There are two cases:

A. Thin, very uniform nebulous veil sometimes hardly visible,
sometimes relatively dense, always without definite detail, but always
producing halo phenomena (halo, sun pillar) round the sun or moon.

B. A white fibrous sheet with more or less clearly defined fibers,
often like a sheet of fibrous cirrus, from which indeed it may be
derived.

This type is found in a typical disturbance immediately in front
of the central area.

8. Cirrostratus not Increasing, and not Covering the Whole Sky
(49, 50).

This is a case of a veil or sheet of cirrostratus reaching the horizon
in one direction but leaving a segment of blue sky in the other direc-
tion; this segment of blue sky does not grow smaller. Generally the
edge of this sheet is clear cut, and does not tail off into scattered
clrrus.

This type is found on the north lateral edge of a disturbance, the
aspect of the cloud being generally very different from that of the
clouds on the south lateral edge. (The terms north and south apply
to the most frequent case of a disturbance moving from west to east.)

If the segment of blue sky is diminishing the cloud must be coded
as C,—6 or C,=5 according as to whether the elevation of the sheet
of cirrostratus is greater or Tess than 45°.

9. Cirrocumulus Predominating, Associated With a Small Quantity
of Cirrus (51, 52).

An absolutely necessary characteristic of cirrocumulus, according
to the new definition of this cloud, is its relationship to cirrus or
cirrostratus.

This type is found on the front or lateral edges of a weak
disturbance.

If the cirrus fibers, or the sheet of cirrostratus merely becomes
slightly corrugated in parts the cirrocumulus is neglected. But
if the bank of cirrus or of cirrostratus degenerates wholly into
cirrocumulus jit is coded as C,=9; this is an important sign of the
weakening of a disturbance.

KEY TO IDENTIFICATION OF CLOUD CODE TYPES

Observers may find the forms and arrangement of clouds in the sky,
as classified in the codes, to be rather confusing, especially in the be- |
ginning. The following key may be helpful in determining the code
numbers to use. However, the observer should not depend upon the
key alone. He should make a thorough study of all the definitions
and classifications and have them clearly in mind, using the key only
as a general guide or reminder.

The first step is to determine whether the clouds are at lower, middle,
or upper levels. If clouds at more than one level are present, the
observer should determine which clouds belong to each of the three
principal levels. He will then determine the code number to use in
each case.

The key is therefore intended merely to supply a hint to the observer
who will then recall the finer distinctions of the various code types.

Although a code number is sent for clouds at each of the three
principal levels, it is evident that certain combinations are rare or even
impossible. For example, with an extensive sheet of dense altocu-
mulus, M/,;,! the observer may be unable to identify the upper cloud
code type even if he knows that there are upper clouds.

1 The symbol Mz is equivalent to Cu=7; [Hs is equivalent to Cu=8; ete.

UPPER CLOUDS '

25)
H = ; co
= i Ays ofoyM oyinb you ynq 4sourje BuLaA0|y a
S | a
—_— 3 r iS
a AYS S[OYUM BuIA) ae
Be
Be ° AyS OFOyA =
e 2 1a ; Z K Ci ja
Ors 5 IdAO0 popudejxo AyTJUDOI SB_T
eo] eh &
® 4 5 °
a nD sy ;
4 ¢ = AYS O[OYM BULA JON | fe
pay mH
oe) o
: |
hom — is
oO io
SS oe. SN{BIYSOLIMD OFUL BULBIOUL SNAITD 3
ae) l
3 8 “A snoiqy Jo IO SN{BIYSOLID JO JOY aaa
% 5 a [AUB UlOIf poatiop ATqeqoig ae
a S (a
S |
s © a
| LY
= . 2 quBpunqy ca
° S) ee
Rn © =|
fe} Comal
S |2i 6 Bee ete
9 8 001BOG aa
on TR
q
Rey [TAUB oP
se > | wtoay poAtsop A[qeqoad osuap sseq 10 RKO) \ a
5 | a
fe Bs)
Sle
ai) o 1
8 As $4jn} Io syooy YyIM ATjensn ‘ozBorp[eCqy ac

1 Tf cirrocumulus predominates, use code [/g.

MIDDLE CLOUDS!

1
eS i ti soBvjIns Jepun uo
8 a2 S| joryet opruygop yy ‘sng c
a ap a -B1}S0}[B YIM poBIOOssB x
i fas snpNwIMso}]B Lo snyuumMooyyy
o
ica td
& a 2 ooBJINS ee
wn eae - | tepun uo jolper oyuyop ON |
~_ —_
Ss ag
= 5
4 Ne)
~ i= “4 fal
= a a5 =
} Ss 3 (snqgeajsoqye uryy yeordaéy,) | S
oO HH Ee
2)
20 3 IOAB] OL SUIS C S
= urs U
eg [epus uy | Ss
a
) - S
Ll
Hes! I oa
= $ IBpnoyua'yT S
S Sm
n
4
S
ra qo
| Se be | : &
8 | eos (snqeyoysey) | &
Sa
mn o
o
i
2) oO
~ Ke) Ie!
& co ak] podapio 10 spueq Joqeavd uy | =
Lal
ao ov
SW ieeeniouss
© | Ll esta
Pa “= of
op soma
= 56 snpnumo wto1y poamog | S
pS Ha s
ovis
DIS

1 Altocumulus in the typical chaotic, thundery type of sky is coded Mg.

33

LOWER CLOUDS

Sheet or layer

Detached masses

MOTOG 1oY}BIM PB JO SpNopd MOT possvI

yA AS 9Y} SuLMeaod snquiruojnumny~s os
JaXB] OY} Buryeajyouod
SNQUTIUO|NUIM) 1Osn[NUINd) YIM 1O MOTE ny)
snpnumns SuleMo, YIM snypnumMd07eBsy4¢
MOTO -
SH|NUINS 1oYyYBOM ITV YYIM SN[NUINIOZBIYS =
yooys Shonuryuo0d =
OVUL posnjy JoyyBVeM peq Jo spnoyp>a MO'T
snpNuIN907zB.148 IO SnyBiys JO IOABT 9[SUIG ny
im roi snquituojnumns tapuyy | QQ
O4> Boe
gat eI Qa
a) aS
o¢g cane aee)
a) See SnyB1jsoq UU 2
re) A = _ “ S
= IO SnjBiySOy[B Jopus)
She
o
=o
= snynuiMd wo1y yno pwaidg ei
i qtururns snoiq =
ac) -J JIM YAMOIS TBOTVAIOA YBoTD
er
@ a
=
Ee qiuauins SNoOIqy =
2 5 QNOYUM YIMOIS [BOIZIOA OATJOW
a @
ae) (od44
= JoyyBoM Arey Teordsy4) yueurdo Sj

-JOAEP [BOTYAOA YYBTTS YAM ‘[ [BUI

34

SKY AND CLOUDS IN A TYPICAL DISTURBANCE

The accompanying diagram, taken from the International Atlas,
shows the distribution of cloud and sky types in the various sectors of
a typical disturbance.

In identifying cloud and sky types, observers in the United States
should be guided by the photographs and not by the position of the
clouds in the disturbance. The following notes are taken from the
International Atlas:

1. This diagram corresponds with a typical disturbance—to be
precise, the first of a family—arriving from the west in western Europe.

\ Zone if
los \ La M 6 5 oe
‘S = L. sFeral 3
Se Ls a mse 2 Les) H,

————— -_

ee ee

2. It sometimes happens that the rear zone is much more extensive
and that it persists for several days over the same region.

3. Fractocumulus may occur practically anywhere in the rear zone.

4. The specifications M/Z, and M, correspond respectively with the
front or lateral sector and with the central or rear sector of a thunder-
storm disturbance. They can find no place in this diagram which is
that of a normal disturbance.

5. Observers should not take the form of cloud corresponding with
the position in this diagram as necessarily determining the form of
cloud to be reported.

3D

PHOTOGRAPHS

Cloud photographs to illustrate the various code specifications are
shown on the succeeding pages. The pictures are accompanied by
descriptions which indicate briefly the reasons for the code figures
assigned. Observers should not rely solely on these pictures and
accompanying descriptions in coding cloud observations. All of
the definitions, descriptions, and code ‘specifications in the preceding
pages must be studied and the criteria for distinguishing the various
cloud forms and coding classifications must be memorized. Observers
must be able to prepare coded messages quickly and without searching
specifications and cloud pictures at time of observation.

Illustrations in the International Atlas should also be studied
carefully by all observers who have access to it.

The cloud photographs reproduced here have been obtained from
various sources. Many of them were selected from the Weather
Bureau collection to which many individuals have contributed. The
following persons furnished photographs especially for use in illus-
trating code specifications: Dr. Charles F. Brooks, Dr. W. J. Hum-
phreys, Mr. H. T. Floreen, and the Reverend C. E. Deppermann, S. J.
Photographs used for plates 61, 62, and 63 were taken by Mrs. Lois
Bowen (T. W. A., Inc.) and photographs for plates 66, 69, and 70 were
taken by the National Advisory Committee for Aeronautics through
courtesy of Pan American Airways.

36

1. Cumulus of Fine Weather, L,.

The clouds are scattered and have definite horizontal bases and
rounded upper surfaces. The diurnal period of convection is well
advanced (3:30 p. m., local time) without much vertical development.
The large cloud in the foreground has its shaded, horizontal base
turned toward the observer. Near the horizon the horizontal bases
and deflated appearance of the clouds are more clearly shown. (Photo
taken at Cheyenne, Wyo., by H. T. Floreen.)

2. Fractocumulus of Fine Weather, L,.

The clouds are broken up by the wind. The shadows are lght,
indicating that the clouds are thin. This form of fractocumulus
must not be confused with the fractocumulus of bad weather, L,
(pls. 15 and 16), forming beneath altostratus or nimbostratus, and
L, (pls. 22, 23, and 24) which appears below heavy and swelling cumulus
or cumulonimbus. (Photo taken at Washington, D. C., by W. J.
Humphreys. )

3. Cumulus, Heavy and Swelling, Without Anvil Top, Z».

The large cloud mass in the center has a horizontal, heavily shaded
base and shows active vertical development, with heaps in a “cauli-
flower” formation. The upper surfaces look hard and are clear cut,
hence the central cloud mass has not yet reached the cumulonimbus
(L;) stage. (Photo at Cheyenne, Wyo., by H. T. Floreen.)

4. Cumulus, Heavy and Swelling Without Anvil Top, L».

These clouds have great vertical development in the form of towers,
showing active convection on a warm afternoon. Their height and
the absence of any other cloud forms indicate that they are of recent
erowth. However, the second cloud from the left shows signs of the
formation of an anvil and is in a transition stage between L, and Ls.
(Photo at Manila, P. I., by C. E. Deppermann.)

40

41

5. Cumulonimbus With Anvil Top Beginning to Form, L;.

The main cloud mass is in process of transformation from L, to
L;. The ‘cauliflower’? summits remain but they are softening and
the summit shows some horizontal spreading. There is a light
shower below the central mass. Cumulus clouds, L.2, also show to
the right in the picture but the cumulonimbus would be given prefer-
ence in coding. (Photo on Mount Wilson, Calif., by F. Ellerman.)

6. Cumulonimbus, L;.

The anvilis partially formed and the ‘cauliflower’ swellings have
dissolved in the upper portion of the cloud mass. There is a small
veil cloud at the left of the summit. Heavy cumuliform masses
appear in the lower part of the picture with some altocumulus and
patches of dark fractocumulus. A heavy shower is visible. (Photo
at Cheyenne, Wyo., by H. T. Floreen.)

‘

43

7. Cumulonimbus, L,.

The anvil is fully developed and its edges are frayed out. Cumulus
clouds of a lighter shade appear against the gray cf the anvil; at the
center, they are merging with the cirrus mass. Below, there is a roller
of dark fractocumulus. The anvil alone, after the cumuliform parts
have disappeared, would be coded /; (pl. 44), but in this picture the
cloud is coded L3. (Photo at Washington, D. C., by P. A. Miller.)

8. Cumulonimbus, L,.

Here the cumulonimbus is shown as a regular ‘‘factory of clouds.”
The cirriform parts (7/;) of anvils extend overhead and beyond the
limits of the photograph. The principal mass appears to be only
one of several cumulonimbus clouds in the vicinity. This one has not
advanced far enough to be coded Ly. (Photo at Pensacola, Fla., by
Air Service, U. S. Navy.)

98863°—38——4 45

9. Stratocumulus Formed by the Flattening of Cumulus Clouds, Z,.

The tops of the cumulus clouds are subsiding and flattening and the
bases have spread out at the end of the period of diurnal convection.
Parts of cumulus bulges are still visible. Stratocumulus is seen in
the upper part of the picture, from degeneration of another mass of
cumulus. There arecrepuscular rays. This type is called stratocumulus
vesperalis. (Anonymous.)

10. Stratocumulus Forming From Cumulus, J,.

In this case the tops of the cumulus clouds are spreading out at the
end of the diurnal period to form opaque layers of stratocumulus.
The final phase in this process is the melting away of the bases of
cumulus or cumulonimbus and thinning of the stratocumulus. This
type is called stratocumulus cumulogenitus. The stratocumulus formed
in this way may show a mammatus structure. (Pheto taken on M.S.
Northern Sun by B. A. Thompson.)

46

11. Layer of Stratocumulus, L;.

This layer is in rather irregular waves or rolls, with lighter semi-
transparent parts. Occasionally stratocumulus appears in rolls which
are very symmetrically arranged; usually the layer is more or less
broken. These clouds are not formed from cumulus; otherwise they
would be coded Ly. (Photo at Washington, D. C., by W. J. Hum-
phreys.)

12. Layer of Stratocumulus, Z,.

These clouds show dark shadows but large patches of sky may be
seen between them. This layer is called stratocumulus. However,
the photograph was taken on Mount Wilson at a height of 1,750
meters. Had the observer been at sea level the clouds would have
appeared smaller and might have been called altocumulus or high
stratocumulus, requiring code Af,. (Photo by F. Ellerman.)

48

CK

49

13. Layer of Stratus, L,.

A uniform layer of stratus is like high fog and often has insufficient
contrasts in its light and dark portions to make a good photograph.
In this picture the stratus is very low and very uniform. The hill
penetrates the cloud layer and shreds of cloud (fractostratus) are
visible along the hillside. (Photo at Aberdeen, Scotland, by G. A.
Clarke.)

14. Layer of Stratus (or Stratocumulus), L;.

Here is a fairly uniform cloud sheet, with some hghter and some
darker parts, corresponding roughly to the rolls of stratocumulus but
appearing somewhat larger. Stratus is often a local cloud and when
it breaks up the blue sky is seen as is evident toward the horizon in
this picture. Sometimes, as in this case, there are transitional forms
between stratus and stratocumulus but in either case the layer is
coded L;. (Photo at Rockland, Maine, by J. Oddin.)

We
<a

Or
pA

15. Low Broken Up Clouds of Bad Weather, L,.

The picture shows patches of dark fractostratus (or fractocumulus)
against a background of thin altostratus. If the low clouds here are
remnants of preexisting cumulus, LZ, or L,, they should be so coded,
but clouds of this type which develop below altostratus or nimbo-
stratus should be coded L,;. (Photo at Cheyenne, Wyo., by H. T.
Floreen.)

16. Low Broken Up Clouds of Bad Weather, L,.

The low clouds show up dark against the lighter background of
nimbostratus which appears in places, chiefly at the upper right.
The low clouds show some relief especially where they are not badly
broken up; they appear to be fractocumulus rather than fracto-
stratus. The clouds in the background may be nimbostratus or alto-
stratus but in either case the proper code numbers are M, for the
middle layer and L, for the low clouds. (Photo at Manila, P. I.,
by C. E. Deppermann.)

52

17. Low Clouds of Bad Weather, L,.

The upper part of this picture shows a layer of nimbostratus clouds
trailing off to dense altocumulus in the distance. Below in the fore-
ground beyond the trees is a layer of dark fractocumulus (LZ) of bad
weather with a band of brighter sky between the two layers. This
photograph, which was taken at Blue Hill Observatory with a minor
front approaching, should be compared with plate 27. (Photo by
C.K. Brooks:)

18. Cumulus of Fine Weather and Stratocumulus, L,.

In this picture a layer of stratocumulus in waves is shown with
openings at the right where the sky is visible. Below this layer there
were cumulus clouds of fine weather. A part of a cumulus cloud is
shown in the right lower margin of the picture; it looks dark against
the stratocumulus. In this code type the stratocumulus is formed
independently and not derived from cumulus. (Photo at Cheyenne,
Wyo., by H. T. Floreen.)

54

ay)

19. Cumulus of Fine Weather and Stratocumulus, L,.

In this photograph there are stratocumulus clouds in large soft
masses with cumulus clouds below; the latter by themselves would be
coded L,. This combination is coded L;. If the cumulus clouds are
sufficiently developed to be coded L, or L; (pls. 20 and 21), the com-
bination of cumulus and stratocumulus should be coded Ls. (Photo
by F. W. Backer in the International Atlas.)

20. Heavy and Swelling Cumulus and Stratocumulus, L;.

Here is a layer of stratocumulus with cumulus clouds below. The
latter are heaped up and do not penetrate the stratocumulus in this
picture but often they do penetrate an existing stratocumulus layer
(pl. 21). In some cases, where the layer above is higher, it may be
altocumulus derived from cumulus clouds; in that case the altocumulus
would be coded Af, (pl. 33) and the cumulus L, or Lz, according to
circumstances. (Anonymous.)

56

57

21. Heavy Cumulus or Cumulonimbus, and Stratocumulus, Z;.

In the upper part of the picture there is a corrugated sheet of strato-
cumulus which, if it were alone, would be coded L;. However, there
are cumulus and cumulonimbus clouds below, some of which pene-
trate the stratocumulus. This combination of cumulus and strato-
cumulus clouds is coded Ly. It should be compared with plates 19
and 20. (Photo by C. J. P. Cave in the International Atlas.)

22. Heavy or Swelling Cumulus (or Cumulonimbus) and Ragged
Low Clouds of Bad Weather, L,.

Cumulus clouds are visible in the distance but a large cumulonimbus
cloud has overspread much of the sky with dark fractocumulus clouds
which occupy the upper portion of the picture. The presence of a
screen of rain falling in the middle distance is shown by the rainbows,
primary and secondary. The proper code number is Ly. (Photo at
Aberdeen Scotland, by G. A. Clarke.)

58

23. Heavy or Swelling Cumulus and Low Ragged Clouds of Bad
Weather, L,.

A large mass of cumulus has covered much of the sky and the ragged
edges of fractocumulus are seen in the direction toward which the
clouds are moving. These low ragged clouds are distinguished by. their
association with a cumulus mass and must not be coded Ls which is
reserved for low ragged. clouds below a sheet of altostratus or nimbo-
stratus (Pls. 15, 16, and 17). (Photo at Washington, D. C., by W. J.
Humphreys. )

24. Heavy or Swelling Cumulonimbus and Low Ragged Clouds of
Bad Weather, Ly.

In this picture the cumulonimbus has spread over the sky with a
shower in the left half of the picture. A ragged mass of dark fracto-
cumulus appears in the upper right corner. Whether the main cloud is
cumulonimbus or a heavy and swelling cumulus, the low ragged clouds
are coded Ly. Clouds with a mammatus structure may appear; after
they have become disassociated with the cumulonimbus, and the
latter is no longer in evidence, code Ly should be used instead of Lo.
(Anonymous. )

60

-
-

ae
NG

oy

~

98863°—38——5 61

25. Typical Altostratus, Thin, M,.

This is a typical sheet of thin altostratus, distinguished from cirro-
stratus by the absence of halo phenomena. Thesun appears as though
shining through ground glass. Below this sheet of altostratus there
are masses of fractostratus which appear very dark. The fracto-
stratus clouds which form below altostratus (pl. 15), should be coded
L, and the altostratus 14, or M;, according to circumstances. (Photo
at Aberdeen, Scotland, by G. A. Clarke.)

26. Typical Altostratus, Thick (or Nimbostratus), ©).

This picture shows a layer of altostratus which covers practically the
whole sky. Its thickness varies considerably. The sun shows in a
lighter patch where the fibrous structure is evident. If the sun were
behind the darker part of the cloud it would be completely hidden.
(Photo by M. Pulvé in the International Atlas.)

62

eae
ay

27. Typical Altostratus, Thick, M).

This view from Mount Washington shows a sheet of altostratus
above, with a band of brighter sky near the horizon. This cloud sheet
is coded M,. When progressive development results in the formation
of fractostratus or fractocumulus clouds below the altostratus (or
nimbostratus) sheet (pl. 17) the lower clouds are then coded Jj.
(Photo by C. F. Brooks.)

28. Altocumulus, or High Stratocumulus, at One Level Only, M3.

The clouds in this photograph would probably be classed as alto-
cumulus by the majority of observers. The masses are rather large,
however, and could properly be called stratocumulus. The criterion
is that the diameters of the smallest of the well-defined and regularly
arranged elements must not exceed the width of three fingers with the
arm extended if they are to be called altocumulus. In either case they
are coded M,. (Photo taken near Crozet, Va., by A. C. Lapsley.)

64

CRED ike:
RR

29. Altocumulus, or High Stratocumulus, at One Level Only, M;.

In this case the clouds are more surely altocumulus. It appears
that the cloud sheet exists at one level only. The cloud masses
here are more nearly globular than in the preceding example. Shading
is plainly visible on the under surfaces of the thicker cloudlets. The
sheet (1/;) does not increase in amount; if it increases materially it
should be coded M;. (Photo on Mount Weather, Va., by A. J.
Weed.)

30. Altocumulus in Small Isolated Patches, More or Less Lenticular
in Shape, M,.

These clouds appear on the lateral edge of a disturbance or in
mountainous districts. This picture was taken in the lee of Mount
Washington during a gale from the northwest. Clouds of this type
are sometimes called ‘‘standing’”’ clouds because they appear sta-
tionary, although they are actually in constant process of formation
on the side toward the wind and evaporation on the other side.

(Photo by C. F. Brooks.)

66

31. Altocumulus in Patches, Showing Signs of Evaporation and Being
More or Less Lenticular in Shape, M,.

This is another type in which small cloudlets are grouped in patches
more or less lenticular in shape (roughly like the cross-section of a
lens). The cloud patches are sometimes at different levels and
usually show signs of evaporation. In this code classification, the
total amount of clouds is not large and they are usually more delicate
than clouds coded /;. (Photo at Cheyenne, Wyo., by H. T. Floreen.)

32. Altocumulus Arranged in More or Less Parallel Bands, Advancing
Over the Sky, M..

In this picture the altocumulus clouds are arranged in more or less
parallel bands. The layer is of irregular thickness with signs of
evaporation in places. To be coded VM; the net result must be an
increase in the amount of cloud. Although there are signs of evapora-
tion, the proportions of the sheet and its regular structure differentiate
it from clouds coded M,. (Photo taken near Crozet, Va., by
A. C. Lapsley.)

69

33. Altocumulus Formed by a Spreading Out of the Tops of Cumulus,
M,.

In the upper part of this picture there are altocumulus clouds which
are evidently derived from the tops of cumulus clouds in the neighbor-
hood. They are fairly thick in places but are thinning out elsewhere.
When there is doubt whether these clouds are altocumulus or strato-
cumulus, they should be coded M/,; so that, as in this case, the cumulus
clouds also present may be reported, using code L, or LZ; as may be
appropriate. (Anonymous.)

34. Altocumulus Associated With Altostratus, M,.

There are two layers of altocumulus in this picture. At the higher
level there are more or less globular cloudlets, especially at the left.
There is a patch of clear sky near the center of the picture. Dark
fibrous veils of altostratus can be seen at the upper right, while the
clouds in the lower part of the photograph are altocumulus at a lower
level with some of the characteristics of altostratus. (Photo at
Cheyenne, Wyo., by H. T. Floreen.)

70

1

35. Altocumulus With a Partially Altostratus Character, M..

This is a layer of altocumulus with a rippled or corrugated under
surface. Formerly this cloud was called “wrinkled altostratus’’ but
the International definitions now place a cloud sheet of this character,
with definite relief on the under surface, in the class of altocumulus
opacus. The layer shown here should be coded M,. (Photo at Fort
Worth, Tex., by E. H. Karrer.)

36. Altocumulus Associated With Altostratus, M..

In this picture there is a layer of altostratus associated with con-
siderable amounts of altocumulus at a lower level. <A sky of this
type is commonly observed in the central region of a weak disturbance.
The photograph was taken shortly after the passage of a cold front.
(Photo at Portland, Oreg., by V. T. Horn.)

“pie ROE

13

37. Altocumulus Castellatus, M,.

In the upper part of the picture there are horizontal bands with
small masses having a vertical development (altocumulus castellatus).
In places the small masses are reduced to thin wisps. There are also
some streaks of cirrus and, near the mountains in the distance, some
swelling cumulus clouds which would be coded L;. (Photo on Mount
Wilson, Calif., by F. Ellermann.)

38. Altocumulus in Scattered Cumuliform Tufts, M;.

In this picture, especially to the left, there are small cumuliform
tufts (castellatus) with a tendency toward domed upper surfaces.
Trails of virga appear like wisps of cirrus. Above and to the right
the clouds are breaking up into poorly defined globular masses (floccus).
These clouds are indicative of steep lapse rates and are precursors of
thunderstorms. (Photo at Blue Hill Observatory, Milton, Mass.,
by C. F. Brooks.)

39. Altocumulus in Several Sheets at Different Levels, Associated
With Thick Fibrous Veils and a Chaotic Appearance of the
Sky, M,.

This picture shows a portion of a thundery sky. At the middle and
right near the upper margin there is a patch of clear sky. The clouds
in the middle and upper levels are poorly defined. These clouds were
preceded by altocumulus castellatus and altocumulus floccus and were
accompanied by thunderstorms in the vicinity. (Photo at Washing-
ton, D. C., by G. Slocum.)

40. Altocumulus in Several Sheets at Different Levels, Associated
With Thick Fibrous Veils and a Chaotic Appearance of the
Sky, M,.

The sky described by this code classification has complex and
varied aspects. In this picture there are cumulus clouds below and at
the upper right. There are also patches of altocumulus and others
with a cirriform aspect. The clouds are chaotic but not ragged.
Altogether the sky has a distinct ‘“thundery”’ tendency which is
characteristic. (Photo at Manila, P. I. , by C. E. Deppermann.)

76

98863°—38——6 WE

41. Cirrus, Fine, not Increasing; Scarce, H,.

This form is distinguished chiefly by its scarceness and also by the
fact that it does not increase in amount or tend to fuse into a cloud
sheet or pass into cirrostratus. The cirrus plume in this picture was
the only one in the sky; it did not increase in amount during the
afternoon and disappeared entirely at sundown. (Photo at Washing-
ton, D. C., by G. Slocum.)

42. Cirrus, Fine, not Increasing; Plentiful but not a Continuous
Layer, H).

These clouds are delicate in texture, although a reproduction in
black and white tends to give too thick an appearance to cirrus. It
does not differ from the type coded as Hy, except that it is more
abundant over much of the sky. It does not increase in amount.
(Photo at Cheyenne, Wyo., by H. T. Floreen.)

78

43. Cirrus of Anvil Clouds, H;.

The cirrus clouds in this picture are quite dense in places. The
wisps of “‘spindrift”’ at the upper left are characteristic of a thundery
tendency. Although the anvil shape is completely lost, these clouds
were probably derived from cumulonimbus clouds at some consider-
able time preceding the taking of this picture. (Photo taken at
Ebro Observatory, Tortosa, Spain.)

44. Cirrus of Anvil Clouds, H.

In this picture the typical anvil is seen, composed entirely of dense
cirrus, frayed out at the upper left. There are cumulus clouds below
with the cirrus of another anvil in the distance at the right. The
cumuliform parts of the main cumulonimbus have degenerated. This
is a transitional stage between L; and H;. (Photo at Cheyenne, Wyo.,
by H. T. Floreen.)

80

81

45. Cirrus Increasing, Generally in the Form of Hooks Ending in
a Point or in a Small Tuft, H,.

The streaks of cirrus in this picture have, here and there, an up-
turned end ora small tuft. It is stipulated that the cirrus in this code
classification shall increase both in time and in a certain direction,
but they shall not merge into cirrostratus, which would call for code
H;. (Photo on Mount Wilson, Calif., by F. Ellermann.)

46. Cirrus, or Cirrostratus, Advancing Over the Sky But Not More
Than 45° Above the Horizon, H;.

The strands of cirrus in this picture are seen to fuse together into a
sheet resembling cirrostratus. The advancing front of the sheet
appears to be less than 45° above the horizon. In other cases, the sheet
may, however, be of cirrostratus without any parts definitely resembl-
ing cirrus. The fact that the clouds are advancing over the sky must
be determined by observation. (Photo on Mount Weather, Va., by
A. J. Weed.)

82

47. Cirrus or Cirrostratus Advancing Over the Sky and More Than
45° Above the Horizon, H,.

Here the cirrus clouds are united in places to form patches like
cirrostratus. This form is like the preceding one except that the
clouds are more than 45° above the horizon. If the sheet covers the
whole sky, however, code H;is used. In this picture parts of the sheet
are cirrocumulus, w hich may be present in any of the code classifica-
tions from HA, to ly. (Photo at Cheyenne, Wyo., by H. T. Floreen.)

48. Veil of Cirrostratus Covering the Whole Sky, H,.

The preceding type, while more than 45° above the horizon, did not
cover the whole sky. The entire sky must be covered if the cloud is
to be coded H;, which is presumably the case in this picture. The
presence of a halo shows this sheet to be definitely cirrostratus. It
has little detail but is plainly seen to be fibrous in places. (Photo at
Aberdeen, Scotland, by G. A. Clarke.)

84

s

YD;
ey
2 a

89

49. Cirrostratus Not Increasing and Not Covering the Whole Sky, H,.

Here is a sheet of cirrostratus (and parts of cirrocumulus) with clear
sky along the horizon in one direction. If the cirrostratus covers most
of the sky and the segment of blue sky is diminishing, it is coded A.
It is only when the segment of blue sky does not diminish that code
H,is used. (Photo at Cheyenne, Wyo., by H. T. Floreen.)

50. Cirrostratus Not Increasing and Not Covering the Whole Sky, H,-

A sheet of dense cirrostratus covers nearly all of the sky. In this
case the strip of clear sky near the horizon persisted for at least 12
hours; the upper clouds moving in a direction parallel to the edge of
the sheet. The lower clouds in this picture should be coded L, and
the upper clouds, Hs. (Photo by G. A. Clarke in the International
Atlas.)

86

87

51. Cirrocumulus Predominating, Associated With a Small Quantity
of Cirrus, H.

In this picture the cirrocumulus predominates but there is some
cirrus, especially in the lower right portion. At the middle of the
lower margin the transition from cirrus to cirrocumulus is evident.
The latter is evolved by a degeneration of a mass of cirrus or a sheet
of cirrostratus. The cloudlets are without definite shadows. (Photo
at Cheyenne, Wyo., by H. T. Floreen.)

52. Cirrocumulus Predominating, Associated With a Mass of Cirrus,
H,.

The individual cloudlets in this picture are without definite shadows
and are associated with tufted cirrus in the lower portion of the picture.
According to the new definition of cirrocumulus, its relationship to
cirrus or cirrostratus is absolutely necessary. Cirrocumulus clouds
may be associated with cirrus or cirrostratus in the other classifications
but in H, the cirrecumulus predominates. (Photo on Mount Wilson,
Calif., by F. Ellermann.)

88

89

AERIAL VIEWS OF CLOUDS

The illustrations which follow have been selected to show the
upper surfaces of clouds as viewed from airplanes. When seen from
above, the clouds at lower and intermediate levels do not present
the same appearance as from below, hence it is not always possible,
from an aerial view alone, to distinguish clouds according to the 10
genera in the international classification which is based on appearance
from the ground. There are certain characteristics, however, which
should be observed.

It is desirable that airplane pilots employed on meteorological
work shall make a systematic study of clouds from the ground as
well as from the air in order that the clouds may be correctly identi-
fied. In some instances, however, it will be necessary to give de-
scriptions in plain language without code numbers or generic names.

Curus, citrostratus, and cirrocumulus clouds are usually above
the levels reached by airplane observers. When lower clouds cover
the sky the airplane observer may supply valuable information as to
high-level clouds which are not visible from the ground. Occasionally
the airplane may pass through cirriform clouds composed of snow-
flakes or ice particles at moderate heights. In some instances ice
particles may be seen sparkling in the sun when the particles are too
widely separated to be visible as a cloud to an observer on the ground.
Such conditions should be recorded.

Clouds in the intermediate and lower levels, when seen from
above, often have the appearance of a continuous horizontal layer
or sheet. Even when there are fairly large openings, the layer may
appear unbroken when viewed at a distance. Near to and below
the observer, the openings will be visible. It is important to note
whether or not there are any openings.

Clouds in sheets or layers nearly always exhibit a rippled or lumpy
surface above. The degree of roughness or turbulence should be
noted. It may range from a slightly rippled effect to a pronounced
lumpiness, with parts extending upward in the form of towers like
the upper portions of cumulus clouds.

Sometimes a layer of clouds may be pierced by cumulus or cumulo-
nimbus which develops below and grows until its summit protrudes
above the layer. This structure should be carefully distinguished
from the development of cumulus clouds in detached but numerous
masses so that their tops when viewed in perspective toward the
horizon will appear as a rippled or lumpy sheet.

The height and thickness of cloud sheets are also important.

In general, aerial observations should include notes as to the follow-
ing features:

1. Date, time, and place.
2. Cloud genera when known.
3. Height of base.

90

. Height of upper surface.

. Description of structure:

a. Detached flat clouds.

b. Detached clouds with vertical development.

c. Layer with openings.

d. Continuous layer.

. Appearance of upper surface of clouds:

a. Smooth or slightly rippled.

b. Lumpy or rough.

c. With cumuliform bulges.

d. With towers and tops resembling cumulonimbus.

. The presence of clouds at two or more levels should be noted
and described separately.

. Presence of ice, snow, or water droplets should be noted in
passing through clouds.

. Separation of the particles of ice, water, or snow, that is,
are they widely separated so that the cloud appears like
a thin mist or does the cloud appear dense when the observer
is passing through it?

91

98863°—

he
ne

so

(60)

95

re

oS

(63)

(66)

98

OE?
ag On PN ie, SE
ii oe

ay Teg?
Giana age.

(69)

100