Climate and accelerated erosion in the arid and semi-arid Southwest, with special reference to the Polacca Wash drainage basin, Arizona


Thornthwaite, C.W.; Stewart Sharpe, C.F.; Dosch, E.F.

US Department of Agriculture Technical Bulletin 808

1942


Accelerated stream trenching or arroyo cutting in the Southwest during the last half century has caused widespread damage to agricultural and range lands. Under natural conditions in the Polacca Wash and in most of the Southwest, the plant cover protected the soil and retarded the flow of water off the lands, but as a result of irritations of various forms, erosion was accelerated, and short discontinuous gullies became numerous. These lengthened and became joined to form continuous trenches. This gully cutting has been most conspicuous and destructive, but sheet erosion and wind erosion are also reducing the value of the lands. Various workers have attributed the current erosion damage to climatic change, but careful analysis of the climate shows that there has been no significant climatic change in the Southwest in the last 2000 yrs. The accelerated erosion that is damaging the lands of the Southwest appears to have been caused by man, and by proper methods man can check it and reclaim the land for his use.

Technical
Bulletin
No.
808
May
1942
UNVI4CD
STATES
DEPARTMENT
AGIttCULTURE.
117
ASIUNGTON,
Climate
and
Accelerated
Erosion
in
the
Arid
and
Semi
-Arid
Southwest
With
Special
Reference
to
the
Polacca
Wash
Drainage
Basin,
Arizona'
By
IVA
I{
EN
T
./
)1i
NTI1
WA
TH.
Ch
('
F.
ti
TE'w.‘RT
SHARPE,
S.V1C
ill
C
C011
-
t
ll
ud
EARL
F.
DOSI
tI,
rr.yYixlrrlrt
:Suit.
coif
vat
fon
fi
d.
CI
bit
ir
(in
rt
ysiogruplar.
(W
Y>
of
1?
.ii
Soil
Conservation
Serrifra
CONTENTS
Page
Page
f
ntroduct
ton
.
I
I
Erosion.
itt
the
Polacea
Wash
••contititted.
(
-
Rotates
of
The
...iouth
west
Basic
cond
it
iosis
governing
erosion
Con.
M'
eteorological
origin
of
['MDR
tic
69
Vegetation
61
Air
-mass
types
in
rho
Southwest
o.
Land
use:
Past
and
present
68
A
ir-111
12..
anti
voriai
ions
in
tempera-
Present
erosion
conditions
.
71
ture
BV
,
ek
Mcsil
stetiro)
72
Air
-masses
and
variations
in
isrtelpitli-
Tusayan
Washes
section.
. .
79
ion
. .
8
I
teteorolciaicx.d
analyses
of
setectett
E'ainted
Desert
swarm
Processes
tend
effects
of
nartuni
and
accel-
84
storms
11
erated
erosion
.
85
Variation
in
monthly
and
annual.
pre-
DeveMputent
of
the
Polaccis
drainage
cipitation
14
prior
to
the
recent
acceleration
of
Excessive
precipitation
in
the
South-
erosion
85
west
22
Recent
accelerated
erosion
91
Rainstorm
frequencies
in
the
•Soitilt-
kccelerated
erosion
in
Elie
Soul
ins
rst
102
west
17
Slate
of
acceleration
of
erosion
102
Drought
fremiencies
in
the
Southwest
ii
The
Southwest
102
Tie
climatic
pattern
in
the
Southwest
The
Polneett
tire
Inoue
basin
104
A
definition
of
climate
35
.
atises
of
acceleration
of
erosion
E07
The
normal
climatic
pattern
Diastro/thisto
107
Variations
in
the
climatic
pall
ern
Agriculture
107
Climatic
fluctuations
in
the
past
47
Climate
Efly
Erosion
in
the
Polacon
Wash
46
Basic'
conditions
coverning
erosion
.
11;
ftrazIng
Summary
and
conclusions
110
125
Climate
of
the
Poi:lent
Wash
40
Literature
cited
.
_
.
129
lealogy
57
INTROIT
UCTION
Accelerated
stream
i
renching
or
arroyo
cutting
in
the
South-
west
has
been
noticed
for
more
than
50
years.
In
that
time
the
gullying
of
channels
in
flat
-floored
valleys
with
the
consequent
dis-
section
of
bottom
land,
lowering
of
the
water
table,
and
loss
of
palatable
gra
,,
ses
hare
become
increasingly
apparent.
-
Wind
action
has
removed
soil
from
tilled
fields
and
other
areas
where
the
vegetal
cover
has
been
depleted
and
has
deposited
the
debris
in
the
form
Submitted
for
IMIA
E
ion
June
21,
1941.
2
TECHNICAL
BULLETIN
SOS,
U.
S.
DEPT.
OF
AGRICULTURE
of
dunes.
Ranges
that
once
carried
10,000
head
of
cattle
can
now
scarcely
support
one
-quarter
as
many.
Valleys
that
the
first
white
settlers
converted
into
prosperous
farms
are
now
deeply
cut
badlands
unsuitable
even
for
grazing.
The
cause
or
causes
of
the
acceleration
of
erosion
in
the
South-
west
is
a
vital
question
on
which
there
is
still
a
lack
of
general
agreement.
If,
as
sonic
workers
believe,
a
progressive
desiccation
of
climate
has
brought
about
the
dissection
of
the
western
lands
there
is
little
hope
that
man
can
stem
the
quickened
erosion.
If,
as
oth-
ers
are
convinced,
misuse
of
the
hind
by
overgrazing
and
imprudent
methods
of
agriculture
has
been
the
cause
there
is
a
good
possi-
bility
of
improving
the
land
by
improving
the
land
use.
Understanding
of
the
problem
depends
primarily
on
a
knowledge
of
the
climate
of
the
Southwest,
of
the
nature
and
extent
of
ac-
celerated
erosion
in
that
region,
and
of
the
correlation
of
accelerated
erosion
and
-land
use.
This
bulletin
contributes
an
analysis
of
the
climate
in
the
South-
west
and
of
its
relation
to
erosion
and
overgrazing.
It
is
written
in
three
main
sections,
any
of
which
may
be
read
separately.
The
fi
rst
deals
with
climat,:,
the
second
with
norned
and
accelerated
erosion
in
a
selected
drainage
basin,
arid
the
third
with
the
history
of
erosion.
The
wide
range
in
precipitation
and
temperature
at
a
singIe
sta-
tion,
the
great
variance
in
sea:onai
arid
annual
precipitation,
and
the
occurrence
of
large
storms
after
dry
periods
or
a
succession
of
several
abnormally
dry
years
are
the
most
critical
features
of
the
climate
from
the
standpoint
of
erosion
and
land
use.
The
fi
eld
studies
of
normal
arid
accelerated
erosion.
reported
in
the
second
section,
were
made
in
the
dritie;;',T
basin
of
the
Polacca
Wash
2
one
of
several
drainage
ways
by
wlikh
;he
rim-olf
from
Black
Mesa.
a
broad.
youthfully
dissected
plateau
in
the
Navajo
country.
flows
toward
the
Little
Colorado
Rive.-
(fig.
1).
In
size.
in
climate.
in
vegetation,
in
sail
and
bedrock.
am
in
past
land
use
the
Polacca
Wash
is
representative
of
conditions
much
of
northeastern
Ari-
zona.
Most
of
the
Navajo
country
ir
open
range.
Only
a
small
percentage
of
the
total
area
is
under
cultivation.
In
the
last
section.
(he
records
of
gully
cutting
and
range
deple-
tion
for
the
Southwest
.
in
general
and
for
the
Polacca
Wash
in
d
particular
are
considered.
The
evidence
for
and
against
tectonic
disturbanc-es,
isturbanees,
agriculture,
overgrazing.
and
climatic
change
as
causes
of
accelerated
erosion
is
weighed.
THE
CLIMATES
OF
THE
SOUTHWEST
METEonoLoGicm,
ORIGIN
OF
CLIM
MU:
FL
MT('
ATioNs
AIR
-MASS
TYPES
IN
THE
SOUTH
WEST
The
climatic
pattern
on
the
earth
(100,
4
as
well
as
the
changes
in
its
position
from
year
to
year.
is
explained
in
terms
of
at
-
The
"wgsii,"
at
eharaeteriszir
nernrai
land
form
of
lite
Southwest,
is
a
liar
-floored
oiliey
generally
one
10
several
tidies
wide.
carries
water
th
ri
ll
told
ly
and
may
or
luny
ma:
contain
FL
steep
-walled
rhennei.
'Assistance
In
the
ellnuttle
analysis
was
giV1.11
ily
David
r,
Riunienstovic,
of
the
Climatic
and
Physingrapide
Haile
ntinthes•
in
parentheses
refer
in
Literature
Cited,
it.
as).
CLIMATE
AND
ACCT:
ERATED
EROSION
LN
THE
SOUTHWEST
3
C
lo
a
at
r
1eq
t.
—•
A
D
V.A111110
PLATEAU
°
vaEli.6P1
PLATEAU
.e
27
I
VIOVUUENT
VALLEY
5EG1
vE5A5
B
..
,./11
A
C
K
:
a
,
,
7
z_l
i
__ L
f_.
,.,
_./
/
F-
I
;9
5
5"
.7
2
4.
'
.
r"
0"
-
ME SA
(
CLF SAY
A
N
I
4'
..,
/ ,
0
7
. r
°-
%`,"
-.'
4
P
C
'‘
i
At
,i•
;
/..
WI
S
k
E
5,
0
L-
p,
L
Acc.
wS
ik
aPEA
v.LE5
0
5
10
so
-
Ff
0
E
5
I.r.ac
Q
A
A
Ftouitr:
Slmtliwest
and
the
p1ipzi(1griiphie
iq
the
PoTiiemi
Wash
arim,
Ariz.
4
TEUELVICAL
BULLETIN
808,
U.
S.
DEPT.
OF
AGRICULTURE
mospheric
circulation.
Over
North
America
the
circulation
consists
of
fl
ows
of
great,
bodies
of
air
which
have
remained
in
their
various
source
regions
long
enough
to
have
acquired
special
individual
prop-
erties.
Outward
movement
of
the
air
from
these
centers
mid
inter-
action
of
the
different
types
of
air
masses
are
chiefly
responsible
for
the
weather
of
the
continent.
These
air
masses
are
of
several
types.
Air
which
fl
ows
south-
ward
from
the
vast
Arctic
tundra
of
northern
Canada
is
cold,
dry,
and
heavy.
The
North
Pacific
Ocean
is
the
source
region
for
cool
to
cold,
moist,
and
mode
.
eately
heavy
air.
Air
over
the
north
At-
lantic
Ocean
also
develops
similar
properties.
Over
the
Gulf
of
Mexico
and
the
Caribbean
region,
and
also
over
the
tropical
waters
of
the
Pacific,
air
bodies
become
warm
to
hot,
very
moist,
and
light.
The
southwestern
part
of
the
United
States,
together
with
the
Mexican
plateau,
is
itself
a
source
region,
where
air
from
upper
levels
sinks
to
the
surface
and
heroines
hot.
light,
and
very
dry
(1
1
5.
112).
The
source
regions
for
these
various
air
masses
are
shown
in
their
approximate
positions
in
fi
gure
2.
The
entire
system
of
air
masses
is
displaced
poleward
in
summer
and
equatorward
in
wi
Titer.
Air
fl
ows
outward
from
all
these
source
regions.
All
three
polar
air
masses
generally
inure
in
a
southeasterly
direction.
The
tropical
air
masses
generally
move
in
a
northeasterly
direction.
The
trajec-
tory
of
the
tropical
air
from
the
Atlantic
characteristically
curves
across
the
Gulf
of
Mexico.
up
the
Mississippi
Valley.
and
thence
eastward
back
to
the
Atlantic.
However.
despite
the
fact
that
these
air
bodies
have
preferred
routes,
all
except
those
which
originate
in
the
north
Atlantic
sometimes
enter
the
Southwest.
It
is
the
invasion
and
interaction
of
air
masses
that
accounts
for
the
day-to-
day
variations
i]1
the
weather
of
that
region.
Cool,
moist
Polar
Pacific
air
may
move
down
the
coast
and
swimur
in
over
the
mountains
to
invade
the
Southwest
aloft.
Such
invasions
are
especially
well
marked
dining
the
winter,
when
the
Aleutian
low-pressure
area
is
well
-Cleve'•
„,2d
and
is
at
its
southernmost
po-
sition.
Cold.
dry
Polar
Continental
air
may
push
equatorward
from
the
Canadian
tundra
and
enter
the
area.
This
is
also
pre-
dominantly
a
winter
phenomenon.
Also
during
the
winter.
warm
and
somewhat
moist
air
may
move
in
at
high
levels
from
its
place
of
origin
over
the
tropical
waters
of
the
Pacific.
Particularly
dur-
ing
late
spring,
summer.
and
early
fall,
invasions
of
warm.
moist
Tropical
Gulf
air
from
the
southeast
may
occur.
At
any
time
of
year,
hot
dry
Tropical
Continental
air
may
descend
to
the
surface
from
aloft.
Air
masses
do
not
follow
one
another
in
any
tlefinite
sequence,
nor
are
any
two
invasions
of
one
air
-mass
type
exactly
the
same.
Air
masses
remain
over
a
source
region
for•
different
lengths
of
time
and
follow
different
trajectories.
Each
invasion,
therefore,
has
a
history
different
from
that
of
any
preceding
or
subsequent
one
so
that
when
an
air
mass
arrives
in
the
Southwest
its
properties
are.
never
exactly
the
same
as
those
of
any
other•
invading
air
mass.
Polar
Pacific
air•
entering
the
Southwest
is
occasionally
colder
than
Polar
Continental
air
or
more
moist
than
Tropical
Gulf
air.
Be-
cause
of
such
variations
in
their
properties,
air
masses
can
be
char-
acterized
only
in
a
relative
sense.
CLIMATE
AND
ACCELERATED
EROSION
E\
-
IRE
SO
CTITWEST
5
AIR
MASSES
AND
VARIATIONS
IN
TEMPERATURE
In
the
interaction
of
air
masses.
of
different
density
the
heavier
air
will
move
along
the
laud
surface
and
will
displace
the
lighter
air
upward.
Since
there
is
p-enerally
a
direct
relation
between
density
of
air
and
temperature
the
replacement
of
one
air
-mass
type
by
another
in
the
Southwest
results
in
lane
variations
in
the
Cod
moist
(Po?or
Allon
Cold
dry
(Polar
Cordfnerdal)
COW
moist
Pow
Pacific)
Hot
di)/
---c
trvcci
,
ial
ConlircrIlol)
t3
0
O
0
ql
Warm
axial
ropiud
Manna
Wawa
mold
(Tropical
1
)
oc/E3c)
FIGURE
2.
—Source
regions
of
North
Anierienn
nil'
niassPs.
6
TECEUNICAL
BULLETIN
SOS,
U.
S.
BF.PT.
OF
AGRICULTURE
surface
temperature
of
the
region.
Superimposed
on
the
diurnal
and
seasonal
temperature
rhythms,
arising
from
the
rotation
of
the
earth
on
its
axis
and
its
revolution
about
the
sun.
are
day-to-day
variations
caused
by
nouperiodic
inundations
of
air
bodies
in
un-
predictable
sequence.
These
three
types
of
temperature
variation
are
well
illustrated
by
the
daily
temperature
data
for
Santa
Fe
for
1870
and
1880.
the
warmest
and.
coldest
years,
respectively,
in
the
period
1850-1939
(fig.
3).
The
temperature
charts
for
these
years
illustrate
the
components
of
the
temperature
regime
and
show
what
lies
behind
temperature
variations
from
year
to
year.
The
length
of
the
bars
indicates
the
span
between
maximum
and
minimum
daily
temperatures.
The
curve
fi
tted
to
the
mean
daily
temperatures
for
a
46
-year
period
is
shown
on
the
chart
for
both
1879
and
1880.
The
march
of
the
daily
means
in
both
these
years
follows
the
trend
of
this
curve,
but
the
deviations
of
the
daily
means
from
the
normal
is
often
large,
a
few
being
more
than
half
the
range
of
the
curve.
These
deviations
reflect
the
variability
introduced
by
the
invasion
of
air
masses.
The
variation
in
the
range
of
diurnal
temperature
from
season
to
season
is
shown
by
the
differences
in
the
lengths
of
the
bars.
During
early
summer
the
range
is
greatest.
insolation
being
at
a
maximum
and
cloudiness
low.
The
mean
Jane
range
for
the
46
-year
period
is
25.8°
F.,
the
December
range.
20.3°.
Changes
in
temperature
from
day
to
day
can
readily
be
seen
by
comparing
the
position
of
adjacent
bars
on
the
temperature
scale.
The
mean
annual
temperature
of
52.5°
F.
in
1879
is
unusually
high.
principally
because
it
includes
unseasonably
high
winter
tempera-
tures
for
that
.
Vea
r.
The
summer
temperatures
also
are
slightly
above
average.
In
1880,
when
the
average
annual
temperature
was
only
45.1°.
winter
and
autumn
temperatures
were
far
below
normal
and
summer
temperatures
were
slightly
so.
The
high
incidence
of
warm
temperatures
during
the
winter
of
1879
was
associated
with
frequent
invasions
of
warm
air:
whereas.
in
the
winter
of
1880.
cold
air
masses
from
the
north
occupied
the
area
much
of
the
time.
Variation
in
the
mean
annual
temperature
from
year
to
year
for
the
period
1874-1939
is
shown
in
figure
4.
Each
of
the
mean
values
should
be
thought
of
as
representing
a
sequence
of
weather
condi-
tions
such
as
that
illustrated
in
figure
3
for
the
years
1.879
and
1880.
Occasionally
warm
years
fol
low
warm.
and
cold
years
may
follow
cold
:
frequently,
as
in
1.879
and
1880,
extreme
shifts
are
displayed
from
one
year
to
the
next.
Since
air-inass
invasions
great
irregularity
to
the
daily
temperature
values
it
is
not
surprising
that
the
annual
fi
gures
made
up
of
these
highly
variable
components
should
themselves
vary
considerably
from
year
to
year.
The
diurnal.
seasonal.
and
annual
variations
displayed
by
the
tem-
perature
of
Santa
Fe
are
representative
of
temperature
variations
elsewhere
in
the
SoutInvest.
There
is,
for
instance,
Flagstaff.
Ariz.
(table
1).
as
at
Santa
Fe,
a
much
greater
range
in
tine
mean
monthly
temperatures
in
winter
than
in
summer
(97).
Furthermore,
the
wide
range
in
temperature
at
Santa
Fe
(fig.
3)
is
character-
istic
of
the
range
at
other
stations
(fig
131
.
In
Arizona.
where
the
absolute
sage
is
greater
than
in
many
other
parts
of
the
country,
the
greatest
difference
between
the
maximum
and
minimum
at
any
100
90
BO
lih
70
44
60
11,
0
r
0
100
90
80
70
60
:
1'
4
'1
50
a
nu
'1
'
12
40
30
20
h
'
0
0
JAN.
FEB.
MAR.
APR.
MAY
JUNE
JULY
AUG.
10
20
J
10
20
20
10
2'0
f
ax
mum
w.
Daily
mean
Standard
average
46
-year
record)
Minimum
-
Daily
precipitation
tO
20
14
ia
is
'
.
(3
I
7
-2"ii
1879
.. 11. Jiiid
4..._.
SEPT.
10
2
1
0
1880
OCT
NOV,
DEC,
10
20
10
90
10
20
J,
Hamm:
3,
—Dally
maximum
and
minim
mm
temperatures
and
daily
precipitation
at
Santa
Fe,
N.
Alex.,
1879
t
nd
1880.
8
TECIli
BULLETIN
SOS,
U.
S.
DEPT.
OF
AGRICULTURE
station
is
134°.
This
is
the
record
for
Keams
Canyon,
where
the
maximum
was
104°
and
the
minimum
—30°.
Though
the
pattern
of
change
is
similar
at
stations
throughout
the
Southwest,
there
is
considerable
difference
in
the
range
of
tempera-
ture
at
different
stations
and
in
th
e
frequency
of
occurrence
of
vari-
ous
temperatures.
The
areal
contrasts
are
brought
about
mainly
by
variations
in
elevation,
cloudiness.
and
the
incidence
of
different
air
masses;
perhaps
to
a
small
extent
they
reflect
differences
in
in-
solation
caused
by
variation
in
altitude.
The
highest
temperatures.
are
experienced
at
Yuma,
which
is
at
the
lowest
elevation
above
sea
level,
and
which
also
has
the
least
cloudiness.
Lowest
temperatures
occur
at
stations
in
the
northern
part
of
the
region,
which
are
at
the
highest
elevations.
5
50
45
48
47
46
41
....
.
L
--'.
.
-
:',.-.'
111
i
FicuRE
4.-..
.A1pvit
:tumid;
ion
pi'vnturo
ii
Saat:i
Fe.
1S74-1939.
TABLE
1.
11
ighesi
awl
fairest
arorage
monthly
and
annual
tempera
/lire
and
ngir.
,
.
fl
agstaff.
151.11
-10.31/
I I
1
,
.Mes.1
Et.
19
lre
I
LE)WeSt
YI
L
"
'
ref
113tY
Yeilr
111Erip
..111
a
i
airy
February
Mardi
April
May
June
34.
4
39.
(1
41.5
5241
411.
0
GI.
7
900
1104
.
9142
rz95
:
99T
At;
12.
Ii
19.
5
`.'9,
9
:37.7
44.5
54.2
M:17
,
1'+:111
'
1917
19151
11111
191171
5!.
9:
9.
7
2,
5
-1.3
11.
5
11.
5
July
.
72.
0
497
5!.
2
:
1912
EE.
9
.1.
ogusE
71.4
W17
1
50.7
i
111041
13.7
September
(13.
1
997
52.:3
I
/912
(I.
S
0.040
ier
.
5.5.
9
991
41,
2
11109
4,
a
November.
.
.
4.1-I
51i7
71.:4
0419
:i.
/
December
as.
D
997
I
K.
9
19115
9_
2
•-
-
A
Initial
•ES
S
'914
42.9
1915
5.9
AIR
I4ASSES
AND
VABIATIONS
IN
/TEC/1'11%1110N
Precipitation
cannot
take
place
unless
the
air
is
cooled
sufficiently
to
release
the
atmospheric
naiistuip.
The
necessary
eool
mg
is
arroni-
plished
by
adiabatic
expansion
wherever
air
flows
up
a
land
slope.
Air
is
similarly
cooled
when
it
encoonterc:
;t
heavier
air
mass
and
is
forced
up
the
air
-mass
slope.
Air
ascends
and
is
cooled
by
convec-
tion
when
an
tilt'
column
is
made
unstable
through
heating
at
the
ground
or
through
cooling
aloft
by
radiation
from
the
tops
of
clouds.
The
principal
cause
of
precipitation
in
the
Southwest.
as
else-
cumAn
AND
ACCELERATED
EROSION
IN
THE
SOUTHWEST
9
where
in
the
United
States.
is
the
lifting
of
air
along
zones
of
dis-
continuity,
or frOntS
between
adjacent
air
masses
of
different
prop-
erties.
Two
general
types
of
fronts
are
recognized.
In
one,
the
colder
air
pushes
actively
into
the
area
occupied
by
the
warmer
air,
forcing
the
warmer
air
aloft.
In
the
other.
the
warm
air
mass
is
the
active
one.
moving
at
a
more
or
less
uniform
rate
up
the
slope
of
a
relatively
stationary
or
slowly
inovin!
,
cold
air
mass.
Whichever
air
mass
may
be
the
active
one,
precipitation
will
usually
occur
if
the
ascending
air
is
moist.
When
the
cold
air
is
advancing
rapidly
its
front
is
relatively
steep.
The
warm
air
therefore
ascends
rapidly
over
a
relatively
narrow
belt.
and
thunderstorms
usually
result.
They
cover
relatively
.s
mall
areas.
are
extremely
spotty.
display
high
intensities.
hut
continue
for
only
a
short
time
at
any
one
place.
When
the
pressure
gradients
are
such
that
warns
air
tletively
pushes
over
cold
air.
the
flow
is
usually
steady
and
uniform
over
wide
areas.
If
rains
result,
they
are
widespread.
(end
to
la
,
homogeneous.
are
of
low
to
moderate
intensity.
and
may
sometimes
continue
for
several
days.
During
summer.
heating
of
the
ground
causes
thermal
convection,
which
ordinarily
does
not
produce
much
rainfall.
However.
when
air
masses
characterized
hr
such
atmospheric
instability
flow
over
denser
air
bodies
or
up
a
slope
of
the
land.
considerable
precipitation
may
result.
Winter
precipitation
in
the
Southwest
is
due
almost
entirely
to
movement
of
relatively
warm
air
over
temporarily
stationary
or
slowly
moving
cold
air
masses.
Motif
commonly.
fresh.
moist
air
from
the
north
Pacific
region
overrides
air
from
the
same
source
which
is
already
in
the
Southwest
and
which
has
been
cooled
by
radiation.
or'
it
may
naive
np
over
colder
and
denser
all
front
Canada,
which
has
moved
in
directly
or
has
swung
in
from
the
east
and
south.
Moist
air
front
tropical
Pacific
waters
mar
yield
precipi-
tatimi
in
the
Southwest
by
overriding
cold
air
from
either
the
north
Pacific
or
Canada.
Also
moisture
is
precipitated
from
fresh
Polar
Pacific
air
masses
directly
in
the
form
of
ii
hi.
scattered
showers
or
snow
flurries.
The
fresh
polar.
air
is
cold
and
becomes
topheavy
after
the
lower
layers
are
heated
during
the
clay.
The
result
ing
convective
currents
manifest
themsel\
-
es
through
the
development
of
cumulus
clouds.
which
by
late
afternoon
Iry
evening
may
release
small
amounts
or
precipitat
ion.
These
small
storms
occur
only
in
the
wake
of
the
pa
ss
ag
e
of
a
cold
front
of
Polar
Pacific.
air
and
contribine
little
to
the
total
winter
precipitation.
Although
the
moisture
content
of
the
air
masses
in
winter
is
rela-
tively
low
because
of
lower
air
temperatures.
winter
is
generally
a
rainy
season
because
of
t
\•igor
of
air
-mass
interactions.
En
summer.
there
i
:
4
much
less
contrast
in
the
properties
of
the
various
air
masses
mid
the
.
fronts
between
successive
advances
cif
(tit
f
ro
m
the
Pacific
are
le,s
pronounced.
Furthermore.
inroads
of
cold.
dry
air
from
Canada
become
less
and
less
frequent
so
that
they
are
of
slight
importance
in
releasing
moisture
as
precipitation.
How-
ever.
the
moisture
content
of
air
masses
.froin
the
ocean
is
higher
I
Il
the
ti
lmaihler
th
an
at
any
other
time.
and.
the
air
'INA
V
he
convectively
immutable
so
that
moisture
is
111[Ire
easi
ly
released.
Durin
!,
the
sum-
mer.
insulation
is
at
a
maximum
and
instability
of
the
air
is
induced
10
TECHNICAL
BULLETIN
808,
U.
S.
DEPT.
OF
AGRICULTURE
7
C6a1
rnorikoi
our
A
A
I
TOril
,
n•
"-•
e
ok`
-
""
Cord_
-
„or,
B
Cold
--'
-
Ot
:r
oir
-Li
.
at
Coidesr
Cool
'Nam
Cool
meo.ren•
rr
1
initnI"e
plegl
,
Cola
FIGURF,
5.--Diagrammatie
representation
of
four
meteorological
situations
‘vilieb
prodnee
precipitation
in
the
Southwest.
Arrows
indicate
direction
of
air
-
mass
fl
ow;
solid
black
cold
front
:
dotted
line,
warm
front;
dashed
line,
upper
front
;
dash
-dot
line,
occluded
front.
A
and
B
are
typical
of
winter,
C
and
D
of
summer,
CLEVL-ITE
AND
ACCELERATED
EROSION
rg
THE
SOUTHWEST
11
by
heating
of
the
ground
during
the
day
and
radiational
cooling
of
moist
layers
aloft
at
night.
As
a
result,
brief,
intense,
local
thun-
derstorms
take
place.
-
Usually
they
occur
along
cold
fronts,
which
exist
between
successively
invading
masses
of
cool,
moist
air
from
the
Pacific
or
where
warm.
moist
air
from
the
Gulf
of
Mexico
is
forced
up
by
cool
air
entering
the
region
from
the
north
Pacific.
Although
these
fronts
are
often
ill-defined.
they
frequently
provide
the
means
of
obtaining
precipitation
from
air
already
made
unstable
by
insole
-
lion
and
radiational
cooling.
For
these
reasons
summer
is
also
a
rainy
season.
Four
meteorological
situations
which
result
in
precipi-
tation
in
the
Southwest
are
represented
diagrammatically
in
much
simplified
form
in
figure
5.
Spring
and
autumn
are
periods
of
transition
in
which
neither
convective
storms
nor
extensive
warm
-front
storms
are
well
-developed.
Both
are
therefore
seasons
of
low
rainfall.
In
late
summer
and
early
autumn
occasional
tropical
cyclones
over
the
Pacific
off
the
coast
of
lower
California
may
induce
an
inflow
of
moist
Tropical
Pacific
air.
which
is
forced
up
over
Polar
Pacific
air
already
occupying
the
Southwest.
Widespread,
heavy
rainfall
generally
results.
September
rainfall
may.
therefore,
occasionally
equal
or
surpass
in
ainotint
and
intensity
that
recorded
during
the
period
of
summer
thunderstorms.
Usually.
however,
September
is
characterized
by
rains
of
less
intensity
and
smaller
total
amounts
than
are
July
and
August.
METEOROLOGICAL.
ANALYSES
OF
SELECTED
STORMS
To
illustrate
the
effects
of
certain
meteorological
conditions.
four
rains'
orzns
in
which
large
,
amounts
of
precipitation
fell
in
Arizona
were
selected
for
study
and
are
presented
in
figure
6.
The
storm
of
December
15-16.
1908,
is
a
typical
general
winter
storm
in
which
the
moisture
was
precipitated
from
warm.
moist
air
forced
upward
by
an
invasion
of
cold
Polar
Pacific
air.
On
the morning
of
the
15th,
the
front
of
a
mass
of
Polar
Pacific
air
extended
across
Arizona
in
a
southwest
-northeast
direction.
The
cold
front
had
become
quasi
stationary.
and
one
of
the
minor
waves
on
the
front
began
to
inten-
sify
and
develop
into
a
well-defined
extra
-tropical
cyclone.
This
caused
widespread
rain
by
forcing
warm,
moist
air
from
the
south
to
ascend
over
the
colder
and
denser
air
to
the
north.
The
cyclone
moved
little
during
the
next
24
hours.
and
by
the
morning
of
the
16th
it
began
to
occlude.
By
evening,
colder
air
had
pushed
into
Arizona,
and
the
rain
ended.
The
storm
of
July
21-24.
1915.
brought
considerable
amounts
of
rainfall
to
northern
Arizona.
The
meteorology
of
this
storm
is
more
characteristic
of
wilder
conditions,
but
since
this
particular
combination
of
air
masses
occurred
in
summer
and
involved
air
which
was
convect
ively
unstable.
the
resulting
pattern
illustrates
both
the
features
of
warm
fr:int
widespread
rain
and
convective
spotty
rain.
At
the
beginning
of
the
storm
perirml
a
widespread
mass
of
Polar
Canadian
air
covered
mast
of
the
United
Stales
and
extended
across
the
Southwest
into
northern
Mexico.
the
from
being
roughly
tuiniyNos
or
thi•Ne
storm%
Wire
1111/(1t•
n=itric
l'itysiographit.
by
lierukimin
1101Aninti,
of
rite
12
TECHNICAL
BULLETIN
808,
U.
S.
DEPT.
OF
AGUICULTURE
F
61,
ff
r
lri
[V I
dG
••••54
-
:
iI
iIL
12
"9s
PRECIPITATION
IN
INCHES
DEC
15
-
16,1909.
c-741
dipta
c)
O
4.9
0
1
1-
1_7
Ii
1St
P.41
a
ii
ra
PRECIPITATION
IN
INCHES
PUG
210914
arm
,
(,):\
12
cS"
VT
2
25
PRECIPITATION
IN
IN=HES
54
Pr
4
-
7_1949
11
4
,
1,
---
25
7:
lwAh
0.50
100
1.50
2.50
350
4.50
INCHES
11.—SP1e.eleil
rainstorms
In
the
Southwest.
CLIMATE
AND
ACCELERATED
EROSION
IN
THE
SOIMIWEST
13
parallel
to
the
international
boundary.
Above
the
Polar
Canadian
air
in
the
Southwest
was
moist
Tropical
Maritime
air.
During
the
evening
of
the
21st,
an
extensive
nn
-1s
of
Polar
Pacific
air
slowly
pushed
eastward
across
Washington.
Oregon.
Nevada,
and
the
South-
west.
This
air.
being
less
dense
than
the
surface.
Polar
Canadian
air
and
more
dense
than
the
Tropical
Maritime,
tended
to
wedge
them
apart.
The
advance
of
the
Polar
Pacific
front
above
the
Polar
Canadian
mass
from
the
west
caused
an
accelerated
movement
of
Tropical.
Maritime
air
aloft
from
the
south
and
resulted
in
wide-
spread
rain.
These
conditions
persisted
through
the
22c1,
and
on
the
23d
the
front
of
the
Polar
Pacific
air
mass
had
moved
only
slightly
eastward.
Not
until
the
24th
did
the
storm
end.
This
storm
illus-
trates
the
generalized
conditions
illustrated
in
fi
gure
5,
B.
The
storm
of
August
28,
1934,
is
typical
of
summer
and
illustrates
the
spottiness
of
rainfall
when
a
cold
front
forces
aloft
air
already
convectively
unstable.
It
was
associated
with
an
invasion
of
a
simple
cold
front.
On
the
morning
of
the
27th,
a
mass
of
Polar
Pacific
air
entered
the
west
coast
States.
This
caused
an
acceleration
in
the
flow
of
warm.
moist
air,
probably
Tropical
Atlantic,
from
the
south.
By
evening
the
surface
winds
at
Flagstaff
and
Phoenix
had
shifted
from
east
to
northwest,
indicat
ing
that
the
front
of
the
Polar
Pacific
:fir
had
passed
these
stations.
During
-
the
morninz
of
the
28th,
the
front
became
quasi
stationary
ill
eastern
Arizona.
Later,
it
continued
eastward
and
by
evening
had
passed
out
of
the
region.
Many
scat-
tered
thunderstorms
were
occasioned
by
the
passage
of
this
front.
The
storm
of
September
4-7.
1939,
owed
its
very
large
amount
of
rainfall
to
a
tropical
cyclone
which
moved
up
the
California
coast
from
the
tropical
waters
oil'
the
west
coast
of
Mexico.
The
average
September
precipit:ttion
for
northern
Arizona,
based
on
a
45
-year
record,
is
1.34
inches.
The
rainfall
of
September
1939
surpassed
all
existing
records.
with
4.87
inches.
For
the
State
as
a
whole.
38
stations
received
the
greatest
total
September
precipitation
on
record,
and
new
records.
exceeding
amounts
for
any
previous
month
of
the
year.
were
esktblished
at
17
stations.
A
large
proport
ion
of
this
precipitation
occurred
in
the
storm
of
September
4-7.
On
September
3d.
there
had
been
an
upper
-air
invasion
of
Polar
Pacific
air
overriding
the
modified
Polar
Pacific
air
already
occupy-
ing
the
region.
On
the
4th.
the
upper
-air
mass
migrated
across
Arizona
in
th
e
early
morning.
General
rains
accompanied
by
con-
siderable
thunderstorm
activity
occurred
throng
-
lumi
the
central
and
northern
sections
of
the
State.
Rainfall
amounts
generally
averaged
over
0.50
inch.
Six
stations
in
northern
Arizona
received
over
1
inch
:
the
largest
amount
;
3.32
inches.
was
reported
at
Truxton.
(-;eneral
rains
continued
on
the
5th.
when
moisture
was
being
precipitated
from
a
Tropical
Pacific
air
mass
that
had
invaded
California
and
Arizona
from
the
south.
The
vigor
of
this
invasion
was
due
to
it
tropical
cyclone
oft
the
Pacifie
coast.
which
of
to
the
southward
of
Acapulco
on
the
5th
and
was
dissipated
over
the
upper
part
of
Lower
California
on
the
12th.
In
northern
Arizona,
14
scat
ions
received
more
than
1
inch
of
rain;
7
stations.
more
than
2
inches:
and
2
stations,
nnre
than
3
inches.
On
the
6th.
the
moist
tropical
air
continued
to
flow
over
the
Southwest
and
general
rains
combined
with
thunderstorms
per-
F.
„14
TECHNICAL
BULLETIN
808,
U.
S.
DEPT.
OF
AGRICULTURE
,
sisted.
In
northern
Arizona,
20
stations
reported
over
1
inch
of
rain
and
6,
over
2
inches.
On
this
day
a
new
mass
of
Polar
Pacific
air
moved
into
the
Pacific
Northwest
and
by
the
7th
had
readied
the
Southwest.
The
moist
tropical
air
flowed
up
the
slope
of
this
dense
air
mass.
Rain
continued
for
several
days
but
never
in
such
large
amounts
as
had
been
recorded
in
the
3
-day
period
from
the
4th
through
the
6th.
The
tropical
cyclone
of
September
5-12
was
only
one
of
three
which
moved
up
the
coast
of
Lower
California
during
the
month.
The
second
was
particularly
violent
over
and
in
the
vicinity
of
the
mouth
of
the
Gulf
of
California,
and
the
third
did
much
damage
in
southern
California.
It
was
the
most
severe
tropical
storm
that
has
ever
been
observed
in
that
region
(68,7).
358).
The
severity
of
tie
storm
along
the
coast
it.;
indiolte.(1
by
a
low
of
45
11ve4
at
sea,
and
a
property
damage
approximating
1;2.000,000,
mostly
to
shipping,
shore
structures,
power,
and
communication
Iitte,
and
to
crops.
Unprece-
dented
September
rains
accompanied
the
storm
along
the
southern
California
coast.
VARIATION
IN
MONTHLY
AND
ANNUAL
PRECIPITATION
Rainfall
totals
for
a
month
or
a
year
are
aggregates
of
individual
rains.
Hence
the
amount
of
variation
from
one
year
to
another
in
monthly
and
annual
rainfall
is
due
entirely
to
variations
in
the
number
of
occurrences
and
the
size
and
position
of
individual
storms.
Similarly,
variations
in
rainfall
from
one
region
to
another
are
to
be
accounted
for
in
terms
of
rainstorm
size.
position.
and
frequency.
Fort
Defiance,
in
1853,
furnished
a
complete
year's
record
of
pre-
cipitation
in
northern
Lrizona.
Records
have
been
continuous
since
1890
at
Natural
Bridge,
since.
1897
at
Flagstaff,
and,
with
an
inter-
ruption
of
only
4
months,
since
1876
at
Prescott.
Within
the
period
for
which
rainfall
observations
have
been.
made
at
these-
stations,
the
range
in
annual
precipitation
amounts
has
been
large
(fig.
7).
The
largest
amount
of
precipitation
recorded
in
a
year
at
any
Weather
Bureau
station
in
northern
Arizona
was
50.17
inches
in
1905
at
Natural
Bridge.
This
station
has
been
in
operation
for
50
years,
and
during
that
time
the
annual
precipitation
in
over
half
of
the
years
was
less
than
20
inches.
The
minimum,
12.28
inches,
fell
in
1900.
At
the
other
extreme
is
Leupp,
whose
greatest
precipitation.
9.09
inches
in
1915,
is
less
than
the
maximum
recorded
at
any
other
station
in
northern
Arizona.
At
Leupp,
the
least
rainfall
received
in
any
year
for
which
there
is
a
record
was
2.52
inches
in
1938.
The
aver-
age
for
the
13
years
of
record
is
6.27
inches.
The
maximum
and
minimum
annual
rainfall
for
stations
in
north-
ern
Arizona
having
a
record
of
at
least
10
years
is
shown
in
fi
gure
8.
Throughout.
Arizona
and
in
western
New
Mexico
there
are
"rainy"
seasons
in
winter
and
summer
and
"dry"
seasons
in
late
spring
and
autumn.
This
double
peak
in
the
distribution
of
rainfall
disappears
both
to
the
west
and
to
the
east.
In
California,
the
only
rainy
Nason
is
in
winter,
summer
being
drier
than
either
spring
or
autumn.
In
eastern
New
Mexico
there
is
no
winter
rainy
period,
the
precipita-
tion
being
at
a
minimum
in
January
and
at
maximum
in
July.
The
transition
from
California
to
New
Mexico
is
shown
in
figure
9.
cuaTATE
AND
ACCELERATED
EROSI'
aN
EN
TEE
SOULELVITEST
15
1870
1880
1850
50
45
Natural
Bridge
4o
351
4
25
20
15
YEA
RS
1900
1910
1926
1930
1940
40
Jerome
I
25
29
Is'
5
0
35
30
z
a
.
••
1-
20'
1
5
a
Prescott
30
I
Walnut
Grove
20
IS
10
5
0
1
30—
--
-
--,Flagstaff
20
-
15
T
Holbrook
to'
5
i
1870
19130
ra
t -t
1890
1900
1910
YEARS
1990
1930
1340
1,—Annuni
prey-ipitation
11t
selected
r.ltions
in
northern
Arizona,
1870-
IVO.
16
TECHNICAL
BULLETIN
SOS,
U.
S.
DEPT.
OF
AGRICULTURE
Despite
large
variations
in
annual
precipitation,
there
is
a
striking
similarity
in
the
pattern
of
monthly
distribution
within
each
of
these
three
rainfall
regions.
PRECIPiTATION
(INCHES)
10
20
30
40
50
60
.7
;
•::
77
MAXIMUM
ANNUAL
PRECIPITATION
En
MINIMUM
ANNUAL
PRECIPITATION
FIGURE
8.
—Maximum
and
In
illin111111
rIITHPith
tion
fa
stafions
iIF
northern
Arizona
having
at
le:nit
10
yenrs
of
record
through
1939.
Precipitation
varies
widely
from
month
to
month.
and
even
at
the
rainiest
of
stations
as
many
as
10
of
the
12
months
have
in
one
year
or
another
experienced
a
complete
absence
of
rain.
For
Natural
Bridge,
the
only
station
in
northern
Arizona
with
a
continuous
CLIMATE
AND
ACCELERATED
EROSION
IN
THE
SOUTEMEST
17
record
for
50
years,
fi
gure.
10
shows
the
frequency
of
occurrence
of
various
monthly
precipitation
amounts.
At
this
station,
the
likeli-
hood
of
complete
absence
of
rain
is
greatest
in
May
and
June
and
least
in
July
and
August.
The
latter
two
are
the
only
months
which
have
not
been
rainless
during
the
50
-year
period.
On
the
other
hand.,
frequencies
of
largest
monthly
amounts
are
greatest
in
Decem-
ber,
January,
and
February.
Monthly
totals
above
3
inches
were
never
experienced
during
May
and
June.
20
0
!0
re
0
20
NELLIE,
CALIF.
47.97
in.
FLAGSTAFF,
ARIZ.
KEAMS
CANYON,
ARIZ.
SANTA
FE,
KM
21.92
in.
12.42.
in.
14.27
in.
4'
4
tit
1
0_
PASADENA,
CALIF.
15.17
in.
20
-
-
s
z
O
10-
a
*!,
NEEDLES,
CALIF.
NATURAL
BRIDGE,
ARIZ.
CLOVIS
N
M.
4.45
In.
23.9!
in.
1859
In.
-r
r.
7
--
SAN
01E60,
CAL
F.
9,67
in.
YUMA
AR
Z.
3.35
In
PHOENIX,
ARIZ.
ROSWELL,
N,M
7.43
in.
14.88
in.
Frouto;
9.—I
i i,
r(•+
-
+ntage
of
the
avenige
annual
precipitation
(shown
in
inches)
rem.ived
ia2
each
month
at
selected
stations
in California,
Arizona,
anti
New
3+1exico.
The
monthly
distribution
of
rainfall
at
Natural
Bridge
is
repre-
sentative
of
that
of
most
of
the
Southwest,
although
the
climate
at
Natural
Bridge
is
moist
subhumid,
whereas
in
most
of
the
Southwest
it
is
semiarid
or
arid.
At
most
stations
rain
may
be
totally
absent
in
July
or
August
as
well
as
in
the
other
months
of
the
year.
The
likelihood
of
rain
during
May
and
June
is
generally
less
in
other
parts
of
Arizona
than
at
Natural
Bridge.
Large
monthly
amounts
may
be
experienced
even
in
the
arid
areas.
222402
°
-
42-2
18
TECHNICAL
BULLETD:
SOS,
U.
S.
DEPT.
OF
AGRICULTURE
Pieeen$
100
20
0
1
,
\
151
(
1
E1
E1:E10-1
\
h
Ik
1 4
2
1
1'4
41
k.
1
IA
1
6
Prec4lotw
In
,n.l.hos
7.00
cod
ave
600-699
500
-
5.99
400
-
499
3.00
-399
2_00
-
9.
100
-
199
.50
-
99
..01
-
49
00
33
0
LEGEND
r
iGui
,
E10
....___
Fr
,
vion
,y
o
f
monthly
orpriiiimtion
of
spovitieli
;Um:milts
Nol1
:11
1001.
Within
much
of
the
ti
outhwest.
1105
was
the
rainiest
year
t'Xperi-
enced
since
observations
have
been
made
;illy
considerable
scale.
At
only
4
of
the
14
stations
in
operatirm
in
northern
Arizona
that
year
hits
the
rainfall
of
005
been
exceeded.
Iii
fact.
the
rainfal
l
311
most
of
these
stations
in
/905
was
several
inches
higher
than
that
of
the
next
rainiest
vear
and
lit
one
station
it
was
seven
times
the
111i11
-
fall
of
I
lly
driest
year
{table
2).
T3Ri.1
-
:
2
Tiu
r1p+137
r
run
rif
11137.1
und
r
rit
-
0,
-
11r10rrI
r
rater(
g
11/
:if
if
rbf
ft
r/i3
t
U,,O
373
1101
1110
111
At
;
4
111
1141
Congress
FIngsznir
rer
r
Holbrook
Jerome
.
:`
,
:n;
oral
13ri1!
t•
Present
i
wal
n
ut
Or.o.:
Yarnell
Young
_
Liqieth
111
•••.0
or
,
i
A
fi
riii:11
pr41•111131.1114/11
hight-q
inride
7.
7
2
,
2
,
21
1:
17.
IG
37
12
:13
.
34
ti l
3.
37.
1.1A31311(13
Ikrh.s
For
soverai
1'1
1
'01•'
prim.
to
RPrp,
alai'
rainrAt
thrcmgimut
Ariznna
was
far
below
the
111
110d-
111131
FOE'
die StatiOn,
for
which
records
exL:4
the
minimum
rainfall
occurred
or
Viar-
chr:qd
V
approxi-
mated
in
one
of
the
year-
between
1N90
and
1004.
Extreme
rari
ability
from
year
io
year
well
illustrated
by
she
maps
showing
annual
precipitat
ion
for
191)4
and
19115
fi",
1.
1).
WhPrea:-
ill
19114
HOL
more
than
3
percem
of
the
area
Of
the
St:1te
as
20
inches.
in
Ig05
fully
70
pereent
received
20
inches
or
more.
During
these
years.
there
were
relatively
few
stal
iothi.
hut
in
11)05
two
of
them
recorded
more
than
50
inches
of
rainfall.
At
no
other
time
ha:,
as
11111(h
NS
50
inches
of
precipitation
been
recorded
any-
where
in
Arizona_
CLIMATE
'AND
ACCELERATED
EROSION
IN
THE
SOUTHWEST
19
The
extreme
variations
in
the
rainfall
of
Arizona
for
1903,
1904,
p..nd
1905
are
presented
graphically
in
figure
12.
In
fi
gure
12,
A
the
days
with
rain
are
shown
for
all
Weather
Bureau
stations
in
operation
in
those
years.
Each
dot
indicates
a
report
of
rain
of
at
least
0.01
inch
on
a
certain
day
at
a
particular
station.
The
number
of
stations
increased
considerably
during
the
period
:
23,
in
1903:
33,
in
1904;
and
41,
in
1905.
The
stations
are
arranged
alphabetically,
just
as
they
appear
in
the
published
reports
of
the
Division
of
Climate
and
Crop
Weather
of
the
'Weather
Bureau.
Station
names
have
been
omitted
because
the
purpose
of
the
chart
is
merely
to
permit
an
over-all
comparison
of
the
rainfall
patterns
in
the
different
years.
r
PRECIPITATION
IN
INCHES
1904
PRECIPITATION
IN
INCHES
1905
0
5
10
15
Alum=
20
30
40
50
INO-IES
FIcrsru
it
—Annual
precipitation
in
Arizona,
1904
and
1005.
Dots
in
vertical
alignment
indicate
the
occurrence
of
precipitation
on
the
same
day
at
various
stations.
Dots
in
horizontal
alignment
show
successive
days
of
rain
al
the
same
station.
Total
rainfall
at
each
station
is
shown
by
the
solid
horizontal
bars
at
the
right.
The
great
contrast
between
the
total
rainfall
in
1905
and
in
the
two
earlier
years
can
be
seen
dearly.
Whereas
in
both
1903
and
1904
only
1
station
received
more
than
20
inches
of
precipitation,
in
1.005
only
9
of
the
41
stations
received
less
than
that
amount.
The
char-
acteristicmint'
seasons
in
winter
and
slimmer
separated
by
dry
periods
in
spring
and
a
utimm
are
apparent
only
in
1905.
Iii
the
other
years
there
was
great
deficiency
of
winter
rain.
From
October
4
to
December
5,
1903,
no
rain
was
recorded
anywhere
in
Arizona.
At
Natural
Bridge,
only
4.35
inches
of
rain
fell
between
October
1,
1903,
and.
July
1,
1904,
whereas
the
average
rainfall
for
this
period
is
15.90
inches.
In
Parker.
only
0.01
inch
fell
between
October
1,
1903,
and
May
1,
1904.
During
July
and
August
1904,
rainfall
20
TEC
-
MN
-
WM.,
EUI,LETIN
SOS,
U.
S.
DEPT.
OF
AGRICULTURE
was
above
normal
throughout
Arizona
despite
the
fact
that
the
summer
rainy
season
was
late.
not
commencing
until
July
19.
The
autumn
months
of
1904
were
considerably
below
normal
in.
most
of
the
State:
in
northern
and
western
Arizona.
rain
was
recorded
at
only
1
station
between
October
9
and
December
2.
Although
1905
was
the
wettest
year
on
record,
5
months
of
the
year
-May.
June.
July.
August.
and
October
-were
below
normal
in
precipitation
at
most
stations.
July
and
August
were
not
lack-
ing
in
storm
periods.
but
storm
amounts
of
rainfall
were
low.
The
rainfall
surplus
for
the
year
is
accounted
for
by
the
unusually
great
number
of
larrre
storms
in
January,
February.
March,
and
November.
Between
19'05
and
the
two
earlier
years
there
is
a
munch
greater
contrast
in
total
rainfall
and
in
the
number
of
rainy
days
than
in
the
number
of
distinct.
storm
periods.
There
were
45
storm
periods
in
1905,
31
in
1903.
and
36
in
1904.
However,
the
amount
of
rain
falling
in
individual
storms
was
much
greater
in
1905
than
in
either
of
the
other
years.
This
is
brought
out
in
table
3,
where
the
num-
ber
of
occurrences
of
1
inch
or
more
of
precipitation
in
24
hours
is
given
for
the
three
yearS.
Because
the
number
of
stations
reporting
was
different
in
each
of
these
vent's
the
data
have
been
adjusted
to
a
common
base
of
100
stations
to
facilitate
comparison.
her
of
oeen
rrencot
of
1
inch
or
ro
ore
of
preeipr
fal
ion
hp
i4
hour.,
fe
Arizoira
du
ring
lime
pretr,L
,
19113,
90
06
1905.
irll
non
11
,
rrdjn.
,
tt
ed
fn
11111
40
I
ion
x
Year
1
SII
I
,
iliill
I-
Vg
.
brit
-
ar
[
I I
A U
..
,Uteri
..,pet
:
1
.1
5y
door
July
10.111-
i
LI
ry
a
ry
tuisl
her
1003
0
211
70
4
4
14
1901-.
0
4
ii
fl
Ill
1
11
190.2
112
152
1211
74
14
ncto-
her
-II
74
7g
0
101
121
14
11O
ti
n•
von].
co
nl-
'1'nt111
her
lit
r
tt
31
r,
2S4
251i
r
902
'Despite
the
tremendous
difference
in
the
amount
of
precipitation
in
1905
and
in
the
two
preceding
year;-,
the
March
(of
temperature
through
the
three
years
was
near
normal
(fig.
12.
B).
The
mean
monthly
temperatores
at
Phoenix
for
10M.
100-I.
and
1005
and
the
average
monthly
temperatures
for
the
period
1876--l030
are
p
r
e_
sented
in
table
4.
The
mean
animal
temperature
of
1.9(1
is
exactly
the
same
as
the
normal
for
the
whole
period.
The
mean
ani
m
a
'
temperatures
of
1C)03
and
1904
are
respeci
ively
only
0.2'
and
0.ti°
F.
above
the
normal.
The
monthly
mean
temperature:,
of
the
three
years
rice
all
close'
to
the
normals.
ca
ound
hilt
1
ern
',erg
it
reP
o/
Phoenix,
for
flrr
!Ica
I'm
100.1.
1904,
rrrrri
100.1.
nod
tt
cern
go
in
ont
it
1
icin
pent
I
n
ees
f
or
/110
itediotl
1,0115
111.1f)
Period
J
1111-
FV1H-11-
kPrLi
ar
y
ary
1
ay
P.
`
c
F.
1870-19.10
:7,0,11
52.
1
'
110.5
07.1
71.7
1903
51.5
13.11
,
50.
3
011.
3
74.8
1004
48.4
AS_
Si
11
2,13
11N.
0
77.0
1902
55.
`I
7f1.1
141.11.
01,
i
70.5
June
J
111y
A
II-
4
P
-
true-
I
(
Mil-
Ni.
Vv111.-
De.
0.111-
Year
111101
her
her
her
her
4
7
,
'
Y.
°
p,
r.
'
F,
34.
4
AO.
7
411.4
82.11
70.3
1
1
.1
2
21.11
59.
5
%0.4
110.
11
911,
4
1,1.7
71.0
62
4
53.4
011.
7
1;0
3
HIS.
11
44.1
39.
11
43.
0
911.
0
40.5
31.
2
70.N
,
71.
11
62.
2
5%,
5
,
v2.
7
70.0
70
2
10
1.5
fn
cis"'
2
JAN
FE6
MAN
APP
MAY
JUNE
,
,
A
1
0
.0
0-
0.
JULY
.0
00
V96
oEPe
Oct
00
DEG
Fit%
rerairoliage%
121
.
1:
••
I i .
li
r
.,
'
'
'7141
A
so,
.
:.7.
-
.
,,
, .
,
r
.
2
i
.
i
:
,
..•
.
'
...q
.
z:7.
,
:.
,
-
--
,',:z
.
..4:
1
.
.
II.
.:
e
°
I
t_
r-
--
.
1•90
f
y
l
yF
yf
.r
JAN
FEB
MAR
APR
MAY
JUNE
JULY
BUD.
SEPT
OCT
NOP
DEC.
'''.-
+0
20
W.
io
w
20
SO
0:0
K.
RP
.9
2.0
9)
SO
.0
T
io
la
,
to
ici
!
io
to
''''
snaseasie
I
1
,
Ai
osity
mean
'
9°,
'
5E4,141rd
auerspe
f2E.yes,
re
oaa
Min
Mgr..
So
9
0
r7
;
g
5
90
1903
t
1904
so
--
-
tl
Mla
SP•
130
B
3
1
t.
7
i
30
la
50
50
1905
Pinang
12.-4,
Daily
precipitation
in
Arizona,
1903,
1904,
and
innni;
B,
daily
maximum,
minimum,
and
mum
temperattires
at
Phoenix,
Aria,
1908,
1904,
and
11305.
In
A
the
station.
are
arranged
alphabetically
by
divisions
of
the
BMW
—northern,
soi
them,
anti
western
—an
they
appear
in
the
Clinlatologleal
Rarornarlet
of
the
Weather
Bureau.
(r11055
D.
20)
CLIMATE
AND
ACCELERATED
EROSION
IN
THE
SOUTHWEST
21
In
fi
gure
12.
B,
the
daily
temperatures
for
1903.
1004.
and
1905
at
Phoenix
are
plotted
for
comparison
with
the
daily
precipitation
Arizona
for
these
years.
It
is
seen
that
the
sequence
of
air
masses
as
revealed
by
the
daily
march
of
temperature
is
in
no
year
particu-
larly
unusual.
However,
on
analyzing
the
daily
weather
maps
for
the
three
years
it
was
found
that
there
was
a
greater
tendency
for
the
cold
fronts
to
stall
over
Arizona
in
1905
titan
in
either
of
the
earlier
years.
Polar
Canadian
air
-mass
invasions
were
also
morn
frequent
in
1905.
In
1903,
there
were
approximately
60
invasions,
of
Polar
Pacific
air
masses,
of
which
6
became
quasi
stationary
over
Arizona.
and
4
invasions
of
Polar
Canadian
air.
1
of
which
became
quasi
stationary.
In
1904.
approximately
79
Polar
Pacific
air
masses
moved
over.
with
stalling
for
a
time.
Of
4
Polar
Canadian
air
masses
crossing
Arizona.
only
1
stalled.
In
1905.
approximately
83
invasions
of
Polar
Pacific
air
masses
occurred.
with
2.3
quasi
stationary.
and
there
were
14
invasions
of
Polar
Canadian
air.
with
4
quasi
stationary.
Conditions
favoring
warm
-front
storms
were
more
numerous
in
1005
than
in
either
of
the
earlier
years.
Otherwise
there
was
little
differ-
ence.
Certainly.
no
one
studying
only
the
meteorological
conditions
of
the
three
years
could
have
determined
that
one
of
these
years
was
the
wettest
on
record
and
the
others
practically
the
driest."
The
influence
of
individual
large
storms
on
monthly
and
annual
rainfall
totals
was
Well
illustrated
in
1930.
Despite
several
large
storms
in
September.
one
of
which
was
discussed
on
pp.
13-14.
most
of
Arizona
was
deficient
in
precipitation
for
the
year.
In
figure
13.
A.
the
days
with
rain
are
shown
for
all
stations
in
operation
in
Arizona
in
1039.
As
in
fi
gure
I
.
each
dot
indicates
a
report
of
rain
of
at
least
0.01
inch
on
a
certain
day
at
a
particular
station.
Widespread
rainfall
is
most
marked
in
die
winter.
early
springy
and
late
.
fall
months.
In
fl
oury
13.
B.
are
shown
the
daily
maximum.
minimum.
and
mean
temperatures
at
Flagstaff
and
Natural
Bridge.
Ariz..
in
19:39
and
the
normal
of
the
means.
The
fluctuations
of
the
line
represent
-
in..
the
daily
mean
result
from
the
temperature
changes
brought
abnit
by
the
influx
of
cool
and
warm
air
IntiN-YS..
The
relation
of
large
changes
of
temperature
to
incidence
of
precipitation
is
clearly
shown.
which
demonstrates
the
control
exercised
by
air
-amass
inter-
actions
on
preeipitation.
Daily
precipitation
amounts
at
Flagstaff
and
at
Natural
Bridge
are
shown
in
figure
13.
1
.
Precipitation
was
lower
than
usual
dlir-
in
,
r
the
early
months
of
the
year.
Fla
.
a-sraff
experienced
below
-normal
rainfall
in
every
month
except
September.
and
despite
large
amounts
in
that
month.
the
deEcienev
for
the
year
was
9,s9
incites.
In
the
ti
month-
prior
to
September.
on
ly
0,66
i
nc
h
e
,
i
h
a
d
fallen.
whereas
the
normal
fsmr
the
period
was
15.47
i
n
c
.
/1a,
At
Natural
Bridge.
where
the
normal
is
13.19
inches.
prior
to
Augr.st
1
there
had
been
only
.7m3
inches
of
preripitation.
nOtt"kr01,
Pgir
31
7
1,
1
,1
13
,,
PS
by
B.njamin
FiryizEraan.
Hof
the
ts•
and
l'hyAiograph
it.
22
TECHNICAL
BULLETIN
SOS,
U.
S.
DEPT.
OF
AGRICULTURE
EXCESSIVE
PRECIPITATION
IN
THE
SOUTHWEST
Both
general
and
local
storms
have
characteristic
patterns
of
structure
and
migration.
both
have
definite
Emits
of
areal
distribu-
tion,
and
both
have
centers
of
maximum
intensity.
If
stations
are
evenly
spaced
throughout
a
storm
area,
it
is
obvious
that
low
intensi-
ties
and
small
amounts
of
precipitation
will
he
recorded
more
frequently
than
high
intensities
and
large
amounts
since
the
periphery
of
a
storm
is
larger
than
its
center.
Similarly.
a
single
station
having
a
reasonably
long
record
of
rain-
fall
will
report
many
occurrences
of
precipitation
of
small
amount
and
low
intensity,
and
few
of
large
amount
and
high
intensity.
Records
of
very
intense
falls
and
large
amounts
will
be
few.
The
location
of
individual
rainstorms
in
a
region
is
more
or
less
random.
Thus.
since
a
storm
center
is
comparatively
small
it
is
more
probable
that
the
station
will
be
in
some
part
of
the
periphery
of
the
storm
rather
than
in
its
center.
Storms
vary
in
size,
also.
and
this
varia-
tion
helps
to
explain
the
frequency
distribution
of
recorded
amounts
of
recipitation
at
an
individual
station
(107.
pp.
480-481).
Santa
Fe
is
the
only
station
in
the
Southwest
having
a
continuous
rainfall
record
exten&ing
back
to
1850.
The
frequency
of
various
24
-hour
amounts
of
precipitation
at
this
station
for
the
90
-year
period
1850-1939
is
given
in
table
5.
The
overwhelming
prepon-
derance
of
precipitation
amounts
under
0.20
inch
does
not
mean
that
rainstorms
are
usually
limited
to
those
amounts
in
the
Southwest,
but
rather
that
the
more
or
less
random
distribution
of
storms
results
in
the
station
at
Santa
Fe
being
much
more
often
marginal
than
central
with
respect
to
the
storm
area.
T
ABLE
5._sunibei•
of
of•eurrences
of
xprrified
..).1
-hour
°Mai!?irk
of
precipitation
at
Sonia.
Fe.
N.
Me.r..
18.70-1894.
18115-1939,
and
1850-1939
Preci
pi
t
r
ton
(inches)
I
5.50-1894
1995-1939
Numbcrof
A
ocu
m
u
cccur*novs
ivied
toil
al
1550-1939
t2-
totati
Number
of
_A.
(TUMU-
()mon-race:-
laled
tntai
Numb
cr
of
A
mum
oc-currenr.vs
lased
4.00-1.99
.
i
1
9
II
i
I
3.00-3-99.
.
I
2
0 0
i
2
2
00-2.99
3 3
3
4
9
1.30-1.99
.8
II
S
11
-
19
.
22
13
6
17
8
39
1.
W-1.39
.....
.
10
'
23
5
22
15
45
1.20-1.29
13
35
4
211
/7
1
62
I
.10-I
-39_
.
......
.
19
51
ii
32
21
I
83
17
£8
'.
18
90
35
i
115
0.90-0-99_
.. .....
.
18
85
i
31
71
i
39
1
197
0.20-0.59
.
.
..
29
I
1
1
!
'hi
92
'
90
[
213
0.704179_
... .....
.
23
137
=
40
719
63
i
276
0.50-0.99
11
1
198
,
641
Ns
,
121
I
397
0.50-0.52
80
.
90
22
597
0.40-0.49
0.30-0.39.
o.5941.2i)
0.10-0.19
_
0.01-0.09
11..;
208.
,
2S7
1.
5fXp
'
1.350
I
.
1
5
12
1
2
1
71
1
4
11
1,
51I
2.891
370232.56
739
2.230
1
1132
559
020
1.
759
3,
,
.*19
I
-
I
7.
1
1.
1:
10
7
3
.!'
3
1
.
T
.1
9
0
.
1
8
-ii
1.271
1,
931
3,270
0,
880
The
90
largest
24
-hour
amounts
of
precipitation
which
have
been
recorded
in
Santa
Fe
during
the
90
years
from
1849
to
1938
are
plotted
in
figure
14.
The
distribution
of
these
storms
in
time
is
ex-
et
1.0
0
z
SOUTHERN
JAN.
FEE.
MAR.
lo
20
10
20
,D
20
:11:1
I
APR.
MAY
JUNE
JULY
lo
20
10
20
ID
20
0
2.9
r
-
AUG.
ZO
20
t
r.
1.1
143.
_
.„
rf:
10
20
.
44'q
".
if
1
.
".
3Ettf
SEPT
OCT.
NOV.
DEC.
10
20
10
20
10
20
JI
t
;"..;
'
maxi/num
r
T
,..
20
--Daily
mean
____.4_.
_____
--Standard
average
70
(E10.351011
33
-
yr
recard
,
I---
Minimum
Natural
Brid
1
ge
/0
-
ye
word)
so
<1
13o
2
20
ID
0
.10
70
so
SO
40
SO
1.0
5
7
0
-1---La
BO
00
t
r
--
-
—Flagstaff
_-F._+
Natural
Br
dge
10C
90
Flagstaff
Natural
Brid
'
e
.1
_
FIGURE
13.—A,
Daily
precipitation
in
Arizona
iu
1939;
B,
daffy
maximum
minimum,
and
mean
temperatu
es
at
Flagstaff
and
Bridge;
C,
daily
precipitation
at
Flagstaff
and
Natural
Bridge.
In
A
the
stations
are
arranged
alphabetically
by
divisions
ul
as
they
appear
in
Climatological
Data,
from
which
the
station
names
can
be
obtained.
Each
dot
in
A
indicates
a
report
0.01
inch
of
rain.
Imams
2
1849
1850
1851
1852
053
1654
1855
1856
1857
1858
1859
MGO
1861
1862
1863
1864
1865 1866
1867
1868
1869
1870
071
072
1813
1874
075
1876
1877
1878
1
1
1
1879
1680
3881
082
1883
1884
1885 1886
1887
1888
1869
1890
1691
1892
1893
1894
9395
1896
1697
1898
1899
MOO
1901
1902
1903
1904
1905
19
06
19
07
1908
1909
1910
1911
1912
1913
1914
1915
1916
4917
1918
1919
1920
1921
1922
192
1924
1925
1926
1927
1928
1929
1930
1931
932
1933
1934
1955
1936
1937
1938
F14URE
14.
—Distribution
of
the
90
maximum
24
-hour
storms
at
Santa
Fe,
N.
Max.,
1649-1938.
traces
n.
22)
CLIMATE
AND
ACCELERATED
F?R08210N
LN
THE
SOUTERVEST
23
tremely
irregular.
Only
3
storms
occurred
in
the
9
years
1930-1938;
on
the
other
hand.
IT
occurred
in
the
4
years
1853-1856.
During
the
7
-year
period.
1896-1902.
1
storm
occurred
each
year.
Of
the
90
maximum
24
-hour
storms.
60
occurred
during
July,
August,
and
September.
Of
the
90
maximum
48
-hour
rainfall
amounts.
53
were
recorded
during
these
months.
The
comparative
data
are
shown
in
table
6.
It
must
be
remembered
that
Santa.
Fe
is
in
the
part
of
the
Southwest
which
has
a
winter
minimum
of
precipitation.
In
western
New
Mexico
and
Arizona.
where
there
are.
2
rainy
seasons,
large
storms;
are
to
be
expected
in
winter
as
well
as
in
summer.
'P
A
It1.E
FM/
le
run) of
50
rrrrrriilluur
x
r
0
rli)
Su
+rlu
Fe,
Y.
Ir4'./!.•
/849—./938
Storm
duration
(hours)
Sono-
1-ehric-
,
Lr
,,,
Nfareh
Apra
May
J
im
..
1,
,,1
1
._
Au-
se
t
,-
ocio-
No-
De
-
„y
gust
torricx
,r
hi
,
r
%Wilber
vernier
Su
err
.V
ul-
No
Hi-
3.
.'n
In
-TA
S
in-
co
",
:11
m
-
her
her
-TCL11;1-
n
ni-
_Vi
her
her
her
brr
her
t
in-
ber
;a-
-
her
fII-
.\
-
ii
ffi-
.1'u
her
her
ber
24
I
1
I
0
0
0 0
51
22
17
S
.1
I
48
r
0
3
0
5
10
10
25
15
13
3
-
4
2
I
Sturm
restrieted
to
1
calendar
klity.
r
Storm
recorded
on
more
than
1
ecileticlar
day
but
not
on
more
than
2.
There
is
no
reason
to
expect
that
years
of
numerous
excessive
storms
will
necessarily
be
followed
by
other
years
of
heavy
rains.
The
Santa
Fe
record
indicates
that
large
storms
may
be
concentrated
in
one
period.
as
during
1s53-56,
mind
that
a
single
large
storm
may
be
isolated.
as
was
the
largest
storm
of
record,
which
took
place
in
February
1861.
Indeed.
as
the
storm
sequence
in
fi
gure
14
shows,
the
:;paei»g
of
hinge
rainfall
amounts
has
been
highly
irregular.
The
annual
rainfall
at
Santa
Fe
for
the
90
-year
period
1850-1939
is
presented
in
figure
15
to
permit
a
comparison
with
the
distribution
of
the
90
greatest
storms
shown
in
figure
14.
25.
17i
20
z.
—.
13
0
,!
10
I
a.
0
1550
1960
[570
1530
1590
1900
1910
1920
1930
1939
FIGURF,
15
—Animal
precipitation,
Santa
Fe,
N.
Mex.,
1874-1939.
The
coincidence
of
the
various
renditions
necessary
for
the
farina-
tion
of
rain
does
not
occur
with
any
regularity.
Not
only
is
mois-
ture
in
the
air
necessary
but
there
must
also
be
present
some
mecha-
nism.
frontal
activity,
convection,
orographic
lifting,
or
a
combina-
tion
of
the
three,
for
releasing
the
moisture.
It
is
not
unusual
for
heavy
thunderstorms
to
occur
in
summer
at
heights
of
2.000
to
5,000
feet
aloft
without
wetting
the
ground.
Sometimes:
the
only
result
of
a
heavy
thunderstorm
will
be
the
falling
of
a
few
hailstones
49,
p.
111).
Tropical
air,
unless
moist,
will
yield
no
precipitation
when
24
TECELNICAL
BULLETIN
SOS,
U.
S.
DEPT.
OF
AGRICULTURE
uplifted
by
cold
fronts
passing
across
the
Southwest.
On
the
occa-
sions
when
all
the
conditions
which
make
for
rains
are
present,
how-
ever,
excessive
amounts
are
possible
even
in
the
arid
parts
of
the
neon,
as
the
storm
of
September
4-7.
1939.
demonstrates
(fig.
6).
Fort
Mohave
7
is
a
desert
station
on
the
Colorado
River
in
western
Arizona
with
an
average
annual
rainfall
of
5.09
inches.
In
1889.
the
total
rainfall
of
the
year
was
21.38
inches.
11.17
inches
of
which
came
during
the
month
of
December.
In
this
month
there
were
8
days
on
which
rain
was
recorded;
the
smallest
daily
amount
being
0.62
inch.
On
5
clays
the
precipitation
exceeded
1
inch.
and
rainfall
amounts
recorded
in
4
storm
periods
were
2.90
inches,
2.70,
2.30,
and
1.90
inches.
In
the
521
scattered
months
of
the
record
at
Fort
Mohave,
monthly
precipitation
totals
exceeded
1.00
inch
only
59
times.
Monthly
rainfall
amounts
exceeding
11.90
inches
are
extremely
few
in
the
Southwest,
yet
the
fact
that
such
an
amount
was
recorded.
in
a
desert
station
indicates
that
the
oeeurrence
of
that
much
rainfall
in
a.
month
is
not
impossible
anywhere.
All
unofficial
excessive
fall
of
precipitation
was
reported
at
Fort
Mohave
for
August
1898
by
Henry
Schlegel
(A!,
p.
d),
the
cooperative
Weather
Bureau
observer.
as
follows:
On
the
2Sth,
we
laid
the
biggest
rain
in
10
or
11
years,
and
to
my
regret.
between
the
rain
.and
furious,
wind,
my
min
guage
was
upset.
To
give
an
idea
of
the
amount
of
rain
that
fell,
and
which
lasted
only
45
minutes,
i
Lad
a
wash
tub
_:et
out
on
the
mesa,
clear
of
everything,
and
the
water
after
the
rain.
measured
8
inches.
Records
of
excessive
precipitation
in
the
Southwest
are
extremely
few.
There
are
only
three
Weather
Bureau
stations
with
long
records
from
automatic
rain
gages.
and
only
within
the
past
6
years
have
other
agencies
installed
self
-recording
gages
in
any
number
in
the
region.
The
records
from
cooperative
Weather
Bureau
stations
do
not
ordinarily
give
the
time
of
beginning
and
ending
of
storms.
Frequently,
too.
the
records
of
large
storms
have
been
lost.
either
through
damage
to
the
gage.
as
at
Fort
Mohave
on
August
28.
1898.
or
through
failure
to
make
an
observation
over
a
period
of
several.
days.
Maddock
and
Leopold
8
have
tabulated
40
storms
of
2
inches
or
more
in
Arizona
and
New
Mexico
for
which
the
times
of
beginning
and
ending
have
been
recorded,
and
the
storm
duration
can
con-
sequently
be
determined.
In
figure
16.
these
40
storms
have
been
plot-
ted
and
the
envelope
curve
drawn.
Thirty-five
of
the
storms
did
not
exceed
I
hours
in
duration,
and,
as
would
be
expected.
nearly
all
of
them
occurred
in
July,
August,
or
September.
More
than
Iwo
-
thirds
of
the
storms
were
reported
from
the
eastern
margin
of
the
Southwest,
where
summer
rainfall
is
proportionately
greater
than
in
the
Southwest
proper.
Consequently.
the
enveloping
curve
repre-
sents
precipitation
intensities
somewhat
larger
than
those
typically
experienced
in
the
Southwest.
Figure
17
reproduces
the
mass
diagrams
of
precipitation
of
six
large
storms
recorded
during
the
simmer
months.
Except
for
Las
7
Prior
to
December
1808
the
station
Oil
rort
3
MAimucx.
Ttlo]1A.•
Iii,.
:ma
LEovor.a,
lArsA
ht.
EviiAlc.turEitircrics
or
WWI
liAINFA[Z•
18
NEW
MEE/CO
AND
ARIZONA.]
11.311DilbliAlea
DELDDS(friilt.]
CLIMATE
AND
ACCELERATED
EtROSION
L
THE
SOUTHWEST
25
7
6
5
3
2
0
2
4
-
6
8
10
/I
/
/
/
I
1
,
i”
I
i
a,*
•5
•7
r
3
I
CERRO,
Km'Ex.,
AUG.
17,
1922
2
EL
FASO,
TEX.,
JULY
9,
1961
3
LAS
CRUCES,
N.5IEX.,
41.10
29,1935
.12
4
62.
1
/
1
,105A,
N.MEX,,
AUG.30,
1925
•13
5
LASE
VALLEY,
N,
PAS
5.,
JULY
1,19/4
6
SIERRA
ANCHA,
ARIZ.,
AUG
5,
1939
7
DESERT
LASICIRATORy,
ARIZ.,
JULY
22,1510
--
6
CROWN
SING,
ARr2.,
AUG.
11,
1527
9
ALBUOIJEROUE,
N.MEX.,
OCT.
5,1
1
355
10
CASs
GRANDE
RUIN,
ARIZ.,
A1.1.0.
1,1906
a
11
CROmm
KING,
ARIZ
,
AUG.
6,
1919
r7
12
RECIP004,
N
MEX.,
OCT.
11,
1532
2
LAS
CRUCES,
N.
MEX.,
AUG.
29,1935
,s.
20
6:9
14
SANTA
MARGUERITA,
AR12.
AUG.
22,1935
el.
•al
15
131ENER,71
MEX.,
JULY
3,19
7
24
CORONA,
N
HEX
,
AUG
FA,
1934
-
IC
•24
17
SANTA
ROSA,
N.
Ore
EX.,
MAY
30,1530
5
;9
SIERRA
ANCHA,
ARIZ
,
SEPT
10,1533
19
PIMA,
ARIZ
,
AUG.
2,1539
027
20
1JARINET7E,
5612,,
AUG
6,1935
21
LAKE
AL105,
N.
MEX.,
AUG.
4,1935
26
29
00,
,
:,
22
uNDRITH,
N.
HEX.,
JULY
30,1931
3.I.
;23;
434
23
SIERRA
ANCHA,
ARIZ.,
SEPT.
IO
1933
3
37
6
24
25
HACHITA,
N
14E3
,
AUG.
6,1931
ROSwELL,
14.1.15:1
,
AUG
8,1965
o•-•
El
m
40
26
ROSWELL,
N.
HEX.,
SEPT.
19,1923
27
01JRAN,
14
MEX.,
SEPT.
16,
ISIS
26
SUPERIOR,
ARIZ.,
AUG.0.7
40E41917
25
GALLtNA
R.
S.,
N.
HEX
,
MAY
29,
1930
30
.2E6E2
SPRINGS,
N.
HEX.,
JULY
7,1523
31
LAKE
ALICE,
N
655.,
AUG.
30,1536
32
ROSwELL,
/I
HEX
,
SEPT.
14,
1523
33
HARVEY'S
UPPER
RANCH,
N
MEX....JULY
27,1915
'34
35
JEMEZ
SPRINGS,
N.
MEX.,
LUG.
3.1533
CAMP
23-A
,5
1
/
1
-HT
LOCK
VALLEY,
ARIZ
,SEPT.
6,1939
-
36
SILVER
CITY,
11
HEX.,
JULY
20,1929
37
PEORIA,
ARIZ.,
AUG.
6,1539
39
JEREZ
SPRINGS,
N.
MEX.,
JULY
12,1910
39
MINERAL
HILL,
N.
MEx.,
APR.
23,
1514
46
LAGUNA,
N.
MEX.,
JULY
I I
22,1912
12
14
16
DuRATKIN
(HOURS)
Froti
10.
-Total
precipitation
and
duration
or
40
large
storms
in
Arizona,
Texas,
and
New
Mexico.
26
TECHNICAL
BULLETLN
805
1
U.
S.
DEPT.
OF
AGRICULTURE
7
6
5
X
4
U
2
2
0
0.
0
3
a_
2
--------
,
i
I
d
d
9'
6
r
''.
CI
d
¢
P
1
.
7
1
3
---a---
••••••••••0--
--10-"--
D.
TUCSON,
ARIZ,
SIERRA
ANCHA,
LAS
CRUCES,
PIMA,
ARIZ.
SIERRA
ANCHA,
FREEMAN
FLAT,
JULY
31,1935
ARIZ.
N.
MEX.
AUG.
2,
1939
ARIZ.
SAFFORD,AFIVZ.
SEPT.
10,1933
AUG.
29,
1935
AUG.5,1939
SEPT,
16,1939
-•
A--
•—V----
2
3
4
HOURS
Munn
h.
—Precipitation
in
six.
large
storms
in
Arizona
and
New
Alexico,
5
6
7
CLIMATE
AND
ACCELERATED
EROSION
IN
THE
SOUTHWEST
27
Cruces,
the
stations
are all
in
Arizona,
and
the
curves
may
be
taken
as representative of
extremes
of
precipitation
intensity
for
short
periods
in
the
Southwest.
RAINSTORNi
EREQuENCIES
IN
THE
sot7THWEST
9
Reliable
estiznates
of
the
probable
frequency
of
large storms
of
specified
precipitation amounts are
difficult
to
make
under
the
most
favorable
conditions
707).
but
in
the
Southwest,
where
rain-
fall
records
ace
fra!Tmentary
and
the
country
is
lack
-
lug
in
topo-
graphic
u.dformity.
the
task
is
doubly
difficult.
However.
despite
diversity
of
surface
features
and
cousL
,
quent
variation
in
average
rainfall
front
place
to
place
and
despite
lack
of
meteorological
homo-
geneity,
considerable
useful
information
on
average
frequencies
of
precipitation amounts
in
the
Southwest
can
be
obtained
by
use
of
the
familiar station
-year
method
of
analysis (M.
In
northern
Arizona. there
:u
-
e
approximately
75
Weather
Bureau
stations
with
precipitation
records
ranging
in
length from only
a
few
,
•ears
to
nearly
70
years.
In
all,
1;269
station-years of
record
are
available for study.
In
these
records
there
were
1.269
reports
of
:4-hour
rains
exceeding
1,42
inches.
This amount
then
represents
the
precipitation
which
might
be
expected
at
each
station
once
every
year. There
were
634
reports
of
24-honr
rains
of
1.73
inches
or
over.
which
amount
could
be
expected
at
each
stat
ion
once
in
,
2
years.
Other
frequencies
for northern
Arizona
are
given
in
the
following
tabu
hit
iiarl
:
EitChek
Ma:611111ln
24-13.1111'
1}F
prPripil
t
r15
i'Xile
4.
14
1
01141'
in
-
I
ysNtr
_
1.42
2
ye:Ws
___.
1.13
3
1
-
l'111
.
8
1.0:1
yvors
2.
1:i
lit
ye21'$
2.55
15
years
.
_
.
_
.
2.74
x
5:3
2u
years
-
-
91
25
years
_
3.00
N)
AWL%
______
.
RIO
yenN
-1.
The largest
recorded
24-hotir
precipitation
in
northern
Arizona
iii
the
period
studied
is
6.46
inches.
However.
the
frequency
of
large
storms
varies
greatly from
place
to
place.
In
northern
Arizona,
the
stat
ions
range
front
350
feet
to
more
than
8,5M
feet
above
sea
level,
and
the
average annual pre-
cipitation ranges
from
less
than
5
inches
to
more
than
3(1
inches.
Large stot
-
ni
amounts
of
rainfall
increase
in
frequency
with
increase
in
average annual tido
-
Fail
iii
areas
meteorologically
similar.
The
frequency of
large storms
also
varies
with
variation
in
ineteoroing
-
ical
conditions. Both
Tuba
City
and
Parker
are
arid,
their
average
annual precipitation
being
6Th
and
5.32
inches.
re-
spectively. Yet
of
a
ll
24
-hour storms
of
0.50
inch
or over, at
Parker,
"
PIIr
it
.
r111
"fretmeney"
here
12;1
8
'
1/11.
SIWC•ini
Dien
Iiing
glveu
IN
,
word
In
hydrologic
and
$.111notie
The
froonooey
of
a
rainfall
of
any
ottionnt
di
rile
1144'44/1
i11/4'1'
5
4M
iw-
twirq•n
/
.
41.11/14
of
that
Amount
awl
greater.
A.
rtlinfflai
of
VA)
'holier..
for
oxiimple,
has
20
-year
frecizOnes,
or
is
1/
10-ye/tr
141.krut,
if
lino
moon
illiorta/
betwoen
rains
of
2
inchos
and
more
1241
10
years.
This
explanation
applies
to
to
of
the
word
In
tabloit,
Illunttittions.
and
Sext,
28
TECHNICAL
BULLETIN
SOS,
U.
S.
DEPT.
OF
AGRICULTURE
33.1
percent
were
1
inch
or
over,
and
at
Tuba
City,
only
12.4
percent.
At
Parker,
two
storms
have
exceeded
3
inches.
but
at
Tuba
City
no
recorded
storm
has
exceeded
2
inches.
Parker
is
only
350
feet
above
sea
level
and,
like
Fort
Mohave,
is
in
the
region
where
convective
storms
reach
their
maximum
intensity.
Tuba
City,
on
the
other
hand,
is
on
the
plateau.
4.500
feet
above
sea
level,
and
although
severe
thunderstorms
are
experienced
(41))
they
do
not
equal
those
at
lower
elevations.
These
two
stations
are
not
meteorologically
homogeneous.
Since
the
most
intense
rains
are
usually
associated
with
thtunder-
storm
activity
a
negative
corm
In
t
IOU
heilveen
storm
intensity
and
retrional
elevation
is
to
be
expected.
The
total
depth
of
moisture
in
the
atmosphere
over
a
station
at
a
low
elevation
is
much
greater
than
that
over
a
station
at
a
high
elevation.
Consequently
equally
in-
tense
convective
activity
will
result
in
greater
storm
intensities
at
the
station
at
low
elevation
than
at
the
other.
No
similar
control
is
exercised
by
elevation
over
the
rainfall
intensities
and
ttmounts
result
ing
from
warm-
front
storms.
Cedar
Glade
and
St.
Alichaels
have
average
annual
precipitation
of
14.10
inches
and
13.15
inches.
respectively.
Both
stations
are
semiarid.
Cedar
Ghule
is
in
a
yid
k
-v
in
the
mountainous
central
part
of
the
State.
4.1310
feet
above
sea
level.
St.
Mielmels
is
nearly
7.000
feet
above
sea
level
in
the
plateau
csaintry
in
the
northeastern
part
of
the
State.
Of
all
24
-hour
storms
of
0.50
inch
or
over,
at
Cedar
Glade.
3;
.3
percent
were
1
inch
or
over,
and
at
St.
Michaels.
only
19.7
percent.
Cedar
Glade
and
St.
:Michaels
are
not
meteorological
homogeneous.
Local
surface
variations
tend
to
cause
differences
in
storm
i
11
ten
-
si
ty.
Even
within
is
small
watershed.
variations
in
the
proportion
of
the
total
rainfall
which
comes
at
high
rates
may
be
considerable.
Extremes
of
meteorological
condit
ions
at
the
stations
in
northern
Arizona
are.
illustrated
by
Itlagstaff
and
Fort
nuhaye.
Although
the
average
annual
precipitm
ion
an
l'
lagstaff
is
211.50
inches.
only
24.0
percent
of
the
24
-hour
rains
of
0..50
inch
or
over
exceed
1
inch.
On
the
other
hatul.
at
Port
Mohave.
with
an
average
animal
precipi-
tation
of
only
5,09
inches.,
42.1
percent
were
in
excess
of
f
mess.
The
average
annum!
number
of
rainy
days
at
Fort
,Mohave
is
only
15,
in
contrast
to
s4
at
Flagstaff.
but
when
precipii
ation
occurs,
large
amounts
are
inure
likely
to
fall
at
Fort
Mohave
than
at
Flagstaff.
It
is
probable
that
most
of
the
stittions
in
northern
Arizona
lie
somewhere
between
the
meteorological
extremes
of
Fort
Mohave'
and
Flagstaff.
and
no
doubt
the
transition
in
meteorological
comli-
lions
from
place
to
place
is
more
or
less
gradual.
To
il
lustrate
the
generaliml
ion
that
rainstorm
intensity
-frequencies
increase
as
the
average
annual
rainfall
inereases
;Ind
also
that
they
vary
with
sur-
face
configuration
.
two
groups
of
stations
have
been
selected.
the
average
24
-hour
storm
frequencies
for
which
are
given
in
table
7.
The
records
are
not
bong,
even
the
combined
records
for
similar
stations,
and
the
reliability
of
the
frequency
determinat
ions
is
low
/On.
In
group
A,
comprising
Crown
King
--final
Minch.
Cedar
Glade.
Winslow—Holhrook,
and
Tuba
City.
the
average
annual
pre-
cipitation
ranges
from
25.59
inches
to
6.78
inches.
In
group
B.
CLIMATE
AND
ACCELERATED
EROSION
IN
THE
SOUTHWEST
29
consisting
of
Natural
Bridge.
Jerome,
and
Kearns
Canyon-Jeddito,
1
°
the
average
annual
precipitation
ranges
from
24.21
inches
to
11.87
inches.
In
each
group
of
stations
the
variation
in
storm
frequencies
is
related
directly
to
variation
in
average
annual
precipitation.
The
fact
that
storm
frequencies
in
one
group
of
stations
are
not
coin
-
parable
to
those
of
the
other
indicates
variations
in
meteorological
conditions
corresponding
to
variations
in
situation
of
the
stations,
such
as
was
illustrated
by
Tuba
City
and
Parker
and
by
Cedar
Glade
and
St.
Michaels
in
the
preceding
paragraphs.
TABLE
t
-
eragc
f
requency
Of
24-horir
precipitolion
amount::
al
.elected
.vtot
ions
irr
Arizona
Orono
Fan!
Stntion.
A
venzge
24
-hour
amoinas
of
rainfall
oeenrring
once
iu-
nnnunF
Keel;
Pit
re
11011
E
yor
2
yors
5
years
10
plus
21)
years
Orono
A.:.
. .
_
Crown
}Zing
-Phial
Main
1
India
India
luchn,
India,
liiditt
Ranch.
_ _
.
25.89
I
2.07
2..8.1
:
3.34
4.
15
h.
10
Cetior
0
lode
14.
10
:
1.2ii
1.59
2.11
2.61
3.22
Winslow
-13o
/
brook
8.52
:
.88
1.
10
1.45
1.711
2.21)
Tizho
City_
6.76
'
.
72
.80
1_
18
1.45
1.70
Group
11:
Natural
Bridge
24.21
'
2.05
2.38
'2
.h1
'
3.4o
;
7.1111
3(10c
-
tie
_
19.20
1.71)
1.97
2.42
!
0.62
I
3.30
Kearns
Canyon-Jecklito
11.87
1.
12
1.31
1.61)
1.67
1
2.
18
In
figure
18
storm
frequencies
for
these
two
groups
of
stations
are
plotted
on
a
logarithmic
scale.
with
the
average
frequencies
in
northern
Arizona
for
comparison.
In
each
group
of
stations,
straight
lines
fit
the
data
and
all
lines
have
the
same
slope.
The
parallelism
of
the
lines
is
interpreted
to
indicizie
meteorological
similitude.
That
.the
lines
which
fi
t
the
data
of
one
grout)
of
stations
are
not
parallel
to
those
which
fit
the
data
of
the
other
group
shows
a
lack
of
similarity
between
groups.
Because
of
the
proportionality
existing
between
average
annual
rainfall
and
storm
frequency
in
these
stations
it
was
possible
to
pre-
pare
simple
nomograms
giving
24
-hour
storm
aniounts
in
terms
of
annual
precipitation
totals.
These
nomograms
alt'
reproduced
in
figure
194
for
meteorological
conditions
prevailing
.
at
the
stations
headed
by
Crown
King
--
final
Ranch,
and
in
19.B
for
meteorological
conditions
at
the
other
group
of
stations.
Frequencies
of
the
storm
amounts
vary
trenwndotw.Iy
from
place
to
place
with
variation
in
meteorological
conditions
and
with
annual
precipitation.
Tinder
one
set
of
meteorological
conditions
the
24
-hour
precipitation
to
be
expected
at
i
station
only
once
in
100
years
ranges
from
3.85
inches
to
9.47
inches
as
annual
precipitation
increases
from
10
inches
to
30
inches.
Under
another
set
of
meteor-
ological
conditions
for
the
same
range
of
annual
precipitation
the
range
for
the
100
-year
storm
is
from
2.70
inches
to
0.8
inches.
Meteorological
conditions
other
than
those
represented
in
the
nomograms
in
fi
gure
19
are
to
he
found
in
northern
Arizona.
Fre-
Tone
spelling
".Tadilor"
f11
priweribt.ti
by
the
i'nlied
Staips
(.;(.0gr11t)i1i
Board
fur
the
ca
n
y
o
n
H
o,'
mpring.
'The
mann.%
of
Lhe
Evading
porn
and
Wotii
her
litsreatt
Ntation
are
atiiI
nreBen
".tetlatto."
30
TECHNICAL
BULLETIN
SOS,
U.
S.
DEPT.
OP
AGRICTIIARTRE
quencies
at
Parker
and
Fort
Mohave
are
higher
than
would
be
de-
termined
from
fi
gure.
19,
A,
and
frequencies
at
Williams
and
Flagstaff
are
lower
than
would
be
indicated
in
figure
19,
B.
Meteor-
ological
conditions
and
consequently
the
rainstorm
amounts
to
be
I0
s
9
7
6
9
3
2
Io
.9
4
3
2
.9
.6
.7
1
2
3
4
5
6
7
9
9
ID
15
20
25
30
40
50
60
70
909010
a
'''
-wall
C,VO
Vi
G
"
00:
0,1C
..-•
-
"'
--.-'.---7--.
0
161
6
o
a
vol•kS
°
ISM'
1'
6
''.
GL
CVS
Y
_r
imnim
4
.
11/
I
M
IIIIIMMEEL
11
41110
1
Med
MIII
1
1" 5"1
0110
.
011
1/
1
1111110
re
t
li
t
NO
NM--
Vt
5
C01101i-
JEDDISO
111111r
2
3
4
5
6
7
9
910
15
FREQUENCY
KARS)
FIGURE
]S.—Fregneney
of
:24-11onr
preeipita
I
ion
amounts
at
selected
stations
in
Arizona
:
.4.
group
A
;
13,
gnaw
B.
0
20
25
30
40
50
60
70
6090100
expected
with
given
frequency
will
vary
from
one
side
of
a
valley
to
another
and
at
different
levels
on
a
slope.
Thus,
it
is
obvious
that
the
nomograms
in
figure
1.9
cannot
be.
used
as
the
basis
for
design
of
erosion
-control
or
flood
-control
structures.
Before
entirely
reliable
frequency
data
can
he
obtained
for
a
single
watershed,
a
detailed
survey
of
its
meteorological
conditions
must
be
made.
The
nothog
-
rams
indulge
that
light
storms
are
relatively
C1
ATE
AND
ACCELERATED
EROSION
MT
TEE
SOWPRWEST
31
numerous
and
heavy
storms
few,
but
that
occasional
storms
of
very
high
intensity
may
be
experienced.
The
probability
of
exceptionally
large
storms
in
Arizona,
as
would
be
expected,
is
greatest
in
duly
or
August
and
least
in
May.
Frequencies
of
24
-hour
precipitation
amounts
for
the
12
months
at
Kearns
Canyon
(table
8),
although
based
on
a
short
record,
appear
to
be
representative
of
the
month
-to
-month
changes
in
the
probability
of
heavy
rainfall
in
the
Southwest.
10
.6
S
s
:44
fr
.
O
15
20
100
90
00
70
0
0
3
Et
tr,
5
a
7
9
B
zr
TOO
90
11
0
7
6g
,
50
6
4
3
2
25
30
0
5
10
I5
20
25
10
AVERAGZ.
0411201-
FIPMF972/7101/
t18C.E54
FinuRE
I9.
—Nomograms
expressing
the
relationship
between
average
annual
rainfall
and
24
-hour
rainfall
for
two
meteorological
conditions
in
northern
Arizona
:
A,
Data
from
Crown
King-1'in:11
Itaarh,
Cedar
Glade,
Winslow
Holbrook,
arid
Tuba
City;
)1,
data
from
Natural
Bridge,
Jerome.
and
Keasms
Canyon—Jeddito.
DROUGHT
FREQUENCIES
IN
THE
SOUTHWEST
In
most
of
the
Southwest,
less
than
1
day
in
6
is
rainy
and
in
the
arid
parts
of
the
region
the
rainy
days
may
not
average
more
than
1
in
30.
At
Fort
Mohave
the
average
annual
number
of
days
with
rain
is
15.
There
is
considerable
monthly
as
well
as
annual
variation
in
the
number
of
days
on
which
precipitation
occurs.
In
fi
gure
20,
the
average
number
of
days
of
rain
in
each
month
is
shown
for
Natural
Bridge,
Meatus
Canyon,
and
Lenpp;
and
also
the
actual
number
of
days
with
rain
at
Natural
Bridge
in
each
month
during
1903,
1004,
and
1905.
By
comparing
the
graphs
of
Natural
Bridge
32
'PEK...h1NICAL
BULLETFN
808, U.
S.
DEPT_
OF
AGRICULTURE
in.
this
figure
'
with
fi
gures
9;10,
and
13,
it
can
be
seen
that
a
rough
parallelism
exists
between
the
number
of
storms
and
the
monthly
and
annual
amounts
of
precipitation.
TABLE
8.-Frequencil
of
24
-hour
precfpitation
amount4,
lineups
Con110a-
-Jeddito,
Ariz.
Month
January.
Febru
n
ry,
March-
April__
.........
_
Islay
hoar-........
_
_
July.
August
_
Se
pteto
her
October.
Noietn
ber.
December.
24
-hour
loom
tntS
of
Mifflin
oceurrin
once
in
-
I
'
no
25
1
20
1
15
10
1
6
3
1
0
'
I
yrar.5
yenr8
!
years
I
years
years
!
years
1
years years
yr.13r5..
Inches
Jun
,
Inches
Inches
;
Inches
,
IncheS
Inches
'
Inches
inch,*
1.:30
1.20
;
1.
12
1
1.90
:
0..57
0
t1S
0_
56
i
0.49
0.116
1.
12
I
I_
03
'
„ra
.
So
I
.74
:
.a7
.47!
.ii
..12
1.22
1.I4
1
1.06
.95
1
.#2
,
.64
.53
:
.49
.70
1.22
1.14
1.06
.
.95
I
.82
1
.54
.53
'
.46
.36
.
9.5
.90
.
81
1
.70
1
.
56
1
.39
[
.30
.
25
.
17
1.20
1.07
.94
;
.711
;
.
62
;
.42
.30
.24
.
16
1.95
1.99
1.82
!
1.70
1
1.00
;
1.33
1.
18
1.05
.94
2.31
.
10
1.
00
1
1.
05
1
1,211
.06
..5
I.
.442
.44
1.45
1.35
1.22
1
1
1.05
i
.86
.
.62
.45
:
.39
.2:s
1.30
1.22
1.
14
:
1.04
:
.
91
.73
.112
I
.54
.44
1,
10
1.1111
.941
:
.s4
:
.69
.
39
.
.32
.
.t
ti.
1.55
1.70
1.50
1.32
'
I.05
.77
.440
.48
.1.5
z
a
x
0
10
NATURAL
BRIDGE,
ARIZ.
Average
54
doys
10
0
NATURAL
BRIDGE,
ARIZ.
1903
44
days
KEAMS
CANYON,
ARIZ.
LEUPP,
ARIZ.
Average
Average
58
days
30
days
NATURAL
BRIDGE,
ARIZ.
NATURAL
BRIDGE,
ARIZ.
1904
1905
36
days
'17
days
rioutu:
21
1
.
--
Average
number
of
days
with rain
ill
each
month
at Natural
Bridge,
KfliMS
Capp
M,
tool
I,cnlgl.
111111
tlet
M11
number
of
days
witti
rain
at
Natural Bridge in
each
month
111
1f103,
11104.
and
1005.
The
years
iIf
record
for
Natural
Bridge,
Keams
Canyon,
:and
Loam)
are
respectively
38,
12,
and
9.
Since
there
is
no
regularity
in
the
occurrence
of
rainfall
there
can
be
none
in
the
intervening
periods
of
drought.
Periods
lacking
in
rainfall
are
brought
about
by
specific
meteorological
conditions
just
as
are
storm
periods
and
are
equally
important
elements
of
the
cli-
mate
of
the
Southwest.
Absence
of
sharp
frontal
passages
and
of
vigorous
invasions
of
moist
unstable
air
resells
in
drought.
Hence.
droughts,
like
large
storms,
do
not
occur
with
any
regularity
but
CLIMATE
AND
ACCELERATED
EIIOSION
IN
THE
SOUTHWEST
33
are
associated
with
nonrhythmic
occurrences
of
air
-mass
invasions.
Despite
the
great,
irregularity
in
the
occurrence
and
length
of
drought
periods,
much
useful
information
can
be
secured
through
analysis
of
mean
drought
frequencies.
There
is
a
minimum
amount
of
precipitation
which
may
be
said
to
break
a
drought
period.
This
varies
from
one
region
to
another
and
from
season
to
season.
Thus.
any
uniform
definition
of
drought
is
necessarily
more
or
less
arbitrary.
On
the
basis
of
experimental
work.
Shreve
(94.
p.
134)
determined
that
under'
desert
conditions
rains
of
less
than
0.15
inch
are
without
influence
on
soil
moisture
al
a
depth
of
15
cm.
except
under
special
conditions.
Other
workers
have
observed
that
small
amounts
of
precipitation
are
of
slight
value
to
crops
unless
they
follow
larger
amounts.
On
the
other
hand,
as
little
as
0.10
inch
in
41i
hours
may
repre-
sent
meteorologic
conditions
which
would
terminate
a
drought
on
the
western
grazing
ranges.
Cloudiness
and
high
atmospheric
hu-
midity
may
cut
daytime
evaporation
hisses
and
bring
about
con-
densation
and.
direct
absorption
at
night
so
that
the
effectiveness
of
the
actual
rainfall
is
greater
than
it
would
be
otherwise.
No
quan-
titive
measurements
of
conde»sation
or
absorption
are
available.
and
it
is
therefore
impossible
to
say
what
is
the
minimum
airiorini
of
rainfall
which
could
be
utilized
by
the
range
grasses.
In
the
analysis
which
follows.
drought
periods
are
considered
to
be
termi-
nated
by
rains
totaling
0.10
inch
in
4
hours,
Drought
frequencies
for
the
four
seits
,
ins
and
the
year
at
four
Arizona
stations
are
given
in
figure
21.
At
Yuma.
drought
periods
of
120
days
duration
mar
be
expected
every
year.
and
once
in
a
10
-
year
period
a
drought
may
exceed
.2.20
days.
Every
taw
of
the
sea-
sons
may
be
entirely
rainless.
Occasionally.
a
drought
period
may
extend
through
two
Consectitive
sellsoils
and
sometimes
through
three.
Spring
drought
may
extend
through
the
entire
00
days
2
years
out
of
4;
autumn
drought.
2
years
out
of
5;
slimmer
drought.
2
years
out
of
7:
and
winter
drought.
2
years
out
of
9.
At
the
other
stations
droughts
are
shorter.
At
Jerrane.
as
at
Yuma,
drought
periods
are
longer
in
spring
and
autumn
than
in
summer
or
winter.
In
Phoenix
and
Tuba
City.
however,
the
autumn
droughts
occurring
once
in
10
year's
are
shorter
than
those
of
s
ummer
o
r
white
r
.
In
figure
22,
the
length
of
the
drought
period
to
be
expected
once
in
5
years
in
each
of
the
.onr
seasons
is
shown
for
the
Southwest.
The.
period
varies
in
the
different
pints
of
the
region
and
with
the
season.
The
mininnun
drought
period
shown,
less
than
30
days,
appears
in
the
mountainous
parrs
of
New
Mexico
and
Colorado
in
all
but
the
autiumi
season.
In
some
isn't
of
the
region
droughts
of
90
days
duration
are
experienced
in
every
season
once
in
5
years
on
the
average.
The
shifting
seasonal
drought
pattern
is
the
obverse
of
the
pattern
of
rainfall
(lisiribution.
sus
may
be
seen
by
comparing
fi
gure
22
with
figure
U.
Excluding
the
arid
areas.
summer
droughts
are
longest-
to
the
west
rind
winter
droughts
are
longest
to
the
east
of
the
region.
Summer
droughts
reach
90
days
in
western
Arizona,
in
virtually
all
of
California.
most
of
Nevada.
and
parts
of
Oregon,
Washington.
Idaho,
and
Vtah.
Winter'
droughts
reach
00
days
in
western
Kansas
and
eastern
Colorado
and
exceed
70
days
in
the
32:1-i
84
TEM:DUCAL
BULLETDT
SOS
3
IL
S.
DEPT.
OF
AGRICULTURE
entire
southern
Great
Plains.
Longer
droughts
are
experienced
in
the
Southwest
1
year
in
10,
on
the
average,
but
the
seasonal
variations
in
pattern
of
distribution
are
similar
to
those
of
the
5
-year
frequency.
Throughout
most
of
the
Southwest,
drought
periods
are
shorter
in
summer
than
in
any
other
season.
In
most
of
the
region,
spring
is
the
season
of
maximum
length
of
drought
and
autumn
stands
Ile
120
t".
E
10
60
r-
140
0'0
160
150
.0
1
OME
40
30
3
2
FR
COMICT
IT
64131
a
to
z
YLFMA
4
6
FR
COOCHCY
1.TE
*P51
130
CO
11:1
S
CO
30
0
t20
5
20
60
1.1
-
10EN1X
3
f1EQUENCT
NEAPS/
sm
,
TUBA
01
TY
4
5
iitCpuCKCY(VC411S11
FIGURE
21.
—Annual
and
seasonal
drought
frerinenoles
Ymna,
Jerome.
Phnenix,
and
Tuba
Cily,
Anix.
second.
In
the
eastern
part
of
the
area,
however.
winter
is
the
season
of
longest,
drought
periods,
since
the
winter
invasions
of
Polar
Pacific
air,
bringing
precipitation
to
the
Southwest,
fall
off
in
frequency
and
become
progressively
drier
from
west
to
east.
In
the
extreme
western
part
of
the
region
summer
droughts
are
the
longest.
This
is
also
related
to
the
incidence
of
air
masses,
in
this
instance
the
relative
infrequency
of
invasion
of
moist
unstable
Tropical
Atlantic
air
into
the
western
part
of
the
region.
I
.
CLIMATE
AND
ACCELERATED
EROSION
DI
TFTJ
SOUTHWEST
35
The
occurrence
of
a
prolonged
drought
does
not
preclude
the
possi-
bility
of
having
annual
or
monthly
rainfall
totals
which
are
equal
to
or
above
normal.
In
1909,
at
Kearns
Canyon,
only
0.15
inch
of
rain
fell
between
March
31
and
July
2.
On
July
3,
1.01
inches
fell.
No
rain
fell
between
September
13
and
November
12,
but
on
November 13
and
14
there
was
a
storm
which
brought
a
total
of
0.60
inch
of
precipitation.
The
total
rainfall
for
1909
was
above
the
normal.
An
abnormally
wet
month
or
year
may
reflect
mainly
the
occurrence
of
a
few
large
storms,
and
low
totals
do
not
necessarily
indicate
that
droughts
have
been
frequent
or
long.
There
is
every
reason
to
expect
at
long
intervals
in
the
Southwest
a
drought
extending
through
most
of
a
year.
Occasionally
such
droughts
may
follow
each
other
in
successive
years.
Such
dry
periods
are
normal
features
of
the
climate
of
the
Southwest.
THE
Cr.tm.vric
PATTERN
IN
THE
SOUTHWEST
A
DEFINITION
OF
CLIMATE
Climate
is
an
integration
of
the
climatic
or
meteorological
ele-
ments
or
factors
which combine
within
a
region
to
give
it
its
char-
acter
and
individuality.
The
catalog
of
the
elements
is
familiar,
consisting
❑f
temperature,
wind,
precipitation.
atmospheric
humidity,
evaporation,
sunshine.
cloudiness,
and
several
others.
These
ele-
ments
are
extremely
diverse,
temperature
being
merely
a
form
of
molecular
energy,
wind
a
form
of
momentum,
and
precipitation
a
material
which
collects
on
the
land
in
varied
forms
(rain,
snow,
hail),
in
various
amounts,
and
at
varying
rates.
Drought
is
an
absence
of
precipitation.
and
evaporation
a
rate
of
loss
of
precipita-
tion.
All
are
complexly
interrelated
:
each
is
dependent
on
the
others,
and
all
are
expressions
of
the
operation
of
meteorological
forces
world
wide
in
their
scope.
IngeiliNts
and
accurate
instruments
have
been
developed
for
meas-
uring
and
recording
the
elements
of
weather
and
climate.
There
are,
however,
no
instruments
for
measuring
the
climatic
complex.
For
the
characterization
of
climate
it
is
necessary
to
select
measured
values
of
the
individual
elements
in
an
effort
to
arrive
at
those
which
arc
correlated
with
the
significant
physical
and
biological
features
of
the
different
parts
of
the
earth.
To
obtain
this
end,
at-
tention
has
at
different
times
been
focused
on
different
sets
of
elements.
A
climatic
index
that
has
been
proved
to
be
highly
significant
is
the
relation
of
precipitation
to
evaporation,
but
it
is
one
that
is
extremely
difficult
to
determine.
Present
measures
of
both
precipita-
tion
and.
evaporation
are
grossly
inadequate,
The
analysis
of
pre-
cipitation
in
earlier
pages
has
shown
that
total
precipitation
expressed
numerically
has
little
significance
since
all
precipitation
is
included,
regardless
of
the
conditions
under
which
it
falls.
The
rainfall
of
a
crop
season
is
a
composite
made
up
of
a
series
of
rains,
each
possessing
an
individual
pattern
of
distribution.
The
rains
may
be
gentle
showers
or
downpours,
long
or
short
in
duration.
Periods
without
rain
may
last
for
a
few
days
or
for
several
months.
Tire
measurements
of
evaporation
are
even
less
satisfactory
than
those
of
precipitation,
and
it
is
only
within
the
past
2
years
that
36
TECHNICAL
BULLET=
8
08
7
U.
S.
DEPT.
OF
AGRICULTURE
1111111 1
r
{11Illlllllr
IIIIIII
t
ti
!
.....
.
1
Ill!
111
11
I
VII
11
101111111
1
1111111
11
11111
111
lb
loll
.1
SUMMER
ilI
Consecutive
Days
•.;
20
40
60
80
10
30
50
70
90
FinritE
Crito4vr.iitivi•
ilnys
f