The Nimbus-4 Backscatter Ultraviolet (BUV) atmospheric ozone experiment — tow years' operation

Abstract

In April 1970 the Backscatter Ultraviolet (BUV) experiment was placed into orbit aboard the Nimbus-4 satellite. This double monochromator experiment measures ultraviolet terrestrial radiance at twelve discrete wavelengths between 2550 and 3400 . Approximately 100 scans covering a 230 km square are made between terminator crossings on the daylight side of the earth. A colinear photometer channel with the same field of view is used to derive the Lambert reflectivity of the lower boundary of the scattering atmosphere. The extraterrestrial solar irradiance is measured at the northern terminator. The instrument has currently produced almost three years of nearly continuous data which are being used to infer the high-level ozone distribution and total ozone on a global basis. The high-level ozone data have been verified by independent coincident rocket ozone soundings, and the total ozone values show good agreement with Dobson spectrophotometer determinations as well as those made with the Infrared Interferometer Spectrometer also on Nimbus-4. An increase has been observed in equatorial radiance at 2550 relative to 2900 , which seems to indicate that the amount of ozone in the upper stratosphere is related to the eleven-year solar cycle.

Full text

-'-
'~
-.' . ~ . --, ·.-.
-.
.:~
' - : , -: ..
~
.;
-: .
;--.,
. . X
,-6
5 1
.:-'7~3:64
-
>i-.
c ~
'~"/ " -:'~ '- "'-~,¥'Y-,' ..
~-
~
,~~
~ ~ ~ ~
~~
.. :.--'
?.I
'?
/ ,
-"
-..
2",~
~T
' :- -"
'~"
'~MT ..
;-'
·
-- '
L'~ :'
:
:'' ''¥
' * -*"* ' '
R'
-/
,·:c
~.' --
~-·-i
,
~~"~/...--.x
" ' ~
'~~~~~~~
-'' ~
·:%
-.
''
- -.
'.,.
' r
').'"%-\,,' ,'- _.
,_-
~, - ,, _.,I '
'' ,~''
_'..~--.-
) .-
',
~
''x:;c
~ . .,
~.
-,
,.-
-.
' ' ' '-'~'
,/'C
.J-I''~.
-~ ...
:--
:.
·~~~~~~~~~~~i~~.
-.
-;
.
~,
.~ ~.,-,
.~ . - x
.~.. -:~ : ~.~ .J -I,_
-;
-., .
··
'-.... ....
7 ~ . . , . -
,.
~-,-
, ~ - /
).~-~
~
x., ''..
... t,.
t
'~
0----'4
'~,~'~ -)~
':~"~`1
'-
-£./R'-~'- ;'
--
,.,:
?`I-
...
cT~' ·I
~-
'
-T,'U
NI
M-BUS
4.,BA;
{ r ~ :I~~.~ '.
'
-' .·~'<~-~~
' /
',
~ , f
,,'--.
'
''"
.
,..~,',~/~'-"' ;;
I,,~
':.'/:'~-
~"
-. ¥ ...
,:..-_
,111T-g,
AVI;O
L~~~~~~~h~f:I
T
(PBUV
ATl":MOSPHER
' -"
~"',~
'1--
-L J-~ ' ~
.,'':
-~
' ?
~--.-,
'->
1.
.-
:..
.
~~~~~~,v
-.-
~.>1,~_ ... ~~ir(~~~:
...
'--...-~">-'
I\
''....
i ~
~~
; yL~':'
-·'"'
'~V:
"
~
...
.f/,.
.. .
I'V~
> -- , - ..
,-,~
.,
·~~~~~~~~~~~~c~
;
;r
· :·
,~~~~~-
~.r~:
,.~, , ,-
..
_.. -~
~_
, .-
;~_
.
,-
~~.
,7
·~~~~~·c ·~~~~~~~~~~'' I· ~~~~~~~~~~~~~
":,~~~~~~~~~~~~=
·~~.
~~~~~~~~~1-1EE·
,'~-
_'~.'.-~ "-'
/. .- ~..~ ; . '
·"
....
, --
,.,)._' ,,., ',,.-,,I.,
-
,,
. ;::<
'./.,:
..
~i----
...
,...-?,
-
:.:-.
.--
q,."'.
, :
/:-'
)
,i_,
'.I
~--~
,
'- v--'
.
>.i~~EP,
\
-E~-''..." '>-', I''.-.-.-·:.:
'% ~ l
~-"','--'': /,1 ''-."';'-"'. :''
-.
',;' ,,.',;.7/
'
,.
-'~',,-k ' ~ : -~1
-. ' - . ,
'/,·
'~~':Z~
~ .
~,~
...
ih:~~~~~~~~~~~
,
%.,.~/.~,.
.
.',.
-L3
~
I~-'
ru"
·.·
.~
:~..._.
.>
.. : ~
/, (.-.,x,'_
I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~··1·~ ,-"~-- ·
..'
~~/,_
-.
~_%
.
x_
,L,.-._
.-
,'l_'
~.-- ~: ~-:{
'
'.d.---~
~
:~
'-~~~~r~~,·
'~,
- . ' ' f 2
:-
, '---I~ _
:--'--/.'., .t-
"
,-~~
,~
'x
Q.,%
q'~,~,
'
::?
.L'
J):
'z
:~,
'>
'l
%-':,
..-
~'--A~"
c:-'-'u ?^':+,u
'
¥.:~'
!-
',..
% ':
·-_..p,-..:~ _~
v
.,,~"~'J
-.
,.
-"-.
~..~..
.~.-~_~'. -~~i
,3
~.~ ,.'.~/
,\
.~~~~~~:· ·^_;i-
J
~
., .....
~.
~
'~
.....
~ - , ..4 ; .
~~~~~
~~~~~~~- ,
~.~
.
,._.-
~ ,,
~j
~
.~<
'%
~
1~~~~~~~~~~~~~~~~~~~~~~~~~~
.~.,
~~.
·
~.~·
.-i
f,
' / · . .. ..
:
·~
_ ~ " ' "
L-
~-
~ -"
'-?
-, ~>
~. .. ~ ; '--.:. x :l .'/~-t- '
,",-
- ' .
,.
' '
",'
~
',,
.- ~ .- , ~
';.r
<,:
-';7~ ~ ~
~~~~~~~~~~~%
,'~*-." ;,
'::'-.
~~~~~~~~~;-~~~~~~~~~~~~~~~~
;" t-
--
a~ '1--'1'
'~

THE
NIMBUS-4 BACKSCATTER
ULTRAVIOLET
(BUV)
ATMOSPHERIC
OZONE
EXPERIMENT-TWO
YEARS'
OPERATION
Donald
F.
Heath'
Carlton
L.
Mateer
2
Arlin
J.
Krueger'
February
1973
1
Goddard
Space
Flight
Center,
Greenbelt, Maryland
20771,
U.S.A.
2
Atmospheric
Environment
Service,
4905
Dufferin
Street,
Downsview,
Ontario,
Canada.
GODDARD
SPACE
FLIGHT
CENTER
Greenbelt,
Maryland
i

Preceding
page
blank
CONTENTS
INTRODUCTION.
PHYSICAL
BASIS
....................................
INSTRU
ME
NTATION
..................................
Optical
........................................
Electronics
............................
..
OBSERVATIONS
.....................................
ACKNOWLEDGEMENTS
................................
REFERENCES
.......................................
PECEDING
PAGE BLANK
NOT
FILME)
iii
Page
1
2
4
4
6
8
12
13

THE
NIMBUS-4
BACKSCATTER
ULTRAVIOLET
(BUV)
ATMOSPHERIC
OZONE
EXPERIMENT
-
TWO
YEARS'
OPERATION
ABSTRACT
In
April
1970
the
Backscatter Ultraviolet
(BUV)
experiment
was
placed
into
orbit
aboard
the
Nimbus-4
satellite.
This
double
monochromator
experiment
measures ultraviolet
terrestrial
radiance at
12
discrete
wavelengths between
2550
A
and
3400
I.
Approximately
100
scans
covering
a
230
kilometer
square
are
made
between
terminator
crossings
on
the
daylight
side
of the
earth.
A
colinear
photometer
channel
with the
same
field
of
view
is used
to
derive
the
Lambert
reflectivity
of
the
lower
boundary
of
the
scattering
atmosphere.
The
extraterrestrial
solar
irradiance
is
measured
at
the
northern
terminator.
The
instrument
has
currently
produced
almost
three
years
of
nearly
continuous
data
which
are
being
used
to
infer
the
high-level
ozone
distribution
and
total
ozone
on
a
global
basis.
The
high-level
ozone
data
have
been
verified
by
independent
coincident
rocket
ozone
soundings,
and
the
total
ozone
values
show
good
agree-
ment
with
Dobson
spectrophotometer
determinations
as
well
as
those
made
with
the
Infrared
Interferometer
Spectrometer
also
on
Nimbus-4.
An
increase
has
af
O
been
observed
in
equatorial
radiance
at
2550
A
relative
to
2900
A,
which
seems
to
indicate
that
the amount
of
ozone
in
the
upper
stratosphere
is
related
to
the
11-year
solar
cycle.

1.
INTRODUCTION
That
satellite
measurements
of
scattered
ultraviolet
solar
radiation
from
the
terrestrial
atmosphere
could
be
used
to
deduce ozone
profiles
on
a
global
basis
was
first
suggested
by
Singer
and
Wentworth
[14].
The
dynamic
range
of
this
measurement
is
quite
severe,
since
the
radiance
of
the
terrestrial
atmosphere
increases
by
about
104
from
2550
X
to
3400
A
and
the
linear
polar-.
ization
of
the
radiation
may
range
from
0
percent
to
nearly
100
percent.
Further-
more,
the
atmospheric
albedo,
(that
is,
atmospheric
radiance/solar
irradiance)
must
be
determined
within
limits
of
error
of
1
to
3
percent
if
reliable
ozone
profiles
are
to
be
inferred.
Subsequent
satellite
applications
of
this
technique
have
been
reported
by
Rawcliffe and
Elliot
[13]
for
a
photometer
experiment;
by
Krasnopol'skiy
[9]
,
Iozenas
[6]
,
Iozenas
et
al.
[7,
8],
and
Anderson
et
al.
[1]
for
monochromator
experiments.
On
April
8,
1970,
the
Nimbus-4
satellite
was
launched
into
a
circular
polar
orbit
at
an
altitude
of
1100
kilometers.
The
100
retrograde
orbit
is
sun
synchro-
nous,
and
every
107
minutes
the
three-axis
stabilized
satellite passes
through
the
ascending
mode
near local
noon.
The
subsatellite
point
crosses
the
equator
in
increments
of
27°
in
longitude between
successive
passes,
and
reaches
a
maximum
latitude
of
800.
The
Backscatter
Ultraviolet
experiment
(BUV)
is
designed
to
measure
the
solar
irradiance
at
the top
of
the
atmosphere
and
the
atmospheric
radiance
in
the
satellite
nadir
direction,
thus providing
data
for
1

determination
of
high-level
ozone
profiles
and
total
ozone
amounts
on
a
global
basis,
with
a
spatial
resolution
of
230
kilometers.
2. PHYSICAL
BASIS
For
a
single-scattering
model
atmosphere
the
atmospheric
radiance
is
I(X,
) = F0
(X)(3,B/1671)(1
+
cos20)
f0
exp
[-
(1
+
sec
e
)(oaxX
p+
PXP)]dp
where
F0
(x)
=
extraterrestrial
solar
irradiance
(ergs/cm
2.
sec
· A)
I(X,9)
=
atmospheric
radiance
(ergs/cm
2*
sec
.·
.
ster)
PA
=
atmospheric
scattering
coefficient
(atm-')
cX=
ozone
absorption
coefficient
(atm
·
cm)
-'
Xp
=
amount
of ozone
above
pressure
p
(atm)
in
atm
· cm
0 =
solar
zenith
angle
The
inversion
of
terrestrial
atmospheric
radiance
to obtain
vertical
ozone
pro-
files
is
possible
since
atmospheric
radiance
at
a
particular
wavelength
originates
in
a
moderately
well-defined
effective
scattering
layer.
The
central
height
and
width
of
a
layer
is
principally
a
function
of
the
ozone
absorption
coefficient,
solar
zenith
angle,
and ozone
distribution.
An
example
of
effective
scattering
layers
is
given
in
Figure
1
for
a
solar
zenith
angle
of
60°
and
a
total
ozone
amount
of
336
m
atm
.
cm.
The
curves
are
normalized
to
unity
at
the
maxi-
mum.
Mateer [11]
provides
a
comprehensive
discussion
of
inversion
of
ultra-
violet
atmospheric
radiances
to
obtain
a
vertical
distribution
of
ozone.
The
basis
for derivation
of
total
ozone
from
the
satellite
observation
of
the
atmospheric
radiance
in
the
Hartly-Huggins
bands
is
provided
by Dave
and
2

Mateer
[3].
The
application
of
this
technique
in
the
BUV
experiment
has
been
described
by
Mateer,
Heath, and
Krueger
[12].
The
inference
of
total
ozone
is
made
from
measurements
at
ultraviolet
wavelengths
which
are
weakly
ab-
sorbed
near
the
long-wavelength
end
of
the
ozone
absorption
band.
The
absorp-
tion
must
be
sufficiently
weak
so
that
the
solar
photons
have
traversed
the
ozone
layer
with
significant
absorption
and
have
been
scattered
by
the
troposphere
and
ground
back
into
space.
A
wavelength
pair,
separated
by
about
200
X to
minimize
scattering
effects,
is
chosen
so
that
one
is
fairly
strongly absorbed
while
the
other
is
only
weakly
absorbed
by ozone.
Two
wavelength
pairs
are
necessary
to
cover
the
range
of
solar
zenith
angles
from
0°
to
90'.
The
two
line
pairs
are:
(A)
3125
A,
3312
A,
and
(B)
3175
A,
3398
A.
The
basic
observation
is
the
relative
logarithmic
attenuation
N(3125,
3312)
=
log,o
{
F/I
} 312 5 -
log
{F
/I}
33
12
where
F0=
extraterrestrial
solar
irradiance
I =
terrestrial
radiance
the
total
ozone
is
estimated
by
comparing
the
observed
value
of
N
with
values
which
have
been
calculated
previously
for
the
following:
a)
Solar
zenith
angles
from
00
to
90°
b)
Two
lower
boundary
pressures
of
400
mb
and
1000
mb
c)
Series
of
standard
ozone
profiles
d)
Range
of
Lambertian
reflectancies
3

This
procedure
minimizes
computer
time
since
a
multiple
scattering
problem
is
involved.
3.
INSTRUMENTATION
a.
Optical
The
BUV
experiment
is
located
in
bay
9
of
the
Nimbus
sensory
ring.
In
this
location
the
experiment
views
the
earth
in
the
satellite
nadir
direction
with
a 12°
field
of
view,
a
square
230
kilometers
on
a
side at
the
surface.
At
the
end
of
a
measurement
cycle
of
32
seconds,
the
subsatellite
point
has
moved
about
200
kilometers.
The
instrument basically consists
of
a
double
(tandem)
Ebert-Fastie
spectro-
photometer
in
conjunction
with
a
narrow-band
interference
filter
photometer.
The
spectrophotometer
is
designed
to
measure
spectral
intensitites
at
12
wave-
lengths
from
2555
A
to
3398
A
with
a
10
A
bandpass
while
the
interference
filter
photometer
records
a
50
A
band
centered
at
3800
R.
This combination
of
in-
struments
measures
the
backscattered
ultraviolet
solar
radiation
from
the
day-
time illuminated
atmosphere,
as
well
as
the
direct
solar
rays reflected
off
a
diffuser
plate.
The
layout
of
the
optical
components
of the double
monochromator
and
interference
filter
photometer
is
shown
in
Figure
2.
Sunlight
diffusely
reflected
from
the
atmosphere
enters
the
two
systems
through
the
optical
horns.
A
double-Lyot
type
depolarizer
is
inserted
in
front
of
the
entrance
slit
of
the
spectrophotometer
in
order
to
render
its
throughput
independent
of
the
state
of
4

polarization
of
the
incident
radiation.
The
double
monochromator
is
composed
of
two
25-cm
focal length
Ebert-Fastie
monochromators.
Light
entering
the
entrance
slit
is
rendered
parallel
by the
spherical
collimator
mirror
and
is
then
diffracted
by
a
52-by-52-mm
grating
of
2400
grooves/mm
(solid angle
is
0.
043
steradians).
The
diffracted
light
returns
to
the
spherical,
collimating
mirror,
passes
through
the
roof
prism,
and
is
imaged
onto
an-
intermediate
slit.
The
light
passing
through
the
intermediate
slit
is
dispersed
again
by
the
second
monochromator.
A
field
lens
at the
exit
slit
is
used
to
image
the
gratings
onto
the photocathode
of
the
photomultiplier.
Both
gratings
are
mounted
on
a
common
rigid
shaft
to
eliminate
wavelength
tracking
errors
between
the
two
monochro-
mators.
A
roof
prism
is
used
to
invert
the
image
in
the
direction
of
dispersion
at
the
intermediate
slit,
in
order
to double
the
dispersion
by
passing
through
the
second
monochromator.
A
slit
aperture
is
placed
in
the
interference
filter
photometer
so
that it
will have
the
same
field
of
view
as
the
spectrophotometer.
Both
of
the
monochromator
and
photometer
slits
are
aligned
at
10°
to
the
space-
craft
velocity
vector.
When
passing
over
the
polar
regions,
diffuser
plates
are
deployed
in
front
of
the
spectrophotometer
and
the
filter
photometer
to
measure
the
extraterrestrial
solar
irradiance.
The
primary
reason
for
using
a
double
monochromator
is
to
obtain
the
de-
sired
spectral
purity,
that
is,
the
elimination
of
scattered
light,
which
permits
one to
use
a
high
quantum
efficiency
(20
percent
for
the
12
wavelengths)
photo-
multiplier
tube
with
a
dark
current
of
the
order
of
10 '
amp
at
a
gain
of
106.
5

The
instrumental
scattered
light
contributes
less
than
1
percent
of
the
atmos-
pheric radiance
signal
at
2550
A.
From
an
extensive
ray-trace
analysis,
which
was
subsequently
confirmed
by
experimental
observations, it
was
found
that
the
optical
aberrations are
less
at
the
exit
slit
of
the
double
monochromator
than
they
are
at
the
exit
slit
of
a
corresponding
single
monochromator.
This
is
related
to
the
fact
that
certain
optical
aberrations
cancel
as
a
result
of
the
second
reflection from
the
spherical
collimating
mirror
in
the
Ebert-Fastie
single monochromator.
A
similar
cancelling
of
aberrations
is observed
in
passing
through
the
second half
of
the
double
monochromator.
The
wavelength
scan
is
accomplished
by
rotating
the
grating
drive
arm
with
a
stepper-motor-driven
cam.
The
cam
has
a
fast
rise
rate
when
moving
from
one
of
the
12
wavelengths
to
the
next,
and
also
maintains
a
constant
radius
so
that
a
given
wavelength
is
displayed
to
an
accuracy
of
0.2
A
at
the
exit
slit.
The
transfer
time from
one
wavelength
to
the
next
is
0.
5
sec,
and
total
scan
cycle
time is
32
sec.
A
cam-driven
system
gives
accurate,
reproducible
wavelengths,
and
it requires
only
one
point
for
wavelength
calibration.
After
two
years
of
operation
on
Nimbus-4, the
BUV
double
monochromator
has
main-
tained
a
wavelength
accuracy
of
better
than
1 A.
b.
Electronics
The
signal-processing
electronics
of
the
monochromator
and
photometer
channels
are
virtually
identical
and only
one
channel
will
be
described.
The
6

nine
decades
of
optical
signal
levels
received
at
the
photomultiplier
are
ac-
commodated
by
a
programmed
high
voltage
power
supply,
a
three-range
electrometer,
and
a
two-mode
digital:
ratemeter.
The
photomultiplier
is
op-
erated
at
two
discrete
gain
states
by
automatically
adjusting
the
high
voltage
power
supply.
In
the
low
gain
state
only
analog
current
signals
are
obtained,
while
in
the
high
gain
state
pulse-count
data
are
available
in
addition
to the
analog
signals.
In
the
analog mode,
the
signal-processing
channel
consists
of
an
electrometer,
an
automatic
range
sequencer,
and
a
logarithmic
analog-to-digital
converter.
The
electrometer
produces
an output
voltage
which
is
proportional
to
the
average
input
current.
The
output voltage
is sensed
by
an
automatic
range
sequencer
and
is
utilized
to
select
any
of the
three
feedback
resistors
and
the
two
high
voltage
levels,
hence
controlling
the
system
gain.
Gain
changes
can
occur
every
200
milli-
seconds
except
during
the
automatic
reference
adjustment
period.
Possible
gain
states
are
107,
2x10
9
and
3x10
10
volts/ampere
in
the
low
voltage
state
and
10 7
volts/ampere
in
the
high
voltage
state.
The
signal-processing
channel
in
the
pulse-counting
mode
consists
of
an
electrometer
(common
to
both analog
and
pulse-counting
modes),
a
pulse
amplifier/shaper,
a
single-channel
pulse-height
selector,
and
a
digital
ratemeter.
In
the
pulse-counting
mode
the
electrometer serves
as
a
wide-band
pulse
7

amplifier.
The
pulses
from
the
electrometer
are
further
amplified
and
ap-
propriately
shaped
for
input
to
the
discriminators.
The
high
and
low
level
discriminators
and
an
anticoincidence
circuit
form
a
single-channel
analyzer.
The
low
level
discriminator
removes noise,
while
the
upper
discriminator
is
utilized
to
compensate
for
radiation-induced
pulses.
The
output
pulses
from
the
anticoincidence
circuit are
counted
for
the 1.8
second
dwell
time
at each
wavelength
step.
4.
OBSERVATIONS
A
computer plot
of
the
experiment
data
from
one
complete
orbit
is
shown
in
Figure
3.
The
radiances
found
with
the
photometer
and the
12
monochromator
channels
are
given
by
the
upper
12
lines
in
this
plot.
The
units
are
ergs/cm
2.
sec
· A .
ster
and
ergs/cm
2.
sec
· A
for
the
atmospheric
radiance
and
solar
irradiance,
respectively.
The
top
of
the
ordinate
scale
is
102.
From
left
to
right
the
data
record
begins
with
earth
radiance
measurements
in
the
northern
hemisphere.
Moving
northward,
the
radiance
decreases
with
increasing
solar
zenith
angles.
At
the
northern
terminator
the
diffuser
plate
is
deployed
for
two
complete
scans.
Then
the
diffuser
is
stowed
and
the
spacecraft
moves
toward
satellite
night.
A
residual
signal
due to
sunlight
reflected
off
the
en-
trance
horns
drops
abruptly
with
passage
into
satellite
night.
The
daytime
equatorial
signal
is
about
100
times
above
the
background;
the
nighttime
back-
ground
signal
is
about
a
factor
of
ten
greater
than the
typical
dark
current
ob-
served
in
the
prelaunch
testing.
The
large
nighttime
increase
in
signal
is
due
8

to
photomultiplier
currents
induced
as
the
spacecraft
passes
through
a
portion
of the
South
Atlantic
anomaly
of
the
Van
Allen
radiation
belts.
A
global
contour
map
of
radiation-belt-induced
dark-currents
is
shown
in
Figure
4.
On
the
figure,
the
units
when
multiplied
by
10
-15
amperes
are
cathode
current.
A
contour
of
5x10-'5
amperes
is
equivalent
to
10
percent
of
the
signal
produced
by
the
equatorial
radiance
at
2550
A.
Consequently,
a
significant
part
of
the
data
reduction
effort
is
directed
towards
establishing
a
dark-current
correction
to
be
applied
to
the
observations.
Some
evidence
for
secular
changes
in
high-level
ozone
has
been
found.
Atmospheric
radiances
in
the
vicinity
of
2900
A
appear
to
have
remained
constant
over
a
two-year
time
span,
implying
that
ozone
in
the
30-
to
40-kilometer
region
is
quite
stable.
At
the
shorter
wavelengths,
from
the
radiances
increased
during
the
same
time
period.
This
could
imply
that
the
high-level
ozone
distribution
is
de-
creasing
with
the
decreasing
phase
of
the
eleven-year
sunspot cycle.
Nimbus-4
was
launched
a
little
more
than
one
year
after
solar
maximum
of
the
current
sunspot
cycle.
Similar
observations
of
a
decrease
in
atmospheric
ozone
in
the
upper
stratosphere
with
declining
solar
activity
have
also been
reported
by
Dutsch
[2]
from
Umkehr
data
and
by
Paetzold
[15,
16]
from
balloon
data.
A
significant
solar-cycle
variation
in
the
solar
irradiance
which
dissociates
02
in
the
upper
stratosphere
and
lower
mesosphere
has been
reported
by
Heath
[5]
from rocket
and
Nimbus-3
and
Nimbus-4
satellite
experiments.
9

The
fundamental
observation
of
the
BUV
experiment
is
the
ultraviolet
terrestrial
albedo,
which
is defined
as
the
ratio
of
earth
radiance
to the
extra-
terrestrial
solar
irradiance,
I/F
0. A
typical
equatorial
albedo
determination
by
the
BUV
experiment
is
shown
in
Figure
5.
The
crosses
represent
the
equa-
torial
albedo
observed
with
the
U.
S.
S.
R.
COSMOS
satellites
using
a
double
monochromator
in
1965
and
1966.
The
observations
have
been
reported
by
Krasnopol'skiy
[9],
Iozenas
[6],
and
Iozenas
et
al.
[7,
8].
These
measure-
ments
occurred
at
solar
minimum
of
the
last
sunspot
cycle.
The
agreement
is
excellent
at
3000
A,
however,
the
curves
diverge
as
one
goes
to
shorter
wave-
lengths.
This
trend
of
increasing
equatorial
albedo
with
decreasing
sunspot
activity
is
the
same
as
that
observed
with
the
BUV
experiment.
One
cannot
dismiss
completely
that
it
is
an
artifact
due
to
instrumental
scattered
light,
although
the
accumulation
of
experimental
evidence
would
seem
to
indicate
otherwise.
Only
the
U.
S.
S.
R.
experiments
on
the
COSMOS
satellites
and
the
BUV
experiment
on
Nimbus-4
measured
the
extraterrestrial
solar
irradiance
in
conjunction
with
terrestrial
atmospheric
radiances.
Others
have
had
to
rely
on
published
values
of
the
ultraviolet
solar
irradiance,
which
may
introduce
serious
errors
in
the
values
for
albedo
if
there
is
an
eleven-year
solar-cycle
variation
in
the
ultraviolet
solar
irradiance
as
reported
by
Heath
[5].
Figure
6
is
an example
of
total
ozone
above
the
10
and
2.8
mb
pressure
levels
displayed
on
northern
and
southern
polar
stereographic
projections.
Gaps
in
the
data
are
the
result
of
failure
of
an
onboard
tape
recorder,
of
the
10

absence
of
a
nearby
tracking
station
to
carry
out
satellite
interrogation,
or
high
dark-current
backgrounds
induced
by
the
inner
regions
of
the
Van
Allen
radia-
tion
belts.
Figure
7
is
an
example
of
derived
total
ozone
for
both
hemispheres.
Prabhakara
et
al.
[17]
has
derived
global
maps
of
total
ozone
from
ob-
servations
in
the
9.6
micron
band of
03,
utilizing
the
Infrared
Interferometer
Spectrometer
(IRIS)
experiment
flown
on
Nimbus-3
and
-4.
An
early
comparison
of
inferred
total
ozone
for
the
BUV
and
IRIS
experiments
on
Nimbus-4
shown
for
a
single
daylight
passage
in
Figure
8.
The
shipboard
Dobson
spectropho-
tometer
measurements
by
White
and
Krueger
[18] shown
in
this
figure
required
54
days,
whereas
the
satellite
observations
were
made in
54
minutes.
The
photometer
channel
at
3800
A
can
be
used
to
infer
approximate
Lambertian
surface
reflectivities
at
the
wavelength
pairs
used
in
the
derivation
of
total
ozone.
An
investigation
of
this
approximation
from
upward
and downward
spectrophotometric
flux
measurements
have
been
made
by
Furukawa
and
Heath
[4]
from
the
NASA/Convair
990
research aircraft
for
the
determination
of
sur-
face
reflectivities
in
the
wavelength
range
of
3000
A
to
4800
A.
Until
these
re-
sults
can
be
incorporated
into
the
BUV
analysis
programs,
an
interim
approach
has
been
to
evaluate
an
equivalent
surface
albedo which
has
been
derived
from
a
regression
analysis
between
Dobson
spectrophotometer
"ground
truth"
measure-
ments
and
the
BUV
data,
for
example,
see
Mateer,
Heath, and
Krueger
[12].
In
general,
the
standard
errors
of
estimate
between the
BUV
and
Dobson
de-
rived
ozone
values
are
about
0.
02
atm-cm.
The
satellite
values
appear
to
be
11

a
little
too
low
or
a
little
too
high,
depending
upon
whether
the
total
ozone
is
low
or
high
respectively.
The
basic
BUV
experiment
data
consist
of
unedited
radiance
data
which
are
stored
on
archive
tapes
at the
World
Data
Center
A.
Production
of
total
ozone
data
and
high-level
ozone
profiles
continues
as
confidence
in
the
validity
of
the
data
and
analysis
increases.
Available
data
will
include
global
atlases
of
total
ozone,
height
distributions
in the
form
of
the amount
above
different
pressure
levels,
and
the
basic atmospheric
radiance
data.
5.
ACKNOWLEDGMENTS
The
success
of the
BUV
experiment
and
its
continued
operation
after
nearly
three years
in
orbit
are
due
in
large
part
to
efforts
of the Beckman
Instrument
Corp.
and
the
Analog
Technology
Corp.
who
built
the
experiment.
These
con-
tributions
are
gratefully
acknowledged.
12

REFERENCES
[11
G.
P.
Anderson
et
al.,
Satellite observations
of the
vertical
ozone
distri-
bution
in
the
upper
stratosphere,
Annales
de.
Geophysique,
25
(1969),
341-345.
[2]
J.
V.
Dave
and
C.
L.
Mateer,
A
preliminary
study
of
the
possibility
of
estimating
total
atmospheric
ozone
from
satellite
measurements,
J.
Atmos.
Sci.,
24
(1967),
414-427.
[3]
H.
U.
Dutsch,
Atmospheric
ozone and
ultraviolet
radiation,
Climate
of
the
Free
Atmosphere,
vol.
4
of
World Survey
of
Climatology,
Elsevier
Publishing
Co.,
1969,
382-432.
[4]
P. Furukawa
and
D.
F.
Heath
(to
be
published).
[5]
D.
F. Heath,
Space
observations
of
the
variability
of
solar
irradiance
in
the
near
and
far
ultraviolet,
J.
Geophys.
Res.
(in
press)
(1973).
[6]
V.
A.
Iozenas,
Determining
the
vertical
ozone
distribution
in
the
upper
atmospheric
layers
from
satellite
measurements
of
ultraviolet
solar
radiation
scattered
by
the
earth's
atmosphere,
Geomag.
Aeron.,
8
(1968),
403-407.
7]
V.A.
Iozenas
et
al.,
Studies
of
the
earth's
ozonosphere
from
satellites,
Izv.
Atmos.
Oceanic
Phys.,
5
(1969),
77-82.
[
8]
V. A.
Iozenas
et
al.,
An
investigation
of
the
planetary
ozone
distribution
from
satellite
measurements
of
ultraviolet
spectra,
Izv.
Atmos.
Oceanic
Phys.,
5
(1969),
219-233.
13

[9]
V.A.
Krasnopol'skiy,
The
ultraviolet
spectrum
of
solar
radiation
re-
flected
by
the
terrestrial
atmosphere
and
its
use
in
determining
the
total
content
and
vertical
distribution
of
atmospheric
ozone,
Geomag.
Aeron.,
6
(1966),
236-242.
[10]
A.
J.
Krueger,
D.
F.
Heath,
and
C.
L.
Mateer,
Variations
in
the
stratospheric
ozone
field
inferred
from
Nimbus
Satellite
observations,
Pure
and Applied
Geophysics,
vol.
19,
1973.
[11i
C.
L.
Mateer,
A
Review
of
Some
Aspects
of
Inferring
the
Ozone
Profile
by
Inversions
of
Ultraviolet
Radiance
Measurement,
in
The
Mathematics
of
Profile
Inversion,
edited
by
L.
Colin,
NASA
TMX-62
(1972),
pp.
2-25.
[12
]
C.
L.
Mateer,
D.
F.
Heath,
and
A.
J.
Krueger,
Estimation
of
total
ozone
from
satellite
measurements
of
backscattered
ultraviolet
earth
radiance,
J. Atmos.
Sci.
28
(1971),
1307-1311.
[13
]
R.
D.
Rawcliffe
and
D.
D.
Elliott,
Latitude
distributions
of
ozone
at
high
altitudes
deduced
from
satellite
measurement
of the
earth's
radiance
at
2840
A.
J.
Geophys.
Res.,
71
(1966),
5077-5089.
[14]
S.
F.
Singer
and
R.
C.
Wentworth,
A
method
for
the
determination
of
the
vertical
ozone
distribution
from
a
satellite,
J.
Geophys.
Res.,
62
(1957),
299-308.
[15]
H.
K.
Paetzold,
Variation
of
the
vertical
ozone
profile
over
middle
Europe
from
1951
to
1968,
Ann.
Geophys.,
25
(1969),
347-349.
14

[16]
H.
K.
Paetzold,
Secular
variations
of
the
atmospheric
ozone
layer
cover
middle
Europe
from
1951
to
1968,
XV
General
Assembly,
I.
U. G.
G.,
Moscow,
1971.
[17]
C.
Prabhakara,
E.
B.
Rogers,
and
V.
V.
Salmonson,
Remote
sensing
of
the
global
distribution
of
total
ozone
and
the
inferred upper-tropospheric
circulation
from
Nimbus
IRIS
experiment,
Pure
and Applied
Geophysics,
vol.
19, 1973.
[18]
W.
C.
White and
A.
J.
Krueger,
Shipboard
observations
of
total
ozone
from
30
N
to
60
S,
J.
Atmos.
Terr. Phys.,
30
(1968),
1615-1622.
15

0.2
0.4
0.6
0.8
CONTRIBUTION
FUNCTION
80
o o3
070
70
4 o
·d
-
C 0
50
40
I 0
, m o
w
F o
.-
4_
cd
-I
0 0
O0
> m cdI
-V
~ ~ 4
1.~~
PZ4W
16
id5
E
w
i) )n
w
L.
1i2
Io-*
10
I0 0

I
IP
!
J
IK a
Q)
-4
o
4>
0)
I
.0
,b
To
0)
.4
0
b©
17

lltl Ii
In11
II
l l
Illll IIII
ll11111
I
~(
lf
I 111
1f
11
I
______
______
I
LX LL1
_
__
_
;______
,,
'I, I
1 I
tl<Jl l' l! ll
i I I i i
ii
,
¥0)
\
\\
'\
1 \ L i
~
~~~~~/
I'~~
N s_ .I
/,i1
/1/iiJ
_ :
I t
1illSl1
II
III
I 1
1
111
i
ii
I
hI1
4-
*4444
A
4jlb/
4
44l]l ii
0
-Ie:
f
]110
IIllliL
UULILL
VILIUAIILUL
- -l
18
!
iI
I
II
ii
I
i
cj
o
Cd
la
o4-
0
4-
-i
0
0
4-
Fr4
I'
2,
Yl
i
2
i
!1
_ol
I

cd
.Xa
Tom
so
a
Ul
'e
O
o
0
To
'-40
100
f Cd
0 0
Q0)
04
0)
4oa
cd
_
0)
4-4
0 -
-a
0 C)
0
-4
'-40
0
i4i
~M
0~4
40,
.0
.- 0
4
0)
U,
19

'
1TI
I0I --
-,
0
0
0
4)
+
4
0
4-
+,
0
40
+0
0
0
BUV,
NIMBUS
4
+
IOZENAS,
et
al.
p -- -
*-
*I
3500
3000
WAVELENGTH
(A)
Figure
5.
Comparison
of
equatorial
albedo
observed
with
the
Nimbus-4
BUV
experiment
in
1970
and
the
U.
S.
S.
R.
experiments
on
the
COSMOS
satellite
series
in
1965
and
1966.
20
-TI
10-2
10-2
co
uA-
10
-3
-=0
2500
I
I
I
I
I
I
LII I I I
I

iTOTAL
OZONE
ABO-VE
10mb
OZONE AMOUNTS
IN
10
-4
ATM-CM
July
5,
1970
Figure
6.
Examples
of
a
global
map
of
the
amount
of
ozone
above
pressure
levels
10
mb
and
2.8
mbfor
the
northern
and
southern
hemispheres.
(Sheet
1)
,
21

TOTAL
OZONE
ABOVE
2.8mb
OZONE
AMOUNTS IN
10-4
ATM-CM
July
5,
1970
Figure
6.
Examples
of
a
global
map
of
the
amount
of
ozone
above
pressure
levels
10
mb
and
2.8
mb
for
the
northern
and
southern
hemispheres.
(Sheet 2)
22

in
mg
0
0
O
0C
o
o
0
.e
z
O
N
O
I--
so
0
43
I-I
m
N
'O
0
4c
o
Iz
0d
0~
23
u
U
i
z
Z
:E
z
4"
&aJ
z
O
z
N
O

ac~~~'c
=
.,
=
0
u,~~
o
-Cooc
~·,-
ac~~~~~~~~~~~~~~~~~~~~~~~~~G
6~4~C
~d=C
r I:
~~~~~~~~~~~c0N
Z
C2
0
cmcf
CA~~~~~~~~~~~~~~~~~~~~~C
MA
c;iS cd,
Lai
4"~~~~~~~~0
QC
-- ) O
"C
X
SC
CD
O
3 t7
Lh.
C,
oc
U
14 cd~~~~~2
NASA-GSFC