The Response of High Altitude lonization Chambers during

JOURNAL OF GEOPI-IY$ICAL RESEAR½I-I VOL. 70, NO. 13 JULY 1, 1965

The Response of High Altitude Ionization Chambers during the 1954-1965 Solar Cycle

R. ]H]. CALLENDER, J. R. MANZANO, 1 AND J. R. WINCKLER

School oi Physics and Astronomy, University of Minnesota, Minneapolis

Abstract. The response of an integrating ionization chamber at 10 g/cm •' depth in the atmosphere to particles of various rigidities is evaluated by using the change of ionization with latitude. This procedure yields the differential response curves at solar minimum and solar maximum and also the mean rigidity of response at any given latitude. For high latitude and Minneapolis, the mean responses are 2.5 bv and 3.2 by, respectively, at solar minimum and 3.6 and 3.8 at solar maximum. The solar cycle effect at 10 g/cm = is evaluated using 250 balloon flights at Minneapolis and at high latitude. Total ionization minimum lags about 18 months behind sunspot maximum. The high altitude ionization when compared with neutron monitors forms a single correlation curve over the solar cycle with significant deviations occurring only for a few months in 1957. We conclude that the relative rigidity effects between 3.6 and 15 by are very similar during the decreasing and increasing phases of the solar cycle. Com- parison of data from ionization chambers of Pioneer 5 and Mariner 2 with data from de- tectors on earth shows no definite gradient effects. The larger intensity changes show a correlation between earth and deep space, but real fluctuations of smaller amplitude are frequently not correlated.

Introduction. Integrating-type ionization chambers have long been used for studying various aspects of energetic particles and other radiation both in the high atmosphere on balloons and, more recently, in free space on satel- lites and space probes [Winckler, 1960; Neher and Anderson, 1962; Arnoldy et al., 1964]. If the general nature of the radiation is known, then this instrument can give the mean energy by measuring the loss rate by ionization inte- grated over the incident spectrum. This information has been of use in the study of aurora.1 and solar X rays and solar cosmic ray protons at high altitudes and in space [Winckler, 1962; Mosley et al., 1962; Winckler, 1963]. If the ionization chamber is supplemented by a Geiger counter having similar stopping power for the radiation under investigation, then the average ionization per count is a significant, quantity for estimating the mean energy of the radiation and, in some cases, its nature [Ho[mann and Winckler, 1963]. Because such ionization chambers are capable of long-term accurate calibra- tion and standardization, they are useful for studies of primary cosmic ray time variations

• NASA Argentine Space Research Committee Fellow on leave from the University of Tucuman.

such as the solar cycle and Forbush-type modulations. Also, attempts have been made to study the space gradient of primary intensity in the solar system near the earth's orbit using ionization chambers on space probes [Neher and Anderson., 1964; Arnoldy et al., 1964]. Since many months of flight data are required, reliability of ion chambers is a great. asset here.

If the ionization chamber is exposed to a complex radiation environment, a detailed in- terpretation usually becomes impossible. For example, attempts to determine the electron flux in some parts of the Van Allen radiation belts have failed because of the mixture of elec-

trons, X rays, and protons [Arnoldy et al., 1962]. Another case is that of an ionization chamber on a balloon at high altitude respond- ing to galactic cosmic rays. The response is due partly to direct primaries but also to secondary particles produced in the atmospheric transition of the primaries, and possibly to re-entrant albedo particles. Attempts to unravel these. effects and obtain any detailed information about the primary particles from a single total ionization rate measured at high altitudes from a balloon seem useless.

However, if high altitude ionization measurements are made over a range of geomagnetic

3189

3190 CALLENDER, MANZANO, AND WINCKLER

latitudes, then with knowledge of the local magnetic rigidity cutoff a response curve for the ionization chamber can be constructed. The

ionization ra.te I is related to the primary rigidity spectrum D incident on the earth by the relation:

I = • f• Dz(P, t)Sz(P, x) dP (1) z

where Z is the atomic number of the incident

pa. rticle, Px is the cutoff rigidity at the instrument, P is the rigidity, x is the atmospheric depth of the ion chamber, and $, (P, x) is its specific yield function.

Differentiating with respect, to cutoff rigidity, we obtain the differential response curve which is a function of depth, time, and rigidity of the primary particles, but no longer has reference

Fig. 1.

300

250

• 200

E

• 150--

z

_o I00--

50-

ION CHAMBER

I0 Mb 1958

O0 I 2 5 4 5 6 7 8 9 I0 II 12 13 14 15 16 17 PX GV

Latitude variation of total cosmic ray ionization at 10 g/cm -• atmospheric depth.

80

60

ION CHAMBER

I0 Mb

1958

0 I I I I I I I I I I I I I I 0 I 2 $ 4 5 6 7 8 9 I0 II 12 I$ 14 15

Px GV

Fig. 2. Differential response curve of an integrating ionization chamber at 10 g/cm 2 atmospheric depth. This response curve is characteristic of sunspot maximum.

IONIZATION CHAMBERS AT HIGH ALTITUDES, 1954-1965 3191

to the geomagnetic field if the Px are correct. is approximately true over a considerable range of rigidities for protons and e• particles [see

OI/OP = - • Dz(P, t)S,(P, x) (2) McDonald and Webber, 1959]) this becomes Z •

If the spectrums of the different Z compo- nents of the primaries are linearly related (this OI/OP = --D(P, t)$(P, x) (3)

6OO

5OO

4OO

300

20O

I00

e\e ION CHAMBER

I0 Mb

0 I I I I 0 1.0 2.0 3.0 4.0 5.0

Px GV Fig. 3. Latitude variation of 10 g/cm" total ionization at sunspot minimum and sunspot

maximum.

Fig. 4.

300

•. ION CHAMBER

• I0 Mb 200 ß

•oo

• • • 1958 •. o i i i i I

0 1.0 2_.0 $.0 • 4.0 5.0 6.0 ,

Px •v .

o•mbe• •t •0 •/om • •osp•e•Jo dept•.


1955

MINNEAPOLIS

--[McDONALD 8• WEBBER]

/

1958 MINNEAPOLIS

D58

[McDONALD 8• WEBBER]

o.ol I I I I I I IIJ I I I I I IIIJ I I I • • •11 0.1 1.0 t0 I00

P GV

Fig. 5. The total primary rigidity spectrum at sunspot minimum and sunspot maximum. Solid curve, direct particle measurements; points on line, solar cycle effect derived from ionization chamber.

where all linear factors are now included in the

S. (0I/OP will be plotted positive; e.g., see Fig- ure 2.) Because S is not expected to be time de- pendent but to depend only on the atmospheric transition phenomena, we can obtain from (3) the relative time variations in the spectrum D once the response curves OI/OP have been measured as a function of time.

If an ion chamber measurement is made at a

single latitude, the mean or effective rigidity of primary particles represented by the total ionization rate can be computed from the response curve by the relation'

- (4)

Since P depends on the spectral shape, it will vary, for example, with solar cycle modulation effects. The entire procedure above is exactly analogous to. the treatment of ground level neutron monitor data, where the details of the atmospheric transition process also are obscure and are never evaluated in detail. The balloon

ionization chambers, however, when flown at high latitude and high altitude, extend the response to considerably lower primary rigidities than are obtainable even with high latitude neutron monitors.

TABLE 1. Mean Rigidity of Response

Instrument

Mean Rigidity, bv

Sunspot Sunspot Min. Max.

(1954) (1958)

Ogo-A ion chamber (free space)

High latitude ion chamber (10 mb)

Minneapolis ion chamber (10 mb)

High latitude neutron monitor

(sea level)

2.1 2.9

2.5 3.6

3.2 3.8

Approx. 15 Approx. 15

Evaluation of ion chamber response curves. In the ideal case, the response curve of a high altitude ion chamber should be obtained from

a set of simultaneous readings distributed from the equator to the pole at the desired atmospheric depth (say, 10 mb pressure). An ap- proximation to this is provided by two sets of observations, one made in 1954 a•d one in 1958 by H. V. Neher and co-workers [Neher, 1956; Neher and Anderson, 1962; data for 1958 were obtained in a preprint from W. R. Webber]. Cutoff rigidities for the location of the flights


- F II FEB, 1958 _ - q- -

15

I0

5

I000 400 2:00 I00 40 2:0 I0

ATMOSPHERIC DEPTH (GM/CM •' )

Fig. 6. Normalized rate of ionization chamber ascending in the atmosphere. The histogram represents every pulse of the ionization chamber and therefore shows all the original data complete with the fluctuations normally seen on such a measurement. The value at 10 mb is estimated by smoothing a line through the histogram.

are based on the new calculated values of

Quenby and Wenk [1962]. The most. complete data are those for 1958,

and Figure I gives the results of this survey. The points indicate the values of the flights giving usable data at 10 mb, and results are given in absolute intensities using Neher's units of ion pairs/cm 8 sec arm. Figure 2 gives the differential response curve for 1958 where, as has been the custom in some papers, the curve has been normalized so that the counting rate at 15 Gv is given the arbitrary value of 100. The absolute rate at 15 Gv is 49.0 ion pairs/cm 8 sec arm. Figure 3 gives the intensity versus cutoff rigidities of 1954 and 1958. Values for 1954 are based on rather limited data., and except for a general idea this curve is probably not very accurate. Figure 4 shows the differential response curves for 1954 and 1958 based on the previous curves, where again the counting rate at 15 Gv has been given the value of 100.

Figure 5 has been constructed to check the accuracy of these curves. The heavy curves are primary rigidity spectrums of protons experi- mentally messured by McDonald and Webber [1959], extrapolated above 3 by. These results

were obtained using a Cerenkov scintillator a•d Geiger counters carried in balloons to an average altitude of 5 rob. Both the July 1955 and the July 1958 rigidity spectrum curves were made at Minneapolis. Using the previ- ously derived differential response curves and the known spectrum of one period, the spectrum of another period can be found, since

Note that the primary spectrum of 1955 has been given, but, as the previous intensity plots have shown, a significant change in the primary spectrum from 1954 to 1955 is not, expected to. have occurred. The light curve in Figure 5 is the spectrum derived using the above formula. The agreement is satisfactory.

Mean response rigidity. The mean rigidity P of response of an ion chamber flown at any latitude can now be computed from the differential response curves of Figures 2 and 4 using equation 4. The results for a number of eases are shown in Table 1. These include balloon ion

chambers at 10 g/em'- atmospheric depth (•10 mb) at Minneapolis (Px = 1.3 by) and at high latitude (Px = •ir cutoff for 10 g/em •.) and a high latitude neutron monitor. The neutron monitor calculations, though not outlined in detail, are analogous to those for the ion chamber.

In addition, the mean response of a free space ion chamber to primary cosmic rays (OgO-A ion chamber, computed by S. R..Kane) is given. The cutoff of this chamber is determined by the wall thickness and corresponds to about 12-Mev protons. The response in this ease is determined both theoretically and from laboratory measurements and will be described in more detail elsewhere (S. R. Kane, Ph.D. thesis).

It is important to note the very significant lowering of the mean rigidity of response achieved by flying an ion chamber on a balloon as compared to a sea level monitor. There is also a smaJ1 but significant gain in flying the balloon at high latitude. However, there is little additional response gained in flying the ion chamber in space as compared to a high latitude balloon. These comments apply, of course, to the 'normal' cosmic ray spectrum. The situation may be very different for solar cosmic ray particles or other radiation of predomi- nately low energy.

319zi CALLENDER, MANZANO, AND WINCKLER

4o

5o

2o

• I • I I I I 54 55 56 57 58 59 60

YEAR

I I I 61 62 65 64

Fig. 7. Flight activity profile for the high altitude monitoring program at Minneapolis. Each bar in the histogram represents the number of usable flights in the six-month period.

25

2o

w

z •o

Oo Io 20 5o 40 50 60 KP VALUE

mi_.J-

Fig. 8. The number of flights for various K• values. For this analysis, activity containing auroral x rays or solar protons was not included. The average K• value is about 2-3.


TABLE 2. Monthly Averages of 10-rob Total Ionization and Correlated Neutron Monitor Values

Year and Mpls. Ions, No. Fits. Chur. Ions, Month Monthly Average in Average Monthly Average

Corr. Ottawa Corr. Climax

Neutrons, Neutrons, Monthly Average Monthly Average

•- 64 •- 128

1954

8 30.9* 2

1955

8 35.4•

1956

7 28.7'[' 8 28.6t

10 29.0 1

1957

4 20.0 1 5 19.5 1 6 18.5 1 7 17.6 3

8 19.4 6 17.9J' 9 18.3 5

10 20.0 1 11 19.6 1 12 17.4 1

1958

I 16.4 2

2 16.2 3

3 15.9 2 4 16.7 2 5 16.5 1 6 17.9 4 7 16.3 9 8 17.6 4 9 16.8 2

10 17.8 4 11 16.9 3 12 15.9 1

1959

i 16.7 2 2 16.5 1 3 16.2 3 4 16.5 1

5 16.4 3 6 18.7 1 7 15.8 2 8 16.0 1 9 16.4 1

10 16.0 2 11 18.4 2 12 16.4 3

1960

i 17.3 4

2 17.6 3 3 17.5 5 4 17.2 12 5 17.4 6 6 17.1 7 17.6 7 17.4 7 18.3

6637

6289 6507

5584 5593 5658 5693 5651 5632 5349 5440 5586 5730 5247 5276 5402 5350 5484 5519

5467 5407

5351 5327

5284 5307 5192 5207 5389 5321 5280 5371 5566 5658 5379 5426 5437 5379 5482 5510 5452 5548

5630 5656 5473 5371

5612 5478 5370 5405 5535 5526 5777 5837 5392 5370 5507 5597 5106 5086 5231 5238 5270 5223 5548 5654 5601 5659 5470 5546

5422 5514 5533 5615. 5562 5621 5437 5486 5391 5457 5459 5538 5544 5605


TABLE 2. (Continued)

Year and Mpls. Ions, No. Flts. Chur. Ions, Month Monthly Average in Average Monthly Average

Corr. Ottawa

Neutrons, Monthly Average

+ 64

Corr. Climax.

Neutrons, Monthly Average

+ 128

8

9 10 11 12

1961

1

2

3 4

5 6

7 8 9

10 11 12

1962

1

2

3 4

5 6

7 8

9 10 11

12

1963

1 2

4

6

7 8 9

10 11

12

1964

1

4

5 7 9

12

18.5 18.6 18.4

19.0 18.4

19.8 20.4 20.3 21.0 21.4

20 4

20 0 20.2 22 6

21.4 23 1 23 6

23.8 22.7 24.0 23.4

23.7 24.1

23.9 23.9 23.3

25.3

24.9

25.8 24.8 24.4 25.6

26.0 25.4 25.6 25.0 26.5

25.9

24.9

27.2

28.4

28.7 30.0

18.8 19.2 18.5

23.3

21.2 21.1

26.1

26.3 24.7

27.6

30.1•

5629 5621

5499 5529 5667

5897 5910 5876 5936 5923 5854

5696 5726 5891

5907 6011 5942

6119

6149 5994 6089 6082 6006 5931 5928 5933 6045 6139

6324

6287 6291

6199 6186 6141

6120 6065 6361

6349

5697 5736 5600 5522

5712

5886 5922 5960 5971 6031 6001 5804 5899 6022 6041 6165 6153

6186 6206 6210 6097 6165

6202 6183

6139 6147

6219 6215

6377 6311

6290 6387 6309 6291 6307 6238 6469

6458

6529 6493 6511

6511 6582

* Bismarck.

t Thule. •: Andenes, Norway.


The solar cycle variation. In this section we compare the results of an extensive series of 10-mb ion chamber observations at Min-

neapolis with neutron monitor observations and a limited number of high latitude 10-mb balloon observations. The University of Minnesota balloon ion chamber program began in 1956 when the solar cycle decrease had already be- gun. Accordingly, we have used certain ion chamber measurements of It. V. Neher of Cal.

Tech. made at Bismarck, N.D. (similar to Min- neapolis), and Thule, Greenland (similar to Churchill), during the period 1954-1957. The Neher results are expressed in ion pairs/cm 8 sec arm air, and this scale is related to. the Minnesota scale of normalized pulses/sec X 108 by the relation

Norm. pulses/sec X 103

= ion pairs/cm 3 sec arm X 0.0683

This relation is determined from balloon

flights intercompared at the Pfotzer maximum and a laboratory comparison using Co © 7 rays. The relation is therefore valid for minimum-

ionizing particles. The rate of the ion chamber is taken as the

balloon first reaches an altitude of 10 mb

(•10 g/cm 2) so that the measurement is made near the launch site. All flights for which ex- cess radiation such as solar protons or auroral X rays were detected were excluded, and thus the 10 mb rates should be representative of the primary cosmic ray level. The rates of individual flights were averaged in one-month in- tervals and in a particular month; the number used varied from zero to twelve. An ionization-

depth curve at Minneapolis typical of those from which the 10-mb rates were read is shown

in Figure 6. 'This curve shows the rate computed for each ion chamber pulse and therefore gives the maximum time resolution with- out smoothing.

Figure 7 is a histogram of the flight distri- bution in time at Minneapolis (including two Neher flights at Bismarck). The number of flights having a given Kp index is given in Fig- ure 8. The Kp is the three-hour value during which the 10-rob ion chamber reading is taken. The median of these readings is about 2 + .

Table 2 summarizes the above data. The

monthly averages of the 250 Minneapolis flights and of the groups of summer flights at Churchill are plotted in Figure 9 for comparison in the usual way with the Zurich sunspot

1.2

1.0 \ •:

.8 •'•-•'-'" MINNEAPOLIS .... z o

m SON SPOT • •

o • .

o .4 •/ • - INNEAPOLIS a 0r ß z

o • 200

- ' ' • -- •00

N

o

_

•.2 I I I I I I I I I I I

54 55 • 57 58 59 60 61 6Z 6• 64

YEAR

•ig. 9. •mcfio•M ch•ge of •he mo•[h]y •vemge of 10 •/cm = •o[•l •o•[•o• • •e•po- 1is •d • h•h l•[•ude (Churchill) e•p•essed • • fmcfio• of [he •ovembe• 1958 •vemge. Al•o •ho• •e •ufich •u•po• •umbe• (•c•le i•c•e•i• do••d).


numbers. Figure 9 and Table 2 also include the Bismarck (1954) and Thule (1955, 1956, 1957) flights of It. V. Neher and the Andenes, Norway (1964), flights under the University of Minnesota 'Pocibo' balloon program.

We see in Figure 9 that at sunspot minimum

the mean response difference between Min- neapolis and Churchill (3.2 bv--• 2.5 bv) pro- duces a 25% effect. At sunspot maximum, the response is very little different (3.8 by -• 3.6 by) and the latitude effect disappears at the two locations. Furthermore, the decrease is

MINNEAPOLIS IONIZATION CHAMBER IO(Mb) 14 16 18 20 22 24 26 28 $0 I I I I I [ ] ]__.(• • [ (•

SEPT.• •;•..© •'

JAN. 62----• • •' 0CT. 56 / e• •e AUG. 54

JAN. 61 • ß ••e

JAN. 59 • e• JAN. 60 • •/ •

ß .ov.

JAN. 58 • • '••• OCT. 57 ' • SEPT. 57

26600 6400

- 6200

- 6000

- 5800

- 5600

- 5400

- 5200

5000

Fig. 10. Correlation plot of 10-g/cm 2 total ionization at Minneapolis and the Climax neutron monitor. Generally a fairly smooth relationship exists except near sunspot maximum, where the points deviate significantly from a smooth curve.

CORRELATED CLIMAX NEUTRON MONITOR

5000 5200 5400 5600 5800 6000 6200 6400 6600

] ] , [ [ [ • ] x/__[6600 JAN. 65 .•.,,,,,,,,,•, / -J6400

JAN. 62 -- e.-•// x

JAN. 61• e• - 6000 • .'•/•

• - •00 APR. 57_ .

xx •i

JAN. 58•••• -5400 JA• 60

- 5200 ß x

5OOO

Fig. 11. Comparison of the Climax and Ottawa neutron monitors. Each point is a monthly average composed of individual bi-hourly readings chosen to be simultaneous (i.e. correlated) with the balloon 10-g/cm 2 measurement. The scatter of points is low enough that we can conclude that deviations shown in Figure 9 are significant.


,oooo , I ' I ' I ' I ' [ ' ß4-,5

"'_o 4-• x I000 ß ' .• - • .4-28

a. I00 ß

ß

g

• 240 N

_

22(

1660 1700 1740 1780 1820 1860 1900

ADJUSTED HOURLY RATE OF THE NEUTRON MONITOR AT DEEP RIVER

Fig. 12. Correlation of an ion chamber in deep space on Pioneer 5 with the Deep River neutron monitor.

rather sharp and the recovery slow. For 3.8-bv mean energy particles characteristic of Figure 9, the intensity minimum lags the sunspot maximum by about 18 months, in agreement with the findings of H. V. Neher [Neher and Ander- so% 1962] for balloon ion chambers. The many fluctuations in Figure 9 are real and represent various Forbush and other short-term varia-

tions, averaged over the flights in each month. Finally, we compare the solar cycle change

for two rigidity ranges. In Figure 10 each point (x, y) is located by plotting the monthly mean 10-g/cm" ion rate at Minneapolis against the 'correlated' monthly mean of the Climax neutron monito. r. The 'correlated' monthly mean is the average of those bi-hourly neutron monitor values taken for the bi-hourly interval during which the balloon 10-g/cm •' ion rate was determined. The number of neutron moni-

tor readings per month is thus the same as the number of balloon measurements as they occurred in a practical sense simultaneously.

As a check on random scatter in the neutron

monitor values, we show in Figure 11 the

'correlated' Climax •' monitor monthly means plotted against the 'correlated' Ottawa s monitor monthly means. An excellent smooth functional relationship exists, with a minimum of scatter of individual points.

The following facts about Figure 9 would thus seem significant: for both decreasing and increasing cosmic ray rates, the relationship between the 3-bv range of mean rigidity and the 15-bv range is the same. This also appears to be true for shorter period fluctuations, with the exception of the fall of 1957, when a real hysteresis loop develops. At this time, the 3-by and 15-bv relation moves off the curve, indi- cating a spectral excursion different from most of the cycle. This was a period of a large decrease preceding sunspot maximum. For data for all the solar cycle other than this period, a smooth functional relationship exists.

In an independent analysis of a more re- stricted sample of the same basic data, Nerur- kar and Webber [1964] found that, the short- term or Forbush changes and the long-term or solar cycle changes followed a single correlation curve. With the exception of the period in the

' ,•OLAR PROTON,•

START

62ol- • ? /

• '•,,• ....

5130

L DEEP RIVER 3450 3500 •550

Fig. 13. Correlation of the Mariner 2 ionization chamber in deep space with the Deep River neutron monitor.

"Data from J. A. Simpson, Univß of Chicago, through World Data Center A for cosmic rays.

8Data from D.C. Rose, National Research Council, Ottawa, through World Data Center A for cosmic rays.


fall of 1957, the present analysis is in general agreement.

Space gradients during the solar cycle. Two analyses have been completed in which data from ion chambers on deep space probes have been compared with data from similar instruments flown simultaneously on high altitude balloons or from neutron monitors on the

ground. In a previous paper [Arnoldy et al., 1964],

a comparison was made for March and April 1960, a period of high solar activity, between the ion chamber on Pioneer 5, balloon ion chambers, and a sea level high latitude neutron monitor. Although the record was dis- turbed by outbursts of solar protons, a clear correlation between primary cosmic ray changes in deep space and on earth was obtained. No evidence for a systematic intensity change with distance from earth of up to 0.1 AU was detected. The correlation persisted during a Forbush decrease of about 4% observed on the neutron monitor. The same correlation was ob-

served between Pioneer 5 and balloon ion cham-

bers and between balloon ion chambers and sea

level neutron monitors (see Figure 12). We see in Table I that at solar maximum the mean

response rigidities are 2.9 bv, 3.8 bv, and 15 bv, respectively, for these three types of measurements.

A similar correlation was attempted for the flight of Mariner 2 to Venus (see Figure 13). The Mariner data were obtained from Neher

and Anderson [1964a]. During this period a small solar proton outburst and a Forbush event of about 3% in sea level neutrons occurred. For this event, an apparent correlation exists. For the rest of the period, the correlation dissolves in the noise level. No definite

evidence for a space gradient can be found (Figure 13).

In the Pioneer 5, but particularly the Mariner 2, correlations in Figures 12 and 13, the noise is frequently too large [Arnoldy et al., 1964] to attribute to counting statistics or other random errors. The fluctuations must represent primary cosmic ray intensity variations with a spatial scale size smaller than the earth-probe separation. The different response energy of a free space ion chamber and a neutron monitor may also be important if the noise is not co- herent over • very wide range of energies.

Using the same data, Neher and Anderson [1964b] have argued for the existence of a space gradient. We believe, however, that the regression type diagram used here has certain advantages in presenting the relationship of the two instruments and that this diagram shows a correlation between the two instruments that is

independent of distance from from earth within the uncertainties which seem to be of random character.

Acknowledgment. We greatly appreciate the cooperation of D.C. Rose of the National t•e- search Council of Canada for forwarding the Ottawa neutron data, and the evaluation of the mean rigidity values by Sharad R. Kane. We have benefited from discussions with Professor W. R. Webber on the neutron monitor data. Amalia

Manzano helped with the computations. This research was supported by the Atmospheric

Sciences Division of the National Science Founda-

tion under grant NSF-G24461.

l•EFERENCES

Arnoldy, R. L., R. A. Hoffman, and J. R. Winck- ler, Observations of the Van Allen radiation regions during August and September, 1959, part 4, The outer-zone electrons, J. Geophys. Res., 67, 2595-2612, 1962.

Arnoldy, R. L., J. R. Winckler, and IL A. Hoff- man, Comparison of the total cosmic radiation in deep space and at the earth during the March- April 1960 events, J. Geophys. Res., 69, 1679- 1690, 1964 (and Proc. Intern. Con•. Cosmic Rays, Jaipur, December 1963).

Holmann, D., and J. R. Winckler, Simultaneous balloon observations at Fort Churchill and Minneapolis during the solar cosmic ray events of July 1961, J. Geophys. Res., 68, 20'67-2098, 1963.

Masley, A. J., T. C. May, and J. I•. Winckler, Analysis of balloon observations during the April 1960 solar cosmic-ray events, J. Geophys. R es., 67, 3243-3268, 1962.

McDonald, F. B., and W. R. Webber, Proton com- ponent of primary cosmic radiation, Phys. Rev., 115, 194-205, 1959.

Neher, H. V., Low energy primary cosmic-ray particles in 1954, Phys. R e v., 103, 228-236, 1956.

Neher, H. V., and H. R. Anderson, Cosmic rays at balloon altitudes and the solar cycle, J. Geophys. Res., 67, 1309-1315, 1962.

Neher, H. V., and H. R. Anderson, Cosmic ray intensity at Thule, Greenland, during 1962 and 1963 and a comparison with data from Mariner 2, J. Geophys. Res., 69, 807-814, 1964a.

Neher, H. V., and H. R. Anderson, Change of cosmic-ray intensity with distance from the sun, J. Geophys. Res., 69, 1911-1913, 1964b.

Nerurkar, N., and W. R. Webber, Observations of primary cosmic ray variations using ion cham-


bers and Geiger counters, 1, A comparison of getic electrons and the geomagnetic field, J. long-term and Forbush variations, J. Geophys. Res. NBS, Radio Propagation, 66D, 127-143, Res., 69, 815-830, 1964. 1902.

Quenby, J. J., and G. J. Wenk, Cosmic ray Winckler, J. R., Energetic X-ray bursts from solar threshold rigidities and the earth's magnetic flares, AAS-NASA Symp. on the Physics of field, Phil. Mag., 7, 1457-1485, 1982. Solar Flares, October 28-30, 1963; NASA Rept.

Winckler, J. R., Balloon study of high altitude SP-50, 1963. radiations during the International Geophysical Year, J. Geophys. Res., 65, 1331-1359, 1960.

Winckler, J. R., Atmospheric phenomena, ener- (Manuscript received March 22, 1965.)

Documents

The Response of High Altitude lonization Chambers during