Pergamon Adv. Space Res. Vol. 20, No. 2, pp. 187-190, 1997

8 1997 COSPAR. Published by Elsevier Science Ltd. All rights reserved Printed in Great Britain

0273-1177/97 $17.00 + 0.00

PII: SO273-1177(97)00532-2

A COMPARISON OF SOLAR EUV FLUX FROM LANGMUIR PROBE PHOTOELECTRON MEASUREMENTS ON THE PIONEER VENUS ORBITER WITH OTHER SOLAR ACTIVITY INDICATORS

K. K. Mahaja.n;* W. R. Hoeav,** W. D. Pesnell,*** L. H. Brace? and N. K. Sethi*

* National Physical Laboratory, New Delhi 110 012, India ** NASA Goddard Space Flight Center, Greenbelt, MD 20771, U.S.A. *** Nomad Research Inc., 2804 Nomad Court, Bowie, MD 20716, U.S.A. t Space Physics Research Laboratov, University of Michigan, Ann Arbor, MI 48109, U.S.A.

ABSTRACT

The electron temperature probe acted as a photo-diode on the Pioneer Venus Orbiter (PVO) and measured the integrated solar EUV flux (I,) over a 13 year period from January 1979 to December 1991, thus covering the declining phase of solar cyle 21 and the rising phase of solar cycle 22. Gross features in the solar activity variations of this flux during the 13 year period have earlier been studied by Brace et al. (1988) and Hoegy et al. (1993). In this paper, we study the fine features by translating the observed Ire to the solar longitude of Earth (to be called as EI,,) and comparing it with other solar activity indicators like F,,,, Lyman alpha and the solar magnetic field. We find that while the daily values of EIpe are highly correlated with F10.7 (correlation coeffi- cient 0.87), there is a large scatter in EIpe for any value of this earth based index. Comparison of EIpe with SME and UARS-SOLSTICE Lyman-alpha measurements taken during the same period indicates that EIpe tracks Lyman-alpha quite faithfully. Similar comparison with the solar magnetic field (I&) shows that EIpe cor- relates better with B, than with F10.7. 0 1997 COSPAR. Published by Elsevier Science Ltd.

INTRODUCTION

It is now well known that solar EUV irradiances are responsible for creating the terrestrial ionosphere and for heating the thermosphere. An accurate knowledge of EUV fluxes is, therefore, necessary for aeronomical studies of the upper atmosphere. Fig. 1 adopted from Tobiska (1996) shows the various satellite measurements of solar EUV from 1960 onwards. There are only a few measurements on daily basis. The largest set of meas- urements is due to the AE-E satellite which covered the period from mid 1977 to the end of 1980. The period 1980 onwards has been named as the “EUV hole” by Donnelly (1987) , because the next full-spectrum daily measurements will not occur until the late 1990’s.

The Langmuir probe, on the spacecraft Pioneer Venus Orbiter, acted as a photodiode during the “EUV hole” period. The probe measured integrated solar EUV flux over the 13 year period from January 1979 to De- cember 1991, thus covering the declining phase of solar cycle 21 and the rising phase of solar cycle 22. Gross features of this flux have been studied earlier by Brace et al. (1988) and Hoegy et al. (1993). In this paper, we study some fine features by comparing the PVO observed flux with other solar activity indices.

LANGMUIR PROBE AS A PHOTODIODE

The Pioneer Venus Orbiter (PVO) had a rhenium-coated Langmuir probe on it to measure the electron density Ne and electron temperature Te in the ionosphere of Venus (Krehbiel et al., 1980). Results of these ionospheric measurements have been reported in several publications (see review by Brace and Kliore, 1991 and references therein). However, when the spacecraft was outside the Venus ionosphere, this probe acted as a

1x7

188 K. K. Mahajan et al.

phot~iode. The solar EUV radiations, roughly between 10 and 150 nm, on striking the rhenium surface, ejected photoelec~ons which produced probe current at the level of 5-20 nA. This current is propo~ional to the total solar EUV flux for wavelengths which create the ionosphere and heat the thermosphere. Elphic et al. (1984) , were the frst to demonstrate that the photoelectron current, Ipe, measured by the probe could be. used as au index of solar EUV flux. They observed solar-rotation related cyclic changes in Ipe coincident with simi- lar cyclic changes in the Venus ionosphere.

Brace et al. (1988) obtained the following relationship between the total EUV flux and Ipe and gave the name Venus EUV (Veuv) to this index:

V EUv = 1.53~10” I,, photons cm3 s”‘; Ipe is in units of 10m9 A.

The amplitude of the photoelectron current is the product of the photoelectron yield and the solar flux summed over EUV wavelengths between 10 nm and 150 mu. About 55% of the Ipe is from Lag-aIpha and 30% from the 30 - 110 nm continium, at both solar minimum and maximum (MahaJan et al., 1990; Hoegy et al., 1993). The remainder is from strong ionizing lines such as He II, He I, C III, etc. This percentage too does not change significantly with solar activity. Hoegy et al. (1993) have estimated that about 10% contribution to Ipe is from coronal regions with wavelengths less than 50 mn, the reminder (90%) is from chromospheric regions with wavelengths more than 50 nm during solar minimum. During solar rn~irn~ these percentages are 14% and 86%.

The Early Studies of Ipe

The daily average values of the photocurrent obtained between 1979 and 1987 were fast ana- lysed and published by Brace et al. (1988) who found that I e exhibited variations related to the solar cycle and solar rotation, as well as to a major 7-month periodic@. & ey compared Ipe measurements for the year 1981 to 1987 with F10.7 and found a good agreement in terms of solar cycle and solar rotation effects, although amplitudes were much larger in F10.7. At times when both Venus and Earth were viewing the same solar disc (for example for the interval near inferior conjunction), the solar rotation components were found to be in phase. These signatures, as expected, were less similar near superior conjunction when earth and Venus viewed opposite sides of the sun. A comp~son by Brace et al. (1988) of Ipe with Lag-alpha flux measured by the SME satellite, showed a better relationship between these two parameters than with F10.7. These Ipe, observa- tions were used by Mahajan et al. (1990) to examine the response of Venus exospheric temperature to changes in solar activity, primarily those related to solar rotation. It was found that the dayside exospheric temperature quite faithfully tracked variations in Ipe. The exospheric temperature was also found to be better correlated with Ipe than with F10.7.

Although the early studies of I were basically related to the ionosphere of Venus, Hpegy and Mahajan (1992) translated I Earth (E,,,). They examine a

to the solar ongitude of Earth (EIpe) and converted these into EUV flux at P the behaviour of ionospheric parameters foE, foF1, foF2 at midlatitude stations

and compared their relationship with E,, and F10.7. They found foFl and foF2 to be better correlated with E,, than with F10.7. However, foE was found to be better correlated with F10.7, because F10.7 is also a proxy for sofi’X-rays which are an important ionizing source in the E-region. Hoegy and Mahajan also provided a Table of the daily values of the EUV flux for the period February 12, 1979 through most of 1991.

Features of the EUV Flux During Solar Cycles 21 and 22

The daily values of Ipe, translated to solar longitude of the Earth by using the ~sfo~ of Hedin et al. (1983) are shown in the left panel of Fig.2 for the period Jan. 1979 to Dec. 1991. This 13 year period covers the peak and later stages of solar cycle 2 1, the intervening minimum and the peak of solar cycle 22. Fig. 2 also contains plots of Lyman alpha, solar magnetic field and Fl0.7. The Lyman alpha values are taken from the SME (Rottman (1985)) and UARS - SOLSTICE (Rottman et al. (1993)) instruments. To bring agreement between the SME and UARS me~uremen~, the SME values have been scaled up by 25% (T. N. Woods, pri- vate comm~i~tion, 1996). The solar magnetic field data is taken from the me~urements at National Solar Observatory, Kitt Peak {e.g. Harvey, 1992). F10.7 is the daily average value of solar radio flux at 10.7 cm re- corded at Ottawa. It can be noted that all the four parameters show large variances on a day-to-day basis, with

EUV Flux fmm Langmuir Probe 189

he largest v&ame occurring during solar maximum and smallest during solar minimum. It is also to be noted that solar magnetic field and F10.7 show larger variance while EIpe shows the minimum variance. Between 1985 and 1987 when EIpe, solar magnetic field and F10.7 had become nearly constant and reached their mini- mum values, the SME measured Lyman alpha was still showing a decreasing trend. This is a very surprising re- sult and a possible cause could be a change in the sensitivity of the SME instrument. We are tempted to believe this to be the case because all the other three measurements (viz. EIl,e, solar magnetic field and F10.7) showed similar trends.

Another alit feature to be noted is that from the beg&g of 1989 to the end of 1991, while the Fl0.7 had reached its maximum value and had become nearly constant, both EIpe and magnetic field were still increasing. A double maxima, however is seen in all the three parameters dnrmg the maximum of SC. 22.. There was a large observational gap in the Lyman alpha measurements during the rising part of solar cycle 22. However, the second maximum seen in the other three parameters is also seen in Lyman alpha. These features can be better identified in the 8 1 -day average plots shown in the right Panel of Figure 2. Here one can also note the longer periodicities (of several months) in these parameters. Spectral analysis of Ipe by Hoegy and Wolff (1989) showed that the dominant periods are the 7-month period at 2 16 days and the solar rotaion at 28 days. These periodicities can also be seen to some extent in Lyman-alpha, magnetic field and F10.7.

t979 1991 Ir--------- PVO F

4-- Reference spectra Proxy models ---)

Fig. 1 Schematic represen~tion of satellite measurements of solar EUV h-radiances i%om 1960 onwards. Langmuir probe measurements of total EUV flux happen to be during the “EUV hole” period (After Tobiska, 1996).

1960 1970 VEAR 1980 1990

300

250 200 150 1w 50 79 81 83 85 87 89 91~7$s81

Fig. 2 A comparison of daily values of EI, with Lyman alpha, Solar magnetic fi;ld and F10.7. (1eR Panel) Units for EIY : 10 A; for Lyman alpha : 10” photon~cm ; for solar mag- netic field : Gauss; for F10.7 : 10” W/m2/Hz. The right panel shows time series of 81 day aver- aged values.

Fig. 3 A scatter plot of daily values of EIp, against Lyman alpha, Solar magnetic field and F10.7. A linear reiationship can be seen with all the three parameters with highest correlation co- efficient and minimum scatter with Lyman alpha. Units as in Fig. 2.

190 K. K. Mahajm et al.

Correlation of Ipe with Lyman-alpha, Solar Magnetic Field and F10.7

Figs. 3 shows plots of daily values of EI &

e against daily values of Lyman-alpha, solar magnetic field and F10.7. We find a strong correlation between pe and L~~-~ph~ with a correlation coefficient of 0.93. This is not a surprising result because EIpe is basically an index of chromospheric emissions and 55% of ~on~ibution to EI compared to the E Y

e is from Lyman-alpha. The EIpe - solar magnetic field plot shows a larger scatter when pe - Lyman alpha plot, though the relationship is nearly linear. The correlation coefftcient is

0.91. In the plot between daily values of EIpe and F10.7, the scatter is considerably larger as compared to the other two plots, However, the correlation coefficient, inspite of the large scatter, is 0.87. This scatter is mainly due to higher excursion of F10.7 during solar rotation, especially during solar maximum. These excursions are mainly due to short-lived bursts from coronal emissions (Donelly, 1993, private communication). It has already been established that the 27-day variations of F10.7 and chromospheric emissions are different (Lean, 1987; Barth et al., 1990; Hoegy and Mahajan, 1992).

CONCLUSIONS

The integrated EUV flux measured by the Langmuir probe on the Spacecraft Pioneer Venus Orbiter during the 13-year period Jan. 1979 to Dec. 1991 has shown the well known solar cycle and solar rota- tion effects. There is a significant variance in the daily flux of all the solar parameters which makes it difftcult to compare daily Earth based observations with the Venus observations that have been translated in solar longi- tude. In addition there are differences in solar flux at different solar longitudes that could , for example, be seen at Venus and not at Earth and vice versa. The largest variance occurs during solar maximum, as also seen in solar Lyman-alpha, solar magnetic field and F10.7. In addition, a periodic@ of 6-7 month exists in the EUV flux, which is very apparent during solar minimum.

ACKNOWLEDGEMENTS

We thank Joe Johnson for his diligent pr~uction of the Ipe data, L.R. Canfield at NIST for his help in calibrating our photodiode surfaces; G.L. Rottman and T. Woods of HAOMCAR, University of Colo- rado for the SME and UARS Lyman-alpha data. The NSOZKitt Peak magnetic data are produced co-operatively by NSF/ NOAO, NASA/GSFC and NOAA&EL. We are thankful to Amar Singh, S. Ghosh and R. K. Choud- hary for their help in preparing this manuscript.

REFERENCES

Barth, C.A., W.K. Tobiska, G.J. RottmanandO.R. White, Geophys. Rex Lett., 17,571 (1990). Brace, L.H. , W.R. Hoegy and R.F. Theis, J. Geuphys. Res., 93,7282 (1988). Donnelly, R.F., in Solar Radiative Output Variations ed. P. Foukal, Proceedings workshop, pp 139-142,

NCAR, Boulder Colorado (1987). Brace, L.H. and A.J. Kliore,, Space&i. Rev. 55,81 (1991). Elphic, R.C., L.H. Brace, R.F. Theis and C.T. Russell, Geophys. Res. Lett., 11, 124 (1984). Harvey, K.L., Proceedings of the Workshop on the solar electromagnetic radiation study for solar

cycle 22 , ed. R.F. Donnelly, pp 113-129, Env~o~en~l Research Laboratory, NOAA, Boulder, USA (1992). Hedin, A.E. , H.B. Niemann, W.T. Kasprzak and A. Seiff, J. Geophys. Res., 88,73 (1983). Hoegy, W.R. and CL. Wolff, J. Geophys. Res., 94,8663 (1989). Hoegy, W.R and K.K. Mahajau, J. Geophys. Res., 97,10525 (1992). Hoegy, W.R , W.D. Pesnell, T.N. Woods and G.J. Rottman, Geophys. Res. Lett. 20, 1335 (1993). Krebbiel, J.P., L.H. Brace, R.F. Theis, J.R. Cutler, W.H. Pinkus and R.B. Kaplan, IEEE, Trans. Geosci. Remote Sens., GE- 18,49 (1980).

Lean, J., J. Geophys. Res., 92,839 (1987). Mahajan, K.K. , W.T. Kasprzak, L.H. Brace, H.B. Niemann and W.R. Hoegy, J. Geophys. Res., 95,109 1

(1990). Rottman, G.J., in A~ospheric Ozone, Proceedings of Qua~ennial Ozone Symposium, edited by C. S. Zerefos,

and A. Ghazi, p. 656, D. Riedel, H&ham, Mass (1985). Rottman, G. J. ,T. N. Woods and T. P. Sparn, J. Geopkys. Res., 98,10667 (1993). Tobiska , W.K., Adv. Space Res, 18, #3,3 (1996). Wolff, C.L. and W.R. Hoegy, Sol. Phys., 123,7 (1984).