On the Major Solar Flare Activity in Solar Cycles 19,20 & 21 (1955-79)

Indian Journal of Radio & Space Physics Vol. 12. August 1983. pp. 124-128

BADRUDDIN, R S YADAV & N R YADAV Department of Physics. Aligarh Muslim University. Aligarh 202001

Received 22 July 1982; re nised received 22 March 1983

Utilizing the major solar flare data for the period 1955-79, which includes solar cycles 19, 20 and 21. a comprehensive analysis of these flares is presented. In this analysis, study is made of their distribution around the sun. both in latitude and longitude, the time latitude occurrence in solar cycles 19, 20 and 21 and their nature and frequency in northern and southern hemispheres in these solar cycles. and the change of occurrence of these flares in the course of years for both halves of the solar disk separately and also for the disk as a whole. Their relationship with sunspot numbers in different solar cycles and their occurrence frequency with increasing comprehensive flare index in general and for cycles 19 and 20 in particular (observations for cycle 21 are not complete) have also been studied.

Introduction longitudinal component of photospheric magnetic Long years of observations of various manifesta- field 19. tions of solar activity indicate that their occurrence in The purpose of the present paper is to study in detail northern and southern hemispheres is not uniform I M. the nature and frequency of major solar flares during In the first quarter of this century, using the sunspot the different periods of solar activity cycle and in data for the period 1874-1913, the latitudinal different zones of solar surface both in latitude and distribution of sunspots was studied in detail" The longitude, using the maximum available data for the heliographic latitude distribution of optical solar flares solar cycles 19, 20 and 21. The peculiarity about the and flares producing magnetic storms", polar cap solar cycles 19 and 21 is that the solar activity was absorption (PCA) producing flares", the coronal line unusually high 17 in solar cycle 19 when compared with intensity" Iotas been studied for varying periods that for any other cycle during the last 350 years and subsequently. The heliographic longitude distribution the solar cycle 21 is the second highest cycle+? during has also been studied, in the past, for flares which the last 280 years. produce energetic solar particles I o. t 1, yellow lines 12 and optical solar flares 13. 2 Analysis Analyses were performed to study the north-south We have carried out an extensive statistical analysis asymmetries in white light flares, major solar flares, of major solar flares fulfilling the critena established complex and non-complex sunspots, and sunspot by Dodson and Hedeman 21' 23 for the period 1955-79, areas". PCA flares", optical flares", photospheric which includes solar cycles 19, 20 and 21. Dodson and magnetic field 14, faculae", prominences 15 and coronal Hedernan P r+' have devised a comprehensive flare line 530.3 nm (Ref. 6). The long term variation in the index (CFI) which classifies a flare as major flare energetic particle emissivity of the sun was examined 16 whenever any of the following criteria is satisfied: short by the use of PCA, solar proton flux and geomagnetic wa ve fade (or sudden ionospheric disturbance) data for 1941-73 period. importance ~ 3; H~ flare importance ~ 3; 10cm flux ~ 500 With limited data of large flares of importance ~ 2 for x 10 -22 Wm -2 Hz -I; type II bursts; type IV radio the solar cycle 19, the relationship between sunspot emission of duration> 10 min. Each individual item is number and frequency occurrence of these flares has scaled from I to 3 to compile the CFI. This index has been investigated recently. Its occurrence frequency the advantage of not excluding small optical events with increasing flare importance has also been with unusually strong ionizing and radio frequency discussed 1 7 . emissions. The relationship between the solar magnetic field Using these major flare data from 1955-79 we have and solar activity has also been extensively studied in studied their latitudinal and longitudinal distribution the past. The negative polarity fields on the solar over the solar disk, the long term variation, their surface are closely related to the development of new occurrence in northern and southern hemispheres of solar activity I 8 and the solar tlare energy output bears the sun and the behaviour of solar activity in the solar a linear relationship with the rate of change of flux of cycle with reference to solar latitude. We have also

124 BADRUDDIN et al.: MAJOR FLARES OF SOLAR CYCLES 19,20 & 21 studied the relationship between the sunspot number south asymmetry is nearly same in all the latitude and the frequency occurrence of major flares. After intervals (of 10) up to 40° beyond which the arbitrarily dividing the major flare events into six observation of major flares is negligible. classes according to increasing Cf'l to establish a The solar longitudinal distribution of major solar criterion for the importance of energy output, the flares is shown in Fig. 2. The distribution is occurrence frequency of these classes in different solar approximately similar in both the hemispheres, east cycles is also studied. and west. Out of the total events (1936) observed during 1955-79, ~49% were observed to occur in 3 Results and Discussion eastern hemisphere and ~ 51% in the western 3.1 Heliographic Latitude and Longitude Di~tribution hemisphere. This heliographic longitudinal distri- We have studied the heliographic distribution of bution of major solar flares may be represented by a major solar flares satisfying the criteria laid down by Gaussian function of the type Dodson and Hedeman using the data for the periods 2 j(¢)d¢ = 7.. exp [ -{J (¢ - fi2(¢ - ¢)2]d¢ 1955-79. The heliolatitudinal distribution of these major flares is shown in Fig. I. The flare location has Here, been summed up over 10 heliolatitudinal interval. The 7.=6.249, {1=0.007, ¢= -0.576, d¢= 10 peaks in the frequency distribution, shown in Fig. I. are located in the interval 10-20 in both the This generated function over the visible hemisphere hemispheres. This clustering in the 10-20' latitude is of the sun is shown in Fig. 2 by dotted line. In this more significant in the northern hemisphere. Out of function R¢)d¢ is the frequency of the flares in the the total events (1936) observed during 1955-79. ::::63: ':";; longitude interval d¢ and IX, {J are constants. The were observed to occur in the northern hemisphere and shallow peak in the middle of the distribution curve ~37% in the southern hemisphere. Further, nearly indicates that the frequency of major solar flares 34'\, of the total were observed in only 10-20C'N latitude around 00 longitude is not much elevated from other interval. Almost similar results were found for optical higher longitude frequencies. Such curve provides a solar flares of all importances and flares associated general idea about the longitudinal distribution of with magnetic storms". As shown in Table I, the north- flares around the solar disk. The shape of curve and the

Table l-North-South Asymmetry 32 in Latitude Intervals of 10'

N-S Latitude A=--- 28 interval N+S deg

0-10 0.291 11-20 0.250 21-30 0.250 ~ 20 31-40 0.245 w 0:: oCt U.-'

U. 9 o -! 0:: ~6 ~ 12 0:: ~ ~ ::> ll...4 z ~ ffi2 CO ~ 4 ::> z 0 '-='=_-'-----J'--~_ _'___:'_:__-=' -90 -60 -30 0 30 60 90 E O~ __-L~~ __ ~ __ w ~=-L---' SOLAR LONGITUDE, deg -90 -60 -30 o 30 60 90 S N Fig. 2- Solar longitudinal distribution of SOLAR LATITUDE,deg major solar flares (The dotted line Fig. I-Solar latitudinal distribution of major solar represents the Gaussian curve approxi- flares mating (he dara.)

125 • INDIAN J RADIO & SPACE PHYS. VOl. 12. AUGUST 1983 small negative value of ~ shows that the frequency 30 Nand 30S. At sunspot maximum, the zone reaches distribution is almost similar in both the hemispheres, ± 15 deg. latitude, and last spot of a cycle appears at east and west, and the flares in western hemisphere are approximately ± 8.. The pattern obtained in Fig. 3 slightly higher than those in the eastern hemisphere. seems to show the details of the Maunder diagram. This result together with the one shown in Fig. 4 3.2 North-South Asymmetry suggests that the 19th solar cycle consisted of two The north-south asymmetry [A =(N - S)/(N + S)] of pronounced peaks of major flare activity, one in 1957 major solar flares from 1955 to 1979 is shown in and the other in 1959-60. The 20th solar cycle consisted Table 2. This asymmetry for these major flares from of two outstanding peaks in 1967 and 1970. The latest 7 1955 to 1974 has been studied earlier . It is clear from solar cycle (up to 1979) has shown one peak of major Table 2 that, generally. the north-south asymmetry is flare activity in 1978-79. Also there are two abrupt positive. with small and large fluctuations in its values, decreases in the northern hemisphere, the first being in except during 1958 (in solar cycle 19), 1972 and 1974 (in 1958 and the second in 1971. Here it may be worth solar cycle 20) and 1976, 1977 (in solar cycle 21). The mentioning that the polarity of the general magnetic negative asymmetry implies that the number of flares field of the sun in the northern hemisphere has changed are larger in southern hemisphere than the northern in 1958 and 1971. The rapid decreases in 1971 in the hemisphere. Earlier results 7 9 have also reported a number of the sunspots. the intensity of the radio negative asymmetry in the case of optical solar flares, emission at 2800 MHz, calcium plages " and 530.3 nm PeA-flares and sunspot magnetic classes, in 1958, 1972 and 1974. Table 2 .-. North-South Asymmetry of Major Flares in Solar The occurrence of major flares in northern and Cycles 19.20 and 21 southern hemispheres of the sun is different in the early N-S Year A =- - periods of the cycle 21 as compared to the same periods N+S of the cycles 19 and 20. Also there does not appear to Solar cycle 19 Solar cycle 20 be any strong relationship between N-S asymmetry 1955 0.143 1968 0.164 and II or 22 yr cycle of occurrence of major flares 1956 0.115 1969 0.638 though the asymmetry reversed (from positive to 1957 0.274 1970 0.5X7 1958 -0.257 1971 0.097 negative value) during or around the periods of 1959 0.778 1972 -0.132 l u4 rcversal . of polarity of solar polar magnetic field 1960 0.516 1973 0.024 (1958 and 1972). 1961 0.367 1974 -0.316 1962 0.391 1975 0.437 1963 0.636 3.3 Long Term Variation 1964 1.000 The distribution of major flares in latitudes shown in Solar cycle 21 Solar cycle 20 Fig.3 is interesting in comparison with Maunder's 1976 -0.500 butterfly diagram. Examining the latitudinal distri- 1965 0.333 1977 -0.250 1%6 0.909 1978 bution of sunspots from 1874-1913, Maunder? showed 0.138 1967 0.579 1979 0.089 that the first spots of a cycle occur at approximately

SOLAR CYCLE 19 SOLAR CYCLE 20 SOLAR CYCLE 21

i 5-60L 1955 1960 1965 1970 1975 1979 YEAR Fig. 3 Heliographic latitudes of major solar flares, 1955-79

126 • BADRUDDIN 1'1 al.: MAJOR FLARES OF SOLAR CYCLES 19,20 & 21 coronal emission line" have also been reported. Another also been found 17 by considering only the flares of observation is that the major flare frequency has high optical importance ~ 2. increased very drastically after 1977. in both the hemispheres. northern and southern. This major flare 3.6 Solar Rotation and Solar Activity activity in 1978-79 is the highest of all the years since Solar activity is basically caused by the interaction 1955. between solar magnetic fields, solar rotation and convective motions? 7. The solar cycle related variation 3..4 Sunspot ",umber-\Iajor Flare Activit~· Relationship of photospheric rotation rate occurs at high latitude. From the limited data of major flares for less than Another cycle related variation observed at latitude three solar cycles. it is not possible to reach at any greater than 65° is the polar field strength+". It was general conclusion regarding the occurrence of major suggested, on the basis of the solar polar rotation fla res with respect to phase of the solar cycle. However. rate"? and the polar magnetic field strength at two for solar cycles 19 and 20, the relation of major flare preceding minima 30, that the maximum of the cycle occurrence to the sunspot numbers has been studied by 21 ma y be ex pected to be very high. us. As shown in Fig. 5, there is a linear relationship The highest observed monthly mean sunspot between the number of observed major flare events number in cycle 19 was 253.8 in October 1957, in cycle and the sunspot number, although there is a great deal 20 it was 135.g in March 1969 (Ref. 31) and the highest of scattering. Using the flare data of H, importance ~ 2 monthly mean sunspot number in cycle 21 was 188.4 in for solar cycle 19 it is shown I 7 that the number of flares 160 x CYCLE 19 • is nearly proportional to the sunspot number for the >- u x z • CYCLE 20 / case in which these numbers are larger than 100 and UJ :::> / that this proportionality breaks down for the cases in a / ~ 120 SOLAR CYCLE 20/ which the number of sunspots are less than 100 LL UJ / : x U . although the number of observed flares tends to x Z / UJ . increase as the sunspot number increases. The linear a: / '5 60 / relationship between the transient geomagnetic u / . SOLAR CYCLE 19 u activity. generated by solar flares or coronal transients. a UJ / " . a: and sunspot number has been reported recently?".

21 After dividing the events Iisted 23 into six classes 700 according to increasing CFI the relationship of flare 1955-79 occurrence frequency and CFI has been studied for solar cycles 19 and 20 separately and for total period 1955-79 whose data for major flares are available. This

relationship is shown in Fig. 6. Fig. 6 shows that flare Vl UJ occurrence frequency first increases for lower values of a:: 400 CFI and then after a certain value it decreases almost LL:3 exponentially to very small value for the highest CFI ~ C''(CL£ 10 An exponential decrease with flare importance has a:: UJ III ,~ ~ ~ '~G- >----« NORTHERN FLARES OJ ~ 200

~ •..• ;0""'" 'jj)ARES iBO:)i~(: ~ .: •• i3 •.••..• ~J 0 -3 4-6 7-9 13-15 ~ r\..~S~·------19'60 1965 19'70 1975-·-~80 ' YEAR COMPREHENSIVE FLARE INDEX Fig.4 /vnnual occurrcnc. frequencies (in numbers) of northern Fig.6 -Relationship of the occurrence of major solar and southern major nares, 1955-79 nares to their comprehensive flare index

]27 INDIAN J RADIO & SPACE PHYS. VOL. 12. AUGUST 1983

September 1979 (Ref. 20). The amplitude of major 13 Badruddin & Yadav R Si tndian J Phvs, 568(1982) 68. solar flare activity is also high, as shown in Fig. 4, and 14 Howard R. Solar l'hvs tNethcrlandsi. 38 (1974) 59. it is highest for the solar cycle 21. These results are in 15 Hansen R T & Hansen S F, Solar Phn (Nether/ands). 44 (1975) 29 JO 225. agreement with the forecast made . on the basis of 16 Hakura Y. Solar Phvs (Sether/ands). 39 (1<)74) 493. solar polar rotation rate and the strength of the polar 17 Sakurai K. Astrophvs & Space s.. (Sether/ands). 63 ( 1979) 369. magnetic fields of preceeding cycles. 18 Ambroz P. Bumba V. Howard R & Sykora J. International A.ltrtJn(lmlca/ Union Symposillm ""11. 4~. (D. Reidel Acknowledgement Publishing Co .. Dordrecht, Holland). 1971, 696. 19 Das Gupta A K. Das T K & Sarkar S K. Solar Phvs The authors are grateful to Dr S P Agrawal. Head. (Nether/ands). 69 (1981) 131. Vikram Space Physics Centre, APS University, Rewa, 20 Waldmeier M. Solar Pins t Nethcrlandsi. 73 (1981) 207. for sending the major solar flares data. Two of the 21 Dodson H W & Hedeman E R. ReI' LAG-14. World Dara authors (Badruddin and N R Y) are thankful to Center-A [or Solar Terrestrial Plivsics. BOll/del'. Colorado. University Grants Commission. New Delhi, for USA. 1971. 22 Dodson H W & Hedernan E R. ReI' L .-IG-52. World Data providing financial assistance. Center-A [or Solar Terrestrial Phvsiis, Boulder. Colorado. (lSA,1975. References 23 Dodson H W & Hedeman E R. ReI' L .-IG-80. ttorld Dora I Maunder E W. Man-Not R Astron So<'(GH). 50(1890) 251. Cellter-A [or Solar Terrestrial Ph nics. Boulder. Colorado. 2 Maunder E W, Mon-Not R Astron Soc (GB). 82 (1922) 534. USA. 1981. 3 Newton H W & Milson A S. MOil-Not R Astron s« (GB). 115 24 Babcock H D. Astrophys J (L SA). 130 (1959) 364. (1955) 398. 25 Dodson H W & Hedeman E R. Solar n.., (.Vether/alltls). 42 4 Bell B. Smithsonian Contr Astrophvs (USA). 5 (1962) 187. (1975) 121. 5 Waldmeier M. Solar Phvs (Ncthcrtandsi, 20 (1971) 332. 26 Legrand J P & Simon P A. Solar Phvs (Sether/al/d.\). 70 ( 1981) 6 Rusin V. Rybansky M & Scheirich L, Solar Phvs i Netherlundss. 173. 61 (1979) 301. 27 Stefno J O. ReI' Prog Pins (G B). 41 (1978) 865. 7 Rene-Roy J. Solar Phys (Netherlands). 52 (1977) 53. 28 Livingston W & Duvall Or) T L Solar Ph cs ( ,"etherla/It/s). 61 8 Yadav R S. Badruddin & Santosh Kumar, Indian J Radio & (1979) 219. Space Phvs, 9 (1980) 155. 29 Howard R. Solar Phvs t Netherlandsi. 59 ( 1978) 243. 9 Hakura Y. NASA Technical Note NASA TND-4473, 1968. 30 Schatten K H. Scherrer P H. Svalgaard L & Wilcox J M. Geophvs 10 Guss D E, Phvs Rei' Lett (USA). J3 (1964) 363. Res Lett (USA). 5 (1978) 411. II Warwick C S, Astrophys J (USA). 141 (1965) 500. 31 Dodson H W. Hedeman E R & Mohler 0 C. s». Geopliv» Splice 12 Trotter D E & Billings D E, Astrophys J(USA), 136(1962) 1140. Phvs (USA). 12 (1974) 329.

•

128