Sixty+ Years of Solar Microwave Flux
Leif Svalgaard
Stanford University
SHINE 2010, Santa Fe, NM
Observations and Observatories• 1947.2-1991.5 Ottawa 2800 MHz (10.7 cm)• 1991.5-2010.5 Penticton 2800 MHz• 1951.8-1994.3 Toyokawa 3750 MHz (8 cm)• 1994.4-2010.5 Nobeyama 3750 MHz• 1957.4-1994.1 Toyokawa 2000 MHz (15 cm)• 1994.4-2010.5 Nobeyama 2000 MHz• 1957.2-1994.4 Toyokawa 1000 MHz (30 cm)• 1994.4-2010.5 Nobeyama 1000 MHz• 1956.3-1994.1 Toyokawa 9400 MHz (3.2 cm)• 1994.4-2010.5 Nobeyama 9400 MHz
All observations are highly correlated, especially the ones [2000, 3750] flanking 2800 MHZ
F2800 = 1.5707 F1000 - 0.9825
R2 = 0.9742
0
50
100
150
200
250
300
0 20 40 60 80 100 120 140 160 180
F2800
F1000
1957-2010
F2800 = 1.177 F2000 + 6.6574
R2 = 0.9933
0
50
100
150
200
250
300
0 50 100 150 200 250
F2800
F2000
1957-2010
F2800 = 1.1728 F3750 - 20.78
R2 = 0.9949
0
50
100
150
200
250
300
350
0 50 100 150 200 250 300
F2800
F3750
1951-2010
F2800 = 1.4215 F9400 - 294.89
R2 = 0.9584
0
50
100
150
200
250
300
0 50 100 150 200 250 300 350 400 450
2800
F9400
1956-2010
Monthly Averages
This allows us to scale all observations to 2800 MHz
0
50
100
150
200
250
300
350
1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010
F9400
F3750
F2000
F2800
F1000
Micro Wave Fluxes Scaled to 2800 MHz Flux
The scaled values match each other very well, except 9400 MHz where the solar activity component is rather noisy.
If the calibrations of the data have not changed over time, the ratios between the scaled fluxes and
the 2800 MHz flux should be constant = 1
0.80
0.85
0.90
0.95
1.00
1.05
1.10
1.15
1.20
1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010
TYKW NBYM
OTWA PENT
Ratio between Scaled Japanese Flux and Canadian 2800 MHz Flux
Within the scatter, the ratios appear to be lower than 1 before 1991 and higher thereafter. We interpret that as a 2% downward jump in Pentiction compared to Ottawa. We therefore reduce the Ottawa 2800 MHz flux by 2%.
Same color associations as on previous Figure
After the correction, we rescale 2000 and 3750 MHz again to 2800 MHz, and construct a composite
F2800 = 4.56239E-06x3 - 7.93355E-04x2 + 1.13869E+00x + 1.10179E+01
R2 = 0.9939
0
50
100
150
200
250
300
0 50 100 150 200 250 300
F2800*
F2000
1957-2010
F2800 = -8.49042E-06x3 + 3.71190E-03x2 + 6.45618E-01x + 1.50641E+00
R2 = 0.9955
0
50
100
150
200
250
300
0 50 100 150 200 250 300
F3750
F2800* 1951-2010
0
50
100
150
200
250
300
1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010
Composite F2800 MHz Solar Microwave Flux
Note how the minimum values do not vary with time (within green box)
0
50
100
150
200
250
300
1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956
Cycle 18
0
50
100
150
200
250
300
1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967
Cycle 19
Individual cycles
0
50
100
150
200
250
300
1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978
Cycle 20
0
50
100
150
200
250
300
1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989
Cycle 21
0
50
100
150
200
250
300
1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999
Cycle 22
0
50
100
150
200
250
300
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Cycle 23
Cycle 24
There is a well-known, strong [slightly non-linear] relationship between the solar flux and the sunspot number (black diamonds). This relationship seems to have changed in solar cycle 23 (red circles).
y = 9.9392E-12x6 - 7.1116E-09x5 + 1.6776E-06x4 - 8.2748E-05x3 - 2.4487E-02x2 + 4.8162E+00x - 2.0983E+02
R2 = 0.97391
0
50
100
150
200
250
300
0 50 100 150 200 250 300
1947-1990
1996-2010
Same conclusion reached by others, e.g. Tapping (2010)
Tapping, SORCE 2010
Plotting the reconstructed Sunspot Number (pink) from the composite 2800 MHz flux using the 1947-1990
relation shows the increasing discrepancy with the SIDC ‘official’ sunspot number (blue) the past ~15 years:
0
50
100
150
200
250
300
1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015
Sunspot Number Observed and Reconstructed from Composite Flux Series
SSN Observed
SSN From 2800 Flux
As the Japanese and Canadian microwave data support each other so well, we must ask: how sure are we of the calibration and stability of the sunspot number?
Some other organizations that have kept a keen eye on the Sun, making
their own sunspot number series:ht t p: / / www. vds- sonne. de/ gem/ r es/ r esul t s. ht mlSIDC: Solar Influences Data Analysis Center, Brussels (International sunspot numbers)SONNE prov.: SONNE network, provisional sunspot numbers SONNE def. : SONNE network, definitive sunspot numbersAAVSO: American Association of Variable Star Observers - Solar DivisionAKS: Arbeitskreis Sonne des Kulturbundes e.V., GermanyBAA: The British Astronomical Association - Solar Section, UKGFOES: G.F.O.E.S. Commission "Nombre de Wolf", FranceGSRSI: GruppoSole Ricerce Solari Italia, ItalyOAA: The Oriental Astronomical Association - Solar Division, JapanRWG: Rudolf Wolf Gesellschaft - Solar Obs. Group of Swiss Astron. SocietyTOS: Towarzystwo Obserwatorow Slonca - Solar Observers Society, PolandVVS: Vereniging voor Sterrenkunde, Werkgroep Zon, Belgium
All these series can be successfully scaled to SIDC before August 2001
0
20
40
60
80
100
120
140
160
180
200
1994 1996 1998 2000 2002 2004 2006 2008 2010 2012
SIDC
RSONNE AAVSO AKS BAA GFOES GSRSI OAA RWG TOS VVS
But the same scaling relation yields a sunspot number after that time that is systematically 12% higher than SIDC’s. Did all these organizations somehow change their procedures and/or or observer cadre? Or did SIDC?
It seems that we increasingly see ‘fewer spots’ for the same amount of microwave flux. We can quantify that by the
ratio between observed spots and expected spots from the pre-1991 relationship:
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015
?
Ratio of Observed Sunspot Number to Simulated SSN
Zurich SIDC
+12% correction
Monthly means when R>4
Adding 12% [assuming that SIDC has a problem in August 2001] does not materially alter this conclusion (red crosses).
1500
2000
2500
3000
3500
4000
1990 1995 2000 2005 2010 2015 2020
0
0.2
0.4
0.6
0.8
1B Gauss Intensity
Year
Livingston & Penn Umbral Data Measurements
1000
1500
2000
2500
3000
3500
4000
1990 1995 2000 2005 2010 2015 2020
B Gauss
Year
Livingston & Penn Umbral Magnetic Field Measurements
Are there other indications of lesser visibility of sunspots?
Livingston and Penn report a decrease of magnetic field strength and hence decrease in contrast (less dark) with the surrounding photosphere.
Is this discrepancy another one?