Transcript The Solar Radio Microwave Flux and the Sunspot Number

The Solar Radio Microwave
Flux and the Sunspot Number
Leif Svalgaard
Stanford University, Stanford, CA, USA.
[http://leif.org/research]
Hugh S Hudson
University of California, Berkeley, Berkeley, CA, USA.
AGU Fall 2009, SH13C-03
Acknowledge input from Kiyoto Shibasaki (Nobeyama) and Ken Tapping (Penticton)
Japanese Observations at Toyokawa
(1951-1994) and Nobeyama (1994-now)
The Observations at 1, 2, and 3.75 GHz straddle the 2.8
GHz frequency of the 10.7 cm flux. The 3.75 GHz series
begins in 1951 and the other frequencies in 1957. We
scale all observations to the longest series (3.75 GHz)
Correlation 3.75 GHz Flux vs. 2 GHZ Flux 1957-2009
300
Monthly means
3.75 GHz
250
y = 0.9989x + 23.819
R2 = 0.9878
200
150
100
50
2 GHz
0
0
50
100
150
200
250
Composite Japanese Microwave Flux
The three (two of them scaled) series agree very
well and it makes sense to construct a composite
series as the simple average
Scaled Solar Radio Fluxes (Toyokawa-Nobeyama)
300
Monthly Averages
250
1 GHz*
2 GHz*
200
3.75 GHz
Average
150
100
50
0
1950
1955
1960
1965
1970
1975
1980
1985
1990
1995
2000
2005
2010
Scaling to the Canadian F10.7 Flux
Correlation F10.7 vs. Composite Japanese Series 1951-2009
300
F10.7 sfu
Monthly means
The next
step is to
scale to
the 10.7
cm flux
250
y = 1.1872x - 22.546
200
R2 = 0.9857
150
100
y = 0.0011x 2 + 0.8884x - 4.11
R2 = 0.9869
50
Composite
0
0
50
100
200
250
300
Average 3.75, 2, and 1 GHz Solar Flux scaled to F10.7 cm Flux
300
250
150
Monthly Averages
Average*
F10.7
200
150
100
50
0
1950
1955
1960
1965
1970
1975
1980
1985
1990
1995
2000
2005
2010
Stability of the Series?
If both series have a stable calibration, their ratio
should be constant in time. There is an indication
that the move from Ottawa to Penticton introduced
a small difference in level. We compensate for this
by dividing the Ottawa values by 1.0314 (and then
rescale)
Ratio Japanese*/F10.7
1.20
1.15
1.10
1.05
1.00
0.95
0.90
Ottawa
0.9904
0.85
0.80
1950
1955
1960
1965
1970
1975
Penticton
1.0215
1980
1985
1990
1995
2000
2005
2010
The Final Composite ‘F10.7’ Flux
The average of the Japanese and the Canadian series is our
final composite, which we shall use in the following. We have
considerable confidence in the stability and calibration of this
series. The constant level at each minimum is notable (green
box) and argues against secular changes
Composite F10.7 Solar Flux
300
sfu
Monthly Averages
250
200
150
100
50
0
1950
1955
1960
1965
1970
1975
1980
1985
1990
1995
2000
2005
2010
The well-known Relationship between the
Sunspot Number and F10.7
Sunspot Number vs. F10.7 Flux Monthly Averages
250
R
y = -1.4940E-11x 6 + 1.6779E-08x 5 - 7.4743E-06x 4 + 1.7030E-03x 3 - 2.1083E-01x 2 + 1.4616E+01x - 4.1029E+02
2
R = 0.9759
200
150
1951-1990
100
50
F10.7 sfu
0
0
50
100
150
200
250
The polynomial formula has no particular physical significance
300
The well-known Relationship between the
Sunspot Number and F10.7
Sunspot Number vs. F10.7 Flux Monthly Averages
250
R
y = -1.4940E-11x 6 + 1.6779E-08x 5 - 7.4743E-06x 4 + 1.7030E-03x 3 - 2.1083E-01x 2 + 1.4616E+01x - 4.1029E+02
2
R = 0.9759
200
150
1951-1990
100
1996-2009
50
F10.7 sfu
0
0
50
100
150
200
250
Changes significantly in solar cycle 23 (Tapping 2009)
300
Comparing the Synthetic Sunspot
Number with Observations
The observed International Sunspot Number, Ri, is
systematically and progressively ‘too low’ compared to
what we would expect from F10.7 starting in ~1991 [the
reason the interval 1951-1990 was used]
Sunspot Number (Observed) and Fitted from F10.7 Flux
300
R
Monthly Averages
250
200
Fit
Obs
150
100
50
0
1950
1955
1960
1965
1970
1975
1980
1985
1990
1995
2000
2005
2010
Comparing Ratios
The ratio between observed and fitted Sunspot Numbers should be one
[avoiding cases where R is too small – and still we have large noise near
solar minima – marked by small m’s on the graph]. The change in SSN
observers from Zurich to Brussels might introduce a small offset (less than
5%), but cannot account for the decrease during solar cycle 23
Observed Rz,i / Calculated Rz,i [for Rz,i >4]
2
SIDC
Zürich
1
?
m
0
1950
1955
m
1960
1965
m
1970
1975
m
1980
1985
m
1990
1995
m
2000
2005
2010
The Fe I line at 1564.8 nm has a very large and
easily measured Zeeman splitting. The Hydroxyl
radical OH is very temperature sensitive and the
lines weaken severely at higher temperatures.
CN
Courtesy Bill Livingston
The Fe I line at 1564.8 nm has a very large and
easily measured Zeeman splitting. The Hydroxyl
radical OH is very temperature sensitive and the
lines weaken severely at higher temperatures.
CN
Courtesy Bill Livingston
The Magnetic Field has Steadily Decreased During SC23. The
Temperature has Steadily Increased. At B = 1500 G, the Spot is
Effectively Invisible.
Decreasing Visibility due to this Effect may lead to an Undercount of
Sunspots and partly Explain the Changed Relationship with the
Microwave Flux
Livingston & Penn Umbral Data
1
0.9
Yearly averages
Intensity
3500
Gauss
0.8
3000
Fe I Line 1568
nmnm
1564.8
0.7
0.6
0.5
2500
0.4
Magnetic field
0.3
2000
0.2
0.1
0
1985
Intensity
1990
OH Line 1565 nm
'Temperature'
1995
2000
Year
2005
2010
1403 measurements since 1998
2015
1500
2020
Wild Speculation
Was the Maunder Minimum Just an
Example of a Strong L&P Effect?
Cosmic Ray proxies show that during both the Maunder
Minimum and the Spörer Minimum, the modulation of
cosmic rays proceeded almost as ‘usual’. So the
Heliosphere was not too different then from now, and
perhaps the spots were there but just much harder to
see because of low contrast because of B ≈ 1500 G.
Conclusions
• The Canadian and Japanese microwave
radiometry is stable, robust, and of high
quality
• The SSN began departing from its usual
correlation in Cycle 23
• The Livingston-Penn sunspot
measurements are consistent with the
SSN change
• The nature of solar activity appears to be
changing as we watch
F10.7 and Geomagnetic Diurnal
Variation Agree in Detail
300
F10.7
250
y = 5.4187x - 129.93
R2 = 0.9815
200
150
100
y = 0.043085x 2.060402
R2 = 0.975948
50
rY
0
30
35
40
45
50
55
60
65
70
Solar Activity From Diurnal Variation of Geomagnetic East Component
250
200
Nine Station Chains
F10.7 sfu
F10.7 calc = 5.42 rY - 130
150
100
12
13
14
15
16
17
18
19
20
21
22
23
1980
1990
2000
50
25+Residuals
0
1880
1890
1900
1910
1920
1930
1940
1950
1960
1970
2010
2020
0.2
0.18 1/MA
0.16
y = 0.0052x + 0.0791
R2 = 0.7153
0.14
1998-2009
0.12
0.1
0.08
y = 0.00538x + 0.06730
R2 = 0.66855
0.06
0.04
1965-1997
0.02
SQRT(RZ )
0
0
2
4
8
10
12
14
16
Alfvenic Mach Number (Bartels Rotation Averages)
18
16
6
The Relationship
between the Alfvenic
Mach number in the
solar wind (at 1AU) and
the sunspot number has
also changed in SC23
MA
70%
missing
14
12
10
8
6
M A = Vn1/2 /(20 B)
4
M A = 10 / [0.673 + 0.0538 Rz 1/2 ]
2
0
1960
1965
1970
1975
1980
1985
1990
1995
2000
2005
2010
Abstract
Since 1947 the flux of microwaves from the Sun at wavelengths between 3 and 30 cm [frequencies
between 10 and 1 GHz] has been routinely measured. This emission comes from both
chromosphere and the corona and has two main sources: thermal bremsstrahlung (free-free
emission) and thermal gyroradiation. These mechanisms give rise to enhanced radiation when the
density and magnetic field increase, so the microwave radiation is a good measure of general solar
activity. Strong magnetic fields occur in the network and can persist for weeks or longer; hence there
is a strong rotational signal in the emission superposed on a solar cycle variation of the background
coronal signal. The radio flux measurements can be calibrated absolutely and are not very sensitive
to observing conditions, and in principle have no personal equation. They may thus be the most
objective measure of solar activity, and our many decades-long flux record could throw light on the
important issue of the long-term variation of solar activity. The longest series of observations F10.7,
begun by Covington in Ottawa, Canada in April 1947 and is maintained to this day. Other
observatories also have long and continuing series of measurements of the microwave flux. One can
now ask how this measure of solar activity compares to other measures, in particular the sunspot
number. We correlate the sunspot number against the F10.7 flux for the interval 1951-1990, and
obtain a good polynomial fit (R^2 = 0.976) up until ~1991.0 after which time the observed sunspot
number falls progressively below the fitted number. Three obvious hypotheses present themselves:
1) The sunspot counting procedure or observers have changed, with resulting artificial changes of
the sunspot number as they have in the past.
2) Physical changes in the corona or chromosphere have occurred.
3) Livingston & Penn’s observations that the sunspots are getting warmer during the last decade,
leading to a decreased contrast with the surrounding photosphere and hence lessened visibility,
possibly resulting in an undercount of sunspots
The near constancy of the flux at minima since 1954 argues against a change of the physical
conditions at the source locations, leaving the exciting possibility that Livingston & Penn may be
correct.