A Comparison Between ESP
30.4 nm and SEM
Measurements
Leonid Didkovsky, Darrell Judge, and Seth Wieman
Space Sciences Center, USC
Why the Continuation of He II (30.4
nm) Irradiance Measurements is
Important
•
•
1.
2.
3.
4.
Long-term (14.5 years: 1996 -- now), practically
uninterrupted absolute solar irradiance
measurements from SEM to study long-term EUV
variations during solar cycle
A proxy for:
The Earth’s ionosphere changes;
Atmosphere neutral density variations;
Thermosphere temperature and composition
variations;
Solar models.
July 7, 2010
Stanford
2
ESP vs SEM
SEM ESP
1st-order:
Made at
USC
USC
Channels
3
9
1st-order
2
4
0th-order
1
QD
Dark
channel
No
Yes
Gain
Correction
No
Yes
Filter
Wheel
No
Yes
36, 25, 19, 30 nm;
0th-order: 4
channels (QD)
with 0 – 7 nm
July 7, 2010
Stanford
3
How ESP Irradiance is Calculated
• See details: Didkovsky, L., D. Judge, S. Wieman, T. Woods, and A.
Jones, "EUV SpectroPhotometer (ESP) in Extreme Ultraviolet Variability
Experiment (EVE): Algorithms and Calibrations", Solar Physics, p. 182,
doi: 10.1007/s11207-009-9485-8, Dec. 2009 (open access) or at
http://www-rcf.usc.edu/~leonid/papers/SolPhys2010.pdf
July 7, 2010
Stanford
4
14.5 Years of SEM EUV Flux With
EVE/ESP Data Overlapping From
DOY=2010120
SEM First Order Daily Average EUV flux, 1996/1/1 - 2010/6/2
ph/cm2*sec (26--34 nm, 1E+10)
4.5
daily ave.
flux
SEM_CLONE
4
RGIC
3.5
EVE/ESP
3
ESP2SEM
2.5
2
1.5
Calendar Year / Data point
1
1996
1997
1998
1999 2000
2001 2002 2003
2004 2005
2006
2007
2008
2009
July 7, 2010
Stanford
5113
4748
4383
4018
3652
3287
2922
2557
2191
1826
1461
1096
730
365
0
0.5
NEXT
slide
5
A comparison of SEM and ESP
Fluxes (Details)
SEM and ESP
7.5E-04
9% now; 4% with
MEGS ref. spectra
Instead of SOLERS22
Irradiance, W/m^2
7.0E-04
2.5% now; -2.5% with
MEGS ref. spectra
Instead of SOLERS22
6% is the SEMflux uncertainty for
the 2008 solar
minimum based
on seven SEM
underflights
SEM
ESP
36.258
6.5E-04
Why the ratio
between SEM
and ESP is not
stable?
6.0E-04
EVE Rocket (NASA 36.258)
5.5E-04
120
125
130
135
140
145
150
155
160
165
170
175
2010, DOY
July 7, 2010
Stanford
6
SEM to ESP Ratios
The ratio (dark
blue diamonds)
shows some
correlation with
solar activity
change (red
squares)
SEM_ESP Ratios
1.300
1.250
Ratio
1.200
SEM/ESP Ratio
18 nm Ratio
ESP Temp Ratio
SEM Temp Ratio
1.150
EEPROM
UPDATE
1.100
1.050
1.000
120
130
140
150
160
170
SEM-detector
temperature
sensitivity
(resolution) is too
low (1.4C/bit) to
detect any Trelated changes.
2010, DOY
July 7, 2010
Stanford
7
A Search for the Sources of SEMESP Differences
Four possible sources:
• Temperature-related change of dark
countrates (SEM only);
•Uncorrected particle-related signal
contamination if any (SEM only);
• Activity-related change of the
second-order influence (both);
• ESP degradation (ESP only)
July 7, 2010
Stanford
8
ESP Measures Darks Daily
Measured ESP dark counts (dark-blue points) show
some small (0.3 cnt) occasional fluctuations around the
thermal proxy (red) used for irradiance calculations
July 7, 2010
Stanford
9
ESP Detector Temperature
ESP temperature changes are very low, about 0.15 C˚
and are mainly corrected by the ESP Lev 1 program.
The uncorrected part is too small to contribute to any
significant change of ESP counts.
July 7, 2010
Stanford
10
ESP and SEM Efficiency
•
We compare below ESP and SEM effective counts with the same variation of
dark counts of 1 cnt/0.25s ( 4 cnt/s).
Parameter
Maximal Efficiency,
cnt/ph
Effective countrate,
1/s for 2010123
Uncertainty with 4
cnt/s, %
ESP (Ch9)
1.62E-6
SEM (Ch1&3)
2.15E-7
1051
149
0.38
2.7
This example shows that thermal variation in SEM darks could be one
of the sources of SEM_ESP difference.
July 7, 2010
Stanford
11
A Proxy for SEM Proton Flux
SOHO/CELIAS/MTOF PM
14
Protons, 1/cm^3
12
10
8
SOHO/CELIAS/MTOF PM
6
4
2
0
120
130
140
150
160
170
Time 2010, DOY
Proton flux at the SOHO location shows some sporadic
fluctuations not correlated to the SEM_ESP changes
July 7, 2010
Stanford
12
Sources of SEM-ESP Differences
That Will be Corrected if…
Activity-related change of the
second-order influence (both);
• ESP will be corrected by resuming the use of MEGS (daily)
spectra. This option in the ESP Lev 1 program was stopped for the
current time while MEGS-B is evaluated;
• SEM would be further improved if modeled spectra of solar EUV
variability for the 1996 – 2010 time period, with the level of MEGS
accuracy, could be available.
ESP degradation (ESP only)
• SEM (if continued operation) may provide the ratio of ESP
degradation till exact measurements on the next EVE SR flight in
2012.
July 7, 2010
Stanford
13
Summary
• ESP is an advanced version of SEM and allows
us to measure solar irradiances with better
accuracy than SEM;
• ESP Ch9 (30.4 nm) provides a SEM-proxy to
continue long-term solar EUV measurements
available from the USC SEM database since
1996;
• SEM/ESP ratio is changing with the solar
activity, mostly due to the use of the SOLERS22 spectrum for SEM flux calculations. If the
SEM calculation would use the MEGS reference
spectrum, the differences between SEM and
ESP would be within 5%. Some other factors
(SEM dark counts and ESP degradation) add
some uncertainty (5 – 6%) to this ratio.
July 7, 2010
Stanford
14
Acknowledgments
• This work was supported by the NASA
grant NNX08AM74G and by the University
of Colorado award 153-5979
July 7, 2010
Stanford
15