Author: Giulio Del Zanna at DAMTP , University of Cambridge.
The EUV variability during the solar cycle causes dramatic changes in the temperature and density of the Earth’s thermosphere-ionosphere, directly affecting satellite drag and possibly also affecting other atmospheric layers, hence our climate.
Several EUV irradiance monitors have been flown in the past 20 years. The Solar & Heliospheric Observatory (SOHO) Solar EUV Monitor, SEM  provided broad-band EUV irradiances from 1996. The Solar EUV Experiment’s EUV Grating Spectrograph, EGS  onboard the Thermosphere Ionosphere Mesosphere Energetics Dynamics (TIMED) spacecraft has provided low-resolution spectral data since 2002, while between 2010 and 2014 the Solar Dynamics Observatory (SDO) Extreme-ultraviolet Variability Experiment EVE , and its prototype flight PEVE in 2008, returned medium-resolution spectral data.
However, measuring EUV irradiances is a challenge. The major difficulty is in characterising the radiometric calibration of the space instruments, and their long-term degradation which is notoriously dramatic.
Between 1998 and 2014, the SOHO Coronal Diagnostic Spectrometer, CDS  Normal Incidence Spectrometer (NIS) obtained full-Sun scan observations about once per month (see Figure 1). These are unique in allowing the study of radiance distributions of various solar features, and how these affect the total irradiance. Here, I summarise the results obtained from these observations as part of a long-term project which I started in 2005.
The SOHO/CDS NIS calibration, radiances and irradiances
Since 1996, I have been working on an improved calibration of the CDS instrument. The standard long-term calibration of the CDS/NIS made some assumptions that turned out to be incorrect, so by 2010 the degradation was overestimated by a factor 2-3. I developed a new, long-term degradation for the 1996-2010 period that showed agreement with observations from several sounding rockets . Over the past two years, I have further improved the calibration by revising the degradation at the line centres (the `burn-in’), and extending the long-term calibration until 2014 . These revised CDS calibrations have prompted similar revisions of several other EUV instruments. I have analysed the full-Sun CDS observations obtained during the 1998-2017 period, and obtained from the radiances the total irradiances [6,7]
The CDS/NIS, which had the Rutherford Appleton Laboratory as PI institute, turned out to be the best EUV spectrometer in space, with a degradation in response of only a factor of two over its lifetime. It is a success story for UK contributions to solar physics.
We used the data until 2010 to describe the characteristics of the radiance distributions of all the spectral lines across the solar cycle . We obtained several new results, in particular new limb-brightening curves which differ from previous ones. We also found that the radiance distributions in transition-region lines do not change significantly with the cycle.
Our final CDS/NIS irradiances have been compared in  with those obtained by the other EUV irradiance monitors. Figure 2 shows some examples.
The strongest EUV line. Was the last solar minimum different ?
As discussed in , most historical radiance and irradiance measurements of the He II 304 Å line, the strongest in the EUV, were incorrect by large factors. With the new CDS calibration, excellent agreement (within 10-20%) is found with the EVE v.5 data.
We have also resolved a question about the EUV irradiance variations between subsequent minima first reported from SOHO/SEM. These observations indicated the possibility that the extended minimum of Cycle 23/24 (2008-2010) had an EUV flux 12% lower than the previous minimum in 1996 . We have found that using CDS to estimate the SEM first-order (SEM 1) count rates gives agreement within 10-20% with the observed SEM count rates during solar minimum, when the He II 304 Å line is the dominant contribution to the band . This good agreement during minimum conditions suggests that it is quite possible that during the last minimum around 2008 the He II 304 Å irradiance was indeed lower, compared to the previous minimum, although the variation is marginal.
However, the solar spectrum in the SEM 1 band changes significantly during maximum conditions, when the He II line only contributes about 40% (the rest of the SEM contribution being due to off-band coronal emission). We originally thought that this off-band contribution was the cause of the high observed SEM count rates during the Cycle-23 maximum, but now conclude that the reasons for this are still unknown.
The analysis of the various EUV irradiances has allowed us to quantify how the irradiances of lines formed at temperatures above 1 MK are affected by the solar cycle. We have also shown that the irradiances of transition-region lines (e.g. O IV) are nearly unaffected by the solar cycle, with the exception of the helium lines.
In addition our results indicate that the CDS/NIS irradiances are the most reliable ones for Cycles 23 and 24, and that the SDO/EVE v.5 data, at least up to 2012, are in good agreement with CDS. However, the calibration of TIMED/EGS data, which are very limited after 2011, is clearly in need of revision. The solar minimum PEVE irradiances of 2008 have also been overestimated in several cases.
Support by STFC is acknowledged. Part of this long-term work was funded by the SOLID EU FP7 network. I acknowledge support and contributions over this long-term project by various colleagues, in particular V. Andretta. SOHO is a mission of international cooperation between ESA and NASA.
-  Hovestadt, D., et al. 1995, Solar Phys., 162, 441
-  Woods, T.N. et al., 2005, JGR (Space Physics), 110, 1312
-  Woods, T.N. et al., 2012, Sol. Phys., 275, 115
-  Harrison, R. et al., 1995, Sol. Phys., 162, 233
-  Del Zanna, G. et al., 2010, A&A, 518, A49
-  Del Zanna, G. & Andretta, V., 2015, A&A, 584, A29
-  Del Zanna, G. & Andretta, V., 2011, A&A, 528, A139
-  Andretta, V. & Del Zanna, G. 2014, A&A, 563, A26
-  Del Zanna, G., Andretta, V., Wieman, S., & Didkovsky, L. 2015, A&A, 581, A25
-  Didkovsky, L. & Wieman, S. 2014, JGR (Space Physics), 119, 4175
-  Wieman, S. R., Didkovsky, L. V., & Judge, D. L. 2014, Sol. Phys., 289, 2907