Renewing our view to past solar activity: the new Sunspot Number series Frédéric Clette, Laure Lefèvre World Data Center SILSO, Observatoire Royal de Belgique, Brussels with contributions from Edward W. CliverNational Solar Observatory, Boulder Leif SvalgaardStanford University José VaqueroUniversidad de Extremadura ROB STCE

Sunspot Number: a composite series Historical reconstruction ( ): Scale adjusted to Wolf’s observations Primary Wolf observations ( ): Standard 83mm refractor Small 40mm travel telescopes Zürich period ( ): 3 reference observers: Wolfer, Brunner, Waldmeier New counting method: 0.6 factor 1957: Zürich + Specola (Locarno) stations 26/11/2015ESWW12, Oostende, Space Climate Session2

Sunspot Number: a composite series Brussels period (1981-now): New pilot station: Specola (Locarno) Statistics over worldwide network 26/11/2015ESWW12, Oostende, Space Climate Session3 Historical reconstruction ( ): Scale adjusted to Wolf’s observations Primary Wolf observations ( ): Standard 83mm refractor Small 40mm travel telescopes Zürich period ( ): 3 reference observers: Wolfer, Brunner, Waldmeier New counting method: 0.6 factor 1957: Zürich + Specola (Locarno) stations

Sunspot Number versus modern measurements 26/11/2015ESWW12, Oostende, Space Climate Session4 Mean mag. flux Sunspot Nb. Stenflo, 2012 STARA catalog; F. Watson, 2012 High correlation with photospheric parameters: – automated counts (MDI,HMI), sunspot area, emerging magnetic flux Mx Quantitative measure Lower correlation with chromospheric and coronal indices (F 10.7cm, CaII K, MgII, TSI) – Additional component: magnetic decay (plages, network) – Non-linearity, time lags Are those relations stable over long durations ?

Two incompatible sunspot records Only alternate series: Group Number (Hoyt & Schatten 1998) – Larger data set, back to 1610 – Only groups: more immune to cruder early observations Large persistent discrepancies between the series (up to 40%) ESWW12, Oostende, Space Climate Session526/11/2015

A necessary revision: Sunspot Number Workshops 4 workshops ( ): Sac. Peak, Brussels, Tucson, Locarno Multiple diagnosed problems in the SN and GN: – Clette, F., Svalgaard, L., Vaquero, J.M., Cliver, E.W.: 2014, Revisiting the Sunspot Number. A 400-Year Perspective on the Solar Cycle. Space Sci. Rev. 186, – Solar Physics topical issue (early 2016): Clette & Lefèvre (New Sunspot Number), Svalgaard & Schatten (New Group Number) 26/11/2015ESWW12, Oostende, Space Climate Session6

Methodology Self-consistent determination of corrections: – Input data: only direct sunspot information – External indices: for comparison and post-validation only – Equivalent to other absolute measurements (TSI) Diverse approaches: – Construction of mean multi-station reference series: Global statistics over many simultaneous stations (cross-normalization, means) Use of long-duration stations ("backbone" observers) – Data recovery: New additional observers and data series (early scarce data) Critical elimination of spurious data – Consultation and critical revision of original historical documents: Evolution of observing and processing practices Re-counting from early drawings (modern sunspot group splitting) – New observing series for testing and quantifying suspected past discrepancies (e.g. double counts) Sunspot and Group numbers corrected independently 26/11/2015ESWW12, Oostende, Space Climate Session7

Sunspot Number corrections: overview ESWW12, Oostende, Space Climate Session826/11/2015 Schwabe - Wolf transition ( ) Waldmeier’s spot weighting ( ) Locarno’s variable drifts ( )

Group Number correction: overview ESWW12, Oostende, Space Climate Session9 Revision of GN database ( ) Elimination of interpolated nulls ( ) Reconstruction: 5 “backbone” observers ( ) Svalgaard & Schatten (2015) “Greenwich” trend ( ) 26/11/2015

Combining all corrections: matching SN and GN Original series: SN / 0.6 GN x 18. ESWW12, Oostende, Space Climate Session1026/11/2015

Combining all corrections: matching SN and GN Close agreement over the entire interval Still significant differences before 1826: (10% – 20%): Target for next upgrade ! ESWW12, Oostende, Space Climate Session11 26/11/2015

Better agreement with modern solar indices Drop of SN versus F 10.7cm after 2000 : up to -20% (Svalgaard & Hudson 2010, Lukianova & Mursula 2011, Clette & Lefèvre 2012) New comparison of F 10.7 with the corrected SN and “backbone” GN ( ) No anomaly after 2000 F 10.7 is too high by 10% after 1983 ESWW12, Oostende, Space Climate Session1226/11/2015

A non-linear S N -G N relation Decline in the average number of spots per group over past cycles, enhanced after after the maximum of solar cycle 23: (Tlatov 2012, Clette, Lefèvre 2012,Svalgaard 2013, Clette et al. 2014) Reconstruction of SN and GN from the same base set of stations (SILSO network) over (Clette et al. 2015) Constant non-linear relation (degree 2 polynomial) : Drop in cycle 24: deficit of large sunspot groups (Kilcik et al. 2014) ESWW12, Oostende, Space Climate Session1326/11/2015

Uniform peak cycle amplitudes over last 3 centuries Original series: strong upward secular trend from the end of the Maunder Minimum to the mid 20 th century (“Modern maximum”, Solanki et al. 2004, Usoskin 2013 ): – GN: + 40% / century (red) SN : + 15% / century (green) New SN and GN= similar very weak upward trend < 5 %/century (blue, purple) Soon after the Maunder Minimum, solar activity returned to high levels equivalent to recent cycles of the 20 th century ESWW12, Oostende, Space Climate Session1426/11/2015

Comparison with indirect indices of solar activity Geomagnetic indices: reconstructed open magnetic flux over the last 180 years (Lockwood et al, 2013) – Recent reconstructions show identical cycle amplitudes between the mid-19 th century and the 20 th century ESWW12, Oostende, Space Climate Session15 Lockwood /11/2015

Conclusions and next steps SN Version new GN released on the new SILSO Web site: July 1st 2015 New conventions: R i S N symbol Elimination of Zürich factor : SN / 0.6 New reference = A. Wolfer ( ) ESWW12, Oostende, Space Climate Session1626/11/2015 Recent wave of interest for the past solar record From a locked archive to a living data series: Open to future improvements (IAU supervision) Archive of past versions

The future: preparation of Version 3 Development of a new operational method – Implementation of a multi-station reference Full re-calculation of the SN since 1981 ESWW12, Oostende, Space Climate Session17 Application of more advanced statistical methods (ARMA, Bayesian regressions, multiscale decomposition, PCA) : – Sparse data and gaps – Time-variable uncertainties Revisiting Wolf’s work on historical data /11/2015

Stay tuned …. SN workshops : 26/11/2015ESWW12, Oostende, Space Climate Session World Data Center – SILSO Sunspot Index and Long-term Solar Observations

26/11/2015ESWW12, Oostende, Space Climate Session19

Better agreement with modern solar indices ESWW12, Oostende, Space Climate Session20 Yeo et al /11/2015 Amplitude and shape of recent solar cycle maxima: – Mismatch between original SN and solar irradiances (TSI, MgII, Lyα, total sunspot magnetic flux) Second peak in cycle 23 (November 2001) now higher than first peak (July 2000) Main unexplained discrepancies are eliminated. Clette & Lefèvre 2015

Conclusions: lessons learned Most diagnosed problems in SN and GN due to: – Improper processing methods – Method changes deviating from an original definition Processing methods have a bigger impact than random errors in the data Defects diagnosed in the SN recalibration may still influence irradiance reconstructions (SATIRE, NRLTSI) : – Trends in RGO photographic group areas and counts (Balmaceda et al. 2009) – Imbedded semi-empirical models based on the uncorrected Hoyt & Schatten Group Number series: Cosmogenic isotopes and open solar magnetic flux (Solanky et al. 2002, Usosking et al. 2005, Lockwood et al. 2014) Magnetic flux transport (Wang, Lean & Sheeley 2005) Using the new SN as direct input parameter is not sufficient 26/11/2015ESWW12, Oostende, Space Climate Session21

Future prospects: beyond the sunspot counts New needs: – Hemispheric SN before 1950 (N/S asymmetries) – Global distributions: latitude, size – Magnetic dipole (width, tilt) The information exists ! Dispersed, paper documents – Need for digitization – Feature extraction software (standardization) Prospects for advanced multi- secular sunspot-based proxies 26/11/ Digitized Carrington drawing, Tlatov 2012 DigiSun software, ROB, Brussels E. Spee, 1895, ROB, Brussels Staudacher Hevelius ESWW12, Oostende, Space Climate Session