SOLAR SOLSPEC Mechanical shaft and motorization FIlter wheel 2d grating Exit slit Detector Mirror Ribbon tungsten or deuterium lamp Iintermediate slit Hollow cathode lamp (UV-VIS) Mirror Coupling Entrance slit Baffle Main shutter Quartz or internal shutter 1st grating Diffusor
SOLAR SOLSPEC Absolute calibration at the PTB Blackbody radiator BB3200pg E BB (λ, T ) = π.l(λ, T ).D 2 4d 2
SOLAR SOLSPEC Launch 8th February 2008 Spectral irradiance in orbit 1 Eλ1 = S lin.r R ( λ ). S < DC > + C (λ ) + C (λ ) E sol (λ ) = net λ Δλ Δt DC (TPTB TISS ).α T (λ ) F ( x, y ). 1 100 TSI estimation: (1357,1 ± ~15) W/m²
SOLAR SOLSPEC IR channel Comparison between space NIR SSI measurements + IRESPERAD (blue dots) Brightness temperature ~350 K below ATLAS3 Ground-based validation campaign (IRESPERAD) to provide new measurements of Top Of Atmosphere (TOA) Solar Spectral Irradiance (SSI) in the NIR Importance of NIR SSI Absorption by the atmosphere (CO2, H2O, O3, O2) & upper layer of the oceans role in the Earth s radiative budget Challenge for the measurements (Bouguer-Langley method ) Instrumentation calibration site of measurements
IRSPERAD campaign The instrumentation IRSPERAD 600 2300 nm 10 nm bandwidth PbS detector Phase sensitive detection Entrance optic: tube designed for SSI measurements (concentric pinholes & frosted quartz) FOV 7.2 Absolute calibration: using the same blackbody than for SOLSPEC, at the same distance Bentham Double monochromator
IRSPERAD campaign The Site High-mountain Izaña Observatory (IZO) Managed by the Izaña Atmospheric Research Centre (IARC), from the State Meteorological Agency of Spain (AEMET) Tenerife (Canary Islands) 28º3 N, 16º5 W 2367 m asl Insolation ~ 3400 h/year - IZO is part of the GAW Precision Filter Radiometer (PFR) network, managed by the World Radiation Centre (Davos, Switzerland) - IZO hosts the reference triad of the WMO-GAW Regional Brewer Calibration Centre for Europe (RBCC-E) - IZO is a direct-sun calibration site of AErosol RObotic NETwork (AERONET)
IRSPERAD campaign The methodology Bouguer-Langley method using WTLS algorithm ( ) log(e (λ ) ) = log E0 (λ ) R 2 m(θ )τ (λ ) Selected wavelengths E0 (λ ) TOA SSI at 1 UA E (λ ) ground-based SSI θ = Solar zenith angle τ = optical depth AMF (m) between 2 and 8 Atmospheric windows (164 channels) Method: Kindel et al., Appl. Opt, 2001, pp 3483-3494
IRSPERAD The campaign Results presented here: from first two weeks (June 2011) optimal weather conditions and strong control of all scales (radiometric, λ) 1-June 18:30 Data selection clear sky, free of visible clouds free of any tin (invisible) cirrus clouds (LIDAR monitoring) low Aerosol Optical Depth (AOD), AERONET data & low variability Bouguer-Langley plots: coefficient of regression > 0.99
IRSPERAD The campaign λ = 870 nm R² = 0.9976 λ = 1070 nm R² = 0.9985 λ = 1640 nm R² = 0.9993 Gaussian fit for the 164 channels: example for λ = 1070 nm Langley plots: 66 available events for the whole campaign one third for the first two weeks
IRSPERAD campaign Results & uncertainties Standard uncertainties calculated using the Law of Propagation of Uncertainties (LPU) for 11 relevant wavelengths E= S Iz + Cλ R.Corr α ΔT 1 T 100 SIz electronic output signal obtained for a given solar measurement Cλ stability of λ scale Denominator thermal correction (if required) for the detector response E BB R= S BB response curve Corr response change (relative units)
Discussion & comparisons IRSPERAD campaign Discussion & comparisons 1) SOLSPEC ATLAS3 SOSP version of SOLSPEC (EURECA 1992 1993) Absolute calibration (blackbody of Heidelberg) Announced standard uncertainty around 3 % 2) Arvesen airborne set data at 12 km of altitude (1969) Calibration using a 1000 W quartz-halogen lamp Standard uncertainty around 3 %
IRSPERAD campaign Discussion & comparisons 3) SIM on SORCE prism spectrometer (2003 ) A SIM-to-ATLAS3 deviation was found and corrected because: the calculated uncertainties reached 8 % for SIM above 1.35 µm the need to achieve both a more plausible TSI and solar brightness temperature values (around 1.6 µm) as ATLAS3 did
IRSPERAD campaign Discussion & comparisons 4) SCHIAMACHY remote-sensing spectrometer with NIR channels (2002 2012) Absolute calibrated Estimated standard uncertainty: 2 % 5) IRSPERAD
IRSPERAD campaign Discussion & comparisons 6) SOLAR/SOLSPEC SOLAR mission on ISS (2008 ---) Renewed NIR channel, absolute calibrated at the PTB (blackbody) Estimated standard uncertainty: lower than 2 % for most of the NIR range
IRSPERAD campaign Discussion & comparisons 7) CAVIAR ESS high-resolution Fourier Spectrometer (2008) TOA SSI retrieved using the Bouguer-Langley method (at sea level) Absolute calibrated at the NPL (blackbody) Estimated standard uncertainty in the range of 3.3 % to 5.9 % Composite results: All sets of data are independent (hardware, methodology, calibration) Except: 1) IRSPERAD SOLAR/SOLSPEC (blackbody) 2) CAVIAR ESS IRSPERAD (Bouguer-Langley)
IRSPERAD campaign Conclusions IRSPERAD Based on a simplified methodology (Bouguer Langley) providing an easy access to the TOA solar spectral irradiance. Stable instrumentation (full radiometric control over the experimental setup and the error sources). Reliability of the calibration using the blackbody of the PTB as primary standard source. Benefit of the ideal atmospheric conditions at IZO This measurement campaign could provide a reference NIR SSI data set from the ground. However, limited to a series of NIR wavelengths that correspond to the atmospheric windows. This work was intended to bring some contributions to the actual discussions on the solar irradiance around 1.6 µm. This work is published in Solar Physics since 17th January 2014 online version DOI 10.1007/s11207-014-0474-1
UV channel SOLAR SOLSPEC Spectral range: 166-371 nm. Improvements for the long term SOLAR mission: PSD, quartz plates (shielding) & 2 internal deuterium lamps. Reliability of the calibration using the blackbody of the PTB and deuterium lamps (below 200 nm, under vacuum). Main objectives Solar UV variability during cycle #24 (for modelling purposes of solar physics, climatology and atmospheric sciences) Spare & common lamps Comparisons with other instruments (SOLSTICE, )
SOLAR SOLSPEC UV measurements Solar cycle #24 Solar cycle #24 Log book of solar UV measurements 6 years of nominal solar meas. (0.6s integration time). 1 ½ year of special UV solar meas. (20s integration time). 1 year of quiet Sun & at 2009.5 loss of D2 lamps power supply. 2 mains fields of activity: short term & long term campaigns
SOLAR SOLSPEC UV measurements Short term campaigns (usual SVW & SVW bridging) Comparisons with SOLSTICE on SORCE& EGS (scaling factors)
SOLAR SOLSPEC UV measurements Short term campaigns (usual SVW & SVW bridging) MgII index: solar proxy between 120 and 400 nm. Scaling factor % change in SSI for 1 % change in MgII index. Also determined for SOLAR SOLSPEC from short term campaigns.
SOLAR SOLSPEC UV measurements Long term campaign (6 years) To build a database of well calibrated solar spectra from SOLSPEC Main topic: the ability of instruments to see the real solar UV variability Evidence of solar UV change in the signal of nominal solar SOLSPEC measurements Evidence of solar UV change in the signal of nominal solar SOLSPEC measurements
SOLAR SOLSPEC UV measurements Long term campaign (6 years) adjustment of the wavelength scale λ ( p) = a sin(b + arcsin( cp + d )) From the internal HCL lamp and Fraunhofer lines wavelength scale well monitored
SOLAR SOLSPEC UV measurements Long term campaign (6 years) adjustment of the wavelength scale SOLSPEC absolute wavelength scale verifica:on H. Slaper algorithm WHI : reference spectrum Composite spectrum (TIMED SEE, SORCE, rocket experiment) 0.1 nm resolu:on binned down to 0.1 nm Spectra taken between 25 and 29 march 2008 - > moderately low solar ac:vity SOLAR SOLSPEC Averaged spectrum for 6 spectra in the quiet- sun period (intercycle 23-24). April may 2008 Beginning of measurement campaign - > virtually no degrada:on H. Slaper et al., Comparing Spectral Solar UV Measurements, Geophys. Res. Le,., 22, 20, 2721-2724, 1995
SOLAR SOLSPEC UV measurements Long term campaign (6 years) adjustment of the wavelength scale Algorithm for convolu:on of spectra to a variable bandpass func:on by C. Fayt and M. Van Roozendael SOLSPEC and WHI have: same bandpass same λ grid
SOLAR SOLSPEC UV measurements Long term campaign (6 years) adjustment of the wavelength scale Qualita:vely: Valida:on of variable BP convolu:on method Valida:on of pre- flight SOLSPEC BP func:on characteriza:on
SOLAR SOLSPEC UV measurements SOLSPEC measurement from the 5th april 2008 Good pre- flight SOLSPEC λ scale characteriza:on: λ shid(t=0) = - 0.011 nm Average wavelength shid with an important dispersion: <λ shid> = - 0.019 nm +- 0.005 nm(24%) Stable around average value (quasi- zero trend) res(filtered curve/points) = 0.0047 nm
SOLAR SOLSPEC Long term campaign (6 years) aging correction of the UV channel response
SOLAR SOLSPEC Long term campaign (6 years) aging correction of the UV channel response Deuterium lamps allow to correct for the UV channel ~ [150,300] nm degradafon Lamps degrada:on model: D1,2(λ,t) = 1+a1,2(λ).T1,2(t) UV channel response degrada:on: d = d(λ,t) 1) different frequencies of lamps u:liza:on 01- oct- 08 02- may- 08 23- apr- 09 2) same aging dynamics: a = ai = a1 = a2
SOLAR SOLSPEC Long term campaign (6 years) aging correction of the UV channel response quartz plate correction no quartz plate correction SOLSPEC op:cal path Deuterium Lamps op:cal path need for correc:on for Quartz plate in SOLSPEC op:cal path method s λ range validity degrada:on as a percentage of UV channel response loss λ (nm) / Fme 3- month 6- month 1- year 14- month* 200 95.9% 85.2% 64.2% 58.7% 210 96.6% 86.2% 64.6% 59.3% 220 98.8% 90.7% 69.7% 69.1% * end of life of Deuterium lamps, due to a techincal failure
SOLAR SOLSPEC Long term campaign (6 years) aging correction of the UV channel response
SOLAR SOLSPEC Long term campaign (6 years) aging correction of the UV channel response: results
SOLAR SOLSPEC Long term campaign (6 years) retrieval of the Solar UV variability
SOLAR SOLSPEC Long term campaign (6 years) new estimation of the scaling factors
SOLAR SOLSPEC - CONCLUSIONS SOLAR SOLSPEC: robust instrument, everything being operational after 6 years of mission (except D2 lamps). Monitoring of all opto-mechanical and electronic parameters. End of the SOLAR mission: en of 2017. NIR channel: new input for the SSI around 1.6 µm. UV channel: Improvements of the aging correction (channel response). Final calculation of scaling factors. Final retrieval of the SSI UV variability during the solar cycle #24 Database of solar spectra, in well maintained absolute radiometric and wavelength scales. Large time scale comparison between SORCE & SOLSPEC for the SSI UV variability. Scaling factor = 1? To perform additional work on the VIS and NIR and provide a SOLAR SOLSPEC reference spectrum.
Thank you for your attention