Revised composite extraterrestrial spectrum based on recent solar irradiance observations

Solar Energy 169 (2018) 434–440

Contents lists available at ScienceDirect

Solar Energy journal homepage: www.elsevier.com/locate/solener

Revised composite extraterrestrial spectrum based on recent solar irradiance observations

T

Christian A. Gueymard1 Solar Consulting Services, Colebrook, NH, USA

A R T I C LE I N FO

A B S T R A C T

Keywords: Extraterrestrial spectrum Solar spectral irradiance Solar physics ASTM

A revision of a previous composite spectrum (Gueymard, 2004) is undertaken here, with a focus on the 200–4000 nm waveband, where most of the sun’s energy is concentrated. The methodology is based on 31 sources of spectral data covering various parts of the spectrum. The data sources include observations from surface observatories, aircraft, high-altitude balloons, satellites, as well as modeled estimates of the solar flux. All existing spectra are downscaled to a single, relatively low spectral resolution to allow their direct comparison. Three successive filtering steps make it possible to eliminate outliers at each of the 2165 wavelengths under scrutiny here. The mean irradiance at each wavelength is then obtained by averaging the values of the surviving data points. The 2004 spectrum is used below 200 nm and above 4000 nm to create a complete spectrum from 0.5 nm to near-infinity. In a final step, the irradiance values are properly scaled so as to integrate to the revised solar constant value of 1361.1 W m−2. Compared to the earlier 2004 spectrum, this new composite differs mostly in the UV and visible parts of the spectrum. Based on the present findings, a revision of the outdated ASTM E490 standard extraterrestrial spectrum is recommended.

1. Introduction Knowledge of the solar spectrum and its variations is required in a wide variety of disciplines, including astronomy, astrophysics, atmospheric physics, climatology, biology, human health, materials degradation, radiometry, and terrestrial or space solar energy applications. In particular, applications in atmospheric sciences and remote sensing typically require the determination of irradiance at many vertical levels of the atmosphere, where the solar spectrum is attenuated in various wavebands due to strong absorbers that are active either in the ultraviolet (UV), such as oxygen or ozone, or in the infrared (IR), such as water vapor or carbon dioxide. At the earth’s surface, the solar spectrum is almost completely attenuated below ≈300 nm and above ≈4000 nm. These limits change somewhat with altitude, to the point where upper stratospheric applications require the complete solar spectrum at all wavelengths, just like space applications. In any case, all calculations of the solar irradiance transmitted by the atmosphere of a planet belonging to the solar system must start with some knowledge of the solar spectrum at the top of that atmosphere. For terrestrial applications, this solar spectral irradiance (SSI) is usually referred to as topof-atmosphere (TOA) spectrum, or extraterrestrial spectrum (ETS). The latter acronym is used here. Spaceborne remote sensing applications rely on the selection of an ETS having as little bias as possible over the

1

E-mail address: [email protected]. ISES Member.

https://doi.org/10.1016/j.solener.2018.04.067 Received 13 March 2018; Received in revised form 24 April 2018; Accepted 30 April 2018 0038-092X/ © 2018 Elsevier Ltd. All rights reserved.

atmospheric windows sensed by the radiometer. Significant uncertainties can occur in the remote-sensed products if the ETS is biased, which makes the selection of the proper ETS critical (Trishchenko, 2006). Atmospheric transmission codes, such as MODTRAN (Berk et al., 1999b), libRadtran (Emde et al., 2016), or SMARTS (Gueymard, 1995, 2001), typically provide various user-selectable ETS options to ultimately simulate the solar irradiance incident at the surface or at various altitudes throughout the atmosphere, potentially using various nominal spectral resolutions. The ETS spectral resolution conditions that of the simulated irradiance. For solar applications, some standard reference terrestrial spectra have been promulgated based on SMARTS predictions, using 2002 wavelengths between 280 and 4000 nm (ASTM, 2003, 2014; IEC, 2016). This type of application is the focus in what follows, with however an extended spectral range from 200 to 4000 nm. In case a higher spectral resolution is required for more demanding applications, it is customary to modulate a low-bias/low-resolution ETS with a high-bias/high-resolution one through appropriate scaling (Bernhard et al., 2004, 2007; Kiedron et al., 2007; Michalsky and Kiedron, 2008). Over the years, many ET spectra, covering at least large parts of the complete solar spectrum, have been proposed in the literature. A partial list, covering the period 1940–2004, was compiled by this author (Gueymard, 2006). This list included the associated value of the solar

Solar Energy 169 (2018) 434–440

C.A. Gueymard

Table 1 Sources of spectral data for the new ETS. The left part is for sources that were previously used to develop the synthetic spectrum in G04 (Gueymard, 2004). The right part is for additional sources, not considered in the latter’s construction. Wavelength limits are expressed in nm. Sources (old)

Lower limit

Upper limit

Sources (new)

Lower limit

Upper limit

Arvesen et al. (1969) ASTM (2000) Burlov-Vasiljev et al. (1995, 1998) Colina et al. (1996) Kitt Peak (Kurucz et al., 1984) Lockwood et al. (1992) MODTRAN-cebchkur (Berk et al., 1999a) MODTRAN-chkur (Berk et al., 1999a) MODTRAN-newkur (Berk et al., 1999a) MODTRAN-oldkur (Berk et al., 1999a) MODTRAN-thkur (Berk et al., 1999a) Neckel and Labs (1984) SOLSPEC-ATLAS1 (Thuillier et al., 2003) Thuillier et al. (2003) UARS-ATLAS2 (Brueckner et al., 1993)

300 119 310 119 296 329 0

2495 Inf. 1070 410 1300 850 Inf.

Bernhard et al. (2006) Bolsée et al. (2014) Dobber(2008) Gueymard (2004) Gurlit et al. (2005) KNMI (Dobber et al., 2008) Meftah et al. (2017a, 2017b)

275 16 202 0 316 250 199

630 2902 600 Inf. 652 520 3000

0

Inf.

Menang et al. (2013)

1000

2500

0

Inf.

Neckel (2003)

330

1099

0

Inf.

Pfeilsticker (2006)

317

653

0

Inf.

PMOD (Egli et al., 2012; Gröbner, 2016; Gröbner et al., 2017)

200

1190

330 0

1250 2398

200 240

1001 2412

200 115

2400 420

SAO (Chance and Kurucz, 2010) SORCE-SIMa (Harder et al., 2010); http://lasp.colorado.edu/lisird/data/sorce_ssi_ l3/) SOLSTICE-ATLAS1 (Woods et al., 1996) SUSIM-ATLAS2 (Andrews and VanHoosier, 1996) (http://wwwsolar.nrl.navy.mil/ susim_atlas_data.html) WHI (Woods et al., 2009)

119 119

410 410

0

2400

a

Average 2003–2015.

measurements have been conducted during the last two decades, using various observational platforms. In an effort to improve accuracy, all this justifies a revision of the ETS that was proposed in (ASTM, 2000) and G04.

constant (i.e., the summation of the spectral irradiance over all wavelengths), whose determination varied between a minimum of 1322 W m−2 and a maximum of 1429.5 W m−2 during that period. Just like the solar constant, all ETS determinations refer to the average sunearth distance (1 ua)2, and are thus directly comparable. Not all spectra are obtained for the exact same intensity of solar activity, however, which causes variance between measurements undertaken over many decades. In 2000, the American Society for Testing and Materials (now known as ASTM International) promulgated Standard E490-00 (ASTM, 2000), which defines a reference spectrum between 119.5 nm and 1000 µm. E490 was reapproved in 2014. The solar constant associated with it is 1366.1 W m−2. In contrast, ISO Standard 21348 (ISO, 2007), reapproved in 2015, promulgates a similar solar constant (1366 W m−2) but does not specify a particular ET spectrum. Rather, it defines the method (experimental or otherwise) that should be followed to define a reference solar spectrum. Hence, it is not a normative standard but a process-based standard. Overall, five different types of solar irradiance products are specifically defined in it. The development below corresponds to “Type 2”, i.e., the process of preparing a reference spectrum product that is derived from multiple measured or modeled data sets. In a previous contribution (Gueymard, 2004), hereafter G04, this author prepared a “Type 2” reference spectrum covering the range 0.5–1000 µm, with a corresponding SC of 1366.1 W m−2. The latter confirmed the ASTM value—as expected, since based on similar datasets. The G04 study, however, underlined the discrepancy between the SC value just mentioned and total solar irradiance measurements made with a new type of radiometer. Based on a 42-year time series of recalibrated irradiance data, a revised SC of 1361.1 W m−2 was recently proposed (Gueymard, 2018), thus confirming other recent determinations that were close to 1361 W m−2 (Gueymard, 2012; Kopp and Lean, 2011). (This SC value has an estimated standard uncertainty of 0.5 W m−2 (or 0.037%), which is far less than the uncertainty of current spectral measurements in space.) In addition, new spectral irradiance

2. Solar spectrum revision: new sources of data Many ETS distributions have been proposed in the literature, based on models (Berk et al., 1999a) or various types of observations and instruments. The observational platforms include terrestrial observatories (Bolsée et al., 2014; Burlov-Vasiljev et al., 1995, 1998; Chance and Kurucz, 2010; Kurucz et al., 1984; Menang et al., 2013; Neckel, 2003), aircraft (Arvesen et al., 1969), stratospheric balloons (Gurlit et al., 2005; Pfeilsticker, 2006), and satellites. In the latter category, the data sources include reference spectra from various instruments: SOLSTICE (ATLAS-1 mission and SIRS-WHI reference spectrum; Woods et al., 1996, 2009), SUSIM (ATLAS-2 mission; Andrews and VanHoosier, 1996), SOLSPEC (Meftah et al., 2017a, 2017b; Thuillier et al., 2003), and SORCE-SIM (Harder et al., 2010). In the latter case, a single spectrum has been derived here by simply calculating an average from the 2003–2015 time series available at http://lasp.colorado.edu/ lisird/data/sorce_ssi_l3/. This averaging is intended to smooth out temporal variations due to solar activity. Additionally, composite or convolved spectra are also used (ASTM, 2000; Bernhard et al., 2006; Dobber et al., 2008; Gröbner et al., 2017; Gueymard, 2004). All sources of data used here consist of a single spectrum, as developed by each author; in parallel, the SORCE-SIM average spectrum is specific to this study. All synthetic spectra that are based, at least partially, on actual measurements rely on a variety of data sources that must be appropriately selected to complement each other. For instance, the current ASTM standard spectrum is based on six data sources between 119.5 nm and 1000 µm. In the case of the G04 synthetic spectrum, 23 sources of ETS data were used in its construction. Eight of them have not been considered for the present study, due to their lower relevance (older and low-resolution data, or arguable surface-based extrapolations). In contrast, 16 new sources have been added, based on a review of the recent literature. Including the G04 synthetic spectrum, a new total of 31 different data sources is reached—most of them typically with a

2 The abbreviation to be used for the astronomical unit is somewhat confusing: ISO stipulates “ua” whereas the International Astronomical Union (IAU) favors “au”, and ASTM reports it as “AU”. The ISO nomenclature is used here.

435

Solar Energy 169 (2018) 434–440

C.A. Gueymard

75th percentiles, respectively, and IQR is the interquartile range, equal to Q75–Q25. After this second group of outliers is eliminated, the mean and standard deviation of all irradiance values from the remaining data sources is calculated for each wavelength. Any spectral source that falls beyond ± 1 standard deviation of the ensemble is also eliminated for that wavelength. The mean is recalculated, which provides the preliminary spectral irradiance at the wavelength under scrutiny. The whole process is illustrated in Fig. 3 for a wavelength of 550 nm. This specific wavelength has particular significance in many fields related to the atmospheric sciences (including climate and solar applications) because it is the most common wavelength used to report the aerosol optical depth obtained from spaceborne remote-sensing observations or atmospheric chemistry models. For that particular wavelength, no obvious outlier was detected visually, but the Lockwood ETS was found an outlier after application of the interquartile range filter. Seven additional ETS sources (ASTM, MODTRAN-cebchkur, SAO, SOLSPECATLAS1, SORCE-SIM, Thuillier, and WHI) failed the mean ± 1 standard deviation criterion and were eliminated. Finally, the 15 surviving ETS sources resulted in a mean of 1.8936 ± 0.0086 W m−2 nm−1. In a final step, scaling coefficients are applied over the main bands identified in G04: 0–200 nm, 200–280 nm, 280–4000 nm, and 4000–∞. The spectrum below 200 nm is not analyzed here, and is thus assumed the same as in the G04 synthetic spectrum, so no scaling is applied to that band. Since its band irradiance amounts to only 0.11 W m−2, the absolute impact of any correction would be small anyway. The irradiance over the 200–280-nm band totals 6.77 W m−2, or 3.2% less than what was found in G04, hence no additional downward scaling is applied. The spectral values above 4000 nm (totaling 10.96 W m−2) are taken from G04 and simply scaled by 1361.1/1366.1 = 0.99634 to accommodate for the revised solar constant value. Finally, a scaling factor of 0.99362 is uniformly applied to all spectral values between 280 and 4000 nm so that the whole spectral summation (0–∞) exactly matches the new solar constant value (1361.1 W m−2) from Gueymard (2018). The whole process just described reflects the fact that the solar constant’s magnitude (based on high-precision broadband absolute radiometers) is much more reliable than that of current spectral measurements. Consequently, the “final scaling” approach is typically used to construct reference spectra and make them consistent with a specific value of the solar constant. (See, e.g., ASTM (2000), Wehrli (1985), or http://rredc.nrel.gov/solar/spectra/am0/.) Table 2 provides a summary of the spectral step and band irradiance corresponding to each of nine important wavebands. For each waveband, the irradiance is obtained using the trapezoidal rule. The differences compared to G04 are also indicated, showing that the largest changes occur over the 280–400-nm (−1.61 W m−2) and 400–700-nm (−2.84 W m−2) wavebands. The resulting ETS is provided in the Supplementary Material for

limited spectral range only, however. A description of these sources is provided in Table 1. The relatively large number of data sources evaluated here allows the development of a “consensus” spectrum, following the methodology detailed in the next section. 3. Methodology

-2

-1

ETS irradiance (W m nm )

It is clear from the previous section that the database assembled here has a large diversity in terms of both method of derivation and spectral range. Moreover, the spectral resolution often varies between different wavebands (e.g., higher resolution in the UV and lower resolution in the IR). Finally, the spectral resolution of the primary spectra also varies widely between them. To make them comparable, a preliminary step consists in downscaling them to the G04 wavelength grid, using a triangular smoothing function with a bandwidth equal to the desired final resolution. For all spectra, that final resolution is 0.5 nm between 200 and 400 nm, 1 nm between 400 and 1705 nm, and 5 nm between 1705 and 4000 nm. Most of the ETS sources reviewed here do not provide a clear indication of their uncertainty over various wavebands. This important information would have helped select the data source(s) with, e.g., the lowest uncertainty over a specific waveband. Since it is not desirable to reject ETS data just because of the lack of precise uncertainty information, an alternate method is devised here, based on a combination of outlier rejection and “consensus”. In a first step, the range of interest (200–4000 nm) is divided into 50-nm intervals below 400 nm, and 100nm intervals above. All available ETS values are plotted in each interval. Two examples are shown in Figs. 1 and 2, for 300–350 nm and 900–1000 nm, respectively. (These wavebands correspond to important terrestrial absorption bands, due to ozone and water vapor, respectively.) In the UV (Fig. 1), most outliers are created by the Arvesen, MODTRAN-chkur, MODTRAN-oldkur, and SORCE-SIM spectra. A likely reason for the latter’s behavior is the limited spectral resolution of the radiometer (≈0.3 nm in that waveband). In the near IR (Fig. 2), only two ETS sources produce outliers: Kitt Peak and, to a much lower extent, Arvesen. It is obvious that the Kitt Peak spectrum, which is derived from terrestrial observatory data (Kurucz et al., 1984), is intensely affected by the strong atmospheric water vapor absorption bands that exist around 940 nm. This problem explains why the very high-resolution (≈0.0001 nm) Kitt Peak ETS is frequently scaled to a lower-resolution ETS with higher accuracy, such as G04 (Bernhard et al., 2007; Kiedron et al., 2007). For each wavelength of the grid defined above, all outliers identified as visually “obvious” are removed from further analysis. Then, a second filter is applied, using the conventional detection method that defines outliers as those values outside of the boundaries defined by Q25–1.5•IQR and Q75+1.5•IQR, where Q25 and Q75 are the 25th and

1.3 1.2 1.1 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3

Arvesen MODTRAN-oldkur

MODTRAN-chkur SORCE-SIM

Wavelength (nm) Fig. 1. Comparison of 24 spectra over the 300–350-nm waveband. The four ETS sources producing the most outliers are shown in color. All other sources are shown as grey lines. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.) 436

Solar Energy 169 (2018) 434–440

C.A. Gueymard

-2

-1

ETS irradiance (W m nm )

1 0.9 0.8

Arvesen

0.7 0.6 0.5

Kitt Peak

0.4 0.3

Wavelength (nm) Fig. 2. Comparison of 17 spectra over the 900–1000-nm waveband. The two ETS sources producing the most outliers are shown in color. All other sources are shown as grey lines. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

further reference. For each wavelength, the mean irradiance value, Eλ, is provided, as well as its standard deviation, SD (calculated with respect to all data sources that have survived the outlier tests), and the coefficient of variation, CV = SD/Eλ. CV is large when substantial disagreement exists between all irradiance values that contribute to the mean Eλ, and thus when the latter is potentially less certain. This occurs mostly around 263, 285, 373, 393, and 397 nm, where CV is larger than 10%. Conversely, CV is low when all the data sources tend to agree after their selection process, which is generally the case. However, this is only true to the extent that a fairly large number of sources is used to construct the mean Eλ. Above 2400 nm, this condition is not fulfilled due to the paucity of actual measurements. Hence, neither SD nor CV is reported beyond 2400 nm. More generally, CV should not be regarded as a real uncertainty estimate. 4. Comparisons with older spectra Fig. 4 provides a wavelength-by-wavelength comparison between the new ETS and that proposed in G04 over the UV–VIS (300–700 nm), where the largest relative differences are found. As could be expected, significant structure occurs in the UV due to the 0.5-nm resolution there and the complex impact of downscaling data sources with widely different resolutions. Compared to G04, the new spectrum shows lower values between 380 and 550 nm, and higher values between 550 and 700 nm. This appears to be caused by a number of old data sources that had importance in the 2004 spectrum, but were either completely excluded or found local outliers in the new spectrum. Preliminary tests have been undertaken to compare ground-based spectroradiometric measurements of direct spectral irradiance in the UV–VIS to modeled irradiance estimates based on either G04 or the new spectrum for the ETS. These tests suggest that the use of the new ETS results in more accurate spectral simulations and aerosol optical depth retrievals, as will be described in forthcoming reports. Similarly to Fig. 4, but considering an extended waveband, 300–2400 nm, Fig. 5 shows the percent difference between the new ETS and either G04 or ASTM (2000). Substantial differences with the latter appear in various parts of this large waveband. Some inadequacies of the ASTM ETS were already discussed in G04. The present study confirms these, which suggests that the reference ETS in the ASTM E490 standard (now outdated) should be revised. Finally, Fig. 6 provides a comparison between the 13 source spectra used in the 1600–1800-nm waveband and the proposed ETS. The latter (blue curve) stands roughly halfway between the Thuillier and Bolsée spectra. This is an important finding in the context of the recent debate initiated by (Bolsée et al., 2014) and continued by (Weber, 2015) about the apparent discrepancies between various ETS determinations beyond 1600 nm, and about the possibly incorrect absolute calibration of the

Fig. 3. Analysis of 23 sources of ETS data at 550 nm. The Lockwood ETS is rejected based on the interquartile range filter. Seven additional ETS sources (orange triangles) are rejected using the mean ± 1 standard deviation filter. The 15-point resulting average ETS value (after outlier filtering but before final scaling) is indicated by the blue horizontal line. Its standard deviation range is indicated by the yellow rectangle. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.) Table 2 Irradiance in nine wavebands and the whole spectrum, compared to previous results from G04. The spectral step used for each waveband is also indicated. Spectral range (nm)

Spectral step (nm)

Band irradiance (W m−2)

Difference vs. G04 (W m−2)

0–200 200–280 280–400 400–700 700–1000 1000–1705 1705–2390 2390–4000 4000–∞ ALL

1.0 0.5 0.5 1.0 1.0 1.0 5.0 5.0 Variable Variable

0.11 6.77 102.15 531.80 307.78 283.89 77.95 39.68 10.96 1361.1

– −0.23 −1.61 −2.84 −0.80 +0.21 −0.15 +0.41 −0.04 −5.0

437

Solar Energy 169 (2018) 434–440

C.A. Gueymard

Fig. 4. Extraterrestrial irradiance at 1 ua based on the present methodology (top panel), and its percent difference against the G04 spectrum (bottom panel). The yellow shade indicates a maximum difference of ± 5%. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Extraterrestrial Irradiance 2.0

This work

-2

-1

ETS Irradiance (W m nm )

2.4

1.6 1.2 0.8 0.4

10 5 0 -5 -10 -15

Difference vs. G04 (%)

20 15

Wavelength (nm)

ETS Irradiance Differences (%)

20 15

Thuillier spectrum there. This finding may prove important for a better determination of the minimum opacity of the solar photosphere, which occurs near 1600 nm (Bolsée et al., 2014; Weber, 2015). The Bolsée spectral irradiance is 0.23431 ± 0.00129 W m−2 nm−1 at 1600 nm. The value obtained here is 0.24797 ± 0.00217 W m−2 nm−1. The two determinations are ≈6% apart and do not agree within 1σ, which suggests that further research is required to improve the ETS accuracy in the IR.

Extraterrestrial Irradiance Differences

10 5 0 -5 -10

Difference vs. G04 (%) Difference vs. ASTM %)

-15

5. Conclusion

-20

In this investigation, a thorough revision of a previous composite solar extraterrestrial spectrum (Gueymard, 2004; G04) has been undertaken to take new sources of irradiance data into consideration. The present study focused on the 200–4000 nm waveband, where 99.2% of the sun’s energy is concentrated. To construct the new spectrum, 31 sources of spectral data covering various spectral ranges have been selected, including observations from surface observatories, aircraft, high-altitude balloons, satellites, as well as modeled estimates of the

Wavelength (nm)

Fig. 5. Percent difference between the present spectrum and both the G04 and ASTM (2000) spectra. The yellow shade indicates a maximum difference of ± 5%. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 6. Analysis of 13 sources of ETS data between 1600 and 2400 nm. The proposed spectrum (blue curve) is roughly halfway between the Bolsée and Thuillier spectra. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.) 438

Solar Energy 169 (2018) 434–440

C.A. Gueymard

solar flux. All these spectra have been downscaled to a single, relatively low spectral resolution to allow their direct comparison. Three successive filtering steps have been devised to eliminate outliers at each of the 2165 wavelengths in the interval 200–4000 nm. The mean irradiance at each wavelength was then obtained by averaging the values of those data points that passed the outlier filtering steps. For completeness, the G04 extraterrestrial spectrum was used below 200 nm and above 4000 nm, thus resulting in a final spectral range from 0.5 nm to near-infinity. In a last step, the irradiance values were properly scaled so as to integrate to the revised solar constant value of 1361.1 W m−2 that was recently proposed. It is found that this new composite differs by up to a few percent compared to the earlier G04 spectrum, mostly in the UV and visible spectral bands. Larger differences are found compared to the outdated ASTM E490 standard extraterrestrial spectrum, whose revision is thus recommended. The new spectrum might also help resolve the current debate about the absolute irradiance value around 1600 nm and beyond, which is of importance in solar physics. It is anticipated that the present spectrum can be used in a large variety of applications that do not demand a high spectral resolution. For those that do, it is possible to modulate it with one of a few existing higher-resolution spectra, whose absolute accuracies are typically lower. Further research, currently underway, is necessary to evaluate the potential improvement that can be expected from the new spectrum in the modeling of surface spectral irradiance. The complete extraterrestrial spectrum is available in the Supplementary Material.

Bolsée, D., Pereira, N., Decuyper, W., Gillotay, D., Yu, H., Sperfeld, P., Pape, S., Cuevas, E., Redondas, A., Hernandez, Y., Weber, M., 2014. Accurate determination of the TOA solar spectral NIR irradiance using a primary standard source and the BouguerLangley technique. Sol. Phys. 289, 2433–2457. Brueckner, G.E., Edlow, K.L., Floyd, L.E., Lean, J.L., VanHoosier, M.E., 1993. The solar ultraviolet spectral irradiance monitor (SUSIM) experiment on board the upper atmosphere research satellite (UARS). J. Geophys. Res. 98D, 10695–10711. Burlov-Vasiljev, K.A., Gurtovenko, E.A., Matvejev, Y.B., 1995. New absolute measurements of the solar spectrum 310–685 nm. Sol. Phys. 157, 51. Burlov-Vasiljev, K.A., Matvejev, Y.B., Lasiljeva, I.E., 1998. New measurements of the solar disk-center spectral intensity in the near IR from 645 nm to 1070 nm. Sol. Phys. 177, 25–40. Chance, K., Kurucz, R.L., 2010. An improved high-resolution solar reference spectrum for earth’s atmosphere measurements in the ultraviolet, visible, and near infrared. J. Quant. Spectros. Rad. Transf. 111, 1289–1295. Colina, L., Bohlin, R.C., Castelli, F., 1996. The 0.12–2.5 µm absolute flux distribution of the sun for comparison with solar analog stars. Astron. J. 112, 307–315. Dobber, M., 2008. Solar reference spectrum, Pers. comm. Dobber, M., Voors, R., Dirksen, R., Kleipool, Q., Levelt, P., 2008. The high-resolution solar reference spectrum between 250 and 550 nm and its application to measurements with the ozone monitoring instrument. Sol. Phys. 249, 281–291. Egli, L., Gröbner, J., Shapiro, A., 2012. Development of a new high resolution extraterrestrial spectrum, in: PMOD/WRC, Annual report 2012, Davos, Switzerland. Emde, C., Buras-Schnell, R., Kylling, A., Mayer, B., Gasteiger, J., Hamann, U., Kylling, J., Richter, B., Pause, C., Dowling, T., Bugliaro, L., 2016. The libRadtran software package for radiative transfer calculations (version 2.0.1). Geosci. Model Dev. 9, 1647–1672. Gröbner, J., 2016. Extraterrestrial spectrum, Pers. comm. Gröbner, J., Kröger, I., Egli, L., Hülsen, G., Riechelmann, S., Sperfeld, P., 2017. The highresolution extraterrestrial solar spectrum (QASUMEFTS) determined from groundbased solar irradiance measurements. Atmos. Meas. Tech. 10, 3375–3383. Gueymard, C.A., 1995. SMARTS2, Simple Model of the Atmospheric Radiative Transfer of Sunshine: Algorithms and performance assessment. Rep. FSEC-PF-270-95. Florida Solar Energy Center, Cocoa, FL. Gueymard, C.A., 2001. Parameterized transmittance model for direct beam and circumsolar spectral irradiance. Sol. Energy 71, 325–346. Gueymard, C.A., 2004. The sun’s total and spectral irradiance for solar energy applications and solar radiation models. Sol. Energy 76, 423–452. Gueymard, C.A., 2006. Reference solar spectra: their evolution, standardization issues, and comparison to recent measurements. Adv. Space Res. 37, 323–340. Gueymard, C.A., 2012. Solar radiation, introduction. In: Meyers, R.A. (Ed.), Encyclopedia of Sustainability Science and Technology. Springer. Gueymard, C.A., 2018. A reevaluation of the solar constant based on a 42-year total solar irradiance time series and a reconciliation of spaceborne observations. Sol. Energy. http://dx.doi.org/10.1016/j.solener.2018.04.001. in press. Gurlit, W., Bösch, H., Bovensmann, H., Burrows, J.P., Butz, A., Camy-Peyret, C., Dorf, M., Gerilowski, K., Lindner, A., Noël, S., Platt, U., Weidner, F., Pfeilsticker, K., 2005. The UV-A and visible solar irradiance spectrum: inter-comparison of absolutely calibrated, spectrally medium resolution solar irradiance spectra from balloon- and satellite-borne measurements. Atmos. Chem. Phys. 5, 1879–1890. Harder, J.W., Thuillier, G., Richard, E.C., Brown, S.W., Lykke, K.R., Snow, M., McClintock, W.E., Fontenla, J.M., Woods, T.N., Pilewskie, P., 2010. The SORCE SIM solar spectrum: comparison with recent observations. Sol. Phys. 263, 3–24. http://dx. doi.org/10.1007/s11207-010-9555-y. IEC, 2016. Photovoltaic devices – Part 3: Measurement principles for terrestrial photovoltaic (PV) solar devices with reference spectral irradiance data. Standard IEC 60904–3:2016. International Electrotechnical Commission, Geneva, Switzerland. ISO, 2007. Space environment (natural and artificial) — Process for determining solar irradiances. ISO 21348 standard. International Organization for Standardization. Kiedron, P.W., Beauharnois, M., Berndt, J., Disterhoft, P., Harrison, L., Michalsky, J.J., Scott, G., Schlemmer, J., Slusser, J., 2007. Calibration, data processing, and maintenance of the United States Department of Agriculture high-resolution ultraviolet spectroradiometers. Opt. Eng. 46, 086201. Kopp, G., Lean, J.L., 2011. A new, lower value of total solar irradiance: evidence and climate significance. Geophys. Res. Lett. 38. http://dx.doi.org/10.1029/ 2010GL045777. Kurucz, R.L., Furenlid, I., Brault, J., Testerman, L., 1984. Solar flux atlas from 296 to 1300 nm. National Solar Observatory Atlas No.1, NOAO, Sunspot, NM. Lockwood, G.W., Tüg, H., White, N.M., 1992. A new solar irradiance calibration from 3295 Å to 8500 Å derived from absolute spectrophotometry of Vega. Astrophys. J. 390, 668–678. Meftah, M., Damé, L., Bolsée, D., Hauchecorne, A., Pereira, N., Sluse, D., Cessateur, G., Irbah, A., Bureau, J., Weber, M., Bramstedt, K., Hilbig, T., Thiéblemont, R., Marchand, M., Lefèvre, F., Sarkissian, A., Bekki, S., 2017a. SOLAR-ISS: a new reference spectrum based on SOLAR/SOLSPEC observations. Astron. Astrophys. http:// dx.doi.org/10.1051/0004-6361/201731316. Meftah, M., Damé, L., Bolsée, D., Pereira, N., Sluse, D., Cessateur, G., Irbah, A., Sarkissian, A., Djafer, D., Hauchecorne, A., Bekki, S., 2017b. A New Solar Spectrum from 656 to 3088 nm. Sol. Phys. 292. http://dx.doi.org/10.1007/s11207-017-1115-2. Menang, K.P., Coleman, M.D., Gardiner, T.D., Ptashnik, I.V., Shine, K.P., 2013. A highresolution near-infrared extraterrestrial solar spectrum derived from ground-based Fourier transform spectrometer measurements. J. Geophys. Res. 118, 5319–5331. http://dx.doi.org/10.1002/jgrd.50425. Michalsky, J.J., Kiedron, P.W., 2008. Comparison of UV-RSS spectral measurements and TUV model runs for clear skies for the May 2003 ARM aerosol intensive observation period. Atmos. Chem. Phys. 8, 1813–1821.

Acknowledgments The author gratefully thanks Drs. Germar Bernhard, David Bolsée, Marcel Dobber, Mustapha Meftah, Julian Gröbner, Kaah Menang, and Klaus Pfeilsticker for their helpful discussions and direct access to their data. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.solener.2018.04.067. References Andrews, M.D., VanHoosier, M.E., 1996. Calibration of solar ultraviolet spectral irradiance monitor (SUSIM) on ATLAS-2. Metrologia 32, 629–632. Arvesen, J.C., Griffin, R.N., Pearson, B.D., 1969. Determination of extraterrestrial solar spectral irradiance from a research aircraft. Appl. Opt. 8, 2215–2232. ASTM, 2000. Solar Constant and Zero Air Mass Solar Spectral Irradiance Tables. Standard E490. American Society for Testing and Materials, West Conshohocken, PA. < http:// www.astm.org > . ASTM, 2003. Standard Tables for Reference Solar Spectral Irradiances: Direct Normal and Hemispherical on 37° Tilted Surface. Standard G173-03. American Society for Testing and Materials, West Conshohocken, PA. < http://www.astm.org > . ASTM, 2014. Reference Solar Spectral Distributions: Direct and Diffuse on 20° Tilted and Vertical Surfaces. Standard G197–14. ASTM International. Berk, A., Anderson, G.P., Acharya, P.K., Chetwynd, J.H., Bernstein, L.S., Shettle, E.P., Matthew, M.W., Adler-Golden, S.M., 1999a. MODTRAN4 User's Manual. Air Force Research Lab. Berk, A., Anderson, G.P., Bernstein, L.S., Acharya, P.K., Dothe, H., Matthew, M.W., AdlerGolden, S.M., Chetwynd, J.H., Richtsmeier, S.C., Pukall, B., Allred, C.L., Jeong, L.S., Hoke, M.L., 1999b. MODTRAN4 radiative transfer modeling for atmospheric correction. In: Proc. Optical Spectroscopic Techniques and Instrumentation for Atmospheric and Space Research III, SPIE, vol. 3756. Society of Photo-optical Instrumentation Engineers. Bernhard, G., Booth, C.R., Ehramjian, J.C., 2004. Version 2 data of the National Science Foundation’s Ultraviolet radiation monitoring network: south Pole. J. Geophys. Res. 109D21207. http://dx.doi.org/10.1029/2004JD004937. Bernhard, G., Booth, C.R., Ehramjian, J.C., Nichol, S.E., 2006. UV climatology at McMurdo station, Antarctica, based on version 2 data of the National Science Foundation’s ultraviolet radiation monitoring network. J. Geophys. Res. 111D11201. http://dx.doi.org/10.1029/2005JD005857. Bernhard, G., Booth, C.R., Ehramjian, J.C., Stone, R., Dutton, E.G., 2007. Ultraviolet and visible radiation at Barrow, Alaska: climatology and influencing factors on the basis of version 2 National Science Foundation network data. J. Geophys. Res. 112D09101. http://dx.doi.org/10.1029/2006JD007865.

439

Solar Energy 169 (2018) 434–440

C.A. Gueymard

Sol. Phys. 290, 1601–1605. http://dx.doi.org/10.1007/s11207-015-0707-y. Wehrli, C., 1985. Extraterrestrial Solar Spectrum. Publ. no. 615. PhysikalischMeteorologisches Observatorium + World Radiation Center (PMO/WRC), Davos Dorf, Switzerland. < ftp://ftp.pmodwrc.ch/pub/publications/pmod615.asc > . Woods, T.N., Chamberlin, P.C., Harder, J.W., Hock, R.A., Snow, M., Eparvier, F.G., Fontenla, J., McClintock, W.E., Richard, E., 2009. Solar irradiance reference spectra (SIRS) for the 2008 whole heliosphere interval (WHI). Geophys. Res. Lett. 36. http:// dx.doi.org/10.1029/2008GL036373. Woods, T.N., Prinz, D.K., Rottman, G.J., London, J., Crane, P.C., Cebula, R.P., Hilsenrath, E., Brueckner, G.E., Andrews, M.D., White, O.R., VanHoosier, M.E., Floyd, L.E., Herring, L.C., Knapp, B.G., Pankratz, C.K., Reiser, P.A., 1996. Validation of the UARS solar ultraviolet irradiances: comparison with the ATLAS 1 and 2 measurements. J. Geophys. Res. 101D, 9541–9569.

Neckel, H., 2003. On the sun's absolute disk-center and mean disk intensities, its limb darkening, and its 'limb temperature' (330 to 1099 nm). Sol. Phys. 212, 239–250. Neckel, H., Labs, D., 1984. The solar radiation between 3300 and 12500 Å. Sol. Phys. 90, 205–258. Pfeilsticker, K., 2006. Updated solar spectrum, Pers. comm. Thuillier, G., Hersé, M., Simon, P.C., Labs, D., Mandel, H., Gillotay, D., Foujols, T., 2003. The solar spectral irradiance from 200 to 2400 nm as measured by the SOLSPEC spectrometer from the ATLAS 1-2-3 and EURECA missions. Sol. Phys. 214, 1–22. Trishchenko, A.P., 2006. Solar irradiance and effective brightness temperature for SWIR channels of AVHRR/NOAA and GOES imagers. J. Atmos. Ocean. Technol. 23, 198–210. Weber, M., 2015. Comment on the Article by Thuillier et al. “The Infrared solar spectrum measured by the SOLSPEC spectrometer onboard the International Space Station”.

440