<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE article PUBLIC "-//NLM//DTD Journal Publishing with OASIS Tables v3.0 20080202//EN" "journalpub-oasis3.dtd">
<article xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:oasis="http://docs.oasis-open.org/ns/oasis-exchange/table" xml:lang="en" dtd-version="3.0">
  <front>
    <journal-meta><journal-id journal-id-type="publisher">AMT</journal-id><journal-title-group>
    <journal-title>Atmospheric Measurement Techniques</journal-title>
    <abbrev-journal-title abbrev-type="publisher">AMT</abbrev-journal-title><abbrev-journal-title abbrev-type="nlm-ta">Atmos. Meas. Tech.</abbrev-journal-title>
  </journal-title-group><issn pub-type="epub">1867-8548</issn><publisher>
    <publisher-name>Copernicus Publications</publisher-name>
    <publisher-loc>Göttingen, Germany</publisher-loc>
  </publisher></journal-meta>
    <article-meta>
      <article-id pub-id-type="doi">10.5194/amt-11-2501-2018</article-id><title-group><article-title>Evaluation of Himawari-8 surface downwelling solar radiation by ground-based
measurements</article-title><alt-title>Evaluation of Himawari-8 surface downwelling solar radiation</alt-title>
      </title-group><?xmltex \runningtitle{Evaluation of Himawari-8 surface downwelling solar radiation}?><?xmltex \runningauthor{A. Damiani et al.}?>
      <contrib-group>
        <contrib contrib-type="author" corresp="yes" rid="aff1">
          <name><surname>Damiani</surname><given-names>Alessandro</given-names></name>
          <email>alecarlo.damiani@gmail.com</email>
        <ext-link>https://orcid.org/0000-0003-3014-6193</ext-link></contrib>
        <contrib contrib-type="author" corresp="no" rid="aff1">
          <name><surname>Irie</surname><given-names>Hitoshi</given-names></name>
          
        </contrib>
        <contrib contrib-type="author" corresp="no" rid="aff1">
          <name><surname>Horio</surname><given-names>Takashi</given-names></name>
          
        </contrib>
        <contrib contrib-type="author" corresp="no" rid="aff1">
          <name><surname>Takamura</surname><given-names>Tamio</given-names></name>
          
        </contrib>
        <contrib contrib-type="author" corresp="no" rid="aff2">
          <name><surname>Khatri</surname><given-names>Pradeep</given-names></name>
          
        </contrib>
        <contrib contrib-type="author" corresp="no" rid="aff3">
          <name><surname>Takenaka</surname><given-names>Hideaki</given-names></name>
          
        </contrib>
        <contrib contrib-type="author" corresp="no" rid="aff3">
          <name><surname>Nagao</surname><given-names>Takashi</given-names></name>
          
        </contrib>
        <contrib contrib-type="author" corresp="no" rid="aff4">
          <name><surname>Nakajima</surname><given-names>Takashi Y.</given-names></name>
          
        </contrib>
        <contrib contrib-type="author" corresp="no" rid="aff5">
          <name><surname>Cordero</surname><given-names>Raul R.</given-names></name>
          
        </contrib>
        <aff id="aff1"><label>1</label><institution>CEReS, Chiba University, Chiba, 263-8522, Japan</institution>
        </aff>
        <aff id="aff2"><label>2</label><institution>Center for Atmospheric and Oceanic Studies, Tohoku University,
Sendai, 980-8578, Japan</institution>
        </aff>
        <aff id="aff3"><label>3</label><institution>Earth Observation Research Center, JAXA, Tsukuba, 305-0047, Japan</institution>
        </aff>
        <aff id="aff4"><label>4</label><institution>Research and Information Center, Tokai University, Tokyo, 151-0063,
Japan</institution>
        </aff>
        <aff id="aff5"><label>5</label><institution>Department of Physics, Santiago University, Santiago de Chile,
8320000, Chile</institution>
        </aff>
      </contrib-group>
      <author-notes><corresp id="corr1">Alessandro Damiani (alecarlo.damiani@gmail.com)</corresp></author-notes><pub-date><day>27</day><month>April</month><year>2018</year></pub-date>
      
      <volume>11</volume>
      <issue>4</issue>
      <fpage>2501</fpage><lpage>2521</lpage>
      <history>
        <date date-type="received"><day>6</day><month>December</month><year>2017</year></date>
           <date date-type="rev-request"><day>2</day><month>January</month><year>2018</year></date>
           <date date-type="rev-recd"><day>21</day><month>March</month><year>2018</year></date>
           <date date-type="accepted"><day>6</day><month>April</month><year>2018</year></date>
      </history>
      <permissions>
        <copyright-statement>Copyright: © 2018 Alessandro Damiani et al.</copyright-statement>
        <copyright-year>2018</copyright-year>
      <license license-type="open-access"><license-p>This work is licensed under the Creative Commons Attribution 4.0 International License. To view a copy of this licence, visit <ext-link ext-link-type="uri" xlink:href="https://creativecommons.org/licenses/by/4.0/">https://creativecommons.org/licenses/by/4.0/</ext-link></license-p></license></permissions><self-uri xlink:href="https://amt.copernicus.org/articles/11/2501/2018/amt-11-2501-2018.html">This article is available from https://amt.copernicus.org/articles/11/2501/2018/amt-11-2501-2018.html</self-uri><self-uri xlink:href="https://amt.copernicus.org/articles/11/2501/2018/amt-11-2501-2018.pdf">The full text article is available as a PDF file from https://amt.copernicus.org/articles/11/2501/2018/amt-11-2501-2018.pdf</self-uri>
      <abstract><title>Abstract</title>
    <p id="d1e181">Observations from the new Japanese
geostationary satellite Himawari-8 permit quasi-real-time estimation of
global shortwave radiation at an unprecedented temporal resolution. However,
accurate comparisons with ground-truthing observations are essential to
assess their uncertainty. In this study, we evaluated the Himawari-8 global
radiation product AMATERASS using observations recorded at four SKYNET
stations in Japan and, for certain analyses, from the surface network of the
Japanese Meteorological Agency in 2016. We found that the spatiotemporal
variability of the satellite estimates was smaller than that of the ground
observations; variability decreased with increases in the time step and
spatial domain. Cloud variability was the main source of uncertainty in the
satellite radiation estimates, followed by direct effects caused by aerosols
and bright albedo. Under all-sky conditions, good agreement was found between
satellite and ground-based data, with a mean bias in the range of
20–30 W m<inline-formula><mml:math id="M1" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> (i.e., AMATERASS overestimated ground observations) and a
root mean square error (RMSE) of approximately 70–80 W m<inline-formula><mml:math id="M2" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula>. However,
results depended on the time step used in the validation exercise, on the
spatial domain, and on the different climatological regions. In particular,
the validation performed at 2.5 min showed largest deviations and RMSE
values ranging from about 110 W m<inline-formula><mml:math id="M3" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> for the mainland to a maximum of
150 W m<inline-formula><mml:math id="M4" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> in the subtropical region. We also detected a limited
overestimation in the number of clear-sky episodes, particularly at the pixel
level. Overall, satellite-based estimates were higher under overcast
conditions, whereas frequent episodes of cloud-induced enhanced surface
radiation (i.e., measured radiation was greater than expected clear-sky
radiation) tended to reduce this difference. Finally, the total mean bias was
approximately 10–15 W m<inline-formula><mml:math id="M5" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> under clear-sky conditions, mainly because
of overall instantaneous direct aerosol forcing efficiency in the range of
120–150 W m<inline-formula><mml:math id="M6" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> per unit of aerosol optical depth (AOD). A seasonal
anticorrelation between AOD and global radiation differences was evident at
all stations and was also observed within the diurnal cycle.</p>
  </abstract>
    </article-meta>
  </front>
<body>
      

<sec id="Ch1.S1" sec-type="intro">
  <title>Introduction</title>
      <p id="d1e264">Following the Tōhoku earthquake of March 2011, governmental policy in
Japan stimulated a broader use of renewable energy sources. However, full
integration of the natural energy resource is still prevented by its
variability. In particular, the instability of solar power caused by
cloudiness is a critical issue for suppliers. One way to overcome this
variability is to develop an appropriate control system for electric power
generated by solar radiation and dispensed into the power grid system, i.e.,
an energy management system (EMS). However, to implement such an EMS system,
surface solar irradiance data must be supplied as accurately as possible.</p>
      <p id="d1e267">Although ground instruments provide the most reliable monitoring results,
they demand special maintenance and are<?pagebreak page2502?> usually sparse. By contrast,
satellites can be easily used to characterize large regions, albeit with
limited temporal and spatial resolution. To date, satellite-based estimates
of surface global solar radiation are routinely performed by different
algorithms which exploit the images recorded by geostationary satellites,
such as the Meteosat Second Generation (MSG) and the Geostationary
Operational Environmental Satellites (GOES), to relate satellite reflectance
to cloud opacity. Among the available algorithms, it is worth mentioning
the Heliosat-2 method (Raschke et al., 1987; Rigollier et al.,
2004; Blanc et al., 2011), the State University of New York (SUNY) model
(Perez et al., 2002), the Surface Insolation under Clear and Cloudy skies
model (SICCS; Greuell et al., 2013), and more sophisticated radiative
transfer models (e.g., Pinker and Laszlo, 1992) and algorithms based on
neural network models (e.g., Takenaka et al., 2011; Taylor et al., 2016).
Because of its unprecedented temporal resolution (up to 2.5 min), the new
Japanese geostationary meteorological satellite Himawari-8 is expected
shortly to attain a key position within this framework, drastically improving
atmospheric research (Bessho et al., 2016) and forecasts of photovoltaic
power generation (e.g., Ohtake et al., 2013, 2015).</p>
      <p id="d1e270">Because punctual ground observations can only partially represent pixel area,
depending on weather (Federico et al., 2017), topography (Gómez et al.,
2016), and surface albedo (Damiani et al., 2012, 2013) conditions, accurate
and extensive comparisons with ground-truthing observations are essential to
evaluating the accuracy of satellite-based estimates of solar radiation. This
validation is performed by ground-based pyranometers usually on an hourly
basis (e.g., Nottrott and Kleissi, 2010; Djebbar et al., 2012; Greuell et
al., 2013; Federico et al., 2017). For example, Federico et al. (2017)
recently evaluated estimates of surface solar irradiance over Italy based on
the observations of MSG's Spinning Enhanced Visible and Infrared Imager
(SEVIRI) and the SICCS model. They found that the root mean square error
(RMSE) ranges between about 70 and 180 W m<inline-formula><mml:math id="M7" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> depending on the sky
conditions. The study of Thomas et al. (2016a) highlighted that the
HelioClim-3 database, derived from images acquired by the MSG, showed a large
correlation coefficient (0.96) with the ground stations over Brazil. However,
it tends to overestimate the observed surface solar radiation with most
values comprised between 1 and 5 %. For the same database, Thomas et
al. (2016b) showed a correlation ranging from <inline-formula><mml:math id="M8" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 0.94 to 0.98 and a
relative RMSE between <inline-formula><mml:math id="M9" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 11 and 30 % with respect to some stations
in Europe. The National Solar Radiation Database, based on visible images
from GOES and the SUNY model, showed an average overprediction of five
percent when compared with surface stations in California due to its tendency
to overestimate solar irradiance under cloudy conditions (Nottrott and
Kleissi, 2010). A following evaluation of the same algorithm using GOES
images over Canada showed an hourly mean bias (MB) error of 5.6 W m<inline-formula><mml:math id="M10" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula>
and a RMSE of 86.5 W m<inline-formula><mml:math id="M11" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> when results were averaged over 18 stations
(Djebbar et al., 2012).</p>
      <p id="d1e323">In the recent literature, only a few validation exercises dealt with data sets
at high temporal resolution (at 15 min, e.g., Ruf et al., 2016; Zo et al.,
2016; Qu et al., 2017; Kosmopoulos et al., 2018). Zo et al. (2016) compared a
satellite-based data set with observational data from 22 solar sites of the
Korean Meteorological Administration (KMA) over Korea and found an overall
correlation coefficient of about 0.95. Satellite-based estimates at 15 min
retrieved by the new Heliosat-4 method (Qu et al., 2017) were recently
validated with observations from stations (mainly in Europe) of the Baseline
Surface Radiation Network (BSRN). In this study, an overestimation of the
surface solar radiation in the range 2–32 W m<inline-formula><mml:math id="M12" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> (with the lower
values over desert or semi-arid locations) and a RMSE between 74 and
94 W m<inline-formula><mml:math id="M13" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> were reported. A somewhat larger RMSE (115.9 W m<inline-formula><mml:math id="M14" display="inline"><mml:mrow><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup><mml:mo>)</mml:mo></mml:mrow></mml:math></inline-formula>
for the same data set has been shown for a single station in Germany (Ruf et
al., 2016). For latitudes higher than 48<inline-formula><mml:math id="M15" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula>, the correlation
coefficient was found in the range of 0.90–0.94, while for lower latitudes
the range was 0.95–0.96; this pointed to difficulty in the cloud
discrimination at high satellite viewing angles. Moreover, Qu et al. (2017)
showed a positive correlation between cloud fraction and RMSE, the extent of
which
was likely affected by the specific climatological features at the different
stations. Overall, in Europe the validation results were better for the
summer months than for winter.</p>
      <p id="d1e375">Although this validation is essential, there is also an increasing need to
evaluate performance at even higher temporal and spatial resolutions (Perez
et al., 2016), because rapid transients can be harmful to photovoltaic
installations. Previous studies have found that the correlation between
sites decreases with distance and increases with temporal aggregation
(Engeland et al., 2017). Although time series of spatially averaged
irradiance fields generally resemble behavior point measurements, their
power spectra are strongly attenuated at higher frequencies and for large
domains (Madhavan et al., 2017). Therefore, variation in spatial averages
(or satellite pixels) and in point measurements is often poorly correlated
at high frequencies (Perez et al., 2016). Indeed, a single pyranometer is
usually representative of the overhead cloud structure only for time
resolutions larger than 1 h; at higher frequencies, sampling within the
satellite pixel helps to reduce discrepancies with remotely sensed estimates
(Nunez et al., 2013).</p>
      <p id="d1e378">Overall, point measurements can deviate strongly from the spatial mean of a
surrounding domain, and satellite–ground comparisons performed at medium–long
timescales tend to smooth variability in the radiation, thus hiding
important issues. One of these is cloud-induced radiation enhancement (RE),
which results in a surface radiation larger than the expected (simulated)
clear-sky radiation (Gueymard et al., 2017). These events usually have only a
small spatial footprint and, although quite frequent under broken cloud
conditions, cannot be reproduced by satellite algorithms based on homogeneous
parallel-plane and single-layer cloud models and may lead to problems in
photovoltaic power stations.<?pagebreak page2503?> Indeed, earth-observing satellites cannot fully
account for cloud inhomogeneity and this points to the necessity of a more
extensive implementation of 3-D radiative transfer in the atmosphere with
inhomogeneous clouds (Okamura et al., 2017; Iwabuchi, 2006). Even whether such
issues are masked by the broad time resolution used in validation exercises,
as we will see, REs can introduce bias into the evaluation of satellite
estimates.</p>
      <p id="d1e381">Many additional factors can introduce a bias in the results of a validation
exercise and make satellite-based estimates more uncertain (Polo et al.,
2016). Among others, it is worth mentioning the important role
played by aerosols in reducing the surface solar irradiance; thus this is an
important parameter to be accounted in the algorithms, especially in polluted
regions or deserts under clear-sky conditions (e.g., Qu et al., 2017). It is
worth mentioning the negative effect of the complex morphology in mountainous
regions which cannot be easily accounted for due to the limited satellite
spatial resolution and the local fast-changing weather conditions (Dürr
et al., 2010; Urraca et al., 2017; Federico et al., 2017). Further, the
negative impact of the high solar zenith angle (SZA) and satellite-viewing
zenith angle on the quality of the estimates must be also mentioned: the
former can cause low clouds to be overshadowed by higher clouds while the
latter leads to parallax effect in the clouds position (Polo et al., 2016; Qu
et al., 2017). Finally, uncertainties in the surface albedo, especially over
bright surfaces (e.g., snow or desert), can make cloud identification hard
(e.g., Tanskanen et al., 2007).</p>
      <p id="d1e384">In this study, we used ground-based observations to evaluate surface
downwelling solar radiation estimates at high temporal resolution made by
applying the EXtreme speed and Approximation Module (EXAM) multiple drive system
algorithm (Takenaka et al., 2011) to Himawari-8 observations.
Following an initial assessment of the spatiotemporal variability of the
radiation fields, we focused on effects caused by cloudiness, aerosols, and
surface albedo. Although cloud-induced variability was expected to be the
main source of uncertainty in satellite estimates of solar radiation, issues
with albedo and aerosols caused further uncertainty under bright albedo and
clear-sky conditions.</p>
</sec>
<sec id="Ch1.S2">
  <title>Data and methods</title>
<sec id="Ch1.S2.SS1">
  <title>Himawari-8 estimates of surface downwelling solar radiation</title>
      <p id="d1e398">The Advanced Himawari Imagers (AHIs) aboard Himawari-8 acquire full-disk
observations in 16 observation bands (three for visible, three for near-infrared, and 10 for infrared wavelengths) every 10 min (and over Japan
every 2.5 min), with a spatial resolution ranging from 0.5 to 2 km (Bessho
et al., 2016). These observations allow quasi-real-time estimation of surface
downwelling global shortwave (SW) radiation over Japan at a temporal
resolution of 2.5 min and a nominal spatial resolution of 1 km using the
EXAM algorithm (Takenaka et al., 2011). The algorithm is based on a
fast neural network, accurately reproducing the radiative transfer model.
This satellite-based radiation data set, AMATERASS solar radiation, uses the
Comprehensive Analysis Program for Cloud Optical Measurement (CAPCOM;
Nakajima and Nakajima, 1995; Kawamoto et al. 2001) algorithm to retrieve
cloud optical thickness and cloud-particle effective radius from Himawari-8
observations by a lookup table (LUT) based approach under a homogeneous
plane-parallel and single-layer cloud model. Additional input information
included in EXAM, such as water vapor and ozone, was acquired from external
data sets (e.g., the Japanese Reanalysis and OMI/Aura satellite), and surface
albedo was computed from Himawari-8 observations using a statistical method.
Although work is ongoing to include aerosol effect in the algorithm, the
current version does not take aerosols into account. Many previous studies
estimated the aerosol impact on surface solar radiation (e.g., Xia et al.,
2007; Cachorro et al., 2008; Di Biagio et al., 2009, 2010; Papadimas et al.,
2012; Huttunen et al., 2014). In particular, under clear-sky conditions, a
daily mean direct aerosol forcing on surface global radiation between <inline-formula><mml:math id="M16" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>8
and <inline-formula><mml:math id="M17" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>64 W m<inline-formula><mml:math id="M18" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> for East Asia (e.g., Table 1 in Kudo et al., 2010) and
in the range <inline-formula><mml:math id="M19" display="inline"><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">8</mml:mn></mml:mrow></mml:math></inline-formula> to <inline-formula><mml:math id="M20" display="inline"><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">23</mml:mn></mml:mrow></mml:math></inline-formula> W m<inline-formula><mml:math id="M21" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> for Japan (in the Kanto region) has
been reported (Kudo et al., 2010). However, under all-sky conditions, it is
generally thought that, depending on relative position and altitude of clouds and
aerosol layer, aerosol effects on global solar radiation are usually small
compared with cloud effects, while their impact on direct solar radiation is
more important (e.g., Qu et al., 2017; Kosmopoulos et al., 2018). Therefore,
under cloudless conditions, AMATERASS estimates could be positively biased,
depending on the actual aerosol load present at each location (Irie et al.,
2017). Note that, in the following analysis, the potential overestimation of
the satellite-based estimates caused by the absence of the aerosol correction
will be investigated under clear-sky conditions, while the study concerning
the RE effects above mentioned, which also affect the bias, will be performed
under all-sky conditions.</p>
      <p id="d1e460">Although assimilation and forecast data sets from the European MACC project or the
Japanese Meteorological Research Institute (MRI/JMA) are suitable sources of operational aerosol optical
properties, aerosol estimations provided by Himawari-8 itself would be the
best products to account for the aerosols effects in a future version of
AMATERASS. Nevertheless, in order to implement such correction, two steps
should be necessary. Firstly, the retrieval of aerosol optical properties
(i.e., aerosol optical thickness and Ångström exponent) must be accomplished
by exploiting Himawari-8 observations. This activity has been initially
performed by using the algorithms of Higurashi and Nakajima (1999) for ocean
and Fukuda et al. (2013) for land and recently further improved in the
retrieval of urban aerosols (Hashimoto and Nakajima, 2017). Preliminary
results are<?pagebreak page2504?> still under evaluation and initial validations with observations
recorded at the Japanese SKYNET stations showed encouraging results.
Therefore, the following step will be the inclusion of Himawari-8 aerosol
parameters into EXAM and the creation of an updated version which will
include the aerosol effects on the solar radiation. As shown in Takenaka et
al. (2011), EXAM was designed to account for aerosols in a neural network for
clear-sky conditions. In the original scheme, three aerosol optical
properties (i.e., aerosol optical
depth (AOD), the imaginary part of the refractive index and the size
distribution) and five additional parameters (i.e., SZA,
surface albedo, surface pressure, ozone, water vapor) were included in the
neural network and the achieved results were satisfactory. A similar approach
will be used in the next version of EXAM.</p>
</sec>
<sec id="Ch1.S2.SS2">
  <title>Surface observations</title>
      <p id="d1e469">We carefully assessed the contribution of cloudiness and aerosols to SW
satellite estimates at four stations in Japan (Fig. 1a) belonging to the
ground-based SKYNET (<uri>http://atmos2.cr.chiba-u.jp/skynet/</uri>) network –
i.e., Chiba (35.625<inline-formula><mml:math id="M22" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> N, 140.104<inline-formula><mml:math id="M23" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> E), Fukue Island
(32.752<inline-formula><mml:math id="M24" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> N, 128.682<inline-formula><mml:math id="M25" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> E), Cape Hedo (26.867<inline-formula><mml:math id="M26" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> N,
128.248<inline-formula><mml:math id="M27" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> E), and Miyako Island (24.737<inline-formula><mml:math id="M28" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> N,
125.327<inline-formula><mml:math id="M29" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> E) – using collocated measurements of surface radiation,
aerosol properties, and total precipitable water (PW) recorded between
January and December 2016. Chiba is an urban site located near Tokyo, and
Cape Hedo, Fukue Island, and Miyako Island are located on relatively small
islands in the East China Sea. There are no major sources of air pollutants
near the latter stations; however, they may be affected by aerosols from the
desert and continental regions in East Asia.</p>
      <p id="d1e548">The stations are equipped with pyranometers and radiometers to measure solar
radiation and cloud and aerosol properties. We used a CM-21 pyranometer (Kipp
&amp; Zonen) to measure global solar (horizontal) irradiance from 285 to
2800 nm. This device was fully compliant with the highest ISO performance
criteria and with the specifications for high-quality instruments as defined
by the World Meteorological Organisation (WMO). It had a combined standard
relative uncertainty of approximately 1.9 % (Kratzenberg et al., 2006).
The temporal resolution of this data set was 10 s.</p>
      <p id="d1e551">Possibly, the reliability of the AMATERASS data set needs to be tested under
various climatological conditions. Therefore, ground-based measurements
should include stations at high altitude, mainland, near the sea, near to
urban sites, etc. Chiba SKYNET station can be considered representative of
urban conditions of the mainland region, while the other SKYNET stations can
be roughly grouped as representative of a subtropical region possibly
affected by desert and continental aerosols. Because of the necessity of
examining other climates and different conditions and to extend our
validation to the whole of Japan, we accomplished further comparisons with
respect to 47 stations in the JMA surface
network of pyranometers, some of them also belonging to the BSRN. The rigorous quality control of the measured data
is regularly performed by JMA. This allowed us to distinguish three main
climatological regions and additional smaller areas presenting a distinct
response.</p>

      <?xmltex \floatpos{t}?><fig id="Ch1.F1"><label>Figure 1</label><caption><p id="d1e556"><bold>(a)</bold> Region of interest showing the four SKYNET stations
used in the validation of Himawari-8 estimates of global radiation. Source:
Google Earth Pro. <bold>(b)</bold> Difference in the coefficient of variation
(CV) of global radiation values obtained by SKYNET (SKY) and Himawari-8 (H-8)
and in the size of the domain for different time averages (instantaneous time
resolution: 2.5 min). Period: August 2016; station: Chiba. See text for
further details.</p></caption>
          <?xmltex \igopts{width=236.157874pt}?><graphic xlink:href="https://amt.copernicus.org/articles/11/2501/2018/amt-11-2501-2018-f01.pdf"/>

        </fig>

      <p id="d1e571">For the SKYNET stations, aerosol optical properties, such as AOD and single-scattering albedo, were retrieved from direct
sun and diffuse sky radiances measured using a sky radiometer (model POM-02, Prede Co. Ltd., Japan). The core retrieval program was
SKYRAD.pack (e.g., Nakajima et al., 1996). In response to recent worldwide
SKYNET activity, we requested the single data set obtained in near real time
using the common algorithm. A new Sky Radiometer analysis program package
from the Center for Environmental Remote Sensing (SR-CEReS, version 1, with
skyrad.pack version 5; Mok et al., 2017) was subsequently developed and used
in this study. We focused on the AOD data set with a temporal resolution of
10 min.</p>
      <p id="d1e574">Information on the total PW at the Chiba station was retrieved by microwave
radiometer (MP1502). These measurements were compared to those of two other
microwave radiometers (WVR-1125, MP1504) during an intensive campaign at
Chiba University in November 2016 and proved to be of reasonable quality. The
temporal resolution of this data set was 1 min. For the other stations, we
used PW estimates from the ECMWF ERA-Interim data set.</p>
</sec>
<sec id="Ch1.S2.SS3">
  <title>Clear-sky screening and clear-sky index</title>
      <p id="d1e583">Screening for clear-sky conditions, which is necessary to evaluate the
influence of aerosols on solar radiation estimates, was conducted in a
two-step process. We first selected<?pagebreak page2505?> only aerosol observations that were
judged to be clear-sky observations by the SKYNET algorithm (Khatri and
Takamura, 2009). From among these observations, we retained radiation
observations within a 10 min time window. Finally, we performed an
additional screening using the standard deviation (SD) of the Himawari-based
global solar irradiance data within a domain area of 20 km around each
station. The SD mirrored the level of cloudiness within the considered area.
We verified experimentally through a sensitivity analysis that an SD
threshold of 10 W m<inline-formula><mml:math id="M30" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> was sufficient to ensure that data were recorded
under clear-sky conditions. For selected days, we further verified the sky
conditions by inspecting images taken using all-sky cameras.</p>
      <p id="d1e598">Some of the following analyses were based on a computation of the so-called
clear-sky index (CSI), here defined as the ratio of recorded irradiance to
the corresponding expected clear-sky and aerosol-free irradiance simulated
under the same conditions (i.e., in theory CSI should range from 0 to 1,
although actual values sometimes exceed 1; see Sect. 3.1). In contrast to
previous studies using empirical formulations (e.g., Piedehierro et al.,
2014), we used an LUT-based approach to simulate clear-sky irradiance that
used simulations run by the LibRadTran–UVSPEC library (Mayer et al., 2005).
We used the DIScrete ORdinate Radiative Transfer (DISORT) solver (Stamnes et
al., 1988) to solve the radiative transfer equations. DISORT was run under
the usual conditions at each station and included observed PW (measured using
a microwave radiometer at Chiba and retrieved by ECMWF ERA-interim at the
other sites); additional parameters (e.g., ozone) remained fixed. In the
following sections, in order to describe the sky conditions, we often refer
to overcast, broken cloud, and clear-sky conditions for low, medium, and high
CSI values, respectively. Although the CSI is a function of both cloud
fraction and cloud optical depth (COD), using the CSI or
clearness index for defining the three main categories of sky conditions is
widely accepted (e.g., Serrano et al., 2006; Bech et al., 2015).</p>
</sec>
<sec id="Ch1.S2.SS4">
  <title>Statistics and comparison</title>
      <p id="d1e607">The comparison of Himawari-8 estimates and ground-based SKYNET observations
was performed at various time resolutions ranging from 2.5 min to 1 day. In contrast, the additional comparison involving JMA measurements was made
at time step of 10 min.</p>
      <p id="d1e610">Consistent with previous validation studies (e.g., Antoìn et al., 2010;
Nottrott and Kleissi, 2012; Damiani et al., 2012, 2013), the statistical
comparison was performed in terms of the correlation coefficient (<inline-formula><mml:math id="M31" display="inline"><mml:mi>r</mml:mi></mml:math></inline-formula>), mean
bias, RMSE, and slope (<inline-formula><mml:math id="M32" display="inline"><mml:mi>a</mml:mi></mml:math></inline-formula>) of the regression
line. In some of the analyses, we compared surface observations with respect
to an average of satellite-based estimates taken within a variable spatial
domain ranging from a single pixel (i.e., 1 km) to an area of 20 km around
each SKYNET station. Unless otherwise noted, we focused on data
characterized by SZAs less than 70<inline-formula><mml:math id="M33" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> to avoid possible cosine response
error in the pyranometer and/or the satellite algorithm and the period from
January to December 2016 (as a reference, at Chiba, under clear-sky
conditions, a global solar radiation of about 300 and 150 W m<inline-formula><mml:math id="M34" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> can be
expected for SZA of 70 and 80<inline-formula><mml:math id="M35" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula>, respectively).</p>

      <?xmltex \floatpos{t}?><fig id="Ch1.F2" specific-use="star"><label>Figure 2</label><caption><p id="d1e659">Scatter plot of ground-based SKYNET observations and satellite-based
estimates of surface global irradiance from Himawari-8 at the Chiba, Fukue,
Miyako, and Cape Hedo stations in 2016. Statistics describing the comparison,
i.e., correlation coefficient (<inline-formula><mml:math id="M36" display="inline"><mml:mi>r</mml:mi></mml:math></inline-formula>), slope of the regression line (<inline-formula><mml:math id="M37" display="inline"><mml:mi>a</mml:mi></mml:math></inline-formula>),
number of samples (<inline-formula><mml:math id="M38" display="inline"><mml:mi>N</mml:mi></mml:math></inline-formula>), mean bias (bias), and root mean square error (RMSE),
are shown in the upper left corner; the dashed line is a 1 : 1 line, and
the regression line is shown in red. Time step: 30 min.</p></caption>
          <?xmltex \igopts{width=341.433071pt}?><graphic xlink:href="https://amt.copernicus.org/articles/11/2501/2018/amt-11-2501-2018-f02.png"/>

        </fig>

</sec>
</sec>
<sec id="Ch1.S3">
  <title>Results</title>
<sec id="Ch1.S3.SS1">
  <title>Spatiotemporal variability of the satellite data set</title>
      <p id="d1e701">In this section, we briefly examine the spatiotemporal variability of the
satellite data set compared to punctual ground-based observations (Fig. 1b).
We exclude the SZA dependence of the irradiance to focus on the variability
of the ground and satellite clear-sky indexes instead of the actual
irradiance values. The variability is represented by the coefficient of
variation CV <inline-formula><mml:math id="M39" display="inline"><mml:mo>=</mml:mo></mml:math></inline-formula> (SD <inline-formula><mml:math id="M40" display="inline"><mml:mo>/</mml:mo></mml:math></inline-formula> mean) <inline-formula><mml:math id="M41" display="inline"><mml:mo>×</mml:mo></mml:math></inline-formula> 100 of the clear-sky index. We
first compared the satellite instantaneous values (recorded every 2.5 min)
with the corresponding surface observations. Then we evaluated the two data
sets for different time step averages spanning from 5 min to 1 day (red to
violet diamonds). Concerning the spatial domain, we started by comparing
ground observations with the satellite pixel closest to the station (i.e.,
the distance from the station was 0 km; Fig. 1); then we focused on
satellite estimates averaged over larger domains (i.e., within 5, 10, and
20 km of the station).</p>
      <p id="d1e725">Figure 1 shows the difference between the CV of the surface SKYNET
observations and that of the Himawari-8 estimates for different time and
spatial domains as recorded at Chiba in August 2016. This difference was
always positive, such that the variability of the surface observations was
always greater than that of the satellite estimates. However, the difference
between satellite and ground variability was smaller when data were averaged
over long time domains (e.g., 1 h or 1 day) than over short time domains
(e.g., 2.5, 5, or 10 min), whereas averaging over larger spatial domains
reduced the variability of satellite-based estimates. These results confirm
those of previous studies (e.g., Perez et al., 2016) and suggest that
comparing satellite and ground observations is likely more challenging at
short timescales than on a hourly or daily basis. Even the variability of
single-satellite pixel results was smoother than the punctual ground
variability, which indicates the small-scale variability of clouds at the
subpixel level (i.e., <inline-formula><mml:math id="M42" display="inline"><mml:mo>&lt;</mml:mo></mml:math></inline-formula> 1 km). The variability became much less
representative of that of the station within only a few (e.g., 5) kilometers.
Finally, it is worth noting that the variability of Himawari-8 estimates
became almost insensitive to the size of the domain at a daily timescale.</p>
</sec>
<sec id="Ch1.S3.SS2">
  <title>Validations under all-sky conditions</title>
      <p id="d1e741">Because a solar radiation data set collected by the Himawari-7 (a previous
Himawari) satellite with the EXAM algorithm had recently been validated using
SKYNET observations<?pagebreak page2506?> (Khatri et al., 2015), we began by examining the results
of our new validation for the same SKYNET stations at the same temporal
resolution (i.e., a time step of 30 min, all available surface observations
averaged within a time window of <inline-formula><mml:math id="M43" display="inline"><mml:mo>±</mml:mo></mml:math></inline-formula>10 min; Fig. 2). Thus, eventual
improvements in the validation results could be more easily detected.
Overall, the new analysis showed that the Himawari-8 estimates reproduced the
observations at all stations very well, with statistics comparable to or
better than those obtained by other geostationary satellites (Fig. 2 and
Table 1; cf. Gómez et al., 2016; Federico et al., 2017). For all
stations, the correlation coefficient was approximately 0.95–0.96, and the
(negative) MB was approximately 20–30 W m<inline-formula><mml:math id="M44" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> (i.e., Himawari-8
overestimated the ground-based observations) and the RMSE ranged from
approximately 80 to 100 W m<inline-formula><mml:math id="M45" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula>. At all stations, the new data set yielded
a lower MB and RMSE than did the previous validation (Table 1). This small
improvement over the results of the previous validation was perhaps due to an
improved signal-to-noise ratio and/or spatial resolution in the new
observations.</p>

<?xmltex \floatpos{t}?><table-wrap id="Ch1.T1" specific-use="star"><label>Table 1</label><caption><p id="d1e778">Comparison of Himawari-8 and Himawari-7 data sets with surface
SKYNET observations at the Chiba, Fukue, Cape Hedo, and Miyako stations in
2016 (see Fig. 2) and 2015, respectively. Results of the statistical analysis
(N is the number of samples, <inline-formula><mml:math id="M46" display="inline"><mml:mi>r</mml:mi></mml:math></inline-formula> correlation coefficient, MB is mean
bias, RMSE is root mean square error, <inline-formula><mml:math id="M47" display="inline"><mml:mi>a</mml:mi></mml:math></inline-formula> is the slope of the regression line)
are reported for a 0.5 h time step.</p></caption><oasis:table frame="topbot"><oasis:tgroup cols="9">
     <oasis:colspec colnum="1" colname="col1" align="left"/>
     <oasis:colspec colnum="2" colname="col2" align="right"/>
     <oasis:colspec colnum="3" colname="col3" align="right"/>
     <oasis:colspec colnum="4" colname="col4" align="right"/>
     <oasis:colspec colnum="5" colname="col5" align="right" colsep="1"/>
     <oasis:colspec colnum="6" colname="col6" align="right"/>
     <oasis:colspec colnum="7" colname="col7" align="right"/>
     <oasis:colspec colnum="8" colname="col8" align="right"/>
     <oasis:colspec colnum="9" colname="col9" align="right"/>
     <oasis:thead>
       <oasis:row rowsep="1">
         <oasis:entry colname="col1">Satellite/year</oasis:entry>
         <oasis:entry namest="col2" nameend="col5" align="center" colsep="1">Himawari-8/2016 </oasis:entry>
         <oasis:entry namest="col6" nameend="col9" align="center">Himawari-7/2015 </oasis:entry>
       </oasis:row>
     </oasis:thead>
     <oasis:tbody>
       <oasis:row>
         <oasis:entry colname="col1">Station</oasis:entry>
         <oasis:entry colname="col2">Chiba</oasis:entry>
         <oasis:entry colname="col3">Fukue</oasis:entry>
         <oasis:entry colname="col4">Hedo</oasis:entry>
         <oasis:entry colname="col5">Miyako</oasis:entry>
         <oasis:entry colname="col6">Chiba</oasis:entry>
         <oasis:entry colname="col7">Fukue</oasis:entry>
         <oasis:entry colname="col8">Hedo</oasis:entry>
         <oasis:entry colname="col9">Miyako</oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Time step</oasis:entry>
         <oasis:entry colname="col2">0.5 h</oasis:entry>
         <oasis:entry colname="col3">0.5 h</oasis:entry>
         <oasis:entry colname="col4">0.5 h</oasis:entry>
         <oasis:entry colname="col5">0.5 h</oasis:entry>
         <oasis:entry colname="col6">0.5 h</oasis:entry>
         <oasis:entry colname="col7">0.5 h</oasis:entry>
         <oasis:entry colname="col8">0.5 h</oasis:entry>
         <oasis:entry colname="col9">0.5 h</oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1"><inline-formula><mml:math id="M48" display="inline"><mml:mi>N</mml:mi></mml:math></inline-formula></oasis:entry>
         <oasis:entry colname="col2">7668</oasis:entry>
         <oasis:entry colname="col3">6664</oasis:entry>
         <oasis:entry colname="col4">6927</oasis:entry>
         <oasis:entry colname="col5">4576</oasis:entry>
         <oasis:entry colname="col6">7508</oasis:entry>
         <oasis:entry colname="col7">6742</oasis:entry>
         <oasis:entry colname="col8">7390</oasis:entry>
         <oasis:entry colname="col9">5563</oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1"><inline-formula><mml:math id="M49" display="inline"><mml:mi>r</mml:mi></mml:math></inline-formula></oasis:entry>
         <oasis:entry colname="col2">0.958</oasis:entry>
         <oasis:entry colname="col3">0.959</oasis:entry>
         <oasis:entry colname="col4">0.961</oasis:entry>
         <oasis:entry colname="col5">0.948</oasis:entry>
         <oasis:entry colname="col6">0.959</oasis:entry>
         <oasis:entry colname="col7">0.952</oasis:entry>
         <oasis:entry colname="col8">0.956</oasis:entry>
         <oasis:entry colname="col9">0.913</oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">MB (W m<inline-formula><mml:math id="M50" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula>)</oasis:entry>
         <oasis:entry colname="col2"><inline-formula><mml:math id="M51" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>21.446</oasis:entry>
         <oasis:entry colname="col3"><inline-formula><mml:math id="M52" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>32.410</oasis:entry>
         <oasis:entry colname="col4"><inline-formula><mml:math id="M53" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>31.443</oasis:entry>
         <oasis:entry colname="col5"><inline-formula><mml:math id="M54" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>30.011</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M55" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>28.974</oasis:entry>
         <oasis:entry colname="col7"><inline-formula><mml:math id="M56" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>44.445</oasis:entry>
         <oasis:entry colname="col8"><inline-formula><mml:math id="M57" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>43.229</oasis:entry>
         <oasis:entry colname="col9"><inline-formula><mml:math id="M58" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>36.725</oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">RMSE (W m<inline-formula><mml:math id="M59" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula>)</oasis:entry>
         <oasis:entry colname="col2">79.411</oasis:entry>
         <oasis:entry colname="col3">88.207</oasis:entry>
         <oasis:entry colname="col4">89.893</oasis:entry>
         <oasis:entry colname="col5">105.008</oasis:entry>
         <oasis:entry colname="col6">83.933</oasis:entry>
         <oasis:entry colname="col7">97.744</oasis:entry>
         <oasis:entry colname="col8">97.369</oasis:entry>
         <oasis:entry colname="col9">131.332</oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1"><inline-formula><mml:math id="M60" display="inline"><mml:mi>a</mml:mi></mml:math></inline-formula></oasis:entry>
         <oasis:entry colname="col2">0.9439</oasis:entry>
         <oasis:entry colname="col3">0.9435</oasis:entry>
         <oasis:entry colname="col4">0.9609</oasis:entry>
         <oasis:entry colname="col5">0.9352</oasis:entry>
         <oasis:entry colname="col6">0.970</oasis:entry>
         <oasis:entry colname="col7">0.982</oasis:entry>
         <oasis:entry colname="col8">0.986</oasis:entry>
         <oasis:entry colname="col9">0.905</oasis:entry>
       </oasis:row>
     </oasis:tbody>
   </oasis:tgroup></oasis:table></table-wrap>

      <?xmltex \floatpos{t}?><fig id="Ch1.F3" specific-use="star"><label>Figure 3</label><caption><p id="d1e1145"><bold>(a)</bold> Influence of cloudiness (i.e., total cloud fraction,
violet to yellow contour map) on the monthly RMSE of ground observations and
Himawari-8 estimates of solar radiation at the 47 stations (points) in the
Japanese Meteorological Agency (JMA) network in May 2016. Bottom inset:
scatter plot of the total cloud fraction and RMSE for the stations in the
central and south regions (i.e., within the area delimited by the black line).
<bold>(b)</bold> Correlation between monthly RMSE and cloud fraction at each JMA
station for 2016 plotted over the Japan Digital Elevation Model derived from
GTOPO-30 (<uri>https://lta.cr.usgs.gov/GTOPO30</uri>). The three main regions,
i.e., north, central, and south, are enclosed by blue, red, and green dashed
lines, respectively; violet circles show the location of the SKYNET stations;
inset: mean RMSE (bold lines) and minimum and maximum RMSE (points) for the
different regions for December to January 2016.</p></caption>
          <?xmltex \igopts{width=497.923228pt}?><graphic xlink:href="https://amt.copernicus.org/articles/11/2501/2018/amt-11-2501-2018-f03.pdf"/>

        </fig>

      <p id="d1e1163">Although local surface features such as a complex morphology and/or surface
albedo can affect the agreement between satellite estimates and ground-based
measurements, cloudiness is expected to be the main factor determining the
magnitude of RMSE. Figure 3a shows the monthly RMSE between Himawari-8
estimates and ground-based JMA observations plotted on a map of the total
cloud fraction in May 2016 retrieved from a Modern-Era Retrospective Analysis
for Research and Applications (MERRA) reanalysis. Despite certain exceptions,
a larger (smaller) RMSE is usually coupled with higher (lower) cloud fraction
levels. This result is evident from the increasing gradient in the cloud
fraction that developed from the south of Japan to the Okinawa Islands and
that resulted in a corresponding increasing gradient in RMSE. Limited to the
central and southern regions of Japan (enclosed by the black line), this
overall influence of cloudiness on RMSE is also well evident in the scatter
plot (inset panel).</p>
      <p id="d1e1166">Figure 3b shows the correlation between monthly RMSE and cloud fraction at
each JMA station for January to December 2016 plotted over the Japan Digital
Elevation Model derived from GTOPO-30
(<uri>https://lta.cr.usgs.gov/GTOPO30</uri>). Overall, the correlation was always
positive but it ranged from more than 0.9 to less than 0.2 for the different
stations. As Japan extends from north to south for about 3000 km, it is
characterized by a variety of climatic regions which affected the pattern of
the correlation. Indeed, according to previous studies (e.g., Ohtake et al.,
2015), we can distinguish at least three main climatological regions and
additional smaller areas. The first region (enclosed by the blue dashed line)
is the northern part of Japan, mostly characterized by a subarctic climate,
which provided a uniform response and small differences between Hokkaido and
the north of the mainland (i.e., Tōhoku). Here, except for two stations
located in the Pacific sector of Japan, usually the correlation was<?pagebreak page2508?> lower
than 0.4. Then, the large central region (within the red dashed line), which
is characterized by humid and temperate climate, presents a more articulated
pattern. The flat and strongly urbanized Kanto region (i.e., around Tokyo)
showed the highest correlation values (<inline-formula><mml:math id="M61" display="inline"><mml:mrow><mml:mi>r</mml:mi><mml:mo>&gt;</mml:mo><mml:mn mathvariant="normal">0.9</mml:mn></mml:mrow></mml:math></inline-formula>). The SKYNET station of Chiba
University is located here and it is supposed to be representative of this
area. Then, correlations became slightly lower toward south with an evident
distinction between the east coast, characterized by higher correlations, and
the west coast, which presents lower values. It is worth noting that in
winter the west coast is usually affected by elevate snowfall levels, while
the Pacific coast usually shows frequent clear-sky conditions. Fukue SKYNET
station is located in the southwest sector of this region.</p>
      <p id="d1e1184">Stations in mountainous regions usually present peculiar features that
distinguish them from the stations located near the sea. A recent validation
of three satellite-based radiation products over an extensive network of 313
pyranometers across Europe (Urraca et al., 2017) showed that stations sited
in the Alps and Pyrenees have errors (i.e., RMSE and MB) two or three times
larger than the ones of other locations. In mountainous regions, the altitude
varies sharply and affects not only surface-related parameters but also the
state of the atmosphere. Therefore, satellite models can fail in such areas
because the spatial and temporal resolutions are not high enough to account
for the sharp terrain and changing weather conditions (Dürr et al., 2010;
Castelli et al., 2014). Accordingly, although not shown here, in  central
and northern Japan we found larger RMSE values for stations in mountainous
regions compared with seaside stations. Finally, we note a somewhat uniform
correlation (<inline-formula><mml:math id="M62" display="inline"><mml:mrow><mml:mi>r</mml:mi><mml:mo>&gt;</mml:mo><mml:mn mathvariant="normal">0.6</mml:mn></mml:mrow></mml:math></inline-formula>–0.8) in the subtropical region (enclosed by the green
dashed line) along the Pacific Ocean where the largest precipitation usually
occurs. The SKYNET stations of Cape Hedo and Miyako are located in this
latter region.</p>
      <p id="d1e1199">The peculiarities of the different regions  manifest when focusing on the
annual cycle of RMSE (see inset in Fig. 3b). Generally, higher RMSE values
were found in summer, the  highest values in the subtropical region,
followed by the central and the northern region. In contrast, in winter
RMSE values in the north were similar to the ones of the subtropical region.
Future analyses, based on longer time series, are expected to further
highlight the satellite uncertainties at the different locations and,
potentially, improve the accuracy of the satellite-derived data set by
site-adaptation methods (Polo et al., 2016).</p>
      <p id="d1e1202">Since reanalysis usually does not assimilate cloud fraction directly from
observations, issues of MERRA in reproducing the cloud fraction for specific
locations and periods can be expected. Moreover, uncertainties in the cloud
identification from the EXAM algorithm should be also taken into account.
Both of them would affect the correlations shown in Fig. 3. We compared the
monthly output of MERRA with visual observations recorded at the JMA stations
and we found that, despite a systematic bias in the amount of cloudiness
(reanalysis underestimates the cloud amount by about 0.1–0.2), MERRA
satisfactory reproduced the month-to-month variability in 2016 in both
central and north Japan (with some difficulty for the subtropical region). However, the EXAM algorithm can face difficulties in the cloud
identification under bright albedo (i.e., snow) conditions or in presence of
small size clouds. For example, it is likely that the bright surface albedo
would play a role in determining the different response on the east and west
coast. Moreover, other factors such as the type of cloudiness or the location
of the station, with possible constrains induced by the morphology and
altitude, likely determined the RMSE.</p>
      <p id="d1e1205">Taking the above results into account, we excluded the stations of the
northern region from the scatter plot (left inset of Fig. 3) and a good
correlation (<inline-formula><mml:math id="M63" display="inline"><mml:mrow><mml:mi>r</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.85</mml:mn></mml:mrow></mml:math></inline-formula>) between cloud fraction and RMSE for May 2016 was
found. For points corresponding to stations in the subtropical region, we
note that a large spread in RMSE was coupled to high cloud fraction (however,
if these points are removed, the remaining ones still show an evident
correlation; <inline-formula><mml:math id="M64" display="inline"><mml:mrow><mml:mi>r</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.74</mml:mn></mml:mrow></mml:math></inline-formula>). A somewhat small RMSE can also be expected when a
stable overcast sky dominates (Qu et al., 2017). Therefore, the larger spread
could be related to the frequency of such extreme conditions at the different
stations of the subtropical region.</p>
      <p id="d1e1233">Determining effects due to the local morphology would require a detailed
analysis beyond the objectives of this study. Figure 4 focuses on the
influence of surface albedo on monthly mean differences (i.e., JMA
measurements minus Himawari-8 estimates) in January 2016, when high (small as
usual) albedo values characterize Japan at latitudes roughly north (south) of
Tokyo, likely because of snow on the ground. The retrieval of the surface
albedo was based on a statistical approach which estimated the land surface
albedo by the second minimum reflectance method over a 30-day time window
(Fukuda et al., 2013). Although satellite estimates of SW radiation
were generally larger than surface observations in snow-free locations
characterized by low albedo values (around 0.1; i.e., in southern Japan), the
opposite trend was observed in the north (e.g., in Hokkaido) as well as in
mountainous regions (e.g., around Nagano). There the satellite estimates of
solar radiation in winter tended to underestimate actual values by about
20 W m<inline-formula><mml:math id="M65" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> (see the scatter plot in the inset panel). As we will see in
the next paragraphs, Himawari-based estimates of global radiation are usually
larger than ground observations under cloudiness conditions; therefore this
opposite trend appears somewhat anomalous. As an example, the bottom-right
inset shows the diurnal variation of the radiation under high surface albedo
conditions at Asahikawa (in Hokkaido, see the pink arrow) on 21 January.
Ground JMA observations suggest that while the morning was generally
characterized by small cloudiness and radiation values close to the average
of the season, in the afternoon a sudden and intense reduction of the
radiation, likely caused by thick clouds, occurred. In contrast,
Himawari-based estimates are extremely low in the morning while much<?pagebreak page2509?> larger
than ground observations in the afternoon. Previous studies showed that the
cloud identification from passive satellite instruments is usually affected
by large uncertainty under bright albedo conditions (e.g., Chan and Comiso,
2013; Damiani et al., 2015) and this can cause biased estimates of surface
solar radiation (Tanskanen et al., 2007). Indeed, a possible explanation of
the anomalous AMATERASS estimates could rely on a potential underestimation
of the surface albedo from the EXAM algorithm which leads to
misinterpretation of the observed bright scene as clouds. However, more
detailed analysis, beyond the objective of the present work, will be
necessary to univocally identify the issue and, eventually, improve the
algorithm.</p>

      <?xmltex \floatpos{t}?><fig id="Ch1.F4"><label>Figure 4</label><caption><p id="d1e1250">Influence of surface albedo (violet to yellow contour map) on the
monthly mean difference between ground observations and Himawari-8 estimates
at the stations (points) of the JMA network in January 2016. Top-left inset:
scatter plot of albedo and difference values for all stations. In the main
panel and the inset negative (positive) differences are shown as red (green)
dots. Bottom-right inset: ground-based JMA observations (black line) and
Himawari-8 estimates (green line) at Asahikawa (Hokkaido, Japan; see the pink
arrow) on 21 January under very high albedo conditions. The original time
step of the comparisons is 10 min.</p></caption>
          <?xmltex \igopts{width=236.157874pt}?><graphic xlink:href="https://amt.copernicus.org/articles/11/2501/2018/amt-11-2501-2018-f04.pdf"/>

        </fig>

      <?xmltex \floatpos{t}?><fig id="Ch1.F5" specific-use="star"><label>Figure 5</label><caption><p id="d1e1261"><bold>(a)</bold> Downwelling global shortwave radiation (SW)
measurements (black line) and clear-sky simulation (red line) at the Chiba
station on 6 August 2016 (top panel). An example of a radiation enhancement
(RE) event lasting about 40 min (green shadow) is indicated by the black
arrow and also shown in the all-sky camera image below around the peak of the
event (bottom). <bold>(b)</bold> Percentage of RE events per month among total
monthly samples for the different SKYNET stations. <bold>(c)</bold> Clear-sky
index (CSI) and solar zenith angle (SZA) at the Chiba station.
<bold>(d)</bold> Global SW measurements (black line) and clear-sky simulation
(red line) at the Chajnantor station (Chile) on 24 January 2017. The arrow
indicates an RE event lasting about 3.5 h (green shadow); the horizontal
dotted line indicates the solar constant (top panel); spectral SW
measurements recorded at the Chajnantor station at 10:30 on 24 January
(i.e., around the peak of the RE event; black line) and under clear-sky
conditions on 29 January 2017 (red line; bottom panel).</p></caption>
          <?xmltex \igopts{width=398.338583pt}?><graphic xlink:href="https://amt.copernicus.org/articles/11/2501/2018/amt-11-2501-2018-f05.png"/>

        </fig>

      <p id="d1e1281">As mentioned in the Introduction, under broken cloud conditions the surface
radiation can be even greater than the corresponding radiation under an ideal
cloud-free sky (i.e., CSI <inline-formula><mml:math id="M66" display="inline"><mml:mo>&gt;</mml:mo></mml:math></inline-formula> 1). This phenomenon is called radiation
enhancement, and it is likely to occur when the sky is partially covered
by thick clouds while the solar disk remains cloud-free (Gu et al., 2001;
Piedehierro et al., 2014) or when the solar disk is partially or fully
obstructed by optically thin clouds (Gueymard, 2017). Previous studies have
shown that RE up to around 50 % can be observed and simulated with
two-dimensional radiative transfer models (Piedehierro et al., 2014; Pecenak
et al., 2016). The left panel of Fig. 5 shows the global solar radiation
measured at the Chiba station on 6 August 2016, as well as the simulated
radiation expected under clear-sky conditions. An RE event lasting for more
than 40 min occurred around noon, and its flux was approximately
100 W m<inline-formula><mml:math id="M67" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> larger than the simulated value (see the pattern of the sky
recorded by the all-sky camera around the peak of the RE). Recall that these
enhancements cannot be reproduced by current satellite-based algorithms.
Figure 5 also shows the frequency distribution of RE events at the four
SKYNET stations examined in this study. A strong seasonality in RE occurrence
is apparent, with the largest frequency of events (about 10–20 % of all
observations) concentrated in July–September, with a lower frequency
(usually <inline-formula><mml:math id="M68" display="inline"><mml:mo>&lt;</mml:mo></mml:math></inline-formula> 5%) during the other seasons. At Chiba, the mean
(additional) flux caused by REs in August was 91 W m<inline-formula><mml:math id="M69" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> and the
SD was 48 W m<inline-formula><mml:math id="M70" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula>. Note that although REs are nearly
equally distributed during the day, the occurrence of the strongest events
was dependent on SZA. Nevertheless, because solar radiation was low under
such conditions, these events did not strongly affect the amount of radiation
at the ground; we therefore did not include them in our analysis (we
considered only data with SZA <inline-formula><mml:math id="M71" display="inline"><mml:mo>&lt;</mml:mo></mml:math></inline-formula> 70<inline-formula><mml:math id="M72" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula>). Previous studies
(Piedehierro et al., 2014) found that at middle–low SZA (roughly
<inline-formula><mml:math id="M73" display="inline"><mml:mo>&lt;</mml:mo></mml:math></inline-formula> 0<inline-formula><mml:math id="M74" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula>) the occurrence of REs is almost equally distributed while,
for larger SZA (roughly <inline-formula><mml:math id="M75" display="inline"><mml:mo>&gt;</mml:mo></mml:math></inline-formula> 40<inline-formula><mml:math id="M76" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula>), their frequency decreases along
with increasing SZA. In contrast, simulations showed that for the
overhead zenith angle the magnitude of the REs is highest for low COD (about
2–3), while their magnitude increases with increasing SZA and reaches the
maximum for high COD values (Pecenak et al., 2016). According to these
previous results, our data set showed that, although REs were nearly equally
distributed during the day, the occurrence of the strongest events occurred
mostly at high SZA. We also checked the relation between COD and REs by
exploiting Himawari-8 COD data (Nakajima and Nakajima, 1995) for Chiba
station. In the greatest majority of the cases, REs were coupled
with COD of about 0.5–5, while only a few of them were associated with higher
COD.</p>
      <p id="d1e1383">RE events are not typically expected to play a main role in the global solar
radiation budget. For example, at Chiba they accounted for only 1.55 % of
the global radiation budget in August. Nevertheless, to exemplify the large
variability in the time length and magnitude of REs, Fig. 5d shows a
record-long RE event recorded at Chajnantor station (Chile, 5100 m a.s.l.),
belonging to the ESR European SKYNET network, under comparable (austral)
summer conditions (i.e., on 24 January 2017). Details on the
state-of-the-art measurements of SW radiation recorded there,  their
comparison with<?pagebreak page2510?> other stations and radiative transfer simulations, and the influence of the altitude on the solar irradiance can be found
in Cordero et al. (2014, 2016). In this location, a single RE episode, with
flux approaching 300 W m<inline-formula><mml:math id="M77" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> greater than the simulated clear-sky flux,
developed between 08:00 and 11:30 and was characterized by an average
irradiance greater than the solar constant. Additional spectral observations
of global irradiance made around the peak of the RE event were compared to
analogous measurements recorded under clear-sky conditions on
29 January 2017. The largest enhancement (up to about
250 W m<inline-formula><mml:math id="M78" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> nm<inline-formula><mml:math id="M79" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">1</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula>) occurring around 450 nm was somewhat lower than
that of the global irradiance, as was expected given the low time resolution
of these spectral observations (Gueymard, 2017). The time length and
magnitude of this episode suggest that large RE effects can sometimes occur;
therefore, it would be desirable for satellite algorithms to take them into
account.</p>

      <?xmltex \floatpos{p}?><fig id="Ch1.F6" specific-use="star"><label>Figure 6</label><caption><p id="d1e1425"><bold>(a)</bold> Distribution of the CSI based on ground-based SKYNET
observations (pink area) and Himawari-8 estimates at the pixel level (red
line) and within a 5 km spatial domain (black line) for the different SKYNET
stations. <bold>(b)</bold> Main panel: distribution of the CSI based on
Himawari-8 estimates as panel <bold>(a)</bold> but only for RE events. Inset:
distribution of the CSI based on SKYNET observations but only for RE events.
Time step: 2.5 min. See text for further details.</p></caption>
          <?xmltex \igopts{width=284.527559pt}?><graphic xlink:href="https://amt.copernicus.org/articles/11/2501/2018/amt-11-2501-2018-f06.pdf"/>

        </fig>

      <?xmltex \floatpos{t}?><fig id="Ch1.F7" specific-use="star"><label>Figure 7</label><caption><p id="d1e1444">Scatter plot of surface global irradiance data obtained by
ground-based SKYNET observations and Himawari-8 estimates at the four SKYNET
stations. Colors indicate different clear-sky index ranges (based on SKYNET
observations): black (0–0.3), red (0.3–0.6), green (0.6–0.9), orange
(0.9–1), light blue (1–1.1), and violet (RE events). Statistics describing
the comparison, i.e., correlation coefficient (<inline-formula><mml:math id="M80" display="inline"><mml:mi>R</mml:mi></mml:math></inline-formula>), slope of the regression
line <inline-formula><mml:math id="M81" display="inline"><mml:mrow><mml:mo>(</mml:mo><mml:mi>a</mml:mi><mml:mo>)</mml:mo></mml:mrow></mml:math></inline-formula>, number of samples (<inline-formula><mml:math id="M82" display="inline"><mml:mi>N</mml:mi></mml:math></inline-formula>), mean bias (bias), and root mean square
error (RMSE), are shown; the dashed line is a 1 : 1 line. Time step:
2.5 min. See text for further details.</p></caption>
          <?xmltex \igopts{width=341.433071pt}?><graphic xlink:href="https://amt.copernicus.org/articles/11/2501/2018/amt-11-2501-2018-f07.png"/>

        </fig>

      <p id="d1e1479">Figure 6a compares the distribution of the CSI (between 0 and 1, we did not
include cases where CSI <inline-formula><mml:math id="M83" display="inline"><mml:mo>&gt;</mml:mo></mml:math></inline-formula> 1) of the ground-based SKYNET observations and
Himawari-8 estimates at different stations. Note that the distributions are
shown at a (satellite) instantaneous temporal resolution of 2.5 min,
collocated with the averages of all available surface observations within a
time window of <inline-formula><mml:math id="M84" display="inline"><mml:mo>±</mml:mo></mml:math></inline-formula>1.25  min. The Himawari-based CSI is shown at the pixel
level (red line) and as the average of the 5 km domain (black line). At the
pixel level, the satellite estimates seem to better reproduce the
distribution of ground observations, although there was an overestimation in
clear-sky cases and an underestimation in cases with substantial cloudiness
(CSI ca. 0.2–0.5), particularly at the southernmost stations (i.e., Cape
Hedo and Miyako), whereas the number of overcast cases (CSI ca. 0.1) was
usually well reproduced. When considering a somewhat larger spatial domain,
the distribution remained almost similar to that of the mid-range CSI values
(ca. 0.2–0.6), although there was a reduction in the size of the peaks for
extreme cases (i.e., clear-sky and overcast conditions). Further expanding
the spatial domain to 10 and 20 km changed the shape of the distribution
only slightly; therefore, we did not include these data in Fig. 6.</p>
      <p id="d1e1496">The insets in Fig. 6b show the distribution of the ground-based CSI for the
RE events (i.e., CSI <inline-formula><mml:math id="M85" display="inline"><mml:mo>&gt;</mml:mo></mml:math></inline-formula> 1). To take into account the combined
uncertainties of the pyranometer and simulation, we plotted only REs with an
absolute difference between observation and simulation of greater than 5%.
The distribution of the RE events in the different locations was quite
similar; however, more cases of extreme CSI (i.e., <inline-formula><mml:math id="M86" display="inline"><mml:mo>&gt;</mml:mo></mml:math></inline-formula> 1.2) occurred at
Chiba than at other stations. Although the satellite algorithm could not
reproduce the REs, it is interesting to examine what the satellite sees under
such conditions. This is shown in the main panels of the right column: at the
pixel level, the large majority of events were interpreted by the EXAM
algorithm as characterized by clear-sky conditions (i.e., CSI ca. 0.9–1).
These cases ranged from 55 to 75 % of the total, with Miyako showing the
largest proportion. The satellite estimates averaged within the 5 km domain
showed a similar distribution, peaking under clear-sky conditions, but with a
more gradual decrease toward more noticeably cloudy conditions.</p>
      <?pagebreak page2512?><p id="d1e1513">Figure 7 shows a scatter plot of the instantaneous global SW radiation at a
time step of 2.5 min between Himawari-8 estimates and ground-based
observations at the four SKYNET stations. We attempted to classify the
cloudiness in six ranges based on previously computed ground-based CSI (0–0.3, 0.3–0.6, 0.6–0.9, 0.9–1, 1–1.1, and RE events as defined
in the previous section), and we applied this classification to the same
scatter plots by highlighting the different CSI ranges with different
colors. Then, we calculated statistics for each range (Fig. 7) and for the
overall data set (Table 2). We confirmed using the all-sky cameras that
overcast conditions were mainly associated with CSI <inline-formula><mml:math id="M87" display="inline"><mml:mo>&lt;</mml:mo></mml:math></inline-formula> 0.3, whereas
clear-sky conditions were mainly associated with 0.9 <inline-formula><mml:math id="M88" display="inline"><mml:mo>&lt;</mml:mo></mml:math></inline-formula> CSI <inline-formula><mml:math id="M89" display="inline"><mml:mo>&lt;</mml:mo></mml:math></inline-formula> 1.1.
Broken clouds were associated with intermediate values. As shown in Table 2,
a somewhat good agreement between satellite and ground-based data was found,
although, as already shown in Fig. 2, Himawari-8 tended to slightly
overestimate the actual surface radiation. This overestimation was larger for
low irradiance values but tended to decrease for higher values and even show
an opposite trend (underestimation) under large/extreme irradiance
conditions. The latter case corresponds to the occurrence of RE events
(violet dots), as reflected in the somewhat large RMSE observed at all
stations. In general, the statistics were slightly worse than those for the
previous comparison (Fig. 2), as expected given the higher temporal
resolution of the data sets used in the new analysis. The correlation between
the two data sets improved with a decrease in cloudiness (as CSI increased).
By contrast, the mean bias (SKYNET minus Himawari-8) was consistently
negative for the supposed overcast conditions and tended to decrease toward
clear-sky conditions.</p>

<?xmltex \floatpos{p}?><table-wrap id="Ch1.T2" orientation="landscape"><label>Table 2</label><caption><p id="d1e1541">Comparison of the Himawari-8 data set with surface SKYNET
observations at the four SKYNET stations. Results of the statistical analysis
(<inline-formula><mml:math id="M90" display="inline"><mml:mi>N</mml:mi></mml:math></inline-formula>, <inline-formula><mml:math id="M91" display="inline"><mml:mi>r</mml:mi></mml:math></inline-formula>, MB, and <inline-formula><mml:math id="M92" display="inline"><mml:mi>a</mml:mi></mml:math></inline-formula>) are reported for time steps of 2.5 min, 1 h, and
1 day. Values in brackets indicate the same data sets excluding radiation
enhancement (RE) events.</p></caption><oasis:table frame="topbot"><?xmltex \begin{scaleboxenv}{.95}[.95]?><oasis:tgroup cols="13">
     <oasis:colspec colnum="1" colname="col1" align="left"/>
     <oasis:colspec colnum="2" colname="col2" align="right"/>
     <oasis:colspec colnum="3" colname="col3" align="right"/>
     <oasis:colspec colnum="4" colname="col4" align="right"/>
     <oasis:colspec colnum="5" colname="col5" align="right" colsep="1"/>
     <oasis:colspec colnum="6" colname="col6" align="right"/>
     <oasis:colspec colnum="7" colname="col7" align="right"/>
     <oasis:colspec colnum="8" colname="col8" align="right"/>
     <oasis:colspec colnum="9" colname="col9" align="right" colsep="1"/>
     <oasis:colspec colnum="10" colname="col10" align="right"/>
     <oasis:colspec colnum="11" colname="col11" align="right"/>
     <oasis:colspec colnum="12" colname="col12" align="right"/>
     <oasis:colspec colnum="13" colname="col13" align="right"/>
     <oasis:thead>
       <oasis:row rowsep="1">
         <oasis:entry colname="col1">Time step</oasis:entry>
         <oasis:entry namest="col2" nameend="col5" align="center" colsep="1">2.5 min </oasis:entry>
         <oasis:entry namest="col6" nameend="col9" align="center" colsep="1">1 h </oasis:entry>
         <oasis:entry namest="col10" nameend="col13" align="center">1 day </oasis:entry>
       </oasis:row>
     </oasis:thead>
     <oasis:tbody>
       <oasis:row rowsep="1">
         <oasis:entry colname="col1">Station</oasis:entry>
         <oasis:entry colname="col2">Chiba</oasis:entry>
         <oasis:entry colname="col3">Fukue</oasis:entry>
         <oasis:entry colname="col4">Hedo</oasis:entry>
         <oasis:entry colname="col5">Miyako</oasis:entry>
         <oasis:entry colname="col6">Chiba</oasis:entry>
         <oasis:entry colname="col7">Fukue</oasis:entry>
         <oasis:entry colname="col8">Hedo</oasis:entry>
         <oasis:entry colname="col9">Miyako</oasis:entry>
         <oasis:entry colname="col10">Chiba</oasis:entry>
         <oasis:entry colname="col11">Fukue</oasis:entry>
         <oasis:entry colname="col12">Hedo</oasis:entry>
         <oasis:entry colname="col13">Miyako</oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1"><inline-formula><mml:math id="M93" display="inline"><mml:mi>N</mml:mi></mml:math></inline-formula></oasis:entry>
         <oasis:entry colname="col2">59 816</oasis:entry>
         <oasis:entry colname="col3">59 128</oasis:entry>
         <oasis:entry colname="col4">60 185</oasis:entry>
         <oasis:entry colname="col5">42 426</oasis:entry>
         <oasis:entry colname="col6">2796</oasis:entry>
         <oasis:entry colname="col7">2687</oasis:entry>
         <oasis:entry colname="col8">2782</oasis:entry>
         <oasis:entry colname="col9">1953</oasis:entry>
         <oasis:entry colname="col10">315</oasis:entry>
         <oasis:entry colname="col11">288</oasis:entry>
         <oasis:entry colname="col12">309</oasis:entry>
         <oasis:entry colname="col13">198</oasis:entry>
       </oasis:row>
       <oasis:row rowsep="1">
         <oasis:entry colname="col1"/>
         <oasis:entry colname="col2">(56 787)</oasis:entry>
         <oasis:entry colname="col3">(56 381)</oasis:entry>
         <oasis:entry colname="col4">(57 315)</oasis:entry>
         <oasis:entry colname="col5">(37 972)</oasis:entry>
         <oasis:entry colname="col6">(2790)</oasis:entry>
         <oasis:entry colname="col7">(2686)</oasis:entry>
         <oasis:entry colname="col8">(2779)</oasis:entry>
         <oasis:entry colname="col9">(1951)</oasis:entry>
         <oasis:entry colname="col10">(315)</oasis:entry>
         <oasis:entry colname="col11">(288)</oasis:entry>
         <oasis:entry colname="col12">(309)</oasis:entry>
         <oasis:entry colname="col13">(198)</oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1"><inline-formula><mml:math id="M94" display="inline"><mml:mi>r</mml:mi></mml:math></inline-formula></oasis:entry>
         <oasis:entry colname="col2">0.909</oasis:entry>
         <oasis:entry colname="col3">0.910</oasis:entry>
         <oasis:entry colname="col4">0.897</oasis:entry>
         <oasis:entry colname="col5">0.879</oasis:entry>
         <oasis:entry colname="col6">0.961</oasis:entry>
         <oasis:entry colname="col7">0.961</oasis:entry>
         <oasis:entry colname="col8">0.956</oasis:entry>
         <oasis:entry colname="col9">0.948</oasis:entry>
         <oasis:entry colname="col10">0.977</oasis:entry>
         <oasis:entry colname="col11">0.976</oasis:entry>
         <oasis:entry colname="col12">0.979</oasis:entry>
         <oasis:entry colname="col13">0.966</oasis:entry>
       </oasis:row>
       <oasis:row rowsep="1">
         <oasis:entry colname="col1"/>
         <oasis:entry colname="col2">(0.925)</oasis:entry>
         <oasis:entry colname="col3">(0.921)</oasis:entry>
         <oasis:entry colname="col4">(0.909)</oasis:entry>
         <oasis:entry colname="col5">(0.885)</oasis:entry>
         <oasis:entry colname="col6">(0.962)</oasis:entry>
         <oasis:entry colname="col7">(0.961)</oasis:entry>
         <oasis:entry colname="col8">(0.962)</oasis:entry>
         <oasis:entry colname="col9">(0.945)</oasis:entry>
         <oasis:entry colname="col10">(0.975)</oasis:entry>
         <oasis:entry colname="col11">(0.977)</oasis:entry>
         <oasis:entry colname="col12">(0.980)</oasis:entry>
         <oasis:entry colname="col13">(0.956)</oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">MB (W m<inline-formula><mml:math id="M95" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula>)</oasis:entry>
         <oasis:entry colname="col2"><inline-formula><mml:math id="M96" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>22.902</oasis:entry>
         <oasis:entry colname="col3"><inline-formula><mml:math id="M97" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>33.798</oasis:entry>
         <oasis:entry colname="col4"><inline-formula><mml:math id="M98" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>34.043</oasis:entry>
         <oasis:entry colname="col5"><inline-formula><mml:math id="M99" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>33.256</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M100" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>20.450</oasis:entry>
         <oasis:entry colname="col7"><inline-formula><mml:math id="M101" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>32.478</oasis:entry>
         <oasis:entry colname="col8"><inline-formula><mml:math id="M102" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>31.998</oasis:entry>
         <oasis:entry colname="col9"><inline-formula><mml:math id="M103" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>31.392</oasis:entry>
         <oasis:entry colname="col10"><inline-formula><mml:math id="M104" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>20.068</oasis:entry>
         <oasis:entry colname="col11"><inline-formula><mml:math id="M105" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>31.425</oasis:entry>
         <oasis:entry colname="col12"><inline-formula><mml:math id="M106" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>31.927</oasis:entry>
         <oasis:entry colname="col13"><inline-formula><mml:math id="M107" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>31.371</oasis:entry>
       </oasis:row>
       <oasis:row rowsep="1">
         <oasis:entry colname="col1"/>
         <oasis:entry colname="col2">(<inline-formula><mml:math id="M108" display="inline"><mml:mo lspace="0mm">-</mml:mo></mml:math></inline-formula>34.477)</oasis:entry>
         <oasis:entry colname="col3">(<inline-formula><mml:math id="M109" display="inline"><mml:mo lspace="0mm">-</mml:mo></mml:math></inline-formula>44.663)</oasis:entry>
         <oasis:entry colname="col4">(<inline-formula><mml:math id="M110" display="inline"><mml:mo lspace="0mm">-</mml:mo></mml:math></inline-formula>45.439)</oasis:entry>
         <oasis:entry colname="col5">(<inline-formula><mml:math id="M111" display="inline"><mml:mo lspace="0mm">-</mml:mo></mml:math></inline-formula>55.204)</oasis:entry>
         <oasis:entry colname="col6">(<inline-formula><mml:math id="M112" display="inline"><mml:mo lspace="0mm">-</mml:mo></mml:math></inline-formula>31.460)</oasis:entry>
         <oasis:entry colname="col7">(<inline-formula><mml:math id="M113" display="inline"><mml:mo lspace="0mm">-</mml:mo></mml:math></inline-formula>41.865)</oasis:entry>
         <oasis:entry colname="col8">(<inline-formula><mml:math id="M114" display="inline"><mml:mo lspace="0mm">-</mml:mo></mml:math></inline-formula>42.223)</oasis:entry>
         <oasis:entry colname="col9">(<inline-formula><mml:math id="M115" display="inline"><mml:mo lspace="0mm">-</mml:mo></mml:math></inline-formula>51.675)</oasis:entry>
         <oasis:entry colname="col10">(<inline-formula><mml:math id="M116" display="inline"><mml:mo lspace="0mm">-</mml:mo></mml:math></inline-formula>30.858)</oasis:entry>
         <oasis:entry colname="col11">(<inline-formula><mml:math id="M117" display="inline"><mml:mo lspace="0mm">-</mml:mo></mml:math></inline-formula>41.421)</oasis:entry>
         <oasis:entry colname="col12">(<inline-formula><mml:math id="M118" display="inline"><mml:mo lspace="0mm">-</mml:mo></mml:math></inline-formula>42.850)</oasis:entry>
         <oasis:entry colname="col13">(<inline-formula><mml:math id="M119" display="inline"><mml:mo lspace="0mm">-</mml:mo></mml:math></inline-formula>52.527)</oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">RMSE (W m<inline-formula><mml:math id="M120" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula>)</oasis:entry>
         <oasis:entry colname="col2">113.317</oasis:entry>
         <oasis:entry colname="col3">125.544</oasis:entry>
         <oasis:entry colname="col4">132.283</oasis:entry>
         <oasis:entry colname="col5">151.303</oasis:entry>
         <oasis:entry colname="col6">70.741</oasis:entry>
         <oasis:entry colname="col7">81.935</oasis:entry>
         <oasis:entry colname="col8">81.579</oasis:entry>
         <oasis:entry colname="col9">95.545</oasis:entry>
         <oasis:entry colname="col10">44.638</oasis:entry>
         <oasis:entry colname="col11">55.729</oasis:entry>
         <oasis:entry colname="col12">52.024</oasis:entry>
         <oasis:entry colname="col13">60.775</oasis:entry>
       </oasis:row>
       <oasis:row rowsep="1">
         <oasis:entry colname="col1"/>
         <oasis:entry colname="col2">(103.827)</oasis:entry>
         <oasis:entry colname="col3">(119.251)</oasis:entry>
         <oasis:entry colname="col4">(126.270)</oasis:entry>
         <oasis:entry colname="col5">(147.870)</oasis:entry>
         <oasis:entry colname="col6">(73.085)</oasis:entry>
         <oasis:entry colname="col7">(84.611)</oasis:entry>
         <oasis:entry colname="col8">(83.829)</oasis:entry>
         <oasis:entry colname="col9">(104.758)</oasis:entry>
         <oasis:entry colname="col10">(51.152)</oasis:entry>
         <oasis:entry colname="col11">(60.770)</oasis:entry>
         <oasis:entry colname="col12">(59.072)</oasis:entry>
         <oasis:entry colname="col13">(77.852)</oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1"><inline-formula><mml:math id="M121" display="inline"><mml:mi>a</mml:mi></mml:math></inline-formula></oasis:entry>
         <oasis:entry colname="col2">0.8498</oasis:entry>
         <oasis:entry colname="col3">0.8497</oasis:entry>
         <oasis:entry colname="col4">0.8422</oasis:entry>
         <oasis:entry colname="col5">0.8000</oasis:entry>
         <oasis:entry colname="col6">0.9528</oasis:entry>
         <oasis:entry colname="col7">0.9463</oasis:entry>
         <oasis:entry colname="col8">0.9679</oasis:entry>
         <oasis:entry colname="col9">0.9327</oasis:entry>
         <oasis:entry colname="col10">0.9613</oasis:entry>
         <oasis:entry colname="col11">0.9550</oasis:entry>
         <oasis:entry colname="col12">1.0003</oasis:entry>
         <oasis:entry colname="col13">0.9456</oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1"/>
         <oasis:entry colname="col2">(0.9030)</oasis:entry>
         <oasis:entry colname="col3">(0.8913)</oasis:entry>
         <oasis:entry colname="col4">(0.8820)</oasis:entry>
         <oasis:entry colname="col5">(0.8592)</oasis:entry>
         <oasis:entry colname="col6">(0.9818)</oasis:entry>
         <oasis:entry colname="col7">(0.961)</oasis:entry>
         <oasis:entry colname="col8">(0.9900)</oasis:entry>
         <oasis:entry colname="col9">(0.9763)</oasis:entry>
         <oasis:entry colname="col10">(0.9812)</oasis:entry>
         <oasis:entry colname="col11">(0.9762)</oasis:entry>
         <oasis:entry colname="col12">(1.0248)</oasis:entry>
         <oasis:entry colname="col13">(0.9846)</oasis:entry>
       </oasis:row>
     </oasis:tbody>
   </oasis:tgroup><?xmltex \end{scaleboxenv}?></oasis:table></table-wrap>

      <p id="d1e2294">Note that this general pattern did not depend on the size of the region
within which the Himawari's pixels were averaged (e.g., there were virtually
no differences between areas extending to 10 or 20 km) but was mainly due to
heterogeneity in the cloudiness within the satellite pixel when data were
taken at nearly instantaneous time steps. Hourly and daily statistics are
also shown in Table 2. As expected, results based on hourly and daily
averages show an improved correlation and reduced RMSE with respect to the
results based on the instantaneous data. By contrast, the mean bias results
were only slightly improved.</p>
      <p id="d1e2297">Because RE events are not expected to be reproduced by the EXAM algorithm, it
is interesting to examine comparison statistics excluding these events. These
additional results are reported in Table 2 in brackets. Excluding REs
slightly improved the correlation coefficient and RMSE between the<?pagebreak page2513?> two data
sets for an instantaneous time step. By contrast, at longer time steps,
although the correlation did not show any substantial change, the RMSE showed
an opposite trend; i.e., the RMSE became somewhat worse at a time step of
1 h and greatly increased at a daily timescale with changes of about
10–30 % over values computed including RE events. By contrast, the mean
bias became about 10 W m<inline-formula><mml:math id="M122" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> larger than that previously computed; this
effect did not depend on the time step. This RE impact was likely
underestimated given the conservative threshold of 5 %, which reduced the
number of identified RE events. We suggest that researchers should take RE
events into account, particularly when conducting satellite validation in
cloudy regions.</p>

      <?xmltex \floatpos{t}?><fig id="Ch1.F8" specific-use="star"><label>Figure 8</label><caption><p id="d1e2314"><bold>(a, b)</bold> Diurnal variation in the difference
(i.e., SKYNET data minus Himawari data) in seasonal global radiation for
spring, fall, winter, and summer (<bold>a</bold> absolute differences,
<bold>b</bold> relative differences). <bold>(c, d)</bold> Diurnal
variation in seasonal CSI for spring, fall, winter, and summer (<bold>c</bold>:
SKYNET-based CSI, <bold>d</bold> Himawari-based CSI). A running average of 1 h
is applied to the instantaneous data. Results are shown for (left to right)
Chiba, Fukue, Cape Hedo, and Miyako.</p></caption>
          <?xmltex \igopts{width=497.923228pt}?><graphic xlink:href="https://amt.copernicus.org/articles/11/2501/2018/amt-11-2501-2018-f08.pdf"/>

        </fig>

      <p id="d1e2340">Figure 8a and b show diurnal variation in the difference (i.e., SKYNET minus
Himawari-8) in seasonal global radiation for spring, fall, winter, and summer
at the four SKYNET stations. Absolute (relative) differences show a satellite
overestimation in the range of 20–80 W m<inline-formula><mml:math id="M123" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> (5–15 %). It is
interesting that the absolute difference did not peak in summer, as one would
expect given the higher irradiance flux, but in spring. Nevertheless, summer
differences became much larger when RE events were not included. This trend was
evident at all stations. Note that, by comparing the SUNY modeled data set
with weather stations in California, a considerable larger bias in the
diurnal variation (annual mean around 18 % and up to about 50 % in
summer) was reported by Nottrott and Kleissl (2010).</p>
      <p id="d1e2356">Figure 8c and d show the CSI diurnal variation. The interseasonal variability
of the ground-based index was very well captured by the satellite-based index
for all stations. Even the diurnal variability, which generally showed
somewhat less cloudiness around noon than in early morning or late afternoon,
was very well reproduced. Although winter was the season with the least
cloudiness at Chiba, it was the cloudiest season at the other stations, where
summer was least cloudy.</p>
</sec>
<sec id="Ch1.S3.SS3">
  <title>Validations under clear-sky conditions</title>
      <p id="d1e2365">Both aerosols and PW strongly reduced the amount of radiation that reached
the ground. Although work is ongoing to include aerosol forcing in the EXAM
algorithm, the current version of the AMATERASS data set does not take
aerosols into account while including PW as estimated by the Japanese
reanalysis. Figure 9 shows the diurnal variation in global irradiance
measured at the ground and estimated by satellite at the Chiba station on
2 days (25 January 2016 and 21 May 2016) mainly characterized by clear-sky
conditions. Although in May Himawari-8 clearly overestimated the actual
radiation at the surface, in January it tended to slightly underestimate it.
Moreover, 21 May was affected by a high aerosol load, with AOD peaking well
above 0.6; 25 January presented very low aerosol amounts, with AOD below
0.05. This clearly suggests direct forcing by aerosols on global surface
radiation.</p>

      <?xmltex \floatpos{t}?><fig id="Ch1.F9"><label>Figure 9</label><caption><p id="d1e2370">Global surface irradiance measured at the ground (red line) and
estimated by Himawari-8 (black line) on 2 days (25 January 2016 and
21 May 2016) mainly characterized by clear-sky conditions at the Chiba
station. Measured SKYNET aerosol optical depth (AOD) values are also shown
for both days (black points, see opposite axis).</p></caption>
          <?xmltex \igopts{width=156.490157pt}?><graphic xlink:href="https://amt.copernicus.org/articles/11/2501/2018/amt-11-2501-2018-f09.pdf"/>

        </fig>

      <?pagebreak page2515?><p id="d1e2379">In the following analysis, we focus only on clear-sky data. As mentioned in
Sect. 2, the SD of the satellite measurements within the area surrounding our
station provided information on cloudiness. Therefore, we used an SD
threshold of 10 W m<inline-formula><mml:math id="M124" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> together with the method of Khatri and
Takamura (2009) to remove data affected by cloudiness. We further collocated
the instantaneous radiation observations collected under clear-sky conditions
with the aerosol property measurements. Figure 10a and b show scatter plots
of the difference between ground SKYNET and Himawari-8 global radiation data
sets with respect to the AOD during the period January to December 2016 at
the Chiba and Fukue stations. The linear regression line clearly shows an
increasing satellite overestimation of solar radiation along with the
increasing aerosol load. The overall instantaneous direct radiative forcing
efficiency estimated by the slope of the linear regression was about
<inline-formula><mml:math id="M125" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>125 W m<inline-formula><mml:math id="M126" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> per AOD unit at the Chiba station. The direct radiative
forcing efficiency of the AOD was slightly larger at Fukue (about
<inline-formula><mml:math id="M127" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>146 W m<inline-formula><mml:math id="M128" display="inline"><mml:mrow><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup><mml:mo>)</mml:mo></mml:mrow></mml:math></inline-formula>. This result is consistent with those of previous
studies showing that Fukue is usually characterized by thicker aerosols (Dim
et al., 2013). Overall, these results are roughly in agreement with findings
reported in modeling studies (Irie et al., 2017) and previous observational
results (e.g., Xia et al., 2007; Huttunen et al., 2014). Indeed, computations
performed by a radiative transfer model under the usual conditions at Chiba
showed that the direct radiative forcing efficiency at SZA of 50<inline-formula><mml:math id="M129" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> is
about <inline-formula><mml:math id="M130" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>150 W m<inline-formula><mml:math id="M131" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> per AOD unit (Irie et al., 2017). A similar value
(<inline-formula><mml:math id="M132" display="inline"><mml:mo lspace="0mm">-</mml:mo></mml:math></inline-formula>146.3 W m<inline-formula><mml:math id="M133" display="inline"><mml:mrow><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup><mml:mo>)</mml:mo></mml:mrow></mml:math></inline-formula> was experimentally estimated from observations
recorded by a CIMEL radiometer at Liaozhong (China) (Xia et al., 2007) and
comparable estimates were highlighted for other AERONET stations around the
world (Huttunen et al., 2014) as well as for the Mediterranean basin (e.g.,
Cachorro et al., 2008; Di Biagio et al., 2009, 2010; Papadimas et al., 2012).
These studies reported a direct radiative forcing efficiency at the surface
of about 100–150 W m<inline-formula><mml:math id="M134" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> (Cachorro et al., 2008), 100–200 W m<inline-formula><mml:math id="M135" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula>
(Di Biagio et al., 2009, 2010) and 100 W m<inline-formula><mml:math id="M136" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> (Papadimas et al., 2012)
per AOD unit. Recall that these values are based on the instantaneous
difference between ground observations and satellite estimates (which do not
take aerosol effects into account) under clear-sky conditions. We verified
that variation in the PW explains approximately 50 % of the AOD variance
at Chiba; therefore, water vapor could potentially create confusion and mask
these results if there are inconsistencies between the actual PW and the
forecast values used in the satellite algorithm. However, we also computed
the forcing efficiency using usual clear-sky simulations including PW
measurements retrieved by the microwave radiometer and verified that, as
expected, differences in the results were not substantial.</p>

      <?xmltex \floatpos{t}?><fig id="Ch1.F10" specific-use="star"><label>Figure 10</label><caption><p id="d1e2526"><bold>(a, b)</bold> Scatter plots of the difference in clear-sky surface
global radiation (SKYNET observations minus Himawari-8 estimates) and
measured SKYNET AOD at 500 nm under clear-sky conditions between January and
December 2016 at Chiba <bold>(a, c)</bold> and Fukue <bold>(b, d)</bold> station. The
regression line (in red), its slope (<inline-formula><mml:math id="M137" display="inline"><mml:mi>a</mml:mi></mml:math></inline-formula>), intercept (<inline-formula><mml:math id="M138" display="inline"><mml:mi>b</mml:mi></mml:math></inline-formula>), and the
correlation coefficient (<inline-formula><mml:math id="M139" display="inline"><mml:mi>R</mml:mi></mml:math></inline-formula>) are also shown. <bold>(c, d)</bold> Scatter plots
as above but for percentage difference in global radiation and aerosol
optical depth slant (AODS).</p></caption>
          <?xmltex \igopts{width=341.433071pt}?><graphic xlink:href="https://amt.copernicus.org/articles/11/2501/2018/amt-11-2501-2018-f10.pdf"/>

        </fig>

      <?xmltex \floatpos{t}?><fig id="Ch1.F11" specific-use="star"><label>Figure 11</label><caption><p id="d1e2570">Diurnal variation in the seasonal global radiation difference (i.e.,
SKYNET data minus Himawari-8 data, black lines) under clear-sky conditions
and AOD (red lines on the opposite axis) for <bold>(a, b)</bold> spring,
<bold>(c, d)</bold> fall, and <bold>(e, f)</bold> winter at the
Chiba <bold>(a, c, e)</bold> and Fukue <bold>(b, d, f)</bold> stations. Error bar
shows the standard deviation.</p></caption>
          <?xmltex \igopts{width=341.433071pt}?><graphic xlink:href="https://amt.copernicus.org/articles/11/2501/2018/amt-11-2501-2018-f11.pdf"/>

        </fig>

      <p id="d1e2594">It is worth noting that the instantaneous forcing efficiency depends also on
air mass (i.e., SZA). In a recent paper (Irie et al., 2017), by running
radiative transfer simulations including aerosol proprieties observed at the
SKYNET Chiba station, we showed that the relative decrease of solar radiation
induced by an increasing AOD is larger at high SZAs. In some previous studies
based on surface measurements (e.g., Xia et al., 2007; Cachorro et al., 2008;
Di Biagio et al., 2009, 2010), the instantaneous forcing efficiency has been
computed for small SZA ranges as a preliminary step to estimate the mean
direct aerosol forcing. However, usually small differences in the forcing
efficiency have been reported for SZA of 10–60<inline-formula><mml:math id="M140" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula>. Thus, because of
the additional constrains due to the satellite acquisition times and clear-sky observations and since data recorded at high SZAs were previously
excluded, in Fig. 10 we showed the forcing efficiency for the whole data set.
An additional issue is due to the fact that the instantaneous forcing
efficiency can be only partially described by a simple linear fit and, for
the same aerosol family, the estimation of the change in solar radiation for
AOD unit is rather different if computed over different AOD ranges. In our
data set, while we have a large range of AOD for mid-SZAs, we have a much
smaller range for high SZAs, which is probably insufficient to extrapolate a
reliable efficiency. Moreover, the slope of the linear regression is also
affected by the different type of aerosols (e.g., Di Biagio et al., 2009,
2010); therefore sampling the data for different SZA could introduce further
issues if episodes of different aerosol types occurred along the year. In
Fig. 10c and d we tried to remove the SZA effect on AOD by computing the
percentage decrease of solar radiation in function of the aerosol optical
depth slant (AODS, i.e., AODS <inline-formula><mml:math id="M141" display="inline"><mml:mo>=</mml:mo></mml:math></inline-formula> AOD <inline-formula><mml:math id="M142" display="inline"><mml:mo>/</mml:mo></mml:math></inline-formula> cos(SZA); e.g., Garcia et al.,
2006) and we obtained <inline-formula><mml:math id="M143" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>14.3 and <inline-formula><mml:math id="M144" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>14.5 % per AODS unit at Chiba and
Fukue, respectively.</p>
      <p id="d1e2634">To highlight the impact of aerosol load on downwelling global radiation on a
seasonal timescale, we plotted the daily cycle of mean radiation differences
(Fig. 11) between SKYNET and Himawari-8 and the AOD for spring, fall, and
winter (not enough clear-sky measurements were available in summer). Overall,
at the Chiba station, the largest negative differences between the ground
SKYNET observations and Himawari-8 estimates occurred in spring (about
20–30 W m<inline-formula><mml:math id="M145" display="inline"><mml:mrow><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup><mml:mo>)</mml:mo></mml:mrow></mml:math></inline-formula>, followed by fall (<inline-formula><mml:math id="M146" display="inline"><mml:mo lspace="0mm">&lt;</mml:mo></mml:math></inline-formula> 10 W m<inline-formula><mml:math id="M147" display="inline"><mml:mrow><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup><mml:mo>)</mml:mo></mml:mrow></mml:math></inline-formula>, whereas the
AMATERASS estimation matched the observations in winter. These differences
reflect the larger AOD amounts in spring, followed by fall and winter. These
results are roughly consistent with a previous investigation (Kudo et al.,
2010), based on ground-based observations carried out in a close location
(Tsukuba city, distance about 50 km), which reported a direct aerosol
forcing larger for spring to summer (<inline-formula><mml:math id="M148" display="inline"><mml:mo lspace="0mm">-</mml:mo></mml:math></inline-formula>22 W m<inline-formula><mml:math id="M149" display="inline"><mml:mrow><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup><mml:mo>)</mml:mo></mml:mrow></mml:math></inline-formula> and small in winter
(<inline-formula><mml:math id="M150" display="inline"><mml:mo lspace="0mm">-</mml:mo></mml:math></inline-formula>8.3 W m<inline-formula><mml:math id="M151" display="inline"><mml:mrow><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup><mml:mo>)</mml:mo></mml:mrow></mml:math></inline-formula>.</p>
      <p id="d1e2719">Although the seasonality of the aerosol load distribution appeared to be
somewhat different at Fukue, the negative differences in spring were close
to those in Chiba and were coupled with a similar AOD amount. There was a
peak in the impact of aerosols in winter, with a reduction of the global
radiation up to about 60 W m<inline-formula><mml:math id="M152" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula>. Finally, the minimum levels of AOD in
fall were coupled with anomalous positive differences, particularly in the
early morning and afternoon. Overall, it is noteworthy that the difference
between SKYNET and Himawari-8 irradiances was anticorrelated with AOD for
spring and winter, when AOD variations were relatively large. This finding
suggests the importance of the impact of aerosol load on global radiation
from the viewpoint of their diurnal variation.</p>
      <p id="d1e2735">A recent study showed that the diurnal variation patterns in Himawari-8 AOD
data were consistent with those seen in SKYNET observations (Irie et al.,
2017). Thus, Himawari-8 aerosol products would provide a unique spatial and
diurnal variation information which, once included in the next version of
the EXAM algorithm, is expected to reduce the positive bias under clear-sky
conditions here highlighted.</p>
      <p id="d1e2738">Since the different weather conditions finally determine the validation, it
is not straightforward to compare our findings with validation results
obtained in previous studies dealing with other satellites or algorithms.
Nevertheless, our<?pagebreak page2516?> results appear to be at least comparable to the ones
previously achieved. Since there is a lack of concurrent operational products
of surface solar radiation with a resolution comparable to AMATERASS (i.e., at
2.5 min), in the introduction we focused on results obtained from data sets
with a resolution of 1 h and 15 min. We noted that the majority of the
previous studies (e.g., Nottrott and Kleissi, 2010; Thomas et al., 2016a, b;
Qu et al., 2017) reported a general overestimation of the estimated solar
radiation by a few tens of W m<inline-formula><mml:math id="M153" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula>. This occurred despite the fact that
aerosols were taken into account in the algorithm. A comparable
overestimation, which did not depend on the adopted temporal resolution, has
been found in our analysis. This suggests that usually the role of aerosols
could be marginal compared with the influence of cloudiness when dealing with
all-sky conditions. In particular, for stations located in the flat and
strongly urbanized Kanto region (e.g., Chiba station), the all-sky
overestimation ranged from a few percent in winter to about 10 % in
summer (with a MB of about 23 W m<inline-formula><mml:math id="M154" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> on annual basis) and roughly
followed the annual cycle of the cloud amount (i.e., smaller in winter and
larger in summer). In contrast, the RMSE between observations and
satellite-based estimates did depend on the resolution (see Table 2) and, in
our analysis, it was reduced to about one-third of its original value
(computed from data at 2.5 min) when statistics were computed from daily
means.</p>
      <p id="d1e2765">We also showed that the RMSE computed from high-resolution (i.e., 10 min)
estimates usually ranged between 60 and 90 W m<inline-formula><mml:math id="M155" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> with larger values
occurring in summer, especially in the subtropical region (see the inset in
Fig. 3b). Such values resulted to be comparable to RMSE values computed at
similar resolution for Europe (i.e., 15 min; Qu et al., 2017) or even to
values from hourly data sets for Canada and Italy (Djebbar et al., 2012;
Federico et al., 2017). In the latter study (Federico et al., 2017) the
authors reported a RMSE in the range of 150–200 W m<inline-formula><mml:math id="M156" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> for three
mountain stations (two of them over 2000 m a.s.l.). It is worth noting that
such values are higher than any RMSE obtained for Japan from the data set at
2.5 min, even for subtropical stations. The authors suggested that in these
locations RMSEs were affected by specific constrains induced by the
morphology and/or the larger amount of cloudy days with respect to other
locations. Among the SKYNET and JMA stations in Japan, not one is located at
such high altitudes so unfortunately we cannot compare this interesting
feature. However, since we found the largest RMSE values for the subtropical
islands, we checked if they were associated with less clear-sky observations.
By using the CSI between 0.9 and 1 as a proxy of clear-sky conditions, it
resulted 5 % more clear-sky observations at Chiba than in the other
subtropical SKYNET locations.</p>
</sec>
</sec>
<sec id="Ch1.S4" sec-type="conclusions">
  <title>Conclusions</title>
      <p id="d1e2799">This study evaluated the AMATERASS global solar radiation data set based on
Himawari-8 observations and the EXAM algorithm with respect to ground-truthing observations and focused on spatiotemporal variability effects
(mainly induced by cloudiness) and the aerosol load. In general, the EXAM
algorithm applied to Himawari-8<?pagebreak page2517?> performed well in reproducing the surface
global irradiance at the four SKYNET stations examined in this study. The MB
was in the range of 20–30 W m<inline-formula><mml:math id="M157" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> (i.e., EXAM overestimated ground
observation), whereas the RMSE was usually around 80 W m<inline-formula><mml:math id="M158" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> (slightly
larger at Miyako).</p>
      <p id="d1e2826">Comparisons with respect to the JMA stations showed that the magnitude of
the RMSE was mainly determined by the level of cloudiness during the period
under investigation. By contrast, bright albedo conditions led to a
reduction or even a reversal of the sign of the mean differences between
ground observations and Himawari-based estimates.</p>
      <p id="d1e2829">The agreement with ground-based observations depended on the time step used
in the validation exercise as well as on the spatial domain and
climatological regions. Worse agreement was found for the instantaneous time
step (2.5 min), with the best RMSE at the daily level (Table 2). In
particular, the RMSE depended heavily on the time step used, whereas the MB
remained roughly constant.</p>
      <p id="d1e2832">At the pixel level, a larger number of samples were interpreted as clear-sky
data by the EXAM algorithm than in the real atmosphere. This trend was only
slightly reduced when a larger spatial domain was considered. Overall,
AMATERASS tended to slightly overestimate actual surface radiation.
Moreover, the overestimation was larger under overcast conditions, whereas
frequent episodes of surface REs (i.e., measured radiation larger than the
expected clear-sky radiation) under broken cloud conditions tended to
compensate the bias.</p>
      <?pagebreak page2518?><p id="d1e2836">The influence of the RE events appears to be substantial, particularly in
summer, when they accounted for 10–20% of the total measurements. The
EXAM algorithm could not reproduce such events and interpreted these
situations as characterized by almost clear-sky conditions. Obviously,
AMATERASS underestimated the global radiation during periods affected by
REs; this tendency balanced the general overestimation associated with
cloudiness. Indeed, removing RE events from the comparison increased the MB
at the instantaneous time step. This finding was also evident at longer time
steps. By contrast, the RMSE was reduced at the instantaneous time step, as
one would expect given the reduced number of spikes, whereas it increased at
the daily scale.</p>
      <p id="d1e2839">Under clear-sky conditions, the influence of aerosols makes the AMATERASS
estimates larger than ground observations. Based on the SKYNET and AMATERASS
data sets in 2016, the overall instantaneous direct radiative forcing
efficiencies were about <inline-formula><mml:math id="M159" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>125 and <inline-formula><mml:math id="M160" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>146 W m<inline-formula><mml:math id="M161" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula> per AOD unit at Chiba
and Fukue, respectively. We found that the diurnal pattern of the difference
between SKYNET observations and AMATERASS estimates in irradiance at these
stations was anticorrelated with AOD in different seasons, with larger
(smaller) differences in periods of higher (lower) AOD. At Chiba, the maximum
differences occurred in spring (about 20–30 W m<inline-formula><mml:math id="M162" display="inline"><mml:mrow><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup><mml:mo>)</mml:mo></mml:mrow></mml:math></inline-formula>, whereas at Fukue
in winter they were greater than 40–50 W m<inline-formula><mml:math id="M163" display="inline"><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:msup></mml:math></inline-formula>. The impact of aerosol
load on global radiation from the viewpoint of their diurnal variation was
also discussed.</p>
      <p id="d1e2895">Overall, our analysis confirmed the good accuracy of the AMATERASS solar
global radiation product at temporal resolution of 1 h and 10 min but showed
larger deviations at 2.5 min. An improved algorithm better accounting for
the cloud effects would certainly alleviate such deviations, which are more
evident under fast-varying gradients induced by changing clouds. For
example, AMATERASS “saw” slightly more clear-sky scenes compared to
surface observations while it tended to overestimate the solar radiation
under cloudiness conditions. A finer satellite spatial resolution would
improve the identification of small cumulus clouds. Moreover, both REs and
aerosols contributed to these deviations but their impact was much smaller.
Nevertheless, the portion of these deviations, which arises from the
intrinsic difference in comparing “punctual” measurements and satellite
pixels, could be hardly removed. When the pixel is partly cloudy, surface
measurements can greatly vary depending on whether clouds are located
along the solar beam path to the sensor possibly, resulting in a satellite
overestimation or underestimation of the irradiance at ground. Due to the
large field of view of the pyranometer and depending on the solar zenith
angle conditions, even clouds located many kilometers from the station can
potentially affect observations. Unless such effects are accounted by a
sufficiently long temporal integration of the surface observations, this
would result in discrepancies with respect to satellite-based estimates.</p>
      <p id="d1e2898">Despite these issues, we expect AMATERASS to play a key role in contributing
to the development of an efficient EMS in Japan.</p>
</sec>

      
      </body>
    <back><notes notes-type="dataavailability"><title>Data availability</title>

      <p id="d1e2906">The data used in this paper are freely available and can be
requested from the corresponding author.</p>
  </notes><notes notes-type="competinginterests"><title>Competing interests</title>

      <p id="d1e2912">The authors declare that they have no conflict of
interest.</p>
  </notes><notes notes-type="sistatement"><title>Special issue statement</title>

      <p id="d1e2918">This article is part of the special issue “SKYNET – the
international network for aerosol, clouds, and solar radiation studies and
their applications”. It is not associated with a conference.</p>
  </notes><ack><title>Acknowledgements</title><p id="d1e2924">The present study was supported by the JST/CREST/EMS/TEEDDA fund, grant
number JPMJCR15K4, Japan. We thank the Japan Meteorological Agency for
providing the ground-based observations used in this work.<?xmltex \hack{\newline}?><?xmltex \hack{\newline}?> Edited by: Stelios Kazadzis<?xmltex \hack{\newline}?> Reviewed by: three
anonymous referees</p></ack><ref-list>
    <title>References</title>

      <ref id="bib1.bib1"><label>1</label><mixed-citation>Antón, M., Cachorro, V. E., Vilaplana, J. M., Toledano, C., Krotkov, N.
A., Arola, A., Serrano, A., and de la Morena, B.: Comparison of UV
irradiances from Aura/Ozone Monitoring Instrument (OMI) with Brewer
measurements at El Arenosillo (Spain) – Part 1: Analysis of parameter
influence, Atmos. Chem. Phys., 10, 5979–5989,
<ext-link xlink:href="https://doi.org/10.5194/acp-10-5979-2010" ext-link-type="DOI">10.5194/acp-10-5979-2010</ext-link>, 2010.</mixed-citation></ref>
      <ref id="bib1.bib2"><label>2</label><mixed-citation>Bech, J., Sola, Y., Ossó, A., and Lorente, J.: Analysis of 14 years of
broadband ground-based solar UV index observations in Barcelona, Int. J.
Climatol., 35, 45–56, <ext-link xlink:href="https://doi.org/10.1002/joc.3961" ext-link-type="DOI">10.1002/joc.3961</ext-link>, 2015.</mixed-citation></ref>
      <ref id="bib1.bib3"><label>3</label><mixed-citation>
Bessho, K., Date, K., Hayashi, M., Ikeda, A., Imai, T., Inoue, H., Kumagai,
Y., Miyakawa, T., Murata, H., Ohno, T., Okuyama, A., Oyama, R., Sasaki, Y.,
Shimazu, Y., Shimoji, K., and Sumida, Y.: An Introduction to Himawari-8/9
— Japan's New-Generation Geostationary Meteorological Satellites, J.
Meteorol. Soc. Jpn., 94, 151–183, 2016.</mixed-citation></ref>
      <ref id="bib1.bib4"><label>4</label><mixed-citation>Blanc, P., Gschwind, B., Lefévre, M., and Wald, L.: The HelioClim
Project: Surface Solar Irradiance Data for Climate Applications, Remote
Sens., 3, 343–361, <ext-link xlink:href="https://doi.org/10.3390/rs3020343" ext-link-type="DOI">10.3390/rs3020343</ext-link>, 2011.</mixed-citation></ref>
      <ref id="bib1.bib5"><label>5</label><mixed-citation>Cachorro, V. E., Toledano, C., Prats, N., Sorribas, M., Mogo, S., Berjón,
A., Torres, B., Rodrigo, R., de la Rosa, J., and De Frutos, A. M.: The
strongest desert dust intrusion mixed with smoke over the Iberian Peninsula
registered with Sun photometry, J. Geophys. Res., 113, D14S04,
<ext-link xlink:href="https://doi.org/10.1029/2007JD009582" ext-link-type="DOI">10.1029/2007JD009582</ext-link>, 2008.</mixed-citation></ref>
      <ref id="bib1.bib6"><label>6</label><mixed-citation>Castelli, M., Stöckli, R., Tetzlaff, A., Wagner, J. E., Belluardo, G.,
Zebisch, M., and Petitta M.: The HelioMont method<?pagebreak page2519?> for assessing solar
irradiance over complex terrain: validation and improvements, Remote Sens.
Environ., 152, 603–613, <ext-link xlink:href="https://doi.org/10.1016/j.rse.2014.07.018" ext-link-type="DOI">10.1016/j.rse.2014.07.018</ext-link>, 2014.</mixed-citation></ref>
      <ref id="bib1.bib7"><label>7</label><mixed-citation>
Chan, M. A. and Comiso, J. C.: Arctic cloud characteristics as derived from
MODIS, CALIPSO, and CloudSat, J. Climate, 26, 3285–3306, 2013.</mixed-citation></ref>
      <ref id="bib1.bib8"><label>8</label><mixed-citation>Cordero, R. R., Damiani, A., Seckmeyer, G., Jorquera, J., Caballero, M.,
Rowe, P., Ferrer, J., Mubarak, R., Carrasco, J., Rondanelli, R., Matus, M.,
and Laroze, D.: The solar spectrum in the Atacama Desert, Sci. Rep.-UK 6,
22457, <ext-link xlink:href="https://doi.org/10.1038/srep22457" ext-link-type="DOI">10.1038/srep22457</ext-link>, 2016.</mixed-citation></ref>
      <ref id="bib1.bib9"><label>9</label><mixed-citation>Cordero, R. R., Seckmeyer, G., Damiani, A., Riechelmann, S., Rayas, J.,
Labbe, F., and Laroze, D.: The world's highest levels of surface UV, Photoch.
Photobio. Sci., 13, 70–81, <ext-link xlink:href="https://doi.org/10.1039/C3PP50221J" ext-link-type="DOI">10.1039/C3PP50221J</ext-link>, 2014.</mixed-citation></ref>
      <ref id="bib1.bib10"><label>10</label><mixed-citation>Damiani, A., De Simone, S., Rafanelli C., Cordero, R. R., and Laurenza, M.:
Three years of ground-based total ozone measurements in Arctic: comparison
with OMI, GOME and SCIAMACHY satellite data, Remote Sens. Environ., 127,
162–180, <ext-link xlink:href="https://doi.org/10.1016/j.rse.2012.08.023" ext-link-type="DOI">10.1016/j.rse.2012.08.023</ext-link>, 2012.</mixed-citation></ref>
      <ref id="bib1.bib11"><label>11</label><mixed-citation>Damiani, A., Cabrera, S., Muñoz, R., Cordero, R. R., and Labbe, F.:
Satellite-derived UV irradiance for a region with complex morphology and
meteorology: comparison against ground measurements in Santiago de Chile,
Int. J. Remote Sens., 34, 5812–5833, <ext-link xlink:href="https://doi.org/10.1080/01431161.2013.796101" ext-link-type="DOI">10.1080/01431161.2013.796101</ext-link>, 2013.</mixed-citation></ref>
      <ref id="bib1.bib12"><label>12</label><mixed-citation>
Damiani, A., Cordero, R., Carrasco, J., Watanabe, S., Kawamiya, M., and
Lagun, V. E.: Changes in the UV Lambertian Equivalent Reflectivity in the
Southern Ocean: influence of sea ice and cloudiness, Remote Sens. Environ.,
169, 75–92, 2015.</mixed-citation></ref>
      <ref id="bib1.bib13"><label>13</label><mixed-citation>
Djebbar, R., Morris, R., Thevenard, D., Perez, R., and Schlemmerd, J.:
Assessment of SUNY version 3 global horizontal and direct normal solar
irradiance in Canada, Enrgy. Proced., 30, 1274–1283, 2012.</mixed-citation></ref>
      <ref id="bib1.bib14"><label>14</label><mixed-citation>Dim, J. R., Takamura, T., Higurashi, A., Kathri, P., Kikuchi, N., and
Nakajima, T. Y.: Validation of Two MODIS Aerosols Algorithms with SKYNET and
Prospects for Future Climate Satellites Such as the GCOM-C/SGLI, Adv.
Meteorol., 2013, 508064, <ext-link xlink:href="https://doi.org/10.1155/2013/508064" ext-link-type="DOI">10.1155/2013/508064</ext-link>, 2013.</mixed-citation></ref>
      <ref id="bib1.bib15"><label>15</label><mixed-citation>Di Biagio, C., di Sarra, A., Meloni, D., Monteleone, F., Piacentino, S., and
Sferlazzo, D.: Measurements of Mediterranean aerosol radiative forcing and
influence of the single scattering albedo, J. Geophys. Res., 114, D06211,
<ext-link xlink:href="https://doi.org/10.1029/2008JD011037" ext-link-type="DOI">10.1029/2008JD011037</ext-link>, 2009.</mixed-citation></ref>
      <ref id="bib1.bib16"><label>16</label><mixed-citation>Di Biagio, C., di Sarra, A., and Meloni, D.: Large atmospheric shortwave
radiative forcing by Mediterranean aerosols derived from simultaneous
ground-based and spaceborne observations and dependence on the aerosol type
and single scattering albedo, J. Geophys. Res., 115, D10209,
<ext-link xlink:href="https://doi.org/10.1029/2009JD012697" ext-link-type="DOI">10.1029/2009JD012697</ext-link>, 2010.</mixed-citation></ref>
      <ref id="bib1.bib17"><label>17</label><mixed-citation>
Dürr, B., Zelenka, A., Müller, R., and Philipona, R.: Verification of
CM-SAF and MeteoSwiss satellite based retrievals of surface shortwave
irradiance over the Alpine region, Int. J. Remote Sens., 31, 4179–4198,
2010.</mixed-citation></ref>
      <ref id="bib1.bib18"><label>18</label><mixed-citation>
Engeland, K., Borga, M., Creutin, J.-D., François, B., Ramos, M.-H., and
Vidal, J.-P.: Space-time variability of climate variables and intermittent
renewable electricity production – A review, Renew. Sustain. Energ. Rev.,
79, 600–617, 2017.</mixed-citation></ref>
      <ref id="bib1.bib19"><label>19</label><mixed-citation>Federico, S., Torcasio, R. C., Sanò, P., Casella, D., Campanelli, M.,
Meirink, J. F., Wang, P., Vergari, S., Diémoz, H., and Dietrich, S.:
Comparison of hourly surface downwelling solar radiation estimated from
MSG–SEVIRI and forecast by the RAMS model with pyranometers over Italy,
Atmos. Meas. Tech., 10, 2337–2352, <ext-link xlink:href="https://doi.org/10.5194/amt-10-2337-2017" ext-link-type="DOI">10.5194/amt-10-2337-2017</ext-link>,
2017.</mixed-citation></ref>
      <ref id="bib1.bib20"><label>20</label><mixed-citation>Fukuda, S., Nakajima, T., Takenaka, H., Higurashi, A., Kikuchi, N., Nakajima,
T. Y., and Ishida, H.: New approaches to removing cloud sha- dows and
evaluating the 380 nm surface reflectance for improved aerosol optical
thickness retrievals from the GOSAT/TANSO-Cloud and Aerosol Imager, J.
Geophys. Res.-Atmos, 118, 13–520, <ext-link xlink:href="https://doi.org/10.1002/2013JD020090" ext-link-type="DOI">10.1002/2013JD020090</ext-link>, 2013.</mixed-citation></ref>
      <ref id="bib1.bib21"><label>21</label><mixed-citation>García, O. E., Díaz, A. M., Expósito, F. J., Díaz, J. P.,
Gröbner, J., and Fioletov, V. E.: Cloudless aerosol forcing efficiency in
the UV region from AERONET and WOUDC databases, Geophys. Res. Lett., 33,
L23803, <ext-link xlink:href="https://doi.org/10.1029/2006GL026794" ext-link-type="DOI">10.1029/2006GL026794</ext-link>, 2006.</mixed-citation></ref>
      <ref id="bib1.bib22"><label>22</label><mixed-citation>Gómez, I., Caselles, V., and Estrela, M. J.: Seasonal Characterization of
Solar Radiation Estimates Obtained from a MSG- SEVIRI-Derived Dataset and a
RAMS-Based Operational Forecasting System over the Western Mediterranean
Coast, Remote Sens., 8, 46, <ext-link xlink:href="https://doi.org/10.3390/rs8010046" ext-link-type="DOI">10.3390/rs8010046</ext-link>, 2016.</mixed-citation></ref>
      <ref id="bib1.bib23"><label>23</label><mixed-citation>Greuell, W., Meirink, J. F., and Wang, P.: Retrieval and validation of
global, direct, and diffuse irradiance derived from SEVIRI satellite
observations, J. Geophys. Res., 118, 2340–2361, <ext-link xlink:href="https://doi.org/10.1002/jgrd.50194" ext-link-type="DOI">10.1002/jgrd.50194</ext-link>,
2013.</mixed-citation></ref>
      <ref id="bib1.bib24"><label>24</label><mixed-citation>
Gu, L., Fuentes, J. D., Garstang, M., da Silva, J. T., Heitz, R., Sigler, J.,
and Shugart, H. H.: Cloud modulation of surface solar irradiance at a pasture
site in southern Brazil, Agric. For. Meteorol., 106, 117–129, 2001.</mixed-citation></ref>
      <ref id="bib1.bib25"><label>25</label><mixed-citation>
Gueymard, C. A.: Cloud and albedo enhancement impacts on solar irradiance
using high- frequency measurements from thermopile and photodiode
radiometers, Part 1: Impacts on global horizontal irradiance, Sol. Energy,
153, 755–765, 2017.</mixed-citation></ref>
      <ref id="bib1.bib26"><label>26</label><mixed-citation>Hashimoto, M. and Nakajima, T.: Development of a remote sensing algorithm to
retrieve atmospheric aerosol properties using multiwave- length and
multipixel information, J. Geophys. Res.-Atmos., 122, 6347–6378,
<ext-link xlink:href="https://doi.org/10.1002/2016JD025698" ext-link-type="DOI">10.1002/2016JD025698</ext-link>, 2017.</mixed-citation></ref>
      <ref id="bib1.bib27"><label>27</label><mixed-citation>
Hoff, T. E. and Perez, R.: Quantifying PV power Output Variability, Sol.
Energy, 84, 1782–1793, 2010.</mixed-citation></ref>
      <ref id="bib1.bib28"><label>28</label><mixed-citation>Huttunen, J., Arola, A., Myhre, G., Lindfors, A. V., Mielonen, T., Mikkonen,
S., Schafer, J. S., Tripathi, S. N., Wild, M., Komppula, M., and Lehtinen, K.
E. J.: Effect of water vapor on the determination of aerosol direct radiative
effect based on the AERONET fluxes, Atmos. Chem. Phys., 14, 6103–6110,
<ext-link xlink:href="https://doi.org/10.5194/acp-14-6103-2014" ext-link-type="DOI">10.5194/acp-14-6103-2014</ext-link>, 2014.</mixed-citation></ref>
      <ref id="bib1.bib29"><label>29</label><mixed-citation>
Higurashi, A. and Nakajima, T.: Development of a two-channel aerosol
retrieval algorithm on global scale using NOAA/AVHRR, J. Atmos. Sci., 56,
924–941, 1999.</mixed-citation></ref>
      <ref id="bib1.bib30"><label>30</label><mixed-citation>
Inman, R. H., Chu, Y., and Coimbra, C. F. M.: Cloud enhancement of global
horizontal irradiance in California and Hawaii, Sol. Energy 130, 128–138,
2016.</mixed-citation></ref>
      <ref id="bib1.bib31"><label>31</label><mixed-citation>
Irie, H., Horio T., Damiani, A., Nakajima, T. Y., Takenaka, H., Kikuchi, M.,
Khatri, P., and Yumimoto, K.: Importance of Himawari-8 aerosol products for
energy management system, Earozoru Kenkyu, 32, 95–100, 2017.</mixed-citation></ref>
      <ref id="bib1.bib32"><label>32</label><mixed-citation>Iwabuchi, H.: Efficient Monte Carlo Methods for Radiative Transfer Modeling,
J. Atmos. Sci., 63, 2324–2339, <ext-link xlink:href="https://doi.org/10.1175/JAS3755.1" ext-link-type="DOI">10.1175/JAS3755.1</ext-link>, 2006.</mixed-citation></ref>
      <?pagebreak page2520?><ref id="bib1.bib33"><label>33</label><mixed-citation>Kathri, P. and Takamura, T.: An Algorithm to Screen Cloud-Affected Data for
Sky Radiometer Data Analysis, J. Meteorol. Soc. Jpn., 87, 189–204,
<ext-link xlink:href="https://doi.org/10.2151/jmsj.87.189" ext-link-type="DOI">10.2151/jmsj.87.189</ext-link>, 2009.</mixed-citation></ref>
      <ref id="bib1.bib34"><label>34</label><mixed-citation>
Khatri, P., Irie, H., Takamura, T., Takenaka, H., and Kurino, T.: Validation
of global fluxes estimated from geostationary satellite HIMAWARI data using
ground-truth data, Autumn meeting of Meteorological Society of Japan, Kyoto,
Japan, 28–30 October 2015.</mixed-citation></ref>
      <ref id="bib1.bib35"><label>35</label><mixed-citation>
Kawamoto, K., Nakajima, T., and Nakajima, T. Y.: A global determination of
cloud microphysics with AVHRR remote sensing, J. Climate, 14, 2054–2068,
2001.</mixed-citation></ref>
      <ref id="bib1.bib36"><label>36</label><mixed-citation>Kosmopoulos, P. G., Kazadzis, S., Taylor, M., Raptis, P. I., Keramitsoglou,
I., Kiranoudis, C., and Bais, A. F.: Assessment of surface solar irradiance
derived from real-time modelling techniques and verification with
ground-based measurements, Atmos. Meas. Tech., 11, 907–924,
<ext-link xlink:href="https://doi.org/10.5194/amt-11-907-2018" ext-link-type="DOI">10.5194/amt-11-907-2018</ext-link>, 2018.</mixed-citation></ref>
      <ref id="bib1.bib37"><label>37</label><mixed-citation>Kudo, R., Uchiyama, A., Yamazaki, A., and Kobayashi, E.: Seasonal
characteristics of aerosol radiative effect estimated from ground-based solar
radiation measurements in Tsukuba, Japan, J. Geophys. Res., 115, D01204,
<ext-link xlink:href="https://doi.org/10.1029/2009JD012487" ext-link-type="DOI">10.1029/2009JD012487</ext-link>, 2010.</mixed-citation></ref>
      <ref id="bib1.bib38"><label>38</label><mixed-citation>Madhavan, B. L., Deneke, H., Witthuhn, J., and Macke, A.: Multiresolution
analysis of the spatiotemporal variability in global radiation observed by a
dense network of 99 pyranometers, Atmos. Chem. Phys., 17, 3317–3338,
<ext-link xlink:href="https://doi.org/10.5194/acp-17-3317-2017" ext-link-type="DOI">10.5194/acp-17-3317-2017</ext-link>, 2017.</mixed-citation></ref>
      <ref id="bib1.bib39"><label>39</label><mixed-citation>Mok, J., Krotkov, N. A., Torres, O., Jethva, H., Li, Z., Kim, J., Koo, J.-H.,
Go, S., Irie, H., Labow, G., Eck, T. F., Holben, B. N., Herman, J., Loughman,
R. P., Spinei, E., Lee, S. S., Khatri, P., and Campanelli, M.: Comparisons of
spectral aerosol absorption in Seoul, South Korea, Atmos. Meas. Tech.
Discuss., <ext-link xlink:href="https://doi.org/10.5194/amt-2017-380" ext-link-type="DOI">10.5194/amt-2017-380</ext-link>, in review, 2017.</mixed-citation></ref>
      <ref id="bib1.bib40"><label>40</label><mixed-citation>
Nakajima, T. Y. and Nakajima, T.: Wide-Area Determination of Cloud
Microphysical Properties from NOAA AVHRR Measurements for FIRE and ASTEX
Regions, J. Atmos. Sci., 52, 4043–4059, 1995.</mixed-citation></ref>
      <ref id="bib1.bib41"><label>41</label><mixed-citation>
Nakajima, T., Tonna, G., Rao, R., Kaufman, Y., and Holben, B.: Use of sky
brightness measurements from ground for remote sensing of particulate
polydispersions, Appl. Opt., 35, 2672–2686, 1996.</mixed-citation></ref>
      <ref id="bib1.bib42"><label>42</label><mixed-citation>
Nottrott, A. and Kleissi, J.: Validation of the NSRDB–SUNY global horizontal
irradiance in California, Sol. Energy, 84, 1816–1827, 2010.</mixed-citation></ref>
      <ref id="bib1.bib43"><label>43</label><mixed-citation>Nunez, M., Serrano, A., and Cancillo, M. L.: Hourly global irradiance from
satellite data in Badajoz, Spain: Spatial and temporal dependence, J.
Geophys. Res.-Atmos., 118, 3650–3661, <ext-link xlink:href="https://doi.org/10.1002/jgrd.50366" ext-link-type="DOI">10.1002/jgrd.50366</ext-link>, 2013.</mixed-citation></ref>
      <ref id="bib1.bib44"><label>44</label><mixed-citation>
Ohtake, H., Shimose, K., Silva-Fonseca Jr., J. G., Takashima, T., Oozeki, T.,
and Yamada, Y.: Accuracy of the solar irradiance forecasts of the Japan
Meteorological Agency mesoscale model for the Kanto region, Japan, Sol.
Energy, 98, 138–152, 2013.</mixed-citation></ref>
      <ref id="bib1.bib45"><label>45</label><mixed-citation>
Ohtake, H., Silva-Fonseca Jr., J. G., Takashima, T., Oozeki, T., Shimose, K.,
and Yamada, Y.: Regional and seasonal characteristics of global horizontal
irradiance forecasts obtained from the Japan Meteorological Agency mesoscale
model, Sol. Energy, 116, 83–99, 2015.</mixed-citation></ref>
      <ref id="bib1.bib46"><label>46</label><mixed-citation>Okamura, R., Iwabuchi, H., and Schmidt, K. S.: Feasibility study of
multi-pixel retrieval of optical thickness and droplet effective radius of
inhomogeneous clouds using deep learning, Atmos. Meas. Tech., 10, 4747–4759,
<ext-link xlink:href="https://doi.org/10.5194/amt-10-4747-2017" ext-link-type="DOI">10.5194/amt-10-4747-2017</ext-link>, 2017.</mixed-citation></ref>
      <ref id="bib1.bib47"><label>47</label><mixed-citation>Papadimas, C. D., Hatzianastassiou, N., Matsoukas, C., Kanakidou, M.,
Mihalopoulos, N., and Vardavas, I.: The direct effect of aerosols on solar
radiation over the broader Mediterranean basin, Atmos. Chem. Phys., 12,
7165–7185, <ext-link xlink:href="https://doi.org/10.5194/acp-12-7165-2012" ext-link-type="DOI">10.5194/acp-12-7165-2012</ext-link>, 2012.</mixed-citation></ref>
      <ref id="bib1.bib48"><label>48</label><mixed-citation>
Pecenak, Z. K., Mejia, F. A., Kurtz, B., Evan, A., and Kleissl, J.:
Simulating irradiance enhancement dependence on cloud optical depth and solar
zenith angle, Sol. Energy, 136, 675–681, 2016.</mixed-citation></ref>
      <ref id="bib1.bib49"><label>49</label><mixed-citation>
Perez, R., Ineichen, P., Moore, K., Kmiecik, M., Chain, C., George, R., and
Vignola, F.: A new operational satellite-to-irradiance model, Sol. Energy,
73, 307–317, 2002.</mixed-citation></ref>
      <ref id="bib1.bib50"><label>50</label><mixed-citation>
Perez, R., David, M., Hoff, T., Jamaly, M., Kivalov, S., Kleissl, J., Lauret,
P., and Perez, M.: Spatial and Temporal Variability of Solar Energy, Found.
Trends Inf. Ret., 1, 1–44, 2016.</mixed-citation></ref>
      <ref id="bib1.bib51"><label>51</label><mixed-citation>
Piedehierro A. A., Antón, M., Cazorla, A., Alados-Arboledas, L., and
Olmo, F. J.: Evaluation of enhancement events of total solar irradiance
during cloudy conditions at Granada (Southeastern Spain), Atmos. Res.,
135–136, 1–7, 2014.</mixed-citation></ref>
      <ref id="bib1.bib52"><label>52</label><mixed-citation>
Pinker, R. T., and Laszlo, I.: Modeling surface solar irradiance for
satellite applications on global scale, J. Appl. Meteorol., 31, 194–211,
1992.</mixed-citation></ref>
      <ref id="bib1.bib53"><label>53</label><mixed-citation>Polo, J., Wilbert, S., Ruiz-Arias, J. A., Meyer, R., Gueymard, C., Súri,
M., Martín, L., Mieslinger, T., Blanc, P., Grant, I., Boland, J.,
Ineichen, P., Remund, J., Escobar, R., Troccoli, A., Sengupta, M., Nielsen,
K. P., Renne, D., Geuder, N., and Cebecauer, T.: Preliminary survey on
site-adaptation techniques for satellite-derived and reanalysis solar
radiation datasets, Sol. Energy, 132, 25–37,
<ext-link xlink:href="https://doi.org/10.1016/j.solener.2016.03.001" ext-link-type="DOI">10.1016/j.solener.2016.03.001</ext-link>, 2016.</mixed-citation></ref>
      <ref id="bib1.bib54"><label>54</label><mixed-citation>
Qu, Z., Oumbe, A., Blanc, P., Espinar, B., Gesell, G., Gschwind, B.,
Gschwind, B., Kluser, L., Lenevre, M., Sabonet, L., Schroedter Homscheidt,
M., and Wald, L.: Fast radiative transfer parameterisation for assessing the
surface solar irradiance: The Heliosat-4 method, Meteorol. Z., 26, 33–57,
2017.</mixed-citation></ref>
      <ref id="bib1.bib55"><label>55</label><mixed-citation>
Raschke, E., Gratzki, A., and Rieland, M.: Estimates of global radiation at
the ground from the reduced data sets of the International Satellite Cloud
Climatology Project, J. Climate, 7, 205–213, 1987.</mixed-citation></ref>
      <ref id="bib1.bib56"><label>56</label><mixed-citation>
Rigollier, C., Lefévre, M., and Wald, L.: The method Heliosat-2 for
deriving shortwave solar radiation from satellite images, Sol. Energy, 77,
159–169, 2004.</mixed-citation></ref>
      <ref id="bib1.bib57"><label>57</label><mixed-citation>
Ruf, H., Schroedter-Homscheidt, M., Heilscher, G., and Beyer, H. G.:
Quantifying residential PV feed-in power in low voltage grids based on
satellite-derived irradiance data with application to power flow
calculations, Sol. Energy, 135, 692–702, 2016.</mixed-citation></ref>
      <ref id="bib1.bib58"><label>58</label><mixed-citation>
Serrano, A., Antón, M., Cancillo, M. L., and Mateos, V. L.: Daily and
annual variations of erythemal ultraviolet radiation in Southwestern Spain,
Ann. Geophys., 24, 427–441, 2006.</mixed-citation></ref>
      <ref id="bib1.bib59"><label>59</label><mixed-citation>
Shettle, E.: Models of aerosols, clouds and precipitation for atmospheric
propagation studies, in: Atmospheric propagation in the uv, visible, ir and
mm-region and related system aspects, no. 454 in AGARD Conference
Proceedings, 1989.</mixed-citation></ref>
      <ref id="bib1.bib60"><label>60</label><mixed-citation>
Stamnes, K., Tsay, S.-C., Wiscombe, W., and Jayaweera, K.: Numerically stable
algorithm for discrete–ordinate–method radiative transfer in multiple
scattering and emitting layered media, Appl. Opt., 27, 2502–2509, 1988.</mixed-citation></ref>
      <?pagebreak page2521?><ref id="bib1.bib61"><label>61</label><mixed-citation>Takenaka, H., Nakajima, T. Y., Higurashi, A., Higuchi, A., Takamura, T.,
Pinker, R. T., and Nakajima, T.: Estimation of solar radiation using a neural
network based on radiative transfer, J. Geophys. Res., 116, D08215,
<ext-link xlink:href="https://doi.org/10.1029/2009JD013337" ext-link-type="DOI">10.1029/2009JD013337</ext-link>, 2011.</mixed-citation></ref>
      <ref id="bib1.bib62"><label>62</label><mixed-citation>
Taylor, M., Kosmopoulos, P. G., Kazadzis, S., Keramitsoglou, I., and
Kiranoudis, C. T.: Neural network radiative transfer solvers for the
generation of high resolution solar irradiance spectra parameterized by cloud
and aerosol parameters, J. Quant. Spectrosc. Ra., 168, 176–192, 2016.</mixed-citation></ref>
      <ref id="bib1.bib63"><label>63</label><mixed-citation>Tanskanen, A., Lindfors, A., Määttä, A., Krotkov, N., Herman, J.,
Kaurola, J., Koskela, T., Lakkala, K., Fioletov, V., Bernhard, J., McKenzie,
R., Kondo, Y., O'Neill, M., Slaper, H., den Outer, P., Bais, A. F., and
Tamminen, J.: Validation of daily erythemal doses from Ozone Monitoring
Instrument with ground-based UV measurement data, J. Geophys. Res., 112,
D24S44, <ext-link xlink:href="https://doi.org/10.1029/2007JD008830" ext-link-type="DOI">10.1029/2007JD008830</ext-link>, 2007.</mixed-citation></ref>
      <ref id="bib1.bib64"><label>64</label><mixed-citation>Thomas, C., Wey, E., Blanc, P., and Wald, L.: Validation of three
satellite-derived databases of surface solar radiation using measurements
performed at 42 stations in Brazil, Adv. Sci. Res., 13, 81–86, 2016a.
 </mixed-citation></ref><?xmltex \hack{\newpage}?>
      <ref id="bib1.bib65"><label>65</label><mixed-citation>
Thomas, C., Wey, E., Blanc, P., Wald, L., and Lefévre, M.: Validation of
HelioClim-3 version 4, HelioClim-3 version 5 and MACC-RAD using 14 BSRN
stations, Enrgy. Proced., 91, 1059–1069, 2016b.</mixed-citation></ref>
      <ref id="bib1.bib66"><label>66</label><mixed-citation>Urraca, R, Gracia-Amillo, A. M., Koubli, E., Huld, T., Trentmann , J.,
Riihelä, A., Lind-fors, A. V., Palmer, D., Gottschalg, R., and
Antonanzas-Torres, F.: Extensive validation of CM SAF surface radiation
products over Europe, Remote Sens. Environ., 199, 171–186,
<ext-link xlink:href="https://doi.org/10.1016/j.rse.2017.07.013" ext-link-type="DOI">10.1016/j.rse.2017.07.013</ext-link>, 2017.</mixed-citation></ref>
      <ref id="bib1.bib67"><label>67</label><mixed-citation>Wang, C., Jeong, G. R., and Mahowald, N.: Particulate absorption of solar
radiation: anthropogenic aerosols vs. dust, Atmos. Chem. Phys., 9,
3935–3945, <ext-link xlink:href="https://doi.org/10.5194/acp-9-3935-2009" ext-link-type="DOI">10.5194/acp-9-3935-2009</ext-link>, 2009.</mixed-citation></ref>
      <ref id="bib1.bib68"><label>68</label><mixed-citation>Xia, X., Chen, H., Li, Z., Wang, P., and Wang, J.: Significant reduction of
surface solar irradiance induced by aerosols in a suburban region in
northeastern China, J. Geophys. Res., 112, D22S02, <ext-link xlink:href="https://doi.org/10.1029/2006JD007562" ext-link-type="DOI">10.1029/2006JD007562</ext-link>,
2007.</mixed-citation></ref>
      <ref id="bib1.bib69"><label>69</label><mixed-citation>
Zo, I.-S., Jee, J.-B., Lee, K.-T., and Kim, B.-Y.: Analysis of Solar
Radiation on the Surface Estimated from GWNU Solar Radiation Model with
Temporal Resolution of Satellite Cloud Fraction, Asia-Pac. J. Atmos. Sci.,
52, 405–412, 2016.</mixed-citation></ref>

  </ref-list></back>
    <!--<article-title-html>Evaluation of Himawari-8 surface downwelling solar radiation by ground-based measurements</article-title-html>
<abstract-html><p>Observations from the new Japanese
geostationary satellite Himawari-8 permit quasi-real-time estimation of
global shortwave radiation at an unprecedented temporal resolution. However,
accurate comparisons with ground-truthing observations are essential to
assess their uncertainty. In this study, we evaluated the Himawari-8 global
radiation product AMATERASS using observations recorded at four SKYNET
stations in Japan and, for certain analyses, from the surface network of the
Japanese Meteorological Agency in 2016. We found that the spatiotemporal
variability of the satellite estimates was smaller than that of the ground
observations; variability decreased with increases in the time step and
spatial domain. Cloud variability was the main source of uncertainty in the
satellite radiation estimates, followed by direct effects caused by aerosols
and bright albedo. Under all-sky conditions, good agreement was found between
satellite and ground-based data, with a mean bias in the range of
20–30&thinsp;W&thinsp;m<sup>−2</sup> (i.e., AMATERASS overestimated ground observations) and a
root mean square error (RMSE) of approximately 70–80&thinsp;W&thinsp;m<sup>−2</sup>. However,
results depended on the time step used in the validation exercise, on the
spatial domain, and on the different climatological regions. In particular,
the validation performed at 2.5&thinsp;min showed largest deviations and RMSE
values ranging from about 110&thinsp;W&thinsp;m<sup>−2</sup> for the mainland to a maximum of
150&thinsp;W&thinsp;m<sup>−2</sup> in the subtropical region. We also detected a limited
overestimation in the number of clear-sky episodes, particularly at the pixel
level. Overall, satellite-based estimates were higher under overcast
conditions, whereas frequent episodes of cloud-induced enhanced surface
radiation (i.e., measured radiation was greater than expected clear-sky
radiation) tended to reduce this difference. Finally, the total mean bias was
approximately 10–15&thinsp;W&thinsp;m<sup>−2</sup> under clear-sky conditions, mainly because
of overall instantaneous direct aerosol forcing efficiency in the range of
120–150&thinsp;W&thinsp;m<sup>−2</sup> per unit of aerosol optical depth (AOD). A seasonal
anticorrelation between AOD and global radiation differences was evident at
all stations and was also observed within the diurnal cycle.</p></abstract-html>
<ref-html id="bib1.bib1"><label>1</label><mixed-citation>
Antón, M., Cachorro, V. E., Vilaplana, J. M., Toledano, C., Krotkov, N.
A., Arola, A., Serrano, A., and de la Morena, B.: Comparison of UV
irradiances from Aura/Ozone Monitoring Instrument (OMI) with Brewer
measurements at El Arenosillo (Spain) – Part 1: Analysis of parameter
influence, Atmos. Chem. Phys., 10, 5979–5989,
<a href="https://doi.org/10.5194/acp-10-5979-2010" target="_blank">https://doi.org/10.5194/acp-10-5979-2010</a>, 2010.
</mixed-citation></ref-html>
<ref-html id="bib1.bib2"><label>2</label><mixed-citation>
Bech, J., Sola, Y., Ossó, A., and Lorente, J.: Analysis of 14 years of
broadband ground-based solar UV index observations in Barcelona, Int. J.
Climatol., 35, 45–56, <a href="https://doi.org/10.1002/joc.3961" target="_blank">https://doi.org/10.1002/joc.3961</a>, 2015.
</mixed-citation></ref-html>
<ref-html id="bib1.bib3"><label>3</label><mixed-citation>
Bessho, K., Date, K., Hayashi, M., Ikeda, A., Imai, T., Inoue, H., Kumagai,
Y., Miyakawa, T., Murata, H., Ohno, T., Okuyama, A., Oyama, R., Sasaki, Y.,
Shimazu, Y., Shimoji, K., and Sumida, Y.: An Introduction to Himawari-8/9
— Japan's New-Generation Geostationary Meteorological Satellites, J.
Meteorol. Soc. Jpn., 94, 151–183, 2016.
</mixed-citation></ref-html>
<ref-html id="bib1.bib4"><label>4</label><mixed-citation>
Blanc, P., Gschwind, B., Lefévre, M., and Wald, L.: The HelioClim
Project: Surface Solar Irradiance Data for Climate Applications, Remote
Sens., 3, 343–361, <a href="https://doi.org/10.3390/rs3020343" target="_blank">https://doi.org/10.3390/rs3020343</a>, 2011.
</mixed-citation></ref-html>
<ref-html id="bib1.bib5"><label>5</label><mixed-citation>
Cachorro, V. E., Toledano, C., Prats, N., Sorribas, M., Mogo, S., Berjón,
A., Torres, B., Rodrigo, R., de la Rosa, J., and De Frutos, A. M.: The
strongest desert dust intrusion mixed with smoke over the Iberian Peninsula
registered with Sun photometry, J. Geophys. Res., 113, D14S04,
<a href="https://doi.org/10.1029/2007JD009582" target="_blank">https://doi.org/10.1029/2007JD009582</a>, 2008.
</mixed-citation></ref-html>
<ref-html id="bib1.bib6"><label>6</label><mixed-citation>
Castelli, M., Stöckli, R., Tetzlaff, A., Wagner, J. E., Belluardo, G.,
Zebisch, M., and Petitta M.: The HelioMont method for assessing solar
irradiance over complex terrain: validation and improvements, Remote Sens.
Environ., 152, 603–613, <a href="https://doi.org/10.1016/j.rse.2014.07.018" target="_blank">https://doi.org/10.1016/j.rse.2014.07.018</a>, 2014.
</mixed-citation></ref-html>
<ref-html id="bib1.bib7"><label>7</label><mixed-citation>
Chan, M. A. and Comiso, J. C.: Arctic cloud characteristics as derived from
MODIS, CALIPSO, and CloudSat, J. Climate, 26, 3285–3306, 2013.
</mixed-citation></ref-html>
<ref-html id="bib1.bib8"><label>8</label><mixed-citation>
Cordero, R. R., Damiani, A., Seckmeyer, G., Jorquera, J., Caballero, M.,
Rowe, P., Ferrer, J., Mubarak, R., Carrasco, J., Rondanelli, R., Matus, M.,
and Laroze, D.: The solar spectrum in the Atacama Desert, Sci. Rep.-UK 6,
22457, <a href="https://doi.org/10.1038/srep22457" target="_blank">https://doi.org/10.1038/srep22457</a>, 2016.
</mixed-citation></ref-html>
<ref-html id="bib1.bib9"><label>9</label><mixed-citation>
Cordero, R. R., Seckmeyer, G., Damiani, A., Riechelmann, S., Rayas, J.,
Labbe, F., and Laroze, D.: The world's highest levels of surface UV, Photoch.
Photobio. Sci., 13, 70–81, <a href="https://doi.org/10.1039/C3PP50221J" target="_blank">https://doi.org/10.1039/C3PP50221J</a>, 2014.
</mixed-citation></ref-html>
<ref-html id="bib1.bib10"><label>10</label><mixed-citation>
Damiani, A., De Simone, S., Rafanelli C., Cordero, R. R., and Laurenza, M.:
Three years of ground-based total ozone measurements in Arctic: comparison
with OMI, GOME and SCIAMACHY satellite data, Remote Sens. Environ., 127,
162–180, <a href="https://doi.org/10.1016/j.rse.2012.08.023" target="_blank">https://doi.org/10.1016/j.rse.2012.08.023</a>, 2012.
</mixed-citation></ref-html>
<ref-html id="bib1.bib11"><label>11</label><mixed-citation>
Damiani, A., Cabrera, S., Muñoz, R., Cordero, R. R., and Labbe, F.:
Satellite-derived UV irradiance for a region with complex morphology and
meteorology: comparison against ground measurements in Santiago de Chile,
Int. J. Remote Sens., 34, 5812–5833, <a href="https://doi.org/10.1080/01431161.2013.796101" target="_blank">https://doi.org/10.1080/01431161.2013.796101</a>, 2013.
</mixed-citation></ref-html>
<ref-html id="bib1.bib12"><label>12</label><mixed-citation>
Damiani, A., Cordero, R., Carrasco, J., Watanabe, S., Kawamiya, M., and
Lagun, V. E.: Changes in the UV Lambertian Equivalent Reflectivity in the
Southern Ocean: influence of sea ice and cloudiness, Remote Sens. Environ.,
169, 75–92, 2015.
</mixed-citation></ref-html>
<ref-html id="bib1.bib13"><label>13</label><mixed-citation>
Djebbar, R., Morris, R., Thevenard, D., Perez, R., and Schlemmerd, J.:
Assessment of SUNY version 3 global horizontal and direct normal solar
irradiance in Canada, Enrgy. Proced., 30, 1274–1283, 2012.
</mixed-citation></ref-html>
<ref-html id="bib1.bib14"><label>14</label><mixed-citation>
Dim, J. R., Takamura, T., Higurashi, A., Kathri, P., Kikuchi, N., and
Nakajima, T. Y.: Validation of Two MODIS Aerosols Algorithms with SKYNET and
Prospects for Future Climate Satellites Such as the GCOM-C/SGLI, Adv.
Meteorol., 2013, 508064, <a href="https://doi.org/10.1155/2013/508064" target="_blank">https://doi.org/10.1155/2013/508064</a>, 2013.
</mixed-citation></ref-html>
<ref-html id="bib1.bib15"><label>15</label><mixed-citation>
Di Biagio, C., di Sarra, A., Meloni, D., Monteleone, F., Piacentino, S., and
Sferlazzo, D.: Measurements of Mediterranean aerosol radiative forcing and
influence of the single scattering albedo, J. Geophys. Res., 114, D06211,
<a href="https://doi.org/10.1029/2008JD011037" target="_blank">https://doi.org/10.1029/2008JD011037</a>, 2009.
</mixed-citation></ref-html>
<ref-html id="bib1.bib16"><label>16</label><mixed-citation>
Di Biagio, C., di Sarra, A., and Meloni, D.: Large atmospheric shortwave
radiative forcing by Mediterranean aerosols derived from simultaneous
ground-based and spaceborne observations and dependence on the aerosol type
and single scattering albedo, J. Geophys. Res., 115, D10209,
<a href="https://doi.org/10.1029/2009JD012697" target="_blank">https://doi.org/10.1029/2009JD012697</a>, 2010.
</mixed-citation></ref-html>
<ref-html id="bib1.bib17"><label>17</label><mixed-citation>
Dürr, B., Zelenka, A., Müller, R., and Philipona, R.: Verification of
CM-SAF and MeteoSwiss satellite based retrievals of surface shortwave
irradiance over the Alpine region, Int. J. Remote Sens., 31, 4179–4198,
2010.
</mixed-citation></ref-html>
<ref-html id="bib1.bib18"><label>18</label><mixed-citation>
Engeland, K., Borga, M., Creutin, J.-D., François, B., Ramos, M.-H., and
Vidal, J.-P.: Space-time variability of climate variables and intermittent
renewable electricity production – A review, Renew. Sustain. Energ. Rev.,
79, 600–617, 2017.
</mixed-citation></ref-html>
<ref-html id="bib1.bib19"><label>19</label><mixed-citation>
Federico, S., Torcasio, R. C., Sanò, P., Casella, D., Campanelli, M.,
Meirink, J. F., Wang, P., Vergari, S., Diémoz, H., and Dietrich, S.:
Comparison of hourly surface downwelling solar radiation estimated from
MSG–SEVIRI and forecast by the RAMS model with pyranometers over Italy,
Atmos. Meas. Tech., 10, 2337–2352, <a href="https://doi.org/10.5194/amt-10-2337-2017" target="_blank">https://doi.org/10.5194/amt-10-2337-2017</a>,
2017.
</mixed-citation></ref-html>
<ref-html id="bib1.bib20"><label>20</label><mixed-citation>
Fukuda, S., Nakajima, T., Takenaka, H., Higurashi, A., Kikuchi, N., Nakajima,
T. Y., and Ishida, H.: New approaches to removing cloud sha- dows and
evaluating the 380 nm surface reflectance for improved aerosol optical
thickness retrievals from the GOSAT/TANSO-Cloud and Aerosol Imager, J.
Geophys. Res.-Atmos, 118, 13–520, <a href="https://doi.org/10.1002/2013JD020090" target="_blank">https://doi.org/10.1002/2013JD020090</a>, 2013.
</mixed-citation></ref-html>
<ref-html id="bib1.bib21"><label>21</label><mixed-citation>
García, O. E., Díaz, A. M., Expósito, F. J., Díaz, J. P.,
Gröbner, J., and Fioletov, V. E.: Cloudless aerosol forcing efficiency in
the UV region from AERONET and WOUDC databases, Geophys. Res. Lett., 33,
L23803, <a href="https://doi.org/10.1029/2006GL026794" target="_blank">https://doi.org/10.1029/2006GL026794</a>, 2006.
</mixed-citation></ref-html>
<ref-html id="bib1.bib22"><label>22</label><mixed-citation>
Gómez, I., Caselles, V., and Estrela, M. J.: Seasonal Characterization of
Solar Radiation Estimates Obtained from a MSG- SEVIRI-Derived Dataset and a
RAMS-Based Operational Forecasting System over the Western Mediterranean
Coast, Remote Sens., 8, 46, <a href="https://doi.org/10.3390/rs8010046" target="_blank">https://doi.org/10.3390/rs8010046</a>, 2016.
</mixed-citation></ref-html>
<ref-html id="bib1.bib23"><label>23</label><mixed-citation>
Greuell, W., Meirink, J. F., and Wang, P.: Retrieval and validation of
global, direct, and diffuse irradiance derived from SEVIRI satellite
observations, J. Geophys. Res., 118, 2340–2361, <a href="https://doi.org/10.1002/jgrd.50194" target="_blank">https://doi.org/10.1002/jgrd.50194</a>,
2013.
</mixed-citation></ref-html>
<ref-html id="bib1.bib24"><label>24</label><mixed-citation>
Gu, L., Fuentes, J. D., Garstang, M., da Silva, J. T., Heitz, R., Sigler, J.,
and Shugart, H. H.: Cloud modulation of surface solar irradiance at a pasture
site in southern Brazil, Agric. For. Meteorol., 106, 117–129, 2001.
</mixed-citation></ref-html>
<ref-html id="bib1.bib25"><label>25</label><mixed-citation>
Gueymard, C. A.: Cloud and albedo enhancement impacts on solar irradiance
using high- frequency measurements from thermopile and photodiode
radiometers, Part 1: Impacts on global horizontal irradiance, Sol. Energy,
153, 755–765, 2017.
</mixed-citation></ref-html>
<ref-html id="bib1.bib26"><label>26</label><mixed-citation>
Hashimoto, M. and Nakajima, T.: Development of a remote sensing algorithm to
retrieve atmospheric aerosol properties using multiwave- length and
multipixel information, J. Geophys. Res.-Atmos., 122, 6347–6378,
<a href="https://doi.org/10.1002/2016JD025698" target="_blank">https://doi.org/10.1002/2016JD025698</a>, 2017.
</mixed-citation></ref-html>
<ref-html id="bib1.bib27"><label>27</label><mixed-citation>
Hoff, T. E. and Perez, R.: Quantifying PV power Output Variability, Sol.
Energy, 84, 1782–1793, 2010.
</mixed-citation></ref-html>
<ref-html id="bib1.bib28"><label>28</label><mixed-citation>
Huttunen, J., Arola, A., Myhre, G., Lindfors, A. V., Mielonen, T., Mikkonen,
S., Schafer, J. S., Tripathi, S. N., Wild, M., Komppula, M., and Lehtinen, K.
E. J.: Effect of water vapor on the determination of aerosol direct radiative
effect based on the AERONET fluxes, Atmos. Chem. Phys., 14, 6103–6110,
<a href="https://doi.org/10.5194/acp-14-6103-2014" target="_blank">https://doi.org/10.5194/acp-14-6103-2014</a>, 2014.
</mixed-citation></ref-html>
<ref-html id="bib1.bib29"><label>29</label><mixed-citation>
Higurashi, A. and Nakajima, T.: Development of a two-channel aerosol
retrieval algorithm on global scale using NOAA/AVHRR, J. Atmos. Sci., 56,
924–941, 1999.
</mixed-citation></ref-html>
<ref-html id="bib1.bib30"><label>30</label><mixed-citation>
Inman, R. H., Chu, Y., and Coimbra, C. F. M.: Cloud enhancement of global
horizontal irradiance in California and Hawaii, Sol. Energy 130, 128–138,
2016.
</mixed-citation></ref-html>
<ref-html id="bib1.bib31"><label>31</label><mixed-citation>
Irie, H., Horio T., Damiani, A., Nakajima, T. Y., Takenaka, H., Kikuchi, M.,
Khatri, P., and Yumimoto, K.: Importance of Himawari-8 aerosol products for
energy management system, Earozoru Kenkyu, 32, 95–100, 2017.
</mixed-citation></ref-html>
<ref-html id="bib1.bib32"><label>32</label><mixed-citation>
Iwabuchi, H.: Efficient Monte Carlo Methods for Radiative Transfer Modeling,
J. Atmos. Sci., 63, 2324–2339, <a href="https://doi.org/10.1175/JAS3755.1" target="_blank">https://doi.org/10.1175/JAS3755.1</a>, 2006.
</mixed-citation></ref-html>
<ref-html id="bib1.bib33"><label>33</label><mixed-citation>
Kathri, P. and Takamura, T.: An Algorithm to Screen Cloud-Affected Data for
Sky Radiometer Data Analysis, J. Meteorol. Soc. Jpn., 87, 189–204,
<a href="https://doi.org/10.2151/jmsj.87.189" target="_blank">https://doi.org/10.2151/jmsj.87.189</a>, 2009.
</mixed-citation></ref-html>
<ref-html id="bib1.bib34"><label>34</label><mixed-citation>
Khatri, P., Irie, H., Takamura, T., Takenaka, H., and Kurino, T.: Validation
of global fluxes estimated from geostationary satellite HIMAWARI data using
ground-truth data, Autumn meeting of Meteorological Society of Japan, Kyoto,
Japan, 28–30 October 2015.
</mixed-citation></ref-html>
<ref-html id="bib1.bib35"><label>35</label><mixed-citation>
Kawamoto, K., Nakajima, T., and Nakajima, T. Y.: A global determination of
cloud microphysics with AVHRR remote sensing, J. Climate, 14, 2054–2068,
2001.
</mixed-citation></ref-html>
<ref-html id="bib1.bib36"><label>36</label><mixed-citation>
Kosmopoulos, P. G., Kazadzis, S., Taylor, M., Raptis, P. I., Keramitsoglou,
I., Kiranoudis, C., and Bais, A. F.: Assessment of surface solar irradiance
derived from real-time modelling techniques and verification with
ground-based measurements, Atmos. Meas. Tech., 11, 907–924,
<a href="https://doi.org/10.5194/amt-11-907-2018" target="_blank">https://doi.org/10.5194/amt-11-907-2018</a>, 2018.
</mixed-citation></ref-html>
<ref-html id="bib1.bib37"><label>37</label><mixed-citation>
Kudo, R., Uchiyama, A., Yamazaki, A., and Kobayashi, E.: Seasonal
characteristics of aerosol radiative effect estimated from ground-based solar
radiation measurements in Tsukuba, Japan, J. Geophys. Res., 115, D01204,
<a href="https://doi.org/10.1029/2009JD012487" target="_blank">https://doi.org/10.1029/2009JD012487</a>, 2010.
</mixed-citation></ref-html>
<ref-html id="bib1.bib38"><label>38</label><mixed-citation>
Madhavan, B. L., Deneke, H., Witthuhn, J., and Macke, A.: Multiresolution
analysis of the spatiotemporal variability in global radiation observed by a
dense network of 99 pyranometers, Atmos. Chem. Phys., 17, 3317–3338,
<a href="https://doi.org/10.5194/acp-17-3317-2017" target="_blank">https://doi.org/10.5194/acp-17-3317-2017</a>, 2017.
</mixed-citation></ref-html>
<ref-html id="bib1.bib39"><label>39</label><mixed-citation>
Mok, J., Krotkov, N. A., Torres, O., Jethva, H., Li, Z., Kim, J., Koo, J.-H.,
Go, S., Irie, H., Labow, G., Eck, T. F., Holben, B. N., Herman, J., Loughman,
R. P., Spinei, E., Lee, S. S., Khatri, P., and Campanelli, M.: Comparisons of
spectral aerosol absorption in Seoul, South Korea, Atmos. Meas. Tech.
Discuss., <a href="https://doi.org/10.5194/amt-2017-380" target="_blank">https://doi.org/10.5194/amt-2017-380</a>, in review, 2017.
</mixed-citation></ref-html>
<ref-html id="bib1.bib40"><label>40</label><mixed-citation>
Nakajima, T. Y. and Nakajima, T.: Wide-Area Determination of Cloud
Microphysical Properties from NOAA AVHRR Measurements for FIRE and ASTEX
Regions, J. Atmos. Sci., 52, 4043–4059, 1995.
</mixed-citation></ref-html>
<ref-html id="bib1.bib41"><label>41</label><mixed-citation>
Nakajima, T., Tonna, G., Rao, R., Kaufman, Y., and Holben, B.: Use of sky
brightness measurements from ground for remote sensing of particulate
polydispersions, Appl. Opt., 35, 2672–2686, 1996.
</mixed-citation></ref-html>
<ref-html id="bib1.bib42"><label>42</label><mixed-citation>
Nottrott, A. and Kleissi, J.: Validation of the NSRDB–SUNY global horizontal
irradiance in California, Sol. Energy, 84, 1816–1827, 2010.
</mixed-citation></ref-html>
<ref-html id="bib1.bib43"><label>43</label><mixed-citation>
Nunez, M., Serrano, A., and Cancillo, M. L.: Hourly global irradiance from
satellite data in Badajoz, Spain: Spatial and temporal dependence, J.
Geophys. Res.-Atmos., 118, 3650–3661, <a href="https://doi.org/10.1002/jgrd.50366" target="_blank">https://doi.org/10.1002/jgrd.50366</a>, 2013.
</mixed-citation></ref-html>
<ref-html id="bib1.bib44"><label>44</label><mixed-citation>
Ohtake, H., Shimose, K., Silva-Fonseca Jr., J. G., Takashima, T., Oozeki, T.,
and Yamada, Y.: Accuracy of the solar irradiance forecasts of the Japan
Meteorological Agency mesoscale model for the Kanto region, Japan, Sol.
Energy, 98, 138–152, 2013.
</mixed-citation></ref-html>
<ref-html id="bib1.bib45"><label>45</label><mixed-citation>
Ohtake, H., Silva-Fonseca Jr., J. G., Takashima, T., Oozeki, T., Shimose, K.,
and Yamada, Y.: Regional and seasonal characteristics of global horizontal
irradiance forecasts obtained from the Japan Meteorological Agency mesoscale
model, Sol. Energy, 116, 83–99, 2015.
</mixed-citation></ref-html>
<ref-html id="bib1.bib46"><label>46</label><mixed-citation>
Okamura, R., Iwabuchi, H., and Schmidt, K. S.: Feasibility study of
multi-pixel retrieval of optical thickness and droplet effective radius of
inhomogeneous clouds using deep learning, Atmos. Meas. Tech., 10, 4747–4759,
<a href="https://doi.org/10.5194/amt-10-4747-2017" target="_blank">https://doi.org/10.5194/amt-10-4747-2017</a>, 2017.
</mixed-citation></ref-html>
<ref-html id="bib1.bib47"><label>47</label><mixed-citation>
Papadimas, C. D., Hatzianastassiou, N., Matsoukas, C., Kanakidou, M.,
Mihalopoulos, N., and Vardavas, I.: The direct effect of aerosols on solar
radiation over the broader Mediterranean basin, Atmos. Chem. Phys., 12,
7165–7185, <a href="https://doi.org/10.5194/acp-12-7165-2012" target="_blank">https://doi.org/10.5194/acp-12-7165-2012</a>, 2012.
</mixed-citation></ref-html>
<ref-html id="bib1.bib48"><label>48</label><mixed-citation>
Pecenak, Z. K., Mejia, F. A., Kurtz, B., Evan, A., and Kleissl, J.:
Simulating irradiance enhancement dependence on cloud optical depth and solar
zenith angle, Sol. Energy, 136, 675–681, 2016.
</mixed-citation></ref-html>
<ref-html id="bib1.bib49"><label>49</label><mixed-citation>
Perez, R., Ineichen, P., Moore, K., Kmiecik, M., Chain, C., George, R., and
Vignola, F.: A new operational satellite-to-irradiance model, Sol. Energy,
73, 307–317, 2002.
</mixed-citation></ref-html>
<ref-html id="bib1.bib50"><label>50</label><mixed-citation>
Perez, R., David, M., Hoff, T., Jamaly, M., Kivalov, S., Kleissl, J., Lauret,
P., and Perez, M.: Spatial and Temporal Variability of Solar Energy, Found.
Trends Inf. Ret., 1, 1–44, 2016.
</mixed-citation></ref-html>
<ref-html id="bib1.bib51"><label>51</label><mixed-citation>
Piedehierro A. A., Antón, M., Cazorla, A., Alados-Arboledas, L., and
Olmo, F. J.: Evaluation of enhancement events of total solar irradiance
during cloudy conditions at Granada (Southeastern Spain), Atmos. Res.,
135–136, 1–7, 2014.
</mixed-citation></ref-html>
<ref-html id="bib1.bib52"><label>52</label><mixed-citation>
Pinker, R. T., and Laszlo, I.: Modeling surface solar irradiance for
satellite applications on global scale, J. Appl. Meteorol., 31, 194–211,
1992.
</mixed-citation></ref-html>
<ref-html id="bib1.bib53"><label>53</label><mixed-citation>
Polo, J., Wilbert, S., Ruiz-Arias, J. A., Meyer, R., Gueymard, C., Súri,
M., Martín, L., Mieslinger, T., Blanc, P., Grant, I., Boland, J.,
Ineichen, P., Remund, J., Escobar, R., Troccoli, A., Sengupta, M., Nielsen,
K. P., Renne, D., Geuder, N., and Cebecauer, T.: Preliminary survey on
site-adaptation techniques for satellite-derived and reanalysis solar
radiation datasets, Sol. Energy, 132, 25–37,
<a href="https://doi.org/10.1016/j.solener.2016.03.001" target="_blank">https://doi.org/10.1016/j.solener.2016.03.001</a>, 2016.
</mixed-citation></ref-html>
<ref-html id="bib1.bib54"><label>54</label><mixed-citation>
Qu, Z., Oumbe, A., Blanc, P., Espinar, B., Gesell, G., Gschwind, B.,
Gschwind, B., Kluser, L., Lenevre, M., Sabonet, L., Schroedter Homscheidt,
M., and Wald, L.: Fast radiative transfer parameterisation for assessing the
surface solar irradiance: The Heliosat-4 method, Meteorol. Z., 26, 33–57,
2017.
</mixed-citation></ref-html>
<ref-html id="bib1.bib55"><label>55</label><mixed-citation>
Raschke, E., Gratzki, A., and Rieland, M.: Estimates of global radiation at
the ground from the reduced data sets of the International Satellite Cloud
Climatology Project, J. Climate, 7, 205–213, 1987.
</mixed-citation></ref-html>
<ref-html id="bib1.bib56"><label>56</label><mixed-citation>
Rigollier, C., Lefévre, M., and Wald, L.: The method Heliosat-2 for
deriving shortwave solar radiation from satellite images, Sol. Energy, 77,
159–169, 2004.
</mixed-citation></ref-html>
<ref-html id="bib1.bib57"><label>57</label><mixed-citation>
Ruf, H., Schroedter-Homscheidt, M., Heilscher, G., and Beyer, H. G.:
Quantifying residential PV feed-in power in low voltage grids based on
satellite-derived irradiance data with application to power flow
calculations, Sol. Energy, 135, 692–702, 2016.
</mixed-citation></ref-html>
<ref-html id="bib1.bib58"><label>58</label><mixed-citation>
Serrano, A., Antón, M., Cancillo, M. L., and Mateos, V. L.: Daily and
annual variations of erythemal ultraviolet radiation in Southwestern Spain,
Ann. Geophys., 24, 427–441, 2006.
</mixed-citation></ref-html>
<ref-html id="bib1.bib59"><label>59</label><mixed-citation>
Shettle, E.: Models of aerosols, clouds and precipitation for atmospheric
propagation studies, in: Atmospheric propagation in the uv, visible, ir and
mm-region and related system aspects, no. 454 in AGARD Conference
Proceedings, 1989.
</mixed-citation></ref-html>
<ref-html id="bib1.bib60"><label>60</label><mixed-citation>
Stamnes, K., Tsay, S.-C., Wiscombe, W., and Jayaweera, K.: Numerically stable
algorithm for discrete–ordinate–method radiative transfer in multiple
scattering and emitting layered media, Appl. Opt., 27, 2502–2509, 1988.
</mixed-citation></ref-html>
<ref-html id="bib1.bib61"><label>61</label><mixed-citation>
Takenaka, H., Nakajima, T. Y., Higurashi, A., Higuchi, A., Takamura, T.,
Pinker, R. T., and Nakajima, T.: Estimation of solar radiation using a neural
network based on radiative transfer, J. Geophys. Res., 116, D08215,
<a href="https://doi.org/10.1029/2009JD013337" target="_blank">https://doi.org/10.1029/2009JD013337</a>, 2011.
</mixed-citation></ref-html>
<ref-html id="bib1.bib62"><label>62</label><mixed-citation>
Taylor, M., Kosmopoulos, P. G., Kazadzis, S., Keramitsoglou, I., and
Kiranoudis, C. T.: Neural network radiative transfer solvers for the
generation of high resolution solar irradiance spectra parameterized by cloud
and aerosol parameters, J. Quant. Spectrosc. Ra., 168, 176–192, 2016.
</mixed-citation></ref-html>
<ref-html id="bib1.bib63"><label>63</label><mixed-citation>
Tanskanen, A., Lindfors, A., Määttä, A., Krotkov, N., Herman, J.,
Kaurola, J., Koskela, T., Lakkala, K., Fioletov, V., Bernhard, J., McKenzie,
R., Kondo, Y., O'Neill, M., Slaper, H., den Outer, P., Bais, A. F., and
Tamminen, J.: Validation of daily erythemal doses from Ozone Monitoring
Instrument with ground-based UV measurement data, J. Geophys. Res., 112,
D24S44, <a href="https://doi.org/10.1029/2007JD008830" target="_blank">https://doi.org/10.1029/2007JD008830</a>, 2007.
</mixed-citation></ref-html>
<ref-html id="bib1.bib64"><label>64</label><mixed-citation>
Thomas, C., Wey, E., Blanc, P., and Wald, L.: Validation of three
satellite-derived databases of surface solar radiation using measurements
performed at 42 stations in Brazil, Adv. Sci. Res., 13, 81–86, 2016a.

</mixed-citation></ref-html>
<ref-html id="bib1.bib65"><label>65</label><mixed-citation>
Thomas, C., Wey, E., Blanc, P., Wald, L., and Lefévre, M.: Validation of
HelioClim-3 version 4, HelioClim-3 version 5 and MACC-RAD using 14 BSRN
stations, Enrgy. Proced., 91, 1059–1069, 2016b.
</mixed-citation></ref-html>
<ref-html id="bib1.bib66"><label>66</label><mixed-citation>
Urraca, R, Gracia-Amillo, A. M., Koubli, E., Huld, T., Trentmann , J.,
Riihelä, A., Lind-fors, A. V., Palmer, D., Gottschalg, R., and
Antonanzas-Torres, F.: Extensive validation of CM SAF surface radiation
products over Europe, Remote Sens. Environ., 199, 171–186,
<a href="https://doi.org/10.1016/j.rse.2017.07.013" target="_blank">https://doi.org/10.1016/j.rse.2017.07.013</a>, 2017.
</mixed-citation></ref-html>
<ref-html id="bib1.bib67"><label>67</label><mixed-citation>
Wang, C., Jeong, G. R., and Mahowald, N.: Particulate absorption of solar
radiation: anthropogenic aerosols vs. dust, Atmos. Chem. Phys., 9,
3935–3945, <a href="https://doi.org/10.5194/acp-9-3935-2009" target="_blank">https://doi.org/10.5194/acp-9-3935-2009</a>, 2009.
</mixed-citation></ref-html>
<ref-html id="bib1.bib68"><label>68</label><mixed-citation>
Xia, X., Chen, H., Li, Z., Wang, P., and Wang, J.: Significant reduction of
surface solar irradiance induced by aerosols in a suburban region in
northeastern China, J. Geophys. Res., 112, D22S02, <a href="https://doi.org/10.1029/2006JD007562" target="_blank">https://doi.org/10.1029/2006JD007562</a>,
2007.
</mixed-citation></ref-html>
<ref-html id="bib1.bib69"><label>69</label><mixed-citation>
Zo, I.-S., Jee, J.-B., Lee, K.-T., and Kim, B.-Y.: Analysis of Solar
Radiation on the Surface Estimated from GWNU Solar Radiation Model with
Temporal Resolution of Satellite Cloud Fraction, Asia-Pac. J. Atmos. Sci.,
52, 405–412, 2016.
</mixed-citation></ref-html>--></article>
