<?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">TC</journal-id><journal-title-group>
    <journal-title>The Cryosphere</journal-title>
    <abbrev-journal-title abbrev-type="publisher">TC</abbrev-journal-title><abbrev-journal-title abbrev-type="nlm-ta">The Cryosphere</abbrev-journal-title>
  </journal-title-group><issn pub-type="epub">1994-0424</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/tc-12-1887-2018</article-id><title-group><article-title>Spring snow albedo feedback over northern Eurasia: <?xmltex \hack{\break}?> Comparing in situ measurements with reanalysis products</article-title><alt-title>Spring snow albedo feedback over northern Eurasia</alt-title>
      </title-group><?xmltex \runningtitle{Spring snow albedo feedback over northern Eurasia}?><?xmltex \runningauthor{M.~Wegmann et al.}?>
      <contrib-group>
        <contrib contrib-type="author" corresp="yes" rid="aff1">
          <name><surname>Wegmann</surname><given-names>Martin</given-names></name>
          <email>martin.wegmann@awi.de</email>
        </contrib>
        <contrib contrib-type="author" corresp="no" rid="aff2">
          <name><surname>Dutra</surname><given-names>Emanuel</given-names></name>
          
        <ext-link>https://orcid.org/0000-0002-0643-2643</ext-link></contrib>
        <contrib contrib-type="author" corresp="no" rid="aff1">
          <name><surname>Jacobi</surname><given-names>Hans-Werner</given-names></name>
          
        <ext-link>https://orcid.org/0000-0003-2230-3136</ext-link></contrib>
        <contrib contrib-type="author" corresp="no" rid="aff1 aff3">
          <name><surname>Zolina</surname><given-names>Olga</given-names></name>
          
        </contrib>
        <aff id="aff1"><label>1</label><institution>Institute for Geosciences and Environmental Research (IGE), Univ. Grenoble Alpes, CNRS, IRD,
Grenoble Institute of Engineering Univ. Grenoble Alpes (INP), Grenoble, France</institution>
        </aff>
        <aff id="aff2"><label>2</label><institution>Instituto Dom Luiz, Faculdade de Ciências, Universidade de Lisboa, Lisbon, Portugal</institution>
        </aff>
        <aff id="aff3"><label>3</label><institution>Shirshov Institute of Oceanology, Moscow, Russia</institution>
        </aff>
      </contrib-group>
      <author-notes><corresp id="corr1">Martin Wegmann (martin.wegmann@awi.de)</corresp></author-notes><pub-date><day>6</day><month>June</month><year>2018</year></pub-date>
      
      <volume>12</volume>
      <issue>6</issue>
      <fpage>1887</fpage><lpage>1898</lpage>
      <history>
        <date date-type="received"><day>28</day><month>November</month><year>2017</year></date>
           <date date-type="rev-request"><day>12</day><month>December</month><year>2017</year></date>
           <date date-type="rev-recd"><day>23</day><month>April</month><year>2018</year></date>
           <date date-type="accepted"><day>27</day><month>April</month><year>2018</year></date>
      </history>
      <permissions>
        
        
      <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://tc.copernicus.org/articles/12/1887/2018/tc-12-1887-2018.html">This article is available from https://tc.copernicus.org/articles/12/1887/2018/tc-12-1887-2018.html</self-uri><self-uri xlink:href="https://tc.copernicus.org/articles/12/1887/2018/tc-12-1887-2018.pdf">The full text article is available as a PDF file from https://tc.copernicus.org/articles/12/1887/2018/tc-12-1887-2018.pdf</self-uri>
      <abstract>
    <p id="d1e123">This study uses daily observations and modern reanalyses in order to evaluate
reanalysis products over northern Eurasia regarding the spring snow albedo
feedback (SAF) during the period from 2000 to 2013. We used the
state-of-the-art reanalyses from ERA-Interim/Land and the Modern-Era
Retrospective Analysis for Research and Applications version 2 (MERRA-2) as
well as an experimental set-up of ERA-Interim/Land with prescribed short
grass as land cover to enhance the comparability with the station data while
underlining the caveats of comparing in situ observations with gridded data.
Snow depth statistics derived from daily station data are well reproduced in
all three reanalyses. However day-to-day albedo variability is notably higher
at the stations than for any reanalysis product. The ERA-Interim grass set-up
shows improved performance when representing albedo variability and generates
comparable estimates for the snow albedo in spring. We find that modern
reanalyses show a physically consistent representation of SAF, with realistic
spatial patterns and area-averaged sensitivity estimates. However,
station-based SAF values are significantly higher than in the reanalyses,
which is mostly driven by the stronger contrast between snow and snow-free
albedo. Switching to grass-only vegetation in ERA-Interim/Land increases the
SAF values up to the level of station-based estimates. We found no
significant trend in the examined 14-year time series of SAF, but interannual
changes of about 0.5 % K<inline-formula><mml:math id="M1" 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> in both station-based and reanalysis
estimates were derived. This interannual variability is primarily dominated
by the variability in the snowmelt sensitivity, which is correctly captured
in reanalysis products. Although modern reanalyses perform well for snow
variables, efforts should be made to improve the representation of dynamic
albedo changes.</p>
  </abstract>
    </article-meta>
  </front>
<body>
      

<sec id="Ch1.S1" sec-type="intro">
  <title>Introduction</title>
      <p id="d1e145">Global warming is enhanced at high northern latitudes, where the Arctic
near-surface air temperature has risen at twice the rate of the global
average in recent decades – a feature called Arctic amplification (Serreze
and Barry, 2011). Climate model experiments for the 21st and 22nd centuries
show that Arctic warming will continue and intensify under all emission
scenarios (Collins et al., 2013). Arctic amplification results from several
processes interacting with each other such as the albedo feedback due to a
reduction in snow and ice cover, enhanced poleward atmospheric and oceanic
heat transport, and changes in humidity (Serreze and Barry, 2011; Pithan and
Mauritsen, 2014).</p>
      <p id="d1e148">As one of the critical factors of the Arctic amplification, the surface
albedo feedback implies a decrease in reflected short-wave radiation at the
top of the atmosphere in conjunction with decreasing surface albedo and
increasing near-surface temperature (Thackeray and Fletcher, 2016). It is
considered to be a positive feedback in the sense that an initial warming is
strengthened over time, being quantified through the change in surface albedo
per unit change in temperature (Robock, 1983; Cess and Potter, 1991; Qu and
Hall, 2007). Snowmelt triggers this feedback via surface absorption of
short-wave radiation followed by conversion to<?pagebreak page1888?> long-wave radiation, warming
the lower layers of the troposphere (Curry et al., 1996). Snow albedo
feedback (SAF) and its impact on climate have been studied for several
decades (Wexler et al., 1953; Budyko, 1967; Schneider and Dickinson, 1974;
Lian and Cess, 1977). It received further attention in the wake of
anthropogenic global warming accompanied by the reduction of snow and ice
cover over the Northern Hemisphere (NH) (Bony et al., 2006; Qu and Hall,
2007, 2014; Fernandes et al., 2009; Flanner et al., 2011; Fletcher et al.,
2015; Thackeray and Fletcher, 2016).</p>
      <p id="d1e151">During 1979–2011, the Arctic snow cover extent in June decreased at a rate
of <inline-formula><mml:math id="M2" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>21 % per decade (Derksen and Brown, 2012). Climate model projections
for the end of the 21st century show an even more reduced Arctic cryosphere
and thus the SAF will continue to modulate Arctic warming (Brutel-Vuilmet
et al., 2013). The SAF is especially effective over the NH since most of it
is covered by snow during boreal wintertime (Groisman et al., 1994). Hall
(2004) found that 50 % of the total NH extratropical SAF caused by global
warming occurs during spring, while Qu and Hall (2014) estimated that the
SAF variability between models accounts for 40–50 % of the spread in the
warming signal over the continents of the NH extratropics.</p>
      <p id="d1e161">Several studies investigated spring NH extratropical SAF based on satellite,
reanalysis and model data sets (Fernandes et al., 2009; Fletcher et al.,
2012, 2015; Qu and Hall, 2014). Satellite-based estimates of SAF vary within
<inline-formula><mml:math id="M3" display="inline"><mml:mrow><mml:mo>±</mml:mo><mml:mn mathvariant="normal">10</mml:mn></mml:mrow></mml:math></inline-formula> % depending on the analysed data set. Hall et al. (2008) used
the International Satellite Cloud Climatology Project (ISCCP) data (Schiffer
and Rossow, 1983) to calculate a SAF strength of <inline-formula><mml:math id="M4" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>1.13 % K<inline-formula><mml:math id="M5" 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>,
whereas Fernandes et al. (2009) using Advanced Very High Resolution
Radiometer (AVHRR) data (Justice et al., 1985) found a slightly weaker SAF of
<inline-formula><mml:math id="M6" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>0.93 % K<inline-formula><mml:math id="M7" 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>. Qu and Hall (2014) determined the SAF using Moderate
Resolution Imaging Spectroradiometer (MODIS) data (Hall et al., 2002) and
found a value of <inline-formula><mml:math id="M8" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>0.87 % K<inline-formula><mml:math id="M9" 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> for springtime. By considering
different spatial and temporal domains as well as the variety of methods
applied, the SAF estimates around <inline-formula><mml:math id="M10" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>1 % K<inline-formula><mml:math id="M11" 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>from satellite data can
be considered quantitatively consistent.</p>
      <p id="d1e252">Model- and reanalysis-based estimates are somewhat higher compared to those
derived from satellite data. Fletcher et al. (2015) investigated Coupled
Model Intercomparison Project 3 and 5 (CMIP3/CMIP5) ensembles to estimate
the SAF for an assortment of global climate models (GCMs). The authors found
a SAF ensemble model mean of <inline-formula><mml:math id="M12" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>1.2 % K<inline-formula><mml:math id="M13" 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> for the NH extratropics,
which is in fair agreement with MODIS values but is higher compared to
ISCCP- and AVHHR-based estimates. Within this comparison Fletcher et al. (2015)
also investigated SAF computations based on ERA-Interim (Dee et al.,
2011), Modern-Era Retrospective Analysis for Research and Applications
(MERRA) (Rienecker et al., 2011) and NCEP-2 (Kanamitsu et al., 2002)
reanalyses, thus providing the most up-to-date assessment of SAF in
reanalysis data sets. While MERRA data resulted in a slightly weaker SAF of
<inline-formula><mml:math id="M14" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>1.17 % K<inline-formula><mml:math id="M15" 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> compared to ERA-Interim (<inline-formula><mml:math id="M16" display="inline"><mml:mo lspace="0mm">-</mml:mo></mml:math></inline-formula>1.23 % K<inline-formula><mml:math id="M17" display="inline"><mml:mrow><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">1</mml:mn></mml:mrow></mml:msup><mml:mo>)</mml:mo></mml:mrow></mml:math></inline-formula>, both
reanalyses show similar SAF values compared to MODIS. That said, most
studies use satellite-derived albedo data in conjunction with temperature
and snow cover data from reanalyses.</p>
      <p id="d1e316">Although satellite products of snow cover and albedo cover large parts of the
NH, they exhibit low temporal resolution and significant uncertainties for
high solar zenith angles as well as complex terrains (e.g. Wang et al.,
2014). Thackeray and Fletcher (2016) compared CMIP3/CMIP5 model families and
found that the models represent the SAF process rather accurately. However,
there are still inherent biases likely related to the use of outdated
parameterizations. In this respect the use of in situ observations would
provide an opportunity for evaluating SAF estimates in different gridded
data sets and especially among reanalyses. However, estimating SAF in the
Arctic using in situ data is challenging, mostly because of the lack of
reliable, relevant observations, both in the temporal and spatial domains.
Furthermore, the lack of in situ SAF estimates hampers the understanding of
SAF in high-latitude climates (Graversen and Wang, 2009; Gravesen et al.,
2014).</p>
      <p id="d1e319">In this study we use a unique data set of daily observations and modern
reanalyses over northern Eurasia in order to (1) evaluate reanalysis
products with respect to radiation and snow properties and (2) determine
the SAF in spring between 2000 and 2013 based on in situ measurements. We
compare different land-reanalysis products with modified vegetation
settings. Specific questions to be addressed in this study are the
following. How well do the modern reanalyses reproduce snow and radiation
features on a daily resolution? What are realistic estimates of the SAF from
the station data over northern Eurasia and how well do they compare to the
gridded reanalyses data? What are the major characteristics of space–time
variability of the SAF in station and reanalysis data?</p>
      <p id="d1e322">The paper is organized as follows. After describing the different data sets
and the methods in Sects. 2 and 3, we evaluate the daily output for snow,
radiation fluxes and temperature within these data sets in Sect. 4.1. In
Sect. 4.2 we assess the results of the SAF computations and the
differences between products including an analysis of the spatial and
temporal variability. Section 5 discusses the results and considers
potential implications for future studies.</p>
</sec>
<sec id="Ch1.S2">
  <title>Data</title>
<sec id="Ch1.S2.SS1">
  <title>Reanalysis data</title>
      <p id="d1e336">To investigate the SAF processes in reanalyses, we evaluated two products:
the ERA-Interim/Land (ERAI-L, Balsamo et al., 2015) and Modern-Era
Retrospective analysis for Research and Applications, version 2 (MERRA-2)
(Gelaro et al., 2017). ERAI-L is a land-surface-only simulation driven<?pagebreak page1889?> by the
near-surface meteorology and fluxes from the ERA-Interim atmospheric
reanalyses (Dee et al., 2011). The land-surface model in ERAI-L (HTESSEL) has
several enhancements compared with the land-surface model used in
ERA-Interim including the snowpack representation (Dutra et al., 2010).
ERAI-L considers the prognostic evolution of snow mass and density, and for
exposed areas there is also a prognostic evolution of snow albedo. For
shaded snow, i.e. snow under high vegetation, the albedo is considered
constant and dependent on vegetation type (see Dutra et al., 2010 for more
details).</p>
      <p id="d1e339">MERRA-2 also includes a dedicated land module for surface variables.
Furthermore, it applies an updated Goddard Earth Observing System (GEOS)
model and analysis scheme and assimilates more observations than its
predecessor MERRA (Rienecker et al., 2011). Finally, MERRA-2 uses
observation-based precipitation data to force its land-surface
parameterizations (Reichle et al., 2017), similar to what formerly was known
as MERRA-land. Unlike ERAI-L, MERRA-2 consists of a full land–atmosphere
reanalysis. Its incremental analysis update (IAU) scheme improves upon 3D-Var
by dampening the analysis increment. In IAU, a correction is applied to the
forecast model gradually, limiting precipitation spin-up in particular.</p>
      <p id="d1e342">For near-surface temperature we use 2 m air temperature for both the
reanalyses and observations. Moreover, we do not use albedo computed by the
reanalysis but calculate it from the radiative flux components consistently
with the observed albedo. For this purpose, we use upward and downward
short-wave radiation at the surface as diagnosed by ERA-Interim and MERRA-2 as
well as surface net and surface incoming radiation from the station
observations. Snow depth is used as inferred by reanalyses and, if needed,
converted to centimetres. More information about general characteristics of reanalysis
products in the Arctic can be found in Lindsay et al. (2014), Dufour et
al. (2016) and Wegmann et al. (2017).</p>
<sec id="Ch1.S2.SS1.SSSx1" specific-use="unnumbered">
  <title>Idealized reanalysis experiment</title>
      <p id="d1e350">Since the in situ measurements in this study are observed over clear-cut
vegetation, idealized simulations prescribing grassland everywhere were
carried out with the ERAI-L configuration (hereafter ERA-Interim/Land grass
only, ERAI-LG). The ERAI-LG simulation was carried out with the same model
and set-up as ERAI-L, differing only in the land cover used. The land-surface
model used in ERAI-L, HTESSEL, accounts for subgrid-scale land cover
variability by representing several land tiles, namely low vegetation, high
vegetation, bare ground, exposed snow (snow on top of bare ground or low
vegetation), shaded snow (snow under high vegetation) and interception. The
land cover is prescribed with four maps: low and high vegetation cover (cvl
and cvh) and low and high vegetation types (tvl and tvh). The bare ground
fraction is computed as cvb <inline-formula><mml:math id="M18" display="inline"><mml:mo>=</mml:mo></mml:math></inline-formula> 1 <inline-formula><mml:math id="M19" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula> cvl <inline-formula><mml:math id="M20" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula> cvh, the snow fraction is a
function of the mean grid-box snow depth and the interception fraction is a
function of the mean interception reservoir water content. For the ERAI-LG
simulation, the high vegetation cover was set to zero (cvh <inline-formula><mml:math id="M21" display="inline"><mml:mo>=</mml:mo></mml:math></inline-formula> 0), the low
vegetation cover to one (cvl <inline-formula><mml:math id="M22" display="inline"><mml:mo>=</mml:mo></mml:math></inline-formula> 1) and the low vegetation type to
grassland. In this idealized simulation the entire globe was covered in grassland
so that only the low vegetation and exposed snow (when snow is present)
tiles were active. The main goal of this simulation is to evaluate the role
of land cover when comparing point observations with gridded reanalysis and
to evaluate pathways to improve reanalyses in representing albedo
processes.</p>
</sec>
</sec>
<sec id="Ch1.S2.SS2">
  <title>Observational in situ data</title>
      <p id="d1e395">To evaluate reanalysis performance, we used newly assembled in situ radiation
observations from Russian meteorological stations. This data set includes
4 h solar radiation and radiation balance data from the World
Meteorological Organization (WMO) World Radiation Network of the World
Radiation Data Center (WRDC) at the Voeikov Main Geophysical Observatory,
Saint Petersburg, Russia. The original WRDC data contains time series from
65 locations. We selected 47 stations for this study because they overlap
with daily snow depth and 2 m temperature observations (see Supplement Table S1).
Of these 47 stations three were attributed by ERAI-L to ocean grid points
and we decided to remove the three coastal stations from the initial
data set, so that the final data set consists of 44 stations. Temperature and
snow depth observations were taken from the All-Russian Research Institute
of Hydrometeorological Information World Data Centre (RIHMI-WDC), Obninsk,
Russia. A detailed description of this data set is provided by Bulygina et
al. (2010). This data set includes snow depth as well as snow cover fraction
around meteorological stations. Snow cover information in this data set is
not stored in percentages but rather on a scale of integers from 0 to 10
(for example, 50 % is assigned a value of 5 but so is 53 %). This makes
these data hardly applicable for precise SAF calculations. Snow depth
information is measured in centimetres with a precision of 1 cm. This
might lead to an underestimation of snow depth in the case of shallow snow
(between 0 and 1 cm). All variables (temperature, snow depth and snow cover,
surface LW radiation budget and surface SW radiation, the sum of the surface
short-wave and long-wave radiation budgets) were represented as daily time
series for the period 2000–2013, which is the time period available for the
radiation observations by the Voeikov Main Geophysical Observatory.</p>
      <p id="d1e398">Figure 1 shows the location of the stations together with the climatological
2000–2013 MAMJ snow depth as computed by ERAI-L. The distribution of
stations is quite heterogeneous, with very few stations located in eastern
Siberia and in the Far East. Moreover, some stations have prolonged periods
of missing values; six stations have more than 50 % missing values in the
daily time series for MAMJ. For monthly<?pagebreak page1890?> means, the total number of missing
values generally decreases from 2000 to 2013 (see Supplement Fig. S1).
However, data for the year 2009 are missing at 44 out of 47 stations during
the MAM period and at three stations in June. Nevertheless, spatial and temporal
coverage of this data set is exceptional for the analysis of albedo in this
region. It is also important to note that neither snow nor radiation from
these stations were assimilated in the reanalysis data sets and therefore
our intercomparisons are completely independent.</p>

      <?xmltex \floatpos{t}?><fig id="Ch1.F1"><caption><p id="d1e403">Station location and snow depth (cm) for the 2000–2013 MAMJ average
taken from ERAI-L. Red-coloured stations are excluded by the land–sea mask of
ERAI-L.</p></caption>
          <?xmltex \igopts{width=236.157874pt}?><graphic xlink:href="https://tc.copernicus.org/articles/12/1887/2018/tc-12-1887-2018-f01.pdf"/>

        </fig>

</sec>
</sec>
<sec id="Ch1.S3">
  <title>Methods</title>
      <p id="d1e419">To evaluate the climatic variables needed for the SAF computation, we first
compared daily values of snow depth, albedo and 2 m temperature from the
meteorological stations with those from the reanalyses. To co-locate
observations with reanalyses, we extracted the information of the grid cell
from the reanalysis in which the station is located. In the case of ERA-Interim/Land,
the horizontal resolution is <inline-formula><mml:math id="M23" display="inline"><mml:mrow><mml:msup><mml:mn mathvariant="normal">0.75</mml:mn><mml:mo>∘</mml:mo></mml:msup><mml:mo>×</mml:mo><mml:msup><mml:mn mathvariant="normal">0.75</mml:mn><mml:mo>∘</mml:mo></mml:msup></mml:mrow></mml:math></inline-formula> degrees,
whereas MERRA-2 has a horizontal resolution of <inline-formula><mml:math id="M24" display="inline"><mml:mrow><mml:msup><mml:mn mathvariant="normal">0.5</mml:mn><mml:mo>∘</mml:mo></mml:msup><mml:mo>×</mml:mo><mml:msup><mml:mn mathvariant="normal">0.625</mml:mn><mml:mo>∘</mml:mo></mml:msup></mml:mrow></mml:math></inline-formula> degrees. That said, the extracted values of the grid cell
are expected to show fewer variability and lower peak values, since they are
integrated over a larger spatial domain, which dampens extreme values. We
then derived long-term differences, performed a correlation analysis and
compared the variability among the data sets for the MAMJ period.</p>
      <p id="d1e458">Since the SAF signals for the seasonal cycle and long-term climate
change are highly correlated (Hall and Qu, 2006), we focus here on the
evaluation of the seasonal cycle. Snow cover is converted from snow depth
following a logarithmic equation according to which 2.5 cm of snow depth was
defined as equivalent to 100 % snow cover (Fletcher et al., 2015). We split
SAF into a snow cover component (SNC) and a temperature/metamorphosis
component (TEM). SNC relates to the decrease in albedo linked to the
earlier melting of snow. TEM concerns the reduction of snow albedo due to
enhanced metamorphism and larger grain sizes at warmer temperatures. In this
study we focus on these two components of the feedback process rather than
the general classic term for net SAF (<inline-formula><mml:math id="M25" display="inline"><mml:mrow><mml:mi mathvariant="normal">Δ</mml:mi><mml:mi mathvariant="italic">α</mml:mi><mml:mo>/</mml:mo><mml:mi mathvariant="normal">Δ</mml:mi><mml:mi>T</mml:mi></mml:mrow></mml:math></inline-formula>),
since our goal is to evaluate differences in the more
intricate terms of SAF. In the following, we assume that SAF <inline-formula><mml:math id="M26" display="inline"><mml:mo>=</mml:mo></mml:math></inline-formula> SNC <inline-formula><mml:math id="M27" display="inline"><mml:mo>+</mml:mo></mml:math></inline-formula> TEM,
which was shown to be true in nearly all cases for the NH (Fletcher et al.,
2012, 2015). Therefore, we compute the two terms as
          <disp-formula id="Ch1.E1" content-type="numbered"><mml:math id="M28" display="block"><mml:mrow><mml:mi mathvariant="normal">SNC</mml:mi><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mo>=</mml:mo><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mo>(</mml:mo><mml:mover accent="true"><mml:mrow><mml:msub><mml:mi mathvariant="italic">α</mml:mi><mml:mi mathvariant="normal">snow</mml:mi></mml:msub></mml:mrow><mml:mo mathvariant="normal">‾</mml:mo></mml:mover><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mo>-</mml:mo><mml:mspace width="0.125em" linebreak="nobreak"/><mml:msub><mml:mi mathvariant="italic">α</mml:mi><mml:mi mathvariant="normal">land</mml:mi></mml:msub><mml:mo>)</mml:mo><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mi mathvariant="normal">Δ</mml:mi><mml:msub><mml:mi>S</mml:mi><mml:mi mathvariant="normal">c</mml:mi></mml:msub><mml:mo>/</mml:mo><mml:mi mathvariant="normal">Δ</mml:mi><mml:msub><mml:mi>T</mml:mi><mml:mrow><mml:mn mathvariant="normal">2</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mi mathvariant="normal">m</mml:mi></mml:mrow></mml:msub></mml:mrow></mml:math></disp-formula>
        and
          <disp-formula id="Ch1.E2" content-type="numbered"><mml:math id="M29" display="block"><mml:mrow><mml:mi mathvariant="normal">TEM</mml:mi><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mo>=</mml:mo><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mover accent="true"><mml:mrow><mml:msub><mml:mi>S</mml:mi><mml:mi mathvariant="normal">c</mml:mi></mml:msub></mml:mrow><mml:mo mathvariant="normal">‾</mml:mo></mml:mover><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mi mathvariant="normal">Δ</mml:mi><mml:msub><mml:mi mathvariant="italic">α</mml:mi><mml:mi mathvariant="normal">snow</mml:mi></mml:msub><mml:mo>/</mml:mo><mml:mi mathvariant="normal">Δ</mml:mi><mml:msub><mml:mi>T</mml:mi><mml:mrow><mml:mn mathvariant="normal">2</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mi mathvariant="normal">m</mml:mi></mml:mrow></mml:msub><mml:mo>,</mml:mo></mml:mrow></mml:math></disp-formula>
        where <inline-formula><mml:math id="M30" display="inline"><mml:mrow><mml:msub><mml:mi mathvariant="italic">α</mml:mi><mml:mi mathvariant="normal">snow</mml:mi></mml:msub></mml:mrow></mml:math></inline-formula> is the snow-covered surface albedo,
<inline-formula><mml:math id="M31" display="inline"><mml:mrow><mml:msub><mml:mi mathvariant="italic">α</mml:mi><mml:mi mathvariant="normal">land</mml:mi></mml:msub></mml:mrow></mml:math></inline-formula> is the snow-free surface albedo,
<inline-formula><mml:math id="M32" display="inline"><mml:mrow><mml:msub><mml:mi>S</mml:mi><mml:mi mathvariant="normal">c</mml:mi></mml:msub></mml:mrow></mml:math></inline-formula> is the snow cover fraction and <inline-formula><mml:math id="M33" display="inline"><mml:mrow><mml:msub><mml:mi>T</mml:mi><mml:mrow><mml:mn mathvariant="normal">2</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mi mathvariant="normal">m</mml:mi></mml:mrow></mml:msub></mml:mrow></mml:math></inline-formula> is
the 2 m temperature. The first term of SNC <inline-formula><mml:math id="M34" display="inline"><mml:mrow><mml:mo>(</mml:mo><mml:mover accent="true"><mml:mrow><mml:msub><mml:mi mathvariant="italic">α</mml:mi><mml:mi mathvariant="normal">snow</mml:mi></mml:msub></mml:mrow><mml:mo mathvariant="normal">‾</mml:mo></mml:mover><mml:mo>-</mml:mo><mml:msub><mml:mi mathvariant="italic">α</mml:mi><mml:mi mathvariant="normal">land</mml:mi></mml:msub><mml:mo>)</mml:mo></mml:mrow></mml:math></inline-formula> is also known as an albedo contrast, whereas the
second term (<inline-formula><mml:math id="M35" display="inline"><mml:mrow><mml:mi mathvariant="normal">Δ</mml:mi><mml:msub><mml:mi>S</mml:mi><mml:mi mathvariant="normal">c</mml:mi></mml:msub><mml:mo>/</mml:mo><mml:mi mathvariant="normal">Δ</mml:mi><mml:msub><mml:mi>T</mml:mi><mml:mrow><mml:mn mathvariant="normal">2</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mi mathvariant="normal">m</mml:mi></mml:mrow></mml:msub></mml:mrow></mml:math></inline-formula>) will be
referred to as the snowmelt sensitivity. In Eqs. (1) and (2) deltas indicate
monthly changes and the overbars indicate means over the two adjacent
months. Note that <inline-formula><mml:math id="M36" display="inline"><mml:mrow><mml:mi mathvariant="normal">Δ</mml:mi><mml:msub><mml:mi>T</mml:mi><mml:mrow><mml:mn mathvariant="normal">2</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mi mathvariant="normal">m</mml:mi></mml:mrow></mml:msub></mml:mrow></mml:math></inline-formula> does not represent a
hemispheric mean but rather the difference at an individual location. It was
found that the contribution of SNC and TEM to the overall SAF is between 60
to 70 and 30 to 40 % for the NH (Fletcher et al., 2015).</p>
      <p id="d1e711">In our SAF assessment, we use 2 m temperature as a surrogate for near-surface air temperature, since the latter variable is not represented by
the stations. Using 2 m temperature introduces some uncertainty to the results
since atmospheric temperature advection can play a role in local temperature
evolution. However, by now multiple studies (Fletcher et al., 2015; Xiao et
al., 2017; Kevin et al., 2017) deal with 2 m temperature in their SAF
assessments, mainly due to the same comparability issues.</p>
      <p id="d1e714">Since daily data are available, we define <inline-formula><mml:math id="M37" display="inline"><mml:mrow><mml:msub><mml:mi mathvariant="italic">α</mml:mi><mml:mi mathvariant="normal">snow</mml:mi></mml:msub></mml:mrow></mml:math></inline-formula>
as the monthly mean over all daily estimates during the specific month when
<inline-formula><mml:math id="M38" display="inline"><mml:mrow><mml:msub><mml:mi>S</mml:mi><mml:mi mathvariant="normal">c</mml:mi></mml:msub><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mo>=</mml:mo><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mn mathvariant="normal">100</mml:mn></mml:mrow></mml:math></inline-formula> %. Moreover, we define
<inline-formula><mml:math id="M39" display="inline"><mml:mrow><mml:msub><mml:mi mathvariant="italic">α</mml:mi><mml:mi mathvariant="normal">land</mml:mi></mml:msub></mml:mrow></mml:math></inline-formula> as the mean over all daily estimates
during MAMJ (in some stations this might only occur in June) when
<inline-formula><mml:math id="M40" display="inline"><mml:mrow><mml:msub><mml:mi>S</mml:mi><mml:mi mathvariant="normal">c</mml:mi></mml:msub><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0</mml:mn></mml:mrow></mml:math></inline-formula> % . This allows for a less
artificial estimation of <inline-formula><mml:math id="M41" display="inline"><mml:mrow><mml:msub><mml:mi mathvariant="italic">α</mml:mi><mml:mi mathvariant="normal">land</mml:mi></mml:msub></mml:mrow></mml:math></inline-formula> than
is conventional using summer (e.g. August) albedo.</p>
</sec>
<sec id="Ch1.S4">
  <title>Results</title>
<sec id="Ch1.S4.SS1">
  <title>Daily data evaluation</title>
      <p id="d1e793">Since 2 m air temperature in reanalyses has been comprehensively evaluated
in previous studies (e.g. Schubert et al., 2014; Lindsay et al., 2014), we
only perform a general comparative assessment of the daily values of albedo
and snow depth in the SAF computations. That said, Lindsay et al. (2014)
found that 2 m temperatures show slight<?pagebreak page1891?> negative biases over Russia in
winter for both ERA-Interim and MERRA-1, whereas in summer ERA-Interim
basically shows no bias and MERRA1 shows slight positive biases. Improvements
in this regard from MERRA-1 to MERRA-2 are to be expected.</p>
      <p id="d1e796">Figure 2 shows an overall comparison between station data and reanalyses in
terms of correlations, differences and magnitude of variability quantified
by the standard deviation for the albedo and snow depths. On a day-to-day
basis MERRA-2 and ERAI-L are underestimating average albedo values compared
to observations by about 0.1 during MAMJ (Fig. 2a). In contrast,
ERAI-LG shows a much smaller average deviation from the station data with
differences close to zero. However, the overall range of the box plot for
ERAI-LG is similar to the other two reanalyses resulting in only slightly
fewer absolute deviations from the observations.</p>

      <?xmltex \floatpos{t}?><fig id="Ch1.F2"><caption><p id="d1e801">Box plot analysis for daily albedo <bold>(a, c, e)</bold> and snow depth
<bold>(b, d, f)</bold> estimates using data from 44 locations over 2000–2013
MAMJ period. <bold>(a, b)</bold> Difference between station and
reanalysis, <bold>(c, d)</bold> linear correlation between station and
reanalysis, <bold>(e, f)</bold> standard deviation. Triangle indicates the mean
value.</p></caption>
          <?xmltex \igopts{width=236.157874pt}?><graphic xlink:href="https://tc.copernicus.org/articles/12/1887/2018/tc-12-1887-2018-f02.pdf"/>

        </fig>

      <p id="d1e825">For snow depth (Fig. 2b), all three reanalysis data sets show an
overestimation of daily values for MAMJ. Interestingly, ERAI-LG shows the
largest deviations from observed values, although the grass better
represents the conditions at the observational sites. This can be caused by
biases in the observations due to surrounding higher vegetation creating a
snowfall shadow or negative instrumental biases (Rasmussen et al., 2012).
Moreover, positive biases in particular for precipitation can occur in
reanalysis products (Brun et al., 2013).</p>
      <p id="d1e829">The analysis of daily correlations (Fig. 2c and d) demonstrates that the
correlations for the albedo are generally weak among all three experiments,
whereas for some stations they can reach correlation coefficients higher
than 0.8. Surprisingly, the correlations between MERRA-2 and station data are
highest for albedo and lowest for snow depth. The observed
difference between MERRA-2 and the ECMWF experiments regarding the
correlation for albedo can likely be explained by the introduction of
aerosols (and their respective deposition) in MERRA-2 (see the Supplement for
a initial investigation). For snow depth, the correlation values are
dominated by snowfall and melting events. Also, in this case, the grass-only
experiment shows no improved performance compared to the classic ERAI
set-up.</p>
      <p id="d1e832">All reanalyses severely underestimate the day-to-day variability of the
albedo (Fig. 2e and f). MERRA-2 and ERAI-L show similar means but reach
the overall station level only in specific grid cells. A clear improvement
is observed in ERAI-LG, which shows the smallest deviation from station
estimates. Nevertheless, all modern reanalyses fail to adequately reproduce
daily variability in the observed albedo. In contrast, for snow depth
the agreement is very good. The mean values of all four products are around 8 to 10 cm, with the grass-only experiment being the closest to
the average station variability.</p>
      <p id="d1e835">In summary, the box plot analysis (Fig. 2) reveals that there is a general
improvement in the agreement between the stations and ERAI-L if vegetation is set to
grass only. However, none of the reanalysis products can accurately
reproduce day-to-day albedo variability. This is likely explained by the
comparison of grid versus point observations, where small-scale variations
are averaged out.</p>
</sec>
<sec id="Ch1.S4.SS2">
  <title>Analysis of feedback components</title>
      <p id="d1e844">To assess regional patterns of key SAF components, we show their spatial
distribution over Russia as revealed by the observations in Fig. 3 (see
Supplement Figs. 2–4 for the respective distribution from the reanalyses
data).</p>

      <?xmltex \floatpos{t}?><fig id="Ch1.F3"><caption><p id="d1e849">Mean SAF components in station data for 2000–2013 MAMJ:
<bold>(a)</bold> SNC, <bold>(b)</bold> TEM, <bold>(c)</bold> snowmelt sensitivity,
<bold>(d)</bold> mean albedo contrast, <bold>(e)</bold> mean albedo,
<bold>(f)</bold> snow depth.</p></caption>
          <?xmltex \igopts{width=236.157874pt}?><graphic xlink:href="https://tc.copernicus.org/articles/12/1887/2018/tc-12-1887-2018-f03.pdf"/>

        </fig>

      <p id="d1e877">Strong SNC (Fig. 3a) responses in the station data are observed in
southern European Russia and Western Siberia as well as over the Far East.
The weaker responses are observed in south-eastern Siberia. TEM (Fig. 3b)
follows a similar distribution but is more homogeneously distributed
with most negative values in central Siberia and towards the Arctic
coastline. Snowmelt sensitivity (Fig. 3c) is strongest in the
midlatitudinal and subpolar regions north of 50<inline-formula><mml:math id="M42" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> N, such as
Finland to the south-east, west and north of Lake Baikal and along the
Pacific coast. Here the temperatures react most strongly to seasonal snowmelt. While there is a broad agreement between the stations and ERAI-LG in
this region, stations show a somewhat stronger snowmelt sensitivity (not
shown). Snowmelt sensitivity is a key factor for the SNC calculations and
thus shapes the spatial variability of SNC.</p>
      <?pagebreak page1892?><p id="d1e889">The other key factor in the SNC calculations is the contrast in albedo
between snow-covered and snow-free periods (Fig. 3d). The observed albedo
contrast is characterized by a relatively homogeneous pattern with somewhat
smaller values in the southern regions, especially over southern
Siberia, east of Lake Baikal. In general, a north–south gradient is
visible, with similar patterns to the SNC. Mean albedo for spring (Fig. 3e)
shows that the highest values are found closer to the Arctic coastline, in
central Siberia and towards the western border. Lower mean albedo values are
mostly located east of Lake Baikal. This distribution is in general
agreement with the reanalyses data sets, especially for the lower values in
the south-east.</p>
      <p id="d1e893">Finally, since TEM closely follows the general MAMJ snow distribution, we
show average snow depth in Fig. 3f. A clear north–south gradient is
visible with hotspots at the Pacific coast and towards the Barents and Kara seas.
Moreover, snow depths from stations closely follow the ERA-L snow depth
distribution shown in Fig. 1.</p>
      <p id="d1e896">To analyse the differences between the data sets and to put the station data
in context, Fig. 4a shows the response for SAF computed for the entire
period 2000–2013 and all 44 locations. Stations show much stronger SAF
(<inline-formula><mml:math id="M43" display="inline"><mml:mo lspace="0mm">-</mml:mo></mml:math></inline-formula>2.5 % K<inline-formula><mml:math id="M44" display="inline"><mml:mrow><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">1</mml:mn></mml:mrow></mml:msup><mml:mo>)</mml:mo></mml:mrow></mml:math></inline-formula> compared to MERRA (<inline-formula><mml:math id="M45" display="inline"><mml:mo lspace="0mm">-</mml:mo></mml:math></inline-formula>1.6 % K<inline-formula><mml:math id="M46" display="inline"><mml:mrow><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">1</mml:mn></mml:mrow></mml:msup><mml:mo>)</mml:mo></mml:mrow></mml:math></inline-formula> and ERAI-L
(<inline-formula><mml:math id="M47" display="inline"><mml:mo lspace="0mm">-</mml:mo></mml:math></inline-formula>1.8 % K<inline-formula><mml:math id="M48" display="inline"><mml:mrow><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">1</mml:mn></mml:mrow></mml:msup><mml:mo>)</mml:mo></mml:mrow></mml:math></inline-formula>. At the same time ERAI-LG shows the SAF estimate close to
that derived from the station data (<inline-formula><mml:math id="M49" display="inline"><mml:mo lspace="0mm">-</mml:mo></mml:math></inline-formula>2.8 % K<inline-formula><mml:math id="M50" display="inline"><mml:mrow><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">1</mml:mn></mml:mrow></mml:msup><mml:mo>)</mml:mo></mml:mrow></mml:math></inline-formula>. Thus, changing the
vegetation to short grass adds an additional 1 % albedo decrease per
degree of warming to the feedback process. Further analysis of the two
components of SAF (SNC and TEM, Fig. 4b and c) shows that ERAI-LG
successfully reproduces the SNC signal derived from the station data (<inline-formula><mml:math id="M51" display="inline"><mml:mo lspace="0mm">-</mml:mo></mml:math></inline-formula>1.6 % K<inline-formula><mml:math id="M52" 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>
mean for stations and <inline-formula><mml:math id="M53" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>1.7 % K<inline-formula><mml:math id="M54" 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> mean for ERAI-LG), whereas
the other two reanalyses show much weaker SNC values. The lowest value of
<inline-formula><mml:math id="M55" display="inline"><mml:mo>-</mml:mo></mml:math></inline-formula>0.56 % K<inline-formula><mml:math id="M56" 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> was obtained from the MERRA-2 data. In general, SNC
responses largely explain differences in SAF (Fig. 4a).</p>

      <?xmltex \floatpos{t}?><fig id="Ch1.F4"><caption><p id="d1e1048">Box plot analysis for MAMJ 2000–2013: <bold>(a)</bold> SNC<inline-formula><mml:math id="M57" display="inline"><mml:mo>+</mml:mo></mml:math></inline-formula>TEM,
<bold>(b)</bold> SNC, <bold>(c)</bold> TEM, <bold>(d)</bold> snowmelt sensitivity,
<bold>(e)</bold> albedo contrast and <bold>(f)</bold> snow albedo. Triangle indicates
the mean value.</p></caption>
          <?xmltex \igopts{width=236.157874pt}?><graphic xlink:href="https://tc.copernicus.org/articles/12/1887/2018/tc-12-1887-2018-f04.pdf"/>

        </fig>

      <p id="d1e1083">For TEM values (Fig. 4c), all three reanalyses are in a good agreement
with the observations, with MERRA-2 showing the best agreement. Changing the
vegetation to grass in ERA-Interim results in a TEM component, which is
0.4–0.5 % K<inline-formula><mml:math id="M58" 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> stronger compared to the standard version of
ERA-Interim. Given that TEM represents the response to snow metamorphosis,
good performance of MERRA-2 is in agreement with findings implied by Fig. 2.
However, it is worth noting that for the station network as well as for
the ECMWF experiments, locations with positive TEM are calculated. This is
due to snow albedo changes being positive in some instances (Fig. 4c).</p>

      <?xmltex \floatpos{t}?><fig id="Ch1.F5"><caption><p id="d1e1100">Box plot analysis for MAMJ 2000–2013: <bold>(a)</bold> snow free albedo,
<bold>(b)</bold> snow cover fraction, where the light grey box plot is the
originally observed snow cover from stations, <bold>(c)</bold> snow depth,
<bold>(d)</bold> 2 m temperature, <bold>(e)</bold> mean albedo and <bold>(f)</bold> snow
albedo change within the season. Triangle indicates the mean value.</p></caption>
          <?xmltex \igopts{width=236.157874pt}?><graphic xlink:href="https://tc.copernicus.org/articles/12/1887/2018/tc-12-1887-2018-f05.pdf"/>

        </fig>

      <p id="d1e1129">To further investigate the nature of the SNC and TEM responses in Fig. 4d we
show the results for snowmelt sensitivity, which is one of the two key
components in the SNC response (Eq. 1). This component is barely influenced
by the underlying vegetation. All three reanalysis data sets agree very well
with the station network, with ERAI-LG<?pagebreak page1893?> showing the closest agreement for both
mean and median values. This indicates an accurate representation of this
relationship in both NASA and ECMWF land-surface modules.</p>
      <p id="d1e1132">Figure 4d implies that the changes in the SNC should stem from the albedo
contrast, the second key component expressed as the average difference
between albedo values for a complete snow cover and snow-free conditions
(Fig. 4e). Indeed, MERRA-2 shows the lowest albedo contrast among all
data sets, resulting in very low SNC values. The albedo contrast in ERAI-L is
higher than MERRA-2 but is on average still lower than the
observations, which show average values around 0.35. ERAI-LG shows the
strongest albedo contrast, which is twice as large as the
experiment with classic vegetation cover. These striking differences among
the data sets mainly drive the SNC results.</p>
      <p id="d1e1135">Snow albedo is well captured by the grass-only experiment, showing the same
average value, around 0.6, as determined from the observations (Fig. 4f).
The standard vegetation schemes used in MERRA-2 and ERAI-L reduce the snow
albedo in the analysed grid cells to 0.33 and 0.37. The differences in snow
albedo between the products is the main driver for the differences in the
albedo contrast, since the snow-free albedo values are remarkably similar for
all reanalysis products (Fig. 5a). Nevertheless, they strongly deviate
from the snow-free albedo determined from the observations, which is roughly
twice as large as the reanalyses, with a mean value of about 0.21,
and is very close to albedo values for grass (see e.g. Betts and Ball,
1997; Wei et al., 2001).</p>
      <p id="d1e1138">To explore the impact of different factors on the TEM estimates, in
Fig. 5 we show mean values of temperature, snow cover and albedo, as well as the
average change in snow albedo during spring. Also, to underline the crucial
role of in situ snow depth information, mean snow depth is shown. Mean
station snow depth lies within the range of reanalyses values, with higher
values reported by ERAI-LG. Moreover, stations have the lowest snow cover
among all data sets (Fig. 5b and c). This difference is likely due to the
conversion of snow depth to snow cover as well as from the precision (in
centimetres) of the Russian snow depth measurement. The precision of snow depth
diagnosed by reanalysis is much finer and the logarithmic conversion here
can be performed more accurately. As a result, TEM values diagnosed by
stations are probably too low. If we consider instead in situ snow cover
information from stations, the average snow cover is quite similar to the
reanalyses (ca. 55 %), and the average TEM value strengthens. However,
replacing converted snow cover with observed snow cover in Eq. (2) is a
questionable procedure, as the remaining terms were computed using snow
depth conversion. Thus, for consistency we show lower values of TEM in
Fig. 4.</p>
      <p id="d1e1141">Temperature is well represented by all data sets with MERRA-2 being about 1 K
colder than at the stations, which is quite notable for such a robust
variable. However, absolute values of temperature do not have a strong
impact on the computation of TEM, since monthly changes in
temperature affect both TEM and SNC computations. For the ERAI-LG albedo
contrast, the effects of the underestimated snow-free albedo and
overestimated snow albedo cancel each other out. Finally, the snow albedo
change during spring (Fig. 5f) is very similar in station data and in
MERRA-2 (<inline-formula><mml:math id="M59" display="inline"><mml:mo lspace="0mm">-</mml:mo></mml:math></inline-formula>0.09 average in both data sets), which points towards an adequate
representation of snow metamorphosis and aerosol deposition in MERRA-2. The
ERAI-LG experiment shows a stronger change in snow albedo during spring than
the standard version. ERAI-L potentially keeps the temperature and therefore
snow metamorphosis more constant throughout spring due to a more stable
local temperature climate induced by the vegetation. Note also that some
stations show an increase in snow albedo during spring. This can be caused
by fresh snow accumulation in late spring in some locations.</p>
      <p id="d1e1151">Figure 6 shows time series (2000–2013) for the mean values for SAF-related
variables. Time series for SNC (Fig. 6a) and TEM (Fig. 6b) show that
interannual variations of up to 0.5 % K<inline-formula><mml:math id="M60" 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> are possible for both
stations and reanalyses. Moreover, for both SNC and TEM, ERAI-LG seems to
successfully reproduce the overall baseline and the magnitude of variability.</p>

      <?xmltex \floatpos{t}?><fig id="Ch1.F6"><caption><p id="d1e1169">Yearly time series of selected MAMJ SAF components averaged over all
44 locations: <bold>(a)</bold> SNC, <bold>(b)</bold> TEM, <bold>(c)</bold> snowmelt
sensitivity, <bold>(d)</bold> snow albedo, <bold>(e)</bold> snow albedo change within
the season, <bold>(f)</bold> snow depth.</p></caption>
          <?xmltex \igopts{width=236.157874pt}?><graphic xlink:href="https://tc.copernicus.org/articles/12/1887/2018/tc-12-1887-2018-f06.pdf"/>

        </fig>

      <p id="d1e1197">For snowmelt sensitivity (Fig. 6c) the agreement among the data sets is
very good when it comes to magnitude and interannual variability, with
MERRA-2 showing an amplified interannual variability (up to 1.5 % K<inline-formula><mml:math id="M61" display="inline"><mml:mrow><mml:msup><mml:mi/><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">1</mml:mn></mml:mrow></mml:msup><mml:mo>)</mml:mo></mml:mrow></mml:math></inline-formula>,
which is<?pagebreak page1894?> beyond the magnitudes observed at the stations. As already
noted above, snowmelt sensitivity seems to be a rather well-reproduced
process in modern reanalyses. Since snow-free albedo is quite constant over
time in the reanalyses, the albedo contrast is dominated by the snow albedo
(Fig. 6d). ERAI-LG and the station network agree very well on the
magnitude of snow albedo, whereas ERAI-L and MERRA-2 fail to reproduce such
high values. Magnitudes of interannual variability can reach up to <inline-formula><mml:math id="M62" display="inline"><mml:mrow><mml:mo>±</mml:mo><mml:mn mathvariant="normal">0.05</mml:mn></mml:mrow></mml:math></inline-formula> in stations, with slightly weaker responses in reanalyses. The correlation
between stations and reanalyses is rather low: only individual years are
captured correctly by ERAI-LG (see Supplement for correlation values).</p>
      <p id="d1e1225">Snow albedo change within spring (Fig. 6e) is well captured by MERRA-2 and
ERAI-LG. Furthermore, ERAI-LG captures the interannual variability well for
this metric. Specifically, variability during 2001–2004 and 2005–2008
periods is quite well represented. In contrast, ERAI-L seems to lack
consistency with observations. Finally, as mentioned in Sect. 4.1,
snow depth variability (Fig. 6f) is very well captured by all
reanalyses. Again, ERAI-LG overestimates snow depth by up to 5 cm, with the
other two reanalyses being on average 1–2 cm above the station values.</p>
      <p id="d1e1228">To further demonstrate the effect of the vegetation changes on the
ERA-Interim/Land reanalysis, Fig. 7 shows anomalies between ERAI-L and
ERAI-LG. The structure follows Fig. 6, with SNC and TEM shown in Fig. 7a and b.
As is clearly visible, both variables are generally less negative in
ERAI-L, a fact already known from time series and box plot analysis. The
largest impact of the vegetation changes is found for northern Russia, the
Pacific coast and the western region between the Black Sea and the Caspian Sea.
Interestingly, but as expected, snowmelt sensitivity (Fig. 6c) is not the
key driver behind this distribution. Since snowmelt sensitivity is not
directly linked to vegetation changes, the anomaly distribution is very
heterogenous, with positive and negative anomalies over the whole domain. As
known from the time series plot, snow sensitivity in ERAI-LG is overall
slightly weaker than in ERAI-L, probably due to positive feedbacks such as
reduction of night-time cooling over higher vegetation types. The main driver
behind the distribution of SNC is the albedo contrast (Fig. 7d). The albedo
contrast is higher overall in ERAI-LG, especially along the borders of the
domain, which are already highlighted for SNC.</p>

      <?xmltex \floatpos{t}?><fig id="Ch1.F7"><caption><p id="d1e1233">Mean SAF components in anomalies of ERAI-L minus ERAI-LG for
2000–2013 MAMJ: <bold>(a)</bold> SNC, <bold>(b)</bold> TEM, <bold>(c)</bold> snowmelt
sensitivity, <bold>(d)</bold> mean albedo contrast, <bold>(e)</bold> mean albedo
and <bold>(f)</bold> snow depth.</p></caption>
          <?xmltex \igopts{width=236.157874pt}?><graphic xlink:href="https://tc.copernicus.org/articles/12/1887/2018/tc-12-1887-2018-f07.pdf"/>

        </fig>

<?xmltex \hack{\newpage}?>
</sec>
</sec>
<?pagebreak page1895?><sec id="Ch1.S5">
  <title>Discussion</title>
      <p id="d1e1271">We compared spring SAF and its components determined from in situ
measurements over Russia for the period 2000–2013 with data derived from
three modern reanalysis products restricted to the grid cells including the
observational sites. This was achieved by using a unique collection of
station measurements of radiation and snow characteristics, investigating observed SAF for
the first time over this broad spatial and temporal domain.
Besides ERAI-L we used a customized version of ERAI-L (ERAI-LG) in
which vegetation was set to grass in all concerned grid cells.</p>
      <p id="d1e1274">All three reanalysis data sets are completely independent from the analysed
station data. While a direct comparison of point measurements with grid cell
output always introduces uncertainties due to the spatial variability of the
surface, this is for now the only way to evaluate reanalyses data using
in situ observations. An alternative option would be satellite data, which
come with their own uncertainties (e.g. Romanov et al., 2002; Foster et al.,
2005; Wang et al., 2014).</p>
      <p id="d1e1277">Snow depth statistics derived from daily station data are reasonably well
reproduced in all three modern reanalyses, which is in agreement with
Wegmann et al. (2017), who investigated April snow depth in ERAI-L. While
snow depth differences between ERAI-L and ERAI-LG are small, ERAI-LG shows
slightly higher deviations from the station data than ERAI-L, which might be
caused by the higher vegetation in the station surroundings and by an
underestimation of snowfall due to instrumentation used at the Russian
station network (Rasmussen et al., 2012).</p>
      <p id="d1e1280">Day-to-day variability of albedo is notably higher in station data compared
to any reanalysis product. Besides spatial averaging over the reanalyses
grid cells, this is potentially caused by land-surface changes due to
weather (e.g. soil moisture change, aerosol deposition), which are not
represented in the reanalyses. However, ERAI-LG demonstrates increasing
albedo variability, nearly doubling the standard deviations diagnosed by
ERAI-L with the standard vegetation scheme.</p>
      <p id="d1e1284">The limitations of the station data imply some constraints for comparisons
with reanalysed data. As near-surface temperature is unavailable in station
data, we used for both stations and reanalyses 2 m air temperature, which
reduces the strength of the SAF feedback. Moreover, using local 2 m air
temperature first and then averaging over our domain later leads to lower SAF
values than if we would have used NH-averaged 2 m air temperature. Since
albedo changes at our stations are much more dramatic (due to the WMO
conditions) than in model or satellite grid cells, using geographically
smoothed temperature data would eventually lead to a much stronger
impact of albedo changes on temperature changes. Thus, our results are not
to be seen as a Northern Hemisphere impact analysis but rather as a
contribution to reanalysis improvement and the investigation of SAF
evolution.</p>
      <p id="d1e1287">Secondly, snow cover is underestimated in station data due to the
measurement precision of 1 cm, which reduces the strength of the TEM
component. The snow albedo and the snow-free albedo are substantially higher
in station data than in the reanalyses with classic vegetation boundary
conditions (MERRA-2 and ERAI-L). Compared to other observation-based studies,
spring snow albedo and grass albedo derived from our station network are
quite realistic (Roesch et al., 2009; Stroeve et al., 2006). Thus, the
difference revealed by reanalyses is likely due to averaging over grid
cells.</p>
      <p id="d1e1290">Results from ERAI-LG clearly demonstrate that SAF and its components are
very close to those in the station data. The largest improvement was found
for the albedo contrast and for snow albedo, which both are more realistic in
ERAI-LG. At the same time snow-free albedo in all three reanalyses
(including ERAI-LG) was found to be lower than in the station data, because
snow-free albedo in all reanalysis data sets is prescribed as a monthly
climatology from MODIS data. As MODIS mostly registers albedo from taiga and
tundra vegetation, a stark difference to the grass albedo from the stations
occurs.</p>
      <p id="d1e1293">MERRA-2 shows the lowest SAF values resulting from a very low albedo
contrast, which is probably a consequence of the vegetation scheme in the
MERRA-2 land module. In contrast, MERRA-2 represents TEM reasonably well,
most likely due to the accurate representation of the intra-seasonal snow
albedo changes. Thus, relative snowpack changes appear to be well
represented in MERRA-2, probably also due to a more accurate representation
of aerosols.</p>
      <p id="d1e1296">In general, we found higher SAF values in ERAI-L than in the recent
CMIP3/CMIP5 analyses of NH SAF by Fletcher et al. (2015). This disagreement
results from a variety of factors. First, our domain is limited to Russia
only, thus excluding considerable parts of Eurasia as well as North America.
In this respect our domain is set within a high SAF region, which may
explain the higher SAF values compared to the NH average by Fletcher et al. (2015).
In contrast, MERRA-2 shows good agreements with the NH CMIP4/5
SAF results but mostly because the albedo contrast is very low.
Furthermore, as we pointed out above, in situ observations used here tend to
slightly overestimate SAF, mainly due to higher snow albedo values. This is
because in situ snow albedo is typically measured by a sensor installed over
a vegetation-free snow pack. The vegetation scheme used in reanalyses gives
lower snow albedo values, implying realistic vegetation cover such as taiga
or tundra. However, our MERRA-2 results agree fairly well with the findings
of Fletcher et al. (2015). Moreover, mean values of the albedo-independent
variable snowmelt sensitivity are very close to the “observational” snowmelt sensitivity computed by Fletcher et al. (2015).</p>
      <?pagebreak page1896?><p id="d1e1299">We also found agreements with Fletcher et al. (2015) in the representation
of the spatial pattern of the SAF components. Fletcher et al. (2015) as well
as Fernandes et al. (2009) have shown maxima in SAF over northern Canada,
northern Siberia and south-western Eurasia. The relation of <inline-formula><mml:math id="M63" display="inline"><mml:mrow><mml:mn mathvariant="normal">60</mml:mn><mml:mo>:</mml:mo><mml:mn mathvariant="normal">40</mml:mn></mml:mrow></mml:math></inline-formula> between SNC
and TEM, which is found in modelled, satellite and reanalysis data, was
replicated by our station network. We found similar spatial patterns for SAF
and its components in both stations and gridded data specifically for
southern Russia, while the pattern of station responses is less homogenous
than the gridded data. Also consistent with Fletcher et al. (2015),
we found higher snowmelt sensitivity north of 50<inline-formula><mml:math id="M64" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> N. Finally, the
albedo contrast distribution, which closely follows the snow albedo pattern,
is in very good agreement with the gridded analysis of snow albedo by
Fletcher et al. (2015).</p>
</sec>
<sec id="Ch1.S6" sec-type="conclusions">
  <title>Conclusions</title>
      <p id="d1e1329">Reanalyses including land-surface modules show a physically consistent
representation of SAF with realistic spatial patterns and area-averaged
sensitivity estimates. ERAI-LG shows a better performance in representing
station-based estimates considering the uncertainty associated with “point
to grid cell” comparisons. Accounting for aerosol-related processes would
likely improve this performance in future reanalysis releases. Thus, for the
analysis and validation of large-scale temporal and spatial averages of SAF,
modern reanalyses seem to be an appropriate tool.</p>
      <p id="d1e1332">However, to analyse processes on smaller scales and at high temporal
resolution, a healthy dense station network is required. The
idealized ERAI-LG simulation also highlights the caveats of comparing
in situ observations with gridded model data. In this study, we show these
discrepancies in terms of albedo and snow depth. Other variables, in
particular 2 m temperature, can be expected to have a similar signal arising
from the differences between the model's grid cell land cover and the actual
station conditions. Our findings show that the experimental approach in
ERAI-LG allows for enhanced use of in situ observations to diagnose the
SAF in non-forested areas.</p>
      <p id="d1e1335">Considering future studies, the extension to other regions and use of other
regional in situ data might give further insights into regional hotspots of
SAF. Cross-validation efforts employing model, reanalysis, satellite and
station data may help to generate blended products to investigate radiation
and albedo feedbacks in the changing Arctic, a region where SAF is
especially strong. Regional modelling, including a variety of multilayer
land-surface models over areas with a relatively dense observation network,
can provide a quantitative estimation of uncertainties among complex
variables such as snow depth, albedo or SAF.</p>
</sec>

      
      </body>
    <back><notes notes-type="dataavailability">

      <p id="d1e1342">ERA-Interim land data (Balsamo et al., 2015) can be
accessed at ECMWF <uri>http://apps.ecmwf.int/datasets/data/interim-land/</uri>
(last access: 1 June 2018). The ERA-Interim land experiment data (created by
Emanuel Dutra) can be made available upon request by email from the lead
author. MERRA2 data (Gelaro et al., 2017) were taken from the Earthdata
repository of NASA <uri>https://disc.gsfc.nasa.gov/daac-bin/FTPSubset2.pl</uri>
(last access: 1 June 2018). Solar radiation data are from the bulletins
“Solar Radiation and Radiation Balance Data. The World Network”
(1964–2015) of WMO World Radiation Data Center (WRDC) at Voeikov Main
Geophysical 55 Observatory, St Petersburg <uri>http://wrdc.mgo.rssi.ru</uri> (last
access: 1 June 2018). Snow depth and 2 m temperature in situ data (Bulygina
et al., 2010) are taken from the All-Russia Research Institute of
Hydrometeorological Information – World Data Centre (RIHMI-WDC)
<uri>http://meteo.ru/english/climate/cl_data.php</uri> (last access: 1 June
2018).</p>
  </notes><app-group>
        <supplementary-material position="anchor"><p id="d1e1357">The supplement related to this article is available online at: <inline-supplementary-material xlink:href="https://doi.org/10.5194/tc-12-1887-2018-supplement" xlink:title="pdf">https://doi.org/10.5194/tc-12-1887-2018-supplement</inline-supplementary-material>.</p></supplementary-material>
        </app-group><notes notes-type="competinginterests">

      <p id="d1e1366">The authors declare that they have no conflict of
interest.</p>
  </notes><ack><title>Acknowledgements</title><p id="d1e1372">This study was supported by the ARCTIC-ERA project funded by the Belmont
Forum Fund through the ANR. Olga Zolina also benefited from support by the
Russian Ministry of Education and Science (project no. 14.613.21.0083, ID
RFMEFI61317X0083). Emanuel Dutra was supported by the Portuguese Science
Foundation (FCT) under the project IF/00817/2015. IGE is part of Labex OSUG@2020
(ANR10 LABX56). We thank Morgan Gray for editorial support and Elena Makhotkina
from MGO for data access support. <?xmltex \hack{\newline}?><?xmltex \hack{\newline}?> Edited by: Michiel van den
Broeke<?xmltex \hack{\newline}?> Reviewed by: two anonymous referees</p></ack><ref-list>
    <title>References</title>

      <ref id="bib1.bib1"><label>1</label><mixed-citation>Balsamo, G., Albergel, C., Beljaars, A., Boussetta, S., Brun, E., Cloke, H.,
Dee, D., Dutra, E., Muñoz-Sabater, J., Pappenberger, F., de Rosnay, P.,
Stockdale, T., and Vitart, F.: ERA-Interim/Land: a global land surface
reanalysis data set, Hydrol. Earth Syst. Sci., 19, 389–407,
<ext-link xlink:href="https://doi.org/10.5194/hess-19-389-2015" ext-link-type="DOI">10.5194/hess-19-389-2015</ext-link>, 2015.</mixed-citation></ref>
      <ref id="bib1.bib2"><label>2</label><mixed-citation>
Betts, A. K. and Ball, J. H.: Albedo over the boreal forest, J. Geophys.
Res.-Atmos., 102, 28901–28909, 1997.</mixed-citation></ref>
      <ref id="bib1.bib3"><label>3</label><mixed-citation>
Bony, S., Colman, R., Kattsov, V. M., Allan, R. P., Bretherton, C. S.,
Dufresne, J., Hall, A., Hallegatte, S., Holland, M. M., Ingram, W., Randall,
D. A., Soden, B. J., Tselioudis, G., and Webb, M. J.: How Well Do We
Understand and Evaluate Climate Change Feedback Processes?, J. Climate, 19,
3445–3482, 2006.</mixed-citation></ref>
      <ref id="bib1.bib4"><label>4</label><mixed-citation>
Brun, E., Vionnet, V., Boone, A., Decharme, B., Peings, Y., Valette, R.,
Karbou, F., and Morin, S.: Simulation of Northern Eurasian Local Snow Depth,
Mass, and Density Using a Detailed Snowpack Model and Meteorological
Reanalyses, J. Hydrometeorol., 14, 203–219, 2013.</mixed-citation></ref>
      <?pagebreak page1897?><ref id="bib1.bib5"><label>5</label><mixed-citation>Brutel-Vuilmet, C., Ménégoz, M., and Krinner, G.: An analysis of
present and future seasonal Northern Hemisphere land snow cover simulated by
CMIP5 coupled climate models, The Cryosphere, 7, 67–80,
<ext-link xlink:href="https://doi.org/10.5194/tc-7-67-2013" ext-link-type="DOI">10.5194/tc-7-67-2013</ext-link>, 2013.</mixed-citation></ref>
      <ref id="bib1.bib6"><label>6</label><mixed-citation>
Budkyo, M. I.: The effect of solar radiation on the climate of the earth,
Tellus, 21, 611–619, 1967.</mixed-citation></ref>
      <ref id="bib1.bib7"><label>7</label><mixed-citation>Bulygina, O. N., Groisman, P. Y., Razuvaev, V. N., and Radionov, V. F.: Snow
cover basal ice layer changes over Northern Eurasia since 1966, Environ. Res.
Lett., 5, 015004, <ext-link xlink:href="https://doi.org/10.1088/1748-9326/5/1/015004" ext-link-type="DOI">10.1088/1748-9326/5/1/015004</ext-link>, 2010.</mixed-citation></ref>
      <ref id="bib1.bib8"><label>8</label><mixed-citation>Cess, R. O. and Potter, G. L.: Interpretation of snow-climate feedback as
produced by 17 general circulation models, Science, 253, 888–892,
<ext-link xlink:href="https://doi.org/10.1126/science.253.5022.888" ext-link-type="DOI">10.1126/science.253.5022.888</ext-link>, 1991.</mixed-citation></ref>
      <ref id="bib1.bib9"><label>9</label><mixed-citation>
Collins, M., Knutti, R., Arblaster, J., Dufresne, J.-L., Fichefet, T.,
Friedlingstein, P., Gao, X., Gutowski, W. J., Johns, T., Krinner, G.,
Shongwe, M., Tebaldi, C., Weaver, A. J., and Wehner, M.: Long-term Climate
Change: Projections, Commitments and Irreversibility, in: Climate Change
2013: The Physical Science Basis. Contribution of Working Group I to the
Fifth Assessment Report of the Intergovernmental Panel on Climate Change,
edited by: Stocker, T. F., Qin, D., Plattner, G.-K., Tignor, M., Allen, S.
K., Boschung, J., Nauels, A., Xia, Y., Bex, V., and Midgley, P. M., Cambridge
University Press, Cambridge, United Kingdom and New York, NY, USA, 2013.</mixed-citation></ref>
      <ref id="bib1.bib10"><label>10</label><mixed-citation>
Curry, J. A., Schramm, J. L., Rossow, W. B., and Randall, D.: Overview of
Arctic cloud and radiation characteristics, J. Climate, 9, 1731–1764, 1996</mixed-citation></ref>
      <ref id="bib1.bib11"><label>11</label><mixed-citation>
Dee, D. P., Uppala, S. M., Simmons, A. J., Berrisford, P., Poli, P.,
Kobayashi, S., Andrae, U., Balmaseda, M. A., Balsamo, G., Bauer, P.,
Bechtold, P., Beljaars, A. C. M., van de Berg, L., Bid- lot, J., Bormann, N.,
Delsol, C., Dragani, R., Fuentes, M., Geer, A. J., Haimberger, L., Healy, S.
B., Hersbach, H., Hólm, E. V., Isaksen, L., Kållberg, P., Köhler,
M., Matricardi, M., McNally, A. P., Monge-Sanz, B. M., Morcrette, J.-J.,
Park, B.-K., Peubey, C., de Rosnay, P., Tavolato, C., Thépaut, J.-N.,
&amp; Vitart, F.: The ERA–interim reanalysis: Configuration and performance
of the data assimilation system, Q. J. Roy. Meteor. Soc., 137, 553–597,
2011.</mixed-citation></ref>
      <ref id="bib1.bib12"><label>12</label><mixed-citation>Derksen, C. and Brown, R.: Spring snow cover extent reductions in the
2008–2012 period exceeding climate model projections, Geophys. Res. Lett.,
39, L19504, <ext-link xlink:href="https://doi.org/10.1029/2012GL053387" ext-link-type="DOI">10.1029/2012GL053387</ext-link>, 2012.</mixed-citation></ref>
      <ref id="bib1.bib13"><label>13</label><mixed-citation>
Dufour, A., Zolina, O., and Gulev, S. K.: Atmospheric moisture transport to
the Arctic: Assessment of reanalyses and analysis of transport components, J.
Climate, 29, 5061–5081, 2016.</mixed-citation></ref>
      <ref id="bib1.bib14"><label>14</label><mixed-citation>
Dutra, E., Balsamo, G., Viterbo, P., Miranda, P. M., Beljaars, A., Schär,
C., and Elder, K.: An improved snow scheme for the ECMWF land surface model:
description and offline validation, J. Hydrometeorol., 11, 899–916, 2010.</mixed-citation></ref>
      <ref id="bib1.bib15"><label>15</label><mixed-citation>Fernandes, R., Zhao, H., Wang, X., Key, J., Qu, X., and Hall, A.: Controls on
Northern Hemisphere snow albedo feedback quantified using satellite earth
observations, Geophys. Res. Lett., 36, L21702, <ext-link xlink:href="https://doi.org/10.1029/2009GL040057" ext-link-type="DOI">10.1029/2009GL040057</ext-link>,
2009.</mixed-citation></ref>
      <ref id="bib1.bib16"><label>16</label><mixed-citation>Flanner, M. G., Shell, K. M., Barlage, M., Perovich, D. K., and Tschudi, M.
A.: Radiative forcing and albedo feedback from the Northern Hemisphere
cryosphere between 1979 and 2008, Nat. Geosci., 4, 151,
<ext-link xlink:href="https://doi.org/10.1038/ngeo1062" ext-link-type="DOI">10.1038/ngeo1062</ext-link>, 2011.</mixed-citation></ref>
      <ref id="bib1.bib17"><label>17</label><mixed-citation>Fletcher, C. G., Zhao, H., Kushner, P. J., and Fernandes, R.: Using models
and satellite observations to evaluate the strength of snow albedo feedback,
J. Geophys. Res.-Atmos., 117, D11117, <ext-link xlink:href="https://doi.org/10.1029/2012JD017724" ext-link-type="DOI">10.1029/2012JD017724</ext-link>, 2012.</mixed-citation></ref>
      <ref id="bib1.bib18"><label>18</label><mixed-citation>
Fletcher, C. G., Thackeray, C. W., and Burgers, T. M.: Evaluating biases in
simulated snow albedo feedback in two generations of climate models, J.
Geophys. Res.-Atmos., 120, 12–26, 2015.</mixed-citation></ref>
      <ref id="bib1.bib19"><label>19</label><mixed-citation>
Foster, J. L., Sun, C., Walker, J. P., Kelly, R., Chang, A., Dong, J., and
Powell, H.: Quantifying the uncertainty in passive microwave snow water
equivalent observations, Remote Sens. Environ., 94, 187–203, 2005.</mixed-citation></ref>
      <ref id="bib1.bib20"><label>20</label><mixed-citation>
Gelaro, R., McCarty, W., Suárez, M. J., Todling, R., Molod, A., Takacs,
L., Randles, C. A., Darmenov, A., Bosilovich, M. G., Reichle, R., Wargan, K.,
Coy, L., Cullather, R., Draper, C., Akella, S., Buchard, V., Conaty, A., da
Silva, A. M., Gu, W., Kim, G., Koster, R., Lucchesi, R., Merkova, D.,
Nielsen, J. E., Partyka, G., Pawson, S., Putman, W., Rienecker, M., Schubert,
S. D., Sienkiewicz, M., and Zhao, B.: The Modern-Era Retrospective Analysis
for Research and Applications, Version 2 (MERRA-2), J. Climate, 30,
5419–5454, 2017.</mixed-citation></ref>
      <ref id="bib1.bib21"><label>21</label><mixed-citation>
Graversen, R. G. and Wang, M.: Polar amplification in a coupled climate model
with locked albedo, Clim. Dynam., 33, 629–643, 2009.</mixed-citation></ref>
      <ref id="bib1.bib22"><label>22</label><mixed-citation>
Graversen, R. G., Langen, P. L., and Mauritsen, T.: Polar amplification in
CCSM4: Contributions from the lapse rate and surface albedo feedbacks, J.
Climate, 27, 4433–4450, 2014.</mixed-citation></ref>
      <ref id="bib1.bib23"><label>23</label><mixed-citation>
Groisman, P. Y., Karl, T. R., Knight, R. W., and Stenchikov, G. L.: Changes
of snow cover, temperature, and radiative heat balance over the Northern
Hemisphere, J. Climate, 7, 1633–1656, 1994.</mixed-citation></ref>
      <ref id="bib1.bib24"><label>24</label><mixed-citation>
Hall, A.: The role of surface albedo feedback in climate, J. Climate, 17,
1550–1568, 2004.</mixed-citation></ref>
      <ref id="bib1.bib25"><label>25</label><mixed-citation>Hall, A. and Qu, X.: Using the current seasonal cycle to constrain snow
albedo feedback in future climate change, Geophys. Res. Lett., 33, L03502,
<ext-link xlink:href="https://doi.org/10.1029/2005GL025127" ext-link-type="DOI">10.1029/2005GL025127</ext-link>, 2006.</mixed-citation></ref>
      <ref id="bib1.bib26"><label>26</label><mixed-citation>
Hall, D. K., Riggs, G. A., Salomonson, V. V., DiGirolamo, N. E., and Bayr, K.
J.: MODIS snow-cover products, Remote Sens. Environ., 83, 181–194, 2002.</mixed-citation></ref>
      <ref id="bib1.bib27"><label>27</label><mixed-citation>Hall, A., Qu, X., and Neelin, J. D.: Improving predictions of summer climate
change in the United States, Geophys. Res. Lett., 35, L01702,
<ext-link xlink:href="https://doi.org/10.1029/2007GL032012" ext-link-type="DOI">10.1029/2007GL032012</ext-link>, 2008.</mixed-citation></ref>
      <ref id="bib1.bib28"><label>28</label><mixed-citation>
Kanamitsu, M., Ebisuzaki, W., Woollen, J., Yang, S. K., Hnilo, J. J.,
Fiorino, M., and Potter, G. L.: Ncep–doe amip-ii reanalysis (r-2), B.
Am. Meteorol. Soc., 83, 1631–1643, 2002.</mixed-citation></ref>
      <ref id="bib1.bib29"><label>29</label><mixed-citation>
Kevin, J. P. W., Kotlarski, S., Scherrer, S. C., and Schär, C.: The
Alpine snow-albedo feedback in regional climate models, Clim. Dynam., 48,
1109–1124, 2017.</mixed-citation></ref>
      <ref id="bib1.bib30"><label>30</label><mixed-citation>
Justice, C. O., Townshend, J. R. G., Holben, B. N., and Tucker, E. C.:
Analysis of the phenology of global vegetation using meteorological satellite
data, Int. J. Remote Sens., 6, 1271–1318, 1985.</mixed-citation></ref>
      <ref id="bib1.bib31"><label>31</label><mixed-citation>
Lian, M. S. and Cess, R. D.: Energy balance climate models: A
reappraisal of ice-albedo feedback, J. Atmos. Sci., 34,
1058–1062,  1977.</mixed-citation></ref>
      <ref id="bib1.bib32"><label>32</label><mixed-citation>
Lindsay, R., Wensnahan, M., Schweiger, A., and Zhang, J.: Evaluation of seven
different atmospheric reanalysis products in the Arctic, J. Climate, 27,
2588–2606, 2014.</mixed-citation></ref>
      <ref id="bib1.bib33"><label>33</label><mixed-citation>Pithan, F. and Mauritsen, T.: Arctic amplification dominated by temperature
feedbacks in contemporary climate models, Nat. Geosci., 7, 181–184,
<ext-link xlink:href="https://doi.org/10.1038/ngeo2071" ext-link-type="DOI">10.1038/ngeo2071</ext-link>, 2014.</mixed-citation></ref>
      <ref id="bib1.bib34"><label>34</label><mixed-citation>
Qu, X. and Hall, A.: What controls the strength of snow-albedo feedback?, J.
Climate, 20, 3971–3981, 2007.</mixed-citation></ref>
      <ref id="bib1.bib35"><label>35</label><mixed-citation>
Qu, X. and Hall, A.: On the persistent spread in snow-albedo feedback, Clim.
Dynam., 42, 69–81, 2014.</mixed-citation></ref>
      <?pagebreak page1898?><ref id="bib1.bib36"><label>36</label><mixed-citation>
Rasmussen, R., Baker, B., Kochendorfer, J., Meyers, T., Landolt, S., Fischer,
A. P., Black, J., Thériault, J. M., Kucera, P., Gochis, D., Smith, C.,
Nitu, R., Hall, M., Ikeda, K., and Gutmann, E.: How Well Are We Measuring
Snow: The NOAA/FAA/NCAR Winter Precipitation Test Bed, Bull. Amer. Meteor.
Soc., 93, 811–829, 2012.</mixed-citation></ref>
      <ref id="bib1.bib37"><label>37</label><mixed-citation>
Reichle, R. H., Draper, C. S., Liu, Q., Girotto, M., Mahanama, S. P., Koster,
R. D., and De Lannoy, G. J.: Assessment of MERRA-2 land surface hydrology
estimates, J. Climate, 30, 2937–2960, 2017.</mixed-citation></ref>
      <ref id="bib1.bib38"><label>38</label><mixed-citation>
Rienecker, M. M., Suarez, M. J., Gelaro, R., Todling, R., Bacmeister, J.,
Liu, E., Bosilovich, M. G., Schubert, S. D., Takacs, L., Kim, G., Bloom, S.,
Chen, J., Collins, D., Conaty, A., da Silva, A., Gu, W., Joiner, J., Koster,
R. D., Lucchesi, R., Molod, A., Owens, T., Pawson, S., Pegion, P., Redder, C.
R., Reichle, R., Robertson, F. R., Ruddick, A. G., Sienkiewicz, M., and
Woollen, J.: MERRA: NASA's Modern-Era Retrospective Analysis for Research and
Applications, J. Climate, 24, 3624–3648, 2011.</mixed-citation></ref>
      <ref id="bib1.bib39"><label>39</label><mixed-citation>
Robock, A.: Ice and snow feedbacks and the latitudinal and seasonal
distribution of climate sensitivity, J. Atmos. Sci., 40, 986–997, 1983.</mixed-citation></ref>
      <ref id="bib1.bib40"><label>40</label><mixed-citation>
Roesch, A., Gilgen, H., Wild, M., and Ohmura, A.: Assessment of GCM simulated
snow albedo using direct observations, Clim. Dynam., 15, 405–418, 1999.</mixed-citation></ref>
      <ref id="bib1.bib41"><label>41</label><mixed-citation>
Romanov, P., Gutman, G., and Csiszar, I.: Satellite-derived snow cover maps
for North America: accuracy assessment, Adv. Space Res., 30, 2455–2460,
2002.</mixed-citation></ref>
      <ref id="bib1.bib42"><label>42</label><mixed-citation>
Schiffer, R. A. and Rossow, W. B.: The International Satellite Cloud
Climatology Project(ISCCP) – The first project of the World Climate Research
Programme, American Meteorological Society, Bulletin, 64, 779–784, 1983.</mixed-citation></ref>
      <ref id="bib1.bib43"><label>43</label><mixed-citation>Schneider, S. H. and Dickinson, R. E.: Climate modeling, Rev. Geophys., 12,
447–493, 1974.
 </mixed-citation></ref><?xmltex \hack{\newpage}?>
      <ref id="bib1.bib44"><label>44</label><mixed-citation>
Schubert, S. D., Wang, H., Koster, R. D., Suarez, M. J., and Groisman, P. Y.:
Northern Eurasian heat waves and droughts, J. Climate, 27, 3169–3207, 2014.</mixed-citation></ref>
      <ref id="bib1.bib45"><label>45</label><mixed-citation>
Serreze, M. C. and Barry, R. G.: Processes and impacts of Arctic
amplification: A research synthesis, Global Planet. Change, 77, 85–96, 2011.</mixed-citation></ref>
      <ref id="bib1.bib46"><label>46</label><mixed-citation>
Stroeve, J. C., Box, J. E., and Haran, T.: Evaluation of the MODIS (MOD10A1)
daily snow albedo product over the Greenland ice sheet, Remote Sens.
Environ., 105, 155–171, 2006.</mixed-citation></ref>
      <ref id="bib1.bib47"><label>47</label><mixed-citation>
Thackeray, C. W. and Fletcher, C. G.: Snow albedo feedback: Current
knowledge, importance, outstanding issues and future directions, Prog. Phys.
Geog., 40, 392–408, 2016.</mixed-citation></ref>
      <ref id="bib1.bib48"><label>48</label><mixed-citation>
Wang, Z., Schaaf, C. B., Strahler, A. H., Chopping, M. J., Román, M. O.,
Shuai, Y., Woodcock, C. E., Hollinger, D. Y., and Fitzjarrald, D. R.:
Evaluation of MODIS albedo product (MCD43A) over grassland, agriculture and
forest surface types during dormant and snow-covered periods, Remote Sens.
Environ., 140, 60–77, 2014.</mixed-citation></ref>
      <ref id="bib1.bib49"><label>49</label><mixed-citation>
Wei, X., Hahmann, A. N., Dickinson, R. E., Yang, Z. L., Zeng, X., Schaudt, K.
J., Schaaf, C. B., and Strugnell, N.: Comparison of albedos computed by land
surface models and evaluation against remotely sensed data, J. Geophys.
Res.-Atmos., 106, 20687–20702, 2001.</mixed-citation></ref>
      <ref id="bib1.bib50"><label>50</label><mixed-citation>Wegmann, M., Orsolini, Y., Dutra, E., Bulygina, O., Sterin, A., and
Brönnimann, S.: Eurasian snow depth in long-term climate reanalyses, The
Cryosphere, 11, 923–935, <ext-link xlink:href="https://doi.org/10.5194/tc-11-923-2017" ext-link-type="DOI">10.5194/tc-11-923-2017</ext-link>, 2017.</mixed-citation></ref>
      <ref id="bib1.bib51"><label>51</label><mixed-citation>
Wexler, H.: Radiation balance of the Earth as a factor in climatic change,
in: Climatic Change, edited by: Shapley, H., Cambridge, Harvard University
Press, 73–105, 1953.</mixed-citation></ref>
      <ref id="bib1.bib52"><label>52</label><mixed-citation>Xiao, L., Che, T., Chen, L., Xie, H., and Dai, L.: Quantifying Snow Albedo
Radiative Forcing and Its Feedback during 2003–2016, Remote Sensing, 9, 883,
<ext-link xlink:href="https://doi.org/10.3390/rs9090883" ext-link-type="DOI">10.3390/rs9090883</ext-link>, 2017.</mixed-citation></ref>

  </ref-list></back>
    <!--<article-title-html>Spring snow albedo feedback over northern Eurasia:  Comparing in situ measurements with reanalysis products</article-title-html>
<abstract-html><p>This study uses daily observations and modern reanalyses in order to evaluate
reanalysis products over northern Eurasia regarding the spring snow albedo
feedback (SAF) during the period from 2000 to 2013. We used the
state-of-the-art reanalyses from ERA-Interim/Land and the Modern-Era
Retrospective Analysis for Research and Applications version 2 (MERRA-2) as
well as an experimental set-up of ERA-Interim/Land with prescribed short
grass as land cover to enhance the comparability with the station data while
underlining the caveats of comparing in situ observations with gridded data.
Snow depth statistics derived from daily station data are well reproduced in
all three reanalyses. However day-to-day albedo variability is notably higher
at the stations than for any reanalysis product. The ERA-Interim grass set-up
shows improved performance when representing albedo variability and generates
comparable estimates for the snow albedo in spring. We find that modern
reanalyses show a physically consistent representation of SAF, with realistic
spatial patterns and area-averaged sensitivity estimates. However,
station-based SAF values are significantly higher than in the reanalyses,
which is mostly driven by the stronger contrast between snow and snow-free
albedo. Switching to grass-only vegetation in ERA-Interim/Land increases the
SAF values up to the level of station-based estimates. We found no
significant trend in the examined 14-year time series of SAF, but interannual
changes of about 0.5&thinsp;%&thinsp;K<sup>−1</sup> in both station-based and reanalysis
estimates were derived. This interannual variability is primarily dominated
by the variability in the snowmelt sensitivity, which is correctly captured
in reanalysis products. Although modern reanalyses perform well for snow
variables, efforts should be made to improve the representation of dynamic
albedo changes.</p></abstract-html>
<ref-html id="bib1.bib1"><label>1</label><mixed-citation>
Balsamo, G., Albergel, C., Beljaars, A., Boussetta, S., Brun, E., Cloke, H.,
Dee, D., Dutra, E., Muñoz-Sabater, J., Pappenberger, F., de Rosnay, P.,
Stockdale, T., and Vitart, F.: ERA-Interim/Land: a global land surface
reanalysis data set, Hydrol. Earth Syst. Sci., 19, 389–407,
<a href="https://doi.org/10.5194/hess-19-389-2015" target="_blank">https://doi.org/10.5194/hess-19-389-2015</a>, 2015.
</mixed-citation></ref-html>
<ref-html id="bib1.bib2"><label>2</label><mixed-citation>
Betts, A. K. and Ball, J. H.: Albedo over the boreal forest, J. Geophys.
Res.-Atmos., 102, 28901–28909, 1997.
</mixed-citation></ref-html>
<ref-html id="bib1.bib3"><label>3</label><mixed-citation>
Bony, S., Colman, R., Kattsov, V. M., Allan, R. P., Bretherton, C. S.,
Dufresne, J., Hall, A., Hallegatte, S., Holland, M. M., Ingram, W., Randall,
D. A., Soden, B. J., Tselioudis, G., and Webb, M. J.: How Well Do We
Understand and Evaluate Climate Change Feedback Processes?, J. Climate, 19,
3445–3482, 2006.
</mixed-citation></ref-html>
<ref-html id="bib1.bib4"><label>4</label><mixed-citation>
Brun, E., Vionnet, V., Boone, A., Decharme, B., Peings, Y., Valette, R.,
Karbou, F., and Morin, S.: Simulation of Northern Eurasian Local Snow Depth,
Mass, and Density Using a Detailed Snowpack Model and Meteorological
Reanalyses, J. Hydrometeorol., 14, 203–219, 2013.
</mixed-citation></ref-html>
<ref-html id="bib1.bib5"><label>5</label><mixed-citation>
Brutel-Vuilmet, C., Ménégoz, M., and Krinner, G.: An analysis of
present and future seasonal Northern Hemisphere land snow cover simulated by
CMIP5 coupled climate models, The Cryosphere, 7, 67–80,
<a href="https://doi.org/10.5194/tc-7-67-2013" target="_blank">https://doi.org/10.5194/tc-7-67-2013</a>, 2013.
</mixed-citation></ref-html>
<ref-html id="bib1.bib6"><label>6</label><mixed-citation>
Budkyo, M. I.: The effect of solar radiation on the climate of the earth,
Tellus, 21, 611–619, 1967.
</mixed-citation></ref-html>
<ref-html id="bib1.bib7"><label>7</label><mixed-citation>
Bulygina, O. N., Groisman, P. Y., Razuvaev, V. N., and Radionov, V. F.: Snow
cover basal ice layer changes over Northern Eurasia since 1966, Environ. Res.
Lett., 5, 015004, <a href="https://doi.org/10.1088/1748-9326/5/1/015004" target="_blank">https://doi.org/10.1088/1748-9326/5/1/015004</a>, 2010.
</mixed-citation></ref-html>
<ref-html id="bib1.bib8"><label>8</label><mixed-citation>
Cess, R. O. and Potter, G. L.: Interpretation of snow-climate feedback as
produced by 17 general circulation models, Science, 253, 888–892,
<a href="https://doi.org/10.1126/science.253.5022.888" target="_blank">https://doi.org/10.1126/science.253.5022.888</a>, 1991.
</mixed-citation></ref-html>
<ref-html id="bib1.bib9"><label>9</label><mixed-citation>
Collins, M., Knutti, R., Arblaster, J., Dufresne, J.-L., Fichefet, T.,
Friedlingstein, P., Gao, X., Gutowski, W. J., Johns, T., Krinner, G.,
Shongwe, M., Tebaldi, C., Weaver, A. J., and Wehner, M.: Long-term Climate
Change: Projections, Commitments and Irreversibility, in: Climate Change
2013: The Physical Science Basis. Contribution of Working Group I to the
Fifth Assessment Report of the Intergovernmental Panel on Climate Change,
edited by: Stocker, T. F., Qin, D., Plattner, G.-K., Tignor, M., Allen, S.
K., Boschung, J., Nauels, A., Xia, Y., Bex, V., and Midgley, P. M., Cambridge
University Press, Cambridge, United Kingdom and New York, NY, USA, 2013.
</mixed-citation></ref-html>
<ref-html id="bib1.bib10"><label>10</label><mixed-citation>
Curry, J. A., Schramm, J. L., Rossow, W. B., and Randall, D.: Overview of
Arctic cloud and radiation characteristics, J. Climate, 9, 1731–1764, 1996
</mixed-citation></ref-html>
<ref-html id="bib1.bib11"><label>11</label><mixed-citation>
Dee, D. P., Uppala, S. M., Simmons, A. J., Berrisford, P., Poli, P.,
Kobayashi, S., Andrae, U., Balmaseda, M. A., Balsamo, G., Bauer, P.,
Bechtold, P., Beljaars, A. C. M., van de Berg, L., Bid- lot, J., Bormann, N.,
Delsol, C., Dragani, R., Fuentes, M., Geer, A. J., Haimberger, L., Healy, S.
B., Hersbach, H., Hólm, E. V., Isaksen, L., Kållberg, P., Köhler,
M., Matricardi, M., McNally, A. P., Monge-Sanz, B. M., Morcrette, J.-J.,
Park, B.-K., Peubey, C., de Rosnay, P., Tavolato, C., Thépaut, J.-N.,
&amp; Vitart, F.: The ERA–interim reanalysis: Configuration and performance
of the data assimilation system, Q. J. Roy. Meteor. Soc., 137, 553–597,
2011.
</mixed-citation></ref-html>
<ref-html id="bib1.bib12"><label>12</label><mixed-citation>
Derksen, C. and Brown, R.: Spring snow cover extent reductions in the
2008–2012 period exceeding climate model projections, Geophys. Res. Lett.,
39, L19504, <a href="https://doi.org/10.1029/2012GL053387" target="_blank">https://doi.org/10.1029/2012GL053387</a>, 2012.
</mixed-citation></ref-html>
<ref-html id="bib1.bib13"><label>13</label><mixed-citation>
Dufour, A., Zolina, O., and Gulev, S. K.: Atmospheric moisture transport to
the Arctic: Assessment of reanalyses and analysis of transport components, J.
Climate, 29, 5061–5081, 2016.
</mixed-citation></ref-html>
<ref-html id="bib1.bib14"><label>14</label><mixed-citation>
Dutra, E., Balsamo, G., Viterbo, P., Miranda, P. M., Beljaars, A., Schär,
C., and Elder, K.: An improved snow scheme for the ECMWF land surface model:
description and offline validation, J. Hydrometeorol., 11, 899–916, 2010.
</mixed-citation></ref-html>
<ref-html id="bib1.bib15"><label>15</label><mixed-citation>
Fernandes, R., Zhao, H., Wang, X., Key, J., Qu, X., and Hall, A.: Controls on
Northern Hemisphere snow albedo feedback quantified using satellite earth
observations, Geophys. Res. Lett., 36, L21702, <a href="https://doi.org/10.1029/2009GL040057" target="_blank">https://doi.org/10.1029/2009GL040057</a>,
2009.
</mixed-citation></ref-html>
<ref-html id="bib1.bib16"><label>16</label><mixed-citation>
Flanner, M. G., Shell, K. M., Barlage, M., Perovich, D. K., and Tschudi, M.
A.: Radiative forcing and albedo feedback from the Northern Hemisphere
cryosphere between 1979 and 2008, Nat. Geosci., 4, 151,
<a href="https://doi.org/10.1038/ngeo1062" target="_blank">https://doi.org/10.1038/ngeo1062</a>, 2011.
</mixed-citation></ref-html>
<ref-html id="bib1.bib17"><label>17</label><mixed-citation>
Fletcher, C. G., Zhao, H., Kushner, P. J., and Fernandes, R.: Using models
and satellite observations to evaluate the strength of snow albedo feedback,
J. Geophys. Res.-Atmos., 117, D11117, <a href="https://doi.org/10.1029/2012JD017724" target="_blank">https://doi.org/10.1029/2012JD017724</a>, 2012.
</mixed-citation></ref-html>
<ref-html id="bib1.bib18"><label>18</label><mixed-citation>
Fletcher, C. G., Thackeray, C. W., and Burgers, T. M.: Evaluating biases in
simulated snow albedo feedback in two generations of climate models, J.
Geophys. Res.-Atmos., 120, 12–26, 2015.
</mixed-citation></ref-html>
<ref-html id="bib1.bib19"><label>19</label><mixed-citation>
Foster, J. L., Sun, C., Walker, J. P., Kelly, R., Chang, A., Dong, J., and
Powell, H.: Quantifying the uncertainty in passive microwave snow water
equivalent observations, Remote Sens. Environ., 94, 187–203, 2005.
</mixed-citation></ref-html>
<ref-html id="bib1.bib20"><label>20</label><mixed-citation>
Gelaro, R., McCarty, W., Suárez, M. J., Todling, R., Molod, A., Takacs,
L., Randles, C. A., Darmenov, A., Bosilovich, M. G., Reichle, R., Wargan, K.,
Coy, L., Cullather, R., Draper, C., Akella, S., Buchard, V., Conaty, A., da
Silva, A. M., Gu, W., Kim, G., Koster, R., Lucchesi, R., Merkova, D.,
Nielsen, J. E., Partyka, G., Pawson, S., Putman, W., Rienecker, M., Schubert,
S. D., Sienkiewicz, M., and Zhao, B.: The Modern-Era Retrospective Analysis
for Research and Applications, Version 2 (MERRA-2), J. Climate, 30,
5419–5454, 2017.
</mixed-citation></ref-html>
<ref-html id="bib1.bib21"><label>21</label><mixed-citation>
Graversen, R. G. and Wang, M.: Polar amplification in a coupled climate model
with locked albedo, Clim. Dynam., 33, 629–643, 2009.
</mixed-citation></ref-html>
<ref-html id="bib1.bib22"><label>22</label><mixed-citation>
Graversen, R. G., Langen, P. L., and Mauritsen, T.: Polar amplification in
CCSM4: Contributions from the lapse rate and surface albedo feedbacks, J.
Climate, 27, 4433–4450, 2014.
</mixed-citation></ref-html>
<ref-html id="bib1.bib23"><label>23</label><mixed-citation>
Groisman, P. Y., Karl, T. R., Knight, R. W., and Stenchikov, G. L.: Changes
of snow cover, temperature, and radiative heat balance over the Northern
Hemisphere, J. Climate, 7, 1633–1656, 1994.
</mixed-citation></ref-html>
<ref-html id="bib1.bib24"><label>24</label><mixed-citation>
Hall, A.: The role of surface albedo feedback in climate, J. Climate, 17,
1550–1568, 2004.
</mixed-citation></ref-html>
<ref-html id="bib1.bib25"><label>25</label><mixed-citation>
Hall, A. and Qu, X.: Using the current seasonal cycle to constrain snow
albedo feedback in future climate change, Geophys. Res. Lett., 33, L03502,
<a href="https://doi.org/10.1029/2005GL025127" target="_blank">https://doi.org/10.1029/2005GL025127</a>, 2006.
</mixed-citation></ref-html>
<ref-html id="bib1.bib26"><label>26</label><mixed-citation>
Hall, D. K., Riggs, G. A., Salomonson, V. V., DiGirolamo, N. E., and Bayr, K.
J.: MODIS snow-cover products, Remote Sens. Environ., 83, 181–194, 2002.
</mixed-citation></ref-html>
<ref-html id="bib1.bib27"><label>27</label><mixed-citation>
Hall, A., Qu, X., and Neelin, J. D.: Improving predictions of summer climate
change in the United States, Geophys. Res. Lett., 35, L01702,
<a href="https://doi.org/10.1029/2007GL032012" target="_blank">https://doi.org/10.1029/2007GL032012</a>, 2008.
</mixed-citation></ref-html>
<ref-html id="bib1.bib28"><label>28</label><mixed-citation>
Kanamitsu, M., Ebisuzaki, W., Woollen, J., Yang, S. K., Hnilo, J. J.,
Fiorino, M., and Potter, G. L.: Ncep–doe amip-ii reanalysis (r-2), B.
Am. Meteorol. Soc., 83, 1631–1643, 2002.
</mixed-citation></ref-html>
<ref-html id="bib1.bib29"><label>29</label><mixed-citation>
Kevin, J. P. W., Kotlarski, S., Scherrer, S. C., and Schär, C.: The
Alpine snow-albedo feedback in regional climate models, Clim. Dynam., 48,
1109–1124, 2017.
</mixed-citation></ref-html>
<ref-html id="bib1.bib30"><label>30</label><mixed-citation>
Justice, C. O., Townshend, J. R. G., Holben, B. N., and Tucker, E. C.:
Analysis of the phenology of global vegetation using meteorological satellite
data, Int. J. Remote Sens., 6, 1271–1318, 1985.
</mixed-citation></ref-html>
<ref-html id="bib1.bib31"><label>31</label><mixed-citation>
Lian, M. S. and Cess, R. D.: Energy balance climate models: A
reappraisal of ice-albedo feedback, J. Atmos. Sci., 34,
1058–1062,  1977.
</mixed-citation></ref-html>
<ref-html id="bib1.bib32"><label>32</label><mixed-citation>
Lindsay, R., Wensnahan, M., Schweiger, A., and Zhang, J.: Evaluation of seven
different atmospheric reanalysis products in the Arctic, J. Climate, 27,
2588–2606, 2014.
</mixed-citation></ref-html>
<ref-html id="bib1.bib33"><label>33</label><mixed-citation>
Pithan, F. and Mauritsen, T.: Arctic amplification dominated by temperature
feedbacks in contemporary climate models, Nat. Geosci., 7, 181–184,
<a href="https://doi.org/10.1038/ngeo2071" target="_blank">https://doi.org/10.1038/ngeo2071</a>, 2014.
</mixed-citation></ref-html>
<ref-html id="bib1.bib34"><label>34</label><mixed-citation>
Qu, X. and Hall, A.: What controls the strength of snow-albedo feedback?, J.
Climate, 20, 3971–3981, 2007.
</mixed-citation></ref-html>
<ref-html id="bib1.bib35"><label>35</label><mixed-citation>
Qu, X. and Hall, A.: On the persistent spread in snow-albedo feedback, Clim.
Dynam., 42, 69–81, 2014.
</mixed-citation></ref-html>
<ref-html id="bib1.bib36"><label>36</label><mixed-citation>
Rasmussen, R., Baker, B., Kochendorfer, J., Meyers, T., Landolt, S., Fischer,
A. P., Black, J., Thériault, J. M., Kucera, P., Gochis, D., Smith, C.,
Nitu, R., Hall, M., Ikeda, K., and Gutmann, E.: How Well Are We Measuring
Snow: The NOAA/FAA/NCAR Winter Precipitation Test Bed, Bull. Amer. Meteor.
Soc., 93, 811–829, 2012.
</mixed-citation></ref-html>
<ref-html id="bib1.bib37"><label>37</label><mixed-citation>
Reichle, R. H., Draper, C. S., Liu, Q., Girotto, M., Mahanama, S. P., Koster,
R. D., and De Lannoy, G. J.: Assessment of MERRA-2 land surface hydrology
estimates, J. Climate, 30, 2937–2960, 2017.
</mixed-citation></ref-html>
<ref-html id="bib1.bib38"><label>38</label><mixed-citation>
Rienecker, M. M., Suarez, M. J., Gelaro, R., Todling, R., Bacmeister, J.,
Liu, E., Bosilovich, M. G., Schubert, S. D., Takacs, L., Kim, G., Bloom, S.,
Chen, J., Collins, D., Conaty, A., da Silva, A., Gu, W., Joiner, J., Koster,
R. D., Lucchesi, R., Molod, A., Owens, T., Pawson, S., Pegion, P., Redder, C.
R., Reichle, R., Robertson, F. R., Ruddick, A. G., Sienkiewicz, M., and
Woollen, J.: MERRA: NASA's Modern-Era Retrospective Analysis for Research and
Applications, J. Climate, 24, 3624–3648, 2011.
</mixed-citation></ref-html>
<ref-html id="bib1.bib39"><label>39</label><mixed-citation>
Robock, A.: Ice and snow feedbacks and the latitudinal and seasonal
distribution of climate sensitivity, J. Atmos. Sci., 40, 986–997, 1983.
</mixed-citation></ref-html>
<ref-html id="bib1.bib40"><label>40</label><mixed-citation>
Roesch, A., Gilgen, H., Wild, M., and Ohmura, A.: Assessment of GCM simulated
snow albedo using direct observations, Clim. Dynam., 15, 405–418, 1999.
</mixed-citation></ref-html>
<ref-html id="bib1.bib41"><label>41</label><mixed-citation>
Romanov, P., Gutman, G., and Csiszar, I.: Satellite-derived snow cover maps
for North America: accuracy assessment, Adv. Space Res., 30, 2455–2460,
2002.
</mixed-citation></ref-html>
<ref-html id="bib1.bib42"><label>42</label><mixed-citation>
Schiffer, R. A. and Rossow, W. B.: The International Satellite Cloud
Climatology Project(ISCCP) – The first project of the World Climate Research
Programme, American Meteorological Society, Bulletin, 64, 779–784, 1983.
</mixed-citation></ref-html>
<ref-html id="bib1.bib43"><label>43</label><mixed-citation>
Schneider, S. H. and Dickinson, R. E.: Climate modeling, Rev. Geophys., 12,
447–493, 1974.

</mixed-citation></ref-html>
<ref-html id="bib1.bib44"><label>44</label><mixed-citation>
Schubert, S. D., Wang, H., Koster, R. D., Suarez, M. J., and Groisman, P. Y.:
Northern Eurasian heat waves and droughts, J. Climate, 27, 3169–3207, 2014.
</mixed-citation></ref-html>
<ref-html id="bib1.bib45"><label>45</label><mixed-citation>
Serreze, M. C. and Barry, R. G.: Processes and impacts of Arctic
amplification: A research synthesis, Global Planet. Change, 77, 85–96, 2011.
</mixed-citation></ref-html>
<ref-html id="bib1.bib46"><label>46</label><mixed-citation>
Stroeve, J. C., Box, J. E., and Haran, T.: Evaluation of the MODIS (MOD10A1)
daily snow albedo product over the Greenland ice sheet, Remote Sens.
Environ., 105, 155–171, 2006.
</mixed-citation></ref-html>
<ref-html id="bib1.bib47"><label>47</label><mixed-citation>
Thackeray, C. W. and Fletcher, C. G.: Snow albedo feedback: Current
knowledge, importance, outstanding issues and future directions, Prog. Phys.
Geog., 40, 392–408, 2016.
</mixed-citation></ref-html>
<ref-html id="bib1.bib48"><label>48</label><mixed-citation>
Wang, Z., Schaaf, C. B., Strahler, A. H., Chopping, M. J., Román, M. O.,
Shuai, Y., Woodcock, C. E., Hollinger, D. Y., and Fitzjarrald, D. R.:
Evaluation of MODIS albedo product (MCD43A) over grassland, agriculture and
forest surface types during dormant and snow-covered periods, Remote Sens.
Environ., 140, 60–77, 2014.
</mixed-citation></ref-html>
<ref-html id="bib1.bib49"><label>49</label><mixed-citation>
Wei, X., Hahmann, A. N., Dickinson, R. E., Yang, Z. L., Zeng, X., Schaudt, K.
J., Schaaf, C. B., and Strugnell, N.: Comparison of albedos computed by land
surface models and evaluation against remotely sensed data, J. Geophys.
Res.-Atmos., 106, 20687–20702, 2001.
</mixed-citation></ref-html>
<ref-html id="bib1.bib50"><label>50</label><mixed-citation>
Wegmann, M., Orsolini, Y., Dutra, E., Bulygina, O., Sterin, A., and
Brönnimann, S.: Eurasian snow depth in long-term climate reanalyses, The
Cryosphere, 11, 923–935, <a href="https://doi.org/10.5194/tc-11-923-2017" target="_blank">https://doi.org/10.5194/tc-11-923-2017</a>, 2017.
</mixed-citation></ref-html>
<ref-html id="bib1.bib51"><label>51</label><mixed-citation>
Wexler, H.: Radiation balance of the Earth as a factor in climatic change,
in: Climatic Change, edited by: Shapley, H., Cambridge, Harvard University
Press, 73–105, 1953.
</mixed-citation></ref-html>
<ref-html id="bib1.bib52"><label>52</label><mixed-citation>
Xiao, L., Che, T., Chen, L., Xie, H., and Dai, L.: Quantifying Snow Albedo
Radiative Forcing and Its Feedback during 2003–2016, Remote Sensing, 9, 883,
<a href="https://doi.org/10.3390/rs9090883" target="_blank">https://doi.org/10.3390/rs9090883</a>, 2017.
</mixed-citation></ref-html>--></article>
