<?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">ESSD</journal-id><journal-title-group>
    <journal-title>Earth System Science Data</journal-title>
    <abbrev-journal-title abbrev-type="publisher">ESSD</abbrev-journal-title><abbrev-journal-title abbrev-type="nlm-ta">Earth Syst. Sci. Data</abbrev-journal-title>
  </journal-title-group><issn pub-type="epub">1866-3516</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/essd-11-1603-2019</article-id><title-group><article-title>High-temporal-resolution water level and storage change data sets for lakes on the Tibetan Plateau during 2000–2017 using multiple altimetric missions and Landsat-derived lake shoreline positions</article-title><alt-title>High-temporal-resolution water level and storage change data sets (Tibetan Plateau, 2000–2017)</alt-title>
      </title-group><?xmltex \runningtitle{High-temporal-resolution water level and storage change data sets (Tibetan Plateau, 2000--2017)}?><?xmltex \runningauthor{X. Li et al.}?>
      <contrib-group>
        <contrib contrib-type="author" corresp="no" rid="aff1">
          <name><surname>Li</surname><given-names>Xingdong</given-names></name>
          
        <ext-link>https://orcid.org/0000-0002-8037-5894</ext-link></contrib>
        <contrib contrib-type="author" corresp="yes" rid="aff1">
          <name><surname>Long</surname><given-names>Di</given-names></name>
          <email>dlong@tsinghua.edu.cn</email>
        <ext-link>https://orcid.org/0000-0001-9033-5039</ext-link></contrib>
        <contrib contrib-type="author" corresp="no" rid="aff1">
          <name><surname>Huang</surname><given-names>Qi</given-names></name>
          
        </contrib>
        <contrib contrib-type="author" corresp="no" rid="aff1">
          <name><surname>Han</surname><given-names>Pengfei</given-names></name>
          
        </contrib>
        <contrib contrib-type="author" corresp="no" rid="aff1">
          <name><surname>Zhao</surname><given-names>Fanyu</given-names></name>
          
        </contrib>
        <contrib contrib-type="author" corresp="no" rid="aff2">
          <name><surname>Wada</surname><given-names>Yoshihide</given-names></name>
          
        <ext-link>https://orcid.org/0000-0003-4770-2539</ext-link></contrib>
        <aff id="aff1"><label>1</label><institution>State Key Laboratory of Hydroscience and Engineering, Department of
Hydraulic Engineering, <?xmltex \hack{\break}?>Tsinghua University, Beijing, China</institution>
        </aff>
        <aff id="aff2"><label>2</label><institution>International Institute for Applied Systems Analysis (IIASA), 2361
Laxenburg, Austria</institution>
        </aff>
      </contrib-group>
      <author-notes><corresp id="corr1">Di Long (dlong@tsinghua.edu.cn)</corresp></author-notes><pub-date><day>28</day><month>October</month><year>2019</year></pub-date>
      
      <volume>11</volume>
      <issue>4</issue>
      <fpage>1603</fpage><lpage>1627</lpage>
      <history>
        <date date-type="received"><day>21</day><month>February</month><year>2019</year></date>
           <date date-type="rev-request"><day>15</day><month>March</month><year>2019</year></date>
           <date date-type="rev-recd"><day>4</day><month>September</month><year>2019</year></date>
           <date date-type="accepted"><day>22</day><month>September</month><year>2019</year></date>
      </history>
      <permissions>
        <copyright-statement>Copyright: © 2019 Xingdong Li et al.</copyright-statement>
        <copyright-year>2019</copyright-year>
      <license license-type="open-access"><license-p>This work is licensed under the Creative Commons Attribution 4.0 International License. To view a copy of this licence, visit <ext-link ext-link-type="uri" xlink:href="https://creativecommons.org/licenses/by/4.0/">https://creativecommons.org/licenses/by/4.0/</ext-link></license-p></license></permissions><self-uri xlink:href="https://essd.copernicus.org/articles/11/1603/2019/essd-11-1603-2019.html">This article is available from https://essd.copernicus.org/articles/11/1603/2019/essd-11-1603-2019.html</self-uri><self-uri xlink:href="https://essd.copernicus.org/articles/11/1603/2019/essd-11-1603-2019.pdf">The full text article is available as a PDF file from https://essd.copernicus.org/articles/11/1603/2019/essd-11-1603-2019.pdf</self-uri>
      <abstract><title>Abstract</title>
    <p id="d1e134">The Tibetan Plateau (TP), known as Asia's water tower, is
quite sensitive to climate change, which is reflected by changes in hydrologic state
variables such as lake water storage. Given the extremely limited ground
observations on the TP due to the harsh environment and complex terrain, we
exploited multiple altimetric missions and Landsat satellite data to create
high-temporal-resolution lake water level and storage change time series at
weekly to monthly timescales for 52 large lakes (50 lakes larger than 150 km<inline-formula><mml:math id="M1" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">2</mml:mn></mml:msup></mml:math></inline-formula> and 2 lakes larger than 100 km<inline-formula><mml:math id="M2" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">2</mml:mn></mml:msup></mml:math></inline-formula>) on the TP during 2000–2017. The data sets are available online at <ext-link xlink:href="https://doi.org/10.1594/PANGAEA.898411" ext-link-type="DOI">10.1594/PANGAEA.898411</ext-link> (Li et al., 2019). With
Landsat archives and altimetry data, we developed water levels from lake
shoreline positions (i.e., Landsat-derived water levels) that cover the
study period and serve as an ideal reference for merging multisource lake
water levels with systematic biases being removed. To validate the
Landsat-derived water levels, field experiments were carried out in two
typical lakes, and theoretical uncertainty analysis was performed based on
high-resolution optical images (0.8 m) as well. The RMSE of the
Landsat-derived water levels is 0.11 m compared with the in situ
measurements, consistent with the magnitude from theoretical analysis
(0.1–0.2 m). The accuracy of the Landsat-derived water levels that can be
derived in relatively small lakes is comparable with most altimetry data.
The resulting merged Landsat-derived and altimetric lake water levels can
provide accurate information on multiyear and short-term monitoring of lake
water levels and storage changes on the TP, and critical information on lake overflow flood monitoring and prediction as the expansion of some TP lakes
becomes a serious threat to surrounding residents and infrastructure.</p>
  </abstract>
    </article-meta>
  </front>
<body>
      

<sec id="Ch1.S1" sec-type="intro">
  <label>1</label><title>Introduction</title>
      <p id="d1e167">The Tibetan Plateau (TP), providing vital water resources for more than a
billion population in Asia, is a sensitive region undergoing rapid climate
change (Field et al., 2014). There are more than 1200 alpine lakes
larger than 1 km<inline-formula><mml:math id="M3" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">2</mml:mn></mml:msup></mml:math></inline-formula> on the TP, where glaciers and permafrost are also
widely distributed. With little disturbance by human activity in this area,
lake storage changes may serve as an important indicator that reflects
changes in regional hydrologic processes and responses to climate change.
Wang et al. (2018) showed that global endorheic basins are experiencing
a decline in water storage, whereas the endorheic basin on the TP is an
exception. Given the fact that TP lakes have been expanding for more than 20 years (Pekel et al., 2016), quality data sets on lake water level and/or
storage could be the basis for investigating its causes (e.g., climate
change/variability) and interactions with the water/energy cycles and human
society (e.g., increasing risks of inundation and overflow floods).</p>
      <?pagebreak page1604?><p id="d1e179">As an important component of the hydrosphere, terrestrial water cycle, and
global water balance, millions of inland water bodies such as lakes,
wetlands, and reservoirs have been investigated globally, and their total
storage was estimated to be <inline-formula><mml:math id="M4" display="inline"><mml:mrow><mml:mn mathvariant="normal">181.9</mml:mn><mml:mo>×</mml:mo><mml:msup><mml:mn mathvariant="normal">10</mml:mn><mml:mn mathvariant="normal">3</mml:mn></mml:msup></mml:mrow></mml:math></inline-formula> km<inline-formula><mml:math id="M5" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">3</mml:mn></mml:msup></mml:math></inline-formula> based on
statistical models (Lehner and Döll, 2004; Messager et al., 2016;
Pekel et al., 2016). Lake storage changes that play an important role in the
regional water balance can be derived from changes in lake water level and
area (Frappart et al., 2005). Lake water levels and areas are mostly
derived from satellite remote sensing due to the scarcity of in situ data
across the TP, where the harsh environment and complex terrain make in situ
measurements difficult to perform and costly (Crétaux et al., 2016;
Song et al., 2013; Yao et al., 2018b; Zhang et al., 2017a). Lake water
levels can be monitored using satellite altimeters initially designed for
sea surface topography or ice sheet/sea ice freeboard height measurements.
Satellite altimeters determine the range between the nadir point and
satellite by analyzing the waveforms of reflected electromagnetic pulses.</p>
      <p id="d1e206">There are two major categories of satellite altimeters, i.e., laser and
radar. Laser altimeters, e.g., the Ice, Cloud, and land Elevation Satellite (ICESat), operating in the near-infrared band have smaller
footprints and generally higher accuracy than radar altimeters, facilitating
applications in glacier/ice mass balance studies (Neckel et al.,
2014; Sørensen et al., 2011). Radar altimeters, operating in the
microwave band, have larger footprints and are more likely to be
contaminated by a signal from complex terrain when applied to inland water
bodies. Nevertheless, it is possible to remove these impacts with waveform
retracking algorithms (Guo et al., 2009; Huang et al., 2018;
Jiang et al., 2017). Zhang et al. (2011) mapped water level changes in
111 TP lakes for the 2003–2009 period using ICESat data that have a
temporal resolution of 91 days. ICESat data have relatively denser ground
tracks but a lower temporal resolution than most of other altimetric
missions. This means that ICESat covers more lakes but provides few water
levels for each lake. After ICESat was decommissioned in 2010, CryoSat-2
data starting from 2010 were adopted in related studies (Jiang
et al., 2017), due to its similar dense ground tracks and competitive
precision compared to ICESat. Other altimetric missions, such as TOPEX/Poseidon
(T/P), Jason-1/2/3, the European Remote Sensing (ERS-1/2) satellite, and
Envisat, also have some but relatively limited applications in monitoring
changes in lake water level on the TP due to sparse ground tracks. In this
study, multisource altimetry data (i.e., Jason-1/2/3, Envisat, ICESat, and
CryoSat-2) were combined if available for lakes in this study, with the
Landsat-derived water levels developed in this study as a critical reference
to increase the water level observations and merging data from multiple
altimetric missions.</p>
      <p id="d1e209">Changes in lake area can be captured by optical or synthetic aperture radar
(SAR) images from medium- or high-spatial-resolution remote sensing data,
such as Landsat and Sentinel series. Extraction of lake water bodies can be
manually (Wan et al., 2016) or automatically (Zhang et al.,
2017b) achieved. Automatic water extraction methods based on the
water index and auto-thresholding are more efficient in dealing with a mass
of remote sensing images. Even so, acquisition and preprocessing of such a
large amount of historical data (<inline-formula><mml:math id="M6" display="inline"><mml:mo lspace="0mm">∼</mml:mo></mml:math></inline-formula> 10 TB) covering TP lakes are
still intractable for researchers with limited computational resources. With
the help of cloud-based platforms, such as the Google Earth Engine (GEE)
that significantly reduces data downloading and preprocessing time, tens of
thousands of images may be processed online in days instead of months
(Gorelick et al., 2017). In this study, more than 20 000
Landsat images were processed online using GEE to extract lake water bodies
based on the water index (McFeeters, 1996) and Otsu algorithm (Otsu,
1979).</p>
      <p id="d1e220">There have been studies focusing on changes in lake water storage on the TP
over recent decades; e.g., Zhang et al. (2017a) examined
changes in water storage for <inline-formula><mml:math id="M7" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 70 lakes from the 1970s to 2015
with ICESat altimetry data and Landsat archives. An individual lake area
data set from the 1970s and annual area data after 1989 were used. Due to
the short time span of ICESat, they used the hypsometric method to convert
lake areas into water levels. Yao et al. (2018b) used digital elevation models (DEMs) and
optical images to develop hypsometric curves for lakes on the central TP
and estimated annual changes in water storage for 871 lakes from 2002 to
2015. These studies have a wide spatial coverage of lakes but relatively
lower temporal resolution and little spaceborne altimetric information,
which may limit the accuracy of trends in lake water level/storage in some
cases and short-term monitoring of lake overflow floods. The Laboratoire
d'Etudes en Géophysique et Océanographie Spatiales (LEGOS) Hydroweb
provides a lake data set, including multisource altimetry-based changes in
lake water level and storage as well as hypsometric curves for 22 TP lakes
(Crétaux et al., 2016, 2011b). The data set
incorporates more spaceborne altimetric information and has a higher
temporal resolution. However, there may be a remaining bias when different
sources of altimetric data are merged, due to the lack of some important
reference that can be derived from optical remote sensing to be shown in
this study. We term the reference data the “Landsat-derived water level”
to be introduced in Sect. 3.2. Here, we list recent studies and data sets
(Table 1) to provide a concise summary on remote sensing monitoring of water
levels and storage changes over lakes on the TP.</p>

<?xmltex \floatpos{t}?><table-wrap id="Ch1.T1" specific-use="star"><?xmltex \currentcnt{1}?><label>Table 1</label><caption><p id="d1e233">Recent studies and data sets on TP lakes. H, A, and V in the table
denote lake water level, area, and volume, respectively.</p></caption><oasis:table frame="topbot"><oasis:tgroup cols="6">
     <oasis:colspec colnum="1" colname="col1" align="justify" colwidth="42.679134pt"/>
     <oasis:colspec colnum="2" colname="col2" align="right"/>
     <oasis:colspec colnum="3" colname="col3" align="justify" colwidth="56.905512pt"/>
     <oasis:colspec colnum="4" colname="col4" align="justify" colwidth="71.13189pt"/>
     <oasis:colspec colnum="5" colname="col5" align="justify" colwidth="85.358268pt"/>
     <oasis:colspec colnum="6" colname="col6" align="justify" colwidth="113.811024pt"/>
     <oasis:thead>
       <oasis:row rowsep="1">
         <oasis:entry colname="col1">Reference</oasis:entry>
         <oasis:entry colname="col2">No. of lakes</oasis:entry>
         <oasis:entry colname="col3">Data type</oasis:entry>
         <oasis:entry colname="col4">Time span</oasis:entry>
         <oasis:entry colname="col5">Temporal resolution</oasis:entry>
         <oasis:entry colname="col6">Source data</oasis:entry>
       </oasis:row>
     </oasis:thead>
     <oasis:tbody>
       <oasis:row rowsep="1">
         <oasis:entry colname="col1">Song et al.<?xmltex \hack{\hfill\break}?>(2013)</oasis:entry>
         <oasis:entry colname="col2">30</oasis:entry>
         <oasis:entry colname="col3">H, A, V, and<?xmltex \hack{\hfill\break}?>hypsometric curve</oasis:entry>
         <oasis:entry colname="col4">4 records for the<?xmltex \hack{\hfill\break}?>1970s, 1990, 2000,<?xmltex \hack{\hfill\break}?>and 2011</oasis:entry>
         <oasis:entry colname="col5">Decadal</oasis:entry>
         <oasis:entry colname="col6">Altimetry data: <?xmltex \hack{\hfill\break}?>ICESat <?xmltex \hack{\hfill\break}?>Optical data: <?xmltex \hack{\hfill\break}?>Landsat 5/7 TM/ETM<inline-formula><mml:math id="M8" display="inline"><mml:mo>+</mml:mo></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row rowsep="1">
         <oasis:entry colname="col1">Crétaux et<?xmltex \hack{\hfill\break}?>al. (2016)</oasis:entry>
         <oasis:entry colname="col2">22</oasis:entry>
         <oasis:entry colname="col3">H, A, V, and<?xmltex \hack{\hfill\break}?>hypsometric curve</oasis:entry>
         <oasis:entry colname="col4">1995–2015 <?xmltex \hack{\hfill\break}?>relative bias partially removed<?xmltex \hack{\hfill\break}?>(only for altimeters with overlapping period)</oasis:entry>
         <oasis:entry colname="col5">Submonthly for lakes<?xmltex \hack{\hfill\break}?>with T/P and Jason-1/2 data, and <inline-formula><mml:math id="M9" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> monthly for lakes without Jason-1/2 or T/P data</oasis:entry>
         <oasis:entry colname="col6">Altimetry data: <?xmltex \hack{\hfill\break}?>T/P, ERS-2, GFO, Envisat,<?xmltex \hack{\hfill\break}?>Jason-1/2, SARAL, ICESat,<?xmltex \hack{\hfill\break}?>and CryoSat-2 <?xmltex \hack{\hfill\break}?>Optical data: <?xmltex \hack{\hfill\break}?>Landsat 5/7/8 TM/ETM<inline-formula><mml:math id="M10" display="inline"><mml:mo>+</mml:mo></mml:math></inline-formula>/OLI and MODIS</oasis:entry>
       </oasis:row>
       <oasis:row rowsep="1">
         <oasis:entry colname="col1">Jiang et al.<?xmltex \hack{\hfill\break}?>(2017)</oasis:entry>
         <oasis:entry colname="col2">70</oasis:entry>
         <oasis:entry colname="col3">H</oasis:entry>
         <oasis:entry colname="col4">2003–2015 <?xmltex \hack{\hfill\break}?>relative bias<?xmltex \hack{\hfill\break}?>between ICESat<?xmltex \hack{\hfill\break}?>and CryoSat-2<?xmltex \hack{\hfill\break}?>unremoved</oasis:entry>
         <oasis:entry colname="col5"><inline-formula><mml:math id="M11" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> Monthly</oasis:entry>
         <oasis:entry colname="col6">Altimetry data: <?xmltex \hack{\hfill\break}?>ICESat and CryoSat-2</oasis:entry>
       </oasis:row>
       <oasis:row rowsep="1">
         <oasis:entry colname="col1">Zhang et al. (2017a)</oasis:entry>
         <oasis:entry colname="col2">60–70</oasis:entry>
         <oasis:entry colname="col3">H, A, V, and<?xmltex \hack{\hfill\break}?>hypsometric curve</oasis:entry>
         <oasis:entry colname="col4">One record for the<?xmltex \hack{\hfill\break}?>1970s, and annual<?xmltex \hack{\hfill\break}?>data for 1989–2015</oasis:entry>
         <oasis:entry colname="col5">Annual</oasis:entry>
         <oasis:entry colname="col6">Altimetry data: <?xmltex \hack{\hfill\break}?>ICESat <?xmltex \hack{\hfill\break}?>Optical data: <?xmltex \hack{\hfill\break}?>Landsat 5/7/8 TM/ETM<inline-formula><mml:math id="M12" display="inline"><mml:mo>+</mml:mo></mml:math></inline-formula>/OLI</oasis:entry>
       </oasis:row>
       <oasis:row rowsep="1">
         <oasis:entry colname="col1">Li et al.<?xmltex \hack{\hfill\break}?>(2017b)</oasis:entry>
         <oasis:entry colname="col2">167</oasis:entry>
         <oasis:entry colname="col3">H</oasis:entry>
         <oasis:entry colname="col4">2002–2012</oasis:entry>
         <oasis:entry colname="col5"><inline-formula><mml:math id="M13" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> Monthly</oasis:entry>
         <oasis:entry colname="col6">Altimetry data: <?xmltex \hack{\hfill\break}?>ICESat and Envisat</oasis:entry>
       </oasis:row>
       <oasis:row rowsep="1">
         <oasis:entry colname="col1">Yao et al.<?xmltex \hack{\hfill\break}?>(2018b)</oasis:entry>
         <oasis:entry colname="col2">871</oasis:entry>
         <oasis:entry colname="col3">H, A, V, and<?xmltex \hack{\hfill\break}?>hypsometric curve</oasis:entry>
         <oasis:entry colname="col4">2002–2015</oasis:entry>
         <oasis:entry colname="col5">Annual</oasis:entry>
         <oasis:entry colname="col6">Optical data: <?xmltex \hack{\hfill\break}?>Landsat 5/7/8 TM/ETM<inline-formula><mml:math id="M14" display="inline"><mml:mo>+</mml:mo></mml:math></inline-formula>/OLI and HJ-1A/1B <?xmltex \hack{\hfill\break}?>DEM data: <?xmltex \hack{\hfill\break}?>SRTM and ASTER</oasis:entry>
       </oasis:row>
       <oasis:row rowsep="1">
         <oasis:entry colname="col1">Hwang et<?xmltex \hack{\hfill\break}?>al. (2019)</oasis:entry>
         <oasis:entry colname="col2">59</oasis:entry>
         <oasis:entry colname="col3">H</oasis:entry>
         <oasis:entry colname="col4">2003–2016 <?xmltex \hack{\hfill\break}?>relative bias partially removed<?xmltex \hack{\hfill\break}?>(only for lakes with<?xmltex \hack{\hfill\break}?>Jason data/in situ<?xmltex \hack{\hfill\break}?>data)</oasis:entry>
         <oasis:entry colname="col5">Submonthly for lakes<?xmltex \hack{\hfill\break}?>with Jason-2 data, and <inline-formula><mml:math id="M15" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> monthly for lakes<?xmltex \hack{\hfill\break}?>without Jason-2 data</oasis:entry>
         <oasis:entry colname="col6">Altimetry data: Jason-2/3,<?xmltex \hack{\hfill\break}?>SARAL, ICESat, and<?xmltex \hack{\hfill\break}?>CryoSat-2 <?xmltex \hack{\hfill\break}?>(Jason-3 data for validation)</oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Our study</oasis:entry>
         <oasis:entry colname="col2">52</oasis:entry>
         <oasis:entry colname="col3">H, A, V, and<?xmltex \hack{\hfill\break}?>hypsometric curve</oasis:entry>
         <oasis:entry colname="col4">2000–2017 <?xmltex \hack{\hfill\break}?>all relative biases<?xmltex \hack{\hfill\break}?>removed</oasis:entry>
         <oasis:entry colname="col5">Submonthly for most<?xmltex \hack{\hfill\break}?>lakes</oasis:entry>
         <oasis:entry colname="col6">Altimetry data: Jason-1/2/3,<?xmltex \hack{\hfill\break}?>Envisat, ICESat, and <?xmltex \hack{\hfill\break}?>CryoSat-2 <?xmltex \hack{\hfill\break}?>Optical data: <?xmltex \hack{\hfill\break}?>Landsat 5/7/8 TM/ETM<inline-formula><mml:math id="M16" display="inline"><mml:mo>+</mml:mo></mml:math></inline-formula>/OLI</oasis:entry>
       </oasis:row>
     </oasis:tbody>
   </oasis:tgroup></oasis:table></table-wrap>

      <p id="d1e628">The overall objective of this study was to examine multiyear and short-term
changes in water level and storage across 52 lakes with surface areas larger
than 150 km<inline-formula><mml:math id="M17" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">2</mml:mn></mml:msup></mml:math></inline-formula> on the TP by merging multisource altimetry and optical
remote sensing images to generate more coherent high-temporal-resolution
lake water level and storage change data sets ranging from weekly to monthly
timescales during 2000–2017 and the hypsometric curve (i.e., the lake-area–water-level relationship) for each lake. To investigate changes in lake
storage,<?pagebreak page1605?> lake water levels and areas need to be derived from multisource
remote sensing.</p>

      <?xmltex \floatpos{t}?><fig id="Ch1.F1" specific-use="star"><?xmltex \currentcnt{1}?><label>Figure 1</label><caption><p id="d1e642">Spatial (the number of lakes covered) and temporal coverage and
their overlap periods of multiple satellite altimetric missions used in this
study, including Jason-1/2/3, Envisat, ICESat, and CryoSat-2.</p></caption>
        <?xmltex \igopts{width=341.433071pt}?><graphic xlink:href="https://essd.copernicus.org/articles/11/1603/2019/essd-11-1603-2019-f01.png"/>

      </fig>

      <p id="d1e651">First, water levels from various satellite altimeters (Fig. 1) for each
lake as well as lake shoreline positions and lake areas from optical remote
sensing images (i.e., Landsat) were derived. Second, systematic biases
between different altimetry data were removed by either comparing the mean
water levels during the overlap period (Fig. 1) or comparing the two water
level time series with lake shoreline positions, depending on the length of
the overlap period (details can be found in Sect. 3.1). Lake-shoreline-position-derived water levels, termed the Landsat-derived water levels in
this study, can serve as a unique source of information reflecting water
levels as well as a data merging reference. We will show that after deriving
two or three regression parameters, lake shoreline positions can well
reflect lake water levels with comparable accuracy to altimetry-derived
water levels. Third, with information on lake water levels and areas derived
from altimetry data and optical remote sensing images, the hypsometric curve
that describes the relationship between the lake<?pagebreak page1606?> water level and storage
changes can be derived. Fourth, the integral of the hypsometric curve was
performed to convert lake water levels into storage changes.</p>
      <p id="d1e655">Results of this study provide a comprehensive and detailed assessment of
changes in lake level and storage on the TP for the recent 2 decades and
short-term monitoring of lake overflow floods for some lakes. This study
could largely benefit more detailed investigations into lakes, lake basins,
and regional climate change, because the generated data sets have the
highest temporal resolution during the study period with systematic biases
well removed. To ensure the data quality, field experiments were carried out
and in situ data were collected to examine the uncertainty in the
Landsat-derived water levels. Users are free to access the data set
described in this paper at <ext-link xlink:href="https://doi.org/10.1594/PANGAEA.898411" ext-link-type="DOI">10.1594/PANGAEA.898411</ext-link> (Li et al., 2019).</p>
</sec>
<sec id="Ch1.S2">
  <label>2</label><title>Study area and data</title>
<sec id="Ch1.S2.SS1">
  <label>2.1</label><title>Study area</title>
      <p id="d1e676">The TP can be generally divided into 12 major basins (Wan et al., 2016;
Zhang et al., 2013), among which the inner/central TP (CP) is the only
endorheic basin and home to most TP lakes, including <inline-formula><mml:math id="M18" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 300 TP
lakes larger than 10 km<inline-formula><mml:math id="M19" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">2</mml:mn></mml:msup></mml:math></inline-formula>. Therefore, it was chosen as the main study
area. The endorheic basin covers an area of <inline-formula><mml:math id="M20" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 710 000 km<inline-formula><mml:math id="M21" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">2</mml:mn></mml:msup></mml:math></inline-formula>
(<inline-formula><mml:math id="M22" display="inline"><mml:mo lspace="0mm">∼</mml:mo></mml:math></inline-formula> 28 % of total TP) with a mean elevation of
<inline-formula><mml:math id="M23" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 4900 m and has a semiarid plateau climate with annual
precipitation ranging from 96 to 295 mm (Li et al., 2017c). Most lakes in
the endorheic basin were expanding under the influence of climate
change/variability as opposed to other areas in the TP, e.g., Selin Co
exceeded Nam Co in area and consequently became the largest lake in the
endorheic basin between 2011 and 2012 and expanded by 26 % over the past 40 years (Zhou et al., 2015), whereas Yamzhog Yumco (also known as Yamdrok Lake; outside the
endorheic basin, 350 km to the southeast of Selin Co) shrunk by
<inline-formula><mml:math id="M24" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 11 % during 2002–2014 according to Wan et al. (2016).
Located in the southeast endorheic basin, the Nam Co basin covering about
10 800 km<inline-formula><mml:math id="M25" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">2</mml:mn></mml:msup></mml:math></inline-formula>, with 19 % of the basin lake water area and a mean lake
elevation of <inline-formula><mml:math id="M26" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 4730 m, was chosen as a field experiment spot.
The mean annual temperature and precipitation of Nam Co are 1.3<inline-formula><mml:math id="M27" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula>
and 486 mm, respectively (Li et al., 2017a). The other
experiment spot was Yamzhog Yumco, which has a mean lake elevation of
<inline-formula><mml:math id="M28" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 4440 m. Subject to steep mountainous terrain, the lake has
a narrow-strip shape with complex shorelines. The basin of Yamzhog Yumco
covers <inline-formula><mml:math id="M29" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 6100 km<inline-formula><mml:math id="M30" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">2</mml:mn></mml:msup></mml:math></inline-formula>, with mean annual temperature and
precipitation of 2.8<inline-formula><mml:math id="M31" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> and <inline-formula><mml:math id="M32" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 360 mm, respectively
(Yu et al., 2011). An overall map of experiment lakes is given in
Fig. 2.</p>

      <?xmltex \floatpos{t}?><fig id="Ch1.F2"><?xmltex \currentcnt{2}?><label>Figure 2</label><caption><p id="d1e800">Experiment locations: Nam Co and Yamzhog Yumco. Nam Co is located
in the endorheic basin of the TP, while Yamzhog Yumco is located in the
Yarlung Zangbo river basin (the upper Brahmaputra River). Both lakes are
close to Lhasa city.</p></caption>
          <?xmltex \igopts{width=236.157874pt}?><graphic xlink:href="https://essd.copernicus.org/articles/11/1603/2019/essd-11-1603-2019-f02.png"/>

        </fig>

<?xmltex \floatpos{t}?><table-wrap id="Ch1.T2" specific-use="star"><?xmltex \currentcnt{2}?><label>Table 2</label><caption><p id="d1e812">Multisource altimetry data used in this study.</p></caption><oasis:table frame="topbot"><?xmltex \begin{scaleboxenv}{.9}[.9]?><oasis:tgroup cols="9">
     <oasis:colspec colnum="1" colname="col1" align="left"/>
     <oasis:colspec colnum="2" colname="col2" align="left"/>
     <oasis:colspec colnum="3" colname="col3" align="left"/>
     <oasis:colspec colnum="4" colname="col4" align="left"/>
     <oasis:colspec colnum="5" colname="col5" align="right"/>
     <oasis:colspec colnum="6" colname="col6" align="right"/>
     <oasis:colspec colnum="7" colname="col7" align="left"/>
     <oasis:colspec colnum="8" colname="col8" align="right"/>
     <oasis:colspec colnum="9" colname="col9" align="left"/>
     <oasis:thead>
       <oasis:row>
         <oasis:entry colname="col1"/>
         <oasis:entry colname="col2"/>
         <oasis:entry colname="col3"/>
         <oasis:entry colname="col4"/>
         <oasis:entry colname="col5">Repeat</oasis:entry>
         <oasis:entry colname="col6">Footprint</oasis:entry>
         <oasis:entry colname="col7">Footprint</oasis:entry>
         <oasis:entry colname="col8">Lake no.</oasis:entry>
         <oasis:entry colname="col9"/>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Mission</oasis:entry>
         <oasis:entry colname="col2">Sensor (type)</oasis:entry>
         <oasis:entry colname="col3">Data record</oasis:entry>
         <oasis:entry colname="col4">Duration</oasis:entry>
         <oasis:entry colname="col5">cycle</oasis:entry>
         <oasis:entry colname="col6">interval</oasis:entry>
         <oasis:entry colname="col7">diameter</oasis:entry>
         <oasis:entry colname="col8">with data</oasis:entry>
         <oasis:entry colname="col9">Data source</oasis:entry>
       </oasis:row>
       <oasis:row rowsep="1">
         <oasis:entry colname="col1"/>
         <oasis:entry colname="col2"/>
         <oasis:entry colname="col3"/>
         <oasis:entry colname="col4"/>
         <oasis:entry colname="col5">(day)</oasis:entry>
         <oasis:entry colname="col6">(m)</oasis:entry>
         <oasis:entry colname="col7">(km)</oasis:entry>
         <oasis:entry colname="col8"/>
         <oasis:entry colname="col9"/>
       </oasis:row>
     </oasis:thead>
     <oasis:tbody>
       <oasis:row>
         <oasis:entry colname="col1">Jason-1</oasis:entry>
         <oasis:entry colname="col2">Poseidon-2 (radar)</oasis:entry>
         <oasis:entry colname="col3">S-GDR</oasis:entry>
         <oasis:entry colname="col4">2002–2013</oasis:entry>
         <oasis:entry colname="col5">10</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M34" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 300</oasis:entry>
         <oasis:entry colname="col7">2–4</oasis:entry>
         <oasis:entry colname="col8">12</oasis:entry>
         <oasis:entry colname="col9">CNES Aviso<inline-formula><mml:math id="M35" display="inline"><mml:mo>+</mml:mo></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Jason-2</oasis:entry>
         <oasis:entry colname="col2">Poseidon-3 (radar)</oasis:entry>
         <oasis:entry colname="col3">S-GDR</oasis:entry>
         <oasis:entry colname="col4">2008–</oasis:entry>
         <oasis:entry colname="col5">10</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M36" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 300</oasis:entry>
         <oasis:entry colname="col7">2–4</oasis:entry>
         <oasis:entry colname="col8">12</oasis:entry>
         <oasis:entry colname="col9">CNES Aviso<inline-formula><mml:math id="M37" display="inline"><mml:mo>+</mml:mo></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Jason-3</oasis:entry>
         <oasis:entry colname="col2">Poseidon-3B (radar)</oasis:entry>
         <oasis:entry colname="col3">S-GDR</oasis:entry>
         <oasis:entry colname="col4">2016–</oasis:entry>
         <oasis:entry colname="col5">10</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M38" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 300</oasis:entry>
         <oasis:entry colname="col7">2–4</oasis:entry>
         <oasis:entry colname="col8">12</oasis:entry>
         <oasis:entry colname="col9">CNES Aviso<inline-formula><mml:math id="M39" display="inline"><mml:mo>+</mml:mo></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Envisat</oasis:entry>
         <oasis:entry colname="col2">RA-2 (radar)</oasis:entry>
         <oasis:entry colname="col3">GDR</oasis:entry>
         <oasis:entry colname="col4">2002–2010</oasis:entry>
         <oasis:entry colname="col5">35</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M40" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 390</oasis:entry>
         <oasis:entry colname="col7">3.4</oasis:entry>
         <oasis:entry colname="col8">35</oasis:entry>
         <oasis:entry colname="col9">ESA</oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">CryoSat-2</oasis:entry>
         <oasis:entry colname="col2">SIRAL (radar)</oasis:entry>
         <oasis:entry colname="col3">InSAR Level 1</oasis:entry>
         <oasis:entry colname="col4">2010–</oasis:entry>
         <oasis:entry colname="col5">369</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M41" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 280</oasis:entry>
         <oasis:entry colname="col7"><inline-formula><mml:math id="M42" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 1.65</oasis:entry>
         <oasis:entry colname="col8">51</oasis:entry>
         <oasis:entry colname="col9">ESA</oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1"/>
         <oasis:entry colname="col2"/>
         <oasis:entry colname="col3"/>
         <oasis:entry colname="col4"/>
         <oasis:entry colname="col5">(subcycle 30)</oasis:entry>
         <oasis:entry colname="col6"/>
         <oasis:entry colname="col7">(across track),</oasis:entry>
         <oasis:entry colname="col8"/>
         <oasis:entry colname="col9"/>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1"/>
         <oasis:entry colname="col2"/>
         <oasis:entry colname="col3"/>
         <oasis:entry colname="col4"/>
         <oasis:entry colname="col5"/>
         <oasis:entry colname="col6"/>
         <oasis:entry colname="col7"><inline-formula><mml:math id="M43" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 0.3</oasis:entry>
         <oasis:entry colname="col8"/>
         <oasis:entry colname="col9"/>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1"/>
         <oasis:entry colname="col2"/>
         <oasis:entry colname="col3"/>
         <oasis:entry colname="col4"/>
         <oasis:entry colname="col5"/>
         <oasis:entry colname="col6"/>
         <oasis:entry colname="col7">(along track)</oasis:entry>
         <oasis:entry colname="col8"/>
         <oasis:entry colname="col9"/>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">ICESat</oasis:entry>
         <oasis:entry colname="col2">GLAS (laser)</oasis:entry>
         <oasis:entry colname="col3">GLAH 14</oasis:entry>
         <oasis:entry colname="col4">2003–2009</oasis:entry>
         <oasis:entry colname="col5">91</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M44" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 170</oasis:entry>
         <oasis:entry colname="col7"><inline-formula><mml:math id="M45" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 0.07</oasis:entry>
         <oasis:entry colname="col8">42</oasis:entry>
         <oasis:entry colname="col9">NASA</oasis:entry>
       </oasis:row>
     </oasis:tbody>
   </oasis:tgroup><?xmltex \end{scaleboxenv}?></oasis:table><?xmltex \begin{scaleboxenv}{.9}[.9]?><table-wrap-foot><p id="d1e815">S-GDR stands for Sensor Geophysical Data Record;
GDR stands for Geophysical Data Record;
GLAH 14 stands for GLAS/ICESat L2 Global Land Surface Altimetry Data
(HDF5), version 34;
CNES stands for Centre National d'Etudes Spatiales;
Aviso stands for Archiving, Validation and Interpretation of Satellite
Oceanographic data;
Aviso<inline-formula><mml:math id="M33" display="inline"><mml:mo>+</mml:mo></mml:math></inline-formula> data set is available via FTP at <uri>http://ftp-access.aviso.altimetry.fr</uri> with a registered username and password (last
access: 18 August 2019);
ESA Envisat products are available via FTP at <uri>http://ra2-ftp-ds.eo.esa.int</uri> with a registered username and password (last access: 18 August 2019);
ESA CryoSat-2 products are available via FTP at <uri>http://calval-pds.cryosat.esa.int</uri> with a registered username and password (last
access: 18 August 2019);
NASA ICESat products are available at <uri>https://nsidc.org/data/icesat/data.html</uri> (last access:
18 August 2019).</p></table-wrap-foot><?xmltex \end{scaleboxenv}?></table-wrap>

</sec>
<sec id="Ch1.S2.SS2">
  <label>2.2</label><title>Data</title>
      <p id="d1e1287">Multisource altimetry data were used in this study as shown in
Table 2. The earliest record dates back to 2002
(i.e., Envisat and Jason-1) and the latest record ends in 2017 (i.e.,
Jason-3 and CryoSat-2, Fig. 1). Most of the 52 lakes examined in this
study were covered by ICESat, Envisat, and CryoSat-2 data. ICESat data
provided by the National Aeronautics and Space Administration (NASA) were
available on 42 lakes in this study. Envisat and CryoSat-2 data provided by
the European Space Agency (ESA) were available on 35 and 51 lakes in this
study. Jason-1/2/3 data provided by the Centre National d'Etudes Spatiales
(CNES) were available only on 12 lakes in this study due to the relatively
sparse ground tracks or data quality issues. Note that Jason-2 inherited the
orbit of Jason-1 after its launch in 2008, whereas Jason-1 was shifted into
an interleaved orbit and continued functioning until 2013, thereby
increasing the spatial coverage of Jason altimetry series to some degree,
e.g.,<?pagebreak page1607?> Jason-1 data in Lake Qinghai, the largest lake on the TP, were only
available after 2008 due to the orbit shift. ICESat and CryoSat-2 data have
the largest spatial coverage but relatively long repeat cycles of 91
and 369 days, respectively (Bouzinac, 2012; Zhang et al., 2011).
The Envisat mission has a lower orbit than Jason-1/2/3 but higher than
ICESat, resulting in a moderate spatial coverage and a temporal resolution
of 35 days (Benveniste et al., 2002). To determine if the altimetry
data fall into the lakes, a lake shape data set generated by Wan et al. (2016) was used. An example of using the lake shape data set to determine
altimetry data falling into the lake boundaries is given in Fig. 3a,
showing that data from all altimeters are available in Zhari Namco.</p>
      <p id="d1e1290">It should be noted that different altimeters vary with wavelengths of
electromagnetic radiation and mechanisms. For instance, Jason-1/2/3 using
the Ku and C bands and Envisat/RA-2 using the Ku and S bands work in the low-resolution mode (LRM). These dual-frequency radar altimeters can provide
more accurate range corrections due to the ionospheric effect (Tournadre,
2004). The LRM is typical for the early version of satellite altimeters such
as TOPEX/Poseidon. There are more advanced modes, such as SAR and
interferometric synthetic aperture radar (InSAR), for recent radar
altimeters, which generally have smaller footprints than the LRM mode.
CryoSat-2/SIRAL working at a single Ku band has three modes, including LRM,
SAR, and InSAR, which were designed to have an increasing resolution in turn
and work in different zones. The InSAR mode uses interference phenomena so
that the shift of the nadir point across the track can be detected, improving
the altimeter's performance on ice sheets with slopes (Bouzinac,
2012). The Geoscience Laser Altimeter System (GLAS) is the laser altimeter
carried by ICESat working in the near-infrared band.</p>
      <p id="d1e1293">We used Landsat 5 TM (2000–2011), Landsat 7 ETM<inline-formula><mml:math id="M46" display="inline"><mml:mo>+</mml:mo></mml:math></inline-formula> (2000–2017), and Landsat
8 OLI (2013–2017) surface reflectance data sets provided by GEE to generate
information on lake shoreline positions and lake areas. Landsat 7 ETM<inline-formula><mml:math id="M47" display="inline"><mml:mo>+</mml:mo></mml:math></inline-formula> was
subject to sensor failure, and all the Landsat 7 ETM<inline-formula><mml:math id="M48" display="inline"><mml:mo>+</mml:mo></mml:math></inline-formula> images contain gaps
after 2003 (Markham et al., 2004). There were more than 20 000 images
processed, and half of them were excluded from the final results due to
cloud contamination or gaps. We collected daily in situ water level
measurements in Yamzhog Yumco for validation purposes with a pressure-type
water level sensor. The in situ water level measurements spanned half a year
from May to<?pagebreak page1608?> October 2018. We also performed unmanned aerial vehicle (UAV or
drone) imaging over a 1 km lake bank in Yamzhog Yumco and Nam Co for
obtaining better knowledge on the experimental environment.</p>
      <p id="d1e1317">In addition, GaoFen-2 (GF-2, the China High-Resolution Earth Observation
System mission) images were used to perform a rigorous statistical
analysis of uncertainty in the Landsat-derived water levels by taking the
GF-2-derived lake shoreline positions as the ground truth to analyze the
subpixel water area ratios of Landsat image pixels (see Sect. 4). GF-2
images have a spatial resolution of 0.8 m for the panchromatic band, and
preprocessing including orthorectification and radiometric calibration was
performed. Before analysis, we performed an image-to-image registration with
manually selected tie points between GF-2 and corresponding Landsat 8 OLI
images until the coregistration error reduced to <inline-formula><mml:math id="M49" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 2 m.</p>
</sec>
</sec>
<sec id="Ch1.S3">
  <label>3</label><title>Method</title>
<sec id="Ch1.S3.SS1">
  <label>3.1</label><title>Satellite altimetry water level</title>
      <p id="d1e1343">The first step of deriving satellite altimetry water levels is to select
correct ground tracks and valid footprints falling on the lakes. Because
there is a random ground track shift at <inline-formula><mml:math id="M50" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 1 km in different
cycles for most altimetry missions, it is uncertain whether valid lake
footprints can be obtained for each cycle, even though the nominal ground
track seems to cross the lake. This problem can be addressed by comparing
geographic coordinates of the footprints with a lake shape data set (Wan
et al., 2016). After picking out the valid footprints, the lake surface
height can be calculated for each footprint. All radar altimetry data share
a relationship:
            <disp-formula id="Ch1.E1" content-type="numbered"><label>1</label><mml:math id="M51" display="block"><mml:mrow><mml:mtext>LSH</mml:mtext><mml:mo>=</mml:mo><mml:mtext>Alt</mml:mtext><mml:mo>-</mml:mo><mml:mo>(</mml:mo><mml:mtext>Range</mml:mtext><mml:mo>+</mml:mo><mml:mtext>cor</mml:mtext><mml:mo>)</mml:mo><mml:mo>,</mml:mo></mml:mrow></mml:math></disp-formula>
          where LSH represents the lake surface height with respect to the geoid; Alt
represents the altimeter height with respect to the reference ellipsoid;
Range represents the distance between the altimeter and lake surface; and cor
represents several range corrections due to atmospheric effects, sensor
design defects, or geophysical effects. Radar altimeters and laser
altimeters need different corrections, given that they are working in
different wavelengths and have different designs. For instance, corrections
for radar altimeters include waveform retracking correction, wet/dry
troposphere correction, ionosphere correction, pole/solid tide correction,
and geoid correction. Laser altimeters also need atmospheric delay
correction, geoid/pole tide correction, and geoid correction. Unlike radar
altimeters, saturation correction instead of waveform retracking correction
is more important to laser altimeters.</p>
      <p id="d1e1379">The retracking correction plays an important role in removing the
contamination of land signal when radar altimetry data are applied to inland
water bodies. In this study, Jason-1/2/3 data were retracked using a
classical waveform retracking algorithm, i.e., the improved threshold method
(ITR), whereas CryoSat-2 data were retracked using the narrow primary peak
threshold (NTTP) method (Birkett and Beckley, 2010; Cheng et al., 2010;
Jain et al., 2015). Retracking corrections were not performed for Envisat
and ICESat data, because the Envisat/RA-2 product already provided a
retracked range using the ICE-1 method, and the ICESat GLAH14 product already
included several corrections (such as saturation correction) that are
sufficient for most applications including studies on inland water bodies
(Zhang et al., 2011). The original idea of the NTTP, ICE-1, and ITR is
quite similar. All of them adopt a threshold defined as the percentage of
the waveform peak to determine the retracking gate and then to convert the
difference between the retracking gate and the nominal gate into range
correction by multiplying the gate range (<inline-formula><mml:math id="M52" display="inline"><mml:mrow><mml:mi>c</mml:mi><mml:mi mathvariant="normal">Δ</mml:mi><mml:mi>t</mml:mi><mml:mo>/</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:mrow></mml:math></inline-formula>, where <inline-formula><mml:math id="M53" display="inline"><mml:mi>c</mml:mi></mml:math></inline-formula> is the speed of light and <inline-formula><mml:math id="M54" display="inline"><mml:mrow><mml:mi mathvariant="normal">Δ</mml:mi><mml:mi>t</mml:mi></mml:mrow></mml:math></inline-formula> is the time duration of a gate). The differences lie in the choice of thresholds as well as the calculation of waveform peaks. For instance, ICE-1 uses a 30 % threshold, whereas ITR uses a 50 % threshold.</p>

      <?xmltex \floatpos{t}?><fig id="Ch1.F3" specific-use="star"><?xmltex \currentcnt{3}?><label>Figure 3</label><caption><p id="d1e1417"><bold>(a)</bold> Ground tracks of multiple altimetric missions over Zhari Namco
and <bold>(b)</bold> the merged altimetry water levels for Zhari Namco. LSH stands for lake surface height.</p></caption>
          <?xmltex \igopts{width=455.244094pt}?><graphic xlink:href="https://essd.copernicus.org/articles/11/1603/2019/essd-11-1603-2019-f03.png"/>

        </fig>

      <p id="d1e1432">For each cycle of an altimeter, it is common that more than one footprint
fall on a lake, thereby providing several lake surface height (LSH)
observations on the same day. After removing outliers with the three-sigma
rule, frequency distributions of the LSHs from the same cycle were
calculated. The mean value of the histogram bin with the highest frequency
was selected to represent the LSH for the cycle. Meanwhile, the frequency of
the chosen histogram bin was reserved to evaluate the data quality for the
cycle; e.g., a cycle was marked as high quality if the frequency was higher
than 0.8, moderate quality if it was only higher than 0.5, and poor quality
if the frequency was lower than 0.5. The LSH from each cycle constituted the
water level time series for a lake. LSHs that were marked as poor quality
and obviously deviated from the moving average were removed from the
altimetry-based lake water level time series.</p>
      <p id="d1e1435">It is not uncommon that systematic biases exist in different altimetry data
sets due to variations in orbit, the discrepancy between correction models,
errors associated with sensors, and even the choice of the reference datum.
After deriving lake water level time series for each altimeter, we first
merged the Envisat and ICESat water levels if both were available for a
lake, because they have the longest overlap period (Fig. 1). We chose
Envisat-derived water levels as the baseline and removed the difference of
the mean values of the two products during the overlap period from the
ICESat data, because Envisat data are generally denser and longer than
ICESat data. A similar process was applied to Jason-1/2/3, as there are two
overlap periods connecting the three altimeters together. Figure 3b shows
a result of merged altimetry data when all sensors are available. There are
tradeoffs between CryoSat-2 and Jason-2/3 data in terms of spatial coverage
and time span. CryoSat-2 data are available for all lakes in this study but
they only have an overlap period with Jason-2/3 data, whereas Jason-2/3 data
are only available for<?pagebreak page1609?> 12 lakes. For most lakes without Jason-2/3 data, we
merged CryoSat-2 data with either ICESat or Envisat using Landsat-derived
water levels spanning from 2000 to 2017, because there is no overlap period
between these altimetry water levels (Fig. 1). Details on how
Landsat-derived water levels aid in merging the altimetry water levels are
shown in Sect. 3.2.</p>
</sec>
<sec id="Ch1.S3.SS2">
  <label>3.2</label><title>Landsat-derived water level</title>
      <p id="d1e1446">For most lake basins, it is possible to find a relatively flat portion of
lake banks with an average slope of <inline-formula><mml:math id="M55" display="inline"><mml:mrow><mml:mn mathvariant="normal">1</mml:mn><mml:mo>/</mml:mo><mml:mn mathvariant="normal">30</mml:mn></mml:mrow></mml:math></inline-formula> or even smaller, where obvious
interannual or intra-annual changes in lake shoreline positions can be
detected using Landsat images (30 m). These locations can be found by
comparing lake images from the first year and the last year of the study
period if the lake shows a clear expanding/shrinking trend. Otherwise, we
can compare images acquired in early summer when the LSH is at its lowest level with
those acquired in late autumn when the lake expands to its limit. In this
study, we assumed that the selected lake bank was flat enough such that the
relationship between the lake water level and shoreline position can be
depicted in a linear or quasilinear (parabolic) way. Thus, we can transform
the lake shoreline positions into Landsat-derived water levels by fitting
with altimetry water levels. The validity of this assumption can be
evaluated with the coefficient of determination (<inline-formula><mml:math id="M56" display="inline"><mml:mrow><mml:msup><mml:mi>R</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup></mml:mrow></mml:math></inline-formula>) for each lake as
shown in Table 3. For most lakes, the goodness of
fit is higher than 0.7, suggesting the generally good fitting relationship
between the lake water levels and shoreline positions.</p>

<?xmltex \floatpos{p}?><table-wrap id="Ch1.T3" specific-use="star"><?xmltex \currentcnt{3}?><label>Table 3</label><caption><p id="d1e1475">Summary of regression analyses and hypsometric function by lake.</p></caption><oasis:table frame="topbot"><oasis:tgroup cols="6">
     <oasis:colspec colnum="1" colname="col1" align="left"/>
     <oasis:colspec colnum="2" colname="col2" align="right"/>
     <oasis:colspec colnum="3" colname="col3" align="right"/>
     <oasis:colspec colnum="4" colname="col4" align="right"/>
     <oasis:colspec colnum="5" colname="col5" align="right"/>
     <oasis:colspec colnum="6" colname="col6" align="justify" colwidth="128.037402pt"/>
     <oasis:thead>
       <oasis:row>
         <oasis:entry colname="col1"/>
         <oasis:entry colname="col2">Lake</oasis:entry>
         <oasis:entry colname="col3">No. of</oasis:entry>
         <oasis:entry colname="col4"><inline-formula><mml:math id="M57" display="inline"><mml:mrow><mml:msup><mml:mi>R</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup></mml:mrow></mml:math></inline-formula> of Landsat-</oasis:entry>
         <oasis:entry colname="col5"><inline-formula><mml:math id="M58" display="inline"><mml:mrow><mml:msup><mml:mi>R</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup></mml:mrow></mml:math></inline-formula> of hypsometric</oasis:entry>
         <oasis:entry colname="col6"/>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Lake name</oasis:entry>
         <oasis:entry colname="col2">area</oasis:entry>
         <oasis:entry colname="col3">Landsat-derived</oasis:entry>
         <oasis:entry colname="col4">derived water level</oasis:entry>
         <oasis:entry colname="col5">curve</oasis:entry>
         <oasis:entry colname="col6">Hypsometric function</oasis:entry>
       </oasis:row>
       <oasis:row rowsep="1">
         <oasis:entry colname="col1"/>
         <oasis:entry colname="col2">(km<inline-formula><mml:math id="M59" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">2</mml:mn></mml:msup></mml:math></inline-formula>)</oasis:entry>
         <oasis:entry colname="col3">water level</oasis:entry>
         <oasis:entry colname="col4">(no. of data pairs)</oasis:entry>
         <oasis:entry colname="col5">(no. of data pairs)</oasis:entry>
         <oasis:entry colname="col6"/>
       </oasis:row>
     </oasis:thead>
     <oasis:tbody>
       <oasis:row>
         <oasis:entry colname="col1">Ake Sayi Lake</oasis:entry>
         <oasis:entry colname="col2">260.74</oasis:entry>
         <oasis:entry colname="col3">113</oasis:entry>
         <oasis:entry colname="col4">0.8951 (14)</oasis:entry>
         <oasis:entry colname="col5">0.9556 (21)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M60" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.45</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">11.26</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">163.97</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M61" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4846</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Lake Aqqikkol</oasis:entry>
         <oasis:entry colname="col2">538.21</oasis:entry>
         <oasis:entry colname="col3">354</oasis:entry>
         <oasis:entry colname="col4">0.9717 (44)</oasis:entry>
         <oasis:entry colname="col5">0.9353 (36)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M62" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">2.36</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">0.21</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">370.29</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M63" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4252</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Lake Ayakum</oasis:entry>
         <oasis:entry colname="col2">987.23</oasis:entry>
         <oasis:entry colname="col3">183</oasis:entry>
         <oasis:entry colname="col4">0.9651 (50)</oasis:entry>
         <oasis:entry colname="col5">0.9695 (57)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M64" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.16</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">65.72</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">658.28</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M65" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">3878</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Bam Co</oasis:entry>
         <oasis:entry colname="col2">255.81</oasis:entry>
         <oasis:entry colname="col3">209</oasis:entry>
         <oasis:entry colname="col4">0.901 (14)</oasis:entry>
         <oasis:entry colname="col5">0.9287 (27)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M66" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.28</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">2.84</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">206.46</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M67" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4560.5</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Bangong Co</oasis:entry>
         <oasis:entry colname="col2">661.64</oasis:entry>
         <oasis:entry colname="col3">232</oasis:entry>
         <oasis:entry colname="col4">0.5164 (172)</oasis:entry>
         <oasis:entry colname="col5">0.7991 (29)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M68" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">1.43</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">15.67</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">619.28</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M69" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4238</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Chibzhang Co</oasis:entry>
         <oasis:entry colname="col2">541.96</oasis:entry>
         <oasis:entry colname="col3">49</oasis:entry>
         <oasis:entry colname="col4">0.8766 (19)</oasis:entry>
         <oasis:entry colname="col5">0.9792 (15)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M70" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.69</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">3.36</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">475.79</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M71" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4930</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Co Ngoin1</oasis:entry>
         <oasis:entry colname="col2">268.37</oasis:entry>
         <oasis:entry colname="col3">174</oasis:entry>
         <oasis:entry colname="col4">0.6637 (15)</oasis:entry>
         <oasis:entry colname="col5">0.8803 (62)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M72" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">3.67</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mo>-</mml:mo><mml:mn mathvariant="normal">1.33</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">263.1</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M73" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4564.5</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Cuona Lake</oasis:entry>
         <oasis:entry colname="col2">192.15</oasis:entry>
         <oasis:entry colname="col3">254</oasis:entry>
         <oasis:entry colname="col4">0.7607 (12)</oasis:entry>
         <oasis:entry colname="col5">0.8876 (27)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M74" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">1.77</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">3.6</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">184.79</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M75" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4585.5</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Dagze Co</oasis:entry>
         <oasis:entry colname="col2">310.8</oasis:entry>
         <oasis:entry colname="col3">192</oasis:entry>
         <oasis:entry colname="col4">0.8334 (67)</oasis:entry>
         <oasis:entry colname="col5">0.8862 (28)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M76" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.08</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">6.14</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">230.51</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M77" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4460</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Dogai Coring</oasis:entry>
         <oasis:entry colname="col2">492.39</oasis:entry>
         <oasis:entry colname="col3">257</oasis:entry>
         <oasis:entry colname="col4">0.8624 (152)</oasis:entry>
         <oasis:entry colname="col5">0.9048 (33)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M78" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">3.2</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">5.66</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">427.17</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M79" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4816</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Dogaicoring Qangco</oasis:entry>
         <oasis:entry colname="col2">403.18</oasis:entry>
         <oasis:entry colname="col3">162</oasis:entry>
         <oasis:entry colname="col4">0.9202 (37)</oasis:entry>
         <oasis:entry colname="col5">0.9218 (47)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M80" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.53</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">3.93</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">279.6</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M81" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4786</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Donggei Cuona Lake</oasis:entry>
         <oasis:entry colname="col2">247.83</oasis:entry>
         <oasis:entry colname="col3">561</oasis:entry>
         <oasis:entry colname="col4">0.8776 (38)</oasis:entry>
         <oasis:entry colname="col5">0.925 (101)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M82" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.54</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">7.22</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">222.19</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M83" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4084</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Dung Co</oasis:entry>
         <oasis:entry colname="col2">139.4</oasis:entry>
         <oasis:entry colname="col3">145</oasis:entry>
         <oasis:entry colname="col4">0.9218 (49)</oasis:entry>
         <oasis:entry colname="col5">0.8652 (28)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M84" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.07</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">2.3</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">137.06</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M85" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4547</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Goren Co</oasis:entry>
         <oasis:entry colname="col2">477.95</oasis:entry>
         <oasis:entry colname="col3">191</oasis:entry>
         <oasis:entry colname="col4">0.6166 (24)</oasis:entry>
         <oasis:entry colname="col5">0.9096 (41)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M86" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">2.91</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mo>-</mml:mo><mml:mn mathvariant="normal">0.03</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">468.33</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M87" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4648.5</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Gozha Co</oasis:entry>
         <oasis:entry colname="col2">246.91</oasis:entry>
         <oasis:entry colname="col3">96</oasis:entry>
         <oasis:entry colname="col4">0.4297 (12)</oasis:entry>
         <oasis:entry colname="col5">0.5564 (19)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M88" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">1.57</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mo>-</mml:mo><mml:mn mathvariant="normal">0.06</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">254.43</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M89" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">5082</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Gyaring Lake</oasis:entry>
         <oasis:entry colname="col2">535.84</oasis:entry>
         <oasis:entry colname="col3">253</oasis:entry>
         <oasis:entry colname="col4">0.6217 (20)</oasis:entry>
         <oasis:entry colname="col5">0.3451 (71)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M90" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">1.99</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">2.8</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">517.18</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M91" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4292</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Har Lake</oasis:entry>
         <oasis:entry colname="col2">621.52</oasis:entry>
         <oasis:entry colname="col3">370</oasis:entry>
         <oasis:entry colname="col4">0.8652 (63)</oasis:entry>
         <oasis:entry colname="col5">0.9893 (50)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M92" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">1.1</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">1.52</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">582.34</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M93" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4075</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Hoh Xil Lake</oasis:entry>
         <oasis:entry colname="col2">351.3</oasis:entry>
         <oasis:entry colname="col3">132</oasis:entry>
         <oasis:entry colname="col4">0.9038 (12)</oasis:entry>
         <oasis:entry colname="col5">0.9355 (27)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M94" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">1</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">5.29</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">300.5</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M95" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4887.1</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Jingyu Lake</oasis:entry>
         <oasis:entry colname="col2">339.69</oasis:entry>
         <oasis:entry colname="col3">224</oasis:entry>
         <oasis:entry colname="col4">0.8978 (51)</oasis:entry>
         <oasis:entry colname="col5">0.989 (34)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M96" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.37</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">4.77</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">238.43</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M97" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4710</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Kusai Lake</oasis:entry>
         <oasis:entry colname="col2">328.8</oasis:entry>
         <oasis:entry colname="col3">295</oasis:entry>
         <oasis:entry colname="col4">0.9787 (151)</oasis:entry>
         <oasis:entry colname="col5">0.8987 (49)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M98" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.52</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">5.04</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">254.67</mml:mn><mml:mo>,</mml:mo></mml:mrow></mml:math></inline-formula> <inline-formula><mml:math id="M99" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4473</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Kyebxang Co</oasis:entry>
         <oasis:entry colname="col2">187.32</oasis:entry>
         <oasis:entry colname="col3">233</oasis:entry>
         <oasis:entry colname="col4">0.75 (12)</oasis:entry>
         <oasis:entry colname="col5">0.8753 (135)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M100" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.16</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">5.4</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">150.9</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M101" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4619</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Langa Co</oasis:entry>
         <oasis:entry colname="col2">256.03</oasis:entry>
         <oasis:entry colname="col3">167</oasis:entry>
         <oasis:entry colname="col4">0.859 (157)</oasis:entry>
         <oasis:entry colname="col5">0.888 (47)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M102" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mo>-</mml:mo><mml:mn mathvariant="normal">0.19</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">4</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">249.28</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M103" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4564</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Lexiewudan Co</oasis:entry>
         <oasis:entry colname="col2">273.63</oasis:entry>
         <oasis:entry colname="col3">286</oasis:entry>
         <oasis:entry colname="col4">0.9216 (49)</oasis:entry>
         <oasis:entry colname="col5">0.9496 (40)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M104" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.13</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">4.63</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">219.65</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M105" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4868</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Lumajiangdong Co</oasis:entry>
         <oasis:entry colname="col2">386.71</oasis:entry>
         <oasis:entry colname="col3">220</oasis:entry>
         <oasis:entry colname="col4">0.9135 (28)</oasis:entry>
         <oasis:entry colname="col5">0.9708 (17)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M106" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.62</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">2.09</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">353.95</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M107" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4812</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Mapam Yumco</oasis:entry>
         <oasis:entry colname="col2">412.69</oasis:entry>
         <oasis:entry colname="col3">163</oasis:entry>
         <oasis:entry colname="col4">0.7096 (30)</oasis:entry>
         <oasis:entry colname="col5">0.9973 (30)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M108" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">1.18</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">5.16</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">399.68</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M109" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4584</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Margai Caka</oasis:entry>
         <oasis:entry colname="col2">137.7</oasis:entry>
         <oasis:entry colname="col3">247</oasis:entry>
         <oasis:entry colname="col4">0.9399 (12)</oasis:entry>
         <oasis:entry colname="col5">0.9955 (35)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M110" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.03</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">5.14</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">112.12</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M111" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4791</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
     </oasis:tbody>
   </oasis:tgroup></oasis:table></table-wrap>

<?xmltex \floatpos{p}?><table-wrap id="Ch1.T4" specific-use="star"><?xmltex \currentcnt{3}?><label>Table 3</label><caption><p id="d1e3532">Continued.</p></caption><oasis:table frame="topbot"><oasis:tgroup cols="6">
     <oasis:colspec colnum="1" colname="col1" align="left"/>
     <oasis:colspec colnum="2" colname="col2" align="right"/>
     <oasis:colspec colnum="3" colname="col3" align="right"/>
     <oasis:colspec colnum="4" colname="col4" align="right"/>
     <oasis:colspec colnum="5" colname="col5" align="right"/>
     <oasis:colspec colnum="6" colname="col6" align="justify" colwidth="128.037402pt"/>
     <oasis:thead>
       <oasis:row>
         <oasis:entry colname="col1"/>
         <oasis:entry colname="col2">Lake</oasis:entry>
         <oasis:entry colname="col3">No. of</oasis:entry>
         <oasis:entry colname="col4"><inline-formula><mml:math id="M112" display="inline"><mml:mrow><mml:msup><mml:mi>R</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup></mml:mrow></mml:math></inline-formula> of Landsat-</oasis:entry>
         <oasis:entry colname="col5"><inline-formula><mml:math id="M113" display="inline"><mml:mrow><mml:msup><mml:mi>R</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup></mml:mrow></mml:math></inline-formula> of hypsometric</oasis:entry>
         <oasis:entry colname="col6"/>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Lake name</oasis:entry>
         <oasis:entry colname="col2">area</oasis:entry>
         <oasis:entry colname="col3">Landsat-derived</oasis:entry>
         <oasis:entry colname="col4">derived water level</oasis:entry>
         <oasis:entry colname="col5">curve</oasis:entry>
         <oasis:entry colname="col6">Hypsometric function</oasis:entry>
       </oasis:row>
       <oasis:row rowsep="1">
         <oasis:entry colname="col1"/>
         <oasis:entry colname="col2">(km<inline-formula><mml:math id="M114" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">2</mml:mn></mml:msup></mml:math></inline-formula>)</oasis:entry>
         <oasis:entry colname="col3">water level</oasis:entry>
         <oasis:entry colname="col4">(no. of data pairs)</oasis:entry>
         <oasis:entry colname="col5">(no. of data pairs)</oasis:entry>
         <oasis:entry colname="col6"/>
       </oasis:row>
     </oasis:thead>
     <oasis:tbody>
       <oasis:row>
         <oasis:entry colname="col1">Memar Co</oasis:entry>
         <oasis:entry colname="col2">167.3</oasis:entry>
         <oasis:entry colname="col3">193</oasis:entry>
         <oasis:entry colname="col4">0.911 (39)</oasis:entry>
         <oasis:entry colname="col5">0.8626 (20)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M115" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.27</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">3.17</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">134.69</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M116" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4923</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Nam Co</oasis:entry>
         <oasis:entry colname="col2">2028.5</oasis:entry>
         <oasis:entry colname="col3">187</oasis:entry>
         <oasis:entry colname="col4">0.9064 (62)</oasis:entry>
         <oasis:entry colname="col5">0.8749 (18)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M117" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">2.43</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">5.55</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">1970.1</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M118" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4724.5</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Ngangla Ringco</oasis:entry>
         <oasis:entry colname="col2">492.86</oasis:entry>
         <oasis:entry colname="col3">88</oasis:entry>
         <oasis:entry colname="col4">0.4652 (25)</oasis:entry>
         <oasis:entry colname="col5">0.9498 (7)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M119" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">3.87</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">3.86</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">490.69</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M120" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4715</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Ngangze Co</oasis:entry>
         <oasis:entry colname="col2">471.21</oasis:entry>
         <oasis:entry colname="col3">245</oasis:entry>
         <oasis:entry colname="col4">0.9538 (236)</oasis:entry>
         <oasis:entry colname="col5">0.9332 (49)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M121" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.2</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">7.03</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">391.21</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M122" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4680</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Ngoring Lake</oasis:entry>
         <oasis:entry colname="col2">656.83</oasis:entry>
         <oasis:entry colname="col3">86</oasis:entry>
         <oasis:entry colname="col4">0.844 (71)</oasis:entry>
         <oasis:entry colname="col5">0.8613 (14)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M123" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">4.69</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mo>-</mml:mo><mml:mn mathvariant="normal">5.04</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">613.66</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M124" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4270</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Paiku Co</oasis:entry>
         <oasis:entry colname="col2">272.85</oasis:entry>
         <oasis:entry colname="col3">231</oasis:entry>
         <oasis:entry colname="col4">0.8341 (21)</oasis:entry>
         <oasis:entry colname="col5">0.9079 (61)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M125" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.91</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">2.64</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">264.89</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M126" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4578.5</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Puma Yumco</oasis:entry>
         <oasis:entry colname="col2">290.98</oasis:entry>
         <oasis:entry colname="col3">250</oasis:entry>
         <oasis:entry colname="col4">0.6871 (18)</oasis:entry>
         <oasis:entry colname="col5">0.5629 (53)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M127" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.48</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">0.8</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">286.34</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M128" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">5011</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Pung Co</oasis:entry>
         <oasis:entry colname="col2">176.93</oasis:entry>
         <oasis:entry colname="col3">187</oasis:entry>
         <oasis:entry colname="col4">0.8017 (12)</oasis:entry>
         <oasis:entry colname="col5">0.9841 (31)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M129" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.03</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">3.75</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">151.66</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M130" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4526</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Qinghai Lake</oasis:entry>
         <oasis:entry colname="col2">4495.33</oasis:entry>
         <oasis:entry colname="col3">323</oasis:entry>
         <oasis:entry colname="col4">0.9011 (151)</oasis:entry>
         <oasis:entry colname="col5">0.8181 (19)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M131" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">3.45</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">155.03</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">4084.73</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M132" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">3193</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Rola Co</oasis:entry>
         <oasis:entry colname="col2">169.83</oasis:entry>
         <oasis:entry colname="col3">347</oasis:entry>
         <oasis:entry colname="col4">0.7842 (18)</oasis:entry>
         <oasis:entry colname="col5">0.9403 (96)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M133" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mo>-</mml:mo><mml:mn mathvariant="normal">0.88</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">14.87</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">115.59</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M134" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4816</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Salt Lake</oasis:entry>
         <oasis:entry colname="col2">144.3</oasis:entry>
         <oasis:entry colname="col3">206</oasis:entry>
         <oasis:entry colname="col4">0.9344 (16)</oasis:entry>
         <oasis:entry colname="col5">0.9858 (32)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M135" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.16</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mo>-</mml:mo><mml:mn mathvariant="normal">0.69</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">37.42</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M136" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4430</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Salt Water Lake</oasis:entry>
         <oasis:entry colname="col2">212.47</oasis:entry>
         <oasis:entry colname="col3">347</oasis:entry>
         <oasis:entry colname="col4">0.9086 (27)</oasis:entry>
         <oasis:entry colname="col5">0.9494 (151)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M137" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mo>-</mml:mo><mml:mn mathvariant="normal">0.82</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">17.31</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">133.71</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M138" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4901</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Selin Co</oasis:entry>
         <oasis:entry colname="col2">2300.49</oasis:entry>
         <oasis:entry colname="col3">179</oasis:entry>
         <oasis:entry colname="col4">0.9777 (70)</oasis:entry>
         <oasis:entry colname="col5">0.945 (22)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M139" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">1.05</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">45.86</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">1754.31</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M140" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4536.4</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Tangra Yumco</oasis:entry>
         <oasis:entry colname="col2">862.94</oasis:entry>
         <oasis:entry colname="col3">100</oasis:entry>
         <oasis:entry colname="col4">0.9155 (18)</oasis:entry>
         <oasis:entry colname="col5">0.8072 (11)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M141" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.94</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mo>-</mml:mo><mml:mn mathvariant="normal">0.28</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">862.94</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M142" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4536</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Taro Co</oasis:entry>
         <oasis:entry colname="col2">485.15</oasis:entry>
         <oasis:entry colname="col3">268</oasis:entry>
         <oasis:entry colname="col4">0.8903 (39)</oasis:entry>
         <oasis:entry colname="col5">0.9576 (19)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M143" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.18</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">4.97</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">477.32</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M144" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4567.3</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Tu Co</oasis:entry>
         <oasis:entry colname="col2">448.64</oasis:entry>
         <oasis:entry colname="col3">257</oasis:entry>
         <oasis:entry colname="col4">0.9276 (41)</oasis:entry>
         <oasis:entry colname="col5">0.9875 (26)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M145" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.02</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">4.91</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">396.59</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M146" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4926</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Urru Co</oasis:entry>
         <oasis:entry colname="col2">356.35</oasis:entry>
         <oasis:entry colname="col3">260</oasis:entry>
         <oasis:entry colname="col4">0.71 (49)</oasis:entry>
         <oasis:entry colname="col5">0.8994 (27)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M147" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">1.35</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">2.67</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">345.34</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M148" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4553</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Wulanwula Lake</oasis:entry>
         <oasis:entry colname="col2">652.08</oasis:entry>
         <oasis:entry colname="col3">225</oasis:entry>
         <oasis:entry colname="col4">0.9679 (81)</oasis:entry>
         <oasis:entry colname="col5">0.9285 (10)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M149" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">2.05</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">16.49</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">513.15</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M150" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4856</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Xijir Ulan Lake</oasis:entry>
         <oasis:entry colname="col2">463.36</oasis:entry>
         <oasis:entry colname="col3">316</oasis:entry>
         <oasis:entry colname="col4">0.9736 (84)</oasis:entry>
         <oasis:entry colname="col5">0.9691 (44)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M151" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.93</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">13.3</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">366.35</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M152" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4770.8</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Xuru Co</oasis:entry>
         <oasis:entry colname="col2">209.87</oasis:entry>
         <oasis:entry colname="col3">144</oasis:entry>
         <oasis:entry colname="col4">0.5984 (9)</oasis:entry>
         <oasis:entry colname="col5">0.5527 (58)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M153" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.12</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">0.22</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">206.53</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M154" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4714</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Yamzho Yumco</oasis:entry>
         <oasis:entry colname="col2">549.61</oasis:entry>
         <oasis:entry colname="col3">398</oasis:entry>
         <oasis:entry colname="col4">0.9215 (140)</oasis:entry>
         <oasis:entry colname="col5">0.9364 (72)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M155" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.51</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">9.63</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">531.79</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M156" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4436</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Yelusu Lake</oasis:entry>
         <oasis:entry colname="col2">203.4</oasis:entry>
         <oasis:entry colname="col3">486</oasis:entry>
         <oasis:entry colname="col4">0.7014 (21)</oasis:entry>
         <oasis:entry colname="col5">0.8352 (92)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M157" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">14.84</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mo>-</mml:mo><mml:mn mathvariant="normal">5.77</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">185.15</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M158" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4686.5</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Yibug Caka</oasis:entry>
         <oasis:entry colname="col2">178.22</oasis:entry>
         <oasis:entry colname="col3">118</oasis:entry>
         <oasis:entry colname="col4">0.9206 (12)</oasis:entry>
         <oasis:entry colname="col5">0.9615 (25)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M159" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mo>-</mml:mo><mml:mn mathvariant="normal">1.25</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">15.79</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">147.03</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M160" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4558.5</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Zhari Namco</oasis:entry>
         <oasis:entry colname="col2">1000.18</oasis:entry>
         <oasis:entry colname="col3">143</oasis:entry>
         <oasis:entry colname="col4">0.9177 (164)</oasis:entry>
         <oasis:entry colname="col5">0.8388 (52)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M161" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">2.66</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">10.07</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">962.57</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M162" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4610</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Zhuonai Lake</oasis:entry>
         <oasis:entry colname="col2">160.1</oasis:entry>
         <oasis:entry colname="col3">260</oasis:entry>
         <oasis:entry colname="col4">0.9528 (11)</oasis:entry>
         <oasis:entry colname="col5">0.973 (21)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M163" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">10.06</mml:mn><mml:mspace linebreak="nobreak" width="0.125em"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">124.29</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M164" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4742</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
       <oasis:row>
         <oasis:entry colname="col1">Zige Tangco</oasis:entry>
         <oasis:entry colname="col2">238.67</oasis:entry>
         <oasis:entry colname="col3">171</oasis:entry>
         <oasis:entry colname="col4">0.9008 (186)</oasis:entry>
         <oasis:entry colname="col5">0.976 (24)</oasis:entry>
         <oasis:entry colname="col6"><inline-formula><mml:math id="M165" display="inline"><mml:mrow><mml:mi>S</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.06</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:msup><mml:mi>h</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:mn mathvariant="normal">4.62</mml:mn><mml:mspace width="0.125em" linebreak="nobreak"/><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>+</mml:mo><mml:mn mathvariant="normal">212.71</mml:mn></mml:mrow></mml:math></inline-formula>,<?xmltex \hack{\hfill\break}?> <inline-formula><mml:math id="M166" display="inline"><mml:mrow><mml:mtext>d</mml:mtext><mml:mi>h</mml:mi><mml:mo>=</mml:mo><mml:mi>H</mml:mi><mml:mo>-</mml:mo><mml:mn mathvariant="normal">4565</mml:mn></mml:mrow></mml:math></inline-formula></oasis:entry>
       </oasis:row>
     </oasis:tbody>
   </oasis:tgroup></oasis:table></table-wrap>

      <p id="d1e5593">Though there were <inline-formula><mml:math id="M167" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 500 Landsat images obtained for the
selected lake banks during the study period, many of them were largely
affected by cloud or cloud shadow. All the images were processed online
using the GEE application programming interface (API). Preprocessing such as
radiometric calibration, atmospheric correction, and geometric
correction was already performed in the production of the data sets. In
addition, the failure of the Landsat 7 sensor SLC left all the Landsat 7 ETM<inline-formula><mml:math id="M168" display="inline"><mml:mo>+</mml:mo></mml:math></inline-formula>
images with gaps after 2003 (Markham et al., 2004), making the available
images even fewer. We managed to make use of some images with gaps in
generating lake shoreline positions. By choosing the region of interest
(ROI) that is parallel to the image gaps, we made most of the Landsat 7
ETM<inline-formula><mml:math id="M169" display="inline"><mml:mo>+</mml:mo></mml:math></inline-formula> images useable. However, the width of ROI must be reduced to avoid
shifting gaps as shown in Fig. 4b. The gaps may vary with time but are more like vibration around the midpoint. The ROI did not fill the interval of gaps, because the wider the ROI, the higher possibility of shifting gaps cross it.</p>
      <p id="d1e5617">Lake shoreline positions were characterized by water area ratios detected in
the ROI. To automatically extract water areas from a mass of Landsat
archives on GEE, the water index and Otsu threshold method were jointly
used. We calculated the normalized difference water index (NDWI) and the
modified normalized difference water index (MNDWI) of the images and
compared their performance in different seasons. It was found that the MNDWI
tends to be more sensitive to shallow water in summer but is less effective
than the NDWI when the lake bank is covered by snow in the cold season as
shown in Fig. 5c. Therefore, the two water indices were
jointly used by applying the MNDWI to images acquired during May to October
and applying the NDWI to the remaining months. The NDWI and MNDWI can be
calculated as follows (McFeeters, 1996; Xu, 2005):

                <disp-formula specific-use="align" content-type="numbered"><mml:math id="M170" display="block"><mml:mtable displaystyle="true"><mml:mlabeledtr id="Ch1.E2"><mml:mtd><mml:mtext>2</mml:mtext></mml:mtd><mml:mtd><mml:mstyle class="stylechange" displaystyle="true"/></mml:mtd><mml:mtd><mml:mrow><mml:mstyle class="stylechange" displaystyle="true"/><mml:mtext>NDWI</mml:mtext><mml:mo>=</mml:mo><mml:mstyle displaystyle="true"><mml:mfrac style="display"><mml:mrow><mml:msub><mml:mi>B</mml:mi><mml:mtext>green</mml:mtext></mml:msub><mml:mo>-</mml:mo><mml:msub><mml:mi>B</mml:mi><mml:mtext>NIR</mml:mtext></mml:msub></mml:mrow><mml:mrow><mml:msub><mml:mi>B</mml:mi><mml:mtext>green</mml:mtext></mml:msub><mml:mo>+</mml:mo><mml:msub><mml:mi>B</mml:mi><mml:mtext>NIR</mml:mtext></mml:msub></mml:mrow></mml:mfrac></mml:mstyle><mml:mo>,</mml:mo></mml:mrow></mml:mtd></mml:mlabeledtr><mml:mlabeledtr id="Ch1.E3"><mml:mtd><mml:mtext>3</mml:mtext></mml:mtd><mml:mtd><mml:mstyle class="stylechange" displaystyle="true"/></mml:mtd><mml:mtd><mml:mrow><mml:mstyle displaystyle="true" class="stylechange"/><mml:mtext>MNDWI</mml:mtext><mml:mo>=</mml:mo><mml:mstyle displaystyle="true"><mml:mfrac style="display"><mml:mrow><mml:msub><mml:mi>B</mml:mi><mml:mtext>green</mml:mtext></mml:msub><mml:mo>-</mml:mo><mml:msub><mml:mi>B</mml:mi><mml:mtext>SIR</mml:mtext></mml:msub></mml:mrow><mml:mrow><mml:msub><mml:mi>B</mml:mi><mml:mtext>green</mml:mtext></mml:msub><mml:mo>+</mml:mo><mml:msub><mml:mi>B</mml:mi><mml:mtext>SIR</mml:mtext></mml:msub></mml:mrow></mml:mfrac></mml:mstyle><mml:mo>,</mml:mo></mml:mrow></mml:mtd></mml:mlabeledtr></mml:mtable></mml:math></disp-formula>

            where <inline-formula><mml:math id="M171" display="inline"><mml:mrow><mml:msub><mml:mi>B</mml:mi><mml:mtext>green</mml:mtext></mml:msub></mml:mrow></mml:math></inline-formula>, <inline-formula><mml:math id="M172" display="inline"><mml:mrow><mml:msub><mml:mi>B</mml:mi><mml:mtext>NIR</mml:mtext></mml:msub></mml:mrow></mml:math></inline-formula>, and <inline-formula><mml:math id="M173" display="inline"><mml:mrow><mml:msub><mml:mi>B</mml:mi><mml:mtext>SIR</mml:mtext></mml:msub></mml:mrow></mml:math></inline-formula> refer to surface reflectance of bands 2, 4, and 5 for Landsat 5/7 TM/ETM<inline-formula><mml:math id="M174" display="inline"><mml:mo>+</mml:mo></mml:math></inline-formula> images and bands 3, 5, and 6 for
Landsat 8 OLI images.</p>

      <?xmltex \floatpos{t}?><fig id="Ch1.F4" specific-use="star"><?xmltex \currentcnt{4}?><label>Figure 4</label><caption><p id="d1e5754"><bold>(a)</bold> Yamzhog Yumco and its surroundings. The DEM was extracted from
the SRTM Global 90 m DEM product (Jarvis et al., 2008). <bold>(b)</bold> Region of
interest (ROI, yellow area) selected from a Landsat 7 ETM<inline-formula><mml:math id="M175" display="inline"><mml:mo>+</mml:mo></mml:math></inline-formula> image for
detecting changes in lake shoreline (black areas represent gaps in the
image). <bold>(c)</bold> Linear regression of lake shoreline positions that are
represented by water area ratios in the ROI and altimetry water levels for
Yamzhog Yumco. <bold>(d)</bold> Landsat-derived water levels and altimetry water
levels for Yamzhog Yumco.</p></caption>
          <?xmltex \igopts{width=426.791339pt}?><graphic xlink:href="https://essd.copernicus.org/articles/11/1603/2019/essd-11-1603-2019-f04.png"/>

        </fig>

      <?xmltex \floatpos{t}?><fig id="Ch1.F5" specific-use="star"><?xmltex \currentcnt{5}?><label>Figure 5</label><caption><p id="d1e5783"><bold>(a)</bold> A Landsat 7 ETM<inline-formula><mml:math id="M176" display="inline"><mml:mo>+</mml:mo></mml:math></inline-formula> image of Lake Aqqikkol acquired in
summer in 2001; <bold>(b)</bold> water area extractions using the modified normalized
difference water index (MNDWI) and the normalized difference water index
(NDWI), showing that the MNDWI performs better in detecting shallow water;
<bold>(c)</bold> a Landsat 8 OLI image of Nam Co acquired in winter in 2015; <bold>(d)</bold> water
area extraction using the NDWI, showing good performance in distinguishing
water from snow; and <bold>(e)</bold> water area extraction using the MNDWI, showing some confusion of water and snow.</p></caption>
          <?xmltex \igopts{width=398.338583pt}?><graphic xlink:href="https://essd.copernicus.org/articles/11/1603/2019/essd-11-1603-2019-f05.png"/>

        </fig>

      <?pagebreak page1612?><p id="d1e5813">After calculating the water index, the grayscale image was binarized using
the Otsu method. If the selected ROI comprises <inline-formula><mml:math id="M177" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 50 % water
and <inline-formula><mml:math id="M178" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 50 % land, the performance of the method is good, as
the distribution of digital numbers of the grayscale image is close to the
assumption of the bimodal histogram implicit in the Otsu algorithm
(Kittler and Illingworth, 1985; Otsu, 1979). The binarized images were
further processed to provide the water area ratio in the ROI, which
represents the lake shoreline position. The lake shoreline position time
series were then converted into Landsat-derived water levels using linear
regression or second-order polynomial fit with altimetry-derived water
levels (Fig. 4c–d). For most cases, we only used linear
regression, and we performed the
second-order polynomial fit only for 2 lakes with Jason-1/2/3 data, because a higher-order regression requires more
input information to ensure the reliability. However, cloud, cloud shadow,
and shifting gaps may contaminate the ROI and cause errors in the
Landsat-derived water levels. Therefore, the QA band of the Landsat surface
reflectance product was used to filter the images. Data were excluded if the
fraction of the cloud or cloud-shadow-covered area in the ROI was higher
than 5 %. For every Landsat 7 ETM<inline-formula><mml:math id="M179" display="inline"><mml:mo>+</mml:mo></mml:math></inline-formula> image acquired after 2003, the pixel
number of the ROI was counted and compared with those acquired before 2003.
If the loss of pixels exceeded 2 %, the ROI was considered to be affected
by a gap and the data were consequently excluded from the subsequent analysis.</p>
      <p id="d1e5838">A critical function of Landsat-derived water levels is to aid in merging
altimetry water levels when there was no overlap period between altimeters
or the overlap period was too short. For lakes without Jason-1/2/3 data,
lake shoreline positions were firstly translated into Landsat-derived water
levels by fitting with CryoSat-2 data functioning as extrapolation of
CryoSat-2 to 1–2 years. Then, we applied the same method of merging
Jason-1/2/3 to merge the extrapolated CryoSat-2 data with either Envisat or
ICESat data. In doing so, we were able to remove all systematic biases
between multisource altimetry water levels. After merging the altimetry
water<?pagebreak page1613?> levels, we performed regression analysis for the second time between
the Landsat-derived water levels and merged altimetry water levels to check
if the linear relationship is stable during the entire study period and at
different elevations. If the linear relationship was stable, the
Landsat-derived water levels would be merged with the altimetry water levels
using the linear fitting parameters from the second regression analysis.
Otherwise, there may have been a change in the lake bank slope, and, therefore, the extrapolation of CryoSat-2 data with Landsat-derived water levels was
not suited. In this case, we reselected the ROI to extract lake shoreline
positions and redid altimetry data merging until the Landsat-derived water
levels and merged altimetry water levels agreed well with one another in the
second linear regression. Detailed analysis about the potential
extrapolation issue can be found in the Supplement.</p>
      <p id="d1e5841">In summary, the basic idea of removing systematic biases of different
altimetry water levels is to calculate the means of two altimetry water
level time series during the overlap period. The difference between the
means is removed from one altimetry water level time series to make both
altimetry water level time series consistent and to form a longer time
series. This process was subsequently applied to all water level time series
with overlap periods to merge them into a single time series for each lake.
However, the overlap period may not be long enough, such as Envisat and
CryoSat-2 (e.g., there are limited data points (e.g., 1–2) during the
overlap period), or does not exist at all, such as ICESat and CryoSat-2. On
these cases, Landsat-derived water levels are used to extend or create an
overlap period that links two altimetry water level time series.</p>

      <?xmltex \floatpos{t}?><fig id="Ch1.F6" specific-use="star"><?xmltex \currentcnt{6}?><label>Figure 6</label><caption><p id="d1e5846">Programming interface of Google Earth Engine©. The red
polygon is the region of interest for lake area change extraction of Selin
Co.</p></caption>
          <?xmltex \igopts{width=497.923228pt}?><graphic xlink:href="https://essd.copernicus.org/articles/11/1603/2019/essd-11-1603-2019-f06.png"/>

        </fig>

</sec>
<sec id="Ch1.S3.SS3">
  <label>3.3</label><title>Hypsometry</title>
      <p id="d1e5863">We derived the hypsometric curve for each lake by polynomial fitting of the
lake area and level time series. The lake area comprises two parts: the
inner invariable part and the outer variable part. As the variable water
area was of more concern in this study, ROIs for extracting changes in lake
area only cover the lake shoreline and its neighboring areas as shown in
Fig. 6. The inner part of the water body was<?pagebreak page1614?> calculated only once and
considered invariant, making the calculation more efficient on GEE.
Meanwhile, more images are available as the area of ROI becomes smaller,
because the possibility of clouds covering the ROI is reduced compared with
an ROI covering the entire lake. Landsat 7 ETM<inline-formula><mml:math id="M180" display="inline"><mml:mo>+</mml:mo></mml:math></inline-formula> images after 2003 were not
included in this part of the calculation as gaps negatively affected the ROI for
lake area extraction. Similar to Sect. 3.2, we selected images with less
than 5 % cloud cover on an ROI to generate time series of changes in lake
area, obtaining 20–30 data points on average for regression. <inline-formula><mml:math id="M181" display="inline"><mml:mrow><mml:msup><mml:mi>R</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup></mml:mrow></mml:math></inline-formula> values
for each lake are listed in Table 3, indicating that most lake basins agree well with the parabolic hypsometric
curve.</p>

      <?xmltex \floatpos{t}?><fig id="Ch1.F7" specific-use="star"><?xmltex \currentcnt{7}?><label>Figure 7</label><caption><p id="d1e5886">Field experiments in two typical lakes: <bold>(a)</bold> an overview map of the
experiment spot; <bold>(b)</bold> pressure-type water level sensor; <bold>(c)</bold> unmanned aerial
vehicle (UAV); <bold>(d)</bold> installation of the water level sensor; and <bold>(e)</bold> UAV image
of a portion of the bank of Nam Co.</p></caption>
          <?xmltex \igopts{width=341.433071pt}?><graphic xlink:href="https://essd.copernicus.org/articles/11/1603/2019/essd-11-1603-2019-f07.png"/>

        </fig>

</sec>
</sec>
<sec id="Ch1.S4">
  <label>4</label><title>Validation of data quality</title>
<sec id="Ch1.S4.SS1">
  <label>4.1</label><title>Field experiment</title>
      <p id="d1e5926">Most Tibetan lakes are located in remote and inaccessible regions, resulting
in the scarcity of ground-based in situ measurements that are
vital for data quality assessment. We made some in situ measurements in two
lakes to validate the data quality of Landsat-derived water levels. The data
quality of satellite altimetry on lakes or rivers has been widely
investigated, and thus it is beyond the scope of our study. Many studies used
in situ water levels to calculate statistical metrics, e.g., root mean
squired error (RMSE). However, results provided by different studies vary,
which could be associated with the cross-section width of the study water
body in the ground track panel (Nielsen et al., 2017). This means that
these results may not be comparable due to their unique applications. In
addition, it is not rigorous to use in situ data of only one lake to
represent the overall performance in the uncertainty assessment for
altimetry water levels. Instead, we used the standard deviation of valid
footprints acquired in the same cycle as an estimate of uncertainty in
satellite altimetry water levels. In contrast, the applicable condition of
Landsat-derived water levels is not so variable as that of satellite
altimetry data. Derivation of Landsat-derived water levels requires
a relatively flat bank as well as some altimetric information, which were
available in all lakes. Since these selected bank slopes were similarly
small (<inline-formula><mml:math id="M182" display="inline"><mml:mo lspace="0mm">∼</mml:mo></mml:math></inline-formula> <inline-formula><mml:math id="M183" display="inline"><mml:mrow><mml:mn mathvariant="normal">1</mml:mn><mml:mo>/</mml:mo><mml:mn mathvariant="normal">30</mml:mn></mml:mrow></mml:math></inline-formula>), it was possible to use a few typical lakes to
represent all lakes. Therefore, we carried out a field experiment (Fig. 7) in Yamzhog
Yumco and Nam Co to validate the Landsat-derived water levels.</p>
      <p id="d1e5948">There were two main goals in our experiments: (1) collecting daily in situ
water level data in a TP lake to validate the corresponding Landsat-derived
water levels statistically and (2) testing the performance of extracting lake
shoreline positions from high-resolution optical images (GF-2) to provide a
theoretical uncertainty analysis of Landsat-derived water levels. On Yamzhog
Yumco, we installed a pressure-type water level sensor (type H5110-DY,
manufactured by<?pagebreak page1615?> Shenzhen Hongdian Technologies Co., Ltd.), which measured
water pressure and temperature at the installation depth that were converted
into water depths with a relative accuracy of <inline-formula><mml:math id="M184" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 0.1 %. The
device was carried onto the lake and put <inline-formula><mml:math id="M185" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 20 m below the
water surface and 0.5 m above the lake bottom, suggesting an absolute error
of <inline-formula><mml:math id="M186" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 2 cm. As for GF-2 images, the spatial resolution of the
panchromatic band is 0.8 m, which is able to provide a very accurate
reference of lake boundaries for assessing water classification results for
Landsat images. We used three GF-2 images acquired at different seasons (two
in July and September 2015 and one in February 2016) and different places on the TP to
better represent the local conditions when extracting Landsat-derived water
levels or areas. Image coregistration was performed to make sure that there
was no obvious spatial shift between the GF-2 images and corresponding
Landsat images. The accuracy of the image coregistration was
<inline-formula><mml:math id="M187" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 2 m.</p>
</sec>
<sec id="Ch1.S4.SS2">
  <label>4.2</label><title>Uncertainty analysis of Landsat-derived water levels</title>
      <p id="d1e5987">Based on the in situ water level measurements made by the pressure-type
water level sensor, we evaluated the accuracy of Landsat-derived water
levels statistically. We first calculated anomalies of in situ water levels
and Landsat-derived water levels, and then water levels from the two sources
acquired on the same day were used for analysis. There were 16
Landsat-derived water level records available for the comparison against the
in situ measurements, indicating an RMSE of the water level anomaly of 0.11 m. The linear fit shows a slope close to 1 and an <inline-formula><mml:math id="M188" display="inline"><mml:mrow><mml:msup><mml:mi>R</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup></mml:mrow></mml:math></inline-formula> of 0.89, suggesting
the consistency between the in situ water level measurements and the
Landsat-derived water levels (Fig. 8b). It should be noted that the
Landsat-derived water levels used for validation here were translated from
lake shoreline positions using parameters derived from fitting with
CryoSat-2 data, i.e., there is no in situ information involved in generating
the Landsat-derived water levels shown in Fig. 8.</p>

      <?xmltex \floatpos{t}?><fig id="Ch1.F8" specific-use="star"><?xmltex \currentcnt{8}?><label>Figure 8</label><caption><p id="d1e6003"><bold>(a)</bold> In situ water level anomaly versus Landsat-derived water level
anomaly in Yamzhog Yumco; <bold>(b)</bold> linear regression between the
Landsat-derived water levels and in situ water levels during the same
period.</p></caption>
          <?xmltex \igopts{width=398.338583pt}?><graphic xlink:href="https://essd.copernicus.org/articles/11/1603/2019/essd-11-1603-2019-f08.png"/>

        </fig>

      <p id="d1e6017">Furthermore, we performed a theoretical uncertainty analysis of the
Landsat-derived water levels by looking at the original optical data and the
generation process with the help of high-resolution GF-2 images. First, we
took GF-2 images (after coregistration with the Landsat image for the same
period and after the coregistration errors were <inline-formula><mml:math id="M189" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 2 m) as the
ground truth to determine the accurate position and shape of the lake
shoreline. Second, we performed water classification from the Landsat 8 OLI
image for the same period jointly using the water index method and Otsu
algorithm to derive the binarized image. Landsat image pixels where the lake
shorelines from the GF-2 images cross were delineated and marked as
shoreline pixels as shown in Fig. 9a. Then the water area in each
shoreline pixel was calculated.</p>

      <?xmltex \floatpos{t}?><fig id="Ch1.F9" specific-use="star"><?xmltex \currentcnt{9}?><label>Figure 9</label><caption><p id="d1e6030"><bold>(a–c)</bold> GF-2 images (upper layer) and corresponding Landsat 8 OLI images (bottom layer) acquired on 7 September 2015, 29 January 2016, and 5 July 2015; <bold>(d)</bold> Landsat 8 OLI shoreline pixels (the background is the GF-2 image) – blue pixels were classified as water, and yellow pixels were classified as
land; <bold>(e)</bold> the relationship between the water area ratio in a pixel and the frequency of the pixel being classified as water. Blue bars are sampled at a 0.04 bin space from the 4128 pixels. The red line shows the fitting curve based on the maximum likelihood method.</p></caption>
          <?xmltex \igopts{width=398.338583pt}?><graphic xlink:href="https://essd.copernicus.org/articles/11/1603/2019/essd-11-1603-2019-f09.png"/>

        </fig>

      <p id="d1e6047">Given that these shoreline pixels were classified as either water or land, a
relationship between the water area ratio of the shoreline pixel and the
probability of the pixel being classified as water can be derived. This
relationship generally describes the function of the water classification
method by telling how likely a pixel is to be determined as water, given the
water area ratio of the pixel. Based on the observations of a total of 4128
Landsat shoreline pixels, a power function<?pagebreak page1616?> was chosen to represent the water
classifier as Eq. (4) shows:
            <disp-formula id="Ch1.E4" content-type="numbered"><label>4</label><mml:math id="M190" display="block"><mml:mrow><mml:mi>f</mml:mi><mml:mo>(</mml:mo><mml:mi>x</mml:mi><mml:mo>)</mml:mo><mml:mo>=</mml:mo><mml:msup><mml:mi>x</mml:mi><mml:mi>n</mml:mi></mml:msup><mml:mo>,</mml:mo></mml:mrow></mml:math></disp-formula>
          where <inline-formula><mml:math id="M191" display="inline"><mml:mi>x</mml:mi></mml:math></inline-formula> represents the water area ratio in the shoreline pixel, <inline-formula><mml:math id="M192" display="inline"><mml:mrow><mml:mi>f</mml:mi><mml:mo>(</mml:mo><mml:mi>x</mml:mi><mml:mo>)</mml:mo></mml:mrow></mml:math></inline-formula> represents the probability of the shoreline pixel being classified as the water pixel, and <inline-formula><mml:math id="M193" display="inline"><mml:mi>n</mml:mi></mml:math></inline-formula> is the parameter that determines the shape of the curve. Parameter <inline-formula><mml:math id="M194" display="inline"><mml:mi>n</mml:mi></mml:math></inline-formula> was determined using the maximum likelihood method
(Fig. 9e).</p>
      <p id="d1e6108">As expected, the probability of the pixel being classified as water increases with
the water area ratio in the pixel (Fig. 9e). The enclosed area of the
fitting curve (<inline-formula><mml:math id="M195" display="inline"><mml:mrow><mml:mi>y</mml:mi><mml:mo>=</mml:mo><mml:msup><mml:mi>x</mml:mi><mml:mn mathvariant="normal">1.43</mml:mn></mml:msup></mml:mrow></mml:math></inline-formula>) is smaller than that of <inline-formula><mml:math id="M196" display="inline"><mml:mrow><mml:mi>y</mml:mi><mml:mo>=</mml:mo><mml:mi>x</mml:mi></mml:mrow></mml:math></inline-formula> on [0, 1],
suggesting that there may be a lower probability of the occurrence of water
pixels that is associated with a systematic bias of the lake shoreline
detection. Note that the systematic bias can be removed when linearly
fitting the lake shoreline positions and altimetry<?pagebreak page1617?> water levels as long as
the bias is stable. Therefore, uncertainty in Landsat-derived water levels
developed in this study arises mainly from the variation in this systematic
bias.</p>

      <?xmltex \floatpos{t}?><fig id="Ch1.F10"><?xmltex \currentcnt{10}?><label>Figure 10</label><caption><p id="d1e6140"><inline-formula><mml:math id="M197" display="inline"><mml:mrow><mml:mi>F</mml:mi><mml:mo>(</mml:mo><mml:mi>X</mml:mi><mml:mo>)</mml:mo></mml:mrow></mml:math></inline-formula>: probability density function of the bias (<inline-formula><mml:math id="M198" display="inline"><mml:mi>X</mml:mi></mml:math></inline-formula>) between the
classified water ratio (<inline-formula><mml:math id="M199" display="inline"><mml:mrow><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">1</mml:mn></mml:msub></mml:mrow></mml:math></inline-formula>) and real water ratio (<inline-formula><mml:math id="M200" display="inline"><mml:mrow><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">0</mml:mn></mml:msub></mml:mrow></mml:math></inline-formula>) in a shoreline
pixel.</p></caption>
          <?xmltex \igopts{width=236.157874pt}?><graphic xlink:href="https://essd.copernicus.org/articles/11/1603/2019/essd-11-1603-2019-f10.png"/>

        </fig>

      <p id="d1e6191">To describe the variation in the systematic bias, a new random variable <inline-formula><mml:math id="M201" display="inline"><mml:mi>X</mml:mi></mml:math></inline-formula> was introduced to represent the bias between the classified water area and the
real water area in a shoreline pixel. Given the shape and position of the
lake shoreline, the real water area in each shoreline pixel is a complex
function of the relative position between the pixel and the shoreline. To
simplify the derivation, we assumed that the water area ratio in a shoreline
pixel is uniformly distributed on [0,1], meaning that the probability of any
value between 0 and 1 is equal. If we use <inline-formula><mml:math id="M202" display="inline"><mml:mrow><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">0</mml:mn></mml:msub></mml:mrow></mml:math></inline-formula> to represent the true water
area ratio in the shoreline pixel and <inline-formula><mml:math id="M203" display="inline"><mml:mrow><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">1</mml:mn></mml:msub></mml:mrow></mml:math></inline-formula> to represent
the classified results based on the water area ratio, the random variable
<inline-formula><mml:math id="M204" display="inline"><mml:mi>X</mml:mi></mml:math></inline-formula> can be expressed as
            <disp-formula id="Ch1.E5" content-type="numbered"><label>5</label><mml:math id="M205" display="block"><mml:mrow><mml:mi>X</mml:mi><mml:mo>=</mml:mo><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">1</mml:mn></mml:msub><mml:mo>-</mml:mo><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">0</mml:mn></mml:msub><mml:mo>,</mml:mo></mml:mrow></mml:math></disp-formula>
          where <inline-formula><mml:math id="M206" display="inline"><mml:mrow><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">1</mml:mn></mml:msub></mml:mrow></mml:math></inline-formula> can take on 1 or 0 (i.e., the classified results only tell us
whether a pixel is a water pixel or not), so <inline-formula><mml:math id="M207" display="inline"><mml:mi>X</mml:mi></mml:math></inline-formula> can only take on either
<inline-formula><mml:math id="M208" display="inline"><mml:mrow><mml:mn mathvariant="normal">1</mml:mn><mml:mo>-</mml:mo><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">0</mml:mn></mml:msub></mml:mrow></mml:math></inline-formula> or <inline-formula><mml:math id="M209" display="inline"><mml:mrow><mml:mo>-</mml:mo><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">0</mml:mn></mml:msub></mml:mrow></mml:math></inline-formula>. Because the range of <inline-formula><mml:math id="M210" display="inline"><mml:mrow><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">0</mml:mn></mml:msub></mml:mrow></mml:math></inline-formula> is [0,1], it
is obvious that the range of <inline-formula><mml:math id="M211" display="inline"><mml:mi>X</mml:mi></mml:math></inline-formula> is [<inline-formula><mml:math id="M212" display="inline"><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">1</mml:mn></mml:mrow></mml:math></inline-formula>,1]. A derivation of <inline-formula><mml:math id="M213" display="inline"><mml:mrow><mml:mi>F</mml:mi><mml:mo>(</mml:mo><mml:mi>X</mml:mi><mml:mo>)</mml:mo></mml:mrow></mml:math></inline-formula>, i.e., the probability density function (PDF) of <inline-formula><mml:math id="M214" display="inline"><mml:mi>X</mml:mi></mml:math></inline-formula>, can be found in the
Supplement (Part 2).</p>
      <p id="d1e6352">Overall, <inline-formula><mml:math id="M215" display="inline"><mml:mrow><mml:mi>F</mml:mi><mml:mo>(</mml:mo><mml:mi>X</mml:mi><mml:mo>)</mml:mo></mml:mrow></mml:math></inline-formula> describes how the bias between the classified water ratio and
real water ratio in shoreline pixels is distributed as shown in Fig. 10. If there are <inline-formula><mml:math id="M216" display="inline"><mml:mi>N</mml:mi></mml:math></inline-formula> shoreline pixels in an ROI, we can take them as
<inline-formula><mml:math id="M217" display="inline"><mml:mi>N</mml:mi></mml:math></inline-formula> independent observations of <inline-formula><mml:math id="M218" display="inline"><mml:mi>X</mml:mi></mml:math></inline-formula> and calculate the mean value <inline-formula><mml:math id="M219" display="inline"><mml:mover accent="true"><mml:mi>X</mml:mi><mml:mo mathvariant="normal">‾</mml:mo></mml:mover></mml:math></inline-formula>. This value <inline-formula><mml:math id="M220" display="inline"><mml:mover accent="true"><mml:mi>X</mml:mi><mml:mo mathvariant="normal">‾</mml:mo></mml:mover></mml:math></inline-formula> can represent an average shift of the detected lake shoreline from the real lake shoreline in the unit of 1-pixel width
(30 m). As we mentioned above, the systematic bias can be removed in the
regression between the lake shoreline positions and altimetry water levels.
As such, it is the variation of the bias that determines the accuracy of the
Landsat-derived water levels. We can calculate the standard variation of
<inline-formula><mml:math id="M221" display="inline"><mml:mover accent="true"><mml:mi>X</mml:mi><mml:mo mathvariant="normal">‾</mml:mo></mml:mover></mml:math></inline-formula> to represent the uncertainty in lake shoreline position. Note that
there is a simple relationship between <inline-formula><mml:math id="M222" display="inline"><mml:mrow><mml:msub><mml:mi mathvariant="italic">σ</mml:mi><mml:mover accent="true"><mml:mi>x</mml:mi><mml:mo mathvariant="normal">‾</mml:mo></mml:mover></mml:msub></mml:mrow></mml:math></inline-formula> and
<inline-formula><mml:math id="M223" display="inline"><mml:mrow><mml:msub><mml:mi mathvariant="italic">σ</mml:mi><mml:mi>x</mml:mi></mml:msub></mml:mrow></mml:math></inline-formula>:
            <disp-formula id="Ch1.E6" content-type="numbered"><label>6</label><mml:math id="M224" display="block"><mml:mrow><mml:msub><mml:mi mathvariant="italic">σ</mml:mi><mml:mover accent="true"><mml:mi>x</mml:mi><mml:mo mathvariant="normal">‾</mml:mo></mml:mover></mml:msub><mml:mo>=</mml:mo><mml:mstyle displaystyle="true"><mml:mfrac style="display"><mml:mrow><mml:msub><mml:mi mathvariant="italic">σ</mml:mi><mml:mi>x</mml:mi></mml:msub></mml:mrow><mml:msqrt><mml:mi>N</mml:mi></mml:msqrt></mml:mfrac></mml:mstyle><mml:mo>.</mml:mo></mml:mrow></mml:math></disp-formula>
          One only needs to calculate <inline-formula><mml:math id="M225" display="inline"><mml:mrow><mml:msub><mml:mi mathvariant="italic">σ</mml:mi><mml:mi>x</mml:mi></mml:msub></mml:mrow></mml:math></inline-formula>:

                <disp-formula specific-use="align" content-type="numbered"><mml:math id="M226" display="block"><mml:mtable displaystyle="true"><mml:mlabeledtr id="Ch1.E7"><mml:mtd><mml:mtext>7</mml:mtext></mml:mtd><mml:mtd><mml:mstyle displaystyle="true" class="stylechange"/></mml:mtd><mml:mtd><mml:mrow><mml:mstyle class="stylechange" displaystyle="true"/><mml:mover accent="true"><mml:mi>X</mml:mi><mml:mo mathvariant="normal">‾</mml:mo></mml:mover><mml:mo>=</mml:mo><mml:munderover><mml:mo movablelimits="false">∫</mml:mo><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">1</mml:mn></mml:mrow><mml:mn mathvariant="normal">1</mml:mn></mml:munderover><mml:mi>F</mml:mi><mml:mo>(</mml:mo><mml:mi>X</mml:mi><mml:mo>)</mml:mo><mml:mo>⋅</mml:mo><mml:mi>X</mml:mi><mml:mi>d</mml:mi><mml:mi>X</mml:mi><mml:mo>≈</mml:mo><mml:mo>-</mml:mo><mml:mn mathvariant="normal">0.09</mml:mn><mml:mo>,</mml:mo></mml:mrow></mml:mtd></mml:mlabeledtr><mml:mlabeledtr id="Ch1.E8"><mml:mtd><mml:mtext>8</mml:mtext></mml:mtd><mml:mtd><mml:mstyle displaystyle="true" class="stylechange"/></mml:mtd><mml:mtd><mml:mrow><mml:mstyle displaystyle="true" class="stylechange"/><mml:msub><mml:mi mathvariant="italic">σ</mml:mi><mml:mi>x</mml:mi></mml:msub><mml:mo>=</mml:mo><mml:msqrt><mml:mrow><mml:munderover><mml:mo movablelimits="false">∫</mml:mo><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">1</mml:mn></mml:mrow><mml:mn mathvariant="normal">1</mml:mn></mml:munderover><mml:mi>F</mml:mi><mml:mo>(</mml:mo><mml:mi>X</mml:mi><mml:mo>)</mml:mo><mml:mo>⋅</mml:mo><mml:mo>(</mml:mo><mml:mi>X</mml:mi><mml:mo>-</mml:mo><mml:mover accent="true"><mml:mi>X</mml:mi><mml:mo mathvariant="normal">‾</mml:mo></mml:mover><mml:msup><mml:mo>)</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mi>d</mml:mi><mml:mi>X</mml:mi></mml:mrow></mml:msqrt><mml:mo>≈</mml:mo><mml:mn mathvariant="normal">0.39</mml:mn><mml:mo>.</mml:mo></mml:mrow></mml:mtd></mml:mlabeledtr></mml:mtable></mml:math></disp-formula>

            Combined with Eqs. (4) and (7), Eq. (8) was resolved numerically,
resulting in <inline-formula><mml:math id="M227" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 0.39-pixel width. Substituting <inline-formula><mml:math id="M228" display="inline"><mml:mrow><mml:msub><mml:mi mathvariant="italic">σ</mml:mi><mml:mi>x</mml:mi></mml:msub></mml:mrow></mml:math></inline-formula>
in Eq. (6) with Eq. (8) gives
            <disp-formula id="Ch1.E9" content-type="numbered"><label>9</label><mml:math id="M229" display="block"><mml:mrow><mml:msub><mml:mi mathvariant="italic">σ</mml:mi><mml:mover accent="true"><mml:mi>x</mml:mi><mml:mo mathvariant="normal">‾</mml:mo></mml:mover></mml:msub><mml:mo>=</mml:mo><mml:mstyle displaystyle="true"><mml:mfrac style="display"><mml:mn mathvariant="normal">0.39</mml:mn><mml:msqrt><mml:mi>N</mml:mi></mml:msqrt></mml:mfrac></mml:mstyle><mml:mo>.</mml:mo></mml:mrow></mml:math></disp-formula>
          If the slope of the shoreline is known, e.g., tan <inline-formula><mml:math id="M230" display="inline"><mml:mi mathvariant="italic">θ</mml:mi></mml:math></inline-formula>, the uncertainty of
the Landsat-derived water level can be expressed as
            <disp-formula id="Ch1.E10" content-type="numbered"><label>10</label><mml:math id="M231" display="block"><mml:mrow><mml:msub><mml:mi mathvariant="italic">σ</mml:mi><mml:mtext>ho</mml:mtext></mml:msub><mml:mo>=</mml:mo><mml:msub><mml:mi mathvariant="italic">σ</mml:mi><mml:mover accent="true"><mml:mi>x</mml:mi><mml:mo mathvariant="normal">‾</mml:mo></mml:mover></mml:msub><mml:mo>⋅</mml:mo><mml:mi>d</mml:mi><mml:mo>⋅</mml:mo><mml:mi>tan⁡</mml:mi><mml:mi mathvariant="italic">θ</mml:mi><mml:mo>=</mml:mo><mml:mstyle displaystyle="true"><mml:mfrac style="display"><mml:mrow><mml:mn mathvariant="normal">0.39</mml:mn><mml:mo>×</mml:mo><mml:mn mathvariant="normal">30</mml:mn><mml:mo>×</mml:mo><mml:mi>tan⁡</mml:mi><mml:mi mathvariant="italic">θ</mml:mi></mml:mrow><mml:msqrt><mml:mi>N</mml:mi></mml:msqrt></mml:mfrac></mml:mstyle><mml:mo>,</mml:mo></mml:mrow></mml:math></disp-formula>
          where <inline-formula><mml:math id="M232" display="inline"><mml:mrow><mml:msub><mml:mi mathvariant="italic">σ</mml:mi><mml:mtext>ho</mml:mtext></mml:msub></mml:mrow></mml:math></inline-formula> is the uncertainty of Landsat-derived water levels
and <inline-formula><mml:math id="M233" display="inline"><mml:mi>d</mml:mi></mml:math></inline-formula> is the spatial resolution of the satellite image (30 m). In this study, a typical width of ROI for deriving Landsat-derived water levels is
<inline-formula><mml:math id="M234" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 10-pixel width, meaning that <inline-formula><mml:math id="M235" display="inline"><mml:mi>N</mml:mi></mml:math></inline-formula> is <inline-formula><mml:math id="M236" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 10. In
addition, lake shores used for generating Landsat-derived water levels here
generally have a relatively mild slope of <inline-formula><mml:math id="M237" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> <inline-formula><mml:math id="M238" display="inline"><mml:mrow><mml:mn mathvariant="normal">1</mml:mn><mml:mo>/</mml:mo><mml:mn mathvariant="normal">30</mml:mn></mml:mrow></mml:math></inline-formula> or even
smaller, which can be roughly estimated from the maximum shoreline change
and altimetry water level change within a year. Here if we use <inline-formula><mml:math id="M239" display="inline"><mml:mrow><mml:mn mathvariant="normal">1</mml:mn><mml:mo>/</mml:mo><mml:mn mathvariant="normal">30</mml:mn></mml:mrow></mml:math></inline-formula> as the
slope tan <inline-formula><mml:math id="M240" display="inline"><mml:mi mathvariant="italic">θ</mml:mi></mml:math></inline-formula>, the uncertainty of the Landsat-derived water levels can
ultimately be estimated to be <inline-formula><mml:math id="M241" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 0.12 m, which is very close to
the RMSE of 0.11 m based on the comparison between the optical water levels
and in situ water level measurements mentioned earlier.</p>
      <p id="d1e6798">However, for most cases we do not know the exact lake bank slope
tan <inline-formula><mml:math id="M242" display="inline"><mml:mi mathvariant="italic">θ</mml:mi></mml:math></inline-formula>, which is the reason why we performed the regression analysis
between the lake shoreline positions and altimetry-derived water levels.
Information on the real lake bank slope is implicitly expressed in the
linear fitting slope <inline-formula><mml:math id="M243" display="inline"><mml:mi mathvariant="italic">β</mml:mi></mml:math></inline-formula> (if the fitting line is <inline-formula><mml:math id="M244" display="inline"><mml:mrow><mml:mi>y</mml:mi><mml:mo>=</mml:mo><mml:mi mathvariant="italic">β</mml:mi><mml:mi>x</mml:mi><mml:mo>+</mml:mo><mml:mi mathvariant="italic">α</mml:mi></mml:mrow></mml:math></inline-formula>). Uncertainty in altimetric information could evolve into the
fitting parameters and impact the accuracy of the generated Landsat-derived
water levels. Given that the observed lake shoreline position is <inline-formula><mml:math id="M245" display="inline"><mml:mrow><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">1</mml:mn></mml:msub></mml:mrow></mml:math></inline-formula>
(e.g., <inline-formula><mml:math id="M246" display="inline"><mml:mrow><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">1</mml:mn></mml:msub><mml:mo>=</mml:mo><mml:mn mathvariant="normal">5.6</mml:mn></mml:mrow></mml:math></inline-formula>, meaning that the observed lake shoreline position is
5.6 Landsat pixels away from the initial position corresponding to the lowest
water level, which is different from Eq. (5), <inline-formula><mml:math id="M247" display="inline"><mml:mrow><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">1</mml:mn></mml:msub></mml:mrow></mml:math></inline-formula> here can be a rational number
because it is determined by averaging all shoreline pixels in the ROI,
whereas<?pagebreak page1618?> in Eq. 5 we focused on only 1 shoreline pixel), combining Eq. (5), the Landsat-derived water level (<inline-formula><mml:math id="M248" display="inline"><mml:mi>y</mml:mi></mml:math></inline-formula>) can be expressed as
            <disp-formula id="Ch1.E11" content-type="numbered"><label>11</label><mml:math id="M249" display="block"><mml:mrow><mml:mi>y</mml:mi><mml:mo>=</mml:mo><mml:mi mathvariant="italic">β</mml:mi><mml:mfenced close=")" open="("><mml:mrow><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">1</mml:mn></mml:msub><mml:mo>-</mml:mo><mml:mover accent="true"><mml:mrow><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">1</mml:mn></mml:msub></mml:mrow><mml:mo mathvariant="normal">‾</mml:mo></mml:mover></mml:mrow></mml:mfenced><mml:mo>+</mml:mo><mml:mover accent="true"><mml:mi>Y</mml:mi><mml:mo mathvariant="normal">‾</mml:mo></mml:mover><mml:mo>=</mml:mo><mml:mi mathvariant="italic">β</mml:mi><mml:mfenced close=")" open="("><mml:mrow><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">0</mml:mn></mml:msub><mml:mo>-</mml:mo><mml:mover accent="true"><mml:mrow><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">0</mml:mn></mml:msub></mml:mrow><mml:mo mathvariant="normal">‾</mml:mo></mml:mover></mml:mrow></mml:mfenced><mml:mo>+</mml:mo><mml:mi mathvariant="italic">β</mml:mi><mml:mfenced open="(" close=")"><mml:mrow><mml:mi>X</mml:mi><mml:mo>-</mml:mo><mml:mover accent="true"><mml:mi>X</mml:mi><mml:mo mathvariant="normal">‾</mml:mo></mml:mover></mml:mrow></mml:mfenced><mml:mo>+</mml:mo><mml:mover accent="true"><mml:mi>Y</mml:mi><mml:mo mathvariant="normal">‾</mml:mo></mml:mover><mml:mo>,</mml:mo></mml:mrow></mml:math></disp-formula>
          where <inline-formula><mml:math id="M250" display="inline"><mml:mrow><mml:mfenced open="(" close=")"><mml:mrow><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">1</mml:mn></mml:msub><mml:mo>-</mml:mo><mml:mover accent="true"><mml:mrow><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">1</mml:mn></mml:msub></mml:mrow><mml:mo mathvariant="normal">‾</mml:mo></mml:mover></mml:mrow></mml:mfenced></mml:mrow></mml:math></inline-formula> denotes the
observed lake shoreline change (in the unit of a Landsat pixel), <inline-formula><mml:math id="M251" display="inline"><mml:mover accent="true"><mml:mrow><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">1</mml:mn></mml:msub></mml:mrow><mml:mo mathvariant="normal">‾</mml:mo></mml:mover></mml:math></inline-formula> denotes the mean of observed lake shoreline positions used for
linear regression, <inline-formula><mml:math id="M252" display="inline"><mml:mover accent="true"><mml:mi>Y</mml:mi><mml:mo mathvariant="normal">‾</mml:mo></mml:mover></mml:math></inline-formula> denotes the mean of altimetry water levels used
for linear regression, <inline-formula><mml:math id="M253" display="inline"><mml:mrow><mml:mfenced close=")" open="("><mml:mrow><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">0</mml:mn></mml:msub><mml:mo>-</mml:mo><mml:mover accent="true"><mml:mrow><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">0</mml:mn></mml:msub></mml:mrow><mml:mo mathvariant="normal">‾</mml:mo></mml:mover></mml:mrow></mml:mfenced></mml:mrow></mml:math></inline-formula>
denotes the real lake shoreline change, <inline-formula><mml:math id="M254" display="inline"><mml:mrow><mml:mo>(</mml:mo><mml:mi>X</mml:mi><mml:mo>-</mml:mo><mml:mover accent="true"><mml:mi>X</mml:mi><mml:mo mathvariant="normal">‾</mml:mo></mml:mover><mml:mo>)</mml:mo></mml:mrow></mml:math></inline-formula>
denotes the variation of the Landsat-derived shoreline position caused by
the water extraction method, and <inline-formula><mml:math id="M255" display="inline"><mml:mi mathvariant="italic">β</mml:mi></mml:math></inline-formula> is the linear fitting slope. It is
obvious that the expected value <inline-formula><mml:math id="M256" display="inline"><mml:mrow><mml:mfenced close=")" open="("><mml:mrow><mml:mi>X</mml:mi><mml:mo>-</mml:mo><mml:mover accent="true"><mml:mi>X</mml:mi><mml:mo mathvariant="normal">‾</mml:mo></mml:mover></mml:mrow></mml:mfenced></mml:mrow></mml:math></inline-formula> is 0. As
we discussed earlier, a systematic bias does not affect the accuracy of the
Landsat-derived water level but the variation of the systematic bias does.</p>
      <p id="d1e7078">Based on Eq. (11), the overall uncertainty of the Landsat-derived water level
<inline-formula><mml:math id="M257" display="inline"><mml:mrow><mml:msub><mml:mi mathvariant="italic">σ</mml:mi><mml:mi>y</mml:mi></mml:msub></mml:mrow></mml:math></inline-formula> can be given as
            <disp-formula id="Ch1.E12" content-type="numbered"><label>12</label><mml:math id="M258" display="block"><mml:mtable columnspacing="1em" class="split" rowspacing="0.2ex" displaystyle="true" columnalign="right left"><mml:mtr><mml:mtd><mml:mrow><mml:msub><mml:mi mathvariant="italic">σ</mml:mi><mml:mi>y</mml:mi></mml:msub><mml:mo>=</mml:mo></mml:mrow></mml:mtd><mml:mtd><mml:mrow><mml:msqrt><mml:mrow><mml:msubsup><mml:mi mathvariant="italic">σ</mml:mi><mml:mi mathvariant="italic">β</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msubsup><mml:msup><mml:mfenced close=")" open="("><mml:mstyle displaystyle="true"><mml:mfrac style="display"><mml:mrow><mml:mo>∂</mml:mo><mml:mi>y</mml:mi></mml:mrow><mml:mrow><mml:mo>∂</mml:mo><mml:mi mathvariant="italic">β</mml:mi></mml:mrow></mml:mfrac></mml:mstyle></mml:mfenced><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:msubsup><mml:mi mathvariant="italic">σ</mml:mi><mml:mover accent="true"><mml:mi>x</mml:mi><mml:mo mathvariant="normal">‾</mml:mo></mml:mover><mml:mn mathvariant="normal">2</mml:mn></mml:msubsup><mml:msup><mml:mfenced close=")" open="("><mml:mstyle displaystyle="true"><mml:mfrac style="display"><mml:mrow><mml:mo>∂</mml:mo><mml:mi>y</mml:mi></mml:mrow><mml:mrow><mml:mo>∂</mml:mo><mml:mfenced close=")" open="("><mml:mrow><mml:mi>X</mml:mi><mml:mo>-</mml:mo><mml:mover accent="true"><mml:mi>X</mml:mi><mml:mo mathvariant="normal">‾</mml:mo></mml:mover></mml:mrow></mml:mfenced></mml:mrow></mml:mfrac></mml:mstyle></mml:mfenced><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:msubsup><mml:mi mathvariant="italic">σ</mml:mi><mml:mover accent="true"><mml:mi>Y</mml:mi><mml:mo mathvariant="normal">‾</mml:mo></mml:mover><mml:mn mathvariant="normal">2</mml:mn></mml:msubsup><mml:msup><mml:mfenced open="(" close=")"><mml:mstyle displaystyle="true"><mml:mfrac style="display"><mml:mrow><mml:mo>∂</mml:mo><mml:mi>y</mml:mi></mml:mrow><mml:mrow><mml:mo>∂</mml:mo><mml:mover accent="true"><mml:mi>Y</mml:mi><mml:mo mathvariant="normal">‾</mml:mo></mml:mover></mml:mrow></mml:mfrac></mml:mstyle></mml:mfenced><mml:mn mathvariant="normal">2</mml:mn></mml:msup></mml:mrow></mml:msqrt></mml:mrow></mml:mtd></mml:mtr><mml:mtr><mml:mtd><mml:mrow><mml:mo>=</mml:mo></mml:mrow></mml:mtd><mml:mtd><mml:mrow><mml:msqrt><mml:mrow><mml:msubsup><mml:mi mathvariant="italic">σ</mml:mi><mml:mi mathvariant="italic">β</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msubsup><mml:mo>(</mml:mo><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">1</mml:mn></mml:msub><mml:mo>-</mml:mo><mml:mover accent="true"><mml:mrow><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">1</mml:mn></mml:msub></mml:mrow><mml:mo mathvariant="normal">‾</mml:mo></mml:mover><mml:msup><mml:mo>)</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:msubsup><mml:mi mathvariant="italic">σ</mml:mi><mml:mover accent="true"><mml:mi>x</mml:mi><mml:mo mathvariant="normal">‾</mml:mo></mml:mover><mml:mn mathvariant="normal">2</mml:mn></mml:msubsup><mml:msup><mml:mi mathvariant="italic">β</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup><mml:mo>+</mml:mo><mml:msubsup><mml:mi mathvariant="italic">σ</mml:mi><mml:mover accent="true"><mml:mi>Y</mml:mi><mml:mo mathvariant="normal">‾</mml:mo></mml:mover><mml:mn mathvariant="normal">2</mml:mn></mml:msubsup></mml:mrow></mml:msqrt><mml:mo>,</mml:mo></mml:mrow></mml:mtd></mml:mtr></mml:mtable></mml:math></disp-formula>
          where <inline-formula><mml:math id="M259" display="inline"><mml:mi mathvariant="italic">β</mml:mi></mml:math></inline-formula> and <inline-formula><mml:math id="M260" display="inline"><mml:mrow><mml:msub><mml:mi mathvariant="italic">σ</mml:mi><mml:mi mathvariant="italic">β</mml:mi></mml:msub></mml:mrow></mml:math></inline-formula> can be derived from the linear
regression analysis, <inline-formula><mml:math id="M261" display="inline"><mml:mrow><mml:msub><mml:mi mathvariant="italic">σ</mml:mi><mml:mover accent="true"><mml:mi>x</mml:mi><mml:mo mathvariant="normal">‾</mml:mo></mml:mover></mml:msub></mml:mrow></mml:math></inline-formula> is given in Eq. (9), and <inline-formula><mml:math id="M262" display="inline"><mml:mrow><mml:msub><mml:mi mathvariant="italic">σ</mml:mi><mml:mover accent="true"><mml:mi>Y</mml:mi><mml:mo mathvariant="normal">‾</mml:mo></mml:mover></mml:msub></mml:mrow></mml:math></inline-formula> is the uncertainty of the mean altimetry water level which can
be estimated from the altimetry data. For a typical lake like Yamzhog Yumco,
<inline-formula><mml:math id="M263" display="inline"><mml:mrow><mml:mi mathvariant="italic">β</mml:mi><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.35</mml:mn></mml:mrow></mml:math></inline-formula> m, <inline-formula><mml:math id="M264" display="inline"><mml:mrow><mml:msub><mml:mi mathvariant="italic">σ</mml:mi><mml:mi mathvariant="italic">β</mml:mi></mml:msub><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.02</mml:mn></mml:mrow></mml:math></inline-formula> m, Max(<inline-formula><mml:math id="M265" display="inline"><mml:mrow><mml:mi mathvariant="normal">|</mml:mi><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">1</mml:mn></mml:msub><mml:mo>-</mml:mo><mml:mover accent="true"><mml:mrow><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">1</mml:mn></mml:msub><mml:mi mathvariant="normal">|</mml:mi></mml:mrow><mml:mo mathvariant="normal">‾</mml:mo></mml:mover><mml:mo>)</mml:mo><mml:mo>=</mml:mo></mml:mrow></mml:math></inline-formula> 11, <inline-formula><mml:math id="M266" display="inline"><mml:mrow><mml:msub><mml:mi mathvariant="italic">σ</mml:mi><mml:mover accent="true"><mml:mi>x</mml:mi><mml:mo mathvariant="normal">‾</mml:mo></mml:mover></mml:msub><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.13</mml:mn></mml:mrow></mml:math></inline-formula>, and <inline-formula><mml:math id="M267" display="inline"><mml:mrow><mml:msub><mml:mi mathvariant="italic">σ</mml:mi><mml:mover accent="true"><mml:mi>Y</mml:mi><mml:mo mathvariant="normal">‾</mml:mo></mml:mover></mml:msub><mml:mo>=</mml:mo><mml:mn mathvariant="normal">0.015</mml:mn></mml:mrow></mml:math></inline-formula> m, which gives a maximum <inline-formula><mml:math id="M268" display="inline"><mml:mrow><mml:msub><mml:mi mathvariant="italic">σ</mml:mi><mml:mi>y</mml:mi></mml:msub></mml:mrow></mml:math></inline-formula>
of 0.22 m. Note that <inline-formula><mml:math id="M269" display="inline"><mml:mrow><mml:mfenced close=")" open="("><mml:mrow><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">0</mml:mn></mml:msub><mml:mo>-</mml:mo><mml:mover accent="true"><mml:mrow><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">0</mml:mn></mml:msub></mml:mrow><mml:mo mathvariant="normal">‾</mml:mo></mml:mover></mml:mrow></mml:mfenced></mml:mrow></mml:math></inline-formula> is
assumed to be the ground truth so there is no error associated with this
term. This relationship shows that the uncertainty in the Landsat-derived
water level increases with the distance from the center point
(<inline-formula><mml:math id="M270" display="inline"><mml:mover accent="true"><mml:mrow><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">1</mml:mn></mml:msub></mml:mrow><mml:mo mathvariant="normal">‾</mml:mo></mml:mover></mml:math></inline-formula>, <inline-formula><mml:math id="M271" display="inline"><mml:mover accent="true"><mml:mi>Y</mml:mi><mml:mo mathvariant="normal">‾</mml:mo></mml:mover></mml:math></inline-formula>) represented by
<inline-formula><mml:math id="M272" display="inline"><mml:mrow><mml:mo>(</mml:mo><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">1</mml:mn></mml:msub><mml:mo>-</mml:mo><mml:mover accent="true"><mml:mrow><mml:msub><mml:mi>X</mml:mi><mml:mn mathvariant="normal">1</mml:mn></mml:msub></mml:mrow><mml:mo mathvariant="normal">‾</mml:mo></mml:mover><mml:msup><mml:mo>)</mml:mo><mml:mn mathvariant="normal">2</mml:mn></mml:msup></mml:mrow></mml:math></inline-formula>.
Interpretation of this phenomenon is that extrapolation of Landsat-derived
water levels (far from the center point) may cause some errors and should be
used with caution. More detailed discussion on the extrapolation can be
found in the Supplement.</p>
      <p id="d1e7502">Overall, the uncertainty quantification of the Landsat-derived water levels
developed in this study indicates clearly that the accuracy of
Landsat-derived water levels depends on the width of an ROI, e.g., the
number of pixels/observations, slope of the lake shore, the effectiveness of
the water classification method, and the uncertainty in the altimetry water
level used for regression. One of the advantages of the Landsat-derived
water level is that an ROI does not necessarily cover a large area of lake
shores, which maximizes the potential of optical remote sensing images to
increase the spatial coverage and temporal resolution of lake water level
estimates that may not be realized by using satellite altimetry alone.
Optical remote sensing images provide important complementary information on
altimetry water levels and can subsequently facilitate lake water storage
estimation.</p>

      <?xmltex \floatpos{t}?><fig id="Ch1.F11" specific-use="star"><?xmltex \currentcnt{11}?><label>Figure 11</label><caption><p id="d1e7507">Cross validation of the TP lake level and storage
changes derived from our study with those provided by the LEGOS Hydroweb
database (Crétaux et al., 2011a): <bold>(a)</bold> trends in lake water levels
from 2003 to 2015 and <bold>(b)</bold> trends in lake water storage from 2003 to 2015.</p></caption>
          <?xmltex \igopts{width=398.338583pt}?><graphic xlink:href="https://essd.copernicus.org/articles/11/1603/2019/essd-11-1603-2019-f11.png"/>

        </fig>

      <?xmltex \floatpos{t}?><fig id="Ch1.F12" specific-use="star"><?xmltex \currentcnt{12}?><label>Figure 12</label><caption><p id="d1e7525">Spatial distribution of trends in lake storage on the TP during
2000–2017. The black line shows the boundary of the endorheic basin of
the TP including 39 lakes in this study. The other 13 lakes are located
outside the endorheic basin.</p></caption>
          <?xmltex \igopts{width=398.338583pt}?><graphic xlink:href="https://essd.copernicus.org/articles/11/1603/2019/essd-11-1603-2019-f12.png"/>

        </fig>

</sec>
<sec id="Ch1.S4.SS3">
  <label>4.3</label><title>Cross validation with similar products</title>
      <p id="d1e7542">We compared our product with a widely used lake water level/storage data set
provided by the LEGOS Hydroweb, indicating that the two products are, on the
whole, consistent with each other (shown in Fig. 11), but our product may
perform better in terms of the temporal continuity as well as the temporal
resolution (shown in Sect. 6.2). Both advantages are important in
improving our understanding of responses of lakes to climate change. There
are 21 lakes that are the same in both our study and LEGOS Hydroweb. Annual trends in
water level and lake storage during 2003–2015 are compared in Fig. 11,
indicating the overall consistency of the two products in terms of <inline-formula><mml:math id="M273" display="inline"><mml:mrow><mml:msup><mml:mi>R</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup></mml:mrow></mml:math></inline-formula> of
the linear fit.</p>
</sec>
<sec id="Ch1.S4.SS4">
  <label>4.4</label><title>Data set description</title>
      <p id="d1e7564">The data sets cover 52 large lakes (50 lakes with a surface area larger
than 150 km<inline-formula><mml:math id="M274" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">2</mml:mn></mml:msup></mml:math></inline-formula> and 2 lakes that are 100–150 km<inline-formula><mml:math id="M275" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">2</mml:mn></mml:msup></mml:math></inline-formula>) on the TP. The data
sets consist of two parts: (1) a table containing hypsometric curves and
corresponding regression statistics (<inline-formula><mml:math id="M276" display="inline"><mml:mrow><mml:msup><mml:mi>R</mml:mi><mml:mn mathvariant="normal">2</mml:mn></mml:msup></mml:mrow></mml:math></inline-formula> and the number of data pairs)
for each lake, with parameters of the hypsometric curves listed in separate
columns for the convenience of batch processing; and (2) time series for each
lake archived as 52 entities with geographic information (i.e., latitude,
longitude, and size of the lake) that can be checked in an online map
provided by PANGAEA, avoiding the confusion of lake names. The time series
of each lake include lake water levels and lake storage changes.</p>
      <p id="d1e7596">For data points in the water level time series, satellite or sensor type is
shown (i.e., from Jason-1/2/3, Envisat, ICESat, CryoSat, or optical images).
Uncertainty was calculated using the standard deviation of valid footprints
in the cycle (only for altimetry data). The lake water storage time series
were transformed from water level time series using the hypsometric
relationship so that they have equal data size. The lake water storage time
series represent changes in lake storage with respect to a reference water
level, which is listed in the corresponding hypsometric curve table as a
parameter. The overall uncertainty of Landsat-derived water levels within
the regression range (the range of altimetry water levels) is 0.1–0.2 m
based on the experiment and analysis in this paper. The extrapolation of
Landsat-derived water levels may occur during the time gap between
altimetric missions and before 2002. The average uncertainty of altimetry
water levels is 0.11 m.</p>
</sec>
</sec>
<sec id="Ch1.S5">
  <label>5</label><title>Applications</title>
<sec id="Ch1.S5.SS1">
  <label>5.1</label><title>Spatiotemporal analysis of changes in lake water storage in the Tibetan Plateau</title>
      <p id="d1e7616">Based on the lake water storage changes we derived, spatial patterns of lake
storage trends during 2000–2017 were<?pagebreak page1619?> shown in Fig. 12. In the endorheic basin of the TP, similar to some reported results (Yao et al., 2018b; Zhang et al., 2017a), most lakes have been
expanding rapidly; e.g., Selin Co (31.80<inline-formula><mml:math id="M277" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> N, 89.00<inline-formula><mml:math id="M278" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> E) gained <inline-formula><mml:math id="M279" display="inline"><mml:mrow><mml:mn mathvariant="normal">19.7</mml:mn><mml:mo>±</mml:mo><mml:mn mathvariant="normal">2.0</mml:mn></mml:mrow></mml:math></inline-formula> km<inline-formula><mml:math id="M280" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">3</mml:mn></mml:msup></mml:math></inline-formula> of water during the study period, and Lake Kusai (35.70<inline-formula><mml:math id="M281" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> N, 92.90<inline-formula><mml:math id="M282" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> E)
experienced an abrupt expansion due to flood and gained <inline-formula><mml:math id="M283" display="inline"><mml:mrow><mml:mn mathvariant="normal">2.2</mml:mn><mml:mo>±</mml:mo><mml:mn mathvariant="normal">0.2</mml:mn></mml:mrow></mml:math></inline-formula> km<inline-formula><mml:math id="M284" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">3</mml:mn></mml:msup></mml:math></inline-formula> of water in 2011, as reported in related work (Yao et al.,
2012). In contrast, some lakes in the southern part of the TP experienced
shrinkage, e.g., Yamzhog Yumco (28.93<inline-formula><mml:math id="M285" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> N, 90.70<inline-formula><mml:math id="M286" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> E) gained a total of <inline-formula><mml:math id="M287" display="inline"><mml:mrow><mml:mn mathvariant="normal">0.8</mml:mn><mml:mo>±</mml:mo><mml:mn mathvariant="normal">0.4</mml:mn></mml:mrow></mml:math></inline-formula> km<inline-formula><mml:math id="M288" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">3</mml:mn></mml:msup></mml:math></inline-formula> water during 2000–2004 but has shrunk during the remaining 13 years (2005–2017) at a rate of <inline-formula><mml:math id="M289" display="inline"><mml:mrow><mml:mo>-</mml:mo><mml:mn mathvariant="normal">0.19</mml:mn><mml:mo>±</mml:mo><mml:mn mathvariant="normal">0.03</mml:mn></mml:mrow></mml:math></inline-formula> km<inline-formula><mml:math id="M290" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">3</mml:mn></mml:msup></mml:math></inline-formula> yr<inline-formula><mml:math id="M291" 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 contrast to Yamzhog Yumco, Lake Qinghai (36.90<inline-formula><mml:math id="M292" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> N, 100.00<inline-formula><mml:math id="M293" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> E) lost <inline-formula><mml:math id="M294" display="inline"><mml:mrow><mml:mn mathvariant="normal">2.2</mml:mn><mml:mo>±</mml:mo><mml:mn mathvariant="normal">0.7</mml:mn></mml:mrow></mml:math></inline-formula> km<inline-formula><mml:math id="M295" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">3</mml:mn></mml:msup></mml:math></inline-formula> water during 2000–2004 but gained <inline-formula><mml:math id="M296" display="inline"><mml:mrow><mml:mn mathvariant="normal">7.7</mml:mn><mml:mo>±</mml:mo><mml:mn mathvariant="normal">0.6</mml:mn></mml:mrow></mml:math></inline-formula> km<inline-formula><mml:math id="M297" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">3</mml:mn></mml:msup></mml:math></inline-formula> of water during 2005–2017. Similar patterns can be detected in adjacent lakes of Lake Qinghai, e.g., Lake Donggei Cuona (35.28<inline-formula><mml:math id="M298" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> N, 98.55<inline-formula><mml:math id="M299" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> E)
and Lake Ngoring (34.90<inline-formula><mml:math id="M300" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> N, 97.70<inline-formula><mml:math id="M301" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> E).</p>

      <?xmltex \floatpos{t}?><fig id="Ch1.F13" specific-use="star"><?xmltex \currentcnt{13}?><label>Figure 13</label><caption><p id="d1e7873">Discrepancy of lake storage trends in Goren Co between
Yao et al. (2018a) and our study.</p></caption>
          <?xmltex \igopts{width=341.433071pt}?><graphic xlink:href="https://essd.copernicus.org/articles/11/1603/2019/essd-11-1603-2019-f13.png"/>

        </fig>

      <?xmltex \floatpos{t}?><fig id="Ch1.F14" specific-use="star"><?xmltex \currentcnt{14}?><label>Figure 14</label><caption><p id="d1e7884"><bold>(a)</bold> Total storage changes in the 52 lakes on the TP,
which can be generally divided into two stages: (1) a rapidly increasing
stage (2000–2011) with a higher increasing rate of 6.71 km<inline-formula><mml:math id="M302" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">3</mml:mn></mml:msup></mml:math></inline-formula> yr<inline-formula><mml:math id="M303" 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> and (2) a mildly increasing stage (2012–2017) with an increasing rate of 1.98 km<inline-formula><mml:math id="M304" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">3</mml:mn></mml:msup></mml:math></inline-formula> yr<inline-formula><mml:math id="M305" 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>. <bold>(b)</bold> Histogram of intra-annual changes in lake water levels of
the 52 lakes on the TP.</p></caption>
          <?xmltex \igopts{width=398.338583pt}?><graphic xlink:href="https://essd.copernicus.org/articles/11/1603/2019/essd-11-1603-2019-f14.png"/>

        </fig>

      <p id="d1e7941">However, spatial proximity cannot fully explain the intricate trend
distribution in the Selin Co basin, where large lakes such as Selin Co were
expanding, whereas smaller adjacent lakes showed an opposite decreasing
trend, e.g., Urru Co (31.70<inline-formula><mml:math id="M306" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> N, 88.00<inline-formula><mml:math id="M307" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> E), Lake Co Ngoin (31.60<inline-formula><mml:math id="M308" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> N, 88.77<inline-formula><mml:math id="M309" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> E),
and Goren Co (31.10<inline-formula><mml:math id="M310" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> N, 88.37<inline-formula><mml:math id="M311" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> E). In fact, we found that the decreasing
trends in some small lakes like Goren Co were not detected in Yao et al. (2018b), which is likely due to the lower temporal resolution as shown in
Fig. 13. The three shrinking lakes are located in
the upstream region and feed Selin Co through two small rivers. One of the
rivers links lakes Goren Co, Urru Co, and Selin Co, whereas the other links
lakes Co Ngoin and Selin Co.</p>
      <?pagebreak page1620?><p id="d1e7999">A possible explanation of the disparity of changes in lake water storage in
the Selin Co basin could be the principle of minimum potential energy. If we
simplify the basin with the tank model and take the upstream small lake as a
tank with a leaking hole, the storage of the small lake is mainly controlled
by the height of the leaking hole. Given that surface water of the small
lake increased, most of the increased water would flow into the large lake
(a lower tank), and the outflow discharge of the small lake at higher
elevations would increase accordingly. The height of the leaking hole would
decline (erosion) so as to increase the overflow capacity, which eventually
results in the decrease in small lake storage. Another possible situation is
that the height of the leaking hole remains the same and the water surface
height of the small lake increases, but this situation is not consistent
with the minimum potential energy principle, as more water potential energy
is stored in the small lake. This phenomenon shows that river-lake
interactions may cause complex patterns of the regional surface water
distribution. Therefore, decreases in small lake water storage and increases
in water storage of Selin Co in the basin detected by our study seem
reasonable. Increases in small lake water storage in this basin reported in
some published studies may be associated with the sparse sampling of lake
water levels.</p>

      <?xmltex \floatpos{p}?><fig id="Ch1.F15" specific-use="star"><?xmltex \currentcnt{15}?><label>Figure 15</label><caption><p id="d1e8004">Similarities and differences between water level time series from
the LEGOS Hydroweb database (Crétaux et al., 2011a) and our study.
<bold>(a)</bold> Taro Co (31.14<inline-formula><mml:math id="M312" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> N, 84.12<inline-formula><mml:math id="M313" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> E). <bold>(b)</bold> Zhari Namco (30.93<inline-formula><mml:math id="M314" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> N, 85.61<inline-formula><mml:math id="M315" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> E).  <bold>(c)</bold> Ngoring Lake (34.90<inline-formula><mml:math id="M316" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> N, 97.70<inline-formula><mml:math id="M317" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> E). Shading areas highlight the differences
between the two data sets. LSH represents lake surface height.</p></caption>
          <?xmltex \igopts{width=398.338583pt}?><graphic xlink:href="https://essd.copernicus.org/articles/11/1603/2019/essd-11-1603-2019-f15.png"/>

        </fig>

      <?pagebreak page1622?><p id="d1e8077">We averaged the total lake water storage change in each season to generate
time series shown in Fig. 14a. The overall
storage change in the 52 lakes is <inline-formula><mml:math id="M318" display="inline"><mml:mrow><mml:mn mathvariant="normal">100.1</mml:mn><mml:mo>±</mml:mo><mml:mn mathvariant="normal">5.7</mml:mn></mml:mrow></mml:math></inline-formula> km<inline-formula><mml:math id="M319" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">3</mml:mn></mml:msup></mml:math></inline-formula>. The total lake
water storage was increasing rapidly during the first 12 years but became
relatively stable since 2012. Intra-annual variation in the TP lakes can
also be investigated using the densified lake water level time series
generated by our study. We removed the linear trend (sometimes there were
multiple linear trends for a lake in different periods, which were removed
in a stepwise fashion) and calculated the mean monthly water level anomaly
for each lake over the study period. Then the intra-annual water level
change was represented by the difference between the maximum and minimum
values of the monthly water level anomaly. The histogram of the intra-annual
water level change in Fig. 14b shows
that most of the TP lakes have water level variations ranging from 0.3 to 0.75 m in a year on average. Similar work was performed by Lei et al. (2017), but only a small number of lakes were investigated in their study.</p>
</sec>
<sec id="Ch1.S5.SS2">
  <label>5.2</label><title>Quality assessment of similar data products</title>
      <p id="d1e8109">Some obvious discrepancies between the two data sources can be noticed,
e.g., water levels of Taro Co. Both Hydroweb data and our estimation used
ICESat and CryoSat-2 data. The difference lies in the fact that our
CryoSat-2 product was more updated with a longer time span but Hydroweb used
an additional altimetry satellite SARAL. Because the systematic biases of
both products were removed, it is possible that we chose different baselines
that resulted in the overall shift as shown in Fig. 15a. For
instance, we may use different sets of ellipsoid and geoid models. In
addition to the overall shift, some time-dependent discrepancy can be found
in Fig. 15, e.g., periods highlighted by shading areas.</p>
      <p id="d1e8112">The black curve shows the Landsat-derived water level we derived, which is a
critical reference for connecting two different altimetry data time series
without an overlap period. The Landsat-derived water level shows that the
last two samples of ICESat data should not be lower than the first few
samples of the CryoSat-2/SARAL data (see the dashed boxes). However, it is
apparent that Hydroweb data display a reverse relationship, showing that the
last two ICESat measurements are smaller than the first few CryoSat/SARAL
measurements. It is likely due to an unremoved systematic bias between
ICESat and CryoSat/SARAL time series from Hydroweb data in Taro Co.</p>
      <p id="d1e8115">Even though the Landsat-derived water levels were generated by linearly
fitting the lake shoreline positions with altimetry data, the relative
magnitude of water levels during different periods should not be largely
affected by the fitting parameters, e.g., if Landsat-derived water levels
show that <inline-formula><mml:math id="M320" display="inline"><mml:mrow><mml:msub><mml:mi>H</mml:mi><mml:mtext>a</mml:mtext></mml:msub><mml:mo>&gt;</mml:mo><mml:mo>=</mml:mo><mml:msub><mml:mi>H</mml:mi><mml:mtext>b</mml:mtext></mml:msub></mml:mrow></mml:math></inline-formula>, where <inline-formula><mml:math id="M321" display="inline"><mml:mrow><mml:msub><mml:mi>H</mml:mi><mml:mtext>a</mml:mtext></mml:msub></mml:mrow></mml:math></inline-formula> (<inline-formula><mml:math id="M322" display="inline"><mml:mrow><mml:msub><mml:mi>H</mml:mi><mml:mtext>b</mml:mtext></mml:msub></mml:mrow></mml:math></inline-formula>) means water
levels acquired in period A (B), the <inline-formula><mml:math id="M323" display="inline"><mml:mrow><mml:msub><mml:mi>H</mml:mi><mml:mtext>a</mml:mtext></mml:msub><mml:mo>&gt;</mml:mo><mml:mo>=</mml:mo><mml:msub><mml:mi>H</mml:mi><mml:mtext>b</mml:mtext></mml:msub></mml:mrow></mml:math></inline-formula>
relationship would not change with the fitting parameters used to generate
the Landsat-derived water levels. This is the main reason for us to use
Landsat-derived water levels as reference. Therefore, Hydroweb data may
overestimate the increasing trends in the water levels of Taro Co as their
ICESat data are <inline-formula><mml:math id="M324" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 0.3 m lower than the SARAL/CryoSat data. A
similar issue can be observed in Zhari Namco and Ngoring lakes shown in
Fig. 15b–c, and the explanation is similar to that of Taro Co. This
problem may also exist in some similar studies when multisource altimetry
data without overlap periods were used.</p>

      <?xmltex \floatpos{t}?><fig id="Ch1.F16" specific-use="star"><?xmltex \currentcnt{16}?><label>Figure 16</label><caption><p id="d1e8190">Lake water level (left <inline-formula><mml:math id="M325" display="inline"><mml:mi>y</mml:mi></mml:math></inline-formula> axis) estimates from our approach for six
TP lakes. Black lines represent optical data and red dots represent
altimetry data. LSH represents lake surface height.</p></caption>
          <?xmltex \igopts{width=398.338583pt}?><graphic xlink:href="https://essd.copernicus.org/articles/11/1603/2019/essd-11-1603-2019-f16.png"/>

        </fig>

      <?xmltex \floatpos{t}?><fig id="Ch1.F17" specific-use="star"><?xmltex \currentcnt{17}?><label>Figure 17</label><caption><p id="d1e8208"><bold>(a)</bold> Lake storage changes in Lake Zhuonai, Lake Kusai, and Lake
Salt corresponding to the outburst event in September 2011 and <bold>(b)</bold> storage changes
in relevant lakes during the outburst event (a magnified plot of the shading
area in <bold>a</bold>).</p></caption>
          <?xmltex \igopts{width=398.338583pt}?><graphic xlink:href="https://essd.copernicus.org/articles/11/1603/2019/essd-11-1603-2019-f17.png"/>

        </fig>

      <?xmltex \floatpos{t}?><fig id="Ch1.F18" specific-use="star"><?xmltex \currentcnt{18}?><label>Figure 18</label><caption><p id="d1e8227"><bold>(a)</bold> Height variations in the outlet of Lake Kusai estimated from
Landsat 8 OLI images indicate that the overflow would occur when the water
level increased from 4483.9 to 4484.1 m. <bold>(b)</bold> Google Earth© image
before the outburst of Lake Zhuonai (December 2010). <bold>(c)</bold> Google Earth© image
after the outburst event (December 2013).</p></caption>
          <?xmltex \igopts{width=341.433071pt}?><graphic xlink:href="https://essd.copernicus.org/articles/11/1603/2019/essd-11-1603-2019-f18.png"/>

        </fig>

      <?xmltex \floatpos{t}?><fig id="Ch1.F19" specific-use="star"><?xmltex \currentcnt{19}?><label>Figure 19</label><caption><p id="d1e8246">Changes in the water level of Lake Kusai after receiving the
outburst flood from Lake Zhuonai. Stage 1 was used to determine the range of
parameter C in Eq. (13). Stage 2 was used to compared the simulated lake
outflow from Kusai Lake based on Eq. (13) with the water gain estimate from
remote sensing of Lake Salt downstream during the same period; and <bold>(b)</bold> changes in water storage of Lake Salt derived from remote sensing using our developed method. There was 0.19 km<inline-formula><mml:math id="M326" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">3</mml:mn></mml:msup></mml:math></inline-formula> of water gained in stage 2, which was comparable to the outflow estimate of Lake Kusai (0.22 km<inline-formula><mml:math id="M327" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">3</mml:mn></mml:msup></mml:math></inline-formula>) based
on Eq. (13).</p></caption>
          <?xmltex \igopts{width=455.244094pt}?><graphic xlink:href="https://essd.copernicus.org/articles/11/1603/2019/essd-11-1603-2019-f19.png"/>

        </fig>

      <p id="d1e8277">As shown in Fig. 16, optical data can be less noisy than altimetry data in
certain lakes and significantly improve the continuity of lake level and
storage change monitoring. In addition, a more apparent seasonality in lake
level change can be seen from the generated lake level time series. These
advantages would largely benefit a better understanding of responses of TP
lakes to climate change and facilitate hydrologic modeling of lake basins,
regional water balance analysis, and even hydrodynamic analysis of lake
water bodies.</p>
</sec>
<sec id="Ch1.S5.SS3">
  <label>5.3</label><title>Lake overflow flood monitoring</title>
      <p id="d1e8288">As mentioned earlier in Sect. 5.1, Lake Kusai experienced an abrupt
expansion in 2011, resulting from the dike break of an upstream lake (Hwang
et al., 2019; Liu et al., 2016; Xiaojun et al., 2012), named Lake Zhuonai
(35.54<inline-formula><mml:math id="M328" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> N, 91.93<inline-formula><mml:math id="M329" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> E). The outburst of Lake Zhuonai occurred on 14 September
(Liu et al., 2016), with <inline-formula><mml:math id="M330" display="inline"><mml:mrow><mml:mn mathvariant="normal">2.47</mml:mn><mml:mo>±</mml:mo><mml:mn mathvariant="normal">0.06</mml:mn></mml:mrow></mml:math></inline-formula> km<inline-formula><mml:math id="M331" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">3</mml:mn></mml:msup></mml:math></inline-formula> of water leaking
into the Kusai River (as shown in Fig. 17b), the main inflow of Lake
Kusai. The water level of Lake Kusai increased by up to <inline-formula><mml:math id="M332" display="inline"><mml:mrow><mml:mn mathvariant="normal">7.9</mml:mn><mml:mo>±</mml:mo><mml:mn mathvariant="normal">0.5</mml:mn></mml:mrow></mml:math></inline-formula> m
within 20 days (from 11 September to 1 October in 2011) based on Jason-2 data and then started to drop as water overflowed from the southeast corner into Lake
Haidingnuoer (35.55<inline-formula><mml:math id="M333" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> N, 93.16<inline-formula><mml:math id="M334" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> E) and Lake Salt (35.52<inline-formula><mml:math id="M335" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> N, 93.40<inline-formula><mml:math id="M336" display="inline"><mml:msup><mml:mi/><mml:mo>∘</mml:mo></mml:msup></mml:math></inline-formula> E). Lake Salt,
the lowest part of the basin close to the basin boundary, has gained
<inline-formula><mml:math id="M337" display="inline"><mml:mrow><mml:mn mathvariant="normal">3.0</mml:mn><mml:mo>±</mml:mo><mml:mn mathvariant="normal">0.1</mml:mn></mml:mrow></mml:math></inline-formula> km<inline-formula><mml:math id="M338" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">3</mml:mn></mml:msup></mml:math></inline-formula> of water since 2011 and has become a critical threat to
the surrounding residents and railway <inline-formula><mml:math id="M339" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 10 km southeast to the
boundary. Note that there are few satellite altimetry data available for
Lake Salt except several CryoSat-2 observations, where Landsat-derived water
levels can provide a near-real-time monitoring of changes in lake water
level and storage that are crucial to flood early warning and risk
management.</p>
      <p id="d1e8408">Aided by the high-temporal-resolution lake water level series, it was
possible to estimate the height of the outlet of Lake Kusai, an important
parameter for overflow estimation. The overflow of Lake Kusai can help
predict the water level rise in Lake Salt and even serve as an indicator of
flood forecast, as Jason-3 data with a 10-day revisit cycle are now
available on Lake Kusai. Several pairs of Landsat 8 OLI images and lake
water levels for the same period were compared to provide a range of
possible outlet heights, which are likely to be 4483.9 to 4484.1 m, as
shown in Fig. 18a. Then we measured the mean width of the outlet from
high-resolution optical images provided by Planet Explorer (Planet, 2017),
which is relatively stable in Dec 2011 at <inline-formula><mml:math id="M340" display="inline"><mml:mrow><mml:mn mathvariant="normal">31.5</mml:mn><mml:mo>±</mml:mo><mml:mn mathvariant="normal">2.3</mml:mn></mml:mrow></mml:math></inline-formula> m in recent years.
Given lake water levels and the outlet height and width, an estimation of overflow
can be made using the broad crest weir formula:
            <disp-formula id="Ch1.E13" content-type="numbered"><label>13</label><mml:math id="M341" display="block"><mml:mrow><mml:mi>Q</mml:mi><mml:mo>=</mml:mo><mml:mi>C</mml:mi><mml:mo>⋅</mml:mo><mml:mi>b</mml:mi><mml:mo>⋅</mml:mo><mml:msup><mml:mi>H</mml:mi><mml:mn mathvariant="normal">1.5</mml:mn></mml:msup><mml:msqrt><mml:mrow><mml:mn mathvariant="normal">2</mml:mn><mml:mi>g</mml:mi></mml:mrow></mml:msqrt><mml:mo>,</mml:mo></mml:mrow></mml:math></disp-formula>
          where <inline-formula><mml:math id="M342" display="inline"><mml:mi>C</mml:mi></mml:math></inline-formula> is a parameter mainly reflecting geometric characteristics of the weir that mainly varies from 0.3 to 0.4, <inline-formula><mml:math id="M343" display="inline"><mml:mi>b</mml:mi></mml:math></inline-formula> is the width of the weir, <inline-formula><mml:math id="M344" display="inline"><mml:mi>H</mml:mi></mml:math></inline-formula> is the water head with respect to the top of the weir, and <inline-formula><mml:math id="M345" display="inline"><mml:mi>g</mml:mi></mml:math></inline-formula> is the acceleration of gravity.</p>
      <p id="d1e8483">We determined <inline-formula><mml:math id="M346" display="inline"><mml:mi>C</mml:mi></mml:math></inline-formula> (<inline-formula><mml:math id="M347" display="inline"><mml:mo lspace="0mm">∼</mml:mo></mml:math></inline-formula> 0.3) by using stage 1 shown in Fig. 19 as
a calibration period. Details can be found in the supplementary file. Then
we applied this result to stage 2 shown in Fig. 19 to estimate the total
overflow from Lake Kusai and compared the overflow with total water gain in
stage 2<?pagebreak page1623?> in Lake Salt. Since Lake Salt mainly relied on the replenishment of
Lake Kusai during that period, with little precipitation input and
negligible glacier meltwater in winter, the outflow of Lake Kusai can be
comparable with the water gain in Lake Salt derived from remote sensing,
though there was a small amount of evaporation loss. This relationship can
provide a straightforward validation of our developed method. However, it
was not available in stage 1, because the outflow of Lake Kusai first
replenished Lake Haidingnuoer until the latter began overflowing. Results
based on Eq. (13) indicate that the total outflow from Lake Kusai in stage 2
ranged from 0.21 to 0.22 km<inline-formula><mml:math id="M348" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">3</mml:mn></mml:msup></mml:math></inline-formula>, whereas the water gain in Lake Salt from
remote sensing was <inline-formula><mml:math id="M349" display="inline"><mml:mrow><mml:mn mathvariant="normal">0.19</mml:mn><mml:mo>±</mml:mo><mml:mn mathvariant="normal">0.01</mml:mn></mml:mrow></mml:math></inline-formula> km<inline-formula><mml:math id="M350" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">3</mml:mn></mml:msup></mml:math></inline-formula>. This indicates that our high-temporal-resolution lake water level time series are valuable in monitoring
and predicting lake outflow flooding that is crucial for the safety of
downstream residents and infrastructure.</p>
</sec>
</sec>
<sec id="Ch1.S6">
  <label>6</label><title>Data availability</title>
      <p id="d1e8539">The derived TP lake water levels, hypsometric curves, and water storage
changes are archived and available at <ext-link xlink:href="https://doi.org/10.1594/PANGAEA.898411" ext-link-type="DOI">10.1594/PANGAEA.898411</ext-link> (Li et al., 2019).</p>
</sec>
<sec id="Ch1.S7" sec-type="conclusions">
  <label>7</label><title>Conclusion</title>
      <p id="d1e8553">In this study, we develop high-temporal-resolution (i.e., weekly to monthly
timescales) water levels and storage change data sets for 52 large lakes on
the TP during 2000–2017 by combining multiple altimetric missions and
optical remote sensing images. Generated from lake shoreline positions and
regression analysis with altimetry data, the Landsat-derived water level
serves as a unique reference covering the entire study period, enabling a
more consistent merging of multisource altimetry time series.<?pagebreak page1624?> Multisource
altimetry water levels are first extracted separately from spaceborne
altimetry products and then combined into a longer and denser altimetry
water level time series with systematic biases well removed using
Landsat-derived water levels as reference. The combined altimetry and
Landsat-derived water levels increase the overall sampling frequency to
submonthly regardless of the lake size.</p>
      <p id="d1e8556">By comparison with a widely used LEGOS Hydroweb data set, we show that
without Landsat-derived water levels as a reference there may be a
remaining bias in the combined<?pagebreak page1625?> altimetry water levels in certain lakes. Our
study has considerably improved the temporal resolution of the monitoring of
lake water level and storage changes in the TP. For most lakes examined in
the published studies, to our best knowledge, the estimates from our study
provide the observations of the highest temporal resolution that can better
reveal the interannual and intra-annual variability and trends in lake water
level and storage, even in some relatively small lakes whose annual trends
may, however, be incorrectly estimated by sparse sampling of lake water
levels. The developed data sets can also facilitate the monitoring of some
rapidly expanding lakes with overflow risks and provide important
information on flood prediction and early warning.</p>
      <p id="d1e8559">We evaluate the uncertainty in the Landsat-derived water levels by field
experiments and rigorous uncertainty analysis. Both methods are consistent
that the magnitude of the uncertainty is <inline-formula><mml:math id="M351" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 0.1 m, which
suggests that Landsat-derived water levels are often more efficient and less
noisy than altimetry data when the altimeter footprints on the lake surface
are insufficient, especially for small lakes. Based on our estimates, 52
large TP lakes accounting for <inline-formula><mml:math id="M352" display="inline"><mml:mo>∼</mml:mo></mml:math></inline-formula> 60 % of the total TP lake
area have gained <inline-formula><mml:math id="M353" display="inline"><mml:mrow><mml:mn mathvariant="normal">100.1</mml:mn><mml:mo>±</mml:mo><mml:mn mathvariant="normal">5.7</mml:mn></mml:mrow></mml:math></inline-formula> km<inline-formula><mml:math id="M354" display="inline"><mml:msup><mml:mi/><mml:mn mathvariant="normal">3</mml:mn></mml:msup></mml:math></inline-formula> of water during the past 18 years.
Lakes in the endorheic basin on the TP have mostly expanded. The complex
spatial pattern of lake storage changes in the Selin Co basin was quantified
and a possible explanation was proposed in this study. Note that the quality
of the Landsat-derived water levels before 2002 may not be as good as those
after 2002, because no altimetry data before 2002 are used in this study.
Extrapolation of the relationship between lake shoreline positions and water
levels may not be stable if the water level during 2000–2001 was much lower
or higher than those from 2002 to 2017. Discussions on how the extrapolation
may affect the data quality can be found in the Supplement.</p>
</sec>

      
      </body>
    <back><app-group>
        <supplementary-material position="anchor"><p id="d1e8596">The supplement related to this article is available online at: <inline-supplementary-material xlink:href="https://doi.org/10.5194/essd-11-1603-2019-supplement" xlink:title="pdf">https://doi.org/10.5194/essd-11-1603-2019-supplement</inline-supplementary-material>.</p></supplementary-material>
        </app-group><notes notes-type="authorcontribution"><title>Author contributions</title>

      <p id="d1e8607">LD and LX designed the research. LX, LD, HQ, and ZF developed the approaches
and data sets. LX, HQ, HP, and LD carried out the field experiment. LX, LD,
and YW contributed to the analysis of results and writing of the paper.</p>
  </notes><notes notes-type="competinginterests"><title>Competing interests</title>

      <p id="d1e8613">The authors declare that they have no conflict of interest.</p>
  </notes><ack><title>Acknowledgements</title><p id="d1e8619">Mingda Du's assistance in the field experiments, discussion on the use of satellite
altimetry data for lake monitoring with Gang Qiao from Tongji
University, and efforts in improving and archiving the data sets made by
Daniela Ransby from PANGAEA data publisher are acknowledged here.</p></ack><notes notes-type="financialsupport"><title>Financial support</title>

      <p id="d1e8624">This research has been supported by the National Natural Science Foundation of China (grant nos. 91547210 and 51722903) and the National Key Research and
Development Program of China (grant no. 2017YFC0405801).</p>
  </notes><notes notes-type="reviewstatement"><title>Review statement</title>

      <p id="d1e8630">This paper was edited by Birgit Heim and reviewed by two anonymous referees.</p>
  </notes><ref-list>
    <title>References</title>

      <ref id="bib1.bib1"><label>1</label><?label 1?><mixed-citation>
Benveniste, J., Baker, S., Bombaci, O., Zeli, C., Venditti, P., Zanife, O.,
Soussi, B., Dumont, J., Stum, J., and Milagro-Perezin, M. P.: Envisat
RA-2/MWR Product Handbook, Eur. Space Agency, Frascati, Italy, 2002.</mixed-citation></ref>
      <ref id="bib1.bib2"><label>2</label><?label 1?><mixed-citation>
Birkett, C. M. and Beckley, B.: Investigating the performance of the
Jason-2/OSTM radar altimeter over lakes and reservoirs, Mar. Geod., 33,
204–238, 2010.</mixed-citation></ref>
      <ref id="bib1.bib3"><label>3</label><?label 1?><mixed-citation>Bouzinac, C.: CryoSat product handbook, ESA, UCL, available at: <uri>https://earth.esa.int/documents/10174/125272/CryoSat_Product_Handbook</uri> (last access: 17 October 2019), 2012.</mixed-citation></ref>
      <ref id="bib1.bib4"><label>4</label><?label 1?><mixed-citation>
Cheng, K.-C., Kuo, C.-Y., Tseng, H.-Z., Yi, Y., and Shum, C.: Lake surface
height calibration of Jason-1 and Jason-2 over the Great Lakes, Mar.
Geod., 33, 186–203, 2010.</mixed-citation></ref>
      <ref id="bib1.bib5"><label>5</label><?label 1?><mixed-citation>
Crétaux, J.-F., Jelinski, W., Calmant, S., Kouraev, A., Vuglinski, V.,
Bergé-Nguyen, M., Gennero, M.-C., Nino, F., Del Rio, R. A., and
Cazenave, A.: SOLS: A lake database to monitor in the Near Real Time water
level and storage variations from remote sensing data, Adv. Space
Res., 47, 1497–1507, 2011a.</mixed-citation></ref>
      <ref id="bib1.bib6"><label>6</label><?label 1?><mixed-citation>
Crétaux, J., Jelinski, W., Calmant, S., Kouraev, A., Vuglinski, V.,
Bergé-Nguyen, M., Nino, G., Del Rio, R., Cazenave, A., and Maisongrande,
P.: Hydrolare/Hydroweb: A lake database to monitor in the Near Real Time
water level and storage variations from remote sensing data, Adv.
Space Res., 47, 1497–1507, 2011b.</mixed-citation></ref>
      <ref id="bib1.bib7"><label>7</label><?label 1?><mixed-citation>
Crétaux, J.-F., Abarca-del-Río, R., Berge-Nguyen, M., Arsen, A.,
Drolon, V., Clos, G., and Maisongrande, P.: Lake volume monitoring from
space, Surv. Geophys., 37, 269–305, 2016.</mixed-citation></ref>
      <ref id="bib1.bib8"><label>8</label><?label 1?><mixed-citation>
Field, C. B., Barros, V. R., Dokken, D., Mach, K., Mastrandrea, M., Bilir,
T., Chatterjee, M., Ebi, K., Estrada, Y., and Genova, R.: IPCC, 2014:
Climate change 2014: Impacts, adaptation, and vulnerability – Part A: Global
and sectoral aspects. Contribution of working group II to the fifth
assessment report of the intergovernmental panel on climate change,
Cambridge University Press, Cambridge, UK, New York, NY, USA,
2014.</mixed-citation></ref>
      <ref id="bib1.bib9"><label>9</label><?label 1?><mixed-citation>
Frappart, F., Seyler, F., Martinez, J.-M., León, J. G., and Cazenave,
A.: Floodplain water storage in the Negro River basin estimated from
microwave remote sensing of inundation area and water levels, Remote Sens. Environ., 99, 387–399, 2005.</mixed-citation></ref>
      <ref id="bib1.bib10"><label>10</label><?label 1?><mixed-citation>
Gorelick, N., Hancher, M., Dixon, M., Ilyushchenko, S., Thau, D., and Moore,
R.: Google Earth Engine: Planetary-scale geospatial analysis for everyone,
Remote Sens. Environ., 202, 18–27, 2017.</mixed-citation></ref>
      <ref id="bib1.bib11"><label>11</label><?label 1?><mixed-citation>
Guo, J., Chang, X., Gao, Y., Sun, J., and Hwang, C.: Lake level variations
monitored with satellite altimetry waveform retracking, IEEE J. Sel. Top. Appl., 2, 80–86,
2009.</mixed-citation></ref>
      <ref id="bib1.bib12"><label>12</label><?label 1?><mixed-citation>
Huang, Q., Long, D., Du, M., Zeng, C., Qiao, G., Li, X., Hou, A., and Hong,
Y.: Discharge estimation in high-mountain regions with improved methods
using multisource remote sensing: A case study of the Upper Brahmaputra
River, Remote Sens. Environ., 219, 115–134, 2018.</mixed-citation></ref>
      <ref id="bib1.bib13"><label>13</label><?label 1?><mixed-citation>
Hwang, C., Cheng, Y.-S., Yang, W.-H., Zhang, G., Huang, Y.-R., Shen, W.-B.,
and Pan, Y.: Lake level changes in the Tibetan Plateau from Cryosat-2,
SARAL, ICESat, and Jason-2 altimeters, Terr. Atmos. Ocean Sci., 30, 1–18, 2019.</mixed-citation></ref>
      <ref id="bib1.bib14"><label>14</label><?label 1?><mixed-citation>
Jain, M., Andersen, O. B., Dall, J., and Stenseng, L.: Sea surface height
determination in the Arctic using Cryosat-2 SAR data from primary peak
empirical retrackers, Adv. Space Res., 55, 40–50, 2015.</mixed-citation></ref>
      <ref id="bib1.bib15"><label>15</label><?label 1?><mixed-citation>Jarvis, A., Reuter, H. I., Nelson, A., and Guevara, E.: Hole-filled SRTM for
the globe Version 4, available from the CGIAR-CSI SRTM 90 m Database,
available at: <uri>http://srtm.csi.cgiar.org</uri> (last access: 17 October 2019), 15, 25–54, 2008.</mixed-citation></ref>
      <ref id="bib1.bib16"><label>16</label><?label 1?><mixed-citation>
Jiang, L., Nielsen, K., Andersen, O. B., and Bauer-Gottwein, P.: Monitoring
recent lake level variations on the Tibetan Plateau using CryoSat-2 SARIn
mode data, J. Hydrol., 544, 109–124, 2017.</mixed-citation></ref>
      <ref id="bib1.bib17"><label>17</label><?label 1?><mixed-citation>
Kittler, J. and Illingworth, J.: On threshold selection using clustering
criteria, IEEE T. Syst. Man. Cyb., September–October 1985, SMC-15, 652–655,
1985.</mixed-citation></ref>
      <ref id="bib1.bib18"><label>18</label><?label 1?><mixed-citation>
Lehner, B. and Döll, P.: Development and validation of a global database
of lakes, reservoirs and wetlands, J. Hydrol., 296, 1–22, 2004.</mixed-citation></ref>
      <ref id="bib1.bib19"><label>19</label><?label 1?><mixed-citation>
Lei, Y., Yao, T., Yang, K., Sheng, Y., Kleinherenbrink, M., Yi, S., Bird, B.
W., Zhang, X., Zhu, L., and Zhang, G.: Lake seasonality across the Tibetan
Plateau and their varying relationship with regional mass changes and local
hydrology, Geophys. Res. Lett., 44, 892–900, 2017.</mixed-citation></ref>
      <ref id="bib1.bib20"><label>20</label><?label 1?><mixed-citation>
Li, B., Zhang, J., Yu, Z., Liang, Z., Chen, L., and Acharya, K.: Climate
change driven water budget dynamics of a Tibetan inland lake, Global
Planet. Change, 150, 70–80, 2017a.</mixed-citation></ref>
      <ref id="bib1.bib21"><label>21</label><?label 1?><mixed-citation>Li, H. W., Qiao, G., Wu, Y. J., Cao, Y. J., and Mi, H.: Water Level Monitoring On Tibetan Lakes Based On Icesat And Envisat Data Series, Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-2/W7, 1529–1533, <ext-link xlink:href="https://doi.org/10.5194/isprs-archives-XLII-2-W7-1529-2017" ext-link-type="DOI">10.5194/isprs-archives-XLII-2-W7-1529-2017</ext-link>, 2017b.</mixed-citation></ref>
      <ref id="bib1.bib22"><label>22</label><?label 1?><mixed-citation>
Li, L., Liu, Q., Zhang, Y., Liu, L., Ding, M., and Changjun, G. U.: Spatial
distribution and variation of precipitation in the Qiangtang Plateau,
Geogr. Res., 36, 2047–2060, 2017c.</mixed-citation></ref>
      <ref id="bib1.bib23"><label>23</label><?label 1?><mixed-citation>Li, X., Long, D., Huang, Q., Han, P., Zhao, F., and Wada, Y.: A high
temporal resolution lake data set from multisource altimetric missions and
Landsat archives of water level and storage changes on the Tibetan Plateau
during 2000–2017,
PANGAEA, <ext-link xlink:href="https://doi.org/10.1594/PANGAEA.898411" ext-link-type="DOI">10.1594/PANGAEA.898411</ext-link>, 2019.</mixed-citation></ref>
      <ref id="bib1.bib24"><label>24</label><?label 1?><mixed-citation>
Liu, B., Lin, L. I., Yue, D. U., Liang, T., Duan, S., Hou, F., and Ren, J.:
Causes of the outburst of Zonag Lake in Hoh Xil, Tibetan Plateau,and its
impact on surrounding environment, J. Glaciol. Geocryol., 38, 305–311, 2016.</mixed-citation></ref>
      <ref id="bib1.bib25"><label>25</label><?label 1?><mixed-citation>
Markham, B. L., Storey, J. C., Williams, D. L., and Irons, J. R.: Landsat
sensor performance: history and current status, IEEE T. Geosci. Remote, 42, 2691–2694, 2004.</mixed-citation></ref>
      <ref id="bib1.bib26"><label>26</label><?label 1?><mixed-citation>
McFeeters, S. K.: The use of the Normalized Difference Water Index (NDWI) in
the delineation of open water features, Int. J. Remote
Sens., 17, 1425–1432, 1996.</mixed-citation></ref>
      <ref id="bib1.bib27"><label>27</label><?label 1?><mixed-citation>Messager, M. L., Lehner, B., Grill, G., Nedeva, I., and Schmitt, O.:
Estimating the volume and age of water stored in global lakes using a
geo-statistical approach, Nat, Commun., 7, 13603, <ext-link xlink:href="https://doi.org/10.1038/ncomms13603" ext-link-type="DOI">10.1038/ncomms13603</ext-link>, 2016.</mixed-citation></ref>
      <ref id="bib1.bib28"><label>28</label><?label 1?><mixed-citation>Neckel, N., Kropáček, J., Bolch, T., and Hochschild, V.: Glacier
mass changes on the Tibetan Plateau 2003–2009 derived from ICESat laser
altimetry measurements, Environ. Res. Lett., 9, 014009, <ext-link xlink:href="https://doi.org/10.1088/1748-9326/9/1/014009" ext-link-type="DOI">10.1088/1748-9326/9/1/014009</ext-link>, 2014.</mixed-citation></ref>
      <ref id="bib1.bib29"><label>29</label><?label 1?><mixed-citation>Nielsen, K., Stenseng, L., Andersen, O. B., and Knudsen, P.: The performance
and potentials of the CryoSat-2 SAR and SARIn modes for lake level
estimation, Water, 9, w9060374, <ext-link xlink:href="https://doi.org/10.3390/w9060374" ext-link-type="DOI">10.3390/w9060374</ext-link>, 2017.</mixed-citation></ref>
      <ref id="bib1.bib30"><label>30</label><?label 1?><mixed-citation>
Otsu, N.: A threshold selection method from gray-level histograms, IEEE T. Syst. Man. Cyb., 9, 62–66, 1979.</mixed-citation></ref>
      <?pagebreak page1627?><ref id="bib1.bib31"><label>31</label><?label 1?><mixed-citation>
Pekel, J.-F., Cottam, A., Gorelick, N., and Belward, A. S.: High-resolution
mapping of global surface water and its long-term changes, Nature, 540, 418–422,
2016.</mixed-citation></ref>
      <ref id="bib1.bib32"><label>32</label><?label 1?><mixed-citation>Planet: Planet Application Program Interface: In Space for Life on Earth,
San Francisco, CA, available at: <uri>https://www.planet.com</uri> (last access: 17 October 2019), 2017.</mixed-citation></ref>
      <ref id="bib1.bib33"><label>33</label><?label 1?><mixed-citation>Sørensen, L. S., Simonsen, S. B., Nielsen, K., Lucas-Picher, P., Spada, G., Adalgeirsdottir, G., Forsberg, R., and Hvidberg, C. S.: Mass balance of the Greenland ice sheet (2003–2008) from ICESat data – the impact of interpolation, sampling and firn density, The Cryosphere, 5, 173–186, <ext-link xlink:href="https://doi.org/10.5194/tc-5-173-2011" ext-link-type="DOI">10.5194/tc-5-173-2011</ext-link>, 2011.</mixed-citation></ref>
      <ref id="bib1.bib34"><label>34</label><?label 1?><mixed-citation>
Song, C., Huang, B., and Ke, L.: Modeling and analysis of lake water storage
changes on the Tibetan Plateau using multi-mission satellite data, Remote
Sens. Environ., 135, 25–35, 2013.</mixed-citation></ref>
      <ref id="bib1.bib35"><label>35</label><?label 1?><mixed-citation>
Tournadre, J.: Validation of Jason and Envisat altimeter dual frequency rain
flags, Mar. Geod., 27, 153–169, 2004.</mixed-citation></ref>
      <ref id="bib1.bib36"><label>36</label><?label 1?><mixed-citation>Wan, W., Long, D., Hong, Y., Ma, Y., Yuan, Y., Xiao, P., Duan, H., Han, Z.,
and Gu, X.: A lake data set for the Tibetan Plateau from the 1960s, 2005,
and 2014, Scientific data, 3, 160039, <ext-link xlink:href="https://doi.org/10.1038/sdata.2016.39" ext-link-type="DOI">10.1038/sdata.2016.39</ext-link>, 2016.</mixed-citation></ref>
      <ref id="bib1.bib37"><label>37</label><?label 1?><mixed-citation>
Wang, J., Song, C., Reager, J. T., Yao, F., Famiglietti, J. S., Sheng, Y.,
MacDonald, G. M., Brun, F., Schmied, H. M., and Marston, R. A.: Recent
global decline in endorheic basin water storages, Nat. Geosci., 11,
926–932, 2018.</mixed-citation></ref>
      <ref id="bib1.bib38"><label>38</label><?label 1?><mixed-citation>
Xiaojun, Y., Shiyin, L., and Meiping, S.: Changes of Kusai Lake in Hoh Xil
region and causes of its water overflowing, Ac. Geogr. Sin., 67,
689–698, 2012.</mixed-citation></ref>
      <ref id="bib1.bib39"><label>39</label><?label 1?><mixed-citation>
Xu, H.: A study on information extraction of water body with the modified
normalized difference water index (MNDWI), J. Remote
Sens.-Beijing, 9, 589–595, 2005.</mixed-citation></ref>
      <ref id="bib1.bib40"><label>40</label><?label 1?><mixed-citation>Yao, F., Wang, J., Yang, K., Wang, C., Walter, B. A., and Crétaux,
J.-F.: High resolution data set of annual lake areas and water storage across the Inner Tibet, 2002–2015, PANGAEA, <ext-link xlink:href="https://doi.org/10.1594/PANGAEA.888706" ext-link-type="DOI">10.1594/PANGAEA.888706</ext-link>, 2018a.
</mixed-citation></ref><?xmltex \hack{\newpage}?>
      <ref id="bib1.bib41"><label>41</label><?label 1?><mixed-citation>Yao, F., Wang, J., Yang, K., Wang, C., Walter, B. A., and Crétaux,
J.-F.: Lake storage variation on the endorheic Tibetan Plateau and its
attribution to climate change since the new millennium, Environ.
Res. Lett., 13, 064011, <ext-link xlink:href="https://doi.org/10.1088/1748-9326/aab5d3" ext-link-type="DOI">10.1088/1748-9326/aab5d3</ext-link>, 2018b.</mixed-citation></ref>
      <ref id="bib1.bib42"><label>42</label><?label 1?><mixed-citation>
Yao, X., Liu, S., and Sun, M.: Changes of Kusai Lake in Hoh Xil region and
causes of its water overflowing, Ac. Geogr. Sin., 67, 689–698, 2012.</mixed-citation></ref>
      <ref id="bib1.bib43"><label>43</label><?label 1?><mixed-citation>
Yu, S., Liu, J., Xu, J., and Wang, H.: Evaporation and energy balance
estimates over a large inland lake in the Tibet-Himalaya, Environ.
Earth Sci., 64, 1169–1176, 2011.</mixed-citation></ref>
      <ref id="bib1.bib44"><label>44</label><?label 1?><mixed-citation>
Zhang, G., Xie, H., Kang, S., Yi, D., and Ackley, S. F.: Monitoring lake
level changes on the Tibetan Plateau using ICESat altimetry data
(2003–2009), Remote Sens. Environ., 115, 1733–1742, 2011.</mixed-citation></ref>
      <ref id="bib1.bib45"><label>45</label><?label 1?><mixed-citation>
Zhang, G., Yao, T., Xie, H., Kang, S., and Lei, Y.: Increased mass over the
Tibetan Plateau: from lakes or glaciers?, Geophys. Res. Lett., 40,
2125–2130, 2013.</mixed-citation></ref>
      <ref id="bib1.bib46"><label>46</label><?label 1?><mixed-citation>
Zhang, G., Yao, T., Shum, C., Yi, S., Yang, K., Xie, H., Feng, W., Bolch,
T., Wang, L., and Behrangi, A.: Lake volume and groundwater storage
variations in Tibetan Plateau's endorheic basin, Geophys. Res.
Lett., 44, 5550–5560, 2017a.</mixed-citation></ref>
      <ref id="bib1.bib47"><label>47</label><?label 1?><mixed-citation>
Zhang, G., Zheng, G., Gao, Y., Xiang, Y., Lei, Y., and Li, J.: Automated
water classification in the Tibetan plateau using Chinese GF-1 WFV data,
Photogramm. Eng. Remote Sens., 83, 509–519, 2017b.</mixed-citation></ref>
      <ref id="bib1.bib48"><label>48</label><?label 1?><mixed-citation>
Zhou, J., Wang, L., Zhang, Y., Guo, Y., Li, X., and Liu, W.: Exploring the
water storage changes in the largest lake (Selin Co) over the Tibetan
Plateau during 2003–2012 from a basin-wide hydrological modeling, Water
Resour. Res., 51, 8060–8086, 2015.</mixed-citation></ref>

  </ref-list></back>
    <!--<article-title-html>High-temporal-resolution water level and storage change data sets for lakes on the Tibetan Plateau during 2000–2017 using multiple altimetric missions and Landsat-derived lake shoreline positions</article-title-html>
<abstract-html><p>The Tibetan Plateau (TP), known as Asia's water tower, is
quite sensitive to climate change, which is reflected by changes in hydrologic state
variables such as lake water storage. Given the extremely limited ground
observations on the TP due to the harsh environment and complex terrain, we
exploited multiple altimetric missions and Landsat satellite data to create
high-temporal-resolution lake water level and storage change time series at
weekly to monthly timescales for 52 large lakes (50 lakes larger than 150&thinsp;km<sup>2</sup> and 2 lakes larger than 100&thinsp;km<sup>2</sup>) on the TP during 2000–2017. The data sets are available online at <a href="https://doi.org/10.1594/PANGAEA.898411" target="_blank">https://doi.org/10.1594/PANGAEA.898411</a> (Li et al., 2019). With
Landsat archives and altimetry data, we developed water levels from lake
shoreline positions (i.e., Landsat-derived water levels) that cover the
study period and serve as an ideal reference for merging multisource lake
water levels with systematic biases being removed. To validate the
Landsat-derived water levels, field experiments were carried out in two
typical lakes, and theoretical uncertainty analysis was performed based on
high-resolution optical images (0.8&thinsp;m) as well. The RMSE of the
Landsat-derived water levels is 0.11&thinsp;m compared with the in situ
measurements, consistent with the magnitude from theoretical analysis
(0.1–0.2&thinsp;m). The accuracy of the Landsat-derived water levels that can be
derived in relatively small lakes is comparable with most altimetry data.
The resulting merged Landsat-derived and altimetric lake water levels can
provide accurate information on multiyear and short-term monitoring of lake
water levels and storage changes on the TP, and critical information on lake overflow flood monitoring and prediction as the expansion of some TP lakes
becomes a serious threat to surrounding residents and infrastructure.</p></abstract-html>
<ref-html id="bib1.bib1"><label>1</label><mixed-citation>
Benveniste, J., Baker, S., Bombaci, O., Zeli, C., Venditti, P., Zanife, O.,
Soussi, B., Dumont, J., Stum, J., and Milagro-Perezin, M. P.: Envisat
RA-2/MWR Product Handbook, Eur. Space Agency, Frascati, Italy, 2002.
</mixed-citation></ref-html>
<ref-html id="bib1.bib2"><label>2</label><mixed-citation>
Birkett, C. M. and Beckley, B.: Investigating the performance of the
Jason-2/OSTM radar altimeter over lakes and reservoirs, Mar. Geod., 33,
204–238, 2010.
</mixed-citation></ref-html>
<ref-html id="bib1.bib3"><label>3</label><mixed-citation>
Bouzinac, C.: CryoSat product handbook, ESA, UCL, available at: <a href="https://earth.esa.int/documents/10174/125272/CryoSat_Product_Handbook" target="_blank">https://earth.esa.int/documents/10174/125272/CryoSat_Product_Handbook</a> (last access: 17 October 2019), 2012.
</mixed-citation></ref-html>
<ref-html id="bib1.bib4"><label>4</label><mixed-citation>
Cheng, K.-C., Kuo, C.-Y., Tseng, H.-Z., Yi, Y., and Shum, C.: Lake surface
height calibration of Jason-1 and Jason-2 over the Great Lakes, Mar.
Geod., 33, 186–203, 2010.
</mixed-citation></ref-html>
<ref-html id="bib1.bib5"><label>5</label><mixed-citation>
Crétaux, J.-F., Jelinski, W., Calmant, S., Kouraev, A., Vuglinski, V.,
Bergé-Nguyen, M., Gennero, M.-C., Nino, F., Del Rio, R. A., and
Cazenave, A.: SOLS: A lake database to monitor in the Near Real Time water
level and storage variations from remote sensing data, Adv. Space
Res., 47, 1497–1507, 2011a.
</mixed-citation></ref-html>
<ref-html id="bib1.bib6"><label>6</label><mixed-citation>
Crétaux, J., Jelinski, W., Calmant, S., Kouraev, A., Vuglinski, V.,
Bergé-Nguyen, M., Nino, G., Del Rio, R., Cazenave, A., and Maisongrande,
P.: Hydrolare/Hydroweb: A lake database to monitor in the Near Real Time
water level and storage variations from remote sensing data, Adv.
Space Res., 47, 1497–1507, 2011b.
</mixed-citation></ref-html>
<ref-html id="bib1.bib7"><label>7</label><mixed-citation>
Crétaux, J.-F., Abarca-del-Río, R., Berge-Nguyen, M., Arsen, A.,
Drolon, V., Clos, G., and Maisongrande, P.: Lake volume monitoring from
space, Surv. Geophys., 37, 269–305, 2016.
</mixed-citation></ref-html>
<ref-html id="bib1.bib8"><label>8</label><mixed-citation>
Field, C. B., Barros, V. R., Dokken, D., Mach, K., Mastrandrea, M., Bilir,
T., Chatterjee, M., Ebi, K., Estrada, Y., and Genova, R.: IPCC, 2014:
Climate change 2014: Impacts, adaptation, and vulnerability – Part A: Global
and sectoral aspects. Contribution of working group II to the fifth
assessment report of the intergovernmental panel on climate change,
Cambridge University Press, Cambridge, UK, New York, NY, USA,
2014.
</mixed-citation></ref-html>
<ref-html id="bib1.bib9"><label>9</label><mixed-citation>
Frappart, F., Seyler, F., Martinez, J.-M., León, J. G., and Cazenave,
A.: Floodplain water storage in the Negro River basin estimated from
microwave remote sensing of inundation area and water levels, Remote Sens. Environ., 99, 387–399, 2005.
</mixed-citation></ref-html>
<ref-html id="bib1.bib10"><label>10</label><mixed-citation>
Gorelick, N., Hancher, M., Dixon, M., Ilyushchenko, S., Thau, D., and Moore,
R.: Google Earth Engine: Planetary-scale geospatial analysis for everyone,
Remote Sens. Environ., 202, 18–27, 2017.
</mixed-citation></ref-html>
<ref-html id="bib1.bib11"><label>11</label><mixed-citation>
Guo, J., Chang, X., Gao, Y., Sun, J., and Hwang, C.: Lake level variations
monitored with satellite altimetry waveform retracking, IEEE J. Sel. Top. Appl., 2, 80–86,
2009.
</mixed-citation></ref-html>
<ref-html id="bib1.bib12"><label>12</label><mixed-citation>
Huang, Q., Long, D., Du, M., Zeng, C., Qiao, G., Li, X., Hou, A., and Hong,
Y.: Discharge estimation in high-mountain regions with improved methods
using multisource remote sensing: A case study of the Upper Brahmaputra
River, Remote Sens. Environ., 219, 115–134, 2018.
</mixed-citation></ref-html>
<ref-html id="bib1.bib13"><label>13</label><mixed-citation>
Hwang, C., Cheng, Y.-S., Yang, W.-H., Zhang, G., Huang, Y.-R., Shen, W.-B.,
and Pan, Y.: Lake level changes in the Tibetan Plateau from Cryosat-2,
SARAL, ICESat, and Jason-2 altimeters, Terr. Atmos. Ocean Sci., 30, 1–18, 2019.
</mixed-citation></ref-html>
<ref-html id="bib1.bib14"><label>14</label><mixed-citation>
Jain, M., Andersen, O. B., Dall, J., and Stenseng, L.: Sea surface height
determination in the Arctic using Cryosat-2 SAR data from primary peak
empirical retrackers, Adv. Space Res., 55, 40–50, 2015.
</mixed-citation></ref-html>
<ref-html id="bib1.bib15"><label>15</label><mixed-citation>
Jarvis, A., Reuter, H. I., Nelson, A., and Guevara, E.: Hole-filled SRTM for
the globe Version 4, available from the CGIAR-CSI SRTM 90&thinsp;m Database,
available at: <a href="http://srtm.csi.cgiar.org" target="_blank">http://srtm.csi.cgiar.org</a> (last access: 17 October 2019), 15, 25–54, 2008.
</mixed-citation></ref-html>
<ref-html id="bib1.bib16"><label>16</label><mixed-citation>
Jiang, L., Nielsen, K., Andersen, O. B., and Bauer-Gottwein, P.: Monitoring
recent lake level variations on the Tibetan Plateau using CryoSat-2 SARIn
mode data, J. Hydrol., 544, 109–124, 2017.
</mixed-citation></ref-html>
<ref-html id="bib1.bib17"><label>17</label><mixed-citation>
Kittler, J. and Illingworth, J.: On threshold selection using clustering
criteria, IEEE T. Syst. Man. Cyb., September–October 1985, SMC-15, 652–655,
1985.
</mixed-citation></ref-html>
<ref-html id="bib1.bib18"><label>18</label><mixed-citation>
Lehner, B. and Döll, P.: Development and validation of a global database
of lakes, reservoirs and wetlands, J. Hydrol., 296, 1–22, 2004.
</mixed-citation></ref-html>
<ref-html id="bib1.bib19"><label>19</label><mixed-citation>
Lei, Y., Yao, T., Yang, K., Sheng, Y., Kleinherenbrink, M., Yi, S., Bird, B.
W., Zhang, X., Zhu, L., and Zhang, G.: Lake seasonality across the Tibetan
Plateau and their varying relationship with regional mass changes and local
hydrology, Geophys. Res. Lett., 44, 892–900, 2017.
</mixed-citation></ref-html>
<ref-html id="bib1.bib20"><label>20</label><mixed-citation>
Li, B., Zhang, J., Yu, Z., Liang, Z., Chen, L., and Acharya, K.: Climate
change driven water budget dynamics of a Tibetan inland lake, Global
Planet. Change, 150, 70–80, 2017a.
</mixed-citation></ref-html>
<ref-html id="bib1.bib21"><label>21</label><mixed-citation>
Li, H. W., Qiao, G., Wu, Y. J., Cao, Y. J., and Mi, H.: Water Level Monitoring On Tibetan Lakes Based On Icesat And Envisat Data Series, Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-2/W7, 1529–1533, <a href="https://doi.org/10.5194/isprs-archives-XLII-2-W7-1529-2017" target="_blank">https://doi.org/10.5194/isprs-archives-XLII-2-W7-1529-2017</a>, 2017b.
</mixed-citation></ref-html>
<ref-html id="bib1.bib22"><label>22</label><mixed-citation>
Li, L., Liu, Q., Zhang, Y., Liu, L., Ding, M., and Changjun, G. U.: Spatial
distribution and variation of precipitation in the Qiangtang Plateau,
Geogr. Res., 36, 2047–2060, 2017c.
</mixed-citation></ref-html>
<ref-html id="bib1.bib23"><label>23</label><mixed-citation>
Li, X., Long, D., Huang, Q., Han, P., Zhao, F., and Wada, Y.: A high
temporal resolution lake data set from multisource altimetric missions and
Landsat archives of water level and storage changes on the Tibetan Plateau
during 2000–2017,
PANGAEA, <a href="https://doi.org/10.1594/PANGAEA.898411" target="_blank">https://doi.org/10.1594/PANGAEA.898411</a>, 2019.
</mixed-citation></ref-html>
<ref-html id="bib1.bib24"><label>24</label><mixed-citation>
Liu, B., Lin, L. I., Yue, D. U., Liang, T., Duan, S., Hou, F., and Ren, J.:
Causes of the outburst of Zonag Lake in Hoh Xil, Tibetan Plateau,and its
impact on surrounding environment, J. Glaciol. Geocryol., 38, 305–311, 2016.
</mixed-citation></ref-html>
<ref-html id="bib1.bib25"><label>25</label><mixed-citation>
Markham, B. L., Storey, J. C., Williams, D. L., and Irons, J. R.: Landsat
sensor performance: history and current status, IEEE T. Geosci. Remote, 42, 2691–2694, 2004.
</mixed-citation></ref-html>
<ref-html id="bib1.bib26"><label>26</label><mixed-citation>
McFeeters, S. K.: The use of the Normalized Difference Water Index (NDWI) in
the delineation of open water features, Int. J. Remote
Sens., 17, 1425–1432, 1996.
</mixed-citation></ref-html>
<ref-html id="bib1.bib27"><label>27</label><mixed-citation>
Messager, M. L., Lehner, B., Grill, G., Nedeva, I., and Schmitt, O.:
Estimating the volume and age of water stored in global lakes using a
geo-statistical approach, Nat, Commun., 7, 13603, <a href="https://doi.org/10.1038/ncomms13603" target="_blank">https://doi.org/10.1038/ncomms13603</a>, 2016.
</mixed-citation></ref-html>
<ref-html id="bib1.bib28"><label>28</label><mixed-citation>
Neckel, N., Kropáček, J., Bolch, T., and Hochschild, V.: Glacier
mass changes on the Tibetan Plateau 2003–2009 derived from ICESat laser
altimetry measurements, Environ. Res. Lett., 9, 014009, <a href="https://doi.org/10.1088/1748-9326/9/1/014009" target="_blank">https://doi.org/10.1088/1748-9326/9/1/014009</a>, 2014.
</mixed-citation></ref-html>
<ref-html id="bib1.bib29"><label>29</label><mixed-citation>
Nielsen, K., Stenseng, L., Andersen, O. B., and Knudsen, P.: The performance
and potentials of the CryoSat-2 SAR and SARIn modes for lake level
estimation, Water, 9, w9060374, <a href="https://doi.org/10.3390/w9060374" target="_blank">https://doi.org/10.3390/w9060374</a>, 2017.
</mixed-citation></ref-html>
<ref-html id="bib1.bib30"><label>30</label><mixed-citation>
Otsu, N.: A threshold selection method from gray-level histograms, IEEE T. Syst. Man. Cyb., 9, 62–66, 1979.
</mixed-citation></ref-html>
<ref-html id="bib1.bib31"><label>31</label><mixed-citation>
Pekel, J.-F., Cottam, A., Gorelick, N., and Belward, A. S.: High-resolution
mapping of global surface water and its long-term changes, Nature, 540, 418–422,
2016.
</mixed-citation></ref-html>
<ref-html id="bib1.bib32"><label>32</label><mixed-citation>
Planet: Planet Application Program Interface: In Space for Life on Earth,
San Francisco, CA, available at: <a href="https://www.planet.com" target="_blank">https://www.planet.com</a> (last access: 17 October 2019), 2017.
</mixed-citation></ref-html>
<ref-html id="bib1.bib33"><label>33</label><mixed-citation>
Sørensen, L. S., Simonsen, S. B., Nielsen, K., Lucas-Picher, P., Spada, G., Adalgeirsdottir, G., Forsberg, R., and Hvidberg, C. S.: Mass balance of the Greenland ice sheet (2003–2008) from ICESat data – the impact of interpolation, sampling and firn density, The Cryosphere, 5, 173–186, <a href="https://doi.org/10.5194/tc-5-173-2011" target="_blank">https://doi.org/10.5194/tc-5-173-2011</a>, 2011.
</mixed-citation></ref-html>
<ref-html id="bib1.bib34"><label>34</label><mixed-citation>
Song, C., Huang, B., and Ke, L.: Modeling and analysis of lake water storage
changes on the Tibetan Plateau using multi-mission satellite data, Remote
Sens. Environ., 135, 25–35, 2013.
</mixed-citation></ref-html>
<ref-html id="bib1.bib35"><label>35</label><mixed-citation>
Tournadre, J.: Validation of Jason and Envisat altimeter dual frequency rain
flags, Mar. Geod., 27, 153–169, 2004.
</mixed-citation></ref-html>
<ref-html id="bib1.bib36"><label>36</label><mixed-citation>
Wan, W., Long, D., Hong, Y., Ma, Y., Yuan, Y., Xiao, P., Duan, H., Han, Z.,
and Gu, X.: A lake data set for the Tibetan Plateau from the 1960s, 2005,
and 2014, Scientific data, 3, 160039, <a href="https://doi.org/10.1038/sdata.2016.39" target="_blank">https://doi.org/10.1038/sdata.2016.39</a>, 2016.
</mixed-citation></ref-html>
<ref-html id="bib1.bib37"><label>37</label><mixed-citation>
Wang, J., Song, C., Reager, J. T., Yao, F., Famiglietti, J. S., Sheng, Y.,
MacDonald, G. M., Brun, F., Schmied, H. M., and Marston, R. A.: Recent
global decline in endorheic basin water storages, Nat. Geosci., 11,
926–932, 2018.
</mixed-citation></ref-html>
<ref-html id="bib1.bib38"><label>38</label><mixed-citation>
Xiaojun, Y., Shiyin, L., and Meiping, S.: Changes of Kusai Lake in Hoh Xil
region and causes of its water overflowing, Ac. Geogr. Sin., 67,
689–698, 2012.
</mixed-citation></ref-html>
<ref-html id="bib1.bib39"><label>39</label><mixed-citation>
Xu, H.: A study on information extraction of water body with the modified
normalized difference water index (MNDWI), J. Remote
Sens.-Beijing, 9, 589–595, 2005.
</mixed-citation></ref-html>
<ref-html id="bib1.bib40"><label>40</label><mixed-citation>
Yao, F., Wang, J., Yang, K., Wang, C., Walter, B. A., and Crétaux,
J.-F.: High resolution data set of annual lake areas and water storage across the Inner Tibet, 2002–2015, PANGAEA, <a href="https://doi.org/10.1594/PANGAEA.888706" target="_blank">https://doi.org/10.1594/PANGAEA.888706</a>, 2018a.

</mixed-citation></ref-html>
<ref-html id="bib1.bib41"><label>41</label><mixed-citation>
Yao, F., Wang, J., Yang, K., Wang, C., Walter, B. A., and Crétaux,
J.-F.: Lake storage variation on the endorheic Tibetan Plateau and its
attribution to climate change since the new millennium, Environ.
Res. Lett., 13, 064011, <a href="https://doi.org/10.1088/1748-9326/aab5d3" target="_blank">https://doi.org/10.1088/1748-9326/aab5d3</a>, 2018b.
</mixed-citation></ref-html>
<ref-html id="bib1.bib42"><label>42</label><mixed-citation>
Yao, X., Liu, S., and Sun, M.: Changes of Kusai Lake in Hoh Xil region and
causes of its water overflowing, Ac. Geogr. Sin., 67, 689–698, 2012.
</mixed-citation></ref-html>
<ref-html id="bib1.bib43"><label>43</label><mixed-citation>
Yu, S., Liu, J., Xu, J., and Wang, H.: Evaporation and energy balance
estimates over a large inland lake in the Tibet-Himalaya, Environ.
Earth Sci., 64, 1169–1176, 2011.
</mixed-citation></ref-html>
<ref-html id="bib1.bib44"><label>44</label><mixed-citation>
Zhang, G., Xie, H., Kang, S., Yi, D., and Ackley, S. F.: Monitoring lake
level changes on the Tibetan Plateau using ICESat altimetry data
(2003–2009), Remote Sens. Environ., 115, 1733–1742, 2011.
</mixed-citation></ref-html>
<ref-html id="bib1.bib45"><label>45</label><mixed-citation>
Zhang, G., Yao, T., Xie, H., Kang, S., and Lei, Y.: Increased mass over the
Tibetan Plateau: from lakes or glaciers?, Geophys. Res. Lett., 40,
2125–2130, 2013.
</mixed-citation></ref-html>
<ref-html id="bib1.bib46"><label>46</label><mixed-citation>
Zhang, G., Yao, T., Shum, C., Yi, S., Yang, K., Xie, H., Feng, W., Bolch,
T., Wang, L., and Behrangi, A.: Lake volume and groundwater storage
variations in Tibetan Plateau's endorheic basin, Geophys. Res.
Lett., 44, 5550–5560, 2017a.
</mixed-citation></ref-html>
<ref-html id="bib1.bib47"><label>47</label><mixed-citation>
Zhang, G., Zheng, G., Gao, Y., Xiang, Y., Lei, Y., and Li, J.: Automated
water classification in the Tibetan plateau using Chinese GF-1 WFV data,
Photogramm. Eng. Remote Sens., 83, 509–519, 2017b.
</mixed-citation></ref-html>
<ref-html id="bib1.bib48"><label>48</label><mixed-citation>
Zhou, J., Wang, L., Zhang, Y., Guo, Y., Li, X., and Liu, W.: Exploring the
water storage changes in the largest lake (Selin Co) over the Tibetan
Plateau during 2003–2012 from a basin-wide hydrological modeling, Water
Resour. Res., 51, 8060–8086, 2015.
</mixed-citation></ref-html>--></article>
