<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE article PUBLIC "-//NLM//DTD Journal Publishing DTD v3.0 20080202//EN" "https://jats.nlm.nih.gov/nlm-dtd/publishing/3.0/journalpublishing3.dtd">
<article xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" article-type="research-article" specific-use="SMUR" dtd-version="3.0" xml:lang="en">
<front>
<journal-meta>
<journal-id journal-id-type="publisher">EGUsphere</journal-id>
<journal-title-group>
<journal-title>EGUsphere</journal-title>
<abbrev-journal-title abbrev-type="publisher">EGUsphere</abbrev-journal-title>
<abbrev-journal-title abbrev-type="nlm-ta">EGUsphere</abbrev-journal-title>
</journal-title-group>
<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/egusphere-2026-6643</article-id>
<title-group>
<article-title>Benchmarking Earth Observation Embeddings for Pan-EU Aboveground Biomass Mapping</article-title>
</title-group>
<contrib-group><contrib contrib-type="author" xlink:type="simple"><name name-style="western"><surname>Ho</surname>
<given-names>Yu-Feng</given-names>
<ext-link>https://orcid.org/0000-0003-0956-4606</ext-link>
</name>
<xref ref-type="aff" rid="aff1">
<sup>1</sup>
</xref>
<xref ref-type="aff" rid="aff2">
<sup>2</sup>
</xref>
</contrib>
<contrib contrib-type="author" xlink:type="simple"><name name-style="western"><surname>Nussbaum</surname>
<given-names>Madlene</given-names>
<ext-link>https://orcid.org/0000-0002-6808-8956</ext-link>
</name>
<xref ref-type="aff" rid="aff2">
<sup>2</sup>
</xref>
</contrib>
<contrib contrib-type="author" xlink:type="simple"><name name-style="western"><surname>Enguehard</surname>
<given-names>Léa</given-names>
<ext-link>https://orcid.org/0000-0002-2144-8264</ext-link>
</name>
<xref ref-type="aff" rid="aff3">
<sup>3</sup>
</xref>
</contrib>
<contrib contrib-type="author" xlink:type="simple"><name name-style="western"><surname>Song</surname>
<given-names>Qian</given-names>
<ext-link>https://orcid.org/0000-0003-2746-6858</ext-link>
</name>
<xref ref-type="aff" rid="aff3">
<sup>3</sup>
</xref>
</contrib>
<contrib contrib-type="author" xlink:type="simple"><name name-style="western"><surname>Serkan Isik</surname>
<given-names>Mustafa</given-names>
<ext-link>https://orcid.org/0000-0003-1769-4451</ext-link>
</name>
<xref ref-type="aff" rid="aff1">
<sup>1</sup>
</xref>
</contrib>
<contrib contrib-type="author" xlink:type="simple"><name name-style="western"><surname>de Lange</surname>
<given-names>Jordy</given-names>
</name>
<xref ref-type="aff" rid="aff4">
<sup>4</sup>
</xref>
</contrib>
<contrib contrib-type="author" xlink:type="simple"><name name-style="western"><surname>Bonannella</surname>
<given-names>Carmelo</given-names>
</name>
<xref ref-type="aff" rid="aff1">
<sup>1</sup>
</xref>
</contrib>
<contrib contrib-type="author" xlink:type="simple"><name name-style="western"><surname>Parente</surname>
<given-names>Leandro</given-names>
<ext-link>https://orcid.org/0000-0003-1589-0467</ext-link>
</name>
<xref ref-type="aff" rid="aff1">
<sup>1</sup>
</xref>
</contrib>
<contrib contrib-type="author" xlink:type="simple"><name name-style="western"><surname>Karssenberg</surname>
<given-names>Derek</given-names>
<ext-link>https://orcid.org/0000-0002-6475-363X</ext-link>
</name>
<xref ref-type="aff" rid="aff2">
<sup>2</sup>
</xref>
</contrib>
<contrib contrib-type="author" xlink:type="simple"><name name-style="western"><surname>Herold</surname>
<given-names>Martin</given-names>
</name>
<xref ref-type="aff" rid="aff3">
<sup>3</sup>
</xref>
</contrib>
<contrib contrib-type="author" xlink:type="simple"><name name-style="western"><surname>Wheeler</surname>
<given-names>Ichsani</given-names>
</name>
<xref ref-type="aff" rid="aff1">
<sup>1</sup>
</xref>
</contrib>
<contrib contrib-type="author" xlink:type="simple"><name name-style="western"><surname>Hengl</surname>
<given-names>Tomislav</given-names>
<ext-link>https://orcid.org/0000-0002-9921-5129</ext-link>
</name>
<xref ref-type="aff" rid="aff1">
<sup>1</sup>
</xref>
</contrib>
</contrib-group><aff id="aff1">
<label>1</label>
<addr-line>OpenGeoHub Foundation, Doorwerth, The Netherlands</addr-line>
</aff>
<aff id="aff2">
<label>2</label>
<addr-line>Department of Physical Geography, Faculty of Geosciences, Utrecht University, Utrecht, The Netherlands</addr-line>
</aff>
<aff id="aff3">
<label>3</label>
<addr-line>Remote Sensing and Geoinformatics, GFZ Helmholtz Centre for Geosciences, Potsdam, Germany</addr-line>
</aff>
<aff id="aff4">
<label>4</label>
<addr-line>Spheer, Groningen, The Netherlands</addr-line>
</aff>
<pub-date pub-type="epub">
<day>17</day>
<month>11</month>
<year>2026</year>
</pub-date>
<volume>2026</volume>
<fpage>1</fpage>
<lpage>39</lpage>
<permissions>
<copyright-statement>Copyright: &#x000a9; 2026 Yu-Feng Ho et al.</copyright-statement>
<copyright-year>2026</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://egusphere.copernicus.org/preprints/2026/egusphere-2026-6643/">This article is available from https://egusphere.copernicus.org/preprints/2026/egusphere-2026-6643/</self-uri>
<self-uri xlink:href="https://egusphere.copernicus.org/preprints/2026/egusphere-2026-6643/egusphere-2026-6643.pdf">The full text article is available as a PDF file from https://egusphere.copernicus.org/preprints/2026/egusphere-2026-6643/egusphere-2026-6643.pdf</self-uri>
<abstract>
<p>The paper presents a comprehensive evaluation of pre-trained Earth Observation (EO) embeddings compared to traditional remote sensing approaches trained with standard multi-sensor satellite data (Sentinel-1, Sentinel-2, and ALOS PALSAR-2, both with and without the addition of multi-scale terrain derivatives) for mapping aboveground biomass density (AGBD). The rank result was evaluated on 4 criteria: point-estimate accuracy, spatial pattern reconstruction, saturation effects, and uncertainty quantification. EO features were specifically used with tree-based (Quantile Random Forest Regression; QRFR) and deep learning (Convolutional Neural Network; CNN) models to map AGBD across varied biogeographical region in Finland (&amp;asymp; 73,000 km&lt;sup&gt;2&lt;/sup&gt;), Spain (&amp;asymp; 273,000 km&lt;sup&gt;2&lt;/sup&gt;), and the Netherlands (&amp;asymp; 2,200 km&lt;sup&gt;2&lt;/sup&gt;). A proportional stratified sampling design using airborne laser scanning (ALS) estimates of AGBD was implemented to generate sample tiles for each country, and country-specific models were trained. The benchmarking results show that EO embeddings achieve predictive accuracy that is highly competitive when compared with models that use Sentinel-1, Sentinel-2, and PALSAR: the lowest RMSE values were recorded by AlphaEarth &amp;ndash; QRFR with terrain variables in the Netherlands (42.1 Mg ha&lt;sup&gt;&amp;minus;1&lt;/sup&gt;), SpheerFM &amp;ndash; CNN in Spain (17.5 Mg ha&lt;sup&gt;&amp;minus;1&lt;/sup&gt;), and TESSERA &amp;ndash; CNN in Finland (13.4 Mg ha&lt;sup&gt;&amp;minus;1&lt;/sup&gt;). Pairing these embeddings with a CNN model yielded clear performance gains: TESSERA improved across all three test countries, while SpheerFM showed distinct enhancements in Spain and Finland. This strong performance was further corroborated by spatial pattern reconstruction metrics. In addition, a rigorous uncertainty assessment of both the QRFR and CNN models using empirical metrics, Prediction Interval Coverage Probability (PICP) and Mean Prediction Interval (MPI), showed that both models derive unbiased, well-calibrated prediction intervals. Ultimately, TESSERA utilized CNN as downstream model comes as the highest-ranked configuration. The result from different predictors set is also marginal, implying no single set dominates. The choice of applying EO embeddings for large-scale AGBD mapping requires consideration also data licensing and transparency, data volume, and their existing spatial temporal coverage.</p>
</abstract>
<counts><page-count count="39"/></counts>
<funding-group>
<award-group id="gs1">
<funding-source>European Commission</funding-source>
<award-id>101218854</award-id>
<award-id>101084481</award-id>
<award-id>101056875</award-id>
</award-group>
<award-group id="gs2">
<funding-source>Helmholtz Association</funding-source>
<award-id>E5430314</award-id>
</award-group>
</funding-group>
</article-meta>
</front>
<body/>
<back>
</back>
</article>