<?xml version="1.0" encoding="utf-8"?>
<TEI xmlns="http://www.tei-c.org/ns/1.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:hal="http://hal.archives-ouvertes.fr/" xmlns:gml="http://www.opengis.net/gml/3.3/" xmlns:gmlce="http://www.opengis.net/gml/3.3/ce" version="1.1" xsi:schemaLocation="http://www.tei-c.org/ns/1.0 http://api.archives-ouvertes.fr/documents/aofr-sword.xsd">
  <teiHeader>
    <fileDesc>
      <titleStmt>
        <title>HAL TEI export of hal-05178352</title>
      </titleStmt>
      <publicationStmt>
        <distributor>CCSD</distributor>
        <availability status="restricted">
          <licence target="https://creativecommons.org/publicdomain/zero/1.0/">CC0 1.0 - Universal</licence>
        </availability>
        <date when="2026-05-15T18:01:21+02:00"/>
      </publicationStmt>
      <sourceDesc>
        <p part="N">HAL API Platform</p>
      </sourceDesc>
    </fileDesc>
  </teiHeader>
  <text>
    <body>
      <listBibl>
        <biblFull>
          <titleStmt>
            <title xml:lang="en">Unsupervised classification of hydrological conditions in non-perennial rivers</title>
            <title xml:lang="fr">Classification automatique des conditions hydrologiques des rivières non pérennes</title>
            <author role="aut">
              <persName>
                <forename type="first">Isabelle</forename>
                <surname>Brichetto</surname>
              </persName>
              <idno type="halauthorid">3667560-0</idno>
              <affiliation ref="#struct-6571"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Carmela</forename>
                <surname>Cavallo</surname>
              </persName>
              <idno type="halauthorid">3667561-0</idno>
              <affiliation ref="#struct-302831"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Giammarco</forename>
                <surname>Manfreda</surname>
              </persName>
              <idno type="halauthorid">3667562-0</idno>
              <affiliation ref="#struct-6571"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Giovanni</forename>
                <surname>Negro</surname>
              </persName>
              <idno type="halauthorid">3667563-0</idno>
              <affiliation ref="#struct-6571"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Maria Nicolina</forename>
                <surname>Papa</surname>
              </persName>
              <idno type="halauthorid">3667564-0</idno>
              <affiliation ref="#struct-302831"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Paolo</forename>
                <surname>Vezza</surname>
              </persName>
              <idno type="halauthorid">1991226-0</idno>
              <affiliation ref="#struct-6571"/>
            </author>
            <editor role="depositor">
              <persName>
                <forename>Sathya</forename>
                <surname>FLORY</surname>
              </persName>
              <email type="md5">cedff8aa932d9681523cccff887cf3a4</email>
              <email type="domain">graie.org</email>
            </editor>
          </titleStmt>
          <editionStmt>
            <edition n="v1" type="current">
              <date type="whenSubmitted">2025-07-23 09:55:48</date>
              <date type="whenWritten">2025</date>
              <date type="whenModified">2025-08-25 12:12:35</date>
              <date type="whenReleased">2025-08-13 11:16:58</date>
              <date type="whenProduced">2025-06-30</date>
              <date type="whenEndEmbargoed">2025-07-23</date>
              <ref type="file" target="https://hal.science/hal-05178352v1/document">
                <date notBefore="2025-07-23"/>
              </ref>
              <ref type="file" subtype="annex" n="1" target="https://hal.science/hal-05178352v1/file/A3-ISR25-217-Brichetto-Isabelle-presentation.pdf" id="file-5178352-4482673">
                <date notBefore="2025-07-23"/>
              </ref>
              <ref type="file" subtype="author" n="0" target="https://hal.science/hal-05178352v1/file/A3-ISR25-217-Brichetto-Isabelle.pdf" id="file-5178352-4482672">
                <date notBefore="2025-07-23"/>
              </ref>
            </edition>
            <respStmt>
              <resp>contributor</resp>
              <name key="1073112">
                <persName>
                  <forename>Sathya</forename>
                  <surname>FLORY</surname>
                </persName>
                <email type="md5">cedff8aa932d9681523cccff887cf3a4</email>
                <email type="domain">graie.org</email>
              </name>
            </respStmt>
          </editionStmt>
          <publicationStmt>
            <distributor>CCSD</distributor>
            <idno type="halId">hal-05178352</idno>
            <idno type="halUri">https://hal.science/hal-05178352</idno>
            <idno type="halBibtex">brichetto:hal-05178352</idno>
            <idno type="halRefHtml">&lt;i&gt;Conférence internationale I.S.Rivers 2025&lt;/i&gt;, Jun 2025, Lyon, France. 2025</idno>
            <idno type="halRef">Conférence internationale I.S.Rivers 2025, Jun 2025, Lyon, France. 2025</idno>
            <availability status="restricted">
              <licence target="https://about.hal.science/hal-authorisation-v1/">HAL Authorization<ref corresp="#file-5178352-4482673"/><ref corresp="#file-5178352-4482672"/></licence>
            </availability>
          </publicationStmt>
          <seriesStmt>
            <idno type="stamp" n="SDE">Sciences De l'Environnement</idno>
            <idno type="stamp" n="GIP-BE">GIP Bretagne Environnement</idno>
            <idno type="stamp" n="ISRIVERS">Conférence internationale I.S.Rivers</idno>
          </seriesStmt>
          <notesStmt>
            <note type="audience" n="2">International</note>
            <note type="invited" n="0">No</note>
            <note type="popular" n="0">No</note>
            <note type="peer" n="1">Yes</note>
            <note type="proceedings" n="1">Yes</note>
          </notesStmt>
          <sourceDesc>
            <biblStruct>
              <analytic>
                <title xml:lang="en">Unsupervised classification of hydrological conditions in non-perennial rivers</title>
                <title xml:lang="fr">Classification automatique des conditions hydrologiques des rivières non pérennes</title>
                <author role="aut">
                  <persName>
                    <forename type="first">Isabelle</forename>
                    <surname>Brichetto</surname>
                  </persName>
                  <idno type="halauthorid">3667560-0</idno>
                  <affiliation ref="#struct-6571"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Carmela</forename>
                    <surname>Cavallo</surname>
                  </persName>
                  <idno type="halauthorid">3667561-0</idno>
                  <affiliation ref="#struct-302831"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Giammarco</forename>
                    <surname>Manfreda</surname>
                  </persName>
                  <idno type="halauthorid">3667562-0</idno>
                  <affiliation ref="#struct-6571"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Giovanni</forename>
                    <surname>Negro</surname>
                  </persName>
                  <idno type="halauthorid">3667563-0</idno>
                  <affiliation ref="#struct-6571"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Maria Nicolina</forename>
                    <surname>Papa</surname>
                  </persName>
                  <idno type="halauthorid">3667564-0</idno>
                  <affiliation ref="#struct-302831"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Paolo</forename>
                    <surname>Vezza</surname>
                  </persName>
                  <idno type="halauthorid">1991226-0</idno>
                  <affiliation ref="#struct-6571"/>
                </author>
              </analytic>
              <monogr>
                <meeting>
                  <title>Conférence internationale I.S.Rivers 2025</title>
                  <date type="start">2025-06-30</date>
                  <date type="end">2025-07-04</date>
                  <settlement>Lyon</settlement>
                  <country key="FR">France</country>
                </meeting>
                <imprint>
                  <date type="datePub">2025-07-09</date>
                </imprint>
              </monogr>
            </biblStruct>
          </sourceDesc>
          <profileDesc>
            <langUsage>
              <language ident="en">English</language>
            </langUsage>
            <textClass>
              <keywords scheme="author">
                <term xml:lang="en">Sentinel-2 satellites.</term>
                <term xml:lang="en">Non-perennial rivers</term>
                <term xml:lang="en">Multispectral images</term>
                <term xml:lang="en">Hydrotypes</term>
                <term xml:lang="en">Convolutional Neural Networks</term>
                <term xml:lang="fr">Satellites Sentinel-2.</term>
                <term xml:lang="fr">Rivières non pérennes</term>
                <term xml:lang="fr">Imagerie multispectrale</term>
                <term xml:lang="fr">Hydrotypes</term>
                <term xml:lang="fr">Réseau de neurones convolutif</term>
              </keywords>
              <classCode scheme="halDomain" n="sde">Environmental Sciences</classCode>
              <classCode scheme="halTypology" n="POSTER">Poster communications</classCode>
              <classCode scheme="halOldTypology" n="POSTER">Poster communications</classCode>
              <classCode scheme="halTreeTypology" n="POSTER">Poster communications</classCode>
            </textClass>
            <abstract xml:lang="en">
              <p>Anthropogenic pressures, such as climate change, land use change and water withdrawals, are leading to significant shifts in hydrological cycles, increasing the spatial extension of the non-perennial rivers (NPRs) network. However, knowledge about the frequency and duration of flow intermittency is severely constrained by the small number of streamflow gauges and the limited reliability of hydrological models in predicting surface water presence when discharge is close or equal to zero. In this context, Sentinel-2 images can provide useful information for studying hydrological processes in NPRs, offering effective ways to observe water surface dynamics at an adequate spatial and temporal resolution. The combination of multispectral satellite imagery and Convolutional Neural Networks (CNNs) can be exploited to distinguish the three hydrological conditions which characterize NPRs automatically: “flowing” (F), “ponding” (P) and “dry” (D). In this study, learning on Sentinel-2 false-color images (B11-B8-B4), different CNNs (2-, 3-, 4-class models) were trained to classify hydrological conditions in five river reaches, considering distinct output classes. In some cases, CNNs were also trained to identify the “cloudy” (C) class. Accuracy of CNNs ranged from 0.87-0.98 for 2- and 3-class models, decreasing to 0.71-0.92 range when the distinction between P and D classes was considered (4-class model). Despite the small size of the dataset (1555 images), the obtained results were promising and deserves further refinement, with the purpose of having a more powerful tool for the automatic classification of hydrological conditions in NPRs.</p>
            </abstract>
            <abstract xml:lang="fr">
              <p>Les pressions anthropiques, comme changements climatique, d'utilisation des sols et prélèvements d'eau, entraînent des modifications significatives des cycles hydrologiques, augmentant l'extension spatiale des rivières non pérennes (NPRs). Cependant, la connaissance de la fréquence et de la durée des intermittences fluviales est fortement limitée par le faible nombre de stations de mesure des débits et la fiabilité réduite des modèles hydrologiques pour prédire la présence d’eau de surface. Dans ce contexte, les images Sentinel-2 peuvent fournir des informations précieuses pour l’étude des processus hydrologiques dans les NPRs, offrant une résolution spatiale et temporelle adéquate. La combinaison des images satellites multispectrales et des réseaux neuronaux convolutionnels (CNNs) peut être exploitée pour distinguer automatiquement les trois conditions hydrologiques des NPRs : “en écoulement” (F), “stagnant” (P) et “asséché” (D). Dans cette étude, à partir des images Sentinel-2 en fausses couleurs (B11-B8-B4), différents modèles de CNNs (à 2, 3 et 4 classes) ont été entraînés pour classifier cinq tronçons de rivière, en considérant classes de prédiction distinctes. Dans certains cas, les CNNs ont été entraînés pour identifier aussi la classe “nuageux” (C). L’exactitude des CNNs variaient en 0.87-0.98 pour les modèles à 2 et 3 classes, et diminuait en 0.71-0.92 avec la distinction des P et D (modèle à 4 classes). Malgré la petite taille du jeu de données (1555 images), les résultats sont prometteurs et méritent améliorations pour être un puissant instrument de classification automatique des conditions hydrologiques en NPRs.</p>
            </abstract>
          </profileDesc>
        </biblFull>
      </listBibl>
    </body>
    <back>
      <listOrg type="structures">
        <org type="institution" xml:id="struct-6571" status="VALID">
          <idno type="IdRef">034390626</idno>
          <idno type="ISNI">0000000419370343</idno>
          <idno type="ROR">https://ror.org/00bgk9508</idno>
          <orgName>Politecnico di Torino = Polytechnic of Turin</orgName>
          <orgName type="acronym">Polito</orgName>
          <date type="start">1859-01-01</date>
          <desc>
            <address>
              <addrLine>Politecnico di Torino - Corso Duca degli Abruzzi, 24 10129 Torino</addrLine>
              <country key="IT"/>
            </address>
            <ref type="url">http://www.polito.it/</ref>
          </desc>
        </org>
        <org type="institution" xml:id="struct-302831" status="VALID">
          <idno type="ISNI">0000000419370335</idno>
          <idno type="ROR">https://ror.org/0192m2k53</idno>
          <orgName>Università degli Studi di Salerno = University of Salerno</orgName>
          <orgName type="acronym">UNISA</orgName>
          <desc>
            <address>
              <addrLine>Via Giovanni Paolo II, 132 - 84084 Fisciano (SA)</addrLine>
              <country key="IT"/>
            </address>
            <ref type="url">https://www.unisa.it/</ref>
          </desc>
        </org>
      </listOrg>
    </back>
  </text>
</TEI>