Scale matters: Spatial resolution impacts tropical leaf phenology characterized by multi-source satellite remote sensing with an ecological-constrained deep learning model

Accurate monitoring of tropical leaf phenology, such as the leaf-on/off status, at both individual and ecosystem scales is essential for understanding and modelling tropical forest carbon and water cycles, and their sensitivity to climate change. The discrepancy between tree-crown size and pixel siz...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Remote sensing of environment 2024-04, Vol.304, p.114027, Article 114027
Hauptverfasser: Song, Guangqin, Wang, Jing, Zhao, Yingyi, Yang, Dedi, Lee, Calvin K.F., Guo, Zhengfei, Detto, Matteo, Alberton, Bruna, Morellato, Patricia, Nelson, Bruce, Wu, Jin
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page 114027
container_title Remote sensing of environment
container_volume 304
creator Song, Guangqin
Wang, Jing
Zhao, Yingyi
Yang, Dedi
Lee, Calvin K.F.
Guo, Zhengfei
Detto, Matteo
Alberton, Bruna
Morellato, Patricia
Nelson, Bruce
Wu, Jin
description Accurate monitoring of tropical leaf phenology, such as the leaf-on/off status, at both individual and ecosystem scales is essential for understanding and modelling tropical forest carbon and water cycles, and their sensitivity to climate change. The discrepancy between tree-crown size and pixel size (i.e., spatial resolution) across orbital sensors can affect the capability of cross-scale phenology monitoring, an aspect that remains understudied. To examine the impact of spatial resolution on tropical leaf phenology monitoring, we applied a spectral index-guided, ecologically constrained autoencoder (IG-ECAE) to automatically generate a deciduousness metric (i.e., percentage of upper canopy area that is leaf-off status within an image pixel) from simulated VIS-NIR PlanetScope data at a range of resolutions from 3 m to 30 m, as well as from VIS-NIR data of three satellite platforms with the same range of spatial resolutions (3 m PlanetScope, 10 m Sentinel-2, and 30 m Landsat-8). We compared the deciduousness metrics derived from the simulated and satellite data to corresponding measurements derived from WorldView-2 (three sites) and local phenocams (four sites) at five tropical forest sites. Our results revealed that: (1) the IG-ECAE model captured the amount of deciduousness across spatial scales, with the highest accuracy obtained from PlanetScope, followed by Sentinel-2 and Landsat-8; (2) coarser spatial resolutions led to lower accuracies in tropical deciduousness monitoring, as demonstrated by both simulated PlanetScope data across various spatial resolutions and real satellite data; and (3) while not as accurate in capturing fine-scale tropical phenological diversity as PlanetScope, Sentinel-2 provided satisfactory monitoring of deciduousness seasonality at the ecosystem level consistently across all phenocam sites, whereas Landsat-8 failed to do so. Collectively, this study provides a robust assessment for advancing cross-scale tropical leaf phenology monitoring with potential for extension to pan-tropical regions and highlights the impact of spatial resolution on such monitoring efforts. •Spatial resolution affects phenology monitoring assessed by simulated and real data.•The spatial resolution's impact was evaluated with WorldView-2 and phenocam data.•Coarser resolution reduces phenology monitoring accuracy using deciduousness metric.•Sentinel-2(10 m) nears PlanetScope(3 m)’s ability for ecosystem phenology monitoring.•PlanetScope(3 m) remains vit
doi_str_mv 10.1016/j.rse.2024.114027
format Article
fullrecord <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_miscellaneous_3153166838</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><els_id>S0034425724000385</els_id><sourcerecordid>3153166838</sourcerecordid><originalsourceid>FETCH-LOGICAL-c282t-c098656bf227e9a06e6e9d3a06f376667d7673790bd8140d49b90ebd4c458a0d3</originalsourceid><addsrcrecordid>eNp9kc1u1DAURi0EEkPhAbrzspsMdn7spKxQRaFSpS5a1pZj33Q8cuzg64Cmr8RL4mhYd3W9ON-x7v0IueRszxkXn4_7hLCvWd3uOW9ZLd-QHe_lUDHJ2rdkx1jTVm3dyffkA-KRMd71ku_I30ejPdBZ5wwJr-njorPTnibA6NfsYqBuXrTJSHOKiys09aAnuhwgRB-fT9QcdCoAJPcClo4nOq8-uwrjmgxQ1Bm8dxmKco5lIAR04Zn-cflAdaBgNs0mrkwMmJN2oXgswLL9lMIGz9GC_0jeTdojfPo_L8jP229PNz-q-4fvdzdf7ytT93WuDBt60YlxqmsJg2YCBAy2KY-pkUIIaaWQjRzYaPtyKtsO48BgtK1pu14z21yQq7N3SfHXCpjV7NCULXSAuKJqeNdwIfqmLyg_oyZFxASTWpKbdTopztRWjDqqUozailHnYkrmyzkDZYffDpJC4yAYsC6BycpG90r6H5D6msw</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>3153166838</pqid></control><display><type>article</type><title>Scale matters: Spatial resolution impacts tropical leaf phenology characterized by multi-source satellite remote sensing with an ecological-constrained deep learning model</title><source>Elsevier ScienceDirect Journals</source><creator>Song, Guangqin ; Wang, Jing ; Zhao, Yingyi ; Yang, Dedi ; Lee, Calvin K.F. ; Guo, Zhengfei ; Detto, Matteo ; Alberton, Bruna ; Morellato, Patricia ; Nelson, Bruce ; Wu, Jin</creator><creatorcontrib>Song, Guangqin ; Wang, Jing ; Zhao, Yingyi ; Yang, Dedi ; Lee, Calvin K.F. ; Guo, Zhengfei ; Detto, Matteo ; Alberton, Bruna ; Morellato, Patricia ; Nelson, Bruce ; Wu, Jin</creatorcontrib><description>Accurate monitoring of tropical leaf phenology, such as the leaf-on/off status, at both individual and ecosystem scales is essential for understanding and modelling tropical forest carbon and water cycles, and their sensitivity to climate change. The discrepancy between tree-crown size and pixel size (i.e., spatial resolution) across orbital sensors can affect the capability of cross-scale phenology monitoring, an aspect that remains understudied. To examine the impact of spatial resolution on tropical leaf phenology monitoring, we applied a spectral index-guided, ecologically constrained autoencoder (IG-ECAE) to automatically generate a deciduousness metric (i.e., percentage of upper canopy area that is leaf-off status within an image pixel) from simulated VIS-NIR PlanetScope data at a range of resolutions from 3 m to 30 m, as well as from VIS-NIR data of three satellite platforms with the same range of spatial resolutions (3 m PlanetScope, 10 m Sentinel-2, and 30 m Landsat-8). We compared the deciduousness metrics derived from the simulated and satellite data to corresponding measurements derived from WorldView-2 (three sites) and local phenocams (four sites) at five tropical forest sites. Our results revealed that: (1) the IG-ECAE model captured the amount of deciduousness across spatial scales, with the highest accuracy obtained from PlanetScope, followed by Sentinel-2 and Landsat-8; (2) coarser spatial resolutions led to lower accuracies in tropical deciduousness monitoring, as demonstrated by both simulated PlanetScope data across various spatial resolutions and real satellite data; and (3) while not as accurate in capturing fine-scale tropical phenological diversity as PlanetScope, Sentinel-2 provided satisfactory monitoring of deciduousness seasonality at the ecosystem level consistently across all phenocam sites, whereas Landsat-8 failed to do so. Collectively, this study provides a robust assessment for advancing cross-scale tropical leaf phenology monitoring with potential for extension to pan-tropical regions and highlights the impact of spatial resolution on such monitoring efforts. •Spatial resolution affects phenology monitoring assessed by simulated and real data.•The spatial resolution's impact was evaluated with WorldView-2 and phenocam data.•Coarser resolution reduces phenology monitoring accuracy using deciduousness metric.•Sentinel-2(10 m) nears PlanetScope(3 m)’s ability for ecosystem phenology monitoring.•PlanetScope(3 m) remains vital for characterizing fine-scale phenological diversity.</description><identifier>ISSN: 0034-4257</identifier><identifier>EISSN: 1879-0704</identifier><identifier>DOI: 10.1016/j.rse.2024.114027</identifier><language>eng</language><publisher>Elsevier Inc</publisher><subject>canopy ; carbon ; climate change ; Deep learning ; Ecosystem deciduousness ; ecosystems ; environment ; Landsat ; Leaf phenology ; leaves ; Phenological diversity ; phenology ; remote sensing ; Satellite remote sensing ; Spatial resolution ; Spectral unmixing ; tree crown ; Tropical forest ; tropical forests</subject><ispartof>Remote sensing of environment, 2024-04, Vol.304, p.114027, Article 114027</ispartof><rights>2024 Elsevier Inc.</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><cites>FETCH-LOGICAL-c282t-c098656bf227e9a06e6e9d3a06f376667d7673790bd8140d49b90ebd4c458a0d3</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://www.sciencedirect.com/science/article/pii/S0034425724000385$$EHTML$$P50$$Gelsevier$$H</linktohtml><link.rule.ids>314,776,780,3537,27901,27902,65306</link.rule.ids></links><search><creatorcontrib>Song, Guangqin</creatorcontrib><creatorcontrib>Wang, Jing</creatorcontrib><creatorcontrib>Zhao, Yingyi</creatorcontrib><creatorcontrib>Yang, Dedi</creatorcontrib><creatorcontrib>Lee, Calvin K.F.</creatorcontrib><creatorcontrib>Guo, Zhengfei</creatorcontrib><creatorcontrib>Detto, Matteo</creatorcontrib><creatorcontrib>Alberton, Bruna</creatorcontrib><creatorcontrib>Morellato, Patricia</creatorcontrib><creatorcontrib>Nelson, Bruce</creatorcontrib><creatorcontrib>Wu, Jin</creatorcontrib><title>Scale matters: Spatial resolution impacts tropical leaf phenology characterized by multi-source satellite remote sensing with an ecological-constrained deep learning model</title><title>Remote sensing of environment</title><description>Accurate monitoring of tropical leaf phenology, such as the leaf-on/off status, at both individual and ecosystem scales is essential for understanding and modelling tropical forest carbon and water cycles, and their sensitivity to climate change. The discrepancy between tree-crown size and pixel size (i.e., spatial resolution) across orbital sensors can affect the capability of cross-scale phenology monitoring, an aspect that remains understudied. To examine the impact of spatial resolution on tropical leaf phenology monitoring, we applied a spectral index-guided, ecologically constrained autoencoder (IG-ECAE) to automatically generate a deciduousness metric (i.e., percentage of upper canopy area that is leaf-off status within an image pixel) from simulated VIS-NIR PlanetScope data at a range of resolutions from 3 m to 30 m, as well as from VIS-NIR data of three satellite platforms with the same range of spatial resolutions (3 m PlanetScope, 10 m Sentinel-2, and 30 m Landsat-8). We compared the deciduousness metrics derived from the simulated and satellite data to corresponding measurements derived from WorldView-2 (three sites) and local phenocams (four sites) at five tropical forest sites. Our results revealed that: (1) the IG-ECAE model captured the amount of deciduousness across spatial scales, with the highest accuracy obtained from PlanetScope, followed by Sentinel-2 and Landsat-8; (2) coarser spatial resolutions led to lower accuracies in tropical deciduousness monitoring, as demonstrated by both simulated PlanetScope data across various spatial resolutions and real satellite data; and (3) while not as accurate in capturing fine-scale tropical phenological diversity as PlanetScope, Sentinel-2 provided satisfactory monitoring of deciduousness seasonality at the ecosystem level consistently across all phenocam sites, whereas Landsat-8 failed to do so. Collectively, this study provides a robust assessment for advancing cross-scale tropical leaf phenology monitoring with potential for extension to pan-tropical regions and highlights the impact of spatial resolution on such monitoring efforts. •Spatial resolution affects phenology monitoring assessed by simulated and real data.•The spatial resolution's impact was evaluated with WorldView-2 and phenocam data.•Coarser resolution reduces phenology monitoring accuracy using deciduousness metric.•Sentinel-2(10 m) nears PlanetScope(3 m)’s ability for ecosystem phenology monitoring.•PlanetScope(3 m) remains vital for characterizing fine-scale phenological diversity.</description><subject>canopy</subject><subject>carbon</subject><subject>climate change</subject><subject>Deep learning</subject><subject>Ecosystem deciduousness</subject><subject>ecosystems</subject><subject>environment</subject><subject>Landsat</subject><subject>Leaf phenology</subject><subject>leaves</subject><subject>Phenological diversity</subject><subject>phenology</subject><subject>remote sensing</subject><subject>Satellite remote sensing</subject><subject>Spatial resolution</subject><subject>Spectral unmixing</subject><subject>tree crown</subject><subject>Tropical forest</subject><subject>tropical forests</subject><issn>0034-4257</issn><issn>1879-0704</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><recordid>eNp9kc1u1DAURi0EEkPhAbrzspsMdn7spKxQRaFSpS5a1pZj33Q8cuzg64Cmr8RL4mhYd3W9ON-x7v0IueRszxkXn4_7hLCvWd3uOW9ZLd-QHe_lUDHJ2rdkx1jTVm3dyffkA-KRMd71ku_I30ejPdBZ5wwJr-njorPTnibA6NfsYqBuXrTJSHOKiys09aAnuhwgRB-fT9QcdCoAJPcClo4nOq8-uwrjmgxQ1Bm8dxmKco5lIAR04Zn-cflAdaBgNs0mrkwMmJN2oXgswLL9lMIGz9GC_0jeTdojfPo_L8jP229PNz-q-4fvdzdf7ytT93WuDBt60YlxqmsJg2YCBAy2KY-pkUIIaaWQjRzYaPtyKtsO48BgtK1pu14z21yQq7N3SfHXCpjV7NCULXSAuKJqeNdwIfqmLyg_oyZFxASTWpKbdTopztRWjDqqUozailHnYkrmyzkDZYffDpJC4yAYsC6BycpG90r6H5D6msw</recordid><startdate>20240401</startdate><enddate>20240401</enddate><creator>Song, Guangqin</creator><creator>Wang, Jing</creator><creator>Zhao, Yingyi</creator><creator>Yang, Dedi</creator><creator>Lee, Calvin K.F.</creator><creator>Guo, Zhengfei</creator><creator>Detto, Matteo</creator><creator>Alberton, Bruna</creator><creator>Morellato, Patricia</creator><creator>Nelson, Bruce</creator><creator>Wu, Jin</creator><general>Elsevier Inc</general><scope>AAYXX</scope><scope>CITATION</scope><scope>7S9</scope><scope>L.6</scope></search><sort><creationdate>20240401</creationdate><title>Scale matters: Spatial resolution impacts tropical leaf phenology characterized by multi-source satellite remote sensing with an ecological-constrained deep learning model</title><author>Song, Guangqin ; Wang, Jing ; Zhao, Yingyi ; Yang, Dedi ; Lee, Calvin K.F. ; Guo, Zhengfei ; Detto, Matteo ; Alberton, Bruna ; Morellato, Patricia ; Nelson, Bruce ; Wu, Jin</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c282t-c098656bf227e9a06e6e9d3a06f376667d7673790bd8140d49b90ebd4c458a0d3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>canopy</topic><topic>carbon</topic><topic>climate change</topic><topic>Deep learning</topic><topic>Ecosystem deciduousness</topic><topic>ecosystems</topic><topic>environment</topic><topic>Landsat</topic><topic>Leaf phenology</topic><topic>leaves</topic><topic>Phenological diversity</topic><topic>phenology</topic><topic>remote sensing</topic><topic>Satellite remote sensing</topic><topic>Spatial resolution</topic><topic>Spectral unmixing</topic><topic>tree crown</topic><topic>Tropical forest</topic><topic>tropical forests</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Song, Guangqin</creatorcontrib><creatorcontrib>Wang, Jing</creatorcontrib><creatorcontrib>Zhao, Yingyi</creatorcontrib><creatorcontrib>Yang, Dedi</creatorcontrib><creatorcontrib>Lee, Calvin K.F.</creatorcontrib><creatorcontrib>Guo, Zhengfei</creatorcontrib><creatorcontrib>Detto, Matteo</creatorcontrib><creatorcontrib>Alberton, Bruna</creatorcontrib><creatorcontrib>Morellato, Patricia</creatorcontrib><creatorcontrib>Nelson, Bruce</creatorcontrib><creatorcontrib>Wu, Jin</creatorcontrib><collection>CrossRef</collection><collection>AGRICOLA</collection><collection>AGRICOLA - Academic</collection><jtitle>Remote sensing of environment</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Song, Guangqin</au><au>Wang, Jing</au><au>Zhao, Yingyi</au><au>Yang, Dedi</au><au>Lee, Calvin K.F.</au><au>Guo, Zhengfei</au><au>Detto, Matteo</au><au>Alberton, Bruna</au><au>Morellato, Patricia</au><au>Nelson, Bruce</au><au>Wu, Jin</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Scale matters: Spatial resolution impacts tropical leaf phenology characterized by multi-source satellite remote sensing with an ecological-constrained deep learning model</atitle><jtitle>Remote sensing of environment</jtitle><date>2024-04-01</date><risdate>2024</risdate><volume>304</volume><spage>114027</spage><pages>114027-</pages><artnum>114027</artnum><issn>0034-4257</issn><eissn>1879-0704</eissn><abstract>Accurate monitoring of tropical leaf phenology, such as the leaf-on/off status, at both individual and ecosystem scales is essential for understanding and modelling tropical forest carbon and water cycles, and their sensitivity to climate change. The discrepancy between tree-crown size and pixel size (i.e., spatial resolution) across orbital sensors can affect the capability of cross-scale phenology monitoring, an aspect that remains understudied. To examine the impact of spatial resolution on tropical leaf phenology monitoring, we applied a spectral index-guided, ecologically constrained autoencoder (IG-ECAE) to automatically generate a deciduousness metric (i.e., percentage of upper canopy area that is leaf-off status within an image pixel) from simulated VIS-NIR PlanetScope data at a range of resolutions from 3 m to 30 m, as well as from VIS-NIR data of three satellite platforms with the same range of spatial resolutions (3 m PlanetScope, 10 m Sentinel-2, and 30 m Landsat-8). We compared the deciduousness metrics derived from the simulated and satellite data to corresponding measurements derived from WorldView-2 (three sites) and local phenocams (four sites) at five tropical forest sites. Our results revealed that: (1) the IG-ECAE model captured the amount of deciduousness across spatial scales, with the highest accuracy obtained from PlanetScope, followed by Sentinel-2 and Landsat-8; (2) coarser spatial resolutions led to lower accuracies in tropical deciduousness monitoring, as demonstrated by both simulated PlanetScope data across various spatial resolutions and real satellite data; and (3) while not as accurate in capturing fine-scale tropical phenological diversity as PlanetScope, Sentinel-2 provided satisfactory monitoring of deciduousness seasonality at the ecosystem level consistently across all phenocam sites, whereas Landsat-8 failed to do so. Collectively, this study provides a robust assessment for advancing cross-scale tropical leaf phenology monitoring with potential for extension to pan-tropical regions and highlights the impact of spatial resolution on such monitoring efforts. •Spatial resolution affects phenology monitoring assessed by simulated and real data.•The spatial resolution's impact was evaluated with WorldView-2 and phenocam data.•Coarser resolution reduces phenology monitoring accuracy using deciduousness metric.•Sentinel-2(10 m) nears PlanetScope(3 m)’s ability for ecosystem phenology monitoring.•PlanetScope(3 m) remains vital for characterizing fine-scale phenological diversity.</abstract><pub>Elsevier Inc</pub><doi>10.1016/j.rse.2024.114027</doi></addata></record>
fulltext fulltext
identifier ISSN: 0034-4257
ispartof Remote sensing of environment, 2024-04, Vol.304, p.114027, Article 114027
issn 0034-4257
1879-0704
language eng
recordid cdi_proquest_miscellaneous_3153166838
source Elsevier ScienceDirect Journals
subjects canopy
carbon
climate change
Deep learning
Ecosystem deciduousness
ecosystems
environment
Landsat
Leaf phenology
leaves
Phenological diversity
phenology
remote sensing
Satellite remote sensing
Spatial resolution
Spectral unmixing
tree crown
Tropical forest
tropical forests
title Scale matters: Spatial resolution impacts tropical leaf phenology characterized by multi-source satellite remote sensing with an ecological-constrained deep learning model
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-07T17%3A24%3A37IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Scale%20matters:%20Spatial%20resolution%20impacts%20tropical%20leaf%20phenology%20characterized%20by%20multi-source%20satellite%20remote%20sensing%20with%20an%20ecological-constrained%20deep%20learning%20model&rft.jtitle=Remote%20sensing%20of%20environment&rft.au=Song,%20Guangqin&rft.date=2024-04-01&rft.volume=304&rft.spage=114027&rft.pages=114027-&rft.artnum=114027&rft.issn=0034-4257&rft.eissn=1879-0704&rft_id=info:doi/10.1016/j.rse.2024.114027&rft_dat=%3Cproquest_cross%3E3153166838%3C/proquest_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=3153166838&rft_id=info:pmid/&rft_els_id=S0034425724000385&rfr_iscdi=true