Correcting GEDI Water Level Estimates for Inland Waterbodies Using Machine Learning
The Global Ecosystem Dynamics Investigation (GEDI) LiDAR on the International Space Station has acquired more than 35 billion shots globally in the period between April 2019 and August 2021. The acquired shots could offer a significant database for the measure and monitoring of inland water levels o...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-05-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/14/10/2361 |
| _version_ | 1797495929642680320 |
|---|---|
| author | Ibrahim Fayad Nicolas Baghdadi Jean-Stéphane Bailly Frédéric Frappart Núria Pantaleoni Reluy |
| author_facet | Ibrahim Fayad Nicolas Baghdadi Jean-Stéphane Bailly Frédéric Frappart Núria Pantaleoni Reluy |
| author_sort | Ibrahim Fayad |
| collection | DOAJ |
| description | The Global Ecosystem Dynamics Investigation (GEDI) LiDAR on the International Space Station has acquired more than 35 billion shots globally in the period between April 2019 and August 2021. The acquired shots could offer a significant database for the measure and monitoring of inland water levels over the Earth’s surface. Nonetheless, previous and current studies have shown that the provided GEDI elevation estimates are significantly less accurate than any available radar or LiDAR altimeter. Indeed, our analysis of GEDI’s altimetric capabilities to retrieve water levels over the five North American Great Lakes presented estimates with a bias that ranged between 0.26 and 0.35 m and a root mean squared error (RMSE) ranging between 0.54 and 0.68 m. Therefore, our objective in this study is to post-process the original GEDI water level estimates from an error model taking instrumental, atmospheric, and lakes surface state factors as proxies, which affect the physical shape of the waveforms, hence introducing uncertainties on the elevation estimates. The first tested model, namely a random forest regressor (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>R</mi><msub><mi>F</mi><mrow><mi>I</mi><mi>C</mi><mi>W</mi></mrow></msub></mrow></semantics></math></inline-formula>) with the instrumental, atmospheric, and water surface state factors as inputs, was validated temporally (trained on a given year and validated on another) and spatially (trained on a given lake and validated on the remaining four). The results showed a significant decrease in elevation estimation errors both temporally and spatially. The temporally validated models showed an RMSE on the corrected elevation estimates of 0.18 m. Concerning the spatially validated model, the results varied based on the lake data used for training. Indeed, the most accurate spatially validated model showed an RMSE of 0.17 m, while the least accurate model showed an RMSE of 0.26 m. Finally, given that an elevation correction model using all the factors (instrumental, atmospheric, and water surface state factors) presents a best-case scenario, as water surface state factors are only available over a selected number of lakes globally, three additional models based on random forest were tested. The first,<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mo> </mo><mi>R</mi><msub><mi>F</mi><mi>I</mi></msub></mrow></semantics></math></inline-formula>, uses only instrumental factors as correction factors, <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>R</mi><msub><mi>F</mi><mrow><mi>I</mi><mi>C</mi></mrow></msub></mrow></semantics></math></inline-formula> uses both instrumental and atmospheric factors, while the third, <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>R</mi><msub><mi>F</mi><mrow><mi>I</mi><mi>W</mi></mrow></msub></mrow></semantics></math></inline-formula>, uses instrumental and water surface state factors. The temporal validation of these models showed that the model using instrumental factors, while less accurate than the remaining two models, was capable of correcting the original GEDI elevation estimates by a factor of two across the five lakes. On the other hand, the <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>R</mi><msub><mi>F</mi><mrow><mi>I</mi><mi>C</mi></mrow></msub></mrow></semantics></math></inline-formula> model was the most accurate between the three, with a slight degradation in comparison to the full model. Indeed, the <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>R</mi><msub><mi>F</mi><mrow><mi>I</mi><mi>C</mi></mrow></msub></mrow></semantics></math></inline-formula> model showed an RMSE on the estimation of water levels of 0.21 m. |
| first_indexed | 2024-03-10T01:56:33Z |
| format | Article |
| id | doaj.art-09f6a37256d441ca94e2ddf153dc6950 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T01:56:33Z |
| publishDate | 2022-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-09f6a37256d441ca94e2ddf153dc69502023-11-23T12:55:00ZengMDPI AGRemote Sensing2072-42922022-05-011410236110.3390/rs14102361Correcting GEDI Water Level Estimates for Inland Waterbodies Using Machine LearningIbrahim Fayad0Nicolas Baghdadi1Jean-Stéphane Bailly2Frédéric Frappart3Núria Pantaleoni Reluy4TETIS, University of Montpellier, AgroParisTech, CIRAD, CNRS, INRAE, CEDEX 5, 34093 Montpellier, FranceTETIS, University of Montpellier, AgroParisTech, CIRAD, CNRS, INRAE, CEDEX 5, 34093 Montpellier, FranceLISAH, University of Montpellier, INRAE, IRD, Institut Agro, 34060 Montpellier, FranceINRAE, Bordeaux Sciences Agro, UMR1391 ISPA, 71 Avenue Edouard Bourlaux, 33140 Villenave d’Ornon, FranceTETIS, University of Montpellier, AgroParisTech, CIRAD, CNRS, INRAE, CEDEX 5, 34093 Montpellier, FranceThe Global Ecosystem Dynamics Investigation (GEDI) LiDAR on the International Space Station has acquired more than 35 billion shots globally in the period between April 2019 and August 2021. The acquired shots could offer a significant database for the measure and monitoring of inland water levels over the Earth’s surface. Nonetheless, previous and current studies have shown that the provided GEDI elevation estimates are significantly less accurate than any available radar or LiDAR altimeter. Indeed, our analysis of GEDI’s altimetric capabilities to retrieve water levels over the five North American Great Lakes presented estimates with a bias that ranged between 0.26 and 0.35 m and a root mean squared error (RMSE) ranging between 0.54 and 0.68 m. Therefore, our objective in this study is to post-process the original GEDI water level estimates from an error model taking instrumental, atmospheric, and lakes surface state factors as proxies, which affect the physical shape of the waveforms, hence introducing uncertainties on the elevation estimates. The first tested model, namely a random forest regressor (<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>R</mi><msub><mi>F</mi><mrow><mi>I</mi><mi>C</mi><mi>W</mi></mrow></msub></mrow></semantics></math></inline-formula>) with the instrumental, atmospheric, and water surface state factors as inputs, was validated temporally (trained on a given year and validated on another) and spatially (trained on a given lake and validated on the remaining four). The results showed a significant decrease in elevation estimation errors both temporally and spatially. The temporally validated models showed an RMSE on the corrected elevation estimates of 0.18 m. Concerning the spatially validated model, the results varied based on the lake data used for training. Indeed, the most accurate spatially validated model showed an RMSE of 0.17 m, while the least accurate model showed an RMSE of 0.26 m. Finally, given that an elevation correction model using all the factors (instrumental, atmospheric, and water surface state factors) presents a best-case scenario, as water surface state factors are only available over a selected number of lakes globally, three additional models based on random forest were tested. The first,<inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mo> </mo><mi>R</mi><msub><mi>F</mi><mi>I</mi></msub></mrow></semantics></math></inline-formula>, uses only instrumental factors as correction factors, <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>R</mi><msub><mi>F</mi><mrow><mi>I</mi><mi>C</mi></mrow></msub></mrow></semantics></math></inline-formula> uses both instrumental and atmospheric factors, while the third, <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>R</mi><msub><mi>F</mi><mrow><mi>I</mi><mi>W</mi></mrow></msub></mrow></semantics></math></inline-formula>, uses instrumental and water surface state factors. The temporal validation of these models showed that the model using instrumental factors, while less accurate than the remaining two models, was capable of correcting the original GEDI elevation estimates by a factor of two across the five lakes. On the other hand, the <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>R</mi><msub><mi>F</mi><mrow><mi>I</mi><mi>C</mi></mrow></msub></mrow></semantics></math></inline-formula> model was the most accurate between the three, with a slight degradation in comparison to the full model. Indeed, the <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mi>R</mi><msub><mi>F</mi><mrow><mi>I</mi><mi>C</mi></mrow></msub></mrow></semantics></math></inline-formula> model showed an RMSE on the estimation of water levels of 0.21 m.https://www.mdpi.com/2072-4292/14/10/2361GEDILiDARgreat lakeswater levels correctionrandom forest |
| spellingShingle | Ibrahim Fayad Nicolas Baghdadi Jean-Stéphane Bailly Frédéric Frappart Núria Pantaleoni Reluy Correcting GEDI Water Level Estimates for Inland Waterbodies Using Machine Learning Remote Sensing GEDI LiDAR great lakes water levels correction random forest |
| title | Correcting GEDI Water Level Estimates for Inland Waterbodies Using Machine Learning |
| title_full | Correcting GEDI Water Level Estimates for Inland Waterbodies Using Machine Learning |
| title_fullStr | Correcting GEDI Water Level Estimates for Inland Waterbodies Using Machine Learning |
| title_full_unstemmed | Correcting GEDI Water Level Estimates for Inland Waterbodies Using Machine Learning |
| title_short | Correcting GEDI Water Level Estimates for Inland Waterbodies Using Machine Learning |
| title_sort | correcting gedi water level estimates for inland waterbodies using machine learning |
| topic | GEDI LiDAR great lakes water levels correction random forest |
| url | https://www.mdpi.com/2072-4292/14/10/2361 |
| work_keys_str_mv | AT ibrahimfayad correctinggediwaterlevelestimatesforinlandwaterbodiesusingmachinelearning AT nicolasbaghdadi correctinggediwaterlevelestimatesforinlandwaterbodiesusingmachinelearning AT jeanstephanebailly correctinggediwaterlevelestimatesforinlandwaterbodiesusingmachinelearning AT fredericfrappart correctinggediwaterlevelestimatesforinlandwaterbodiesusingmachinelearning AT nuriapantaleonireluy correctinggediwaterlevelestimatesforinlandwaterbodiesusingmachinelearning |