| |
|
|
|
Registro completo
|
|
Biblioteca (s) : |
INIA Las Brujas. |
|
Fecha : |
15/05/2024 |
|
Actualizado : |
15/05/2024 |
|
Tipo de producción científica : |
Artículos en Revistas Indexadas Internacionales |
|
Autor : |
GASO, D.; PAUDEL, D.; DE WIT, A.; PUNTEL, L.A.; MULLISSA, A.; KOOISTRA, L. |
|
Afiliación : |
DEBORAH VIVIANA GASO MELGAR, INIA (Instituto Nacional de Investigación Agropecuaria), Uruguay; Laboratory of Geo-Information Science and Remote Sensing, Wageningen University and Research, Wageningen, 6708PB, Netherlands; DILLI PAUDEL, Laboratory of Geo-Information Science and Remote Sensing, Wageningen University and Research, Wageningen, 6708PB, Netherlands; ALLARD DE WIT, Wageningen Environmental Research, Wageningen, 6708PB, Netherlands; LAILA A. PUNTEL, Department of Agronomy and Horticulture, University of Nebraska-Lincoln, Keim Hall, 1825N 38th Street, Lincoln, 68583-0915, NE, United States; ADUGNA MULLISSA, Laboratory of Geo-Information Science and Remote Sensing, Wageningen University and Research, Wageningen, 6708PB, Netherlands; LAMMERT KOOISTRA, Laboratory of Geo-Information Science and Remote Sensing, Wageningen University and Research, Wageningen, 6708PB, Netherlands. |
|
Título : |
Beyond assimilation of leaf area index: Leveraging additional spectral information using machine learning for site-specific soybean yield prediction. |
|
Fecha de publicación : |
2024 |
|
Fuente / Imprenta : |
Agricultural and Forest Meteorology. 2024, Volume 35, article 110022. https://doi.org/10.1016/j.agrformet.2024.110022 -- OPEN ACCESS. |
|
ISSN : |
0168-1923 |
|
DOI : |
10.1016/j.agrformet.2024.110022 |
|
Idioma : |
Inglés |
|
Notas : |
Article history: Received 24 October 2023, Revised 6 February 2024, Accepted 18 April 2024, Available online 21 April 2024, Version of Record 21 April 2024. -- Correspondence: Gaso, D.V.; Laboratory of Geo-Information Science and Remote Sensing, Wageningen University and Research, Wageningen, Netherlands; email:deborah.gasomelgar@wur.nl -- Funding: This research was funded by the Instituto Nacional de Investigación Agropecuaria de Uruguay and a Ph.D. fellowship provided by Agencia Nacional de Investigación e Innovación (ANII, scholarship code: POS_EXT_2017_1_147121). We would like to thank ProNutrition Agrotecnologías, USDA-NRCS Conservation Innovation Grant (Award Number NR213A7500013G021) and USDA NIFA-AFRI Food Security Program Coordinated Agricultural Project for sharing the field data. -- License: This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). |
|
Contenido : |
ABSTRACT.- Assimilating external observations of crop state in cropping system models is essential for making spatially explicit predictions of crop variables relevant in precision agriculture. Satellite-based leaf area index (LAI) estimates have been the most frequent variable used as a proxy of actual crop growth. However, additional information beyond LAI, like canopy N content, water content, and structure, can be retrieved from satellite observations. Including such variables by data assimilation directly is difficult because many crop models do not have corresponding state variables or the relationship between the observations and the process that regulates crop growth is unclear. Therefore, other approaches are required to include such information. In this study, we investigate the improvement in the predicted yield and feature impact on model outputs by using a hybrid approach that combines observations from Sentinel-1 and 2 time-series with the outputs from a process-based model embedded in a data assimilation framework and uses the Gradient-boosted trees regressor (GBTR) as predictive model. We used two regions with soybean fields: the US (13 K points) and Uruguay (400 K points). We found an advantage when using the GBTR as the predictive model (reduced RRMSE by ~16%) compared to data assimilation. Adding the vegetation indices had a marginal improvement (reduced RRMSE by ~1%), while the impact of adding reflectance and backscatter values was negative. The satellite-based features had a very small importance score, while features' impact on prediction was predominantly unclear, explaining the marginal predictive power added by satellite-based features. We found that features from the reproductive stages had the highest importance, while the importance of an index related to drought stress (NMDI) across the growing season provided insights for further improvement of data assimilation methods. However, more studies are required to better disentangle pathways towards further improvement in constraining crop models by ingesting satellite observations. © 2024 MenosABSTRACT.- Assimilating external observations of crop state in cropping system models is essential for making spatially explicit predictions of crop variables relevant in precision agriculture. Satellite-based leaf area index (LAI) estimates have been the most frequent variable used as a proxy of actual crop growth. However, additional information beyond LAI, like canopy N content, water content, and structure, can be retrieved from satellite observations. Including such variables by data assimilation directly is difficult because many crop models do not have corresponding state variables or the relationship between the observations and the process that regulates crop growth is unclear. Therefore, other approaches are required to include such information. In this study, we investigate the improvement in the predicted yield and feature impact on model outputs by using a hybrid approach that combines observations from Sentinel-1 and 2 time-series with the outputs from a process-based model embedded in a data assimilation framework and uses the Gradient-boosted trees regressor (GBTR) as predictive model. We used two regions with soybean fields: the US (13 K points) and Uruguay (400 K points). We found an advantage when using the GBTR as the predictive model (reduced RRMSE by ~16%) compared to data assimilation. Adding the vegetation indices had a marginal improvement (reduced RRMSE by ~1%), while the impact of adding reflectance and backscatter values was negative. The satellit... Presentar Todo |
|
Palabras claves : |
Crop modeling; Crops models; Data assimilation; Machine learning; Partnership for the goals - Goal 17; Remote sensing; Soybean; Sustainable Development Goals (SDGs); Zero hunger - Goal 2. |
|
Asunto categoría : |
F01 Cultivo |
|
URL : |
https://www.sciencedirect.com/science/article/pii/S0168192324001370/pdf
|
|
Marc : |
LEADER 04106naa a2200325 a 4500
001 1064621
005 2024-05-15
008 2024 bl uuuu u00u1 u #d
022 $a0168-1923
024 7 $a10.1016/j.agrformet.2024.110022$2DOI
100 1 $aGASO, D.
245 $aBeyond assimilation of leaf area index$bLeveraging additional spectral information using machine learning for site-specific soybean yield prediction.$h[electronic resource]
260 $c2024
500 $aArticle history: Received 24 October 2023, Revised 6 February 2024, Accepted 18 April 2024, Available online 21 April 2024, Version of Record 21 April 2024. -- Correspondence: Gaso, D.V.; Laboratory of Geo-Information Science and Remote Sensing, Wageningen University and Research, Wageningen, Netherlands; email:deborah.gasomelgar@wur.nl -- Funding: This research was funded by the Instituto Nacional de Investigación Agropecuaria de Uruguay and a Ph.D. fellowship provided by Agencia Nacional de Investigación e Innovación (ANII, scholarship code: POS_EXT_2017_1_147121). We would like to thank ProNutrition Agrotecnologías, USDA-NRCS Conservation Innovation Grant (Award Number NR213A7500013G021) and USDA NIFA-AFRI Food Security Program Coordinated Agricultural Project for sharing the field data. -- License: This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
520 $aABSTRACT.- Assimilating external observations of crop state in cropping system models is essential for making spatially explicit predictions of crop variables relevant in precision agriculture. Satellite-based leaf area index (LAI) estimates have been the most frequent variable used as a proxy of actual crop growth. However, additional information beyond LAI, like canopy N content, water content, and structure, can be retrieved from satellite observations. Including such variables by data assimilation directly is difficult because many crop models do not have corresponding state variables or the relationship between the observations and the process that regulates crop growth is unclear. Therefore, other approaches are required to include such information. In this study, we investigate the improvement in the predicted yield and feature impact on model outputs by using a hybrid approach that combines observations from Sentinel-1 and 2 time-series with the outputs from a process-based model embedded in a data assimilation framework and uses the Gradient-boosted trees regressor (GBTR) as predictive model. We used two regions with soybean fields: the US (13 K points) and Uruguay (400 K points). We found an advantage when using the GBTR as the predictive model (reduced RRMSE by ~16%) compared to data assimilation. Adding the vegetation indices had a marginal improvement (reduced RRMSE by ~1%), while the impact of adding reflectance and backscatter values was negative. The satellite-based features had a very small importance score, while features' impact on prediction was predominantly unclear, explaining the marginal predictive power added by satellite-based features. We found that features from the reproductive stages had the highest importance, while the importance of an index related to drought stress (NMDI) across the growing season provided insights for further improvement of data assimilation methods. However, more studies are required to better disentangle pathways towards further improvement in constraining crop models by ingesting satellite observations. © 2024
653 $aCrop modeling
653 $aCrops models
653 $aData assimilation
653 $aMachine learning
653 $aPartnership for the goals - Goal 17
653 $aRemote sensing
653 $aSoybean
653 $aSustainable Development Goals (SDGs)
653 $aZero hunger - Goal 2
700 1 $aPAUDEL, D.
700 1 $aDE WIT, A.
700 1 $aPUNTEL, L.A.
700 1 $aMULLISSA, A.
700 1 $aKOOISTRA, L.
773 $tAgricultural and Forest Meteorology. 2024, Volume 35, article 110022. https://doi.org/10.1016/j.agrformet.2024.110022 -- OPEN ACCESS.
Descargar
Esconder MarcPresentar Marc Completo |
|
Registro original : |
INIA Las Brujas (LB) |
|
|
Biblioteca
|
Identificación
|
Origen
|
Tipo / Formato
|
Clasificación
|
Cutter
|
Registro
|
Volumen
|
Estado
|
Volver
|
|
|
| Registros recuperados : 4 | |
| 1. |  | GASO, D.; PAUDEL, D.; DE WIT, A.; PUNTEL, L.A.; MULLISSA, A.; KOOISTRA, L. Beyond assimilation of leaf area index: Leveraging additional spectral information using machine learning for site-specific soybean yield prediction. Agricultural and Forest Meteorology. 2024, Volume 35, article 110022. https://doi.org/10.1016/j.agrformet.2024.110022 -- OPEN ACCESS. Article history: Received 24 October 2023, Revised 6 February 2024, Accepted 18 April 2024, Available online 21 April 2024, Version of Record 21 April 2024. -- Correspondence: Gaso, D.V.; Laboratory of Geo-Information Science and Remote...| Tipo: Artículos en Revistas Indexadas Internacionales | Circulación / Nivel : Internacional - -- |
| Biblioteca(s): INIA Las Brujas. |
|    |
| 2. | | GASO, D.; DE WIT, A.; DE BRUIN, S.; PUNTEL, L.A.; BERGER, A.; KOOISTRA, L. Efficiency of assimilating leaf area index into a soybean model to assess within-field yield variability. European Journal of Agronomy, February 2023, Volume 143, 126718. OPEN ACCESS. doi: https://doi.org/10.1016/j.eja.2022.126718 Article history: Received 7 March 2022, Revised 17 October 2022, Accepted 5 December 2022, Available online 22 December 2022, Version of Record 22 December 2022. -- Corresponding author: Deborah Gaso, E-mail addresses:...| Tipo: Artículos en Revistas Indexadas Internacionales | Circulación / Nivel : Internacional - -- |
| Biblioteca(s): INIA Las Brujas. |
| |
| 3. | | PUNTEL, L.A.; BOLFE, E.L.; MELCHIORI, R.J.M.; ORTEGA, R.; TISCORNIA, G.; ROEL, A.; SCARAMUZZA, F.; BEST, S.; BERGER, A.; HANSEL, D.S.S.; PALACIOS, D.; BALBOA, G. How digital is agriculture in South America? Adoption and limitations. [PC- ICPA 2022]. In: INTERNATIONAL CONFERENCE ON PRECISION AGRICULTURE, 15., 2022, Minneapolis. Proceedings... [Monticello]: International Society of Precision Agriculture, 2022 p. 1-10. ICPA 2022.| Tipo: Trabajos en Congresos/Conferencias |
| Biblioteca(s): INIA Treinta y Tres. |
|  |
| 4. | | PUNTEL, L.A.; BOLFE, E.L.; MELCHIORI, R.J.M.; ORTEGA, R.; TISCORNIA, G.; ROEL, A.; SCARAMUZZA, F.; BEST, S.; BERGER, A.; HANSEL, D.S.S.; PALACIOS DURÁN, D.; BALBOA, G.R. How digital is agriculture in a subset of countries from South America? Adoption and limitations. Crop and Pasture Science, 2022, Special Issue, Review. CP21759. Open Access. doi: https://doi.org/10.1071/CP21759 Article history: Submitted: 9 November 2021 Accepted: 13 July 2022 Published online: 16 September 2022.
Correspondence to: L.A. Puntel Department of Agronomy and Horticulture, University of Nebraska, Lincoln, NE, USA Email:...| Tipo: Artículos en Revistas Indexadas Internacionales | Circulación / Nivel : Internacional - -- |
| Biblioteca(s): INIA Treinta y Tres. |
| |
| Registros recuperados : 4 | |
|
| Expresión de búsqueda válido. Check! |
|
|
