Oil palm age estimation based on vegetation and moisture indices using satellite imagery and machine learning
Article Sidebar
Main Article Content
Abstract
Palm oil (Elaeis guineensis) is one of the most widely consumed vegetable oils in the world and is a key ingredient in profitable global value chains. In Costa Rica, it is one of the three crops with the largest occupied agricultural area. The objective of this study is to estimate the age of the palm oil crop based on vegetation and humidity indices generated with satellite images using machine learning techniques, for this information will be taken from 20 years of a palm oil crop located in the Central Pacific of Costa Rica and its LANDSAT satellite images. Two models were generated, the Random Forest (RF) model presented better results when using a ntree of 10000 in training, obtaining the following coefficients; NSE=0,9161, RMSE=1,6485 years, MAE=1,0178 years and r2=0,9476; and the XGBoost model using a max.depth of 4, nrounds of 50 and nthread of 10 in the training for which it was obtained: NSE=0,9530, RMSE=1,23 years, MAE=0,89 years and r2=0,9577. For the data obtained in the validation, the RF coefficients were: NSE=0,5395, RMSE=3,76 years, MAE= 2,46 years and r2=0,5533; for the XGBoost model NSE=0,6204, RMSE=3,4114 years, MAE= 2,23 years and r2=0,6210. The most important variable in both models was the NDMI 75th percentile. The most suitable model with the best performance is XGBoost, however it is not recommended when the age of the crop exceeds 25 years.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Los autores conservan los derechos de autor y ceden a la revista el derecho de la primera publicación y pueda editarlo, reproducirlo, distribuirlo, exhibirlo y comunicarlo en el país y en el extranjero mediante medios impresos y electrónicos. Asimismo, asumen el compromiso sobre cualquier litigio o reclamación relacionada con derechos de propiedad intelectual, exonerando de responsabilidad a la Editorial Tecnológica de Costa Rica. Además, se establece que los autores pueden realizar otros acuerdos contractuales independientes y adicionales para la distribución no exclusiva de la versión del artículo publicado en esta revista (p. ej., incluirlo en un repositorio institucional o publicarlo en un libro) siempre que indiquen claramente que el trabajo se publicó por primera vez en esta revista.
References
[1] A. Descals, D. L. A. Gaveau, S. Wich, Z. Szantoi, y E. Meijaard, “Global mapping of oil palm planting year from 1990 to 2021”, Earth Syst Sci Data, vol. 16, núm. 11, pp. 5111–5129, nov. 2024, doi: 10.5194/essd-16-5111-2024.
[2] P. Clare, “El cultivo de la palma aceitera en Costa Rica en el contexto del TLC con los Estados unidos de Norteamérica”, Revista Espiga, vol. 5, núm. 9, pp. 95–124, 2004, doi: 10.22458/re.v5i9.1108.
[3] P. N. Chowdhury, P. Shivakumara, L. Nandanwar, F. Samiron, U. Pal, y T. Lu, “Oil palm tree counting in drone images”, Pattern Recognit Lett, vol. 153, pp. 1–9, 2022, doi: 10.1016/j.patrec.2021.11.016.
[4] K. L. Chong, K. D. Kanniah, C. Pohl, y K. P. Tan, “A review of remote sensing applications for oil palm studies”, Geo-Spatial Information Science, vol. 20, núm. 2, pp. 184–200, 2017, doi: 10.1080/10095020.2017.1337317.
[5] C. Li et al., “Integrating Multi-Temporal Landsat and Sentinel Data for Enhanced Oil Palm Plantation Mapping and Age Estimation in Malaysia”, Remote Sens (Basel), vol. 17, núm. 16, p. 2908, ago. 2025, doi: 10.3390/rs17162908.
[6] C. CHARLOQ, A. S. THOHA, A. YAZID, A. M. SIDIQ, y O. A. LUBIS, “Estimating oil palm (Elaeis guineensis) production potential using Unmanned Aerial Vehicle”, Biodiversitas, vol. 25, núm. 9, sep. 2024, doi: 10.13057/biodiv/d250919.
[7] A. C. Morel, J. B. Fisher, y Y. Malhi, “Evaluating the potential to monitor aboveground biomass in forest and oil palm in Sabah, Malaysia, for 2000–2008 with Landsat ETM+ and ALOS-PALSAR”, Int J Remote Sens, vol. 33, núm. 11, pp. 3614–3639, 2012, doi: 10.1080/01431161.2011.631949.
[8] A. Rodríguez et al., “Implications of crop model ensemble size and composition for estimates of adaptation effects and agreement of recommendations”, Agric For Meteorol, vol. 264, pp. 351–362, 2019, doi: 10.1016/j.agrformet.2018.09.018.
[9] S. Khanal, J. Fulton, A. Klopfenstein, N. Douridas, y S. Shearer, “Integration of high resolution remotely sensed data and machine learning techniques for spatial prediction of soil properties and corn yield”, Comput Electron Agric, vol. 153, pp. 213–225, 2018, doi: 10.1016/j.compag.2018.07.016.
[10] A. D. Amirruddin, F. M. Muharam, M. H. Ismail, N. P. Tan, y M. F. Ismail, “Hyperspectral spectroscopy and imbalance data approaches for classification of oil palm’s macronutrients observed from frond 9 and 17”, Comput Electron Agric, vol. 178, p. 105768, 2020, doi: 10.1016/j.compag.2020.105768.
[11] J. Zheng et al., “Growing status observation for oil palm trees using Unmanned Aerial Vehicle (UAV) images”, ISPRS Journal of Photogrammetry and Remote Sensing, vol. 173, pp. 95–121, 2021, doi: 10.1016/j.isprsjprs.2021.01.008.
[12] J.-H. RYU, D. OH, y J. CHO, “Simple method for extracting the seasonal signals of photochemical reflectance index and normalized difference vegetation index measured using a spectral reflectance sensor”, J Integr Agric, vol. 20, núm. 7, pp. 1969–1986, 2021, doi: 10.1016/S2095-3119(20)63410-4.
[13] Q. Zhao, L. Yu, Z. Du, y K. Kanniah, “Towards an Annually Updated Oil Palm Age Database (OPAD) with Multimodal Remote Sensing Features: Design and Preliminary Result”, Journal of Advanced Geospatial Science & Technology, vol. 4, núm. 1, pp. 16–42, mar. 2024, doi: 10.11113/jagst.v4n1.85.
[14] F. Watson-Hernández, N. Gómez-Calderón, y R. P. da Silva, “Oil Palm Yield Estimation Based on Vegetation and Humidity Indices Generated from Satellite Images and Machine Learning Techniques”, AgriEngineering, vol. 4, núm. 1, pp. 279–291, mar. 2022, doi: 10.3390/agriengineering4010019.
[15] INEC, VI Censo Nacional Agropecuario: Características de las Fincas y de las Personas Productoras. San José, Costa Rica, 2015.
[16] G. Chander, B. L. Markham, y D. L. Helder, “Summary of current radiometric calibration coefficients for Landsat MSS, TM, ETM+, and EO-1 ALI sensors”, Remote Sens Environ, vol. 113, núm. 5, pp. 893–903, 2009, doi: 10.1016/j.rse.2009.01.007.
[17] D. Ramos, “Combinación de bandas en imágenes de satélite Landsat y Sentinel”, 2021. Consultado: el 16 de diciembre de 2021. [En línea]. Disponible en: https://mappinggis.com/2019/05/combinaciones-de-bandas-en-imagenes-de-satelite-landsat-y-sentinel
[18] EOS (Earth Observing System), “LANDSAT 7”, 1999. Consultado: el 14 de diciembre de 2021. [En línea]. Disponible en: https://eos.com/find-satellite/landsat-7
[19] EOS (Earth Observing System), “LANDSAT 8”, 2013. Consultado: el 14 de diciembre de 2021. [En línea]. Disponible en: https://eos.com/find-satellite/landsat-8
[20] EOS (Earth Observing System), “LANDSAT 5 (TM)”, 2012. Consultado: el 14 de diciembre de 2021. [En línea]. Disponible en: https://eos.com/find-satellite/landsat-5-tm
[21] T. Dube, S. Pandit, C. Shoko, A. Ramoelo, D. Mazvimavi, y T. Dalu, “Numerical assessments of leaf area index in tropical savanna rangelands, South Africa using Landsat 8 OLI derived metrics and in-situ measurements”, Remote Sens (Basel), vol. 11, núm. 7, 2019, doi: 10.3390/rs11070829.
[22] D. Sykas, “Remote Sensing Satellite Data and Spectral Indices. Spectral Indices with multispectral satellite data”, 2020. Consultado: el 16 de diciembre de 2021. [En línea]. Disponible en: https://www.geo.university/pages/blog?p=spectral-indices-with-multispectral-satellite-data#:~:text=Bare
[23] C. Tan et al., “Predicting grain protein content of field-grown winter wheat with satellite images and partial least square algorithm”, PLoS One, vol. 15, núm. 3, pp. 1–15, 2020, doi: 10.1371/journal.pone.0228500.
[24] M. Zakir, “GIS and RS Based Spatio-Temporal Analysis of Soil Moisture/Water Content Variation in Southern Irrigated Part of Sindh, Pakistan”, Journal of Remote Sensing & GIS, vol. 07, núm. 04, 2018, doi: 10.4172/2469-4134.1000250.
[25] R Studio, “R Studio”, 2021. Consultado: el 9 de diciembre de 2021. [En línea]. Disponible en: https://www.rstudio.com
[26] D. Moriasi, J. Arnold, M. Van-Liew, R. Bingner, R. Harmel, y T. Veith, “Model Evaluation Guidelines for Systematic Quantification of Accuracy in Watershed Simulations”, Trans ASABE, vol. 50, núm. 3, pp. 885–900, 2007, doi: 10.13031/2013.23153.
[27] A. Chemura, I. van Duren, y L. M. van Leeuwen, “Determination of the age of oil palm from crown projection area detected from WorldView-2 multispectral remote sensing data: The case of Ejisu-Juaben district, Ghana”, ISPRS Journal of Photogrammetry and Remote Sensing, vol. 100, pp. 118–127, 2015, doi: 10.1016/j.isprsjprs.2014.07.013.
[28] K. P. Tan, K. D. Kanniah, y A. P. Cracknell, “Use of UK-DMC 2 and ALOS PALSAR for studying the age of oil palm trees in southern peninsular Malaysia”, Int J Remote Sens, vol. 34, núm. 20, pp. 7424–7446, 2013, doi: 10.1080/01431161.2013.822601.
[29] J. McMorrow, “Linear regression modelling for the estimation of oil palm age from Landsat TM”, Int J Remote Sens, vol. 22, núm. 12, pp. 2243–2264, 2001, doi: 10.1080/01431160117188.
[30] H. M. Rizeei, H. Z. M. Shafri, M. A. Mohamoud, B. Pradhan, y B. Kalantar, “Oil Palm Counting and Age Estimation from WorldView-3 Imagery and LiDAR Data Using an Integrated OBIA Height Model and Regression Analysis”, J Sens, pp. 1–13, 2018, doi: 10.1155/2018/2536327.